We recommend you address the following areas when preparing a protocol to modify the indication for use to include testing of cadaveric blood specimens.
- What data about specificity and sensitivity are recommended when matched pairs of pre- and post-mortem serum/plasma specimens are available?
- Specificity
We recommend that you test at least 50 paired specimens (1 pre- and 1 post-mortem specimen from the same donor).
- Clinical Specificity
Clinical specificity is a measure of how often the test is negative in non-diseased donors.
We recommend that you determine if a statistically significant difference exists between pre- and post-mortem specimens based on frequency of false positive results.
- Analytical Specificity
Analytical specificity measures a test's ability to exclusively identify a target substance rather than different substances.
We recommend that you determine if a statistically significant difference exists between pre- and post-mortem specimens based on signal strength.
- Sensitivity
We recommend that you test at least 50 paired reactive specimens (1 pre- and 1 post-mortem specimen from the same donor).
- Clinical Sensitivity
Clinical sensitivity is a measure of how often the test is positive in diseased donors.
We recommend that you determine if a statistically significant difference exists between pre- and post-mortem specimens based on frequency of false negative results.
- Analytical Sensitivity
Analytical sensitivity measures a test's ability to detect a low concentration of a given substance.
We recommend that you determine if a statistically significant difference exists between pre- and post-mortem specimens based on signal strength and endpoint dilutions of positive specimens.
- What data about specificity and sensitivity are recommended when matched pairs of pre- and post-mortem serum/plasma specimens are not available?
- Specificity
We recommend that you concurrently test at least 50 cadaveric (post-mortem) specimens from 50 different cadaveric donors and an equal number of random living donor specimens (unmatched pre-mortem specimen) with the same test kit lots.
- Clinical Specificity
We recommend that you determine if a statistically significant difference exists between the cadaveric specimens and the random living donor specimens based on the frequency of false positive results, i.e., the number of pre-mortem nonreactives/post mortem reactives.
- Analytical Specificity
We recommend that you determine if a statistically significant difference exists between the cadaveric specimens and the random living donor specimens based on signal strength.
- Analytical Sensitivity
We recommend that you concurrently test at least 50 nonreactive cadaveric specimens from 50 different cadaveric donors with an equal number of random living donor specimens with the same kit lots, with both types of specimens spiked with the infectious disease marker at a potency near the assay's cutoff. You should use a minimum of 5 individual positive sources for the spiking experiment.
We recommend that you determine if a statistically significant difference exists between the spiked living donor specimens and the spiked cadaveric specimens based on signal strength.
- What is an example of a recommended reproducibility study?
We recommend that you conduct a reproducibility study to determine if a statistically significant difference exists between the coefficients of variations of cadaveric specimens compared to those of living donors. One possible experimental design might include comparing at least 20 random cadaveric specimens with at least 20 random living donor specimens (confirmed true positives may be excluded) spiked to be reactive near the cutoff. Test each specimen individually, in 6 separate runs on 6 separate days using each of 3 different kit lots (18 data points per specimen). We also recommend that the specimens to be tested on 6 separate days be stored at 4° Centigrade to avoid repeated freezing and thawing.
- How many test kit lots are recommended to be included in the studies?
We recommend that you include at least three test kit lots in all studies.
- What plasma dilution issues are recommended for consideration?
Prior to including a cadaveric blood specimen in these studies, we recommend that you determine whether the specimen has been appropriately evaluated for plasma dilution. You can obtain information about plasma dilution from FDA's "Guidance for Industry: Screening and Testing of Donors of Human Tissue Intended for Transplantation," dated July 1997 (the July 1997 guidance). This document can be found on the Internet at www.fda.gov/cber/tissue/docs.htm. As stated in the July 1997 guidance, plasma dilution is due to the transfusion or infusion of fluids into the donor prior to specimen collection, and can result in false negative test results. In an adult donor, if blood loss is known or suspected to have occurred and there was transfusion/infusion of more than 2000 mL of blood or colloids within 48 hours, or more than 2000 mL of crystalloids within 1 hour, or any combination thereof, prior to the collection of the blood specimen, plasma dilution may have occurred. In this case, we recommend that you use a specimen taken from the donor prior to transfusion or infusion, or an algorithm designed to evaluate volumes administered in the 48 hours prior to specimen collection, to ensure that plasma dilution sufficient to affect test results has not occurred. An example of an algorithm can be found in the July 1997 guidance.
- What information about specimen collection times does FDA recommend I note?
- We recommend that you note the time between death and specimen collection. In order to accurately document test kit performance, we recommend that the time at which cadaveric specimens are taken incorporate the full range of time points typically encountered during tissue recovery, e.g., 0-24 hours.
- We recommend that you include hemolyzed specimens in the study, since a large percentage of cadaveric specimens are hemolyzed due to biological processes that occur immediately post-mortem. You should quantify the degree of hemolysis, if possible, of the cadaveric specimens.
- We recommend that you note information about storage and handling conditions of both living donor specimens and cadaveric specimens.
- Where do I submit my data?
We intend to review any applications for cadaveric blood specimens jointly in the Offices of Cellular, Tissue and Gene Therapies and Blood Research and Review.
- If you seek an indication for use of cadaveric blood specimens and blood donor specimens, you may submit an Investigational New Drug Application (IND), or a Biologics License Application (BLA), as appropriate, with data for both intended uses, to:
Center for Biologics Evaluation and Research
Attn: Office of Blood Research and Review
HFM-99, Suite 200N
1401 Rockville Pike
Rockville, MD 20852-1448
- If an IND or BLA has already been submitted for blood donor screening, and you seek to modify the indication for use to include testing of cadaveric blood specimens, you may submit an amendment to the IND, or a supplement to the BLA with cadaveric blood specimen data, as appropriate, to the same address as above.
- If you seek to modify the indication for use to include testing of cadaveric blood specimens only, you may submit an IND or BLA with cadaveric blood specimen data, to:
Center for Biologics Evaluation and Research
Attn: Office of Cellular, Tissue and Gene Therapies
HFM-99, Suite 200N
1401 Rockville Pike
Rockville, MD 20852-1448
If you have questions about this guidance, please contact Melissa Greenwald in the Office of Cellular, Tissue and Gene Therapies at 301-827-2002.