Guidance for Industry
Handling and Retention of BA
and BE Testing Samples
This
guidance represents the Food and Drug Administration's (FDA's)
current thinking on this topic. It does not create or confer
any rights for or on any person and does not operate to bind
FDA or the public. You can use an alternative approach if
that approach satisfies the requirements of the applicable
statutes and regulations. If you want to discuss an
alternative approach, contact the FDA staff responsible for
implementing this guidance. If you cannot identify the
appropriate FDA staff, call the appropriate number listed on
the title page of this guidance.
I.
INTRODUCTION
This guidance is intended to provide
recommendations for study sponsors and/or drug manufacturers,
contract research organizations (CROs), site management
organizations (SMOs), clinical investigators, and independent
third parties regarding the procedure for handling reserve
samples from relevant bioavailability (BA) and bioequivalence
(BE) studies, as required by 21 CFR 320.38 and 320.63. The
guidance highlights (1) how the test article and reference
standard for BA and BE studies should be distributed to the
testing facilities, (2) how testing facilities should randomly
select samples for testing and material to maintain as reserve
samples, and (3) how the reserve samples should be retained.
The guidance also clarifies and emphasizes points addressed in
§§ 320.38 and 320.63.
FDA's guidance documents, including this
guidance, do not establish legally enforceable
responsibilities. Instead, guidances describe the Agency's
current thinking on a topic and should be viewed only as
recommendations, unless specific regulatory or statutory
requirements are cited. The use of the word should in
Agency guidances means that something is suggested or
recommended, but not required.
Following the generic drug scandal in the
1980s, the FDA issued an interim rule in the Federal Register
of November 8, 1990,
on the retention of BA and BE testing samples. The intent of
the interim rule was to deter possible bias and fraud in BA and
BE testing by study sponsors and/or drug manufacturers.
Following public comments, a final rule was issued in the
Federal Register of April 28, 1993.
Implementing regulations are located in 21 CFR 312.57(d),
314.125(b)(17), 314.127(b), 314.150(b)(9), 320.31(d)(1), 320.38,
and 320.63.
In the preamble to the final rule, the
Agency stated that the study sponsor and/or drug manufacturer
should not separate out the reserve samples of the test article
and reference standard before sending the drug product to the
testing facility.
This is to ensure that the reserve samples are in fact
representative of the batches provided by the study sponsor
and/or drug manufacturer for the testing. The study sponsor
and/or drug manufacturer should send to the testing facility
batches of the test article and reference standard so that the
testing facility can randomly select samples for
testing, and material to maintain as reserve samples. The drug
product should also be maintained in the sponsor’s and/or
manufacturer's original container (see section III).
Also in the preamble to the final rule, the
Agency noted that reserve sample retention is the responsibility
of the organization that conducts the BA or BE study.
The intent is to eliminate the possibility of sample
substitution by the study sponsor and/or drug manufacturer, or
prevent the alteration of any reserve samples from a study
conducted by a contractor before release of drug product samples
to the FDA.
FDA’s Division of Scientific Investigations
(DSI) and field investigators from the Office of Regulatory
Affairs (ORA) conduct inspections of clinical and analytical
sites that perform BA and BE studies for study sponsors and/or
drug manufacturers seeking approval of generic and new drug
products. A frequent finding from these inspections is the
absence of reserve samples at the testing facilities where the
studies are conducted. In many cases, DSI finds that testing
facilities return reserve samples to the study sponsors and/or
drug manufacturers, against the direction of the regulations in
21 CFR 320.38 and 320.63. In other cases, study sponsors and/or
drug manufacturers, SMOs, or contract packaging facilities
designate the study test article and reference standard for each
subject, and preclude the testing facilities from randomly
selecting representative reserve samples from the supplies. DSI
also finds that deviations from the regulations more often occur
in BE studies with pharmacodynamic or clinical endpoints in
which the studies are confused with clinical safety or efficacy
studies. The pharmacodynamic or clinical endpoint BE studies
are usually multisite, blinded studies conducted under contract
(either directly with the study sponsor or drug manufacturer or
through an SMO) by physicians or clinical investigators who use
their own clinics or offices to conduct the studies. Moreover,
some clinical investigators believe that they are not CROs and
are not required to retain reserve samples. This guidance
clarifies the responsibilities for retention of reserve samples.
We recommend that the study sponsor and/or
drug manufacturer send to the testing facility batches of the
test article and reference standard packaged in such a way that
the testing facility can randomly select samples for
bioequivalence testing and samples to maintain as reserve
samples. This will ensure that the reserve samples are in fact
representative of the batches provided by the study sponsor
and/or drug manufacturer and that they are retained in the study
sponsor’s original container. Because the study sponsor
and/or drug manufacturer may provide a testing facility with a
variety of container sizes and packaging, FDA is flexible in
applying the representativeness requirement described in 21 CFR
320.38. For example, any of the following random sampling
techniques might be used by the testing facility for the
container size and packaging described
(bolded text is particularly relevant).
Single Container – If a single
container of the test article and reference standard are
provided to the testing facility, the testing facility should
remove a sufficient quantity of the test article and reference
standard from their respective containers to conduct the study;
the remainder in each container should be retained as reserve
samples in the original containers.
Multiple Containers – If multiple
containers of the test article and reference standard are
provided to the testing facility, the testing facility should
randomly select enough containers of the test article
and reference standard to conduct the study; the remaining
containers of the test article and reference standard should be
retained as the reserve sample in the original containers.
Generally, multiple open bottles are discouraged. We encourage
testing facilities to limit the number of open containers
retained as study reserves.
Unit Dose – If the test article and
reference standard are provided to the testing facility in unit
dose packaging, the testing facility should randomly
select a sufficient quantity of unit doses of the test
article and reference standard to conduct the study; the
remaining unit doses of the test article and of the reference
standard should be retained as the reserve samples in the
original unit dose packaging. Therefore, it would be
inappropriate to provide the study medications in unit dose
packaging and all the reserve samples in bulk containers.
Blinded Study – If the study is to
be blinded and the test article and reference standard are
provided to the testing facility in unit dose packaging with
each unit dose labeled with a randomization code, the
study sponsor and/or drug manufacturer should provide the
testing facility with a labeled set of the test article and
reference standard sufficient to conduct the study and with
additional, identically labeled sets sufficient to retain the
“five times quantity” (see section V). The testing facility
should randomly select a labeled set to conduct the study; the
remaining labeled sets would be retained in their unit dose
packaging as the reserve samples. For a blinded study,
we recommend that the study sponsor and/or drug manufacturer
also provide to the testing facility a sealed code for use by
FDA should it be necessary to break the code. The sealed code
should be maintained at the testing facility.
If the same batches of the test articles
and reference standards initially provided to the testing
facility are used in performing more than one study, only one
reserve sample of the test article and reference standard in
sufficient quantity need to be retained. The reserve samples
should be identified as having come from the same batches as
used in each study. However, if additional supplies of the test
article and reference standard will be used by a testing
facility to perform the same study or additional studies, the
testing facility should retain a sufficient quantity of reserve
samples from the subsequent shipment. If a CRO with multiple
testing facilities conducts more than one BE study (e.g., fed
and fasted studies) for the same drug product, and the study
test article and reference standard are sent to the testing
facilities in different shipments, we recommend that sufficient
quantity of reserve samples be kept for each study at each
testing facility. These approaches are to ensure that the
reserve samples are in fact representative of the batch provided
by the study sponsor and/or drug manufacturer to the testing
facility.
The quantity of reserve samples should be
sufficient to permit the Agency to perform five times all of the
release tests required in the application or supplemental
application. The rationale for requiring the five times
quantity is provided in the final rule. The clinical
investigator can obtain the amount that constitutes the five
times quantity from the sponsor and/or drug manufacturer. For
solid oral dosage forms (e.g., tablets, capsules), an upper
limit of 300 units each for the test article and reference
standard can be considered sufficient to meet the five times
quantity. Because the Agency has limited experience with the
retention and testing of non-solid oral dosage forms, the Agency
is unable to recommend an upper limit for the retention of
non-solid oral dosage forms at this time. In the case of a
reference standard that is an extemporaneously compounded
solution or suspension or a reconstitutable powder, we recommend
that the pure active ingredient and the unconstituted powder be
retained. For a multisite BA or BE study, we recommend that the
total amount of reserve samples to be retained across all
testing facilities satisfy the five times quantity requirement.
Each site is asked to retain a reasonable amount of test article
and reference standard to be determined by considering (1) the
total number of testing facilities participating in the study,
(2) the number of subjects expected to be enrolled at each
testing facility, and (3) a minimum limit (e.g., 5 dose units)
for each of the test articles and reference standards. If the
reserve samples from more than one testing facility are
transferred to an independent third party for storage, we
recommend that the independent third party segregate the reserve
samples from the various testing facilities so that any given
reserve sample can be unambiguously associated with the testing
facility from which it came.
Because of the variety of study settings
potentially involved in conducting BA and BE studies, several
examples are provided here. These examples are not the only
possible study settings. However, in all
instances, the chain of custody of the reserve samples used in
the study should be preserved. The sponsor and/or manufacturer
and any storage facility should document and maintain the
transfer records for Agency verification.
CROs are the most common study sites. Many
BA/BE studies of oral dosage forms are conducted at CROs to
support approval of abbreviated new drug applications (ANDAs),
new drug applications (NDAs), and NDA supplements. CROs
typically conduct single-site, open-label, crossover design
studies with healthy volunteers as participants.
Study sponsors and drug manufacturers
sometimes conduct BA and BE studies through a CRO, university
faculty, hospitals, or clinical investigators in private
practice. The testing facilities are usually clinical study
units in universities, hospitals, or clinics run by physicians.
The responsibilities of the study sponsor
and/or drug manufacturer include:
·
Packaging, distributing, and shipping the test
article and reference standard to the testing facility
·
Monitoring the study if it is conducted under an
investigational new drug application (IND) (rarely needed for
most ANDA studies)
The responsibilities of the testing
facility are as follows:
·
The clinical investigator or designee (such as the
study coordinator or research pharmacist of the testing
facility) should randomly select sufficient test article and
reference standard to conduct the study from the supplies
received from the sponsor and/or drug manufacturer, and retain
the remaining study samples as study reserves.
·
The testing facility or the pharmacy of the
testing facility should retain the reserve samples.
·
If the testing facility does not have adequate
storage, or goes out of business, the reserve samples can be
transferred to an independent third party with an adequate
facility for storage under conditions consistent with product
labeling.
Note: When studies are
conducted at universities, hospitals, or physicians’ offices,
the clinical investigator or physician conducting the study
should not send the reserve samples back to the
study sponsor and/or drug manufacturer. The goal is to
eliminate the possibility for sample substitution by the study
sponsor and/or drug manufacturer, and to preclude the alteration
of a reserve sample from a study conducted by another entity
before the release of the reserve sample to the FDA.
When BA or BE studies are conducted by an
SMO, they are frequently multisite, open-label studies of oral
dosage forms in patients, or multisite, open-label studies of
nonoral dosage forms with pharmacodynamic or clinical
endpoints. Often, the study sponsor and/or drug manufacturer
contracts with an SMO to recruit clinical investigators and to
monitor a study. The SMO is involved directly or indirectly
(i.e., by subcontracting to another party) in packaging and
shipping of study test articles and reference standards to the
testing facilities. The testing facilities are usually the
clinical study units of CROs, universities, hospitals, or
clinics run by physicians.
The responsibility of the study sponsor or
drug manufacturer is to ship the test article and reference
standard to the SMO under contract, or to the packaging facility
under subcontract to the SMO.
The responsibilities of the SMO include:
·
Packaging, distributing, and shipping the test
article and reference standard to all testing facilities (or
subcontracting a packaging facility to perform this function)
·
Monitoring the study at different sites if it is
conducted under an IND (rarely needed for most ANDA studies)
The SMO should not randomly
select and retain reserve study samples. As explained in the
preamble to the final rule, the Agency intended that sufficient
test article and reference standard to conduct the study should
be randomly selected at each testing facility, and that each
testing facility should retain the remaining study samples as
reserves.
[7]
The responsibilities of the testing
facilities are as follows:
·
The clinical investigator or designee (such as the
study coordinator or the research pharmacist of each testing
facility) should randomly select sufficient test article and
reference standard to conduct the study from the supplies
received from the SMO under contract, or from the packaging
facility under subcontract with the SMO, and retain the
remaining study samples as study reserves.
·
Each testing facility or the pharmacy of each
testing facility should retain the reserve samples.
·
Following the completion of the study, if one or
more of the testing facilities do not have adequate storage,
reserve samples can be transferred to an independent third party
with an adequate facility for storage under conditions
consistent with product labeling. The reserve samples should
not be transferred back to an SMO or any other organization that
deals with packaging the test articles and reference standard
for storage. This is to eliminate the possibility of
commingling reserve samples from packaging activities (21 CFR
211.84 and 211.170) and bioequivalence studies (21 CFR 320.38
and 320.63). As stated in subsection VI.A above, the reserve
samples should not be shipped back to the sponsor
or manufacturer.
C. Blinded Studies With
Pharmacodynamic or Clinical Endpoints Involving an SMO
Blinded BE studies are often conducted at
multiple sites and involve nonoral dosage forms with
pharmacodynamic or clinical endpoints. Often, the study sponsor
and/or drug manufacturer contracts with an SMO to recruit
clinical investigators and monitor the study. The SMO is
involved directly or indirectly (i.e., by subcontracting to
another party) in packaging and shipping study test articles and
reference standards to the testing facilities. The testing
facilities are usually the clinical study units of CROs,
universities, hospitals, or clinics run by physicians. In
multisite, blinded BE studies, the sponsor and/or drug
manufacturer needs to consider whether the study design will
allow for selection and retention of reserve samples in
accordance with 21 CFR 320.38 and 320.63 and the final rule. If
the study design is too complex to meet the regulatory
requirements for reserve samples, the study design may need to
be reconsidered.
The responsibility of the study sponsor
and/or drug manufacturer is to ship the test article and
reference standard to the SMO under contract, or to the
packaging facility under subcontract to the SMO.
The responsibilities of the SMO include:
·
Packaging, distributing, and shipping test article
and reference standard to all testing facilities (or
subcontracting a packaging facility to perform this function).
We recommend that the SMO provide the testing facilities with
enough code-labeled sets to conduct the study and to retain the
five times quantity. Based on inspection experience, DSI does
not recommend that test article and reference standard be
prenumbered for subjects, because assigning unit doses to a
designated subject number precludes the random selection of drug
used for dosing and drug used for reserve samples (see example
below for illustration).
·
Monitoring the study at different sites if it is
conducted under an IND (rarely needed for most ANDA studies)
Note: The SMO should not select reserve
samples. In addition, the reserve samples should not be
transferred by the testing facility back to an SMO or any other
organization that deals with packaging the test articles and
reference standard for storage.
The responsibilities of the testing
facilities are as follows:
·
The clinical investigator or designee (such as the
study coordinator or the research pharmacist) of each testing
facility should randomly select sufficient test article and
reference standard to conduct the study from the supplies
received from the SMO under contract, or from the packaging
facility under subcontract with the SMO, and retain the
remaining study samples as study reserves. The clinical
investigator should be aware of the sampling techniques used for
blinded studies as described in section III.
·
Each testing facility or the pharmacy of each
testing facility should retain the reserve samples. Please note
that if a placebo is used in blinded BE studies, reserve samples
for the placebo should be retained along with the test article
and reference standard reserves. The sealed treatment code of
the study should be kept at the testing facility. This is
applicable even if the reserve samples are forwarded to an
independent third party (see paragraph below).
3. If one or more of the testing facilities do not have
adequate storage, or go out of business, the reserve samples can
be forwarded to an independent third party with an adequate
facility for storage under conditions consistent with product
labeling.
Below is a suggested packaging and random
selection plan for a blinded, multisite study of a
dermatological cream product involving a SMO:
The study enrolls 300 subjects with
approximately 60 subjects at five testing facilities. The five
times quantity for the test article and reference standard is 50
tubes for each product. In preparation for conducting the
study, the SMO prepares 200 boxes that contain one code-labeled
tube of test article and one code-labeled tube of reference
standard in each box. The SMO randomly distributes 40 boxes to
each clinical testing facility. The clinical facility randomly
selects 30 of the boxes to dose 60 subjects. The remaining 10
boxes serve as the reserve samples. In this example, staff
(e.g., a pharmacist) not involved with the study may be
recommended to ensure the study remains blinded. This packaging
system ensures that an equal number of test article and
reference standard are administered to the subjects at each
site, and that an equal number of test article and reference
standard will be maintained as reserve samples. Since 10 boxes
are kept at each of 5 testing facilities, 50 tubes each of test
article and reference standard are retained and the five times
quantity reserve sample requirement is met. In addition, the
requirement of random selection by each testing facility is also
met.
Only about 7 percent of all sites inspected
by DSI from 1997 to 2002 conducted in-house BA and BE studies.
If a study sponsor and/or drug manufacturer conducts such a
study, manufacturing reserve samples (21 CFR 211.170) and BE
study reserve samples (21 CFR 320.38 and 320.63) should be
separated. The in-house clinical research unit should operate
as an independent unit for the purposes of sample retention.
All matters (e.g., manufacturing, purchasing, packaging,
transfer records) concerning the test article and reference
standard should be clearly documented and available to FDA
investigators during an inspection. Standard procedures
concerning security and accountability of the test article and
reference standard for each study should be established to
eliminate the possibility of sample substitution. Sponsors
conducting in-house studies can engage an independent third
party to store reserve samples. If an independent third party
is not used, there should be (1) a totally segregated and fully
compliant in-house storage area; (2) procedures and policies in
place to show that adequate test article and reference standard
are retained; (3) controlled access to the reserve samples; (4)
a rigorous and unbroken chain of custody for the reserve
samples.
The study sponsor and/or drug manufacturer
(clinical research department) should be responsible for
packaging and transferring the test article and reference
standard to the in-house clinical study unit.
The testing facility (in-house clinical
study unit) should be responsible for:
·
Documentation of all matters concerning the
transfer and receipt of the test article and reference standard
·
Random selection of sufficient test article and
reference standard to conduct the study, and retention of the
remaining study samples as reserves. The selection is generally
made by the clinical investigator, study coordinator, or
research pharmacist (if available) in the clinical study unit.
We recommend that a staff member (e.g., a study nurse) witness
the random selection process and dosing.
·
Retention of reserve samples in a secure area. To
ensure the authenticity of the reserve samples, access to this
area should be limited. We encourage maintenance of an entry
log to the storage area.
·
Preparation for adequate storage of reserve
samples. If the in-house testing facilities do not have
adequate storage, or go out of business, the reserve samples can
be forwarded to an independent third party with an adequate
facility for secure storage under conditions consistent with
product labeling.
21 CFR 320.63 states:
The applicant of an abbreviated
application or a supplemental application submitted under
section 505 of the Federal Food, Drug, and Cosmetic Act, or, if
bioequivalence testing was performed under contract, the
contract research organization shall retain reserve samples of
any test article and reference standard used in conducting an in
vivo or in vitro bioequivalence study required for approval of
the abbreviated application or supplemental application.
Thus, the regulations for reserve samples
apply to in vitro BE studies. The in vitro BE studies required
for approval of nasal aerosols and nasal sprays for local action
are an example of this. Note that in vitro studies conducted to
compare dissolution rates for different strengths of the same
formulation are not subject to the reserve sample regulations.
For an in vitro BE study, the roles of the study sponsor and/or
drug manufacturer and the testing facility are similar to those
described for in vivo BE studies conducted by CROs and in the
examples of in vivo BE studies conducted in-house by a study
sponsor and/or drug manufacturer.
As stated in 21 CFR 320.38(c), each reserve
sample shall consist of a sufficient quantity of samples to
permit FDA to perform five times all of the release tests
required in the application or supplemental application. Dose
content uniformity or spray content uniformity release tests
alone usually take 30 units (canisters or bottles) per batch.
Performance of other release tests can suggest a need for
additional units. The number of reserve sample units to be
retained for three batches of test article and reference
standard could exceed 1000 units (up to 250 units for each batch
of the test article and reference standard) based on the five
times quantity requirement. The Agency has determined that in
lieu of the “five times quantity” requirement, the quantity of
inhalant (nasal aerosol or nasal spray) test article and
reference standard retained for testing and analyses should be
at least 50 units for each batch (see the preamble to the final
rule).[8]
For NDAs, at least 50 units of each of the
three batches of nasal aerosol or nasal spray needed for BA
studies should be retained. However, where the reference
standard is another nasal aerosol or nasal spray, at least 50
units of that batch should also be retained. For ANDAs, at
least 50 units of each of three batches should be retained for
each of the test articles and reference standards used for in
vivo or in vitro BE studies. If multiple testing facilities are
used in a BA or BE study, the total amount of reserves for each
product across all testing facilities would be at
least 50 units, and each testing facility should retain a
reasonable amount of test articles and reference standards (see
section V for more details). For NDAs and ANDAs, if the in vivo
or in vitro studies include placebo aerosols or sprays, at least
50 units of each placebo batch should also be retained. These
recommendations apply only to nasal aerosol and nasal sprays for
local action that are to be marketed as multiple dose products,
typically labeled to deliver 30 or more actuations per canister
or bottle.