Additional copies of this guidance are available from the Office of Communication, Outreach and Development (OCOD), (HFM-40), 1401 Rockville Pike, Suite 200N, Rockville, MD 20852-1448, or by calling 1-800-835-4709 or 301-827-1800, or e-mail email@example.com , or from the Internet at http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/default.htm .
For questions on the content of this guidance, contact OCOD at the phone numbers or e-mail address listed above.
Guidance for IndustryPre-Storage Leukocyte Reduction of Whole Blood and Blood
Components Intended for Transfusion
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 the approach satisfies the requirements of the applicable statutes and regulations. If you want to discuss an alternative approach, contact the appropriate FDA staff. If you cannot identify the appropriate FDA staff, call the appropriate number listed on the title page of this guidance.
We, FDA, are issuing this guidance document to provide you, blood establishments, with recommendations for pre-storage leukocyte reduction of Whole Blood and blood components intended for transfusion, including recommendations for validation and quality control monitoring of the leukocyte reduction process. We also provide information to licensed blood establishments for submitting biologics license application supplements to include leukocytes reduced components. This guidance applies to Whole Blood, Red Blood Cells, Plasma, and Plateletsmanufactured from Whole Blood or collected by automated methods. This guidance document finalizes the draft guidance of the same title dated January 2011 and supersedes the FDA memorandum issued on May 29, 1996, entitled “Recommendations and Licensure Requirements for Leukocyte-Reduced Blood Products.”
We support the use of leukocytes reduced blood and blood components for specific indications (see section II.B) and seek to streamline the licensing procedure for leukocytes reduced blood components to assist blood establishments in making pre-storage leukocytes reduced blood components more widely available.
Although there have been reports of adverse events associated with leukocyte reduction by filtration (Refs. 1 and 2), advances in blood cell separation technology generally enable the safe reduction of leukocytes. We believe that increased availability of pre-storage leukocytes reduced blood components for non-targeted recipients will support the treatment of recipients who may benefit from receiving leukocytes reduced components but have not been identified as falling within the indications identified in section II.B. We suggest that consideration should be given to making leukocytes reduced blood components more widely available.
FDA’s guidance documents, including this guidance, 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. The use of the word should in FDA’s guidances means that something is suggested or recommended, but not required.
A. Changes from FDA Memorandum, “Recommendations and Licensure Requirements for Leukocyte-Reduced Blood Products,” May 29, 1996
In a May 29, 1996 memorandum, we issued recommendations on leukocyte reduction, a manufacturing step performed under controlled and monitored laboratory conditions. Under those recommendations, leukocytes reduced components were to contain < 5.0 x 106 residual white blood cells (WBCs) per each Whole Blood, Red Blood Cells or Platelets, Pheresis collection, and < 8.3 x 105 residual WBCs per each Platelets derived from Whole Blood, with at least an 85 percent recovery of the original component. The memorandum was consistent with the outcome of a public workshop held in March 1995 on leukocytes reduced products. This guidance document maintains those standards for residual WBCs and modifies the 1996 recommendations as follows:
B. Established Benefits of Pre-Storage Leukocyte Reduction
Leukocytes reduced Whole Blood and blood components have been shown to reduce the following:
We believe it is advantageous to provide leukocytes reduced components to reduce such adverse events. In addition, we believe it is advantageous to provide pre-storage leukocytes reduced components in preference to bedside filtered blood components based on quality and safety considerations (Ref. 14). Routine use of leukocytes reduced blood components further protects patients who have increased risk, but whose conditions have not yet been identified.
C. Potential Benefits of Pre-Storage Leukocyte Reduction
Potential, but not established, benefits of leukocyte reduction include reduction of:
Limited scientific data have suggested the possibility that leukocyte reduction may be a useful measure against the risk of transfusion-transmitted vCJD (Refs. 19 through 23). Some international blood authorities have implemented or are considering the implementation of a regulatory requirement to leukocyte reduce all Red Blood Cells and Platelets intended for transfusion (universal leukocyte reduction), as well as plasma for further manufacturing to reduce the risk of vCJD transmission.
Similarly, the potential to reduce infection by pathogens that are primarily WBC associated may extend beyond CMV (Refs. and ) to include other known and unknown agents potentially transmitted by WBCs (Ref. 19).
Leukocyte reduction is not considered appropriate for the prevention of transfusion-associated graft-versus-host disease. At this time, irradiation of blood products is the only definitive method available to prevent this serious and often fatal transfusion outcome in patients at risk (Ref. 24).
D. Safety Concerns Related to Bedside Leukocyte Reduction Filtration
Bedside filtration remains available as a leukocyte reduction method to physicians prescribing transfusion therapy. As a post-storage procedure, however, bedside filtration has been associated with precipitous hypotension in the transfusion recipient, an infrequent yet serious adverse effect not associated with pre-storage leukocyte reduction (Ref. 25). Patients on medications that inhibit the angiotensin converting enzyme (ACE inhibitors) appear to be particularly susceptible. Pre-storage leukocyte reduction allows the leukocyte reduction process to be monitored under controlled conditions that assure component purity, consistency, and safety and is therefore generally preferable to bedside filtration (Refs. 25 through 27).
For the purposes of the terms used in this guidance, the following definitions apply:
Automated Blood Cell Separator (ABCS): A device that uses a centrifugal or filtration separation principle to automatically withdraw Whole Blood from a donor, separate the Whole Blood into blood components, collect one or more of the blood components and return to the donor the remainder of the Whole Blood and blood components. The ABCS device is intended for routine collection of blood and blood components for transfusion or further manufacturing use.
Non-process failure: Failure of a leukocyte reduction process due to a non-controllable parameter (i.e., donor specific characteristic such as HbS). Note: In your sampling plan for validation or quality control (QC) testing, you may exclude (and replace) non-process failures.
Percent recovery (of the original component):Ratio of the post-filtration to the pre-filtration content of the component expressed as a percent.
Process failure:Failure of a leukocyte reduction process due to an error that was avoidable if proper procedures, precautions and/or instructions were followed (i.e., not following manufacturer’s directions) or a product/device defect.
Process validation: In brief, is the collection and evaluation of data from the process design stage through production. This validation establishes scientific evidence that a process is capable of consistently delivering quality product (Ref. 28).
Residual White Blood Cell (WBC) content: The number of WBCs remaining in a leukocytes reduced component, calculated by multiplying the WBC count from a sample of the component times the volume of the component.
Yield: The quantity that is actually produced during manufacturing.
The Current Good Manufacturing Practices (cGMP) regulations described in 21 CFR Parts 210 and 211 contain the minimum requirements for methods to be used in, and the facilities or controls to be used for, the manufacture, processing, packing or holding of a drug to assure that the drug meets the safety requirements of the Federal Food, Drug, and Cosmetic Act, and has the identity and strength and meets the quality and purity characteristics that it purports or is represented to possess (21 CFR 210.1(a)). These cGMP regulations apply to Whole Blood and blood components (21 CFR 210.2(a) and 211.1(b)) and supplement the cGMP regulations for blood and blood components contained in 21 CFR Part 606. As an element of cGMP, process validation establishes “scientific evidence that a process is capable of consistently delivering quality product”(Ref. 28). We recommend that establishing documentation of process validation include, but not be limited to, equipment installation qualification, validation protocol development, process operator performance qualification and product performance qualification (Ref. 28).
A. Methods of Leukocyte Reduction
Whole Blood and blood components (including components collected by automated methods) intended for transfusion may be leukocyte reduced using any of the closed system or functionally closed methods as follows:
B. Equipment Installation Qualification
Section 606.60(a) (21 CFR 606.60(a)) requires that equipment be observed, standardized and calibrated on a regularly scheduled basis as prescribed in the Standard Operating Procedures Manual and must perform in the manner for which it was designed. Upon initiation of an automated leukocyte reduction method, we recommend that the equipment used for leukocyte reduction (e.g., leukocyte reduction filter, ABCS device) be qualified as described in the operator’s manual or manufacturer’s directions for use.
C. Validation Protocol Development
An integral element of the performance and documentation of process validation is the development of a validation protocol. You should refer to FDA’s “Guidance for Industry: Process Validation: General Principles and Practices” (Ref. 28) as an outline for developing your validation protocol. We recommend that the validation protocol include, but may not be limited to:
D. Process Performance Qualification (Operator)
Each person engaged in the manufacture of leukocytes reduced components must have adequate education, training, or experience to assure their competent use of the devices involved (21 CFR 606.20(b)).
We recommend that personnel training include the successful consecutive performance, under supervision, of an appropriate number of procedures, as defined by your facility. These procedures should result in leukocytes reduced components meeting the specifications for the specific leukocyte reduction methodology being qualified.
E. Product Performance Qualification for the Leukocyte Reduction Process
Various factors may adversely influence the leukocyte reduction process, e.g., donor specific factors or improper use of an ABCS device. The objective of product performance qualification is to verify that the leukocyte reduction method conforms to the claims of the device manufacturer and the blood establishment. In addition, appropriate testing will establish confidence that the final component meets all release requirements for safety, purity, and potency (Refs. 28 through 31). All components collected during the qualification process can be released for transfusion provided that they meet minimum specifications as defined by the device manufacturer, are labeled appropriately, and are otherwise suitable.
We believe that, generally, product performance qualification only needs to be performed when the leukocyte reduction method (either by ABCS devices or filtration) is first put into use at an establishment. We note the following:
Product performance qualification should include testing for the residual WBC count and percent recovery of the original component as recommended in section III.C, and testing of components processed by all trained operators. In addition, we recommend you perform residual WBC count testing within 48 hours of collection (Ref. 33) or per the manufacturer’s directions for the cell counting methodology.
You must conduct an investigation of each product performance qualification failure, and when appropriate, initiate corrective action and follow-up measures (21 CFR 211.192 and 606.100(c)). There should be ongoing assurance that the process continues to perform at the level established during the product performance qualification. Evaluating the performance of the process identifies problems and determines whether action must be taken to correct, anticipate, and prevent problems (21 CFR 211.180(e), 211.192, and 606.100(c)). We understand that some failures may occur due to conditions not resulting from a failure of the process (e.g., failure due to donor specific characteristic such as HbS).
F. Leukocyte Reduction Performance Qualification Criteria
Conformance to product standards must be assessed by a statistically valid method (see 21 CFR 211.160(b)). In the absence of a validation method (plan) provided by the manufacturer, you should develop a statistically valid plan based on 95% confidence that more than 95% of the components will meet the recommended results. One possible approach to process validation is a two-stage sampling plan based upon the binomial distribution. In this approach, an initial representative sample of components with a pre-determined sample size should be tested consecutively (validation) or randomly (QC monitoring). Identified non-process failures may be excluded from the sample size and replaced. Conformance can be demonstrated if an acceptable number of process failures are observed in this sample. In the event that there is one more process failure than permitted in the initial sample, a second sample of components may be tested consecutively. Conformance can be demonstrated if no further process failures are observed in the second sample. The size of the second sample will depend on the size of the first sample. We provide two examples to illustrate the use of this approach:
Note that the sample sizes are pre-determined and fixed based on the required statistical criteria. For example, if you plan to test 94 samples and encounter no failure in the first 60 samples, changing your plan to 60 samples is not permitted.
In some cases, you may decide on a one-stage sampling strategy rather than the two-stage strategy described above. Based on the binomial distribution, conformance can be demonstrated by observing zero process failures in a fixed sample of 60 components, one or fewer failures in a fixed sample of 94 components, or two or fewer failures in a fixed sample of 124 components.
We recommend that the following parameters as listed in Table 1 should be assessed to assure an overall statistically-based level of product conformance:
Plasma components, when manufactured under cGMP, may inherently have a residual WBC content of < 5.0 x 106 per unit. Currently accepted and validated cell counting procedures may need to be modified for plasma components. Plasma components (including Fresh Frozen Plasma and Cryoprecipitated Antihemophilic Factor) may be labeled “Leukocytes-Reduced” provided that manufacturing process validation as described in this section (section III) includes these components and the ABCS device is cleared to manufacture the product.
G. Investigation of Product Performance Qualification Failure
Section 211.192 (21 CFR 211.192) requires that the failure of a batch or any of its components to meet any of its specifications be thoroughly investigated. If a post-filtration residual WBC count is > 5 x 106 (> 8.3 x 105 for unpooled Platelets) and/or fails to meet the percent recovery of the original component during validation, you must conduct an investigation of the product performance qualification failure (21 CFR 211.192 and 606.100(c)).
You must investigate collections that fail to meet the percent Red Blood Cell recovery or Platelet recovery criteria (21 CFR 211.192 and 606.100(c)). However, the component may be released if the actual platelet yield or Red Blood Cell volume is determined to be suitable by the quality control unit and the component is labeled appropriately.
Product performance qualification failures may occur due to conditions not resulting from a failure of the process. For example, if the observed failure is investigated and found to be due to a donor-specific factor, it does not constitute a process failure and should be excluded from the performance qualification evaluation.
When a failure during product performance qualification is investigated and found to be due to an identified donor-specific factor, we encourage you to flag the donor record. Upon a subsequent occurrence of incomplete filtration or inadequate WBC removal, we encourage you to consider not using this donor for future donations of leukocytes reduced components.
Quality assurance (QA) is the sum of activities planned and performed to provide confidence that all systems and system elements that influence the quality of the component are functioning as expected (Ref. 30). When this is demonstrated, the process is considered to be in a state of control. The determination of whether a process is operating in a state of control is made by analyzing the day-to-day process, performance variability and the data for conformance with the manufacturer’s specifications. Please refer to FDA’s “Guideline for Quality Assurance in Blood Establishments” (Ref. 30) for assistance in developing a QA and monitoring program. This program is distinct from the initial validation described in section III.
You must have a QC unit that has the responsibility and authority to approve or reject all components, containers, closures, in-process materials, packing materials, labeling and drug products and the authority to review production records to assure that no errors have occurred or, if errors have occurred, that they have been fully investigated (21 CFR 211.22(a)). Thus, the QC unit’s responsibilities include the review of production records, and the review of complaints involving the possible failure of a product to meet its specifications (See, for example,
A. Component Testing
A QA program should include in-process monitoring of the manufacturing procedures and QC testing (Ref. 16).
Under 21 CFR 211.160(b), laboratory controls must include the establishment of scientifically sound and appropriate specifications, standards, sampling plans and test procedures. We will consider statistical plans that confirm a < 5% non-conformance rate with 95% confidence.
One example of a scientifically sound statistical sampling and analytic plan is based on a two-stage binomial approach. The sampling sizes described in section III.F will confirm with 95% confidence a < 5% non-conformance rate for residual WBC counts. Another example of a scientifically sound plan is based on the hypergeometric distribution (see Appendix). The hypergeometric plan is only for quality control sampling plans and cannot be used for validation. Other statistical plans may also be appropriate, such as the use of scan statistics (Ref. 32) or a lower confidence limit (e.g., for platelet yields we will consider statistical plans that confirm a < 5% non-conformance rate with 75% confidence).
As part of your QC procedures, we recommend that you:
B. Equipment/Ancillary Supplies
Equipment must be observed, standardized, and calibrated on a regularly scheduled basis as prescribed in the Standard Operating Procedures Manual (21 CFR 606.60(a)). Such equipment includes, but may not be limited to the ABCS, cell counting instrument(s), scales, blood shaker and STCD.
Supplies and reagents must be used in a manner consistent with instructions provided by the manufacturer (21 CFR 606.65(e)). In addition, all supplies (including filters) and reagents must meet all of the requirements described in 21 CFR 606.65.
C. Operator Training
Operators must have adequate training, education, experience, or combination thereof, to assure competent performance of their assigned functions (21 CFR 606.20(b)). Control over operational variables should be maintained through initial and continued staff training, routine staff participation in leukocyte reduction and periodic assessment of operator performance. We recommend that assessment of operators include scheduled competency assessment and proficiency testing. In addition, we recommend that you develop appropriate training on leukocyte reduction processes and procedures and/or equipment maintenance as updated information becomes available.
D. Standard Operating Procedures
A circular of information must be available for distribution if the product is intended for transfusion (21 CFR 606.122).
Your container labels must comply with 21 CFR 606.121 and 610.60. The container label must include the proper name of the product, in a prominent position, with any appropriate modifier(s) and attributes (21 CFR 606.121(c)(1)). One way to comply with the applicable labeling requirements is through of use of standard terminology. For example, Table 3 lists the proper names of the major leukocytes reduced blood components, followed by the corresponding International Society of Blood Transfusion (ISBT) Code 128 name (Ref. 39). The component labeling should include the phrase “Leukocytes Reduced.” The phrases “Leukocytes Removed,” “Leukocyte Poor,” “Leukocytes Depleted,” and other similar terms should not be used in component labeling.
Only those components meeting the recommended residual WBC content for leukocytes reduced blood components or prepared by a validated method known to meet the recommended residual leukocyte content for leukocytes reduced blood components may be labeled as “Leukocytes Reduced.”
Those components (except Whole Blood derived Platelets) that have been tested and found to have a residual WBC count of < 1.0 x 106 may be labeled either with the actual residual leukocyte content via a supplemental label or a tie-tag stating: “The residual white blood cell count of this component has been determined to be < 1.0 x 106.
Table 3: Component Names for Container Labels
An establishment that distributes leukocytes reduced blood components in interstate commerce must have an approved BLA, including an approval for leukocytes reduced blood components, in accordance with section 351 of the Public Health Service Act.Licensed establishments must report changes to their approved application(s) in accordance with 21 CFR 601.12. For assistance in reporting your changes see FDA’s “Guidance for Industry: Changes to an Approved Application: Biological Products: Human Blood and Blood Components Intended for Transfusion or for Further Manufacture” dated July 2001. The information below is intended to assist you in determining which reporting mechanism is appropriate for a change to your approved BLA, as it applies to the manufacture of leukocytes
Under 21 CFR 601.12(b), changes that have a substantial potential to have an adverse effect on the identity, strength, quality, purity, or potency of the product as they may relate to the safety or effectiveness of the product must be reported to FDA in a Prior Approval Supplement (PAS).Under this regulation, the following kinds of manufacturing changes would fall within this category, warranting submission of your request to implement the following changes to your approved BLA as a PAS:
A Comparability Protocol is appropriate, but not required, if you wish to implement this change at multiple collection facilities under your direction and control, using the same process to manufacture automated leukocytes reduced components prepared by apheresis. We consider the recommendations in this guidance document to provide appropriate criteria for a biologics license supplement for leukocytes reduced blood or blood components. You may use an alternate approach if such approach satisfies the requirements of the applicable statutes and regulations. Your alternative procedure(s) may be acceptable if you demonstrate that the resulting leukocytes reduced blood or blood components are equivalent to components manufactured according to the procedures described in this guidance. You must not distribute in interstate commerce blood components made using a changed manufacturing process requiring a PAS until you have received our approval of your supplement (21 CFR 601.12(b)(3)).
B. Changes Being Effected in 30 Days (CBE-30) Supplement: Changes Requiring Supplement Submission at Least 30 Days Prior to Distribution of the Product Made Using the Change (21 CFR 601.12(c))
Under 21 CFR 601.12(c), changes that have a moderate potential to have an adverse effect on the identity, strength, quality, purity, or potency of the product as they may relate to the safety or effectiveness of the product must be reported to FDA in a Changes Being Effected in 30 days (CBE-30) supplement.You must submit your request to implement manufacturing changes with a moderate potential for an adverse effect to your approved BLA as a CBE-30 supplement under
You may distribute your blood components made using the change requested in your CBE-30 supplement in interstate commerce 30 days after we receive your supplement, unless we notify you otherwise (21 CFR 601.12(c)(4)).
C. Changes to be Described in an Annual Report (Minor Changes) (21 CFR 601.12(d))
Under 21 CFR 601.12(d), changes in the product, production process, quality controls, equipment, facilities, or responsible personnel that have a minimal potential to have an adverse effect on the identity, strength, quality, purity, or potency of the product as they may relate to the safety or effectiveness of the product must be documented in an annual report submitted each year within 60 days of the anniversary date of approval of the application. Changes from one type of FDA approved or cleared leukocyte reduction filter to another type of FDA approved or cleared leukocyte reduction filter is an example of a manufacturing change that we believe falls within this category, if implemented according to the instructions in the package insert without modification.
D. Submission of Documents
Additionally, we recommend two months of QC data for component volume and residual WBC count per component.
In addition, you should include the submission tracking number (STN) of the approved Comparability Protocol, and the STN(s) of changes to the SOPs associated with an approved Comparability Protocol.
Questions about licensing and specific review questions about leukocyte reduction should be directed to the Office of Blood Research and Review, Division of Blood Applications (DBA). Submit all registration forms (Form FDA 2830) and licensure applications/supplements to the Director, Center for Biologics Evaluation and Research (CBER). Table 4 presents FDA contact information regarding leukocyte reduction.Table 4: FDA Contact Information
1. Alvarado-Ramy, F., Kuehnert, M., Alonso-Echanove, J., Sledge, L., Haley, R., Epstein, J., Vostal, J., and Pearson, M. A multistate cluster of red blood cell transfusion reactions associated with use of a leukocyte reduction filter. Transfusion Medicine 2006; 16:41-48.
QC testing and sample size calculations based on the hypergeometric distributionSection III.F of this document describes a two-stage sampling approach to process validation based on the binomial distribution. As discussed in Section IV.A, we consider this approach to be a scientifically sound statistical sampling and analytic plan for QC testing as well as for process validation. One limitation of the binomial distribution-based approach, however, is that it assumes that the population of components being sampled from is infinite. This assumption may not be ideal for sampling performed for monthly QC testing.A two-stage sampling approach based on the hypergeometric distribution may be used for monthly QC. Unlike the binomial approach, the hypergeometric approach assumes that a finite number of components will be manufactured during a monthly QC period. In general, the sample size for QC testing under the hypergeometric approach will be smaller than that for binomial distribution-based sampling. However, you may prefer to use binomial distribution-based sampling for other reasons, including familiarity and simplicity. The decrease in sample size for the hypergeometric approach relative to the binomial approach is most noticeable when the total number of collections to be performed during a given QC period is small.The two-stage hypergeometric approach is similar in concept to the two-stage binomial approach described in Section III.F. In the first stage, an initial representative sample of components with a predetermined sample size should be tested consecutively. Identified non-process failures may be excluded from the sample size and replaced. Conformance can be demonstrated if an acceptable number of process failures are observed in this sample. In the event that there is one more process failure than permitted in the initial sample, a second sample of components may be tested consecutively. Conformance can be demonstrated if no further process failures are observed in the second sample. The size of the second sample will depend on the size of the first sample, and may encompass the entire remaining population of components to be tested during a given QC period. Table A details possible sampling plans and sample sizes for two-stage QC testing based on the hypergeometric distribution for components other than platelets. Table B provides sampling plans and sample sizes for platelet yield QC testing. Each of the sampling plans contained in the tables satisfies the statistical criteria for successful QC testing provided in Section IV.A. The tables may be used as follows:
As an example of this approach, suppose that you intend to collect 100 leukocyte-reduced plasma components on a given instrument at a given site during the next month. Note first that you may test every one of these 100 units and, if 4 or fewer process failures are observed, the process failure rate is less than 5% for that QC period. Adopting instead a statistical sampling strategy, you may consult Table A and choose one of the three sampling plans provided in the row corresponding to population size of 100. Suppose you choose the “zero failures allowed” sampling plan in this row, which specifies a first stage sample size of 45 components and a second stage sample size of 34 components. You would therefore test the first 45 consecutive components for that month. If no process failures were observed among these 45 components, the process would be considered to be in control and no further QC testing would be required during that month. If one process failure is observed among the first 45 components, you would test an additional 34 consecutive components. If no further process failures were observed among the second stage sample of 34 components, the process would be considered to be in control and no further testing would be required. If two or more process failures were observed during the first stage sample of 45 components or if one or more process failures were observed during the second stage sample of 34, the process would not be in control and QC testing would have failed.Statistical derivation of the sample sizes in Tables A and BIn Table A, for each fixed population size, N, and number of allowed failures, m,the first stage sample size was calculated as the minimum number n1 such the probability of observing m or fewer process failures in a sample of size n1 from a population of size N is at most 5% when the true process failure rate is 5% or more. For example, assume that 100 components will be collected during a QC period. Using the “zero failures allowed” first stage sampling plan, it can be calculated based on the hypergeometric probability distribution that the probability of observing zero failures in a sample of 45 components is 0.0462 if the true failure rate is at least 5% (that is, if there are at least 5 process failures among the 100 components to be collected). On the other hand, the probability of observing zero failures in a sample of 44 components is 0.0507. Therefore, 45 is the minimum sample size such that the probability of observing zero failures is less than 5% when the true failure rate of at least 5%.Once the first stage sample size has been determined, to obtain the second stage sample size, we first calculated 0.05 minus the probability of observing m or fewer process failures in the first stage when the true failure rate is at least 5%. Continuing our example above, this quantity would be 0.05 – 0.0462 = 0.0038. The second stage sample size was then calculated as the minimum number n2 such that the probability of observing (i) one failure among the first n1 samples at stage one and (ii) no failures among the n2 samples at stage two is less than 0.05 minus the probability of observing m or fewer process failures in the first stage when the true failure rate is at least 5%. In the example here, with n2 = 33, this probability is 0.0044, which is not less than 0.0038, whereas with n2 = 34, this probability is 0.0036 which is less than 0.0038. Therefore, the required sample size at stage two for this example is 34. Romeu (Ref. 41) adopts the same algorithm for an infinite population size based on the binomial distribution.The derivation of the sample sizes in Table B followed the same process, but was based on ruling out a true failure rate of at least 25% rather than 5% as in Table A. Note that hypergeometric sampling schemes other than those presented in Tables A and B are possible. For instance, in some cases it is possible to use a larger first stage sample size and smaller second stage than that provided in the tables. We will consider any sampling plan that meets the criteria described in Section IV.A.
Table A. Sampling scheme for quality control testing based on hypergeometric distribution (95%/95%).
1The inference that with 95% confidence more than 95% of the components meet the standard is only applicable to the finite population from which the samples are drawn.
Table B. Sampling scheme for quality control testing based on hypergeometric distribution (95%/75%).
1The inference that with 95% confidence more than 75% of the components meet the standard is only applicable to the finite population from which the samples are drawn.
The guidance document entitled “Guidance for Industry and FDA Review Staff: Collection of Platelets by Automated Methods” dated December 2007 contains our recommendations for validating; quality assurance and monitoring; labeling and licensure for leukocyte reduction of Platelet, Pheresis products. http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/Blood/ucm073382.htm .
The guidance document entitled “Guidance for Industry: Recommendations for Collecting Red Blood Cells by Automated Apheresis Methods” Technical Correction dated February 2001. http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guidances/Blood/ucm076756.htm .
The requirement for process control is set forth in general terms in 21 CFR 211.100.
See footnote 1.
See footnote 1.