28 BLA 125742 0 07 20 2021 Memo Review

Pfizer Documents (PHMPT/FDA)

Pfizer Bla Submission

Pfizer 16 Plus Documents

5

Document text

BLA 125742/EU A   27034 D ata validation  Report  Summary and Subsequent  follow -up with  
Pfizer  
Our Reference:  BLA 125742/E UA 27034 (studied under  IND 19736) 
 
Sponsor: Pfizer -BioNTech  SE 
 
Product:  COVID- 19 (BNT162b2)  
 
Proposed Indication:  Prevention  of COVID- 19 in individuals  16 years  of age and older.  
 
On May  7, 2021 the BLA was submitted  and included datasets  for C4591001  and BNT162- 01. 
However,  prior  to this time  we had also received  part of the datasets  or all of the datasets  in 
the EUA  27034 as described below.  
On November  20, 2020,  the sponsor  submitted  the Emergency  Use Authorization  for individuals  
16 years  of age and older.  Two  clinical trial datasets  were  submitted:  
• BNT162- 01 - Phase  1/2, 2-Part,  Dose -Escalation  Trial Investigating  the Safety  and 
Immunogenicity  of Four  Prophylactic  SARS -CoV-2 RNA  Vaccines  Against  COVID- 19 Using  
Different  Dosing  Regimens  in Healthy  Adults  
• C4591001  - Phase  1/2/3 Study  to Evaluate  the Safety,  Tolerability,  Immunogenicity,  and 
Efficacy  of RNA  Vaccine  Candidates  Against  COVID- 19 in Healthy  Individuals.  The 
datasets  included analyses  for follow -up up to 2 months  after  Dose  2. 
 
On November  23, 2020,  the eDATA  team  discussed  the validation  results  with  the review  
committee for EUA  27034.  Three  study  datasets  were  validated  – BNT162- 01, C4591001- ia- 
efficacy  (IA efficacy  cutoff  of Nov 4, 2020)  and C4591001- safety -fa-eff (which  is the final 
efficacy/EUA  safety  analysis - data cutoff  of Nov 14, 2020) . 
On November  24, 2020 an IR was sent  regarding  clarification on discrepancies  already  found 
either  by the clinical/s tats reviewers  or by the validation.   On November  25, 2020,  a 
teleconference was held  with  Pfizer  in which  they  provided an explanation  for the 
discrepancies.  
Based  on the validation  results  (beginning  page 6 below)  and a deeper dive into the study  
datasets,  the following  limita tions were  identified:  
1. The datasets  are not clean  so results  may  not be entirely  accurate.  
2. There  are events  in CE that maybe  should be in AE and there  are events  in AE that  
should be in CE 
3. Temperature  was not always  reported  in VS to back  up “fever.”  Temperature  was also 
missing  for several  of the days.  Data  is missing  for the other  solicited  events  also.  
FDA-CBER-2021-5683-1058204
4. The signs  and symptoms  for those  subjects  with  COVID- like symptoms  but were  
determined  to be negative  for COVID were  also reported  in the FACE dataset.  This is ok 
except  many  of them  overlap  events  that could  be considered systemic  reactions  to the 
vaccine.   This is not reported  as an AE and thus  would  not be included in the analysis.  
5. In matching  up COVID positive  subjects  with  MH with  BMI I have  noticed  that obesity  is 
not consistently  used across  the board,  e.g. one subject  with  BMI of 31.5 is marked  as 
obese  whereas  another  with  BMI of 40.1 is not, so I don’t  think  we can rely solely  on the 
MH terms.  Vital signs  may  be more accurate.  
 
Even  with  these  limitations,  the datasets  were  still able  to be used to confirm  the safety  and 
efficacy  results  following  several  IRs to Pfizer.  The EUA  for adults  16 years  of age and older  was 
authorized on December  11, 2020.  
On March  15, 2021,  Pfizer  sent  via email the following  question regarding  the planned BLA 
submission:  
Pfizer/BioNTech  would  like to make a further  clarificat ion on the SDTM  mapping  for 
reactogenicity  data for C4591001.   Specifically,  Pfizer/BioNTech  will provide  the same  
mapping  logic as in the EUA  and also stated  in the pre -BLA briefing  document  Section  
1.3 whereby  th e SDTM  mapping  rules  will be applied  to provide a flat model  (FACE,  CE, 
VS, etc.)  solely  for reactogenicity  data collected  via e-diary  per study  design  and data 
collection  (e.g.,  not from  an AE form).  Pfizer/BioNTech  will not be submitting  a 
supplemental  format  schema  as part of the BLA as most  recently  described in the 
Response  to CBER  Comment  #1 of the 08 January  2021 Information  Request  regarding  
Trumenba  (Meningococcal  Group  B Vaccine)  submitted  on 12 March  2021 (STN  
125549/ 737).  If required,  Pfizer/BioNTech  can submit  tables  using  the new  mapping  if 
required.  Does  CBER  agree?  
 
On April 1, 2021,  we responded to their  question with  the following:  
 
Your  reference to a supplemental  format  schema  as described under  STN 125549 /737  is 
unclear.  We recommend  standardizing  the CE dataset  format,  as previously  
communicated  in correspondences  under  IND 19736 and IND  and summarized  
below: 
• Reactogenicity  events  that begin  within  the prespecified  assessment  period  but are 
currently  reported  in the AE dataset  should be transferred  from  AE to FACE,  or be 
flagged  in AE so that we know  that it is being  included in the CE dataset.  
• Records  covering  the entire  event  duration (which could  go beyond the protocol - 
defined assessment  period)  be created  in the ADaM3  ADCEVD  dataset  with  start/end  
dates  and durations  (based  on first and last days  the symptom  was present  as recorded  
(b) (4)
FDA-CBER-2021-5683-1058205
in the e-diary  and/or  Symptom  Resolved  Dates  CRF,  ignoring  any gaps)  derived  from  
both  the e -diary  and CRF data.  
• Include  all individual  supporting  assessments (daily  e-diary  and any unplanned 
assessments)  in the FACE domain  and the assessor for each  finding  can be identified  
using  FAEVAL  (STUDY SUBJECT  or INVESTIGATOR).  
• Maintain  one row per subject/vaccination/symptom  in the CE domain,  with  CE 
summarizing  the duration of the event  and maximum severity.  Maximum  severity  
(CESEV)  should be based  on the highest-levelseverity reported  by the subject  (via e- 
diary)  or investigator  (in the unplanned assessment  CRF).  
• Use SUPPCE  CESEV1  and CEDIFFRS  to show  assessment  of severity  by study  subject  
and reason  investigator’s  assessment  of severity  differed  from  study  subject  as needed  
for events  reported  in CE. 
On April 8, 2021,  a teleconference was held  to discuss  our request  for the BLA submission.  
Pfizer  agreed  to include  a separate set of datasets  (supplemental  tabulation  and analysis)  in 
which  all “solicited  events”  whether it be from  the diary  or investigator  are combined in the 
upcoming  BLA.  
On April 9, 2021,  the sponsor  submitted  an amendment  (132)  to the EUA  to extend  the 
authorization  to individuals  12-15 years  of age.  The data presented  in this EUA  amendment  
from  pivotal study  C4591001   (conducted under  IND 19736) consists of the immune  response  
measured  by SARS -CoV-2 neutralizing  antibody  titers  in this adolescent  group  compared  to 
young  adults  16-25 years  of age,  which  serves  as immunobridging  for adolescents;  efficacy  data 
in the 12-15 years  of age group;  and safety  data in approximately  2200 adolescents  with  a 
median  follow -up time  of at least  2 months  after  Dose  2. Additionally,  safety  data in 
adolescents  are compared  to the larger  data safety  data set of individuals  16-55 years  of age.  
The d atasets  also contain  the information  from  the adult  portion  of the study.  
 
On April 14, 2021,  the eDATA  team  discussed  the validation  results  with  the review  committee  
for EUA  27034 amendment  132.  One study’s  datasets  were  validated  – C4591001 which  
included all 48,901 subjects.  There  were  no major  dataset  concerns  as the data were  much  
cleaner.   The datasets  were  additionally  validated  for only  those  subjects  55 years  of age and 
younger  (n=29,  417) .  See validation  results  for Part II (all subjects)  and Part III (16 to 55 years  of 
age)  (beginning  page  … below).  
The EUA  for adolescents  12 to 15 years  of age was authorized on May  10, 2021.  
FDA-CBER-2021-5683-1058206
On May  7, 2021 the BLA was submitted  and included datasets  for C4591001  and BNT162- 01. 
This BLA (STN  125742/ 0)   contains  cumulative  follow -up from  Dose  1 to 6 months  after  Dose  2, 
as well as updated Efficacy  analyses  in blinded placebo -controlled  follow -up evaluated  duration 
of protection  (data cutoff  date:  13 March  2021),  and immunogenicity  analyses  of adults  (18 to 
85 years  of age)  including  data up to 1 month  after  Dose  2 in Phase  2, and up to 6 months  after  
Dose  2 in Phase  1.  The updated datasets  for C4591001  were  also submitted  to EUA  27034  
amendment  132 so they  do not need  to be validated  again.   The updated datasets  for BNT162- 
01 (6-Month  Follow -Up Data  from  subjects  16 Years  of Age and Older)  were  submitted  in EUA  
27034 amendment  174; however,  they  were  not validated  at that time.  
For C4591001:   As agreed  by Pfizer,  they  also submitted  a set of SDTM -SUPPL  and ADaM -SUPPL,  
which  is for: Supplemental  analysis  package  created  as a supplement  to the BLA esub package   
to revise  the reactogenicity  SDTM  and ADaM  data to address  the agreements  made  with  CBER.  
SDTM  includes  DM,  EX and updated domains  (AE, CE, FACE,  VS, SUPPAE, SUPPCE, SUPPFACE,  
SUPPVS,  RELREC),  define  and aCRF.  ADaM  includes  ADSL  and updated datasets  (ADAE,  ADCEVD,  
ADFACEVD).  Based  on what  needed  to be reviewed  with  this data we determined  that the 
supplemental  datasets  did not need  to be validated.  
On M ay  18, 2021,  an IR was sent  regarding  the C4591001  datasets  submitted  thus  far and Pfizer  
responded with  the following:  
1. Pfizer/BioNTech  confirm  that the datasets  submitted  in the EUA  27034- amendment  132 
are identical to the datasets  submitted  to BLA STN 125742/ 0. 
2. Pfizer/BioNTech  confirm  that the datasets  submitted  as SDTM -SUPPL  and ADaM -SUPPL  
contain  the reactogenicity  data changes  as requested  (including  flags)  in the 
teleconference held  on April 8, 2021 and nothing  additional.  
3. No data were  changed in the supplemental  EX and DM datasets.  They  were  included in 
the supplemental  package because  these  two datasets  would  be required  for pinnacle  
21 check  in case  FDA wants  to run its own  check.  
For study  BNT162- 01, we only  needed  to look  at immunogenicity  data for 24 subjects  that  
received  BNT162b2  (12 subjects  in the 18-55 age group  and 12 subjects  in the 56-85 yr age 
group  who  received  the 30mcg  dose)  so validation  is not necessary.  I looked  at the DM,  IS and 
EX datasets  to determine  if data was acceptable.   Immunogenicity  data was not provided for 12 
of the 24 subjects.  
On June  8, 2021,  an IR was sent  requesting  the immunogenicity  data for the 12 subjects  in 
BNT162- 01.  Pfizer  responded on June  16, 2021 (amendment  6). They  indicated  that the data 
were  not available  at the time  of immunogenicity  cut-off for this report  because  the samples  
were  put on hold  due to necessary  testing  prioritizations  at the Pfizer  labs (eg C4591001  6- 
month  stability  and booster;  C4591007) .  Pfizer has now  resumed  testing  of these  samples  and 
an updated BNT162- 01 study  report  will be provided once  it is available.  Pfizer  did not believe  
FDA-CBER-2021-5683-1058207
these  data to be material  to the review  of the BLA.   Please see clinical review  regarding  the 
immunogenicity  data results.  
FDA-CBER-2021-5683-1058208