• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Vaccines, Blood & Biologics

  • Print
  • Share
  • E-mail

Section Contents Menu

Vaccines

Review of pertussis assay validation for STN: 125300, March 6, 2009 - Menveo

MEMORANDUM  DEPARTMENT OF HEALTH AND HUMAN SERVICES
  Public Health Service
   
  Food and Drug Administration
Center for Biologics Evaluation and Research
Date: March 6, 2009 
From: Drusilla Burns, DBPAP, OVRR, CBER
Subject: Review of pertussis assay validation for STN: 125300, Novartis Meningococcal ACWY Conjugate Vaccine.
To: File
  Through: Milan Blake, Director, DBPAP, OVRR, CBER

FOCUS OF REVIEW

Novartis Vaccines and Diagnostics, Inc. submitted a biologics license application for their Menigococcal ACWY Conjugate Vaccine.  This vaccine is a quadravalent conjugate meningococcal vaccine representing serogroups A, C, W, and Y and is hereafter referred to as MenACWY.  This vaccine uses pre-sized oligosaccharides from each of the primary pathogenic serogroups (A, C, W, and Y) conjugated to the CRM197 protein carrier.  The following indication has been requested for the vaccine:  for use in individuals 11 through 55 years of age to prevent invasive meningococcal disease caused by Neisseria meningitides serogroups A,D, W-135 and Y.

The focus of this review is the pertussis assays used to measure pertussis immunogenicity in studies to assess the effect of concomitant vaccination with both MenACWY and Tetanus Toxoid, Reduced Diphtheria Toxoid and Acellular Pertussis Vaccine Adsorbed (Tdap) on the immunogenicity of individual vaccine components.

EXECUTIVE SUMMARY OF REVIEWER’S CONCLUSIONS

Based on my review, I consider the pertussis ELISA assays performed at b(4) --------------------------- for Study V59P18 to be sufficiently validated for their intended purposes.  I do not consider the pertussis ELISAs performed at the ---b(4)------------------------------- for Study V59P11 to be adequately validated and thus the pertussis ELISA data from V59P11 cannot be interpreted.

REVIEW of 125300.0:

The original BLA submission contained assay validation studies for both --b(4)---------------------------- which performed the pertussis assays for V59P18 and the --b(4)------------------------ which performed the pertussis assays for V59P11.

Assay validation from ---b(4)--------------------------:

The original validation of the pertussis ELISAs at --b(4)------------------- was carried out from 1994-1998.  In 2003-2004, --b(4)------------ added experiments that addressed run length.  A description of these validation studies and their results was submitted in the original BLA submission (125300.0).  Assay attributes that were addressed included repeatability, intermediate precision, accuracy, linearity, parallelism, specificity, impact of freeze thaw cycles, and analytical run-length.  Insufficient information and data were provided to support the conclusion that the assays were validated for their intended purpose.

Assay validation from the --b(4)----------------------------------

In the original BLA submission (125300.0), Novartis also submitted validation data for the pertussis assays performed at the --b(4)------------------------------.  However, this information was very incomplete and insufficient to support the conclusion that the assays were validated for their intended purpose.

Based on my review of the assay validation information provided for --b(4)-------------- and the --b(4)------------------, I had the following questions and concerns.

Deficiencies identified:

In regards to the Pertussis ELISAs:

1.  At this time, the validation reports for the pertussis ELISAs conducted at --b(4)----------------------- and the --b(4)--------------------------------- are missing critical information such that the validity and soundness of the data generated using these assays cannot be adequately evaluated.  A considerable amount of addition information and data are needed to demonstrate that the assays are appropriately validated.  Examples of critical information and data that are missing from the reports are provided below.

  1. a.  a detailed description of the methods for each of the three pertussis assays including system suitability and acceptance criteria
  2. b.  a description of the methods used to calculate ELISA units/ml of test samples and representative sample calculations
  3. c.  a description of the critical reagents used in the assay and a summary of the data generated to qualify the batches used
  4. d.  for each coating antigen, the source and a summary of the testing done to ensure purity
  5. e.  a definition of the working range of the assay and data demonstrating that the assay has acceptable precision and accuracy over the entire working range
  6. f.  data supporting the precision and accuracy of the assay for samples at or near the lower limit of quantitation
  7. g.  data demonstrating that critical assay parameters did not change from the time of assay validation to the time the clinical assays reported in this submission were conducted
  8. h.  data demonstrating acceptable performance of the assay over the time period that the clinical samples were analyzed

2.  The validation reports did not indicate pre-defined acceptance criteria.  Please comment.

In regards to the -b(4)- pertussis assays:

3.  The dilutional linearity studies for each of the pertussis antigen ELISAs shows considerable bias as samples are diluted.  Such a bias might affect interpretation of the data generated by these assays.  Please comment.

In regards to the assays performed by the --b(4)-------------------------:

4.  Please provide a detailed description of the dilutional linearity study described including a description of the samples and diluent used.  Please include a more detailed description of the labels for the x- and y-axes (currently noted as “concentration” and “reading”, respectively).

5.  Please provide more detail regarding how the lower limit of quantification for each assay was determined.  You indicate limits of quantification for each of the three assays, however the units are expressed only as “concentration”.  Please specifically define the units of these limits of quantification.

6.  You indicate that prior to September 2007, the lower limits of quantification for PT, FHA, and PRN --b(4)--------------, respectively.  These limits were based on the lowest titer obtained in each assay with a Coefficient of Variation less than or equal to 0.15.  Please provide a rationale for changing these limits of quantification.

7.  The test for specificity of the assays is inadequate.  ---b(4)--------------------------------------------------------- --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------.  No further investigation was conducted to determine whether this result was due to cross-reactivity as suggested or rather it was due to lack of specificity of the assay.  

In regards to V59P11:

8.  What value was assigned to samples with titers below the assay limit of quantification for purposes of calculating GMC and percent responders?

SPONSOR RESPONSES:

In submissions dated December 19, 2008 (125300.3) and January 15, 2009 (125300.4), Novartis submitted additional data and information to address the above concerns pertaining specifically to --b(4)-------.  After reviewing these submissions, I have determined that they have adequately responded to my concerns.

In response to comment 1a, they submitted adequate descriptions of the assay methods including system suitability and acceptance criteria.

In response to comment 1b, they submitted an adequate description of the calculation methods.

In response to comment 1c and 1d, they submitted a description of the critical reagents used in the assay and a summary of the data generated to qualify the batches used.  They provided the source for each coating antigen.  The source is known by CBER to produce antigen of sufficient quality for the intended purpose of the assays. 

In response to comments 1e and 1f, they defined the working range of the assay and demonstrated that the assays have acceptable precision and accuracy, with one exception, over their entire working range.  The one exception was that the accuracy of the pertactin assay was not within pre-specified criteria.  They found that the fold-differences between the measured and nominal standard concentration at ----b(4)------------------------------------------------------------.  They judged these to be acceptable, given the assays intended purpose. Nonetheless, given these data, they indicated in their submission of January 15, 2009 (125300.4) that they intended to raise the lower limit of quantitation of the assay from --b(4)-----------------------, with results below -b(4)---- being reported as such.  Because of the change in the lower limit of quantitation for the pertactin ELISA, they indicated that they would re-run the statistical analysis for V59P18 taking this change into account and submit the amended data in the final version of the clinical study report. I concur with their decision to raise the lower limit of quantitation of the PRN assay to -b(4)---- and prepared the following to be conveyed to Novartis.

In your submission of January 15, 2009, you provided results generated by -b(4)-------------------------- to re-assess the accuracy over the entire working range of the pertussis ELISAs.  The pre-specified acceptance criteria for accuracy—that the fold-difference between the measured and nominal concentration be within -b(4)----—was satisfied for the PT and FHA assays over their entire working range.  The pre-specified accuracy for the PRN assay was met at nominal values of --b(4)-----; however, the ratio of the measured value to nominal value was --b(4)-------------------------------------------.  You judged this level of accuracy exhibited at --b(4)---------------- to be acceptable given the assays intended purpose of quantitating antibody titers to PRN for study V59P18 clinical samples.  Nonetheless, you indicated that you intend to temporarily reset the lower limit of quantitation (LLQ) for the PRN assay to be --b(4)----- pending improvements in the assay’s accuracy in this range and re-establishment of the LLQ at --b(4)----.  You also indicated in this submission that, as a consequence of the upward revision of LLQ for PRN from --b(4)----, you propose to rerun the statistical analyses for PRN that were presented in the interim V59P18 Clinical Study Report using the new LLQ and to include these revised results in the final version of the Clinical Study Report.  CBER concurs with this decision.

CBER supports --b(4)--------------- commitment to conduct further work to improve the accuracy of this assay in the lower range of quantitation and reminds both Novartis and   --b(4)-------------- that, while the PRN assay—as it currently stands—is deemed acceptable for the intended purposes of evaluating clinical samples from Study V59P18, it may not be deemed acceptable for other clinical studies that involve different study populations, different end points, etc.

In response to comments 1g and 1h, they submitted assay trending data demonstrating that critical assay parameters did not change from the time of assay validation to the time the clinical assays reported in this submission were conducted.  Moreover, they provided data supporting acceptable performance of the assay over the time period that the clinical samples were analyzed.

In regards to comment 2, the sponsor indicated that at the time the ELISAs were established and validated at ---b(4)---------, it was not yet common practice to preset acceptance criteria.  They did, however, provide information regarding pre-defined acceptance criteria for the re-qualification study performed in 2006 to re-evaluate the performance of the pertussis ELISAs to evolving standards.

In response to comment 3, they provided, in their submission of January 15, data that demonstrates adequate linearity of the assay.

In regards to comments concerning the assays conducted at the --b(4)--------------------------------------, the sponsor indicated in their submission of December 19, 2008, that they did not intend to respond to these issues and that they accept the resulting consequences in regards to the review of the BLA.  Thus, the pertussis ELISAs conducted for V59P11 cannot be considered adequately validated.  Therefore the pertussis ELISA data from that trial cannot be interpreted.  I made the BLA Review Committee Chair, the Clinical Reviewer, and the Regulatory Program Manager aware of these conclusions on January 13, 2009.

SUMMARY AND CONCLUSIONS:

In summary, based on my review, I consider the pertussis ELISA assays performed at       -b(4)----------------- for Study V59P18 to be sufficiently validated for their intended purposes.  I do not consider the pertussis ELISAs performed at the ---b(4)----------------------------------for Study V59P11 to be adequately validated and thus the pertussis ELISA data from V59P11 cannot be interpreted.