FDA Workshop

Licensure of Apheresis Blood Products

August 15, 2007

Center for Biologics Evaluation and Research
HHS Office of the Secretary/Office of Public Health and Science
AABB
America's Blood Centers

Lister Hill Auditorium
National Institutes of Health
Bethesda, Maryland

Printable Transcript (PDF, 396 KB)
Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®.

Table of Contents

Welcoming Remarks - Jesse Goodman
The Regulatory Paradigm and Managed Review - Alan Williams
FDA Review of Apheresis Submissions - Judy Ciaraldi
Top Ten Pitfalls with Submissions - Rosia Nesbitt
Preparing Apheresis License Submission - Stephen Kassapian
Blood Systems' Licensure of 5 and 7-Day Platelets - Kathleen Hopping
FDA Guidance on Collection of Platelets by Automated Methods - Lore Fields
Revisions to the Requirements Applicable to Blood, Blood Components, and Source Plasma - Elizabeth Callaghan
Impact of Very Frequent Plateletpheresis on Donor Platelet Counts - Louis Katz
FDA Regulations and Recommendations for Failure Investigations - Hoi-May Wong
Experiences with a Failure Investigation Program for Apheresis Blood Products - Faye Kugele
Device Manufacturer's Forum
Merilyn Wiler, Gambro BCT
Sharyn Orton, Fenwal
Sue Finneran, Haemonetics
Frequently Asked Questions and Answers

P R O C E E D I N G S

MS. FIELDS: Good morning. My name is Lore Fields, and I want to welcome everybody to the Licensure of Apheresis Blood Products Workshop.

I need to start with a few housekeeping issues.

[Administrative announcements]

We are going to have welcoming remarks this morning by Dr. Goodman. Dr. Goodman is the director of the Center for Biologics Evaluation and Research.

Agenda Item: Welcoming Remarks

DR. GOODMAN: I will try to be very brief, because I know you have a lot of technical material to cover here today, and I think that's the main reason for being here.

On this slide we always kind of show the vision. I mostly want to take the opportunity, in showing this, to thank you folks in the blood community who maintain quality and work with us and deliver this blood system to the American people. We usually are only interacting when there are problems. But I think you can feel pretty proud about the accomplishments and the way we have generally worked together well in achieving your part of these goals: a healthy public and a blood supply that contributes to their health, and where we have avoided or prevented a number of threats or mishaps from occurring.

In addition, you really have worked with us, I think, to improve what we do and improve the industry and the technologies. I think this workshop is in that spirit. I know it's primarily to explain what we do and how you can comply with various guidances and regulations, but I think we also always want to hear from you and understand ways we can achieve the same ends that might be better. So we are very open to that. That helps keep us strong, of course.

Again, this workshop is focused on the processes and procedures in pheresis. I don't have to go through this for you. I do realize, and have heard from the people in OBRR, that identifying the need and agenda for this meeting and some of the questions -- and I also note that several people from the industry are contributing case studies, which is very important to us. There are probably other people involved, but I know at least these folks are involved: HHS, the ABB, America's Blood Centers, and the staff from CBER, who have taken the effort to work with you and put this together.

I know there is a workshop planning committee. The people are recognized here.

Again, I thank all of you on the outside for working with us. That's the way we can succeed, and not just succeed, but do better.

I think that's all I have to say. I am actually very privileged to be able to stay and listen to one or two talks and educate myself some about what you are doing.

So thank you very much.

MS. FIELDS: Thank you, Dr. Goodman.

Our next speaker today is going to be Dr. Williams. Dr. Williams is going to speak on the regulatory paradigm and the managed-review process.

Dr. Williams is the associate director for regulatory affairs in the Office of Blood Research and Review at CBER. In his position, he also serves as the ombudsman, as well as the principal manager responsible for oversight of quality assurance in the department.

Dr. Williams joined the FDA in 2001 as the director of the Division of Blood Applications, where he managed the processing of all regulatory applications reviewed by OBRR, as well as the scientific review of the blood and plasma establishment license applications and supplements and device applications.

Prior to joining FDA, Dr. Williams enjoyed a highly successful 18-year career with the American Red Cross as a senior research scientist.

Return to Table of Contents

Agenda Item: The Regulatory Paradigm and Managed Review

DR. WILLIAMS: Thank you very much, Lore. Good morning.

In addition to those folks that Jesse mentioned as helping to organize the workshop -- and it was a great effort, putting this together, and there was a lot of cooperation between our steering committee and our own internal staff -- I would also like to thank our administrative support staff, who work behind the scenes to make these things run smoothly virtually all the time. So a big thanks to everyone who helped put this together.

The goal of the workshop really is to educate those of you in the audience -- and thank you all for coming -- about what constitutes a successful license application or license supplement allowing you to make licensed apheresis products. Just in very broad outline, we will talk a little bit about the authorities which underlie FDA's licensure process; a little bit about the content and the expectations that FDA has for submissions, as well as the hallmarks of a successful submission; a little bit about failure investigations, because that is very much an important part of the manufacturing process; and a chance to meet and interact a little bit with your consumer safety officers and with some of the device manufacturing folks. This interaction, particularly with our CSOs, is critical if you have questions. They are always happy to receive your calls.

This is really a very important topic, not only for you, but for the agency as well.

This is a slide particular to platelets, pheresis. It just shows some of the growth in not only the overall use of platelets therapeutically since 1989, but the proportion of platelets which have been derived from single-donor or apheresis procedures. In 1989, the total number of platelets administered was about 7.3 million, and about 2.1 million, or 29 percent of those, were apheresis.

I will mention that the data through the years has derived from the National Blood Collection and Utilization Surveys. These were first done by Wallace and Serjiner [phonetic], going back to the early 1980s, then assumed by Marian Sullivan in the National Blood Data Resource Center, and most recently, under DHHS support, the program is managed by the AABB. So it has provided us a very important consistent tracking of our collection and use patterns for various components.

Moving from 1989 through the years up to 2004, you can see that platelets, pheresis became about half of the nation's blood supply around 1994. Now, with a total platelet administration of better than 13 million doses per year, about 9 million-plus, or close to 69 percent of that, is apheresis platelets. So a real growth in the production of this product.

Although the data aren't quite as concise with respect to the collection of red cells and double red cells, I am aware that in at least one major blood collection community, apheresis red cells account for about 30 percent of the red cell collections. This is anticipated to grow immensely in the coming years, particularly as this becomes associated with additional plasma collections.

This is also critical from our standpoint, because we receive a lot of license supplements for review. This simply reflects the incoming applications in this year, 2007, in our Blood and Plasma Branch which are licensed supplements. Typically, we receive about 800 supplements per year in this branch related to blood and plasma establishment licenses. This year we had 314 prior approval supplements, and 14 percent were related to apheresis procedures; CBE-30s -- and I will explain these terms in a moment, if you are not familiar with them -- 120 submissions, about 44 percent of our overall workload; and overall, about 20 percent of our overall workload is represented in apheresis-related applications.

So the smoother we can make this process, the better it is for those of you who intend to manufacture these components and the better it is for us, because it gives us a better application review, less back-and-forth trying to work out problems, and overall just makes it a more efficient and more successful process.

I am going to say a few words about some of the licensure authorities.

For those of you who may not be familiar with our structure, the Office of Blood Research and Review is one of the offices in CBER. We have three divisions. The Division of Blood Application is the home of our Blood and Plasma Branch. This is the area where these applications related to SOPs for blood and plasma establishments are reviewed.

We also have a Device Review Branch within the Division of Blood Applications. But that actually is almost exclusively focused on blood establishment computer software and in vitro diagnostics for immunohematology reagents. The devices related to apheresis are reviewed by our Division of Hematology.

In the first part of the talk, I am going to discuss some of the authorities for licensing blood products; some of the regulations related to licensing; some characteristics of the license process itself and its significance; a very important area, changes to an approved application, how to modify your SOPs once you are actually licensed for a procedure; and alternative procedures.

I wanted to say just a few words about some of the nomenclature. We have found over time in discussing with some of you that some of the distinctions aren't necessarily clear.

The regulatory authorities: The highest level of authority is what is known as a statute, sometimes called an act. These are laws passed by Congress. They can't be violated by you or, in fact, by us, as FDA. This would be the Food, Drug, and Cosmetic Act, the Public Health Service Act, the user fee acts for medical devices and prescription drugs. Some of the areas like safety, purity, and potency of a biologic or a drug are defined in the statute. So we actually don't have the authority to say, "Yes, we recognize that this glitch happened in the manufacturing process, but that's okay; you can go ahead and use it." If it's something that affects the safety, purity, and potency, we just don't have the authority to make that determination.

Then there are the regulations, also known as rules. These are generally promulgated by an agency as rules that carry the force of law. They tend to be more specific than the statutes. There is some provision for flexibility in the regulations and how one needs to carry out one's procedures in response to a regulation. For instance, within the regs themselves, we have a 640.120 regulation, which allows for one-time exceptions and alternative procedures to existing regs.

There are guidance documents, which are the next lower level. These are current thinking of FDA on a particular subject and may be how FDA interprets the carrying out of a regulation, or it might be recommendations related to other areas reflecting the manufacturing process. You are all familiar with guidance documents. It's the most common way that FDA transmits its current thinking.

Then we have our own standard operating procedures. CBER has operating procedures, as do the individual offices.

All of these materials can be found on the CBER Web site. If you look under "Manufacturers," you will see a very organized listing of all of these. So if you want to know what happens to a BLA or a BLS, license supplement, when it's received, you can go there and find administrative processing of BLA. It's there; it's an SOP. It is a very important resource to be aware of.

Specific to some of the statutes, the Public Health Service Act, specifically 351(a), indicates that a license is required for shipment in interstate commerce. This is a key requirement. You don't need a license to produce and distribute within a state, but as soon as you ship interstate, the federal statutes kick in and require a license.

In addition, licenses are required for shipment into or out of the United States.

The Food, Drug, and Cosmetic Act requires registration of drug and device manufacturers, and prohibits the adulteration and misbranding of drugs and biologics.

Within the regulations, the relevant regulations are found in Title 21, Parts 600 and 601. These two reflect the general licensing requirements. Again, under Title 21-606 and Part 211, generally known as the drug regulations -- the 200s -- these cover good manufacturing practice. 610 covers labeling, labeling being everything associated with a product and the information that accompanies that product -- the product insert, as well as the actual label attached to the product itself. Then there are additional standards covered in 630 and 660.

I am not going to read all of this text. This is some of the text out of the regulation. Basically, what this does is to define the elements that are necessary for an application for a license. These include the format of the license application; the data to show safety, purity, and potency of the product; a description of the manufacturing methods; studies to demonstrate stability of the product. In some cases, a sample of the product is required, and in some cases there is a waiver of that sample submission. It provides for submission of the results of lab testing, labeling, and addresses of all of the manufacturing sites.

With respect to licensure, we look for a description of the establishment, including the facilities, the responsible personnel within the facilities who are key to the manufacturing process. Each facility agrees to open up its site for inspection to ensure that the firm is compliant with all regulations and FDA guidances.

The significance of license is such that it signifies FDA's approval of a product under the existing regulations. The license number must appear on all approved products. This, of course, permits interstate shipment of the product.

One thing of note is that while most biological products carry their own individual licenses, blood products are a little bit different. This is an evolution from an old ELA/PLA system, where there was a separate license procedure for an establishment and for a product. Because many of these establishment licenses went way back for the blood establishments, this was converted into what is known as a multiple-product BLA for each establishment. So while the BLA doesn't reflect an establishment license per se, what it reflects is a multiple-product license. Changes in SOPs or a desire to manufacture a new product are accomplished through supplements to an existing BLA license. So it's a little different paradigm than either the old one or the one that is in place for most other biological products.

As part of the application process, we do require a separate application -- in most cases, a supplement -- for each product, a separate application for each operating location, for each manufacturing method. These need to contain a description of the manufacturing methods, an example of the labeling, the products to be manufactured, and the safety and efficacy data if this is a new product.

I mentioned that changes to an approved application are an important element here. This is provided under the regulations, 21 CFR 601.12. This regulation requires a manufacturer to inform FDA about each change in a product, in the production process, in the quality controls, equipment, facilities, responsible personnel, or labeling that is changed in an approved license application or their facility.

The level at which a change in an application is reported to FDA falls into four different categories. These are classified by not only the extent of the manufacturing change, but the level of risk to the product to compromise safety, purity, or potency. For instance, something with a potentially higher risk to affect product safety, purity, potency we would require to come in as a prior-approval supplement, commonly known as a PAS.

Another category is a change being effective in 30 days, a CBE-30. A third one is a CBE and a fourth is an annual report.

There will be more details on this in the next couple of slides.

A prior-approval supplement is changes which require supplement submission and approval prior to distribution of the product that is made using the proposed change. These are generally major changes. This supplement is submitted for any change in the product, in the production process, in the quality controls, equipment, facilities, or responsible personnel, with a substantial potential to have an adverse effect on the identity, strength, quality, purity, or potency of a product, as that relates to the safety or effectiveness of the product.

For a change under a PAS category, manufacturers must submit a supplement to their approved license application, including the following:

A detailed description of the proposed change.
The products that will be involved with the change.
The manufacturing sites or areas affected.
A description of the methods used and studies performed to evaluate the effect of the change on the product safety or effectiveness.
The data from those studies.
Relevant validation protocols and data.
Appropriate labeling.
The relevant SOPs or a list referencing previously approved SOPs for that manufacturing site.

Included on this slide are some examples which FDA would consider to be major changes, for which prior approval of a supplement must be made before the product is distributed. This would include:

Implementation of a new manufacturing process, including but not necessarily limited to addition or revision of SOPs, if the change is less restrictive than a previously approved SOP or not addressed in published FDA guidance documents. It's less restrictive.
Changing from manufacturing a sole product by automated apheresis to manufacturing additional products as byproducts of the apheresis procedure, or the request to manufacture additional products.
We will have a lot of discussion about comparability protocols related to apheresis products. The initial request for approval of the comparability protocol comes in as a prior-approval supplement.

Equipment changes also require a PAS:

A conversion from manual to automated collection of components, including platelets, plasma, both fresh-frozen and source plasma, red blood cells, and source leukocytes.
Also included are changes or upgrades in automated apheresis equipment that affect the purity, potency, or quality of the products. These include but aren't limited to increase in product yield, change in storage conditions, change in anticoagulants, leukocyte reduction, and collection of an additional or a different product.
Also a change in manufacturing of automated apheresis equipment used in the collection of red cells or platelets.

The next category is CBE-30. CBE-30 is changes being effected within 30 days. At the FDA side, what this means is that we have 30 days to look at the submission and determine if that level of reporting category is appropriate. If we determine that it is not an appropriate category, we will contact you and ask that the category of the submission be changed.

That said, we have six months to review that application for content. The implication of that is, if a manufacturer makes a manufacturing change before the final review is completed, it's kind of at the manufacturer's risk, that they have submitted adequate data and validation to support that change. So that is one important implication of going ahead and manufacturing a product under a CBE-30.

Under a CBE-30, 601.12(c), changes requiring supplement submission at least 30 days prior to distribution of the product made using the change -- this is generally for any change in the product, production process, quality controls, equipment, facilities, or personnel, with a moderate potential to have an adverse effect on the identity, strength, quality, purity, or potency of the product.

I went through the content requirements for a PAS. These are identical for a CBE-30.

Here are some examples of CBE-30 submissions in which we would like to see 30 days before the change occurs. These are generally manufacturing or procedural changes. Examples would be:

Addition of the collection of plasma as a byproduct to an approved plateletpheresis program, provided the applicant is otherwise approved to manufacture the plasma product.
Request for an alternative procedure under 640.120 for which published guidance is available and implementation conforms with the guidance.
Implementation of recommendations described in FDA guidance documents, if followed without modifications and directed to be reported in this manner by the guidance document.

In most cases, an area that a guidance document discusses will have a section in the guidance document referring to what reporting category is appropriate for changes in response to that guidance.

The third category is changes to an approved application, commonly known as a CBE. This is simply a notice to FDA that a manufacturing change is being done. You have the option to implement this change immediately, before hearing back from the FDA. But by the same token, we do have six months to approve this application. Again, it is, to a certain extent, your own responsibility that this change is accompanied by adequate validation and supporting SOPs.

So in certain circumstances, FDA may determine that, based on experience with a particular type of change, the product using the change may be distributed immediately upon receipt of the supplement by FDA. Often these are circumstances which have substantial similarity to a type of change regularly involving a CBE supplement or a situation in which the applicant presents evidence that the proposed change has been validated according to a protocol that has already been approved.

It's important that you retain a record of the date that we received your CBE submission, in case ensuing discussion occurs with the agency as far as the date on which you went ahead and implemented that change.

Here are some examples of a supplement appropriate for a CBE category. These include:

Implementation of another manufacturer's previously approved SOP, with written permission from the manufacturer.
Implementation of recommendations described in final FDA guidance documents, if followed without any modifications and if directed within the guidance document to be reported as a CBE.

The final category is the annual report. This is the lowest risk-related level of supplement. This involves changes in the product, process, quality controls, equipment, facilities, or responsible personnel with a minimal potential to have an adverse effect on the identity, strength, quality, purity, or potency of the product.

It's important that manufacturers document any changes to a product, process, quality controls, equipment, or facilities. The annual report must include a list of all licensed products involved, a full description of the manufacturing and controls changes, which include the manufacturing site or the areas involved, the date each change was made, and a cross-reference to your relevant validation protocols or your previously approved SOP.

Examples here would be, under manufacturing/procedural changes relevant for an annual report:

Revision of an SOP for the categories if the change is more restrictive than previously approved or not described in published guidance documents.
A change in the quality-control method if the procedure is consistent with the manufacturer's directions.

Under equipment changes:

Changes or upgrades by the device manufacturer of automated apheresis equipment that does not affect the purity, potency, or quality of the product, if the facility is already approved for the original procedure.
Finally, use of a sterile connecting device to manipulate a product in a sterile manner, if used according to the SCD manufacturer's instructions and use of the device is consistent with that insert.

All of this is spelled out -- 601.12 is a broader category for biologics, but actually blood components have their own guidance related to this. It's a July 2001 guidance, entitled "Changes to an Approved Application: Biological Products: Human Blood and Blood Components Intended for Transfusion or for Further Manufacture." That, of course, is on the Web.

In all instances, if you have any questions, if there is something you don't understand and you are wondering what level of submission to put in, it's always a good idea to contact your consumer safety officer. Each manufacturing facility will generally interact mostly with a primary CSO. That should be your point of contact. You should make sure that you go up and say hello to your CSO. If you don't know who it is, just ask around a little bit. Each CSO is assigned a manufacturer whom they work with and oversee, and who serve as a primary facilitator of the license supplement that is submitted to us.

A little bit about alternative procedures. This is defined in 640.120 in the regulations. This is a provision that the director of CBER may approve an exception or alternative to any requirement in Subchapter F of Chapter I of Title 21, regarding blood, blood components, or blood products. This is kind of a special consideration for blood components. Some of it is related to preserving adequacy of supply, particularly under stressed or emergency conditions.

Normally, such requests for exceptions or alternative procedures should be submitted in writing. I will note, however, that there is a process for a verbal request and approval of alternative procedures, commonly known as variances, in the regulations. You should follow 601.12 of the guidance in terms of the category requested. Again, this can be accomplished through the consumer safety officer.

There is a list of previously approved alternative procedures on the FDA Web site. Judy Ciaraldi very diligently maintains this site on a regular basis.

One word of caution about this. While, generally, the titles for the approved variances are listed on the Web site, often the conditions that resulted in the approval or disapproval of that alternate-procedure exception is not given in detail. So don't assume that all situations are similar.

But this is provided as a reference, so that you can see the types of variances that have been approved.

One of the most common requests for variances that has been received in the past is, "Oh, my gosh, we left 100 red blood cell units out on the bench overnight, but the air conditioning was on. Is it okay to go ahead and use this product?" That raises a difficult situation for us, not the least of which is that we have very little information to go on in terms of whether the safety, purity, or potency of that product was compromised. We don't actually have the authority to say, "Go ahead and use that product," if that is not really shown to be safe, pure, and potent through rather compelling data.

We do look very hard at these variances, particularly with respect to he status of alternate supplies, but in many cases we just don't have the ability to say that a product is safe, pure, and potent, in the absence of data to support that determination. So that is an important consideration when requesting variances, particularly after the fact.

I close with just a few words about the managed-review process. When you submit a license application or supplement to us, that starts a chain of events, which is going to be described in more detail by Judy and others in the next few talks. We, of course, have a responsibility to you, as one of our major stakeholders, to help you bring products into the medical community that will help improve public health. So we need to make sure that we are under a managed process to try to get that determination made and help bring you to a successfully approved application as soon as possible.

We work under a system known as managed review. I am going to tell you a little bit about the program and how it applies to blood establishments.

You have heard about the user-fee programs. The two programs currently in place under statute are the Prescription Drug User Fee Act and the Medical Device User Fee Modernization Act. Both of these, you may have seen from news reports, are currently undergoing final deliberations in Congress, now delayed until the end of the summer recess. These are major programs that establish timelines for review processes for the products covered by the user-fee programs.

Blood and blood components are not covered by a user-fee program, but we do use managed review for the applications that you submit to us. What we do is voluntarily follow the PDUFA II standards for these products.

Under PDUFA II, the goal is to review and act on about 85 percent of complete blood bank and source plasma BLA submissions or 90 percent of BLA submissions within 12 months after the submission date. In practice, although we receive a lot of submissions, with a limited staff, I think our average time for turning around something like a PAS ranges between six to eight months. So we are well ahead of these goals. We have never even come close to missing one of these voluntary goals.

So that is the general expectation.

I think one take-home message, if nothing else, from this workshop is that your highest likelihood of having an approved application in the shortest amount of time is to come in with a high-quality application. That is very predictive. Particularly from an oversight management standpoint, where there are multiple sites involved, the more consistency and high quality of applications, the faster things will go through the process.

What are some of the elements of managed review? Some of these are unique to our office. Some of them are CBER-wide.

We have instituted milestones for the review process. An important one that we use is called mid-cycle review. If you submit an application for which we have a 12-month or a 90-day timeframe for review, the midpoint of that review period is the time by which we expect that the submission will be substantially reviewed. In other words, each of the discipline reviewers needs to go through that application and identify any key hurdles which would prevent approval of that submission.

In the second half of the cycle, what we do is work interactively with the sponsor to iron out any of the perceived deficiencies in that application and try to arrive, before or at the cycle endpoint, with an approved application.

So the goal is a one-cycle approval, to do the review process, interact with the manufacturer, and not have to go through the process of issuing what is known as a complete review letter, which cites deficiencies and then puts the ball back in your court to respond in writing to us. That is our goal. It's a worthy one for you, as well, to be responsive when we do have interactions following that mid-cycle review process.

Another major milestone we have -- I know from signing many of these applications, it's very hard to review an entire application and make sure all the documents are in order, if you have very little time to do that and a number of applications to sign off on in an afternoon. So we have instituted a metric for T minus 10. In other words, that application needs to be put into final order and start submitting for final review signoff 10 days before the target date for finishing that application. That just saves us from compressing everything down to the last few days. It really does result in a higher-quality process on our end.

We have improved reviewer and supervisor accountability. Each of the reviews is signed off by a branch or laboratory chief, who needs to concur with that review or else iron out any differences between management and the reviewer in a systematic way.

We try to adhere to timely final actions and to a quality-assurance program of our own at all steps of the application, including content review, the meeting of timelines, and the completion of the final archival package.

In closing, I will just say once again that the Division of Blood Applications is your primary point of interaction with the blood establishment part of the application process. The CSO assigned to your firm is your primary contact point for calling in. You can get hold of them through that number. Virtually all of our CSOs are here today. We invite you to introduce yourself and start establishing a working relationship if you don't already have one. I expect many of you do.

We look forward to a successful workshop and hearing from those of you who have had successful experiences with submissions, defining some of the best practices associated with those submissions, and also from our colleagues who actually manufacture the devices, which are critical, of course.

Thank you, and thank you all for coming.

MS. FIELDS: Thank you, Alan. That was a very comprehensive review of the regulations associated with the licensing process, and also the managed-review process, which a lot of people from the blood banks and blood centers are not aware of.

The next speaker today is Ms. Judy Ciaraldi, who will be presenting an overview of the review process that we do at the Division of Blood Applications for the apheresis products.

Ms. Ciaraldi is a consumer safety officer at the Division of Blood Applications, the Blood and Plasma Branch. She actually holds two BS degrees from the University of Maryland, one in biology and one in computer and information sciences. She has worked as a blood bank medical technologist, a blood bank educational coordinator, and an assistant supervisor at the Washington Hospital Center in Washington, D.C., for about 25 years before coming to FDA.

At FDA, she has been very involved with the blood and plasma review duties, as well as with the inspection process. She also is the trainer for the field inspectors that are deployed from the Division of Blood Applications. She has also won numerous CBER and FDA awards.

It's my pleasure to present Ms. Judy Ciaraldi.

Return to Table of Contents

Agenda Item: FDA Review of Apheresis Submissions

MS. CIARALDI: Good morning, and thank you very much for attending.

It's nice to know that some things are still free. We welcome your attendance at this workshop.

I am going to explain what goes into our review of apheresis submissions. I am going to do this by listing the resources that we use for our reviews, introducing you to our review checklist, going over the general content of apheresis submissions, and then drill down to more specific content for specific products. I will tell you how to ship products to CBER, and then I will explain the comparability protocol submissions.

We base our reviews on regulations in the CFR, recommendations in FDA guidance documents, device operator manuals and package inserts for reagents and supplies, and published scientific literature. I am going to go into more detail for the first four items.

These are the regulations that we use for our apheresis review. It's not an all-inclusive list, but they are the ones that pop up most often during our reviews. I am going to spend a little bit of time highlighting a few of the regulations in each of the sections.

Under donor selection:

640.3 contains the general donor suitability criteria, information such as the vital sign requirements and hemoglobin/hematocrit criteria.
640.21 contains additional information for apheresis donors.

Under collection procedure:

640.4 says that collection must be done by a trained staff, that there must be records kept of the collections, there should be positive identification of the donor, and arm preparation should be done to prevent contamination of the final blood component.
640.22 talks about apheresis collections.

Under testing:

610.40 contains requirements for infectious disease testing.
640.5 talks about testing for group and type.

Under product standards and quality control:

600.15 contains the requirements for shipping blood components, the shipping temperatures.
610.53 contains the requirements for storage and storage temperatures of the different blood components, as well as the expiration dates.

Under labeling:

606.121 has the requirements for the contents of the container label.
606.122 has the minimum contents for the circular of information.

Under GMPs:

210.1 says that if a manufacturer fails to comply with good manufacturing practices, FDA considers the product to be adulterated, and the manufacturer could be subject to regulatory action.
211.22 talks about the quality-control unit and their responsibilities.

Lastly, under submission content, 601.12 -- Alan talked about that, so I won't go into any detail.

These are the GMP regulations that we use for our reviews. Again, it's not an all-inclusive list. I am just going to spend a few moments going over some examples in each of them:

211.25 and 211.28 say that the staff and the supervisors must be trained in their responsibilities.
606.20 says that there should be sufficient numbers of staff and that they, through education and training, must have a thorough understanding of their activities and responsibilities.

Under equipment, supplies, and reagents:

211.68 says that automated electronic, mechanical, and computer systems must be calibrated and checked according to a written program.
606.60 says that equipment must be standardized and calibrated on a regular basis to ensure that they perform as they are designed to.
606.65, for supplies and reagents, requires you to follow the package insert.

Under SOPs:

211.100 says that you must have written SOPs and you must follow them. If you need to deviate for any reason from the SOPs, this deviation must be justified and documented.
606.100 says that SOPs must contain all the steps of a manufacturing process. They must be available to staff in the area where the work is performed. You are required to review records before products are distributed. If there is any discrepancy, you are required to thoroughly investigate it and document this investigation.

Under lab controls:

211.160 requires you to establish specifications to determine the acceptability of your product. This procedure or process should include a sampling plan and procedures. If, again, you need to deviate from any of the specifications and procedures you have set up, you need to justify it and document this in writing.
In 606.140, it says that your standards must be scientifically sound and you should have SOPs for monitoring the accuracy of your procedures and your devices.

Under records:

211.186 says that the records must be dated and signed by the preparer of the record and checked and dated and signed by a second person.
211.192 says that the records must be reviewed by the quality-assurance unit, and if there are any discrepancies, they must be investigated.
606.160 says that you must maintain your records concurrently with the performance of your activities, and the records need to identify the date when they were created and the person who created that record. They need to be as detailed as necessary to show a complete history of the work that was done.

Lastly, under adverse events:

606.170 says you must maintain an adverse-event file and you must report fatalities of donors or of patients, where that death is related to a transfusion, to FDA.
606.171, which you are all familiar with, is the biological product deviation report regulation.

Besides regulations, we also use guidance documents to base our reviews on. The guidance documents on this particular slide can be found at this Web site. I am going to just briefly go over the contents of each of these guidance documents.

The first one, which is for automated RBC collection, contains information about the validation procedures, quality-control procedures, the recommendations we have for these; also information about donor suitability and how to send in a submission if you want approval for a license for automated red blood cells.

The next one is for the sterile connecting device. It contains information on when it is appropriate to use this device.

The next one is our intrepid plateletpheresis guidance. It contains information on donor suitability and product quality control.

Lastly, the guideline for quality assurance talks about the scope and content of a quality-assurance program. Again, it's only the minimum scope and content. You would probably have to adjust it for your own needs.

This slide contains our labeling guidance documents.

In the first one, we recognize the ISBT as an acceptable standard for our machine-readable requirement. In this particular one -- it's our second one out recognizing an ISBT standard -- this specific one recognizes the new guide, the 2.0.0 guide for ISBT.

The next one contains information for blood and blood component manufacturers about how the barcode regulation applies to these products.

After that, it's the guidance where we have accepted the AABB circular of information as an acceptable labeling to meet the regulations in 606.122.

The last one here is our guidance for recognizing the Codabar labeling standard.

This slide contains guidance documents to help you submit your application or your supplement to FDA. I am going to start with the bottom one first. This is the one that Dr. Williams went over that describes how to submit an original application for a license, but it also contains general information for the content of a supplement, if you want to change that.

The top one, as he also talked about, contains information for licensed manufacturers who want to in some way amend that license -- change it to do something different from the way they were originally approved.

These are additional manufacturing guidance documents. I want to point out that they are found on a different Web site, this one that contains our old blood memoranda.

The first one is for the licensure of leukoreduced products. It contains information on product standards and how to send in information if you want approval for this.

The next one is for infrequent plasmapheresis donors. It contains a description of what this is and how to submit to FDA.

The next one is for the red cell loss during source plasma collection. Even though this is specifically titled for source plasma, a lot of the content of this has been seen in operator's manuals for collecting other products.

Lastly, the volume-limits guidance document is our guidance document of our nomogram. While the title says "Source Plasma," we have seen that a lot of operator's manuals do include the FDA nomogram. I wanted to point that out.

The last set of guidelines that I will go over, or guidance documents, are the ones that we reference for validation procedures.

The first one is for the validation of software. It includes a small section on on-site user validation. It can be found on the CBER Web site.

The next one is for general principles of process validation. This can be found on the CDER Web site. So that is a little bit different, but it is out there for us to use. It was a joint guidance document prepared by CBER and by CDER, but CDER has posted it on its Web site instead.

We also rely on operator's manuals and package inserts for the information that we don't have in our regulations or have in our guidance documents. Examples of operator's manuals or PIs, package inserts, are the ones that accompany apheresis instruments, sterile connecting devices, platelet and white cell counters, collection bags, and leukoreduction filters. This, of course, is not an all-inclusive list.

We use the operator's manuals and package inserts to get information on donor selection, collection and processing procedures, and product specifications. Just as an example, we will use the package insert that comes with a collection bag to determine what the volume capacity is for that bag, because you are not allowed to exceed that limit. Secondly, we will look at that same package insert to see what the storage temperature and the shelf-life is for product stored in that bag.

So that is just an example.

These are details that are not contained in regulations or in guidance documents, and they are unique and specific for each supply or each apheresis instrument. These are valuable documents to reference.

We used all of the documents and all the regulations that I mentioned and put it all together and created our apheresis review checklists. These checklists are located on the CBER Web site at this link. I am going to introduce you to what is available there. They are actually available in the handouts right behind my talk.

We have one for apheresis red cell review, one for leukoreduction, plateletpheresis review, one for infrequent plasma collection. We have ones for SOP and labeling, talking about the minimum content of these, and ones that talk about the minimum content of a quality-control form.

We also have individual checklists that are based on each device manufacturer's directions or specifications. The ones that are posted on the Web are listed on this slide.

I want to let you know that, while it is not listed on the slide, we are prepared to review submissions for the Haemonetics Cymbal device.

We use the checklists to ensure a complete, correct, and consistent review. That way, everybody is being reviewed against the same standards.

The lists that are on the Web were current at the time that they were posted, which is October 2006. But these that are posted may not represent our current or our future policies. These are living documents, and they are being updated constantly, as we use them in the branch.

Submissions and approvals are site-specific, device-specific, and product-specific. But you can bunch multiple sites and products in one submission, especially if they are all being collected with the same device. After the approval of that initial submission, if you want to implement collection at a different site or you want to use a different instrument or you want to collect a different product, you would need to put a new submission into FDA. But you may be able to streamline this process with the use of a comparability protocol, which I am going to talk about later.

The general content for any apheresis submission is to include:

The FDA form 356h. It is basically a cover letter that helps direct your submission to the proper office.
The cover letter.
SOPs that are appropriate for what you are requesting.
Records and forms, both completed and blank.
Product quality-control logs.
Labeling.
Products (for some submissions, not for all).
You may reference your previously approved SOPs, forms, and labelings, but please include the STN number or the reference number under which those previous documents were approved.

Let's talk about each component of this and go into a little more detail, so you know what to include.

For the cover letter:

It should contain information about the products that you have requested for licensure.
Include the anticoagulants that you are using.
Include whether or not it's going to be leukoreduced or you are going to irradiate it.
Include a list of collection, processing, and testing devices, and identify what they are. That will help tell us if this is a new device or one that we have previously approved that you are just implementing at a new center.
Include the list of blood centers that will be preparing these products. It would be helpful to include the address and registration number, because then we can be very specific about our reviews.
Also include a description of your operator training. You don't need to include the actual training records. We will review those during FDA inspections.

The SOPs that we would like to see to help facilitate your review are included on this slide. These are called our "Big Six" SOPs. I know some of you have heard that term. It's because they cover almost all aspects of the manufacturing process.

Under donor suitability, we would like to see how you determine that a donor is acceptable or unacceptable. It would include your donor-deferral procedures and donation intervals.
Collection procedures include not only how you collect the blood component, but how you prepare the arm and how you monitor the donor for reactions.
Donor history forms, how you facilitate the interview process, including any informed consent that you would do.
The product manufacturing procedures. This is probably one of the biggest sections. It includes quality control, labeling, the splitting of the production, leukoreduction, storage and shipping instructions, equipment calibrations. It could also include use of your sterile connecting device, apheresis machines, and irradiation, if that is something that you are submitting for the first time.
We would also like to see procedures for adverse-event and failure investigation. This would include monitoring and investigation of donor reactions and the quality-assurance oversight program.
Lastly, the quarantine and disposition of unsuitable products.

Under records and forms, we like to see the records that you use and forms for your donor-selection process, including your donor questionnaire, your informed consent, and your high-risk-education materials. Again, if they have been previously submitted to us and they are unchanged at the time of the new submission, you don't need to report them, but reference the submission under which they were previously approved.

We would also like to see records and forms that you are going to be using for processing your product and for doing the collection. Part of the reason we look at blank forms is so that we can make sure you are capturing all the important data.

Next we like to look at your quality-control logs. In some cases these may be blank forms, and in some cases they will be completed. I am going to go into a little more detail on the content of the QC logs.

The completed records and forms that we like to see submitted to us that will help us facilitate your review are two consecutive months of quality control. These are quality control for the apheresis red cells and the plateletpheresis products. These right now are the only ones that we are looking at for quality control.

The quality-control records should represent each device at each center. They should be a sample of all the product types that you are making. For instance, if you are making single and double red blood cells, automated apheresis red blood cells, we like to see a sampling of the singles and a sampling of the doubles. In plateletpheresis, we like to see a sampling of singles, doubles, and triples, depending on what you are asking for in your submission. It should be for each instrument and each facility.

If you also want leukoreduction, those two consecutive months of quality control should include leukoreduced products as well.

Also we want to see, not your whole validation data, but just a summary of the results of your validation. That should include failure investigation that you have done if it was needed.

The type of information we feel is important to be included in your QC logs:

The product description, such as the name of the product, whether or not it is leukoreduced.
The type of collection (a single or a double, for instance).
Collection and testing dates.
Product specifications.
The product testing results, such as the white cell counts, platelet yields, absolute red cell volume, the pH, red cell recovery, product volume, and so forth. Again, that is not an all-inclusive list.
We would also like to see notation of the collection device, identify which one is involved in that quality control.
Product identification number, collection, which is your unit number.
The collection center where this quality control took place.
The identification of the person who performed the quality-control testing.
Evidence of quality-assurance oversight.

Some of the submissions may need to have labeling sent in.

If that's the case, you will also be asked to include a Form 2567. It's basically a cover letter for the labeling submission. It is returned to you with the results of the label review.
A circular of information should be included, unless it has been previously approved.
A base label, with product overlay labels for each product that you are requesting approval for.

Our newest revision to our labeling regulation is one that requires certain information on the label to be machine-readable. This is the unique facility identifier, the lot number, the product code, and the group and type of the donor. These all must be in a machine-readable format.

We have recognized two labeling standards that meet this particular requirement: the ISBT 128 and the Codabar. If you want to implement the ISBT, you must request a variance, a 640.120 alternative procedure to 606.121(e)(1)(ii). The reason for this is that the ISBT labels have the anticoagulant in a different position from what is specified in the regulations. So to use ISBT, you would have to deviate from the regulations and get approval for that.

Also what is unique to apheresis products is that the term "apheresis" is included within the product name or the attribute section of the label. To get specific, in Codabar labeling format, they term "apheresis" as an attribute to the red cell label, and right now only to the red cell label. The word "pheresis" is included in the plateletpheresis label.

Under ISBT, the term "apheresis" is a modifier in front of all product names.

Let's spend some time talking about the specific products and additional information that we would expect to see in a submission for that particular product.

The type of information we would expect to see in SOPs for the apheresis red cells would be information about donation interval between apheresis red cell donations. After donating one unit of red cells, with or without platelets and/or plasma, we would expect to see an eight-week deferral, and after donation of a double-unit red cell collection, we would expect to see a 16-week deferral.

We would also expect to see information about deferral for red cell loss during an incomplete procedure. If there has been less than 200 mL of red cells lost during the procedure, there is no deferral required or recommended. However, if there is a second red cell loss within eight weeks, deferral will probably be needed. If it's less than 100 mL within an eight-week period, the deferral would be eight weeks from the second red cell loss. If the red cell loss is 100 mL or greater, and again it's within the eight weeks, we would expect to see a 16-week deferral from the second red cell loss; if the red cell loss is between 200 and 300 mL, an eight-week deferral; and more than 300 mL, a 16-week deferral.

Our guidance document for apheresis red cells recommends two testing protocols, one for initial validation and one for monthly quality control.

For the validation, or what is in the guidance document identified as a Phase 1 testing, we recommend 100 consecutive units from all devices at the site, to include single and double red cell collection. We mention in the guidance that you can use both units of a double red cell collection to count towards the 100. We have recommended in the guidance document that 95 percent meet the product standards or, again, we recommend that you repeat the validation to assure that you are making product that is consistent with product standards.

For monthly quality control, two months of which are included in your submission for this product, we recommend 50 units at each site, to include single and double red cell collection, in all devices at the site. We recommend in our guidance document that 95 percent meet product standards or repeat the quality control to identify the problem.

Some firms have told us that this may be burdensome to their sites. They have asked if they could send in an alternative approach or procedure. For guidance documents, recommendations in guidance documents, you are certainly allowed to do that. If you want to revise your procedures to include a different testing paradigm, please contact your consumer safety officer. We will review it to determine if it's better than, or at least equivalent to, the one that we have recommended, so that there is no impact on the safety, purity, and potency of the products that you are manufacturing.

What we would expect to see in the SOPs for plateletpheresis is information about the donation intervals. Plateletpheresis can be collected every 48 hours, but no more than twice in a week and no more than 24 per year.

Also, after a 450 whole blood loss during a plateletpheresis procedure, we would expect to see an eight-week deferral, unless the extracorporeal red cell loss is less than 100 mL. The maximum red cell loss per year should not exceed the red cell loss that is allowed for a year's worth of whole blood donations.

For determining if a donor is eligible to be a plateletpheresis donor, you may use a pre-donation platelet count of 150,000 platelets/μL or greater. You may also use post-count from the previous donation or as directed by your device operator's manual.

For your submission for your plateletpheresis, you should include a summary of your validation and any sterility testing that you have performed during this validation. That should be included within the summary for the implementation validation.

The plateletpheresis procedure requires a monthly quality control, two months of which is included in your submission. The quality control is four units per machine type, per product type, at each site. Per product type -- the way that we identify this is, if you are collecting singles, doubles, and triples, we would expect to see four single units, four doubles, and four triples. The platelet counts should be greater than or equal to 3 x 10 11, with 75 percent of the products tested meeting this standard. The pH should be greater than or equal to 6, with all the units meeting this standard. You records should include the product volume.

The quality control can be done at the expiration of the plateletpheresis or at time of issue. We ask that you test one unit of a multiple-bag collection.

Let's go on a little bit more about plateletpheresis, the newest product, the 7-day plateletpheresis.

We have currently approved two systems for 7-day storage of plateletpheresis products, the Gambro and Baxter systems that you see on this slide. Products from both of these systems must be tested with the BacT/ALERT microbial detection system using both the aerobic and the anaerobic bottles. The particular type of product that can be prepared for a 7-day is the leukoreduced plateletpheresis only. Approval is required before you distribute the product.

Licensed firms requesting this product need to request an exemption to 610.53(c) under the provisions of 610.53(d). Unlicensed firms requesting this product need to request an alternative procedure to 610.53(c) under 640.120, the variance regulation that Dr. Williams mentioned. The reason for this is that currently the regulations only allow a five-day expiration date for platelets and plateletpheresis. Therefore, to go beyond that, you are deviating from the regulations and you need approval for that.

Both licensed and unlicensed firms should participate in a post-market surveillance study. Both should submit SOPs for us to review, and both should submit data to FDA one year after the approval.

Licensed firms also have to submit labels and products to CBER.

These are the review criteria if you want either the red cells or the platelets that you prepare by apheresis techniques to be leukoreduced. The product standards are greater than or equal to 85 percent product recovery, the platelet recovery or the red cell recovery. The white cell count for the plateletpheresis and for the red cells has to be less than 5 x 10 6 white cells per container. If the collection container itself meets the standard, less than 5 x 10 6, you don't need to collect any baby bags or the split bags that you prepared. However, if the collection container is greater than 5 x 10 6, you should do an investigation to determine where the process failed. If you still want to give those baby bags, you should count each individual baby bag. If the count of any of them is less than 5 x 10 6, then you can label those bags as leukoreduced for distribution.

If you want to get approval for this, you should submit two months of quality control, but you should continue to perform quality control:

One percent of your collection, or four units if you prepare less than 400 units per month.
Test each product type each month.
Ninety-five percent should meet the product standards or another standard that is specified by the device manufacturer.

One reminder: If you want to submit and prepare products for leukoreduction and apheresis products, you should follow the appropriate guidances. In other words, if you want to prepare leukocyte-reduced apheresis red blood cells, both guidance documents should be used and followed, and become part of your SOPs. If you want to prepare and submit for licensure the leukoreduced plateletpheresis, again the plateletpheresis guidance and the leukoreduction guidance should be included within your manufacturing process.

What we would expect to see in the SOPs for fresh-frozen plasma collected by apheresis is the collection volume. Most often we see the FDA nomogram, partly because that is included in a lot of the operator's manuals for the devices. The maximum amount of plasma collected per year should not exceed the following requirements: If you are less than or equal to 175 pounds, 12 liters, and if you are greater than 175 pounds, 14.4 liters. That is the annual plasma loss.

The collection volume should also not exceed the capacity of the bag.

If you choose to incorporate an infrequent plasma-collection procedure, you must request this as an alternative procedure to 640.3(a), under the provisions of the variance regulation. This is because these donors donate plasma every 28 days or less frequently, but they do it without undergoing a physical exam or having their total protein measured. That is what the exception is for.

The guidance document for this recommends that the donors who undergo infrequent plasma collection have a donor weight of at least 110 pounds.

If you want to collect plasma on a frequent basis, more frequent than every 28 days, you must follow the donor criteria for the source plasma donor, which is included in 640.63 and 640.65.

FDA inspections are another important part of our review process. They are done for centers that are applying for a brand-new license, a license applicant with an approved apheresis program, but they are collecting product at an additional or a new site, and as requested by us go facilitate our review.

They are done to confirm that the center is in compliance with the laws, the FD&C Act and the PHS Act, regulations, and commitments made in the submissions.

Let's go on to a new subject, and that is the comparability protocol.

What is it? First, it's described in our regulations under 601.12(e). It is an extension of the regulation that Dr. Williams described in reporting changes to a license application. It is an option for a submission. By no means is it required. It is an option that you may employ.

Specifically, it is a highly specific, well-defined plan for implementing a specific change in manufacturing. It is specific for that change and it is specific for the specific applicant or manufacturer. A comparability protocol is not appropriate for all changes, as we will learn in a moment. But once you get your comparability protocol approved, you may be able to implement that same change at additional sites, but do it in a reduced reporting category. That will allow you to ship products much sooner than if you submit everything using the original reporting date.

It is used if the product manufactured using that change will meet approved product standards, if the manufacturing process has been validated and all equipment has been qualified, and if appropriate validated assays are available to determine the effect of the change on the product.

The reason that all these things must be in place is that we need assurances that products that are being reported in a lower category meet the same safety, purity, and potency standards as a product that has been reported in a higher category.

The most common use for a comparability protocol for blood and blood components is implementing a single change in multiple locations that are operating under the same license, using the same SOPs, forms, and labels that were approved within that comparability protocol.

When should a comparability protocol not be used?

It's not appropriate for a broad plan covering every conceivable change. Remember, it must be specific.
Also if the change has a potential to adversely affect the product. Remember, anything that could have a major impact or, on a major scale, adversely affect the product needs to come in as a prior-approval supplement.
If pre-specified acceptance criteria are not available to determine the effect on the product, we would not expect to see submissions like these as a comparability protocol.
If the change results in a newly defined product that hasn't been evaluated by FDA, then again, that is something we would not expect to see.
Use of a new manufacturing facility that needs to be inspected.
A change in process, equipment, or a facility that may need to be inspected would not be an opportune submission to come in as a comparability protocol.

Submit your original comparability protocol, or CP, as I am going to call it, as a prior-approval submission. Once it is approved, it must be implemented exactly as it is described in the comparability protocol. That is the way we approved it; that is the way we have agreed that it could come in as a lower reporting category. Subsequent supplements reporting implementation of the change described in the comparability protocol may be reported in a lower reporting category. We will include which category is most appropriate in our approval letter. The most common downgrading is from a prior-approval supplement to a CBE-30. As Dr. Williams described, on a CBE-30 supplement, you can distribute a product within 30 days if you don't hear from us that there are problems with your submission.

The subsequent supplements reporting the implementation should include all the information committed to in the approved comparability protocol.

If we determine that the subsequent submission that has been reported in the lower reporting category has been implemented in a manner that is not consistent with your approved protocol, we will most likely require that all future submissions reporting that change would have to come in as a prior-approval supplement. In other words, if you don't comply with your approved comparability protocol, you will lose the advantage of having it and submitting it, which is to get the lower reporting category.

If you have a new comparability protocol, again those would come in as a prior-approval supplement. If you change an approved comparability protocol, you should discuss this with us, because we don't know what the impact of that change would be on the safety, purity, and potency of the product, and we will probably have to review it and determine if that change will require a different reporting category.

The general contents of a comparability protocol submission:

It should include a description of the change, what you want to implement in the future under the lower reporting category.
It should include your full implementation plan, including a description of your training.
Specific tests and validation protocols, such as used to determine the impact on the product that you are making.
Product acceptance criteria.
The supportive data obtained from the testing and the validation.
Your quality-control testing procedures, including your sampling plan for that testing.
A description of actions that you would take if the acceptable results are not achieved -- in other words, your failure investigation procedure.
Your quality-assurance program oversight of this manufacturing process.

To go on to the next topic, for some products, we do require that they be sent in to us as part of your continuation of your review. The products so far that we are asking to be sent in are the plateletpheresis products.

Ship the products for the following situations:

If you are applying for a brand-new license, we would like to see the platelets to determine if you are making the product according to recommendations, requirements, and according to the product standards described in operator's manuals.
Centers that have an approved license but they are supplementing it to newly include apheresis products.
As requested by us to facilitate our reviews.

The products should be shipped after you have completed your validation and your two months of quality control. They should be sent after you have completed all the testing and the labeling of the product.

Before shipping products to CBER, call the Laboratory of Cellular Hematology in the Division of Hematology, at (301)827-3413, to schedule your shipment. The lab is open from Monday through Friday between 8:30 and 4:00, excluding federal holidays.

Follow your existing SOPs for collecting the product, processing and testing it, for packing and shipping it to anyplace that you might ship any other products intended for transfusion.

The product should arrive at the laboratory during operating hours and before they expire. To ensure this last particular requirement, the product should not expire on Friday or Saturday. It's a good idea to include the reference number or the STN number in the documentation with your platelets so that it can be traced to the proper submission. Also include all the information that we have listed for the content of the quality-control log.

Send products to this address. I won't read it; it will be on your slides.

If you want us to return any of the shipping boxes or coolants that you sent us, please include a prepaid, self-addressed shipping label with your platelets.

In summary, CBER reviews are based on regulations and guidance documents, operator's manuals, and package inserts. Your submission should contain enough information for us to do a substantive review. If it's not complete and not accurate, it could delay the approval of your submission.

Merilyn Wiler was just talking to me about this before I came up for my talk. She said, "How much is enough? How much should we include in the submission?" As a general statement, what I can answer to that is that you have to not make any assumptions about our knowledge. In other words, a lot of us come to the branch with different backgrounds. Don't assume that we have performed these procedures on a regular basis. Some of us have; not all of us have. You should be writing your documentation for an outsider to read and understand. That will tell you how much information should be included.

Consult our checklist and your operator's manuals and package inserts for specific information and standards that should be included in your submission.

Some reviews may require platelets to be sent in to us for testing and some may require facility inspections.

A comparability protocol is a submission option. It's something that you may elect to do if you want. It may allow implementation of the change that is described in the comparability protocol in a reduced category. The CP must be implemented as approved. If you have any changes in your approved CP, please contact us as soon as possible. It's not appropriate for all types of submissions.

The approvals are specific for the apheresis instrument, the products that are collected, and the collection facility.

If you want more information, you can write us at this address. You can also call us at (301)827-3543 or fax us at (301)827-3534. You may also speak to the consumer safety officers in the Blood and Plasma Branch.

I am proud to introduce you to the Blood and Plasma Branch consumer safety officers. I am also proud to consider them my colleagues in this branch. I would like to ask all of them to stand and remain standing until we are all done. That way, it will give everybody a chance to see who you are and where you are sitting.

Karan Blum. I am already standing. Marla Cohen will start next week. She is brand-new. Lore Fields you have already met. Diane Hall is in the back. She is also brand-new to us. Jennifer Jones, another new employee; Rosia Nesbitt; Faye Vigue. Faye is recovering from surgery. I wasn't sure if she was going to make it. Cecilia Watson is manning the fort back at the office. Hoi May Wong; Monica Yu; Ken Zemann, our elder statesman. Our acting branch chief is Dr. Les Holness.

One other important individual is not on this slide, but I want to tell you how important she is. We do rely on her for a lot of guidance. That is Ms. Elizabeth Callaghan, acting division director.

Thank you very much for being part of my branch and working with me. Thank all of you for attending this workshop.

MS. FIELDS: I just want to thank Judy again. That was a very thorough review of our review process in the Blood and Plasma Branch.

We began posting these review checklists on the CBER Web site about a year ago. I think it has really made an impact on how the apheresis submissions are coming in and allowing industry to really understand our thought processes behind how we do the review.

The other key point that Judy brought up -- she explained a lot about the CP process. I think that is something that a lot of people can take advantage of that will streamline the review process.

We are actually running a little early. I can't tell you the last time that that happened. But I did ensure that they do have all of the refreshments out in the break area. Maybe we could return here at 10:00. That will give us 20 minutes.

Thank you.

(Brief recess)

MS. FIELDS: I would like to make a few housekeeping announcements.

[Administrative announcements]

The next speaker today is going to be Rosia Nesbitt. Rosia is going to be outlining some of the issues that we come up with when submissions come in. Basically, the whole group in Blood and Plasma got together and we came up with the top 10 reasons that we find problems with submissions. She is going to go through all of them. I think it will be a good learning lesson for everybody. As you are going through, you can make sure that your submission, right before you send it in, doesn't contain any of what we are calling pitfalls.

Rosia Nesbitt is a consumer safety officer in the Division of Blood Applications in the Blood and Plasma Branch. She has been with us for about two years. Prior to that, she worked as a reference laboratory technician, a regulatory affairs associate, a manager and an error/accident regulatory person at the American Red Cross, and a senior regulatory affairs associate at Blood Systems. We stole her.

Ms. Rosia Nesbitt.

Return to Table of Contents

Agenda Item: Top Ten Pitfalls with Submissions

MS. NESBITT: Good morning.

Judy has reviewed what should be included in a submission. My topic is "Top Ten Pitfalls with Submissions." These pitfalls are what we have observed in prior submissions to CBER. These pitfalls are not in any order or according to rank or weight. All of these pitfalls could potentially be the number-one pitfall.

I will list the top 10 pitfalls first and then I will go back and go into more detail on each one of the pitfalls.

The 10 top pitfalls with submissions:

Donor informed consent and applicable SOPs do not include all possible adverse reactions cautioned by device manufacturers.
Applicable SOPs are not submitted with the application.
SOPs are incomplete.
Failure investigation SOPs are either incomplete or they are missing from the submission.
Quality-control data in the submission is either missing or inadequate.
Quality-control forms are incomplete.
There are general content problems that we see.
Comparability protocol content problems and subsequent submissions.
Label submissions are incomplete sometimes.
Platelet product submission problems.

Donor informed consent and applicable SOPs do not include all possible reactions cautioned by device manufacturers. The following are often not included in the informed consent: anxiety, allergic symptoms, unusual taste or smell.

Donor adverse events unique to apheresis should be included in the informed consent and are often missing. Certain manufacturers require additional recommendations. For example, improper operating conditions may cause complications such as blood loss, hemolysis, air embolism, and blood clotting. If you are using a device from a manufacturer that makes these recommendations in their operator's manuals, your informed consent should reflect these recommendations.

Applicable SOPs are not submitted with the application. SOPs that have an impact on the manufacture of the components being requested for licensure should be included, such as your quality-control SOPs, failure investigation SOPs, donor reaction SOPs, approved SOPs that have undergone major revisions. These SOPs are vital when we are reviewing submissions. When SOPs are missing, this slows down the review process. Please note that previously approved SOPs do not have to be submitted unless they have undergone major revisions. However, the form needs to reference the previously approved submission tracking number if you are submitting previously approved SOPs.

SOPs are incomplete. SOPs, including donor deferral criteria, are often lacking pertinent information, such as deferral information for red blood cell loss. Judy has already talked about red blood cell loss deferrals. If you need to, refer to her slides for the appropriate deferral category.

SOPs do not state what to do with platelets, pheresis, or red blood cell components that have been flagged for QC by the collection device for possible leukocyte reduction problems. The SOP should clearly state what to do with the product. Should the products be relabeled as non-leukocyte-reduced or should they be discarded?

The SOPs for component collection do not contain all of the operation specifications described in the device operator's manual, such as the type of sample, timeframe for performing WBC and platelet counts, what to do if the timeframe for testing is not met. I would caution you here that the firm should be very careful about paraphrasing steps from the device manufacturer operator's manual, because important steps in the process may be left out and the intent of the SOPs may be missed.

Reference in your SOP to follow the operator's manual is acceptable. However, the firm should be aware that if they are relying only on the operator's manual, it may not be as thorough as an SOP. For instance, the operator's manual may not have all information for GMPs, such as documentation.

SOPs describing monthly QC do not always include random or representative selection. The SOP should state that the product for QC should be randomly selected and that each product type should be represented.

Time limits for testing: Your SOP should state the time limit that your product should be tested, whether it's within 24 hours, 48 hours -- whatever the time limit is. Your SOP should reflect that.

Your SOP should reflect the acceptance/rejection criteria -- what is acceptable, and when that product is not acceptable, that it should be rejected.

The disposition of unsuitable units: Your SOP should state what you are going to do with the unsuitable units if they are not within the acceptable criteria.

Failure investigation SOPs often do not include when to initiate a failure investigation or root-cause analysis. Failure investigation SOPs often do not include methods for investigation and correction. How to handle and investigate failures should be clearly stated in the firm's SOPs, so that when a failure does occur, staff will know exactly what to do. Later on today, Hoi-May will cover failure investigations.

Quality-control data in submissions is either missing or inadequate. Monthly QC data for platelets, pheresis does not always include the required collection per machine type, product type, per site. Judy has in her talk this morning discussed what that monthly QC data should include.

Oftentimes, the two consecutive months of QC data are not submitted with the application. Monthly QC data for apheresis red blood cells do not always include 50 units per site with at least one single component included. Oftentimes, we see your monthly QC data and it might have all double products. We need to see at least one single product in that.

The measured total volume is not within the volume limits.

Another example of inadequate QC data is that sometimes the firm requests approval for single, double, and triple platelets. However, the QC data only reflects single and double units. Remember, your QC data should reflect what the firm is requesting approval for.

When the QC data is missing or inadequate, the firm is contacted and requested to submit additional QC data. This in turn slows down the review process.

Monthly QC forms do not include all of the required information. Your firm should state the facility you are requesting approval for. Oftentimes, when there is a facility that has multiple facilities, you see the main facility on the QC form, but they are requesting approval for one of the other facilities. You need to either write that facility on the form or stamp it on. The facility name should be on the form that is requesting the approval.

I know sometimes the blood unit numbers can tell you what the facility is. But if that is the case, you need to have some type of legend for us to know.

Your QC forms should have device manufacturer and type. It should have your blood unit number, date of collection, date of testing. That is exactly when that test was performed, whether it was the WBC count or platelet count or whatever. It should have appropriate collection types, whether it's for a singles, doubles, or triples, or for singles only, doubles only.

On your QC form, you should have interpretation of results, whether your QC passed or failed, was satisfactory or not satisfactory. You should have your yield on there, what the acceptable criteria are. That should be on the form, so that people that are reviewing can just look right on the form and know that it either passed or did not pass. The initials of the person who is performing the test should be on your QC form. You should also have evidence of review by a second person that has reviewed that form, and records of calculation. This does not necessarily have to be on your QC form, but it should be in your submission packet, so we can see that the calculations were done appropriately.

If you are using a summary sheet, copies of the raw data should be also included.

General submission content problems:

Oftentimes we get submissions that do not include the 356h. The 356h should be included with each submission.
The cover letter does not accurately or clearly state what the firm is requesting. For instance, if the firm is requesting approval to manufacture platelets, pheresis, leukocyte-reduced, singles, doubles, and triples, all three should be listed in your cover letter. Oftentimes we see where the firms, in addition to the singles, doubles, triples, want infrequent plasma, but they never ask for that. Your cover letter should clearly state the reporting category of the submission, whether you are submitting your request as a PAS, a CBE-30, whatever your reporting category, or comparability protocol, if you are requesting that.

When your cover letter is incomplete, we call the firm and we ask you to submit another cover letter requesting exactly what your approval is for.

Comparability protocol submissions often lack:

A summary of validation, including performance/acceptance criteria.
A summary of results.
A description of actions taken if acceptable results are not achieved.
Proposed change in reporting category.
A description of training.
Subsequent submissions do not refer to the approved STN number. For instance, if you are requesting a change to an approved CP, you should reference that STN number of the approved CP, so we can easily go back and look at the original CP.

Label submissions:

Often, your label submissions do not include a Form FDA 2567 when you are submitting your labels. Any time you submit labels, you need to make sure that you include your 2567 with the submission.
Labels lack conformity with the Codabar or ISBT format. You can refer to the guidelines, to uniform labeling of blood and blood components, for the Codabar labels and the United States industry consensus standards for the uniform labeling of blood and blood components using ISBT for your ISBT labels. That should give you an idea of the format that you should be using with your labels.

Problems with product being submitted:

Improper shipment of products to the Laboratory of Cellular Hematology.
Because of the improper shipment, we see that the temperature is not being maintained during the shipping process. They receive the product either too warm or too cold. So make sure that you are shipping your products properly.
Products are not received between Monday and Friday. Judy has already reviewed for us how and when to ship products to the Laboratory of Cellular Hematology.
Products expiring over the weekend. So make sure, when you are shipping your products, that they have a good date on them, so that they won't expire over the weekend.

In summary, this is not an all-inclusive list of deficiencies that we see in submissions. In addition to following the CFR, firms should follow the manufacturer's operator's manuals and package inserts. Submitting a complete application may ensure a comprehensive review in a shorter timeframe.

Thank you very much.

MS. FIELDS: Thank you, Rosia. I think there was a lot of helpful information in that presentation.

Our next presenter is Steve Kassapian. Steve is the director of regulatory affairs at the American Red Cross. He has been there since 1986. He has been in his position for approximately the last 10 years. He is responsible for the preparation, submission, and follow-up of all the license applications and supplements for over 290 of the American Red Cross' facilities. In approximately a 12-month period, his group will submit about 100 applications to the Blood and Plasma Branch.

In 1999, the Red Cross did receive their first comparability protocol. He will be lecturing on how that happened and how that has improved his processes.

Steve received his degree in medical technology from the State University of New York at Albany in 1978.

Return to Table of Contents

Agenda Item: Preparing Apheresis License Submission

MR. KASSAPIAN: Thanks, Lore. Good morning, everyone.

On today's agenda, this section has been billed as "Successful Submissions." How do you define a successful submission? I generally use two measurements for success: number one, limited or no follow-up needed once a supplement is submitted to FDA, and then, of course, number two, quick approval.

Dr. Williams mentioned that approval times average, for PAS, between six and eight months; quick approval, two to three months. That's what we want to get.

Successful submissions are all about communication, communication with the FDA and, just as importantly, communication with your own organization.

The fact is, the process of developing, submitting, and, most importantly, getting approval of license supplement requests is easier than most people think. We use a three-step approach. The first two steps are the same. Steps 1 and 2 are, call the FDA. Call them early, when you first become aware of the need for a license supplement. That's why they are there. In the long run, you will be simplifying the process for everyone involved. Why try guessing at what they want to see in your supplement when you can just ask? By calling early in the process, before you actually begin developing the submission package, you are not only giving your reviewer a heads-up, but you are also ensuring that it will contain all of the appropriate information and that it will be in the format preferred by the reviewer.

Obtaining this information early is also beneficial to your organization's project planning.

Calling early doesn't only mean calling the FDA, but it means calling on the decision makers and project managers within your own establishment. Early involvement within your own organization is critical in the development of a successful submission. Far too often, the submission of a license supplement is the last thought in an organization's implementation plan. That approach will almost certainly cause delays in licensure.

The call-early approach will get and keep you and your organization ahead of the curve. Armed with the information provided by your CSO, you can now get involved with your own organization at the earliest stages of product development. That is the best time to develop the licensure strategy.

Let's use the example of triple platelets. If you get yourself involved at the very beginning of the project, you will be able to provide your organization with invaluable information about what types and how much data you will need to collect and evaluate, because this is exactly what you have been talking about with FDA. For example, what kind of data are they looking for -- validation data, QC data? How much data do they want to see? In what format should it be provided?

Experience tells me that the answers to these questions usually come as the result of discussions between the blood establishment and the FDA. (Okay, maybe it's a little heavier on the FDA side, but it's still discussion.)

Every submission is truly a case-by-case situation. There is room for an honest difference of opinion. By calling early, you allow for this discussion to happen before you collect data that turns out to be useless. More importantly, you provide FDA with the time to consider your proposals for what you think needs to be part of the submission. This time factor can only help your case.

Finally, once the answers to these questions are determined, the establishment can move forward with confidence, knowing that they will be evaluating the process and collecting the data in a fashion that will expedite licensure.

I have heard it said more than once how unfair it is that different reviewers sometimes have different expectations for the same type of submission. Sometimes there might be very good reasons for this. But that is really not the point. We can hit any target required, so long as it is well defined and not moving. By calling early, you ensure definition of the target.

An additional byproduct of this process may also be to help standardize the guidance provided by the CSOs, because by calling early, you allow them some time to engage in internal discussions on the subject and to reach consensus.

By the way, in this real-life example, ARC received approval to a triples prior-approval supplement for a facility and approval for a triples comparability protocol in less than two months from submission. That's being successful.

The second step in the process: Call the FDA. Call them often. Believe me, they don't mind. They would rather you call them before the submission than have to have them call you after the submission.

What is "often"? "Often" is as many times as needed to ensure that everyone involved has a clear understanding of what the FDA reviewer expects to see in the submission package. It clearly establishes and then reinforces the requirements for the submission, nailing down the target, making it easier and quicker to hit.

I am not suggesting that you make a nuisance of yourself, but keeping the lines of communication open before the actual submission will pay huge dividends. As I mentioned before, quick approval is all about communication. Effective communication will prevent misunderstandings between the agency and your establishment regarding your supplement. As a result, the FDA should only have very minor comments and questions, if any, about the contents of the submission after their review. Usually these issues can be resolved via phone calls, emails, and/or faxes. After that, the approval letter will be on its way to your office.

Just one recommendation. Decide who in your organization is the key point of contact and only allow that person to manage the communication. You don't want 14 different people calling them. They don't want that either.

Finally, the third step is to listen and do. Do what you have agreed to do, and do it perfectly. Follow the directions and guidance you have talked about with your CSO to the letter. Ensure that the required data is complete and accurate. Explain any failures or unexpected results. Describe all failure investigations that were performed. If the explanations are complete and thorough, fine, no problem; an approval will be on its way. On the other hand, missing data, failed data, lack of clarity as to the request and/or information provided will surely cause delays in approval.

Obviously, it is in the organization's best interest to submit an application that is organized, clear, concise, formatted in a way that is easy to understand, neat, and complete.

That is all there really is to submitting a successful license supplement.

To summarize, appropriate content plus expected format yields fast approvals. Put another way, give them what they want and need, in a way and manner they expect, and you will be a happy camper.

Now that we have discussed the approach, let's talk a little bit about the content and the structure of an apheresis license submission. The purpose of this submission is basically to demonstrate to FDA that you can manufacture a product in a controlled manner that will consistently meet predetermined specifications -- that is, to ensure the safety, quality, identity, potency, and purity of the product. You achieve this through your SOPs, validation, QC, et cetera.

If your package can sell this, then you will get your license approval. So what should the apheresis licensure package contain?

Before I go any further, I need to make two disclaimers. First, I am going to be talking about and giving examples of how ARC puts together these license supplements. They are not FDA recommendations, but they do work for us.

Secondly, I am going to be showing a lot of ARC forms as examples. While the forms themselves are copyrighted and shouldn't be used as is, there is nothing proprietary or confidential about the information contained within them. In fact, that's why I am using them as examples.

Finally, as an aside, I am not the artist behind these PowerPoint slides. They are the handiwork of Illeana Rivera. Thanks, Illeana.

So what is in an apheresis BLA submission? Routinely, it includes a cover letter, a Form 356h, and supporting information, called chemistry and manufacturing controls, such as SOPs, labelings, and data, and a summary review of that data. In certain submissions, it may contain additional information -- for example, in a situation where a blood establishment wants to expand the submission to include a comparability protocol. We will talk about that in a little bit.

The cover letter is the backbone of the submission. It tells the FDA what you are requesting, for whom, and where. It acts as a table of contents. The cover letter includes a list of products for which the license supplement is being requested, the location and registration number of the collection center, the location and registration of the testing facility, and the location and registration of where QC testing is being performed.

It may also include a list or summarization of supporting information, which may be attached as enclosures, to substantiate the blood establishment's claims that they can consistently manufacture an acceptable and controlled product.

The cover letter should also reference applicable SOPs and labels previously approved. It may also reference previously approved related product supplements.

In addition, if a facility is a backup laboratory for component or donor screening, the name and address of that facility should be included.

Here is an example of a cover letter ARC might use in a BLA package.

That is the address of the FDA. That can be found in 21 CFR 600.2.
This is the type of category and what we are asking for.
A list of products.
We have some mention of comparability protocol information.
Where they are being manufactured.
A list of attachments.
Some information about validation.
Some information about following manufacturer's instructions and package inserts.
SOPs and when they were approved.
Labels, again referencing the approvals.

As I mentioned, the cover letter should include a list of all relevant SOPs and labels. Relevant SOPs include such areas as donor screening, collection, manufacturing SOPs -- the "Big Six" that Judy was mentioning before.

Obviously, any product for which licensure is being requested should be included in the list of labels. If a blood establishment has multiple facilities or has a previously approved comparability protocol for the products for which licensure is being requested, chances are good that these SOPs and labels have already been submitted and approved. If so, simply include the STN number and the date of the approval as part of the list.

Otherwise, if the SOPs and/or labels are new or have been revised since last being approved, include them as part of the submission package itself. Remember, if you include labels, you must also include a Form 2567 for each label.

This is the Form 356h. It must also be included as part of the submission package. It is relatively easy and straightforward. Be sure to include the facility name and address, the product description -- you usually reference your cover letter there -- the type of submission, the reporting category, the reason for the submission, and you should also include any relevant cross-references, such as the STN of an approved comparability protocol that lowers your reporting category or any other previously approved BLA submissions that are relevant.

All the boxes don't need to be filled in. We put "NA" in the boxes that don't apply, for completeness.

This is page 2. We check off the chemistry section there. We check off "Other." We reference our cover letter and attachments, and, of course, sign and date it.

Enclosures and attachments are probably the most critical part of your submission package. It is this information and data that provides the proof that you have the manufacturing process under control and that you are meeting product specifications. So it stands to reason that this is where many supplements get hung up at FDA. Since these are the documents that demonstrate that your manufacturing process is under control and that your products are meeting stated specifications, it's vital to ensure that these documents are complete and accurate. Review these documents with a critical eye. Your FDA reviewer certainly will.

One of the easiest ways to demonstrate that your process is in control is the use of QC data. Routinely, we provide two months of data in the form of daily log sheets and monthly summary reports. As you are about to see, ARC is big on using forms to provide this information in a consistent and standardized way.

To license products, you usually need to submit two months of QC data. For platelets, that would be pH, color, yield, and volume; for red cells, volume and hemoglobin. For plasma that is concurrent with plateletpheresis, we submit a verification that the displayed volume is accurate within 10 percent of the actual volume and a statement that the volume collected is within the manufacturer's recommendations.

Leukoreduction QC can be either linked to specific products being used for QC or can simply be a statistical sampling of units. We have found that statistical sampling preferable because the platelet QC and the leukoreduction QC are performed at different times in the manufacturing stream.

When reviewing QC data, you should ensure not only that the individual data points pass each specification, but that the entire data set is sufficient. In other words, are there enough units, or data points, under the submission guidelines to demonstrate a statistical level of control?

You should also review for appropriate signatures and dating by both staff and quality-assurance reviewers.

This is the platelet QC log that we use that is standardized at all facilities. This log is designed so it can be used to document platelet QC information for singles, doubles, and triples, and for any of the devices that we use. It shows information about the calculation. It includes the acceptable results. It shows whether or not they passed or failed. It indicates who did the work, who reviewed the work. It also provides a space for additional comments.

This is an example of the form that ARC uses to summarize that platelet QC form. Like the platelet QC log, this form is standardized for documentation of the QC data, regardless of the instrument or technology used. We have columns for each of the devices. It includes the acceptance criteria, it shows the cumulative results of the QC data, and it has final sign-off by QA.

This is an example of the red blood cell QC log. It, too, has been standardized and "genericized" to be used by facilities with all technologies and devices. For traceability purposes, this form contains information about the collection site and the type of equipment being used to prepare the components. It also has information about the calculation. It shows the acceptable requirements and whether or not the component passed or failed. It also has final sign-off.

This is a summary form ARC uses for red blood cells. The red blood cell QC summary form is used as a cover to summarize the individual RBC log sheets previously shown. Standardization allows staff to document QC summary data for the three devices used at ARC. Like our other forms, it includes information about the acceptance criteria, calculations and results, the percentage that have acceptable results, and whether or not they passed or failed.

Note that, although all three instruments are apheresis RBC collection devices, their acceptance criteria are not all the same. This form allows for the acceptance criteria of all three instruments to be listed.

This is the counting log for leukoreduced plateletpheresis. This is a form we use to record our leukoreduction counts. It has a similar construct to our other QC logs.

This is the summary form that we use for leukoreduction. Like our other forms, the leukoreduction summary form contains acceptance criteria for the Amicus, Trima, and Spectra apheresis instruments.

At this point, if I haven't totally put you to sleep, you should be noticing a recurring theme. Regardless of what QC test is being performed, we attempt to design a QC log sheet for all devices that reflects the QC process flow, and then we design a summary form to demonstrate the additional control of a quality review. The individual QC forms are designed to flow into the summary report, and it is that summary report that demonstrates the actual control of the process. It's these summary reports that serve as the documentation to demonstrate that the process is, in fact, in control.

So why do we document our QC in this manner? First, we want to demonstrate consistency of process. Second, we want to show consistency of products. Finally, we want to simplify the documentation process. By doing this, we show consistency of the manufacturing process, and therefore have a high level of confidence that we can consistently produce quality products that meet or exceed product specifications.

Validation of apheresis equipment is another critical element in ensuring that manufactured products consistently meet specifications. A summary of the validation performed should include:

Pass/fail criteria.
A summary of the performance qualification showing which instruments were validated, for what products, how many procedures were performed, and the percentage of successful procedures. The performance quality procedures should be outlined in an organization's validation plan by type of instrument.
The validation summary should also include information about any discrepancies noted during the validation process, including a description of the failure and the subsequent investigation and corrective action taken.

The entire validation package should be evaluated to ensure that it is complete and that all of the discrepancies have been addressed. The decision to accept a device for use is made based on the requirements set in the validation protocol and the number and types of failures. Obviously, if the validation needs to be repeated, the submission package should not be submitted until corrective action is taken and the device has been successfully validated.

When validation discrepancies do occur, a complete and thorough investigation is required. An example of a discrepancy would be if a platelet product is collected, everything appears to go as expected, but when the product gets to the QC lab, the yield is only 2.6. What happened? An investigation of a discrepancy of this type is definitely in order. This is extremely important, because, depending upon the outcome of the investigation, a decision must be made as to the validity of the data point, and ultimately about the usability of the device in its current state.

If, for example, the low platelet yield was obtained because the donor was uncomfortable and the process was prematurely terminated, this data point can be discarded and replaced with another one. On the other hand, a platelet yield with less-than-acceptable results where no problem with equipment, donor, or staff was identified can't simply be discarded and replaced with another procedure. It counts as a true process failure. That's why we evaluate each case individually and as thoroughly as possible.

The results of the investigation should be concisely presented in a report, along with an explanation for the resolution.

Validation discrepancies with no impact are usually related to donor or staff issues. In these cases, the procedures are usually discarded and the data point not used as part of the PQ data set. Donor issues tend to be associated with donors becoming uncomfortable for some reason, resulting in the collection process being terminated earlier than expected. Staff issues tend to be associated with not following the procedures.

Bottom line: You want to be able to demonstrate that the equipment is operating as expected and that you have a high degree of confidence that it will consistently do so. This, then, is the justification for moving on to statistical sampling for monthly QC.

This is the ARC discrepancy resolution form. It provides a concise and standardized way for staff to document discrepancies that occur during the validation process. Required information includes:

A description of the discrepancy.
Notification of the vendor or BHQ technical support, if required.
A description of the corrective action.
A conclusion of the discrepancy. By conclusion, we mean a determination as to whether or not the discrepancy may have led to an impact in the collection of products.

The validation summary report provides a means for staff to ensure and document that all aspects of the validation plan were executed properly for each device. This is one of our reviewers' favorite forms. This report identifies and summarizes the facility and equipment, the validation protocol followed, the dates of execution, and the total number of discrepancies. It also includes the pass/fail criteria for the execution.

Page 2 of this form is basically for operational and QA management staff approval. The approval section of the validation summary form is used to document approval by operations and quality-assurance staff. After both groups have approved the data and execution of the plan, it then goes to the CEO of the facility for final sign-off.

Some other supporting information:

Sterility testing. This is somewhat passé and may only be necessary at the initial establishment phase or for a first-time user of an apheresis device.
With regard to submission of product samples, once an organization has an approved comparability protocol for a specific technology, the submission of QC product is no longer necessary. It has actually been quite a while since ARC has submitted products for QC as part of a license application.
Finally, a comparability protocol. This may be used by a blood establishment to reduce the reporting category of subsequent submissions of the same type.

A little bit more about the comparability protocol. When 21 CFR 601.12(e) was first put into the regs, you needed a pretty significant supplement to get an approval for a CP. It's a lot easier today (but we won't tell the FDA). Today we usually submit a comparability protocol along with the first prior-approval supplement for a new technology or product. The comparability protocol basically states that all future submissions of this nature will essentially be identical, because the process and procedure will be the same for the entire organization. Nothing will change or be different. All facilities are expected to submit or request data and information in the same manner.

The comparability protocol may also include other information about such things as internal audits and implementation schedules. But the key elements are that the procedures and processes -- that is, the SOPs -- are standardized, the labels are standardized, the training is standardized. Of course, the beauty of an approved CP is that it allows for products to be shipped across state lines 30 days after receipt by FDA of subsequent submissions.

These are the basic elements of an apheresis BLA supplement. Keep in mind that each supplement is unique, and getting FDA input and guidance prior to submitting a package is the key to success.

Thank you.

I would be happy to take any questions now. Are there any questions?

PARTICIPANT: Do you have a standard template for your different organizations when they send information to you -- here's all the stuff that you ought to put in here? Or are they just the summary reports, like you have here, when it relates to a license submission?

MR. KASSAPIAN: A lot of our SOP and QC documents, the QC SOPs, have the standardized log forms. They send us that information. Generally, we put together the summary forms. In some cases, it's put together by the field; in some cases, it's put together by the regulatory affairs staff. But the log forms are standardized. They are part of the QC SOPs.

Other questions?

[No response]

Thank you.

MS. FIELDS: Thanks, Steve. Steve and his group do put together a lot of submissions every year, and they get them through in a very timely manner, based on some of the documents and ideas that he did present today. I think "call the FDA" was the primary lesson we learned today. And I like the one, "Follow what the FDA says." That's always helpful in getting everything done.

Our next speaker this morning is Kathleen Hopping. Kathleen is going to speak to you today about their submissions for 5-day and 7-day platelets. I know that is something that a lot of people have been struggling with in the past. She is going to give the example of how this happened at BSI, which is Blood Systems.

Kathleen is the regulatory manager for Blood Systems and has over eight years of experience in this highly regulated area, and is the liaison and contact with all regulatory agencies. She attained a Bachelor of Science degree in sociology and chemistry from North Dakota State University in 1993. She is responsible for the approval and submission of all regulated documents. Kathleen provides direction to operations and quality management on regulatory issues for 80 collection facilities and two national testing laboratories.

I believe her group has also published a paper on this, several years ago, at the AABB.

Return to Table of Contents

Agenda Item: Blood Systems' Licensure of 5- and 7-Day Platelets

MS. HOPPING: Thank you, Lore. Good morning.

Today I am going to share with you some of our experiences with regard to product licensure. Blood Systems, Incorporated, or BSI, was founded in 1943 as the Salt River Valley Blood Bank in Phoenix, Arizona. We have grown considerably, and in 2006, our nonprofit community blood centers collected more than 1 million blood components and provided these components and other special services to patients in 550 hospitals, in 18 states nationwide. In our 63 years of business, BSI has grown to 76 regional and community blood centers, with two centralized national laboratories.

As the manager of regulatory affairs for Blood Systems, I have the opportunity to work with different regulating agencies, including the FDA. The Regulatory Affairs Department is responsible for submitting requests for changes to our blood license application to the FDA. So far in 2007, Blood Systems has submitted more than 30 blood license application change requests to the FDA. These communications were to report changes to our approved license application, changes in products, labeling, production process, quality control, and equipment or facilities.

Submitting for licensure is just one part of a much larger process. A request for licensure is a request to make a change to the organization's approved blood license application, or BLA. If an organization is going to implement a new cleared apheresis device for collection, it will be important to evaluate, through a change control process, the impact of the change on the organization to ensure a smooth transition.

The plan should include a description of the proposed change. It should describe the process that is to be used for requesting and managing changes. It should facilitate communication about the requested changes among all stakeholders, provide a common process for resolving problems, and reduce the uncertainty around the outcome of the change.

How the organization will implement the change, what equipment will be required, what the costs are to the organization, and what the benefits are all need to be addressed.

It is also important to consider the number of affected locations and how the change will be rolled out to those locations. Different actions may be appropriate if only a few locations will implement the change rather than the implementation affecting the system over time.

When the plan is rolled out, and after all qualifications and validations have been completed, and the process has been established and is in control, it is time to gather the data to submit for licensure submission. This requires two consecutive months of quality-control data and a submission packet.

This slide shows the organizational structure of Blood Systems. The three tan boxes represent the Blood Centers Division of Blood Systems. These blood centers operate under three different license numbers: license number 183 for Blood Systems doing business as United Blood Services; license number 1706 for Tri-Counties Blood Bank, also doing business as United Blood Services; and the third, 1249, for Blood Centers of the Pacific.

The number of collection sites under the different license numbers varies between six facilities and 58 facilities.

Because of our size and the fact that we operate under three different license numbers, BSI has faced many challenges, managing 76 collection sites and the product licensure for each site.

As indicated earlier, Blood Systems has submitted more than 30 change requests to the FDA so far this year. This is just a summary of some of those changes. Most of them will fall in the categories of changes to a product, labeling, or production processes.

Blood Systems identifies, through project planning, when to take advantage of different options for reporting changes to the FDA and the significance of utilizing a comparability protocol to better manage our product licensure submissions.

21 CFR 601.12(e) describes the use of the comparability protocol, or CP, in reporting a change in manufacturing. The CFR describes how a change will be managed -- that is, developed and provided to the FDA -- prior to implementation of the change. A CP is an alternative option for reporting a change and a mechanism to reduce the reporting category -- again, usually from a prior-approval supplement to a CBE-30.

A CP allows for more expedient distribution of product by permitting the applicant to submit a protocol for a change which, if approved, will justify reporting at the time of implementation. It is requested, again, as a prior-approval supplement.

The reporting category will be established by the FDA at the time the comparability protocol is reviewed and approved. FDA is able to evaluate the applicant's ability to effect change that has greater assurance that the change is being properly evaluated. There is less potential for the change to have an adverse effect, and subsequent submissions under the CP will have a lower reporting category. The use of the comparability protocol is an optional choice and is specific to a change.

The use of comparability protocols has been a benefit for our organization because it meant that we can implement and easily report a change across multiple facilities at different times. The individual submission packets are considerably easier to compile. They are uncomplicated, and the time it takes to implement the change after submission is greatly reduced.

Next, I will go ahead and present our example for the Blood Systems Gambro Trima protocol. The slide lists some of the things that we included with the packet. Again, this was specific to Gambro Trima, but the same method could be applied to any apheresis collection device.

Today BSI has 33 collection facilities licensed for products collected using the Trima. That number includes facilities operating under license number 183 and 1249, BSI and Blood Centers of the Pacific, respectively. Tri-Counties Blood Bank does not use the Trima automated collection system.

Initially, when the device was brought up, one center was selected as the pilot center. After the center completed the validation phase and collected two months of quality-control data, BSI submitted the change to the FDA as a CP. The original CP contained a cover letter. The use of a cover letter is strongly recommended and encouraged to introduce and summarize the supplement.

Two forms are included. The 356h is one. This form should be included with all submissions to the FDA regarding a biologics license, including supplements for review and approval, as well as for notifications. This form allows proper identification, routing, and filing of the submission at the FDA.

All FDA forms and instructions are available on the FDA Web site.

Additional information regarding specific information to the CP will be addressed in a follow-up slide.

Relevant SOPs were included and contained all of the processes used to manufacture the final product. The documentation should demonstrate compliance with applicable laws, regulations, and guidance documents. Information that is unchanged from previously approved supplements does not need to be submitted again. But applicable submission tracking numbers should be referenced. Form FDA 2567, the Transmittal of Labels and Circulars, is only indicated if labels are being submitted with the supplement. If labels are necessary, do include the form and an actual label or a proof of the label. The proof is nice, because it does show the font type and the size.

Lastly, we included two consecutive months of QC data.

Specific to BSI Trima CP, the following information was summarized in the cover letter, and the actual information was contained in the supplement as an attachment.

We started our cover letter by clearly and concisely describing the change and stating the purpose of the request. This immediately identifies what it is that we hope to accomplish.

In the section on implementation, we discussed our rollout plan. This included the list of centers that would eventually submit for licensure to manufacture the product specified in the request. It also included the overall plan for implementing the new manufacturing process using this device and the description of the manufacturing changes.

The validation process included equipment, installation, training, instrument validation, product qualification, validation protocol documents and forms, acceptance criteria of the product and actions to take if the results were not achieved, revalidation criteria, and a summary statement.

For the training program, we included task training requirements and an example of the scheduled training that took place.

The quality-assurance and quality-control program documented quality management's involvement and their responsibility for ensuring that all regulatory and quality-assurance issues were addressed in policies and procedures in training. Quality management approved the validation protocol and audited the processes.

The product submission sampling plan identified when we would send products to the Division of Hematology. In the cover letter, BSI also indicated that future submissions would be filed as a CBE-30 once the CP was approved. Follow-up submissions contain a cover letter referencing the STN of the approved CP, FDA form 356h, two consecutive months of QC data for each center seeking licensure, and we also have continued to include summary statements with each follow-up submission.

This is a sample of the summary statement that we will include with each of those subsequent CBE submissions under the CP. The three areas that we address are training, validation, and implementation.

We do continue to use this template, but we do amend it so that it's appropriate for the facility that we are submitting for.

Utilizing a comparability protocol does have some advantages. The initial Trima center that was linked to the CP had to wait for review and written approval from the FDA prior to distributing their products as licensed. The follow-up centers had to wait only 30 days. The 30 days is after receipt of the supplement at the FDA. Therefore, it was at the time and still is important to have a mechanism in place to track that 30-day time period.

Each subsequent facility submitted QC data to the FDA as a CBE-30 and was not required to wait for written FDA approval prior to distributing their products as licensed.

The CP also allows more flexibility for each individual center. Local operations can still choose when to implement a change and wait until they are ready.

The use of a CP for our organization, with our multiple locations, has simplified the reporting process. With the initial CP submission, the center was required to submit platelet products to the FDA, to the Division of Hematology. Our experience has been that additional platelet submissions are not necessary for the subsequent CBE submissions under the CP.

Also the CP -- or, more specifically, the initial change control process which would determine whether or not a CP would be a viable option for you to use -- is an important and useful tool to assist in how to roll out the change across your organization.

There are also some disadvantages when utilizing a comparability protocol. The comparability protocol is useful for only discrete, specific manufacturing changes. The CP would not be appropriate for changes that are too critical, complex, or too large. That said, they do work well for a change to use a cleared apheresis device.

Our approval letters from the FDA for the follow-up CBE-30 Trima submissions uses the language, "The approval of this supplement is based on strict adherence to your currently approved comparability protocol." This means that the approval is in effect only as long as the facility follows the terms outlined in the CP. If there is a change to a manufacturing step, a new CP would need to be submitted as a prior-approval supplement to capture this change.

Technical innovations may also render a CP obsolete. An example would be a software version upgrade. Blood Systems needed to submit a new CP as a follow-up to the original BSI Trima CP after we upgraded from software version 4.0 to software version 5.1. Now, going forward, when we submit a follow-up CBE-30 request, we do list both of those STN numbers in our cover letter.

The next example I will share with you is Blood Systems' experience with 7-day platelets and the variance request to extend the dating period of plateletpheresis from five days to seven days. These requests were submitted as prior-approval supplements. The list on the slide is very similar to the previous example for the comparability protocol. The information specific will be discussed in the next slide.

For the 7-day platelet request, our cover letter to the FDA included additional documentation specific to this request. The cover letter served as an introduction and summarized the requested change. We specifically added verbiage to the request to request an exemption under 21 CFR 610.53(d) to 610.53(c), which is the table of dating periods. We chose this verbiage because we were submitting as a licensed facility that was previously approved for Trima collection.

Our initial submission and all that followed but one applied to collection facilities already approved for five days on the specified instrument, either a Trima or Amicus.

We stated that testing would be with the BacT/ALERT microbial detection systems and would apply only to plateletpheresis, leukocyte-reduced. In the cover letter, we listed all submission tracking numbers of any relevant submissions where approvals had been granted for 5-day platelets specific to the locations and the instruments listed in the new request.

The final additional item was to state that we would agree to participate in a post-marketing surveillance study.

This language carried over to our approval letters. Our approval requires that Blood Systems participate in a post-marketing surveillance protocol and supply the FDA with two consecutive months of QC data from the affected locations via a post-marketing commitment annual report. This annual report is different and separate from our regular annual report, which is submitted every year within 60 days of our anniversary date.

Lastly, we were required to submit platelets, pheresis, leukocyte-reduced to the Division of Hematology for QC testing. We will be required to do this for each license number that has submitted now for 7-day platelets for each instrument that they request.

To date, Blood Systems' Regulatory Affairs has submitted six separate change requests to the FDA for 7-day platelets. The table summarizes some of the highlights.

Again, remember that we are operating under three different license numbers. That is part of the reason for the multiple submissions. Our facilities are currently using both Trima and Amicus.

One thing I would like to point out is the timeframe from request to approval. The first submission was received by the FDA in early September 2005, and written approval was granted December 28, 2005. The total time in review was less than 120 days, or approximately four months. Four of the five subsequent submissions were all in review less than 50 days, and one took only 27 days. All submissions were very similar, with modifications made to specific locations, to the license numbers and their instrumentation. All but one of these submissions pertained to facilities previously licensed for 5-day platelets.

There do exist different regulatory pathways for 7-day platelets. They are for firms not previously approved or for firms previously approved. The submissions do vary. The examples that I discussed earlier were specific to previously approved locations.

The last submission listed on the table is still pending in-house. We have not yet submitted it to the FDA. As just a note that I wanted to add, the submission is ready to go, but we will hold off on submitting it to the FDA because we have an Amicus 5-day licensure request that is pending, and we have been told that as long as we have a pending application or supplement that is outstanding, we will hold off on submitting for 7-day until that time.

The purpose of the previous table was to demonstrate submissions that were reviewed and approved in a relatively short period of time. This can be achieved by keeping an open communication with FDA staff. It is important to establish and maintain a positive working environment and relationship with your consumer safety officer and other FDA staff. If you have a question or you are not sure of which reporting category to report something, just ask. If a deficiency is received and it is not understood, ask for clarification to understand what the expectation is.

It is also important to follow up on all deficiencies and/or questions in a timely manner. This will help to keep the supplement moving through its review.

The standard operating procedures we submitted with our initial 7-day platelet request went through a revision change before the final approval. Yet it still was a very quick turnaround time of four months. To accomplish this, Blood Systems' Regulatory Affairs, the FDA, and the BSI subject-matter experts worked together to address the issues and concerns -- again, making sure that everybody is kept in the loop and is aware of what is going on.

Product licensure can be a cumbersome and sometimes confusing process, even when it is a task that is frequently performed. It is important to make sure that all submissions are clear, complete, accurate, and neatly organized. If you are struggling, make sure you utilize your resources at the FDA. It will make the submission easier to compile and also easier to review.

Thank you.

MS. FIELDS: Kathleen has agreed to take questions if anyone has any for her.

Wow. Two big directors of regulatory affairs, and no questions for them.

We are running very early. We have notified the cafeteria that we will be letting out at 11:30 for lunch. Please be back by 12:45. We have a very extensive agenda for this afternoon.

We are also going to add a surprise presentation this afternoon. Things that don't happen very often at the FDA apparently happen on workshop day. Elizabeth Callaghan will be presenting some stuff for you this afternoon.

Thank you.

(Thereupon, at 11:25 a.m., the meeting was adjourned, to reconvene at 12:45 p.m., the same day.)

AFTERNOON SESSION

DR. WILLIAMS: Welcome back, everyone.

Just a reminder from Lore. There were evaluations passed out. If you need early, please make sure you do that and drop it off somewhere in the front or to Lore directly. That would be really helpful.

The next presentation is entitled "Collection of Platelets by Automated Methods Guidance Documents." It's going to be presented by our own Lore Fields.

Lore is a consumer safety officer in the Division of Blood Applications. She earned her BS degree in biology from Westfield State College and her MT (ASCP) certification as a specialist in blood banking from the Transfusion Medicine Academic Center at the University of South Florida.

She has, overall, 15 years of experience in the field, first as a donor test lab manager and a blood bank manager. She joined us here at FDA in 2005, where she reviews apheresis comparability protocols and does pre-license inspections for source plasma facilities, and also is very active in guidance development -- and, as we see, also is a superb workshop coordinator.

Lore has the next talk. Again, it's "Collection of Platelets by Automated Methods Guidance Documents."

Return to Table of Contents

Agenda Item: FDA Guidance on Collection of Platelets by Automated Methods

MS. FIELDS: Good afternoon.

Today we are going to talk about the collection of platelets by automated methods guidance documents. First, I want to give a little bit of background on how we define guidance and how it works within the center.

FDA guidance provides blood establishments and FDA staff with FDA's current thinking on acceptable ways to comply with regulatory requirements. Guidance is intended to help ensure donor safety and the safety, purity, and potency of the products.

We consider the recommendations in the guidance documents to provide appropriate criteria for a biologics license application or supplement for manufacturing and to provide guidance on preparing a manufacturing supplement under Title 21 of the CFR, Section 601.12.

FDA's guidance documents, including guidance discussed today, do not establish legally enforceable responsibilities. Instead, guidances describe the FDA's current thinking on a topic and should be viewed only as recommendations, unless specific regulatory or statutory requirements are cited. Use of the word "should" in FDA guidances means that something is suggested or recommended, but not required.

There are three things that we are going to talk about today. I want you to be very careful as you are going through the slides. We have the guidance from 1988. (Please don't throw any tomatoes at me.) We have the draft guidance from September 2005. We are going to very briefly ‑‑ because there is not a lot that I can say about it -- tell you some of the topics that we are thinking about changing or thinking about the issues, based on the numerous comments that we received on this document. [Laughter]

We will talk about things that are under current consideration. As you know, I can't give you details of what those are, but, hopefully, just the fact that we are considering them will help ease some fears.

The first thing that is a requirement is the donor weight. The one thing I do want you to remember about the donor weight here is that, although we recommend it, a lot of operator's manuals for the apheresis equipment actually have a weight that the donor has to be, or several weights, depending on the algorithm. If you look at the CFR 606.60(a), it requires that you follow the operator's manuals, package inserts, et cetera. So you want to make sure that even though this is a recommendation, you are following the instrument's requirement for the weight.

We did state in the 2005 guidance that one of the issues under consideration at FDA is a minimum donor weight of 110. We think this will standardize some of the things that are going on in industry.

The second issue in donor selection is that plateletpheresis donors must meet all donor suitability requirements described in 21 CFR 640.3. For example, you must have the medical history to determine that a donor is well and healthy, the normal temperature, a demonstration that you have a BP within normal limits, and a blood hemoglobin level which shall be demonstrated to be no less than 12.5 grams of hemoglobin per 100 mL of blood, or a hematocrit value of 38, depending on what you use at your facility.

Additionally, we always recommend that all AIDS-related deferral procedures are followed.

One of the other hot issues right now is that we still currently recommend that persons who have recently ingested medicines containing aspirin, especially within 36 hours, should not donate platelets. There has been a lot of talk about this, and this is what I am prepared to say. [Laughter]

At the March 2006 BPAC meeting, and on the FDA docket, public comments were made on the medication deferrals in the draft guidance. The issues and data are currently under consideration at FDA. We are taking all of those comments very seriously.

The next recommendation is that the platelet count should be greater than 150,000. There are several ways to qualify the donor for this recommendation. Some people use pre-counts, platelet counts obtained from the end of the previous procedure. Some manufacturers do have algorithms that are acceptable to use in their operator's manuals and associated publications.

Additionally, donors with platelet counts of less than 150,000 should be deferred until the count has been returned to normal. This means that if you do get a count that you see is less than 150,000, you can't say, at the next donation, "Oh, we think, based on previous experience, that this donor's count is going to go back up to 150,000." Sometimes we see issues with that in submissions. You really need to ensure that that count is above 150,000.

For donation intervals, FDA currently recommends that a donor should not undergo more than two procedures in a seven-day period, with a donation interval of 48 hours between procedures. Additionally, a donor should not generally undergo a total of more than 28 plateletpheresis procedures during a calendar year.

The donation frequency and donation intervals from the draft guidance are issues currently under consideration at FDA. Based on comments that we received, we are considering the following issues: the collections per year; donation interval for doubles and triples; and minimum post-donation platelet count.

We did ask Dr. Katz to come in today. He has a recent paper out on some of the studies that he has done at his blood center. He is going to speak this afternoon on that data.

FDA also recommends that any person who has donated a unit of whole blood or who has lost the equivalent amount of whole blood during a plateletpheresis procedure should not serve as a pheresis donor for eight weeks if the extracorporeal volume is greater than 100. You need to determine this based on your instrument. Your instrument manufacturer should be able to provide that information for you.

FDA also recommends that the red blood cell loss recommendations include:

Maximum total red blood cell loss for one calendar year should not exceed the loss of red blood cells permitted by FDA regulations for whole blood collections. That is the 640.3. You basically have to do a calculation based on what we require to come up with what your allowable red cell loss is for the year. We would be more than happy to help you with that, if you have any questions.
Red blood cell loss for pheresis procedures that exceed the frequency limit stated for whole blood should be monitored. This is keeping track of the amount of blood that you are losing in each donation for the year.
Complete and accurate records of this information should be maintained.

For plasma, FDA recommends that the total volume, excluding anticoagulant, of all products should not exceed 500 mL or 600 mL for donors weighing more than 175 pounds. Collection of plasma byproducts resulting in a larger total volume will be considered upon receipt of request.

We are also working on this issue. I think people saw that in the 2005 guidance. We are dealing with some issues under consideration to help alleviate this.

Additionally, FDA recommends for donor selection to include:

The accumulated laboratory data for donors should be monitored by qualified personnel and reviewed every four months by a licensed physician.
Donors should be questioned about adverse reactions.
Pheresis instruments employed should be approved, or cleared, by CBER, depending on which category they are in.

Informed consent is an important process for any medical procedure. FDA currently recommends that the following be included in your consent process and procedures:

Informed consent should be obtained from new donors.
On subsequent donations, a simple consent is adequate, if you want to have two separate consent forms.
Appropriate consent should be obtained by a physician or an allied health person experienced in the procedure to be performed. Whoever is doing the informed consent -- it's usually someone from your donor room -- should totally understand the procedure and what is going to happen to the donor and how they could possibly react.

The following information should also be provided to each donor in the informed consent:

A description of the procedure.
A description of the possible adverse effects and side effects of the procedure, and solutions and/or drugs the donor will receive.

This should include here anything that is in the operator's manual as an adverse effect. A lot of times the instruments will have their own little section that will say, "In addition to whatever the FDA says in their recommendations that you should have in your procedure, we also think that these may be risks if you use our instrument." We see a lot of times that people forget to include those in their consent. We understand that consent forms are painful to get approved through your blood center, because they usually have to go through a legal process or whatever. So when you are writing the consent form, you definitely want to remember that not only do you want to follow our recommendations, but also the manufacturer's, and the solutions and drugs. You want to make sure that any drugs or solutions you are giving -- anticoagulant, et cetera -- that you have listed all of their possible side effects.

The person should have a clear opportunity to refuse consent.
A statement that participation is voluntary and that the donor may withdraw their consent at any time.
A statement informing the donor of his or her right to ask questions of and discuss the procedure with a physician. You want to make sure that's clear and they don't feel like they are rushed, if they really have questions about it. Apheresis can be invasive, and they really should have that opportunity.
A statement that the long-term effects of the reduction of lymphocytes is not clear. This is still in the 1988 guidance.
A statement that the donor has received and understands the information provided to him or her regarding the spread of AIDS virus by donated blood and plasma.

Your collection protocols should include: The equipment shall be observed, standardized, and calibrated on a regularly scheduled basis. This is 21 CFR 606.60. This is a huge statement, just in one line here. You really need to make sure that you follow the operator's manual and other applicable standards when you are writing out your information here to standardize or calibrate or QC your instrument.

We also recommend for your collection protocol that you include:

A qualified physician who is familiar with the procedure should be available to attend the donor within 15 minutes.
Personnel engaged in plateletpheresis should obtain specialized training for the use of the instruments. There should be documentation of that.
During the procedure, you should visually inspect the plasma for hemolysis.
There should be a written procedure for management of cardiopulmonary emergencies that contains steps for contacting physician, obtaining an emergency rescue squad, and transport of the donor to the hospital.

For processing and testing the plateletpheresis, we recommend that the procedures include:

The platelets, pheresis should be tested as a whole blood donation. They should have the same infectious disease testing and those sorts of things.
A hematocrit should be performed on final products containing visibly apparent red blood cells to determine the total packed red cell volume. If the final product contains more than 2 mL of the packed red cells, a sample of the donor blood should be attached to the container for compatibility testing.

If your blood center doesn't want to do this, you should have instructions in your procedure on the disposal of those products, so that when we are reviewing these submissions, we know that you have identified this and you have done something about it.

The dating period for platelets, pheresis collected with an approved instrument is 24 hours from the termination of the procedure, unless the system used for collection has been specifically approved for longer storage. As you know, most plateletpheresis bags have been approved for five, and even seven, days. You should review all of your package inserts for the storage approval of your bag.

I think this one is self-explanatory. Platelets, pheresis should be stored at 20 to 24 degrees C, with gentle agitation.

During the establishment phase or validation of the procedure, the following determinations should be made for each unit: There should be a hematocrit, a platelet count, white cell count, pH, total product volume, and a sterility test at outdate, if applicable. Additionally, you should add to your protocol any validation requirements provided by the manufacturer of the instrument.

Once the procedure is operating satisfactorily, for QC you must test four units per month at issue for actual plasma volume, pH, and platelet count. You can find that in CFR 650.25. Currently, all units must have a pH of not less than 6.0.

There are a lot of questions on how to do this. We will be answering those later this afternoon during the question-and-answer session.

Additionally, we recommend for QC:

At least 75 percent of the units tested should contain a minimum of 3.0 x 10 11 platelets.
A white cell count should be performed, if applicable.
The product volume should be determined after sampling.
The standard operating procedure should contain the acceptable limits for your instruments.

The draft guidance addresses the importance of statistically significant sampling plans for quality control. We received many comments from industry and are considering several issues for the final guidance.

FDA recommends also, for processing and testing, that the actual platelet yields be determined on every unit. This data should be part of the issue records. This doesn't mean it has to be on the unit as such; it can just be in your records.

For labeling, we have an extensive list of requirements that have to be done for plateletpheresis:

The first is a statement that this product may transmit infectious disease agents.
The name, address, registration number, and, if a licensed product, the license number of each manufacturer.
The donor unit number.
The expiration date, including the day, month, and year, and, if the dating period for the product is 72 hours or less, the hour of expiration.
The donor's ABO and Rh blood groups.
A statement indicating the kind and volume of anticoagulant solution present.
The proper name of the product in a prominent position, and modifiers, if applicable.
The appropriate donor classification statement -- for example, "paid donor" or "volunteer donor."
The volume of the product, accurate to within plus or minus 10 percent, or a volume range within reasonable limits. (This is pretty hard to do with apheresis.)
The recommended storage temperature, in degrees Celsius.
There should be an "Rx only" for that, so there is a prescription.
"See circular of information for indications, contraindications, cautions, and methods of infusion."
Properly identify intended recipients.

Additionally, we recommend that the instruction circular, or the circular of information, is considered labeling and should include the following specific information, in addition to the general requirements for all blood products:

Instructions to begin administration as soon as possible, but should not exceed four hours after entering the container.
A statement that he actual platelet content is available from the manufacturer upon request.
Instructions to maintain continuous gentle agitation during storage.

The instruction circular must also include the following specific information, in addition to the general requirements for all blood products: There should definitely be an instruction to use a filter. You can find that requirement here in 606.122(b).

I just want everyone to keep in mind that this is kind of a summary of the big topics of the recommendations in the guidance documents and should not be used as all-inclusive. There are several recommendations from the 1988 guidance that I did not include, because it's not a hot topic or it's not something that people use every day. If you have anything out of the ordinary, you should really check the guidance document when you are writing your SOPs.

If you have any questions about any of the guidance you have heard today, feel free to contact me or your consumer safety officer in the Division of Blood Applications. I have put our phone numbers. That is my personal phone number, and then the phone number for the main branch if you would like to talk to any of the other CSOs.

Thank you.

Next we are going to have our surprise presentation, which isn't much of a surprise now that all the handouts have been put out. But we really felt that this was a significant topic and very timely, and it has some implications for the blood centers. Elizabeth Callaghan, who is the acting director of the Division of Blood Applications, will go through a short summary for you.

Return to Table of Contents

Agenda Item: Revisions to the Requirements Applicable to Blood, Blood Components, and Source Plasma

MS. CALLAGHAN: Sorry to barge in like this. If you are like me, you are waiting to hear Dr. Katz's presentation, which will be much better than mine, I'm sure.

Up until about 10:00 this morning, we weren't sure that this presentation was going forward, because we weren't sure if the rule was going to get displayed. So we didn't put it on the agenda, just in case it didn't. But, fortunately, it is displayed. We now have a new regulation that changes some of the plateletpheresis CFR sites. So I figured it would be nice if I went over them with you, so you don't get too surprised after hearing the old ones today.

The regulation is called "Revisions to the Requirements Application to Blood, Blood Components, and Source Plasma." The regulation is going out as a direct final rule and a companion proposed rule. We put the regulations out this way because we do not feel that the changes that are in the CFR will be controversial. Actually, most of them, we feel, are more applicable to the way industry is conducting business now. We didn't want anything to be too hard on anybody to do.

If we receive no significant comments to this rule, it will go into effect February 16, 2008.

The first regulation for platelets that is changed is 610.53, the dating period. Presently, this section reads "72 hours from the time of collection of the source blood, providing labeling recommends storage at 20 to 24 degrees C, between 1 and 6 degrees C." What we have added is "or as specified in the directions for the blood collecting, processing, and storage system approved for such use by the director." This is allowing 5- and 7-day platelets to not be against the law. I'm sure that makes you all happy.

640.20 Source: "The source material for platelets is plasma, which may be obtained by whole blood collection or by plateletpheresis." What we have taken out of that part of the reg is "by plasmapheresis." I really do not think that any of you are collecting two units of platelet concentrates by manual plasmapheresis. I would not think that would be monetarily acceptable at this point in time. So we have taken out "by plasmapheresis." We have made consistent the "by plasmapheresis" changes in the following two regs.

640.21 Suitability of donors: We have removed Section (b), which is "by plasmapheresis," and taken the regulations under plasmapheresis and moved them into plateletpheresis. It now reads, "Plateletpheresis donors must meet the criteria for suitability as prescribed in 640.3 and 640.63(c)(6), or as described in an approved biologics license application or BLA supplement. Informed consent must be obtained as prescribed in 640.61."

What this does is codifies the requirement for 110-pound minimum weight for plateletpheresis donors. The way it is written now, it ensures that registered facilities are held to the same standards as licensed facilities.

640.22 Collection of Source Material: Again, we have removed the plasmapheresis section and moved the requirements into the plateletpheresis section. It says, "If plateletpheresis is used, the procedure for collection must be as prescribed as in 640.62, 640.64" -- except paragraph (c), which is anticoagulants, because the 640.60s are source plasma regs and use different anticoagulants for transfusion than they use for the collection of source plasma -- "and 640.65, or as described in an approved biologics license application or biologics license supplement."

The way this is written, it again ensures that registered facilities are held to the same standards as the licensed facilities.

640.25 is amended by changing the requirement for the pH at the end of the storage period to be not less than 6.2, instead of the 6.0 which is now in the regulations.

So for your late-summer reading enjoyment, this will be published tomorrow. You can get it at this Web site. It does contain other changes applicable to other parts of the blood regs and plasma regs. But I figured the platelet regs were what you wanted to hear about. So have fun.

MS. FIELDS: Thank you, Elizabeth.

The next speaker will be Dr. Katz, who will be presenting his recent paper. His presentation is entitled "Impact of Very Frequent Plateletpheresis on Donor Platelet Counts."

Dr. Katz trained in infectious disease and hospital infection control at the University of Iowa Hospitals and Clinics in Iowa City, Iowa, and served as medical director of the Quad Cities Regional Virology Center in Davenport, Iowa. Currently, he is executive vice president, medical affairs, at Mississippi Valley Regional Blood Center in Davenport, Iowa. Dr. Katz is a member and past chair of the Transfusion-Transmitted Diseases Committee and a member of the Clinical Transfusion Medicine Committee of AABB. He has also served as a member and chair of the ABC Scientific, Medical and Technical Committee. Dr. Katz currently serves as the industry representative on the FDA Blood Products Advisory Committee.

Return to Table of Contents

Agenda Item: Impact of Very Frequent Plateletpheresis on Donor Platelet Counts

DR. KATZ: Thanks, Lore.

It's really odd when you get off the plane in Washington, D.C., in August and comment on how cool it is. That's where I live, out in Iowa. It keeps the Californians out, which is, after all, the object.

Lore, thanks.

I am going to take the invitation to show some of this data as a sign that, in fact, the FDA is listening very carefully to comments from the blood community on the draft guidance.

How many people here took nitroglycerine when they read this guidance? [Laughter]

I certainly did.

All kidding aside, there were proposals in the guidance that we felt, as soon as we read it, would have a deleterious impact on our ability to supply platelets to people in hospitals, both those who need them and those that the heart surgeons take care of. [Laughter]

So I was having angina and talking to Sharyn Orton, who was still at the FDA at that point, at the AABB meeting. She mumbled something about, "Show me some data. Quit whining." So that's what this little study that we did at our center was an attempt to do.

To protect the safety of the donor -- and this was specifically stated in the draft guidance -- FDA was suggesting that some restrictions -- not so much restrictions on the number of collections; 24 a year has been the maximum for some period of time, certainly as long as I have been involved in blood banking, in an operational sense. But the real change would be that the number of components would go down. So rather than 24 collections, really, we were talking about a maximum of 24 products a year.

In Iowa, where there are substantially more pigs than people, that would mean that we would have to start drawing pigs, in addition to people. So we thought that we needed to look very carefully at the impact of this.

In addition, there were some changes in interval with regard to the number of platelets produced at an individual setting. Intervals get longer if you do a double rather than a single, or a triple rather than a double. We thought that might have some impact and weren't sure, because of data that we have been collecting for a long time, that there was a need to protect the safety of the donor in that manner.

Post-donation platelet count -- I think Lore referred to some issues about non-steroidal anti-inflammatory drugs that I am not going to address in what I shows you here.

So this was the genesis in the guidance, this study from NIH, Lazarus et al., published in Transfusion, representing data from 1994 to 1998, which suggested -- and I think credibly suggested -- something that most of us had not observed in our apheresis programs. That is, with increasing amounts of apheresis over some period of time, donor platelet counts fell. We just had never observed this phenomenon. So 21-30, over a period of four years, associated with a 30,000 decline in baseline platelet count. I believe these were pre-counts.

The conclusion of the paper was that clinically significant thrombocytopenia is unusual. But FDA, appropriately, looked at this data and asked the question in the draft guidance of whether some caution was appropriate.

This describes our system. The important thing is, we are doing 13,000, 14,000 apheresis platelets this year; six different sites spread over an area of about 250 by 300 miles; five hours from the main center to the furthest site. That had to do with the requirement for a physician to be available within 15 minutes.

I have been told by my insurance agent not to get any more speeding tickets, so that was going to be a problem for us.

We use Trima, primarily, but also Amicus. We have a very interesting group of donors that we call our 24-karat donors, who donate 24 times a year, every year, some for as long as 10 years and more. There are about 100 of them. They are a little crazy, I think, but that's a different story.

So this is the distribution in 2004, frequency. Here are the once-in-a-year platelet donors; here are our 24-karat donors. I am sure that most places have a distribution something like this. Again, our population density is a little bit small, and so maybe we push on our donor base a little harder than some places, but I don't think out of whack.

So that would have reduced our collections by 12.5 percent, based on the mean, 1.47 products per procedure, that we had across the entire donor base. In fact, these donors produce more like 1.7, 1.8 products per procedure.

We used the data from our fixed site in Davenport. These are paper records. We didn't want to pull our paper records back from the other sites, because then we couldn't draw these donors without their prior records being available. So we just restricted to 60 out of the 100 or so, but they are representative.

They gave, in 2005, 2,439 platelets -- that is, 20 percent of our total collections were from 24-karat donors at our main site -- 40.7 products, and we would have lost 16.7 components with a 24-component limit. That is 41 percent of the platelets at the Davenport fixed site. So that is the maximum estimate.

Across the system, we did just a little bit of modeling to compare. At the average number of products in the <16 and the >16, we would have lost 2,000 products from everybody at 16 and above. The calculation is in your thing. You can read it and argue with me if you want. Assuming no increased frequency amongst the current donors, we would have had to replace 33 percent of the current donors to make up the difference.

We look at that -- obviously, duh -- as a major challenge, on a date when a guidance becomes effective, to need to increase our donor base by 33 percent.

This is the American Red Cross. It gives you kind of a similar feel. Anne Eder and Ed Notari sent me this. This is six regions of the American Red Cross where donors gave more than 24 components -- not 24 times a year, but 24 components in the year. They would have lost 28 percent with the limits as published in the draft guidance.

These are our 60 24-karat donors, platelet pre-count prior to the first donation of the year in 2005. You can see the distribution centers around 250,000, 260,000 -- males a little bit lower than the females. Our donors have higher platelet counts than community donors. I didn't do a statistical analysis of this, because I can't tell you that these are two comparable groups.

The point is that we select our donors to have high platelet counts. We start out trying as hard as possible, as whole blood donors come in, to recruit the donors that are 200,000 and up from the whole blood donor base and try to talk those people into apheresis, because they will be good donors, in our historical experience.

These are the pre-counts in 1,440 donations during 2005 for those 60 24-time-a-year donors. The striking thing is, number one, no donor had a single platelet count at or below the threshold of 150,000. There were 151,000 in donor 1 there.

The other thing that is quite striking is that these boxplots are the 25 th-75 th percentile, the range, and the outlier. So you can see that these donors regulate their platelet counts within relatively narrow ranges. Particularly for those in the lower end of our range, you have really very tight distributions. For those of you who have biological bachelor's degrees and whatnot, this is what we call homeostasis. They have a set point and they look to maintain their set point pretty easily.

This is across 1 through 24, the 24 donations in 2005, and it shows you what the pre-counts look like for all 60 of those donors. There is no significant change in platelet counts in linear regression from donation 1 to donation 24.

These were their first-three-ever donations in the system. Some of those go back as far as 10 years. For some, we no longer have the records, so the sample size is about half of the 60 donors here. I don't think it's real that their platelet counts went up as a result of being platelet donors. We have gone from counting platelets one at a time with a magnifying glass to an actual instrument and into the second or third iteration of our hematology analyzer during this. It's just to show you that even back as far as five to eight years, there is really no discernible trend. These people maintain their platelet counts quite nicely.

So -1 to -3 of their first-three-ever counts for those donors for whom we still have records. So it looks like their counts hang in there pretty well.

This is data that German Leparc at Tampa shared with me. It's not exactly the same. Again, these are donors up to 23 times, 4/07 to 1/05, in their program. Again, there is no discernible trend that would suggest that a relatively frequent plateletpheresis over a period of time causes a deterioration in baseline pre-counts.

This is our data again, but this is displayed according to the number of products made. These are single donations, double donations, triple donations, according to the pre-count, at each setting. And -- duh -- people who triple have higher platelet counts; that's why they triple, and double and single. Also there is no statistically significant trend in the regression analysis. So people with high platelet counts tend to maintain high platelet counts; people with moderate platelet counts tend to maintain them. It's just homeostasis.

This is from Hemacare. This was looking at post-counts. This is the relevant data here. The index pre-counts, you can see here whether they singled, doubled, or tripled. Here are the index post-counts and subsequent pre-counts. They do drop to below 150,000 on the post-count. I don't think that is terribly surprising. None of them got to 100,000 on the post-count and all had rebounded, statistically identical to their prior pre-counts.

We have asked a question of FDA as to whether a post-count really adds anything to a pre-count, when you put the donor on the machine.

That is not our data; that is Hemacare's. We don't do post-counts.

This is the pre-count versus the inter-donation interval. If you look at an inter-donation of two or more weeks, these are statistically identical pre-counts on our 60 24-time donors during 2005. This is statistically different than this. If you draw donors at one-week intervals, they do have lower pre-counts than if you wait for two weeks. Once again, none at or below 150,000.

Why is this important to us? Those of you who have been to the Upper Midwest know that when you get past 30, you become a snowbird and you spend your summers in Florida donating for German Leparc at Tampa. Nobody wants to be in Iowa in the winter. So we have a number of these 24-karat donors who, in fact, work very hard to get their 24 donations in before it gets cold. That means July and August. [Laughter]

It really means, they come home in March and they go back to Florida or McAllen, Texas, where they get dengue fever, in November.

We have a substantial number of very short -- one-week -- inter-donation intervals. But once again, none of their pre-counts were ever at or below the threshold as an acceptable donor. That is a subset of the 1,440 donations I have been describing.

This is a little complicated. I apologize. We don't believe that we need to vary the inter-donation intervals by the number of products produced, even though some of these numbers are a little small. These are some leaky bags, so ignore them; there were no products.

This has got one product, got two product, got three product, and then the inter-donation interval before the next was one, two, three, four, five, six weeks. As you can see, since the donors who triple start with such high platelet counts, they do just fine. They have very nice pre-counts. Even at a one-week interval, the median pre-count is about 280,000.

So we don't think that we have to complicate the issue, especially in Iowa, where they haven't invented computers yet, by having several strata of inter-donation intervals.

There was concern about plasma volume loss and its impact on plasma proteins. You have seen these requirements. They basically look very similar to plasmapheresis guidelines. The concern was, if we are really pushing on these people hard, could we be dropping their plasma protein levels? I have to admit, we kind of ignored this until Sharyn Orton said, "You have to do it if you want us to take you seriously." So we did.

This is a little study that we did. We took 30 of our 24-karat donors and 30 six-time-a-year, during the past 12 months, whole blood donors, matched for gender and age within three years, and looked at albumin and total protein. Here is apheresis males and females, control males and females: albumin, statistically identical, and total proteins, the same displays, statistically identical.

So even though we are beating up on these donors pretty good, their plasma proteins are very stable.

So our conclusions are:

Platelet homeostasis works. I expect the Nobel Prize for that one.
We don't think that the number of components needs to be restricted as published in the guidance.
The inter-donation intervals do not need to become complicated in terms of the number of products produced from the index donation. If a donor has a platelet count high enough to triple, their platelet count will be very nicely maintained prior to the next collection.
Post-counts -- we don't see the need, since our donors maintain their platelet counts. You saw the data from Hemacare that suggests that as well.
Plasma volume restrictions will not enhance the safety compared to whole blood donors with regard to total protein and albumin.

These folks contributed: Kim Palmer, who is here ‑‑ she and her nursing staff actually pulled most of the records -- and the people from the Red Cross and elsewhere that gave me data.

Susan Leitman I put on here because we had a couple of conversations about why their observations appear so much different than our observations. The only thing we could come up with is that that was 10 years ago, and maybe the machines have changed, maybe some materials have changed -- the manufacturers tell us nothing in the materials that they would be aware of. Personally, I think what it is, is that we select our donors from whole blood donors who have already qualified to be donors by virtue of their physical exam and infectious disease testing, and then we look at a platelet count and say, "Ah, that looks like somebody who is likely to do quite well." Generally, that means a baseline platelet count that is around 200,000, when we try to pull them into the platelet donation program, and particularly if we try to get them to donate very frequently.

So that's it. I would be happy to answer questions.

I am going to make the assumption that since Lore asked to see this data, this and the other comments are being taken very seriously and that it will be easier to live with the final guidance than it would have been with the original. But I don't know that.

Susan?

PARTICIPANT: Do you do any collections at less than one-week intervals?

DR. KATZ: Only if we have what appears to be an alloimmunized or, for some reason, refractory recipient, and we find somebody who just absolutely is a donor that works. That is not very frequent.

Thank you for your attention.

MS. FIELDS: Thank you, Dr. Katz. I know I can say thank you from everybody at the FDA to you and Dr. Leparc and the Red Cross and everybody else who has provided this data for us. It really is helpful when we are considering the recommendations that we are going to make in guidance documents.

I also want to thank Dr. Katz. He did allow us to produce his paper, so you don't have to go online and try to find it. It's right there in your book.

The next set of presentations is going to be on failure investigations. These presentations contain some important information on investigation of apheresis failures. We have asked both the Red Cross -- they did present some of this information at PhRMA, but we felt that it was important that they get a chance to explain what their processes are at the Red Cross for doing failure investigations on apheresis, so that you can potentially make a template for yourself. Hoi-May Wong, who is a consumer safety officer in the Division of Blood Applications, is going to go through some of the regulations of failure investigations.

Hoi-May Wong is a consumer safety officer in the Division of Blood Applications, in the Office of Blood Research and Review. She has worked as a blood bank supervisor and an HPC cell processing specialist before she joined the FDA in 2005. She is a medical technologist, with a specialty in blood banking. In addition to her blood and plasma review duties, she performs pre-license inspections of blood banks and source plasma facilities and has served as the FDA liaison to the Circular of Information Task Force.

Return to Table of Contents

Agenda Item: FDA Regulations and Recommendations for Failure Investigations

MS. WONG: Thank you, Lore.

Rosia Nesbitt included in her presentation that lacking a thorough failure investigation or dimensions of the failure investigation is one of the top 10 pitfalls in submissions that we have reviewed. So we will spend the next few minutes going over this topic.

Why is failure investigation important in the manufacturing process? It's because the federal regulations require you to do so in order to show product safety, purity, and potency.

Now let's review what the federal regulation says. 21 CFR 211.192 states, "All drug product production and control records shall be reviewed before a batch is released or distributed. Any unexplained discrepancy or failure should be thoroughly investigated. The investigation should extend to other batches that may have been associated with the specific failure or discrepancy. A written record of the investigation shall include the conclusions and follow-up."

In 21 CFR 606.100(c), "All records pertinent to the lot shall be reviewed before the release or distribution of the final product. The review may be performed at appropriate periods during or after collecting, processing, or compatibility testing and storing. A thorough investigation, including the conclusions and follow-up, of any unexplained discrepancy or failure shall be made and recorded."

FDA has not issued any official documents on how you should do your failure investigation. My co-workers and I have put together some of the things that we saw during review as helpful hints to share with you today. These are not the official FDA recommendations.

Here are some examples of the manufacturing process deviations that warrant investigation:

Equipment validation.
Donor suitability or eligibility.
The device that you are using.
QC
Components not meeting product specifications.

Remember, the goal of failure investigations is not to find out who did or did not do certain things. The goal is to identify the problem and then prevent it from recurring.

What is a thorough investigation? It's an investigation of the critical steps or areas where an unexplained discrepancy or failure resulted. Some may want to call it root-cause analysis. The most important element before one can begin an investigation is to recognize that a discrepancy or a problem has occurred. From there, one can gather information to find out how it happened. A common mistake in an investigation is generalizing the problem. There are four common generalizations: (1) human error; (2) procedure not followed; (3) equipment failure; the last one, inadequate training.

Don't stop at any of these four conclusions of investigations. If you ever arrive at one of these four results, ask more "why" questions. As the cause of deviation becomes more specific, your corrective action to prevent it from recurring will be more specific. Along the way, document your findings and corrective actions.

Examples of some critical steps or areas: donor, operator, supplies and reagents, device, components, environment, SOPs, and samples.

If your donor may have a history of clotting problems, you may have low yield in your cell count. Donor specifications: If your blood center used historical platelet count to qualify a platelet donor for his or her donation, it does not accurately predict the platelet yield in your current collections.

Operator performance depends on how well he or she is trained. Make sure you provide adequate training and periodic competency evaluation.

Supplies and reagents: Check if you have faulty collection kits. Make sure your cell counter reagents are properly prepared.

Devices: This includes your collection equipment, your cell-counting device, if you are using flow cytometer for your white count or using a manual WBC enumeration method.

Evaluate and review equipment-maintenance records to make sure that recommended periodic maintenance is properly performed. Make sure the device is properly calibrated or validated.

Examples of components: Different devices will have different component specifications. If you are targeting a triple platelet collection from a donor with low count, on that day you may not get what you need.

The environment: Make sure the products are properly stored and shipped in a proper container at the proper temperature.

SOPs: Make sure that your operators are following the proper SOP. Make sure they are adequately trained. Make sure your SOP includes recommendations from your device operator's manual.

Samples: Make sure the samples are collected in the right type of tubing as containers and that samples are properly stored before testing.

In closing, I want to remind you that any unexplained discrepancy may have more than a single cause. An apparent deviation may have multiple impacts on the final products. A sound QA system will identify shifts and trends that will enable the detection of unexplained discrepancies or failures.

MS. FIELDS: Thanks, Hoi.

The next speaker is going to be Ms. Faye Kugele, who will be describing the failure investigation process for apheresis products used at the American Red Cross.

Ms. Kugele graduated from Western Kentucky University with a bachelor's degree in medical technology and from the University of Louisville with a master's of science degree in systems science, both with honors. She is certified by the ASCP for medical technology and is a specialist in blood banking. She is also a certified quality manager by the American Association of Quality.

Faye's extensive working experience includes working in the clinical hospital laboratories, primarily in blood banks and transfusion services. She was also the laboratory administrative director for a 350-bed suburban hospital and for the university hospital, before joining the Red Cross in 1990.

With the Red Cross, Faye has assumed many roles, including the management and technical responsibilities of the reference laboratory, and later, taking on the additional duties of the QC lab and the management of all equipment and QC processes.

In 1999, Faye expanded her focus and joined the regional Quality Assurance Department and then became the regional quality director at the ARC River Valley Blood Services Region in 2000.

In 2005, Faye joined the Process Design-Product QC Department for ARC Biomedical headquarters and was given the responsibility for writing procedures for product QC, hematology analyzers and irradiation.

Return to Table of Contents

Agenda Item: Experiences with a Failure Investigation Program for Apheresis Blood Products

MS. KUGELE: Thank you very much.

If you came to hear Kathy Waldman, you are out of luck. You have me today.

What we are going to do is talk about a little overview of our apheresis program. We are going to talk about the sources of product quality failures and look at the components of the QC program and the importance of monitoring that QC on an ongoing basis. Of course, we will spend quite a bit of our time talking about the failure investigation itself. To finish the discussion, we want to talk a little bit about communicating the results, once you have them, and lastly, look at some challenges to a good quality program.

I want to tell you a little bit about the Red Cross. We have over 262 registered facilities that annually collect and distribute 630,000 apheresis platelets, 312,000 apheresis red cells, and over 1,000 of the granulocytes.

We use devices from three different manufacturers:

From Fenwal, we use the ALYX leukoreduced double reds; the Amicus, for leukoreduced platelets and concurrent plasma. We use the Auto-C for plasma and the CS3000 for granulocytes.
From Gambro, we are currently using the Spectra for leukoreduced platelets, concurrent plasma, and granulocytes, and the Trima for leukoreduced platelets, plasma, concurrent red cells, and also for double reds.
From Haemonetics, we use their device with the 832 Alpha[?] collection set for double reds.

The volume of testing at Red Cross for apheresis red cells continues to increase as we implement at more registered sites, and we are collecting more volume at mobile collection operations. We currently test over 100,000 red cell products each year for the volume and component hemoglobin, and we also monitor the mean total component hemoglobin each month. We test over 3,200 for residual WBC counts and percent recovery for our leukoreduced products.

For apheresis platelets, we are testing between 30,000 and 40,000 of our plateletpheresis products each year for volume, pH, platelet yield, and color, and over 4,000 residual counts each year. We test that on the initial sample and then, if that fails the 5 x 10 6, then we will test individual products to determine their leukoreduction status.

For the average results -- I want to share a little bit of that with you, for the different technologies ‑‑ the Trima is a non-leukoreduced red cell, and we are averaging a 66-gram hemoglobin. The Haemonetics ‑‑ that is, the in-line filtration that is performed in the component laboratory -- those are averaging a 53-gram hemoglobin. The ALYX, which is leukoreduced component by filtration, but is performed by the machine, with no intervention by the staff, has a 57-gram hemoglobin average.

For residual leukocyte counts on our leukoreduced components, all of them are passing the requirement to meet less than 5 x 10 6, but with our ALYX we are not seeing any residual WBCs in about 99 percent of the products tested. We do not see cells in about 94 percent of the Haemonetics products. We are doing the Nageotte count.

For apheresis platelets, the average pH is about 7.4, plus or minus .27. These results are on about 3,300 of our products, which is about one month's testing. The volume will vary. The average is 255, plus or minus 69 mL. The platelet concentration is 1,545,000, plus or minus 255,000. Platelet yield varies, 3.8, plus or minus 0.6, x 10 11.

Failures in your product quality can occur and be detected at several points throughout the lifecycle of your product. We have discovered that during validation and the implementation of a new procedure, we are particularly susceptible to failures. But it can also occur during routine collection, routine processing, and it can be discovered as a result of a customer complaint, as a result of QC testing. We are going to discuss each of those in a little more detail.

During the validation: We had a situation in late 2004, early 2005, with the implementation of a new technology. It was noted that we did not perform our monthly QC for plateletpheresis. Part of the investigation ‑‑ we involved staff in the interviews. We made sure we asked a lot of "whys." We did a lot of brainstorming. It boiled down to where we looked at a lot of the timelines and the requirements, and also our change implementation steps. It was important to understand what was going on in the thought process of everyone involved, to see where the gaps had occurred.

We missed the QC for one particular category of product, plateletpheresis, as I said. The staff had been very diligent to perform all the validation testing, all the process control testing, and forgot that those particular products were tested later, at a different time in the manufacturing process, for many of the same parameters that had already been tested.

In the resolution, we determined that all the subsequent implementations -- we put the word out and we called for a reminder to staff that the products did need to be considered for any subsequent testing that would be qualified for that product type. We also tried to be more inclusive in providing our guidance for implementation for other new technologies. The validation plans could also be used to call out a reminder for staff to remember the other QC requirements for the products.

During routine processing, the problems can be related to the collection. This can be from using the default settings for the equipment when you have a new donor. If you have a lot of alarms during the procedure, that can indicate either an equipment or a donor problem. Usually you have difficulty with the flow rate. If the procedure was ended early or if the donor had a reaction, you may end up with a complete product, but it may not be of the same quality that it would have if it had been collected normally. Also, was there a failure to recognize a machine error message that would require a product to be evaluated further?

As we move on through the process, when it gets to manufacturing, during the handling and storage of a product, you may have occasions where the product quality may be affected. As a result of incorrect decisions during your production process, when it comes time to splitting the component, you may make the wrong decision. You could have miscalculations or if, during weighing, you use the wrong tare wrong or use the wrong technique, that will introduce an opportunity for quality to be impacted.

For routine QC, many of the same issues that you see in the manufacturing of the product could also occur -- the wrong tare bag, miscalculations, wrong decisions. You can also have trouble with analyzer or clerical or transcription errors.

For multiple processes or technologies, sometimes we cause confusion with our staff when we ask them to do multiple technologies. One example is when we ask whole blood collection staff out on the mobile to also collect double reds. There are some requirements that are different between those two processes, such as recording the volume on apheresis red cells, whereas on the whole blood collection, collection staff may not be recording volume. Then, when the product gets into the manufacturing lab -- and this is especially true of Trima, where we are going to leukoreduce it using a sterile dock method -- those red cells, again after leukoreduction, have to have the volume recorded on the label, and the whole blood-derived red cells and whole blood through leukoreduction do not. So it does create some confusion.

Over the past few years, we as an organization have experienced process failures from most of the failure categories. With over 30 regions and, as I said, over 250 sites, and using almost every technology available, we have had the opportunity to see the good, the bad, and the ugly. Fortunately, we have learned to communicate, so that when we do detect an issue or a problem, we do communicate it to our other sites so that we hopefully avoid a recurrence.

It is hoped that failure is discovered prior to a product being distributed or transfused. But we must be alert to the possibility that a customer complaint, once it is investigated -- we want to make sure that we handle the situation and potential impact appropriately.

Some past experiences that you may remember: White particulate matter or red eye syndrome did occur, and each of those had quite a bit of investigation and actions associated with them.

Today we might see the impact of a possible bacterial-contaminated adverse reaction. That would require us to look at the donor records, maybe interview the donor for subsequent status of how they are feeling. Then the donor history, the process, steps would be reviewed, and any staff that were involved in the production of that component would be reviewed, to ensure that if there was a process issue going on, it would be handled.

Also if a consignee calls you and tells you they have hemolyzed red cells in their inventory, we may have to work with the consignee to investigate that issue. They will be looking at receipt records and their storage and handling, and we would participate by looking at our distribution procedure, looking to see if we have had any other customer complaints that would indicate that it's a broader problem. We would definitely look at our current inventory to see if we have any other hemolyzed units.

Now let's talk a little bit about the components of what you need to do to ensure that you are looking at product quality on an ongoing basis.

As you know, it can be a challenge to ensure that the various testing requirements for all the different products are completed within the required timeframes. We use a standardized tracking tool to help the regions make those decisions. It lists all the different products that they make, the device that they use, and the collection sites or manufacturing sites where those products are maintained.

We had a specific situation where a region did have difficulty in collecting an adequate number of samples. This involved the Trima process. The facility was using the Trima to collect primarily the double reds, but they would use it occasionally, if they had a donor who was not able to give a double red, to collect a plateletpheresis and plasma. That made them have to do QC for four singles, four doubles, and four triples, for instance they were manufactured at that site. Their practice was to collect one of their four singles, four doubles, four triples each week, so at the end of the month, you have your products and everyone is happy.

However, they found that they did not always collect a double and a triple every week. Therefore, comes the end of the month, and they didn't have enough.

So you do have to plan based on how you are using your equipment at the different sites and try to build in some ability for you to adjust to either erratic production volumes or to the way you are using your equipment.

Of course, the procedures must be comprehensive and written in a clear and concise manner. It's very important for staff to understand the steps that they need to perform, and also the things that they do that will influence the end product results.

When you are monitoring the quality of the finished product over an extended period of time, that provides your staff and management a certain level of confidence that the process is stable. They can then evaluate and detect any trends, correct any failures or adverse trends, and you can also look for opportunities for process improvement.

We are going to talk a little bit about the failure investigation and the corrective actions. I want to talk about it in a minute.

One of the areas that we have found is very important -- and I think the larger the operation, the more important it can be -- is the responsibility and accountability. If you are a one-man operation and you are basically responsible and accountable for everything, sometimes it's easy to know who to go to if you have a question. However, you have to remember, if that process isn't written and something happens to that person, then you could end up losing your program. So even if you have a small operation or if it is a large one, you want to define who is responsible for what.

The steps:

Maintaining the tracking the schedule. How are you going to ensure that you collect sufficient products and test them?
Looking at your actual results to see if they are acceptable and you are meeting your pass-rate criteria.
What staff do you notify if you have a failure? That should be defined.
The compiling and initiating of any summary reports or monthly reports.
Routing them for review.
Who is going to review? We require the medical director and QA to review all of our quality results. That is at a minimum.

The individual product documentation we review carefully as well. The reviewer looks to see if information was transcribed completely and correctly, that the calculations are correct, and that any interpretations that are made based on those results are correct.

If you have a failure, the staff notifies a supervisor, so that if the failure throws them into, say, a monthly failure for QC, for the month, they can go ahead and start looking at that process instead of waiting until the end of the month and getting their monthly data.

The supervisor also works with the staff to continue any steps to investigate of that individual product failure. If the product can be evaluated for release, that is done as well. We will talk a little bit more about how that is done in a minute.

In addition to reviewing and submitting the documentation for the individual QC results, for the monthly QC, we look to make sure that an adequate number of samples were tested by whatever required site collection site or manufacturing site, and whether or not the QC met the specified criteria and the pass rate for the month. We look at results for process improvement.

One of the things that is outside of apheresis, but that the Red Cross has been doing, is taking cryoprecipitate results for AHF, Factor VIII, and fibrinogen, and we are looking at the individual manufacturing sites and comparing them against the system mean. If one region is always running higher than anyone else, we look for maybe best practices. If someone is very erratic and beginning to trend downward, they get a flag. So it sort of alerts them to the fact that something may be going on in their process, and they can take action before they fail QC. This is an area that we want to put into place in some of our other product QC, such as pH for the platelets.

In our failure investigation, as Steve said, we use a lot of forms. We use a standard failure investigation or checklist. The one you see here is for apheresis red cells. In the back of the talk, there are several examples of checklists that we use for different product categories.

Besides looking at the obvious product exceptions or error messages that the machine may flag for certain products or calculation errors, we also look at the sample: What may have happened to it that might affect the accuracy of the sample? Is it clotted or hemolyzed? In those cases, you may need to collect a new sample. So that would need to be documented, that you were getting a new sample.

Also, if the accuracy of the test result is in question, you may have gotten a flag on your analyzer, or you could have other equipment that has not passed QC. Your scales may not have been working properly. All of those devices and your hematology analyzer can be evaluated.

Other process-related steps: You want to look at those and see if, either intentionally or unintentionally, something has been changed or you are getting more variability than normal, which would tell you that it's no longer stable. Again, tare bags are a real good example of something to look for, to make sure that those are the appropriate ones, especially if you have multiple technologies and different bag types that component staff or collection staff would be working with.

I want to go back to the product for just a minute. We talked about product suitability for release. Depending on the results of the evaluation, you may find that it may qualify to be released as a non-standard product. A plateletpheresis that has less than 3.0 x 10 11 platelet yield may qualify as a non-standard product, as long as it has at least 2.2 x 10 11 platelets per bag. Or it may be something that you can relabel as a different product type. If it fails leukoreduction QC, it could be relabeled as a non-leukoreduced component and released.

There are situations where you may use your material review board to release a product that has failed QC, but that is used only for situations where there may be a special need or a medical reason for releasing that component.

Other than that, if no rework is appropriate, such as a failed component hemoglobin or pH of a platelet, then those should be discarded.

When we get into the actual investigation, when a monthly requirement for a QC parameter is failed, we use the same checklist that we use for an individual product failure checklist. We will get those out and look at them to see what has already been looked at when that product failed initially.

We also look at historical results. We look at the past couple of months to see if we have had any other failures. If those had failures, we want to look at the checklist and make sure that we look for commonalities between them.

We also look to see what wasn't investigated, that perhaps we need to expand our investigation to include a broader scope.

For the critical components or the critical parts of the process that we want to specifically look at -- and you want to be very open when you make these observations. Sometimes that is a difficult thing when you are really close and familiar with it. Sometimes it's hard to question, because you know how it's done. So be sure to be very critical during this part of the investigation.

You want to look at the procedures and make sure they are accurate. Have there been changes to either package inserts or the manufacturer's procedures, where things have maybe changed? Maybe staff are doing things differently than what the procedures say, or maybe they should be doing it differently, but the procedures have not been updated.

You look at the equipment that is being used and look at the QC and PM. Are they acceptable?

Look for historical trends with some of your equipment, like the centrifuges used in component manufacturing. Sometimes you may see the speed on the centrifuge deteriorating over time, and what you see is the tip of failures that might occur before you see the equipment fail its QC.

You want to also look at equipment used in testing. Have you put in a new piece of equipment, and all of a sudden you are having failures? We experienced this with a new hematology analyzer, where a region implemented the technology. Everything appeared to be working fine, but, for some reason, their monthly QC on plateletpheresis was consistently showing the yields significantly lower than what they had originally. In looking at it, they were testing their sample very quickly after the sample was taken from the product. After a quite lengthy investigation and assistance from the Holland Laboratories and a study that was done there, it was noted that, for this particular analyzer and the way it detected the platelets, the sample had to set for a while before it would count all of them. So the time that it had spent in the anticoagulant in the bag, and then you take the sample ‑‑ then that sample needed to set. EDTA would start sphering the platelets, and there is reagent that is used with the analyzer that spheres the cells so it can count them.

The analyzer manufacturer told us to refrigerate the sample for an hour before we test it. We did that, brought it to room temperature, tested it. The results then were very comparable to what we had seen with the sample that had been collected routinely and that we had made our decisions on.

If you notice something and the analyzer is not acting appropriately, this is something to consider.

If you want to know the analyzer, I will see you after the meeting.

Supplies: You do want to make sure that any supplies -- also that the QC, if applicable, is acceptable and that it is in date.

You do want to observe staff during a failure investigation and make sure that they are doing the process as it was intended and that they are following the procedure. Look to see if you have either new or experienced staff. Sometimes having experienced staff with a new procedures is as bad as having new staff with an old procedure. So keep that in mind.

Also make sure that your staff understand why they are doing what they do and the impact that it can have on the quality of the product down the line.

A couple of other things, such as changes in the environment or facility. If the temperature or lighting changes, this may make it more difficult for the staff to do their job. Also electrical interference, especially in a new facility -- you might find that you are not getting a stable electrical source or there may be some interference.

Equipment software updates or equipment repairs ‑‑ keep that in mind, that you can have adverse product quality after you have conducted that.

Now that you have your information from gathering your data, talking to staff, you want to do some brainstorming and start identifying some possible causes. There are many ways you can do this. You can use some flow-charting, cause-and-effect diagramming. There are several problem-solving tools. There is no one that is better than another. Some are better, depending on the complexity of the issues you are looking at. But the main thing is that you get everything out on the table and then you start ranking it into the most probable causes.

As mentioned earlier, you may have more than one, and you may have contributing causes. So you do want to get all those out.

Your failure mode and effects analysis is a tool that can help you rank those sometimes in severity.

After you evaluate for the root cause, then you determine, is this a process failure or not? Did you find something in the donor procedure record that told you it was an isolated situation? Were the donor parameters incorrect? Does the donor have a biological variation? There are some donors who, when you collect them, their platelets clump and it's irreversible. So you may have a situation where the donor is actually your problem, and it is not really part of your process.

Your sampling technique variation -- if your staff are interrupted temporarily or they are distracted during a particular critical part of the process, that can impact your product quality.

Your product safety and risk evaluation, going back to the product: We have looked at what you can do with individual products. Even though it may only the impact the failure of one, you do want to look at the scope, depending on what cause you came up with, to see if it impacts, say, back to the last time you had acceptable QC results. You also want to look to see if it's a supply or equipment problem that came up. You will want to look to see if it impacts other product types. Your scope and impact over time and product type may be extended. It may no longer be just the one product failure. You want to look at the potential risk to the recipient for any product that may have been distributed. Depending on the parameter, that can vary.

The final action you want to be sure to determine, looking at the scope and the impact, is the potential for regulatory risk. Here we include the need for a BPD report to be filed. You may have to recall or withdraw products. The Red Cross uses a risk evaluation where we have a corporate material review board that discusses, along with regional or divisional quality and operational staff, the need to take additional actions. Sometimes we suspend activity at one or more sites, if it's indicated. Red Cross also has some other reporting requirements because of the consent decree.

Investigation documentation: We should all understand how important documentation is. It does have to stand on its own, so that over time it's just as clear as to what happened as the day it happened. You want to make sure it's independent of the audience that is reading the documentation, to make sure that they understand and can see the timeline and the actions as they were taken. At the same time, you want to be clear and concise in the information that is there.

You do have to say what happened, what actions were taken, and why -- anytime you do take an action, it needs to be evident what was going on in your mind and the things you considered in making your decision -- and then any follow-up with outcomes of actions that were taken.

After you do take corrective action, you will want to do extra testing to make sure you have no unexpected outcome. This is not QC. This is just additional testing if you have made a change to a process, to make sure that this has not caused an adverse impact. It may be just testing a few products or you may repeat process control if you have changed a critical process step.

The quality-assurance department and the medical director play an active role in assessing impact and in determining when the routine QC can be implemented and reinstated.

We want to ensure that, when changes are made to the process, SOPs are changed to reflect those changes.

What happens when you can't determine the cause? As we talked about, you go through the investigation steps and you rule out all the process causes. You look for isolated reasons that an individual product might fail. But once you go through and you are unable to pinpoint a cause, then you are kind of stuck. If it's the first failure you have had, that's a good thing. If it's not the first, you probably do have another problem going on.

So if it's the first time it has happened and you have checked everything you can think of, then document that. Document all the things that you have ruled out, what you looked at. That's why we use our checklist. Then that will give you a picture, so if it does occur later, then you can go back and try to figure out why it is happening.

That is, again, where documentation is very important.

We also get the question often of when additional testing is required or needed. Again, when you need to collect data. When we had the problem with the hematology analyzer, we did a study and we pulled a lot of samples to try to see what was going on and to ensure that we were not creating a change that was artificial by doing what the manufacturer said. So there were a lot of controls in that study.

Anytime you need additional data or if you have made changes and you want to make sure they are effective, that is the time you test additional samples.

Communicating failure information could be important to the vendor or supplier. It can also help get you information on the potential impact for transfusion recipients. Your vendor or supplier may have other information about other complaints that they have received. So that is always a good thing to follow through on.

Communicating to staff is, I believe, a critical part and a responsibility of our organization, so that when a change is made, they understand why, or even when a change is not made, they understand what the investigation was involving and the status of their product QC and potential things that can impact product quality.

Of course, you have to remember to communicate with FDA when it's appropriate.

We routinely also want to share our product QC outcomes with other departments. It's not just product QC staff or component lab staff where these products come from. You do want to make sure that other departments, or even other divisions of your organization, know what the quality performance of the organization is. It also opens up the opportunity for you to look at process improvements.

It is a constant challenge, I think, to plan and execute the strategies for meeting product QC requirements for all the different products that are manufactured today. We are familiar with the difficulty in sampling and testing plateletpheresis at the time of distribution and getting the results reviewed before the product is distributed. It can be difficult to perform effective random sampling when you have one collection site that may only produce 10 triples and you have another collection site that manufactures 40 triples, and you are going to submit four form each facility. Sometimes that, in and of itself, makes it difficult.

But when we went back and were looking at some of our data, we are testing about 6 percent of our apheresis products just using the four singles, four doubles, and four triples at each collection site, by technology. So we are probably getting a pretty good sampling of our products. But we are going to continue to monitor that.

Timing is very difficult when you have time-sensitive products, like platelets, or for deglycerolized or washed red cells, when you are performing QC, doing the testing. In order to maintain low costs and testing economies, I know a lot of places are looking at consolidation of testing. I think we are going to have to be very careful and keep in mind that, as we do that, these time-sensitive products we will have to make allowances for.

The last concern I want to mention is the difference, again, as I mentioned, in whole blood and red cell requirements -- as I mentioned before, in the labeling requirements. Also there is no standard available for whole blood-derived products as far as hemoglobin content, whereas we do have it for apheresis products.

I would like to challenge us to rethink some of the leukoreduction requirements, such as the percent recovery. For a lot of the equipment that we have today, you do not have a lot of operator or staff intervention. Your red cell recovery I know has been a very basic part of the QC requirements for leukoreduction and was very, very appropriate, and is very appropriate, where you have filtration and it requires staff manipulating it, and it can be impacted by the way it's handled. But a lot of the technology today is much more mechanically controlled. The controls within the equipment ensure the effectiveness of the processing for the donor blood. We monitor the donor blood as far as red cell loss, and we are monitoring our components as to their quality and content with the component hemoglobin.

So what is the purpose of red cell recovery in those situations? What does it tell us about product quality? Even if it failed percent recovery, would that signal us before we would notice something -- maybe our donors losing too many red cells first or having failures due to product content? I think we would see that before we will notice a red cell recovery pointing us to a failure. But that can be a discussion for another time.

Basically, in summary, there are many pieces to putting together a successful program to ensure quality and safe products. A failure investigation is just one piece, although a very important one to stay in control and prevent future failures. It has been our experience that apheresis QC failures tend to be more often a result of the poor planning or human mistakes that occur during implementation of a new technology. So pay special attention.

True process failures are usually related more to the products that require a lot of manipulation, like random platelets or cryo.

For apheresis, as we continue to update our methods with the latest software capabilities and we utilize the research that occurs for collection and storage of our products, we continue to provide the safest and highest-quality products for our customers as possible.

MS. FIELDS: I want to thank Faye for coming and sharing her process on failure investigation. As most of you either know or have found out very painfully, FDA considers failure investigation a big part of manufacturing, and we are always looking for you to find out what went wrong, especially with your apheresis devices, because most of them have a less-than-1-percent chance of actually making an error. So when it does, we really want to drill down and figure out what the problem was.

Since we are running ahead of schedule, the manufacturers have agreed to bump up their schedule. We will still hit break probably around 3:15.

Our first speaker is going to be Merilyn Wiler. Merilyn is a medical technologist and a blood bank specialist, with over 25 years of experience in transfusion medicine, education, and the medical devices industry. During her career, she has worked in a hospital transfusion service, taught in both a university-based medical technology and community college-based medical technology program; taught anatomy and physiology in a university-based gifted young children's program; and served as the director of quality and regulatory affairs for a large, full-service blood center. She is currently a customer regulatory support specialist for Gambro BCT.

Return to Table of Contents

Agenda Item: Device Manufacturer's Forum - Merilyn Wiler

MS. WILER: I just want to say thank you to the FDA for putting this together. Obviously from the attendance, it is needed, it is wanted, and there has been a lot of useful information that has been provided today. So I do thank you for putting this on for everyone. I'm sorry that you couldn't get everyone in, being oversubscribed. Perhaps we need part 2, to do it again for the rest of the folks.

What I would like to do today in this presentation:

To just briefly review the list of support materials that Gambro provides and the intended use for each of those support materials.
To review the Trima section of the licensing checklist that FDA has posted on their Web page.
To highlight some revisions that have been made to that checklist.
To briefly review a short list of the Trima device-specific specifications that need to be taken into consideration when you are implementing and when you are licensing your products. These requirements are in addition to what FDA and accrediting agencies require.
To highlight a couple of best practices that are associated with the devices.
To include a few comments on 7-day platelets, on labeling.
To provide some helpful Web addresses where you can access some of this information that I am going to be discussing.

In your handout, what we have provided to you are copies of the presentation slides and we have provided to you a revised section for that licensing checklist as it relates to the Trima system. We have also provided a table that lists the various software versions for Trima and the licensing paths that are associated with each of the software upgrades.

Gambro BCT provides a variety of documents. In this case I am going to be talking primarily about Trima Accel. We provide similar documents for the Spectra system that also collects platelets, but I am focusing primarily on Trima in this presentation.

Of course, we have the operator's manual. We use this as the primary training reference. It contains the requirements and recommendations that are associated with the device. It describes the configurations, the use, the care and the maintenance of the Trima system. The manuals may be ordered separately and ahead of time, so that you can begin to revise your SOPs prior to even receiving the device, if you care to. In addition, you can order additional ones if you would like to, so that you can provide them to the various locations throughout your organization.

These are available in hard copy and they are also available on the Gambro Web site, so they can be accessed at any time.

We also have a system administrator's guide that is provided. It adds additional detail on how to configure the system. It includes donor settings, procedure parameters, product definitions for the Trima. These also are available in hard copy and are orderable.

With each software upgrade or revision, we provide customer letters. In that, we describe what the new features are and the changes that have taken place between the various software versions. Also in that letter, we try to include whether or not it is a major, minor, or moderate change, which is an indicator as to the extent of validation that might be considered when you are going to validate the process. We try to get a discussion with the FDA, so that they concur with whether or not it is major, minor, or moderate, so that we are all on the same page and starting from the same standpoint. If there are specific validation requirements, we would include that in the letter, if that is known at the time the letter is composed.

We do have other documents that are not in the manual because they go across various device lines. The bags that store platelets on the Spectra system, as well as the Trima system, are exactly the same. So the storage requirements are irrespective of which device it is collected on. We do have a separate document that describes the platelet bag criteria for storage of platelets.

When it comes to 7-day platelets, we have a packet of information that describes how to request the variance to allow you to extend the expiration date to 7-day platelets. This is available not only on the Gambro Web site, but also on the Fenwal Web site, because you can do that with the Fenwal device. Also we have a common Web site called passportstudy.com, where that information can be obtained.

These documents that I have just described are the primary references where Gambro includes recommendations and what needs to be done, associated with the different devices.

We also provide additional literature to help support a successful implementation. These are just to help the customer plan and execute, and give you a list of things to consider when you are trying to put together your implementation plan. These are not intended to include additional requirements that are associated with the devices, unless it is specifically so stated in the document itself.

In this, we include things like example training lists. This is something the customer can take and customize for the intended audience that they are training. There are quick reference guides that have abbreviated setup and operational instructions. We also have a chart of height and weight that can be used to help you predict whether or not a specific donor might be eligible for double red cell donations, because the double red cell donations require a total blood volume of 4,500 mL, which is unique, compared to our other procedures that can be collected on the Trima system.

We also provide example validation plans and sample SOPs. Again, these are provided as templates and starting points for developing facility-specific SOPs and for the validation plans and for helping to design procedure run records.

These documents are categorized as examples. They are not intended to dictate specific content. They are available also on the Web site.

We have a couple of documents as well: the Red Cell Process Control Guide and the Leukoreduction Red Cell Process Control Guide, and the Guide to Platelet and Plasmapheresis.

Some of these are quite old. We are in the process of revising them all and updating them and standardizing them. But they do provide some example SOPs, and they describe various methods by which you may take samples from the various components, so that you can do your product quality control for your validation and licensing procedures. In some cases, they provide formulas and conversion factors that can be used when you are starting to qualify your products.

Now moving on to the FDA licensing checklist that they have posted on their Web site. As soon as that was published, we arranged to have a teleconference with FDA, and we reviewed them and mutually agreed to several edits for the Trima-specific section of that document. We were told during the conference that FDA would make the edits that we discussed internally to the document, but that the document on the Web site might not get updated for some time.

What I have done is gone through the Trima section and rewritten it and included it in your handout with the revisions that were agreed to. I am just going to briefly summarize the modifications that I have made.

In the Trima section, there was a list of products that can be collected with Trima. That has been updated to reflect the products and the combinations that can be collected on the Trima Accel system.

In the red cell loss section, when it dealt with version 5, what we had was a statement that talked about platelet-only procedures. Actually, that would be the same as if you were doing an LRS red cell, platelet, plasma procedure, because the same set is used. So I have deleted the section that dealt with platelet-only.

There was also a statement in the red cell section when it dealt with leukoreduction that said that the red cells had to be stored at 4 degrees centigrade for eight hours prior to leukoreduction. That has been removed. What we suggest is that you follow the IFU, or the information for use, that is associated with the leukoreduction filter that you are using.

In the plasma section, we suggested a clarification. The maximum plasma volume of 600 mL is the maximum amount of plasma that can be collected and frozen in one of our plasma bags. It's not the maximum amount of plasma that can be collected during a procedure. The Trima is currently cleared to collect up to 15 percent of the donor's total blood volume, and up to a liter of plasma can be placed into a single collect bag when it is liquid. However, when you go to freeze it, you would have to transfer a portion of it to a different bag prior to freezing, so that you wouldn't exceed the 600-mL limit for freezing the bag. That's just so the bag won't rupture during the freezing process.

In the red cell section, we suggested that it be clarified that the limitation for a total blood volume of greater than 4,500 mL prior to doing the procedure was restricted only to the double red cell procedures and does not apply to the single procedure when you are doing a red cell or when you are doing platelet or plasma collections.

In the donor profile section, there was a statement that dealt with red cell loss. We asked that that be excluded, because that is generic and it applies to all systems. It's not unique to Trima.

Then there was one statement that we would like modified. It's outside the Trima section. It fell in the section that dealt with leukoreduction review checklist, the general section. There was a statement that talked about, "Trima may include a minimum red cell volume of 160 mL." That is not one of our criteria, so we asked that that be taken out of that section as well.

This concludes the revisions that were made to the checklist. The revised checklist is in the section that I have provided in your handouts.

The next thing we were asked to do is to go into some specific requirements that are associated with either our device or our procedure. What we have done here is provided the storage requirements or limitations on the apheresis platelet product bag itself. There is a volume limit per bag that you can see, a range. There is the platelet concentration range for the bag. There is a total number of platelets per bag -- a higher limit here. Then the temperature requirement is generic across all different platelet products.

As far as the plasma bag, we have already talked about that. The maximum volume that can be placed into the bag during collection is a liter. However, the maximum volume that can be frozen is 600 mL.

Then the collection-specific criterion: The maximum volume that can be removed from a donor during a single sitting is up to 15 percent of the total blood volume for the donor. For your larger donors, this 15 percent TBV per donation may exceed the 500- and 600-mL limit that a lot of the guidance documents talk about, removing plasma at a single donation. But the device is cleared for this purpose. Therefore, it should be okay to do that. As long as your annual volume limits don't exceed your 12 liters for a donor weighing less than 175 pounds or 14.4 liters for donors that weigh greater than 175 pounds, these donors can still be collected with the higher volume. If they are collected infrequently, with the appropriate -- more frequently than every 28 days -- they can still be classified as an infrequent donor.

The criteria for donors to be collected on the Trima are the same as those for whole blood donors, plus what FDA has stated as criteria for donation interval and for frequency and for blood loss. In addition, for the double red cells, as we have said, you must have a total blood volume of greater than 4,500 mL.

We also have the claim that the leukoreduction claim for the device is less than 5 x 10 6 white blood cells per collection.

Per request, I have included a table in your handout that lists the various Trima software versions and a summary of the licensing paths that are typically used by customers when they are upgrading from one to another. It includes upgrading from our Spectra system or a competitor's device to a Trima Accel system or upgrading from one version of Trima to a Trima Accel.

During validation, one thing that customers can look at to validate their procedure is that the actual plasma and platelet volumes should be within plus or minus 10 percent of the displayed volume at the end of the procedure. For the red cell product volumes, we suggest that you take the end-run summary red cell volume, add 100 mL, and compare that to the actual volume of your red cell product that was collected after the storage solution has been added to it. Those should match within plus or minus 10 percent.

Gambro does provide the example validation plans. We are in the process of revising those, for those who are interested, not so much to change the content of them as to try to standardize them across all of the different product lines that we have, in the way of format and content. I do emphasize the fact that these are example validation plans and templates, and they are only one option or a starting point from which you can begin to develop your own individual facility validation plans. It is not Gambro's intent in these plans to interject additional requirements for your process. We would very much appreciate customer feedback if we do something like that, so that we can get it out of there.

As for the number of products to test and which assays to perform, we do refer you to the FDA guidance documents and to industry standards, such as AABB. If FDA and AABB are silent, Gambro will work with you to help you identify and clarify what your needs are and provide information on device functionality, so that you can come up with something that you are comfortable with in your validation procedure.

I was also asked to discuss a couple of the best practices that we typically share with our customers. Here we are trying to share with you things that can impact variability in the product that you might get from the Trima system.

One of the most important ones for donor safety and for product quality is the pre-hematocrit, which is typically done on the day of donation, and prior to the donation is the ideal. Based upon the information that you enter into Trima on your pre-hematocrit, the red cell and plasma interface is established in the Trima collection channel, based upon that value that you enter. So it's critical for Trima to have accurate information so that you know where this interface lies, because that, then, helps the machine know where the platelet-rich plasma lies, so that it can harvest the platelets and the plasma and the red cells appropriately.

If you enter an artificially low hematocrit, this could result in red cells actually spilling over into your platelet and plasma product, because the interface is riding too high in the channel. If you enter an artificially too high hematocrit into the system, you might wind up with a platelet product that doesn't meet your expectations and the actual yield will be too low, because the interface is riding too low and Trima is actually harvesting above the platelet-rich plasma line. So you are not getting all the platelets that you thought you would get.

So the pre-hematocrit is very important when you are performing the procedure.

A second important value is the pre-platelet count. Based on this value, the Trim actually determines how much blood has to be processed, which dictates how long the procedure will actually run, so that you can get your desired product. If you enter a pre-platelet count that is less than actual, there is a chance that the final product will have more platelets in it than you desired. If this is the case, it could result in a high concentration, which would result in its being outside the storage limits for the actual bag that you have. If you enter a pre-platelet count that is too high, there is a chance that he final product will have fewer platelets in it than is expected. You could wind up with a product that doesn't make yield, and so you would have a substandard dose of platelets in that particular product.

On this screen what I have provided is a table describing the various methods of obtaining a pre-platelet count that are currently used and their relative effectiveness in providing the desired platelet product when each is used. As you can see, the ideal is taking a venous sample day of, and entering it prior to or near the beginning of the procedure. The variance in the finished product, which is in the far right-hand column, is usually just around 5 percent.

We realize that there are a lot of places that just simply can't do that. Sometimes it's not possible to even get that pre-platelet count back before the procedure is actually finished. So other options are available for entering information for the pre-platelet count.

One option that is provided is to average the previous three pre-platelet counts and use that average as the pre-platelet count that is entered into the machine. As you can see in this grid here, the variability in the product you are going to get is wider, and so you will have more products that won't meet your anticipated expectations. That could be attributable to the pre-platelet count that you enter.

If you don't have three prior donations, the single previous pre-donation is acceptable to be used.

The least favorable one and the one that provides the most variability is if you use the machine default. In some cases, facilities have gone across and obtained a mean pre-platelet count for their donor population and used that in lieu of a machine default. They seem to think that they get more reliable products from what they targeted if they use the mean rather than a machine default.

The option to extend apheresis platelet expiration date to seven days is available when you are collecting products on either the Gambro BCT equipment or on the Fenwal apheresis equipment. This option is available to both registered and licensed facilities. Some people feel if they are registered, they are not eligible. They are.

However, prior to implementing, whether you are registered or you are licensed, you still must submit a request in the form of a prior-approval supplement for an exception or permission to use an alternate procedure, so that you can extend the expiration date from five days to seven days.

The details of what to include in this prior-approval supplement and how to submit are outlined on the PASSPORT study Web site.

Please be aware, for registered facilities, submitting a prior-approval supplement doesn't mean you are submitting for licensure. Some people get confused. A lot of people think, prior-approval supplement; therefore, you are licensed. That's not the case. The prior-approval supplement is a method to obtain the variance or approval to extend the expiration date. So if you are a registered-only facility, the 7-day prior-approval supplement approval only allows you to extend the expiration date; it does not allow you to ship across state lines.

If you are a blood establishment that is already licensed for 5-day platelets at the time that you submit for your 7-day prior-approval supplement, then, once you receive the 7-day approval, you can ship those across state lines as well.

When folks call and ask about labeling, what we try to do is refer them to the AABB Web site for the Codabar list of product labels. When we are asked about ISBT Code 128, we refer folks to the ICCBBA Web site.

If you have ever gone there and looked, there are over 5,000 labels on that blood product code list. It's very lengthy. They are not clustered together by device type, so it's a little tough sometimes to figure out which product code might be appropriate for the particular device that you are looking for. What we have done for the Trima products is to actually go through the list and, for the more commonly produced products off the Trima system, created a crosswalk. We have included the current Codabar codes that most people are using and then what your options are for ISBT Code 128 for the Trima Accel anticoagulant additive combination products for red cells, platelets, and plasma. It's not all-inclusive, but it's a good start. At least it gets you looking further into the list. A lot of our product codes are toward the end of the list. They are not up front, even though some up front look like they could be used for our products.

So that is available on the Web site.

For facilities that are preparing for a 7-day platelet license or prior-approval supplements, there is a list of the Codabar codes and ISBT codes that are appropriate for those products on the Web site, as well as example labels that actually have the eye-readable text on them for Codabar. We do not have them yet for ISBT Code 128. These are all available on the PASSPORT study Web site, the Gambro Web site, or the Fenwal Web site.

Here is a list of the Web sites that I have used to gather my information. If you would like additional information about them, this is where you can go to get that. If you have any questions about what I have presented here or anything else related to our systems, please contact Gambro BCT at either this phone number or the email address provided here.

I thank you for your attention.

MS. FIELDS: It's heading on towards 3:00, so I think we are going to take this opportunity to have our afternoon break. This is also our "Meet Your CSO" session, so please introduce yourself to your CSO and get to know us a little bit. Most of the CSOs have business cards if you want direct contact information.

Let's plan to be back by 3:30.

(Brief recess)

MS. FIELDS: I want to thank everyone for staying this afternoon. I do want to reiterate that the evaluations are in the front of the notebooks, in the administrative section. It really is a good tool for us to understand how the workshop went and what you are looking for in the future, if you can just take a couple of minutes to fill that out before you leave and put them at the front desk.

We are going to get started. The first presentation this afternoon is going to be from Dr. Sharyn Orton from Fenwal, who is going to discuss her automated devices.

Return to Table of Contents

Agenda Item: Device Manufacturer's Forum - Sharyn Orton

DR. ORTON: I want to thank the agency for inviting me. Obviously, I must have left on good terms, if they had me come back.

I sort of want to apologize to Lou Katz, but not really. One of the things I learned the hard way in the government is that you, in fact, have to take a very serious position seriously. I struggled with that for five years. So I waited five years to get him back. [Laughter]

The other thing I would like to say is that quite a few of the staff from the FDA that presented today were staff that worked for me. I want to tell them, good job. You guys did a great job.

[Applause]

When I was asked to do this licensing workshop for Fenwal -- I was actually one of the inventors of the infamous checklist and struggled through it for all the different devices early on. So I thought I would be going back to the operator's manual to see if I could, in a concrete way, put together any additional information that Fenwal requires, and where this information could be found. The actual information that is additional to anything that the FDA requires or recommends was not much. That part was easy.

But when I started going back to try to find the same information that you as the user would need, I struggled quite a bit to do that.

Cheryl Rocher [phonetic] works in Regulatory at Fenwal, and we are in the process right now of reviewing our operator's manuals to ensure that the information that you need to provide in your procedures is, in fact, in an easily accessible place.

Agenda-wise, I am going to talk a little bit about these devices and these products. The ALYX is cleared for double leukoreduced red cells; leukoreduced red cells with plasma; non-leukoreduced double red cells is in what we call a limited launch; and plasma-only, which is actually not in distribution.

Amicus has single need with optional red cells and plasma, double need with optional plasma, 5-day and 7-day platelet storages. We are part of the PASSPORT program.

The CS-3000 is still in use, and Auto-C for plasmapheresis.

For donor eligibility, we don't ask for much. We ask that you follow the FDA regulations or guidance. You can follow AABB standards if you choose. There are some additional Fenwal criteria for ALYX. When you are dealing with double red cells, the sex, height, weight, minimum hemoglobin or hematocrit, to determine the appropriate red cell target and volume, need to be used. Similarly, for the red cells and the plasma.

For validation, quality control, and acceptance criteria, again we don't ask for much. We do get some questions about, "An FDA guidance document doesn't make any kind of recommendation for validation, particularly some of the old ones. How come we have to do it? We have gone to submit our SOPs to the agency and they say we have to do validation."

In fact, there is a GMP talking about sound and appropriate sampling plans required in laboratory controls. In fact, in changes to an approved application, you have to submit evidence that your procedure works, and that qualifies as validation.

Any AABB standards can be used as well.

One of the things I kind of struggled with as I was putting this presentation together -- and in the interest of now being able to be jovial, since I don't work for the government anymore -- I wanted to give you some other kind of helpful information.

To keep critters from eating flowers or vegetables form your garden, are you all aware that you can use shave-scented soap or big cat manure, and they both work great? [Laughter]

My brother used to own a landscaping company, and he told me to add this in. That's useful tip number one.

So for product validation, on the ALYX, one of the pieces of information that is given is that for double red cells, the volume versus the target is a comparison that is made, and it can be either hemoglobin or total red cell volume. This is really not a product validation per se. It really is taking a look at the performance of the device.

As far as the product goes, for the leukoreduced red cells, residual white blood cells need to be done, and percent recovery.

Some of you may be aware that our version 1.5 software for ALYX displayed the percent recovery calculation on the screen for every collection. When version 2.1 came out, for reasons I am not going to get into -- and I can still say, before my time -- that screen was removed. So it did not enable the users who were moving forward with this software to have a percent recovery. There was no way for them to actually calculate it. They didn't know what the software was calculating.

So for version 2.1, we ended up sending out a customer letter that provided the equation that is used by the software, so that the blood centers could do 1 percent QC.

Having been an end user, I personally would prefer to have the machine calculate it and show it on the screen all the time, rather than having to sit and calculate it and hope I didn't make a mistake. Our original plan was for the next version of the software, sometime next year, to reinstitute that screen. But, in fact, we are in the process of getting that screen back available for any of our end users who either do not want to move forward with the 2.1 software because of the calculation or are using the calculation and would prefer not to.

But there are no additional Fenwal-specific requirements for validation.

For quality control, again no specific requirements. I just went over the information about the percent recovery screen.

For the ALYX red cells themselves, this is a specification that the target versus the actual volume -- red cell volume plus or minus 10 percent of the target. This is really a device performance metric. The leukoreduction claim is less than 5 x 10 6 with 99 percent confidence, 99 percent of the time. In fact, the device will collect leukoreduced red cells with 85 percent recovery.

But I do want to note that non-leukoreduced red cells from these double collections are not cleared on this device. So if you do not get two leukoreduced red cells from an ALYX double red cell procedure, the products that you do get are not cleared for the device yet.

For ALYX plasma, there are no specifications.

For Amicus, the device will collect equal to or greater than 3 x 10 11. The maximum storage for the container is here. Standard storage volumes are collected by a default volume set on the machine, although lower volumes have been validated. It's cleared for 700 mL per storage container, depending on the weight of the donor. The residual white count is here. The singles and doubles are cleared for collection. However, the triples are cleared for component. So a triple collection that fails 5 x 10 6 is, in fact, not a performance failure, if the individual components meet the criteria.

For Amicus red cells, there really are no specifications, generally. But the hematocrit range tends to be between 80 and 85 percent, and a volume range of 220 to 240.

Amicus plasma -- no real specific things, outside of this volume being within 20 mL or plus or minus 10 percent of the target. Again, this is a device performance metric rather than a plasma product metric.

Auto-C specifications are that plasma for transfusion must be transfused or processed as FFP within six hours of collection and source plasma can be collected on this device.

So useful tip number two: If you are in the mood for some sun and sand, and don't feel like going to the beach, one of the most breathtaking sights in the world is, in fact, the pyramids in Egypt. This is yours truly, along with Dr. Dean Alfath [phonetic], but I don't recommend going in August. We were on the Giza plateau this day and it was 118 degrees.

So the next thing is, where is the information that you, as the end user, need? Where is it located? For the most part, it's in the operator's manual or kit disposable's directions for use.

The first thing I went to look for was, what is each of these devices cleared for?

Has anybody gone through any of our operator's manuals to see exactly what our devices are cleared for? You are not going to find it. In fact, it isn't listed anywhere. It's there by chapter, but it's not listed anywhere. It is described in the separate chapters.

As far as donor eligibility, the ALYX operator's manual appendix contains the nomograms that you need. For specifications, for ALYX, it's in the operator's manual, Chapter 1, "Introduction." For Amicus, it's in Chapters 2 and 4.

Hence, you can see why we are revising this to try to make it more user-friendly for you.

For validation, the operator's manual, in fact, doesn't contain anything. For ALYX, there are documents that Fenwal has, called Guidelines and Suggestions for Implementation, Core Documents. It includes this list. You can use these documents if you wish.

For Amicus, in fact, in the operator's manual, Chapter 10, there is some operational qualification for different software versions, and then there are some additional documents that you can obtain from Fenwal to do your validation. There is a document called Product Quality Validation that does refer to any FDA requirements or recommendations or AABB standards.

Calibration and maintenance are both in Chapter 6 for both of these devices.

As FDA has stated, failure investigations really have become quite important. I would like to say, I had a large part in that, because I certainly feel that it is very important, not just for the agency, but certainly for the user, not to be so concerned when things go right, but to be concerned when things go wrong.

So I went looking for any information that we could provide you, as the end user, on what to do in a failure investigation. In fact, I was a little hard-pressed to find where the information was.

In ALYX, there is a section in the introduction on things that might impact the quality of the product if you are not cognizant of them. I wouldn't have looked there for it normally, but I went page by page.

In Amicus, it's in Chapter 2, called "Before Using Amicus," which is kind of odd. You do your failure investigation after you have used Amicus. There is an operator's workbook that has a Session 5 that is called "Common Mistakes and Consequences."

Product mixing, which is a big issue, particularly for getting adequate samples off -- Amicus does have some references to product mixing. ALYX does not. ALYX and Amicus do both have some information on sampling. For ALYX, it's what they call the core documents, which you can get from Fenwal. There is an SOP for sampling final product. In Amicus, in Chapter 4, there are procedures for sampling.

Additional risks that were brought up several times during the FDA presentations, additional risks to include in informed consent, for ALYX are found in Chapter 1, and in Amicus, in Chapter 2.

Residual red cells and plasma for collection and annual loss calculations are, in fact, in the appendices for all of the Fenwal devices.

So useful tip number three: Did you have any idea that you could use DW-40 for all of these things? This is only about a quarter of the list that my brother sent me (he's a wealth of knowledge): protect your silver from tarnishing, remove road tar grime or bugs, keep pigeons away -- I need it right now. I live in a cornfield, with the pigeons.

They actually work. I tried these. So don't say I didn't give you any information.

So the bottom line is, what services are provided by Fenwal? They have training and training documents and train-the-trainer documents. They do have validation document support, as I have noted. They will provide you with draft SOPs.

One of the things that I found somewhat alarming when I came to Fenwal -- I was talking about the fact that I was a little bit concerned when I had my FDA hat on that when the blood centers needed help, and they were using some of my competitor's devices, they could call those competitors and really get the information they needed from them. The complaint was, if they called Baxter, they didn't get any information. So, in fact, they would call me at the FDA.

In fact, that was a Baxter culture, that you provided with the documents and really didn't give too much more information. There were certain people who did. Rick, who is from Fenwal, is here, and he is one of those few people that was always willing to help the end user. I am working with him and with Dr. Rocher. We are going to begin providing not just answers to your questions, but regulatory support.

I certainly think I have a little bit of experience with what FDA is looking for when they want SOPs to come in from the end users. Whether it's a registered use-only facility or you are preparing licensing applications and you have questions, if you need help in the development of SOPs, if you want to know what failure investigation information you need, what would be helpful ‑‑ is the QC form, in fact, going to meet their criteria -- we would be more than glad to help you in any way. Certainly there are situations where the best thing to do is to call your consumer safety officer for an opinion on something, but there is definitely a certain amount that we at Fenwal can help you with.

Any additional information requests -- you have already submitted something to FDA; they want additional information; you are confused, because perhaps Fenwal has not provided the information that you need -- do not hesitate to call, and any post-inspection questions.

So useful tip number four: To hear some of the best blues played in the United States, come to Chicago. It's the best part of living there.

For this regulatory support, you can call me. Here is my office phone number. It's easier to email me. I am not often in my office. I like to think that when I was at the FDA, I was always available to anybody who had a question and needed help. That certainly is the same with myself and my staff at Fenwal. So any help we can provide, we would be glad to do that.

Thank you.

MS. FIELDS: Our next speaker today is from Haemonetics. Her name is Susan Finneran. She is the director of regulatory affairs for Haemonetics Corporation. In this role, she is responsible for the global regulatory strategy for Haemonetics apheresis equipment and solutions.

She works closely with the research and development group to support the development of new products. She is also responsible for submitting the premarket notifications, investigational device exemptions, as well as the new drug application supplements pertaining to solutions.

In her role as director of regulatory affairs, she works closely with the Haemonetics clinical group to develop clinical protocols in support of these submissions. In addition, she supports the Haemonetics field staff by providing advice on validating and licensing strategies.

Return to Table of Contents

Agenda Item: Device Manufacturer's Forum - Sue Finneran

MS. FINNERAN: Good afternoon. As Lore said, I'm Sue Finneran, director of regulatory affairs for Haemonetics Corporation.

Today I am going to provide an overview of some of the key points to consider when licensing products that are manufactured using Haemonetics devices.

First, I just want to say, this is a great forum for manufacturers and FDA and blood centers to get together and hear the same requirements. I have certainly learned a lot today from all the presentations.

I would also like to find out how, next time, I can not be the last presenter, and also not be after Dr. Orton. So if I have made any enemies at FDA, let me know what I can do to rectify that. [Laughter]

I am going to cover the MCS+8150 device. That is our two-unit red cell collection system that has been on the market for a number of years. I am also going to talk about the Cymbal Automated Blood Collection System, which is our newest device for the collection of two units of filtered red cells. I will touch upon collection of platelets and plasma with the MCS+9000 device and briefly just talk about the ACP 215 device in terms of the requirements, although it's not technically an apheresis device.

The 8150 device, as I said, has been on the market for many years. It has been on the market since the late 1990s for the collection of two units of filtered red cells. The target volume of red cells is based on FDA-cleared nomograms for autologous and allogeneic donors. For autologous donors, there are pre-donation hematocrit requirements, as well as a weight requirement. For allogeneic donors, there is a pre-donation hematocrit requirement, a weight and a height requirement. Those are all detailed in the operator's manual, as well as the FDA checklist.

The two-unit filtered protocol has a standard target collection. Pre-filtration is 360 mL of red cells. The filtration is completed with the 2RBCF protocol. It's completed via gravities in the RC2H filter manufactured by PALL Corporation. It is integrated into the disposable set.

A single red cell unit can be saved from an aborted procedure from a 2RBCF procedure, but quality control needs to be completed -- residual white cell counts, red cell volume -- and this is not considered to be a licensable unit. That information is detailed in the FDA checklist that has been included.

As far as the QC criteria, the 2RBC and RBC protocols, non-filtered red cells -- our criterion is that 95 percent of the red cells should be within plus or minus 15 of the target. There is, again, a variable target. If you are targeting at least 200 mL, you can apply the AABB criteria -- that is, a mean red cell volume of 180 mL and at least 95 percent of the units having a minimum red cell mass of 150 mL.

For the 2RBCF protocol, the AABB criteria can be used. You also need to demonstrate that you are meeting greater than 85 percent recovery, looking at the pre-filtration red cell bag, and that you have less than 5 x 10 6 residual leukocytes.

As far as filtration, warm filtration can be completed at room temperature or it can be completed cold. With room temperature, it must be completed within eight hours of collection, and cold filtration must be completed within 72 hours of collection. I have also listed the normal timeframes for filtration. This is based on the clinical data that we have collected over the years.

In the FDA checklist, it states that if the product fails to filter within these timeframes, the unit should be evaluated for residual leukocytes. Recently, a 510k was filed to modify the statement. That is to be in line with PALL Corporation's recommendations, not to be so prescriptive. Basically, now it's stating in our manual that filtration times can be influenced by collection and processing conditions and biological variability of donors. Experimental data with some filter products indicate that a prolonged filtration time can be an indication of suboptimal leukocyte reduction.

So it's not so prescriptive that each product must be tested, but it's more based on your center's validation.

In terms of the tips for success:

Certainly people have mentioned stripping, mixing, proper sampling technique. We detail those in our operator's manual. We recommend that testing be completed very close to the time when sampling is done, to get an appropriate measurement.
The daily check of the weigher to make sure the scales are functioning properly is important, as detailed in the manual.
Maintenance of equipment. We have a detailed section of daily and weekly and monthly maintenance activities that must be conducted to make sure that the machine is functioning properly.
We also recommend that cell counters and scales that you are using to complete these measurements are calibrated. In terms of failure investigations, those are areas where we sometimes find problems when we are asked to come in and look at the process to see why the red cell volumes are not what they were expected to be.
It's also important to maintain the stability of the device at mobile sites, not only to avoid damage, but to make sure that the weighers are functioning properly and that the device is level.

Now I am going to switch gears and talk about the Cymbal, which is our two-unit filtered red cell collection device:

It's a portable device, with optional battery power. So it can be run off of AC or battery power.
It is equipped with a data-transfer system.
It's a handheld device that is used to collect and transfer procedure information to a PC.
It has a standard target collection of 360 mL. That is pre-filtration.
It also uses the RC2H filter.
The filtration is done after each cycle. There are four cycles of collection. There is on mechanism to look at recovery.
Unlike the 8150, at this time there is no option to save one unit of red cells in the case of an interrupted procedure. But we have put together a clinical protocol and we have gotten feedback from FDA, and will soon be starting to conduct that trial, so we can offer that option as well.

The QC criteria as specified in the operation manual: We basically follow the AABB criteria of mean red cell volume of 153 mL, with at least 95 percent of the units having greater than 128 mL. Also we specify that all the units should have less than 5 x 10 6 residual leukocytes.

Where the device displays QC products, in the operation manual we also have a volume range that is given. Since we just recently released that device, look at some of the validation data that has come back. We base that range off of the clinical data. As the clinical results were pretty close to what we expected, and not getting down to that lower range, looking at that range -- that may need to be expanded, because it's a little too restrictive, given our expectations for the potential red cell mass that could be acceptable.

An interesting thing about this slide -- it has gone through a few iterations. The first iteration of the slide said that recovery was not required. Then we got a call from Dr. Vostal and he said, "What do you mean, it's not required?" Actually, as far as recovery, just like the ALYX device, there isn't a reservoir bag, so there really isn't a way to complete red cell recovery.

We have talked to Dr. Vostal in the Division of Hematology, and because I am optimistic and I know that they are going to agree with me eventually, I wanted to leave this note as, "It's currently being discussed with FDA."

I just want to say, I didn't talk to Faye. I didn't talk to Faye at all. She made a good point about the recovery. If you would consider that, I would appreciate it.

But we will be advising our customers -- certainly it's important for licensing. People need to know what the requirements are, so not to make light of it. But FDA is requesting that we use 180 or 360 as the pre-filtration red cell mass. So in the event that we need to complete that recovery calculation, we can use the centers' information that they have already collected and pretty easily demonstrate that you will meet that requirement.

I am hoping to advise our customers in the coming weeks as to the appropriate action to take.

I am going to switch gears and talk about the MCS+9000. That is our mobile platelet collection system. It's used to collect platelets and plasma, with or without saline compensation. Again, I have updated the FDA checklist. There are a number of different disposable sets that are noted on there, but a couple of them are obsolete. Right now, as far as I know, most centers are using the 994CF-CPP set. The other sets, which use the CLX bag, which is a slightly different type of PBC, are obsolete. I am pretty sure that there aren't any centers using those.

This is interesting. "CF" stands for "continuous filtration." As noted on the FDA checklist, there is no requirement for recovery. It also notes on the checklist that the device displays recovery, but it actually does not. That is currently listed on the checklist.

The platelets can be stored up to five days. I didn't want to confuse anybody. We are clear for 7-day storage, but we are currently not participating in the PASSPORT study. What I can say is that we are putting together a proposal to float by FDA to see how we can validate our platelets for use with the BacT/ALERT system. Right now you cannot store our platelets for seven days, but as far as the quality of the platelets, we are cleared for that. We just can't take advantage of it.

The 994CF-CPP bag is cleared for storage of up to 5 x 10 11 for five days. We have some concentration requirements there: a maximum concentration of 2,600 x 10 6 platelets per mL.

I will just make note that the sets with the CLX bags are no longer available. They have lower requirements for storage, 3.5 x 10 11, and the maximum concentration is also a bit lower.

The typical volume range for a single platelet is 250 to 300 mL, with a minimum of 200 mL. The typical range for a double platelet product is 350 to 450. Our leukoreduction claim is that we have 95 percent confidence that 95 percent of the products have less than 5 x 10 6 residual leukocytes.

In terms of tips for success, we recommend that you may want to target a bit higher -- for example, targeting at 4 x 10 11 to get to a platelet count of 3 or targeting at 7 to get to a platelet count of 6. Certainly, the actual pre-donation counts provide the best results. When you have those accurate results and you plug them into the algorithm, it works quite well and it is very accurate. But, also as noted in some prior presentations, the average historical pre-donation counts also give very reliable results.

Again, we have detailed instructions for product storage, product sampling, just to ensure that you have the highest accuracy in terms of your sampling and platelet counts.

Just a quick word on ACP 215. This is our system that is used for cell processing, for glycerolization and deglycerolization of red cells. It's a closed system, which provides for extended storage of deglycerolized red cells for up to 14 days post-deglycerolization. The red cells can be frozen for up to 10 years prior to deglycerolizing them.

The criteria that are included in our premarket notification and that are the basis for validation:

We have 95 percent confidence that 95 percent of the units have less than 1 percent hemolysis after 14 days of storage post-deglycerolization.
In terms of recovery, 95 percent confidence that 90 percent of the units have a recovery of at least 80 percent post-deglycerolization compared to the pre-glycerolization unit.

That's all I have.

MS. FIELDS: Thank you, Sue.

I just want to make a comment on a couple of the issues that were brought up during the manufacturers' presentations. I think there was a common theme that some of the checklists do not get updated on the Web as often as we update them internally. We are aware of that. There is actually a big note on the Web site that announces this to the public.

We have not updated it recently, because the next change or revision is probably going to be pretty significant, and so we would rather wait until we can make all the changes at once, because of the time it takes to get them up on the Web.

Our last presentation of the day -- we want from "The Top 20 Questions" that you ask FDA to "Frequently Asked Questions," because we didn't actually get 20 questions. We did get questions through the docket. We have taken a shot at answering most of them today for you. Some of them are slightly repetitive, but we want to answer everybody's questions. You may say, "Why is that up there twice?" But we did want to be fair to everybody.

There are a couple of slides -- and I apologize to whoever put them on there -- that were not included, because some of the things we just can't talk about yet, if there are products that haven't been approved or touchy subjects like that. So I did delete about four of the questions. I do apologize. If they were your questions, please feel free to ask them privately.

Can we have the panel come down?

We should probably apologize to Eileen today, because this is normally her job. But since we are asking the questions from industry, we are going to let our associate director of policy ask the questions for the panel.

Return to Table of Contents

Agenda Item: Frequently Asked Questions and Answers

MS. SCHARPF: Thanks, Lore.

Question number 1: What is an appropriate number of procedures to perform for validation of a double red cell procedure for licensure by the FDA?

MS. NESBITT: FDA regulations require blood establishments to perform appropriate tests before release of their products. Those requirements can be found in 21 CFR 210, 211.10, 606.100, 606.140, 610.1, 610.10.

FDA believes that the quality-control program described below would satisfy this requirement.

Phase 1 validation: Evaluate 100 consecutive red blood cell units in the following manner. Determine the expected or target red cell volume, either total red blood cell product volume or absolute red blood cell mass volume, as determined by the device operator's manual. Also determine any other target parameters specified in the device operator's manual. Compare the target value and the actual value to determine product acceptability. Use the product acceptability ranges described in the device operator's manual to determine if the product meets the volume specification. If ranges are not available in the operator's manual, the product must at least meet the volume range described in the labeling requirement, 21 CFR 606.121(c)(6).

PARTICIPANT: Can we ask follow-up questions? Why 100?

MS. CIARALDI: At the time that the guidance document was written, the consultants to the guidance document, which were statisticians, felt that this was a sufficient number to get a good idea of the operation. As I mentioned in my discussion, if you would like to use an alternative plan for validating, you can certainly send that in, as an alternative approach to the recommendations we have in a guidance document. That option is always open to you.

But that was advice that we received from the statisticians, looking at types of collections, looking at the number of blood centers. It was their recommendation that we include that in our guidance.

MS. SCHARPF: Question number 2: If a collection site collects less than 50 automated red cell collections per site per month, is it possible to obtain a license, if QC is performed on all the products each month?

MS. NESBITT: The answer is yes. The Phase 2 QC section recommends, if fewer than 50 units were manufactured at the center during the month, all units should be tested. However, when applying for licensure, if your monthly manufacturing totals are significantly less than 50 units, you should ensure that you submit an adequate number of units for us to evaluate for your manufacturing process.

PARTICIPANT: What would the FDA consider that number to be? I think that's where the confusion is coming from the field. You are saying 100, and then you are also saying an adequate number of it's less than 50.

MS. FIELDS: Question number 2 has to do with quality control, not validation. Phase 1 quality control is pretty much validation. The Phase 2 quality control, which is the 50 units, is really QC.

I think what we are really looking for here is -- obviously, some centers aren't going to get 50 if they don't collect that many, if they have 40 or 30 or something like that. We may deem that an appropriate number for a total month.

We do get phone calls, though, that say, "I collected 20 one month and I collected two the next month. Can I send that in?" That is where we get into whether that is really enough to evaluate the manufacturing process at that facility. We judge those on a case-by-case basis, when they are that low.

MS. SCHARPF: Question 3: If we are assessing individual instruments -- i.e., if we use three Trimas and one Spectra -- do we collect four products using each Trima and four from the Spectra? Or are we assessing the process? In other words, do we QC four products total from the Trimas, ensuring that at least one product collected on each machine is included in the four?

MS. FIELDS: Currently, the FDA recommends that four units of each device type per product type -- singles, doubles, triples -- at each collection center be tested. In the example that they asked here, this would mean that they should QC four of each product type, singles, doubles, and triples, from the Trima and four of each product type, singles, doubles, and triples, from the Spectra at each of their facilities.

They should also have a sampling plan to ensure that at each of these facilities, each instrument is tested ‑‑ some sort of plan that they have on how they are going to do their quality control.

MS. SCHARPF: Question 4: What can a facility do if their collection total for a particular month is insufficient to meet the total number of products required for QC?

MS. JONES: The firm should include documentation that four units were not tested for the month, but that 100 percent quality control was done, due to the number of units collected.

MS. SCHARPF: Number 5: Are singles and doubles considered different product types for QC purposes; i.e., must we treat singles as one product type and doubles as another?

MS. FIELDS: Yes, currently, the singles, doubles, and triples apheresis platelets are considered a different product type for quality control. For red blood cell QC, however, you need to do a total of 50 units and ensure that a sampling of singles and doubles, if manufactured, are tested.

PARTICIPANT: Could you repeat that?

MS. FIELDS: The question was basically for, I think, plateletpheresis, but I also wanted to answer for red cell apheresis, because those are different. With the red cells, you have to do a total of 50 for your monthly QC ‑‑ not 50 singles and 50 doubles, 50 total. But you need to ensure that you have some singles and some doubles, if you manufacture both items.

MS. SCHARPF: Question 6: When a blood establishment is requesting licensure for red cells, platelets, singles, doubles, and/or triple collections, and plasma, does CBER expect to see every possible product combination appear in the two months of QC data submitted with the license amendment or supplement?

MS. JONES: The answer is, no, at this time we are not requesting that all product combinations are represented. We are licensing the apheresis products individually.

MS. SCHARPF: Question 7: When licensing multiple apheresis platelet products, what criteria does FDA use to determine what will be included in the approval? If a submission is for triple apheresis platelets, will double products at the same facility be automatically included in the approval, or must a center request them to be included separately?

MS. FIELDS: The firm should indicate in the cover letter what products they are requesting to be licensed. Products listed will be the only ones reviewed, unless you contact your consumer safety officer after you send in your submission, to amend your submission to include additional products.

Additionally, we must received evidence -- for example, two months of QC for platelets -- of acceptable manufacturing for each of the products.

MS. SCHARPF: Number 8: When submitting for a red blood cell/platelet procedure, how does a blood center know what products are subject to licensure? If a collection yields a double red cell and a platelet product that has a yield sufficient for a triple platelet product, can the triple platelet be made, and are all products subject to licensure?

DR. HOLNESS: The FDA will only license product combinations that the apheresis device was cleared to manufacture. You should ensure that your firm only manufactures products that the device was cleared for by the FDA. You should refer to the operator's manual or the package insert for a complete list.

MS. SCHARPF: Number 9: Why must products from each collection site be submitted to FDA when apheresis collection sites are located within a small geographic area, all products are transported identically, SOPs and training are identical, sometimes the same staff even work out of multiple locations, and all product quality-control testing is performed at one lab? This appears to result in unnecessary loss of volunteer donations without a clear benefit.

MS. CIARALDI: Products are sent in from each facility as part of the review and approval for the manufacture of the products at that facility. It is done to confirm that the products are being made according to product standards, again at that facility. As we saw on the slides, there are products being sent in for limited situations much less than in the past. However, licensed blood establishments with an approved comparability protocol do not have to submit products or platelets for each facility that is using that approved comparability protocol.

The submission of a comparability protocol implementation plan provides information for us and assurances to us that you are manufacturing the product in a manner that will not adversely affect the product. You must implement the manufacturing process as it is approved in your comparability protocol.

PARTICIPANT: So, Judy, it's the regulatory pathway that is available to accomplish what they want to accomplish.

MS. CIARALDI: I don't understand your question.

PARTICIPANT: If it doesn't go in as a comparability protocol, then, yes, it would need to be products from each site.

MS. CIARALDI: In most cases, that would be true. For the specific contents of the submission and all the activities that are involved in the submission, the manufacturer can certainly contact their CSO. As a general sense, that may be necessary, but there may be some exceptions where products may not be needed.

MS. SCHARPF: Number 10: According to the May 29, 1996 recommendations and licensure requirements for leukocyte-reduced products, it must be determined that a product labeled as leukocyte-reduced must have fewer than 5 x 10 6 total residual white blood cells, but there is no specific requirement to determine the actual white blood cell yield. Is it necessary to actually calculate the final white blood cell yield on a leukoreduction QC data sheet submitted in a licensure supplement if we provide the acceptability parameters for product volume and cells counted?

DR. WILLIAMS: There are actually two ways to interpret this question. I will try to cover both of them.

The 1996 guidance states that leukocyte-reduced blood products should contain fewer than 5 x 10 6 residual leukocytes per container, and establishments should include this yield value in their quality-control data and have documentation at the facility -- in other words, not just the parameters of the determination, but the final outcome of the determination.

That said, one could also interpret this as meaning, does the final outcome need to be an actual count of white cells versus a plus/minus determination as to whether it met the 5 x 10 6 criterion.

The answer to that is the latter. You can have a plus/minus criterion. We actually encourage methods, given the labor-intensity of the Nageotte method, to try alternate procedures, such as pooling or other techniques, which would give a reliable result as to whether it meets that criterion. There really is not a recommendation to provide an actual count.

MS. SCHARPF: Can a license submission for concurrent plasma be sent in for any apheresis process that is already licensed? Are there any limits -- i.e., concurrent plasma with a double platelet or red blood cell procedure?

DR. HOLNESS: The answer to the question is, yes, a firm may send in a submission for only plasma products. However, the FDA will only license the product combinations that the apheresis was cleared to manufacture.

MS. SCHARPF: Please explain the CBER policy on the regulation of apheresis activities in registered hospital-based transfusion services that collected products for in-house transfusion only, and the statutory authority behind that policy.

DR. HOLNESS: Blood and blood products are regulated as drugs under the Federal Food, Drug, and Cosmetic Act. Manufacturing means collection, preparation, processing, or compatibility testing of any blood product that meets the definition of a drug, as defined in Section 201(g) of the act. The regulations that apply to establishments that manufacture -- that is, collect -- apheresis blood components are contained in Title 21 of the Code of Federal Regulations, Part 211 and Parts 600 and 640.

MS. SCHARPF: Number 13: Does FDA require that a new apheresis collection site be inspected by the local FDA district prior to submission for apheresis products?

MS. CIARALDI: FDA inspections are conducted for the following situations: a blood establishment that is applying for a brand-new license, licensed blood centers that are adding new collection sites, brand-new collection sites, and as we need to to facilitate our reviews. As part of our requirements in the Code of Federal Regulations, we are required to determine that an establishment and a product meet applicable regulations to ensure the safety, purity, potency of the products, and we are required to do this before we approve the product. So for some submissions, we may need an inspection prior to the approval of that submission.

The real question here says, do we need to get inspected prior to the submission? The answer to that is, no, the inspection is not needed prior to the submission. But if the inspection has not been conducted before we are ready to approve, it could hold up the approval, because we will need evidence of that compliance with the regulations.

Again, the inspection is a part of our review for many of our products. It's a confirmation that the facility is operating in compliance with the regulations. For brand-new facilities, there is no compliance history on file for us to base that determination on. Therefore, an inspection may be needed.

MS. SCHARPF: Number 14: In reference to the FDA Web site on plateletpheresis for doubles and triples, when the collection residual white blood cell count is greater than or equal to 5 x 10 6, the device has failed to leukoreduce and a failure investigation should be initiated. Counting each of the baby bags and labeling as leukoreduced if the residual white blood cell count is less than 5 x 10 6 is appropriate, providing a failure investigation of the collection is completed.

So the question is, does this mean the failure investigation must be completed prior to labeling and release for distribution? It would be very difficult to complete the investigation -- investigate, document, review, and sign-off by all -- prior to the expiration date of the product. Or is it that a failure investigation must be performed as part of the monthly QC, but not before distribution?

DR. WILLIAMS: That is part A of the question. The regulation governing that is 21 CFR 606.100(c). Hoi-May went through this, but it's short, so I will read it again.

With respect to failure investigation, all records pertinent to the lot or unit maintained pursuant to these regulations shall be reviewed before the release or distribution of a lot or unit of a final product. The review or portions of the review may be performed at appropriate periods during or after blood collection, processing, compatibility testing, and storing. A thorough investigation, including the conclusions and follow-up of any unexplained discrepancy or the failure of a lot or unit to meet any of its specifications, shall be made and recorded.

In this particular case, the white cell count would need to be performed to ensure that the unit is leukocyte-reduced and labeled appropriately prior to distribution of the product. The investigation should take place in a timely manner, but the regulation does not specify that this has to be fully completed prior to distribution of the unit.

There is a part B to that question.

MS. SCHARPF: Yes, there is a B: How can we reduce the QC requirements for automated red cell collections? The 50 collections per site per month are overwhelming and hold little value. An automated instrument does a better standardized collection than a manual method -- tip scales.

DR. WILLIAMS: The QC recommendations are based in guidance. One of the statements included in every guidance is that guidance is nonbinding and manufacturers are free to submit alternate procedures for review by FDA staff. You can also certainly contact in advance to determine how to go about doing that.

So the answer to the question is, anyone wishing to submit an alternate procedure should contact their CSO to discuss the procedure and how to submit the submission to the FDA.

I want to make one additional comment, following up on the spirit of Steve Kassapian's talk. Yes, we are very happy and encourage you to contact us to help you put together your submissions, but we would urge you to queue up your questions so that this interaction can be efficient. Keep in mind that any time that a CSO is spending on the phone answering individual questions or playing telephone tag is time that isn't going into finishing the approval of a supplement. So please help us be efficient as well. We are happy to work with you, but keep that under consideration.

PARTICIPANT: For part A, could you cite the reference to the CFR?

DR. WILLIAMS: 21 CFR 606.100(c).

MS. SCHARPF: Question 15: In reference to Haemonetics Cymbal device for two leukocyte red cell products, during the validation process of this machine in blood centers, I understand from FDA the following: Any failure -- i.e., not able to manufacture two units -- means that we must start validation over again from the beginning, and not just add on another unit or two. The Haemonetics Cymbal was cleared for two units of leukocyte-reduced red blood cells, so we must successfully manufacture two leukocyte-reduced red blood cells with each procedure. If a unit is not leukoreduced or the second unit is not a usable unit, then the validation failed and the process must begin again.

If we have to discontinue the procedure because the donor has a reaction or something like this, is it considered a failure and we would have to start the validation over from the beginning?

MS. FIELDS: I just want to read something out of the guidance here. It says, if more than 5 percent of the units fail to meet the product specification, the cause of the deviation should be investigated and corrected as part of the overall quality-assurance program, and the Phase 1 qualification process should be repeated. Units that do not meet the acceptable range as specified in the device operator's manual or in the product labeling should be evaluated to determine their suitability for the distribution.

So the guidance does allow for five or your 100 units to fail without your actually failing your entire validation thing. As I mentioned earlier, usually your apheresis instruments have a less-than-1-percent failure rate.

That being said, in this example, where they were talking about -- in the question they talked about if the donor had a reaction. A donor having a reaction wouldn't be a unit that we would normally see included in those validation numbers. That is not really a device failure.

What we usually tell people is that they should write their validation plan and include reasons that they are going to extract any data out. So if you have very specific reasons why you don't think your units should be included in your 100 or 95 percent compliance, all of that should be stated before you even start your validation. If your donor has a reaction, no matter what the instrument did, there is no way you will probably get a good unit.

So all these things should be thought of ahead of time, to determine how you are going to run your validation protocol.

MS. SCHARPF: Number 16: Can someone talk about the electronic submissions gateway system for submitting submissions? I am just in the process of registering and looking to start submitting electronically.

Please discuss acceptable submission binding and show good and bad examples. What is an ACCO-type binder?

Please discuss packaging and labeling of shipping box and envelopes?

DR. WILLIAMS: I guess the core answer here is, spend some time with the CBER Web site, because everything that we are telling you here is documented on the Web site, either as a guidance document or an SOP or some sort of procedure that is documented there.

Specific to the electronic submissions, there are guidances for electronic BLA, electronic IND submissions. You can find general information at www.fda.gov/cber/esub/esubguid.htm.

Simply by doing a search on the Web site, you can easily find anything pertaining to electronic submissions.

With respect to the binder and the submission of paper documents, there is an SOPP that covers that also on the Web. That is SOPP 81-10, Investigational and Marketable Application Submissions of Regulatory Documents to CBER, and an additional SOPP, 800-7, DCC Binding Procedures for Regulatory Documents. They are both fascinating documents. [Laughter]

An ACCO-type binder is a two-piece paper fastener used with submissions that are hole-punched. Submissions should be three-hole-punched. Hard binders -- i.e., notebooks -- should not be used, because they could open during shipping and handling by the carrier and are not easily stored.

Bad examples of bindings of submissions -- we have seen them. We didn't take photos of them to show you, but we have had huge stacks of submissions with a single rubber band around them, and they didn't arrive in very good shape. So there are reasons for some of these procedures.

A couple of words about some of the e-government initiatives, particularly in our office. I think it's fair to say that our office is kind of taking the lead within CBER, and to some extent within FDA, on some of the electronic initiatives. We are routinely now using electronic signature on documents. In some of the letters you have received the signatures might look identical. That's because we are using Adobe Acrobat for a lot of the signatures. We are making documents available electronically. Instead of running up and down Rockville Pike with paper documents, we are posting them in what's known as an e-room, which everyone can access and work with the documents. We are archiving them into an electronic document room, which is a CBER utility that was developed.

Similarly, we are working on ways to improve our internal and our sponsor meeting efficiency. The fellow crouched behind the laptop there is Jeff Smith. He is a real IT guru and has helped develop a procedure for a simple Web-based Web-cam procedure. This conference is actually being simulcast to our building simply through that PC.

We are taking, I think, some very positive steps to help make us more efficient, in the spirit of e-government. I think a lot of these things are really helping us do our job better and more efficiently.

I heard the other day that an ultimate goal for FDA is to require all submissions to come through the gateway in an ECTD, or electronic common technical document, format. Blood establishments are a long way from that. We seldom get an electronic submission from the blood community. But it's something to keep in mind, particularly if you are putting in a new BLA or a supplement. Give the electronic submission a try.

PARTICIPANT: Can you repeat the references to the SOPs? And did you say binders were or were not acceptable?

DR. WILLIAMS: The ACCO-type binder is mentioned in the SOP. That's simply a cardboard binder with clips that go through a hole-punched document, not a hard notebook.

It's SOPP 81-10 and 800-7, both relating to paper submission of documents. It's important to follow those. You want your document to look nice for our reviewers.

MS. CIARALDI: I just want to add something about what Jeff Smith is doing. As you know, this workshop was overflowing, and there were several CBER people that agreed to hold back and give up their seats so that a lot of other people could attend this meeting. It is those people that are sitting back at headquarters that are receiving this simulcast feed. So I want to give credit to Jeff for setting this up, and thank you to the people that stayed back home, so to speak.

MS. SCHARPF: Our final question: When licensing multiple apheresis products, what criteria does FDA look for in a comparability protocol?

MS. CIARALDI: Another thing I want to say is, I'm glad I got the small questions with the big print. That's always good for me.

First, the content of a comparability protocol is essentially the same as the content of a regular submission that has more than one product. All the information that we talked about on the slides that is available in our checklist that deals with each of those products would be included in a general submission.

On top of what is included in a general submission for our comparability protocol, you would have a written description of how the specific manufacturing change will be implemented at the additional centers. It would include the testing, the validation, the training -- everything that you would be doing to determine that, by implementing it at another center in a reduced reporting category, you have put all the steps in place to ensure that there is no adverse effect on the product.

Again, if you want multiple products to be included in the comparability protocol, just with a regular submission, you would include all the information that we have been requesting for all of the different products. For instance, if you want platelets and red cells, you would need to include labels for both of them and two months of quality control for both of them. If you want to get a comparability protocol that includes both of those products, it would be that information, on top of which would be added the implementation plan that you have heard described today.

Thank you.

MS. SCHARPF: Finally, we provided the phone number to contact consumer safety officers within the branch at DBA.

I would just like to remind you that this program is being transcribed, so if you missed any of the citations or reg cites, you can always go to the transcript and pull that up. That will be available in a few weeks on the Web site.

MS. FIELDS: I just have a few things to say in closing, before everybody runs out.

I just want to thank once again CBER, HHS, AABB, and ABC for cosponsoring this workshop. AABB has kindly provided the compendiums for you, your notebooks with all the slides and everything in them and all the additional information that was included. ABC provided all the catering. We really couldn't make these workshops happen without them.

[Applause]

One last kudos: The workshop planning committee included Dr. Bianco, Sue Finneran, Joe Giglio from AABB, Steve Kassapian from ARC, Sharyn Orton, and Merilyn Wiler, from our industry side. From inside, it included Dr. Williams, Elizabeth Callaghan, Jennifer Scharpf, Rhonda Dawson, who is our policy analyst, who did everything -- we could not have put this together without her -- myself, Rosia, and Hoi-May.

So thank you to everybody who helped make this workshop work out well today.

(Thereupon, at 4:40 p.m., the meeting was concluded.)

Return to Table of Contents