125742 S1 M5 5351 c4591001 interim mth6 sap

Pfizer Documents (PHMPT/FDA)

Pfizer Bla Submission

Pfizer 16 Plus Documents

96

Document text

Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 1
TMF Doc ID: 98.03Protocol C4591001
A PHASE 1/2/3, PLACEBO -CONTROLLED, RANDOMIZED, OBSERVER -BLIND , 
DOSE -FINDING STUDY T O EVALUATE THE SAFETY, TOLERAB ILITY, 
IMMUNOGENICITY, AND EFFICACY OF SARS -COV -2RNA VACCINE 
CANDIDATES AGAINST C OVID -19 IN HEALTHY INDIVIDUALS
Statistical Analysis Plan 
(SAP)
Version: 5
Date: 17Mar 2021
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221083
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 2
TMF Doc ID: 98.03TABLE OF CONTENTS
LIST OF TABLES ................................ ................................ ................................ ..................... 7
LIST OF FIGURES ................................ ................................ ................................ ................... 7
APPENDI CES ................................ ................................ ................................ ........................... 8
1. VERSI ON HISTORY ................................ ................................ ................................ ............ 9
2. INTRODUCTION ................................ ................................ ................................ ................. 9
2.1. Study  Objectives, Endpoints, and Estimands ................................ ............................ 9
2.2. Study  Design ................................ ................................ ................................ ........... 18
2.2.1. Overall Design ................................ ................................ ............................ 18
2.2.2. Phase 1 ................................ ................................ ................................ ........ 19
2.2.3. Phase 2/3................................ ................................ ................................ .....21
3. ENDPOINTS AND BASELINE VARIABLES: DEF INIT IONS AND 
CONVENTIONS ................................ ................................ ................................ ................ 24
3.1. Primary  Endpoints ................................ ................................ ................................ ...24
3.1.1. Safet y Endpoints ................................ ................................ ......................... 24
3.1.1.1. L ocal Reactions ................................ ................................ ......... 25
3.1.1.2. Sy stemic Events (Sy stemic Event S ymptoms and Fever) ......... 27
3.1.1.3. Use of Antip yretic Medication ................................ .................. 28
3.1.1.4. Adverse Events ................................ ................................ .......... 29
3.1.1.5. Serious Adverse Events ................................ ............................. 29
3.1.1.6. Hematology  and Chemistry  Laboratory  Parameters (for 
Phase 1 Only ) ................................ ................................ .................... 30
3.1.2. I mmunogenicit y Endpoints (for the Phase 2/3 Subset for Evaluation 
of Boostability and Protection Against Emerging VOCs Only )...................... 31
3.1.3. Vaccine Efficacy  Endpoints (for Phase 2/3 Only )................................ ......32
3.2. Secondary  Endpoints ................................ ................................ ............................... 32
3.2.1. I mmunogenicit y Endpoints ................................ ................................ ......... 32
3.2.2. Vaccine Efficacy  Endpoints (for Phase 2/3 Only )................................ ......33
3.3. Exploratory  Endpoints ................................ ................................ ............................. 34
3.3.1. Safet y Endpoints (for Phase 1 Boostability Assessment Only) .................. 34
3.3.2. Vaccine Efficacy  Endpoints (for Phase 2/3 Only )................................ ......34
3.3.3. I mmunogenicit y Endpoints ................................ ................................ ......... 34
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221084
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 3
TMF Doc ID: 98.033.3.4. Additional Endpoints (for Phase 2/3 Onl y)................................ ................ 35
3.4. Baseline and Other Variables ................................ ................................ .................. 35
3.4.1. Demographics, Medical History , and Phy sical Examination ..................... 35
3.4.2. E- Diary  Completion ................................ ................................ .................... 36
3.4.3. Prior/Concomitant Vaccines and Concomitant Medications...................... 36
3.5. Safet y Endpoints ................................ ................................ ................................ .....37
4. ANALYSIS SETS (PO PUL ATIONS FOR ANALYS IS)................................ ................... 37
5. GENERAL METHODOLOGY AND CONVENTIONS ................................ .................... 39
5.1. Hy potheses and Decision Rules ................................ ................................ .............. 39
5.1.1. Vaccine Efficacy  Hypothesis ................................ ................................ ......39
5.1.2. I mmunogenicit y Hypothesis ................................ ................................ .......40
5.1.2.1. Hy pothesis for Immunogenicit y Bridging of 12 to 15 
Years to 16 to 25 Years ................................ ................................ .....40
5.1.2.2. Hy pothesis for Boostability  and Protection Against 
Emerging SARS -CoV -2 VOCs ................................ ......................... 40
5.1.3. Sample Size ................................ ................................ ................................ 41
5.1.3.1. Phase 1................................ ................................ ....................... 41
5.1.3.2. Efficacy  Against COVID -19................................ ..................... 41
5.1.3.3. Efficacy  Against Asy mptomatic Infection ................................ 41
5.1.3.4. I mmunogenicit y Bridging of 12 to 15 Years to 16 to 25 
Years ................................ ................................ ................................ ..41
5.1.3.5. Boostability  and Protection Against Emerging SARS -
CoV -2 VOCs ................................ ................................ ..................... 42
5.1.3.6. Safet y................................ ................................ ......................... 43
5.1.4. Multiplicity  Considerations ................................ ................................ ........ 44
5.1.4.1. Phase 1 ................................ ................................ ....................... 44
5.1.4.2. Phase 2/3 Vaccine Efficacy ................................ ....................... 44
5.1.4.3. Phase 2/3 I mmunogenicity ................................ ........................ 44
5.2. General Methods ................................ ................................ ................................ .....45
5.2.1. Analy ses for Binary  Data ................................ ................................ ............ 45
5.2.2. Analy ses for Count Data ................................ ................................ ............. 45
5.2.3. Analy ses for Continuous Data ................................ ................................ ....45
5.2.3.1. Geometric Means ................................ ................................ ......46
5.2.3.2. Geometric Mean Fold Rises................................ ...................... 46
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221085
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 4
TMF Doc ID: 98.035.2.3.3. Geometric Mean Ratios ................................ ............................. 46
5.2.3.4. Geometric Mean Fold Rise Ratios ................................ ............ 47
5.2.3.5. Reverse Cumulative Distribution Curves ................................ ..47
5.3. Methods to Manage Missing Data ................................ ................................ .......... 47
6. ANALYSES AND SUMM ARIES................................ ................................ ...................... 47
6.1. Primary  Endpoint(s) ................................ ................................ ................................ 47
6.1.1. S afety Endpoints ................................ ................................ ......................... 47
6.1.1.1. L ocal Reactions ................................ ................................ ......... 48
6.1.1.2. Sy stemic Events ................................ ................................ ........ 48
6.1.1.3. Adverse Events ................................ ................................ .......... 49
6.1.1.4. S erious Adverse Events ................................ ............................. 50
6.1.1.5. Hematology  and Chemistry  Parameters (for Phase 1 Onl y).....51
6.1.2. I mmunogenicit y Endpoints (for the Phase 2/3 Subset for Evaluation 
of Boostability and Protection Against Emerging VOCs Only )...................... 51
6.1.2.1. SARS -CoV -2 Reference Strain NT and SA NT at 1 
Month After Dose 3 vs Reference Strain NT at 1 Month After 
Dose 2 in BNT162b2 -Experienced Participants................................ 51
6.1.2.2. Seroresponse to the Reference Strain and SA Strain at 1 
Month After Dose 3 vs Seroresponse to the Reference Strain at 
1 Month After Dose 2 in BNT162b2 -Experienced Participants .......52
6.1.2.3. SARS -CoV -2 SA NT at 1 Month After Dose 2 vs 
Reference Strain NT a t 1 Month After Dose 2 in BNT162b2-
Naïve Participants ................................ ................................ .............. 53
6.1.2.4. Seroresponse to the SA Strain at 1 Month After Dose 2 vs 
Serorespon se to the Reference Strain at 1 Month After Dose 2 
in BNT162b2- Naïve Participants ................................ ...................... 54
6.1.3. Vaccine Efficacy  Endpoints (for Phase 2/3 Only )................................ ......54
6.1.3.1. COVID -19 Incidence per 1000 Person -Years of Follow -
up ................................ ................................ ................................ .......54
6.2. Secondary  Endpoints ................................ ................................ ............................... 56
6.2.1. I mmunogenicit y Endpoints ................................ ................................ ......... 56
6.2.1.1. SARS -CoV -2 Neutralizing Titers (Phase 1) ............................. 56
6.2.1.2. S1- Binding IgG Levels and RBD -Binding IgG Levels 
(Phase 1) ................................ ................................ ............................ 57
6.2.1.3. SARS -CoV -2 Neutralizing Titers vs SARS -CoV -2 S1-
Binding IgG Levels and RBD -Binding IgG Levels (Phase 1) .......... 58
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221086
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 5
TMF Doc ID: 98.036.2.1.4. SARS -CoV -2 Neutralizing Titers in Participants 12 to 15 
Years of Age vs Those 16 to 25 Years of Age (Phase 2/3) ............... 59
6.2.1.5. SARS -CoV -2 SA NT and Reference Strain NT at 1 
Month After Dose 3 vs Reference Strain NT at 1 Month After 
Dose 2 in BNT162b2 -Experienced Participants................................ 60
6.2.1.6. Seroresponse to the SA Strain and Reference Strain at 1 
Month After Dose 3 vs Seroresponse to the Reference Strain at 
1 Month After Dose 2 in BNT162b2 -Experienced Participants .......61
6.2.1.7. SARS -CoV -2 SA NT After Dose 3 (BNT162b2 -
Experienced Participants)................................ ................................ ..61
6.2.1.8. SARS -CoV -2 SA NT at 1 Month After Dose 4 vs 
Reference Strain NT at 1 Month After Dose 2 in BNT162b2 -
Experienced Participants ................................ ................................ ...62
6.2.1.9. SARS -CoV -2 SA NT at 1 Month After Dose 2 
(BNT162b2 -Naïve Participants) ................................ ........................ 63
6.2.1.10. Seroresponse to the SA Strain at 1 Month After Dose 2 
(BNT162b2 -Naïve Participants) ................................ ........................ 64
6.2.1.11. Reference Strain NT at 1 Month After Dose 2 
(BNT162b2 -Naïve Participants) ................................ ........................ 64
6.2.2. Vaccine E fficacy  Endpoints (for Phase 2/3 Only )................................ ......65
6.2.2.1. COVID -19 Incidence per 1000 Person -Years of Follow -
up ................................ ................................ ................................ .......65
6.2.2.2. Confirmed Severe COVID -19 Incidence per 1000 Person -
Years of Follow -up................................ ................................ ............ 66
6.2.2.3. Confirmed COVID -19 Incidence per 1000 Person -Years 
of Follow -up (According to the CDC -Defined S ymptoms) .............. 67
6.2.2.4. I ncidence of As ymptomatic SARS -CoV -2 Infection per 
1000 Person- Years of Follow -up (According to the N -Binding 
Antibody  Seroconversion) ................................ ................................ .68
6.2.2.5. I ncidence of As ymptomatic SARS -CoV -2 Infection per 
1000 Person- Years of Follow -up (According to the Central 
Laboratory –Confirmed NAAT) ................................ ......................... 69
6.3. Exploratory  Endpoints ................................ ................................ ............................. 69
6.3.1. Safet y Endpoints (for Phase 1 Boostability Assessment Only) .................. 69
6.3.1.1. L ocal Reactions ................................ ................................ ......... 69
6.3.1.2. Sy stemic Events ................................ ................................ ........ 70
6.3.1.3. Adverse Events ................................ ................................ .......... 71
6.3.1.4. Serious Adverse Events ................................ ............................. 72
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221087
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 6
TMF Doc ID: 98.036.3.2. Vaccine Efficacy  Endpoints (for Phase 2/3 Only )................................ ......72
6.3.2.1. COVID -19 Incidence per 1000 Person -Years of Blinded 
Follow -up ................................ ................................ .......................... 72
6.3.2.2. COVID -19 Incidence per 1000 Person -Years of Follow -
up ................................ ................................ ................................ .......73
6.3.2.3. I ncidence of As ymptomatic SARS -CoV -2 Infection per 
1000 Person- Years of Follow -up (According to the N -Binding 
Antibody  Seroconversion) ................................ ................................ .74
6.3.2.4. I ncidence of As ymptomatic SARS -CoV -2 Infection per 
1000 Person- Years of Follow -up (According to the Central 
Laboratory –Confirmed NAAT) ................................ ......................... 75
6.3.3. I mmunogenicit y Endpoints ................................ ................................ ......... 75
6.3.3.1. SARS -CoV -2 Reference -Strain Neutralizing Titers, 
SARS -CoV -2 SA -Variant Neutralizing Titers, and Full -Length 
S-Binding or S1 -Binding IgG Levels (Phase 1 ) ................................ 75
6.3.3.2. SARS -CoV -2 Reference -Strain Neutralizing Titers and 
SARS -CoV -2 SA -Variant Neutralizing Titers at 1 Month After 
Dose 3 vs SARS -CoV -2 Reference -Strain Neutralizing Titers 
at 1 Month After Dose 2 (Phase 1) ................................ .................... 76
6.3.3.3. SARS -CoV -2 Neutralizing Titers, and F ull-length S -
Binding or S1 -Binding IgG Levels (Phase 2/3) ................................ 77
6.3.3.4. Serological Responses in Participants With Confirmed 
COVID -19, Confirmed Severe COVID -19, and SARS -CoV -2 
Infection Without Confirmed COVID-19 ................................ ......... 78
6.3.3.5. SARS -CoV -2 NTs for An y VOCs (Phase 3, Boostability  
and Protection Against Emerging VOCs) ................................ ......... 78
6.3.4. Additional Analy sis................................ ................................ .................... 78
6.4. Subgroup Analy sis................................ ................................ ................................ ..79
6.5. Baseline and Other Summaries and Anal yses................................ ......................... 79
6.5.1. Baseline Summaries ................................ ................................ .................... 79
6.5.1.1. Demographic Characteristics ................................ .................... 79
6.5.1.2. Medical History ................................ ................................ ......... 79
6.5.2. Study  Conduct and Participant Disposition ................................ ................ 79
6.5.2.1. Participant Disposition ................................ .............................. 79
6.5.2.2. Blood Samples for Assay ................................ .......................... 79
6.5.2.3. E- Diaries ................................ ................................ .................... 80
6.5.3. Study  Vaccination Exposure ................................ ................................ .......80
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221088
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 7
TMF Doc ID: 98.036.5.3.1. Vaccination Timing and Administration ................................ ...80
6.5.4. Prior/Concomitant Vaccination and Concomitant Medications ................. 80
6.6. Safet y Summaries and Analyses ................................ ................................ ............. 80
7. ANALYSES TIMING ................................ ................................ ................................ ......... 80
7.1. I ntroduction of Interim Analy sis................................ ................................ ............. 80
7.2. I nterim Anal yses and Summaries ................................ ................................ ............ 83
7.2.1. Data Monitoring Committee ................................ ................................ .......84
8. REFERENCES ................................ ................................ ................................ .................... 85
9. APPENDI CES ................................ ................................ ................................ ..................... 86
LIST OF TABLES
Table 1. Summary  of Changes ................................ ................................ .................. 9
Table 2. List of Primary  and Secondary  Objectives, Estimands, and 
Endpoints for Phase 1 ................................ ................................ ............... 10
Table 3. List of Primary , Secondary, and Tertiary /Exploratory  Objectives, 
Estimands, and Endpoints for Phase 2/3 ................................ ................... 11
Table 4. Derived Variables for Presence of Each and An y Local Reaction 
Within 7 Days for Each Dose ................................ ................................ ...25
Table 5. Local Reaction Grading Scale ................................ ................................ ..26
Table 6. Systemic Event Grading Scale ................................ ................................ ..27
Table 7. Scale for Fever ................................ ................................ .......................... 28
Table 8. Laboratory  Abnormality  Grading Scale ................................ ................... 30
Table 9. Power Anal ysis for Noninferiorit y Assessment ................................ .......42
Table 10. Probability  of Observing at Least 1 AE by  Assumed True Event 
Rates With Different Sample Sizes ................................ .......................... 43
Table 11. Interim Anal ysis Plan and Boundaries for Efficacy  and Futility .............. 82
Table 12. Statistical Design Operating Characteristics: Probability  of Success 
or Failure for Interim Analy ses................................ ................................ .82
Table 13. Statistical Design Operating Characteristics: Probability  of Success 
for Final Anal ysis and Overall ................................ ................................ ..83
LIST OF FIGURES
Figure 1. Multiplicity  Schema ................................ ................................ .................. 44
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221089
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 8
TMF Doc ID: 98.03APPENDICES
Appendix 1. List of Abbreviations ................................ ................................ ........................... 86
Appendix 2. Details for Bayesian Design ................................ ................................ ................ 88
Appendix 3. IRR and VE Derivation ................................ ................................ ....................... 90
Appendix 4. Asy mptomatic Case Based on N -Binding Antibod y Seroconv ersion ................. 95
Appendix 5. Asy mptomatic Case Based on Central L aboratory –Confirmed NAAT.............. 96
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221090
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 9
TMF Doc ID: 98.031.VERSION HISTORY
Table 1.Summary of C hanges
Version /
DateAssociated Protocol 
AmendmentSummary and Rationale for Changes
1/
20 May 2020Protocol amendment 1 ,
13May 2020N/A
2/
30Jul2020Protocol amendment 5,
24July 2020Implem ented the changes made in p rotocol amendment s 2through 
5.
3/
02Nov 2020Protocol amendment 9,
29Oct 2020Implem ented the changes made in protocol amendments 6 through 
9.
4/
08Jan2021Protocol amendment 11 , 
04Jan2021Implem ented the changes made in protocol amendments 10 and
11.
5/
17Mar 2021Protocol amendment 14,
02 Mar 2021Implem ented the changes made in protocol amendments 12 
through 14.
2.INTRODUCTION
This SAP provides the detailed methodology  for summary and statistical analy ses of the data 
collected in Study  C4591001 . This document may modify  the plans outlined in the protocol; 
however, an y major modifications of the primary endpoint definition or it s anal ysis will also 
be reflected in a protocol amendment.
2.1. Study Objectives , Endpoints ,and Estimands
The estimands corresponding to each primary , secondary , and tertiary /exploratory  objective 
are described in Table 2and Table 3below.
In the primary  safet y objective evaluations, missing e -diary  data will not be imputed. 
Missing AE dates will be imputed according to Pfizer safet y rules. No other missing 
information will be imputed in the safet y anal ysis.
The estimands to evaluate the immunogenicity objectives are based on evaluable populations 
for immunogenicity  (seeSection 4for definition). These estimands estimate vaccine effect 
in the hy pothetical setting where participants follow the study  schedules and protocol 
requirements as directed. Missing antibody results will not be imputed. Immunogenicit y 
results that are below the LLOQ will be set to 0.5 × LLOQ in the anal ysis; this may  be 
adjusted once additional data on the assay  characteristics become availabl e.
The estimands to evaluate the efficacy objectives are based on evaluable populations for 
efficacy (seeSection 4for definition). These estimands estimate vaccine effect in the 
hypothetical setting where participants follow the study  schedules and protocol requirements 
as directed. In addition, VE w ill be anal yzed by the all -available efficacy population s.  
Missing laboratory results will not be imputed for the primary  analy sis, but missing data 
imputation for the efficacy endpoint may  be performed as a sensitivity  analy sis.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221091
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 10
TMF Doc ID: 98.03Table 2.List of Primary andSecondary Objectives, Estimands, and Endpoints for 
Phase 1
Objectives Estimands Endpoints
Prim ary: Prim ary: Prim ary: 
To describe the safety and tolerability 
profiles of prophylactic BNT162 
vaccines in healthy adults after 1 or 2 
dosesIn participants receiving at least 1 dose 
of study intervention, the percentage of 
participants reporting:
 Local reactions for up to 7 days 
following each dose 
 Systemic events for up to 7 days 
following each dose
 Adverse events (AEs) from 
Dose 1 to 1 month after the last 
dose
 Serious AEs (SAEs) from Dose 1 
to 6 months after the last dose Local reactions (pain at the 
injection site, redness, and 
swelling)
 Systemic events (fever, fatigue, 
headache, chills, vomiting, 
diarrhea, new or worsened musc le 
pain, and new or worsened joint 
pain)
 AEs
 SAEs
In addition, the percentage of 
participants with:
 Abnormal hematology and 
chemistry laboratory values 1 and 
7 days after Dose 1; and 7 days 
after Dose 2
 Grading shifts in hematology and 
chemistry laboratory assessments 
between baseline and 1 and 7 days 
after Dose 1; and before Dose 2 
and 7 days after Dose 2Hematology and chemistry laboratory 
parameters detailed in the protocol, 
Section 10.2
Secondary: Secondary: Secondary: 
To describe the immune responses 
elicited by prophylactic BNT162 
vaccines in healthy adults after 1 or 2 
dosesIn participants complying with the key 
protocol criteria (evaluable 
participants) at the following time 
points after receipt of study 
intervention :7 and 21 days after Dose 
1; 7and 14 days and1, 6, 12 ,and 24
months after Dose 2
 Geometric mean titers (GMTs) at 
each time point
 Geometric mean fold rise (GMFR) 
from before vaccination to each 
subsequent time point after 
vaccination
 Proportion of participants 
achieving ≥4-fold rise from before 
vaccination to each subsequent 
time point after vaccinationSARS -CoV -2 neutralizing titers
 Geometric mean concentrations 
(GMCs) at each time point
 GMFR from prior to first dose of 
study intervention to each 
subsequent time point
 Proportion of participants 
achieving ≥4-fold rise from before 
vaccination to each subsequent 
time point after vaccinationS1-binding IgGlevels and 
RBD -binding IgGlevels
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221092
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 11
TMF Doc ID: 98.03Table 2.List of Primary andSecondary Objectives, Estimands, and Endpoints for 
Phase 1
Objectives Estimands Endpoints
 Geometric mean ratio (GMR), 
estimated by the ratio of the 
geometric mean of SARS -CoV -2 
neutralizing titers to the geometric 
mean of binding IgGlevels at each 
time point SARS -CoV -2 neutralizing titers
 S1-binding IgGlevels
 RBD -binding IgGlevels
Exploratory : Exploratory : Exploratory : 
To describe the immune responses 
elicited by a third dose of prophylactic 
BNT162 b2administered to healthy 
adults 6 to 12 months after the second 
dose of either BNT162 b1 or 
BNT162 b2 GMC/ GMT at the time of Dose 3 
and 7 days and 1 month after 
Dose 3, and GMFR f rom before 
Dose 3 to 7 days and 1 month after 
Dose 3 SARS -CoV -2 reference -strain 
neutralizing titers
 SARS -CoV -2 SA-variant 
neutralizing titers
 Full-length S -binding or 
S1-binding IgGlevels
 GMR of SARS-CoV -2reference -
strain neutralizing titers 1 month 
after Dose 3 to 1 month after 
Dose 2 SARS -CoV -2 reference -strain 
neutralizing titers
 GMR of SARS-CoV -2SA-variant
neutralizing titers 1 month after 
Dose 3 to SARS -CoV -2reference -
strain neutralizing titers 1 month 
after Dose 2 SARS -CoV -2 reference -strain 
neutralizing titers
 SARS -CoV -2 SA-variant 
neutralizing titers
To de scribe the safety profile of a third 
dose of prophylactic BNT162 b2
administered to healthy adults 6 to 12 
months after the second dose of either 
BNT162 b1 or BNT162 b2In pa rticipants receiving a third dose of 
BNT162b2 , the percentage of 
participants reporting:
 Local reactions for up to 7 days 
after Dose 3
 Systemic events for up to 7 days 
after Dose 3
 AEs and SAEs from Dose 3to 
1month after Dose 3 Local reactions (pain at the 
injection site, redness, and 
swelling)
 Systemic events (fever, fatigue, 
headache, chills, vomiting, 
diarrhea, new or worsened muscle 
pain, and new or worsened joint 
pain)
 AEs
 SAEs
Table 3. List of Primary, Secondary, and Tertiary/Exploratory Objectives, 
Estimands, and Endpoints for Phase 2/3
ObjectivesaEstimands Endpoints
Prim ary Efficacy
To evaluate the efficacy of prophylactic 
BNT162b2 against confirmed 
COVID -19 occurring from 7 days after 
the second dose in participants without 
evidence of infection before 
vaccinationIn participants complying with the key 
protocol criteria (evaluable 
participants) at least 7 days after receipt 
of the second dose of study 
interve ntion:
100 × (1 –IRR) [ratio of active vaccine 
to placebo]COVID -19 incidence per 1000 
person -years of follow -up based on 
central laboratory or locally confirmed 
NAAT in participants with no 
serological or virological evidence (up 
to 7 days after receipt of the second 
dose) of past SARS-CoV -2 infection
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221093
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 12
TMF Doc ID: 98.03Table 3. List of Primary, Secondary, and Tertiary/Exploratory Objectives, 
Estimands, and Endpoints for Phase 2/3
ObjectivesaEstimands Endpoints
To evaluate the efficacy of prophylactic 
BNT162b2 against confirmed 
COVID -19 occurring from 7 days after 
the second dose in participants with 
and without evidence of infection 
before vaccinationIn participants complying with the key 
protocol criteria (evaluable 
participants) at least 7 days after receipt 
of the second dose of study 
intervention:
100 × (1 –IRR) [ratio of active vaccine 
to placebo]COVID -19 incidence per 1000 
person -years of follow -up based on 
central laboratory or locally confirmed 
NAAT
Prim ary Safety
To define the safety profile of 
prophylactic BNT162b2 in the first 360 
participants randomized (Phase 2)In participants receiving at least 1 dose 
of study intervention, the percentage of 
participants reporting:
 Local reactions for up to 7 days 
following each dose
 Systemic events for up to 7 days 
following each dose
 AEs from Dose 1 to 7 days after 
the second dose
 SAEs from Dose 1 to 7 days after 
the second dose Local reactions (pain at the 
injection site, redness, and 
swelling)
 Systemic events (fever, fatigue, 
headache, chills, vomiting, 
diarrhea, new or worsened muscle 
pain, and new or worsened joint 
pain)
 AEs
 SAEs
To define the safety profile of 
prophylactic BNT162b2 in all 
participants randomized in Phase 2/3In participants receiving at least 1 dose 
of study intervention, the percentage of 
participants reporting:
 Local reactions for up to 7 days 
following each dose
 Systemic events for up to 7 days 
following each dose
 AEs from Dose 1 to 1 month after 
the second dose
 SAEs from Dose 1 to 6 months 
after the second dose AEs
 SAEs
 In a subset of at least 6000 
participants:
o Local reactions (pain at the 
injection site, redness, and 
swelling)
o Systemic events (fever, 
fatigue, hea dache, chills, 
vomiting, diarrhea, new or 
worsened muscle pain, and 
new or worsened joint pain)
To define the safety profile of 
prophylactic BNT162b2 in participants 
12 to 15 years of age in Phase 3In participants receiving at least 1 dose 
of study intervention, the percentage of 
participants reporting:
 Local reactions for up to 7 days 
following each dose
 Systemic events for up to 7 days 
following each dose
 AEs from Dose 1 to 1 month after 
the second dose
 SAEs from Dose 1 to 6 months 
after the second dose Local reactions (pain at the 
injection site, redness, and 
swelling)
 Systemic events (fever, fatigue, 
headache, chills, vomiting, 
diarrhea, new or worsened muscle 
pain, and new or worsened joint 
pain)
 AEs
 SAEs
To describe the safety and tolerability 
profile of BNT162 b2SAgiven as 1 or 2 
doses to BNT162b2 -experienced
participants , or as 2 doses to 
BNT162b2 -naïve participants
To describe the safety and tolerability 
profile of BNT162 b2given as a third 
dose to BNT162b2-experienced
participantsIn participants receiving at least 1 dose 
of study intervention , the percentage of 
participants reporting:
Local reactions for up to 7 days 
following each dose 
Systemic events for up to 7 days 
following each dose
AEs from Dose 1 to 1 month after 
the last dose
SAEs from Dose 1 to 5 or 6 months 
after the last doseLocal reactions (pain at the 
injection site , redness , and 
swelling )
Systemic events (fever, fatigue, 
headache, chills, vomiting, 
diarrhea, new or worsened muscle 
pain, and new or worsened joint 
pain)
AEs
SAEs
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221094
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 13
TMF Doc ID: 98.03Table 3. List of Primary, Secondary, and Tertiary/Exploratory Objectives, 
Estimands, and Endpoints for Phase 2/3
ObjectivesaEstimands Endpoints
Prim ary Immunogenicity
BNT162b2 -experienced participants
Todemonstrate the noninferiority of 
the anti –reference strain immune 
response after a third dose of 
BNT162b2 compared to after 2 doses 
of BNT162b2, in the same individualsGMR of reference strain NT 1 month 
after the third dose of BNT162b2 to 
1month after the second dose of 
BNT162b2
The difference in percentages of 
participants with seroresponse to the 
reference strain at 1 month after the 
third dose of BNT162b2 and 1 month 
after the second dose of BNT162b2SARS -CoV -2 reference strain NTs in 
participants with no serological or 
virological evidence (up to 1 month
after receipt of the third dose of 
BNT162b2 ) of past SARS -CoV -2 
infection
Todemonstrate the noninferiority of 
the anti-SA immune response after 
1dose of BNT162b2 SAcompared to the 
anti–reference strain immune response 
after 2 doses of BNT162b2, in the same 
individualsGMR of SA NT 1month after 1 dose 
of BNT162b2 SAto the reference strain 
NT 1 month after the second dose of 
BNT162b2
The difference in percentages of 
participants with seroresponse to the 
SAstrain at 1 month after 1 dose of 
BNT162b2 SAand seroresponse to the 
reference strain at 1 month after the 
second dose of BNT162b2SARS -CoV -2 SA and reference strain 
NTs in participants with no serological 
or virological evidence (up to 1 month
after receipt of 1dose of BNT162b2 SA) 
of past SARS -CoV -2 infection
BNT162b2 -naïve participants
Todemonstrate the noninferiority of 
the anti-SA immune response after 2 
doses of BNT162b2 SAcompared to the 
anti–reference strain immune response 
after 2 doses of BNT162b2 GMR of SA NT 1month after the 
second dose of BNT162b2 SAto the 
reference strain NT 1 month after the 
second dose of BNT162b2
The difference in percentages of 
participants with seroresponse to the 
SAstrain at 1 month after the second 
dose of BNT162b2 SAand seroresponse 
to the reference strai n at 1 month after 
the second dose of BNT162b2SARS -CoV -2 SA and reference strain 
NTs in participants with no serological 
or virological evidence (up to 1 month
after receipt of the second dose of 
BNT162b2 SAor BNT162b2 as 
appropriate ) of past SARS -CoV -2 
infection
Secondary Efficacy
To evaluate the efficacy of prophylactic 
BNT162b2 against confirmed 
COVID -19 occurring from 14days 
after the second dose in participants 
without evidence of infection before 
vaccinationIn participants complying with the key 
protocol criteria (evaluable 
participants) at least 14days after 
receipt of the second dose of study 
intervention:
100 × (1 –IRR) [ratio of active vaccine 
to placebo]COVID -19 incidence per 
1000 person -years of foll ow-up based 
on central laboratory or locally 
confirmed NAAT in participants with 
no serological or virological evidence 
(up to 14days after receipt of the 
second dose) of past SARS-CoV -2 
infection
To evaluate the efficacy of prophylactic 
BNT162b2 against confirmed 
COVID -19 occurring from 14days 
after the second dose in participants 
with and without evidence of infection 
before vaccinationIn participants complying with the key 
protocol criteria (evaluable 
participants) at least 14days after 
receipt of t he second dose of study 
intervention:
100 × (1 –IRR) [ratio of active vaccine 
to placebo]COVID -19 incidence per 
1000 person -years of follow -up based 
on central laboratory or locally 
confirmed NAAT
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221095
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 14
TMF Doc ID: 98.03Table 3. List of Primary, Secondary, and Tertiary/Exploratory Objectives, 
Estimands, and Endpoints for Phase 2/3
ObjectivesaEstimands Endpoints
To evaluate the efficacy of prophylactic 
BNT162b2 against confirmed severe 
COVID -19 occurring from 7 days and 
from 14 days after the second dose in 
participants without evidence of 
infection before vaccinationIn participants complying with the key 
protoco l criteria (evaluable 
participants) 
 at least 7 days 
and 
 at least 14 days 
after receipt of the second dose of study 
intervention:
100 × (1 –IRR) [ratio of active vaccine 
to placebo]Confirmed severe COVID -19 incidence 
per 1000 person -years of follow -up in 
participants with no serological or 
virological evidence (up to 7 days and 
up to 14 days after receipt of the second 
dose) of past SARS -CoV -2 infection
To evaluate the efficacy of prophylactic 
BNT162b2 against confirmed severe 
COVID -19 occurring from 7 days and 
from 14 days after the second dose in 
participants with and without evidence 
of infection before vaccinationIn participants complying with the ke y 
protocol criteria (evaluable 
participants) 
 at least 7 days 
and
 at least 14 days
after receipt of the second dose of study 
intervention:
100 × (1 –IRR) [ratio of active vaccine 
to placebo]Confirmed severe COVID -19 incidence 
per 1000 person -years of fo llow-up
To describe the efficacy of prophylactic 
BNT162b2 against confirmed 
COVID -19 (according to the 
CDC -defined symptoms) occurring 
from 7 days and from 14 days after the 
second dose in participants without 
evidence of infection before 
vaccinationIn participants complying with the key 
protocol criteria (evaluable 
participants) 
 at least 7 days 
and 
 at least 14 days
after receipt of the second dose of study 
intervention:
100 × (1 –IRR) [ratio of active vaccine 
to placebo]COVID -19 incidence per 10 00 
person -years of follow -up based on 
central laboratory or locally confirmed 
NAAT in participants with no 
serological or virological evidence (up 
to 7 days and up to 14 days after receipt 
of the second dose) of past SARS -CoV -
2 infection
To describe the efficacy of prophylactic 
BNT162b2 against confirmed 
COVID -19 (according to the 
CDC -defined symptoms) occurring 
from 7 days and from 14 days after the 
second dose in participants with and 
without evidence of infection before 
vaccinationIn participants comp lying with the key 
protocol criteria (evaluable 
participants) 
 at least 7 days 
and
 at least 14 days
after receipt of the second dose of study 
intervention:
100 × (1 –IRR) [ratio of active vaccine 
to placebo]COVID -19 incidence per 1000 
person -years of f ollow -up based on 
central laboratory or locally confirmed 
NAAT
To evaluate the efficacy of prophylactic 
BNT162b2 against non -S 
seroconversion to SARS -CoV -2 in 
participants without evidence of 
infection or confirmed COVID-19 In participants complying with the key 
protocol criteria (evaluable 
participants):
100 × (1 –IRR) [ratio of active vaccine 
to placebo]Incidence of asymptomatic 
SARS -CoV -2 infection per 1000 
person -years of follow -up based on N -
binding antibody seroconversion in 
participants with no serological or 
virological evidence of past 
SARS -CoV -2 infection or confirmed 
COVID -19 
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221096
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 15
TMF Doc ID: 98.03Table 3. List of Primary, Secondary, and Tertiary/Exploratory Objectives, 
Estimands, and Endpoints for Phase 2/3
ObjectivesaEstimands Endpoints
To evaluate the efficacy of prophylactic 
BNT162b 2against asymptomatic 
SARS -CoV -2 infection in participants 
without evidence of infection up to the 
start of the asymptomatic surveillance 
periodIn participants complying with the key 
protocol criteria (evaluable 
participants):
100 × (1 –IRR) [ratio of active vaccine 
to placebo]Incidence of asymptomatic 
SARS -CoV -2 infection per 1000 
person -years of follow -up based on 
central laboratory –confirmed NAAT in 
participants with no serological or 
virological evidence (up to the start of 
the asymptomatic surveillance period) 
of past SARS -CoV -2 infection
Secondary Immunogenicity
Todemonstrate the noninferiority of 
the immune response to prophylactic 
BNT162b2 in participants 12 to 15 
years of age compared to participants 
16 to 25 years of ageGMR, estimated by the ratio of the 
geometric mean of SARS -CoV -2 
neutralizing titers in the 2 age groups 
(12-15 years of age to 16 -25 years of 
age) 1 month after completion of 
vaccinationSARS -CoV -2 neutralizing titers in 
participants with no serological or 
virological evidence (up to 1 month
after receipt of the second dose) of past 
SARS -CoV -2 infection
BNT162b2 -experienced participants
Todemonstrate the noninferiority of 
the anti-SA immune response after a 
third dose of BNT162b2 compared to 
the anti –reference strain immune 
response after 2 doses of BNT162b2, in 
the same individuals GMR of SA NT 1month after the third 
dose of BNT162b2 to the reference 
strain NT 1 month after the second 
dose of BNT162b2
The difference in percentages of 
participants with seroresponse to the 
SAstrain at 1 month after the third 
dose of BNT162b2 and seroresponse to 
the reference strain at 1 month after the 
second dose of BNT162b2SARS -CoV -2 SA and reference strain 
NTs in participants with no serological 
or virological evidence (up to 1 month
after receipt of the third dose of 
BNT162b2 ) of past SARS -CoV-2 
infection
Todemonstrate the noninferiority of 
the anti –reference strain immune 
response after 1 dose of BNT162b2 SA
compared to after 2 doses of 
BNT162b2, in the same individuals GMR of reference strain NT 1 month 
after 1 dose of BNT162b2 SAto 1mont h 
after the second dose of BNT162b2
The difference in percentages of 
participants with seroresponse to the 
reference strain at 1 month after 1 dose 
of BNT162b2 SAand 1 month after the 
second dose of BNT162b2SARS -CoV -2 reference strain NTs in 
participants with no serological or 
virological evidence (up to 1 month
after receipt of 1 dose of BNT162b2 SA) 
of past SARS -CoV -2 infection
To descriptively compare the anti-SA
immune response after 1 dose of 
BNT162b2 SAanda third dose of 
BNT162b2GMR of SANT 1 month after 1 dose 
of BNT162b2 SAto1month after the 
third dose of BNT162b2
The difference in percentages of 
participants with seroresponse to the 
SA strain at 1 month after 1 dose of 
BNT162b2 SAand 1 month after the 
third dose of BNT162b2SARS -CoV -2 SA NT in participants 
with no serological or virological 
evidence (up to 1 month after receipt of 
1dose of BNT162b2 SAor the third 
dose of BNT162b2 ) of past 
SARS -CoV -2 infection
Todescriptively compare the anti-SA 
immune response after 2 doses of 
BNT162b2 SAand the anti –reference 
strain immune response after 2 doses of 
BNT162b2, in the same individuals GMR of SA NT 1month after the 
second dose of BNT162b2 SAto the 
reference strain NT 1 month after the 
second dose of BNT162b2
The difference in percentages of 
participants with seroresponse to the 
SA strain at 1 month after the second SARS -CoV -2 SA and reference strain 
NTs in participants with no serological 
or virological evidence (up to 1 month
after receipt of the second dose of 
BNT162b2 SA) of past SARS -CoV -2 
infection
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221097
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 16
TMF Doc ID: 98.03Table 3. List of Primary, Secondary, and Tertiary/Exploratory Objectives, 
Estimands, and Endpoints for Phase 2/3
ObjectivesaEstimands Endpoints
dose of BNT162b2 SAand seroresponse 
to the reference strain at 1 month after 
the second dose of BNT162b2
BNT162b2 -naïve participants
Todemonstrate a statistically greater 
anti-SA immune response after 2 doses 
of BNT162b2 SAcompared to after 2 
doses of BNT162b2 GMR of SA NT 1month after the 
second dose of BNT162b2 SAto 
1month after the second dose of 
BNT162b2
The difference in percentages of 
participants with seroresponse to the 
SA strain at 1 month after the second 
dose of BNT162b2 SAand 1 month after 
the second dose of BNT162b2SARS -CoV -2 SANTs in participants 
with no serological or virological 
evidence (up to 1 month after receipt of 
the second dose of BNT162b2 SAor 
BNT162b2 as appropriate) of past 
SARS -CoV -2 infection
Todescriptively compare the anti–
reference strain immune response after 
2 doses of BNT162b2 SAand after 2 
doses of BNT162b2 GMR of reference strain NT 1 month 
after the second dose of BNT162b2 SA
to 1month after the second dose of 
BNT162b2
The difference in percentages of 
participants with seroresponse to the 
reference strain at 1 month after the 
second dose of BNT162b2 SAand 1 
month after the second dose of 
BNT162b2SARS -CoV -2 reference strain NTs in 
participants with no serological or 
virolo gical evidence (up to 1 month
after receipt of the second d oseof 
BNT162b2 SAor BNT162b2 as 
appropriate ) of past SARS -CoV -2 
infection
Exploratory
To describe the efficac yof prophylactic 
BNT162b 2against confirmed 
COVID -19 occurring from 7 days after 
the second dose through the blinded 
follow -up period in participants 
without, and with and without ,
evidence of infection before 
vaccinationIn participants complying with the key 
protocol criteria (evaluable 
participants) after receipt of the second 
dose of study intervention:
100 × (1 –IRR) [ratio of active vaccine 
to placebo] COVID -19 incidence per 1000 
person -years of blinded follow -up 
based on central laboratory or 
locally confirmed NAAT
To describe the incidence of confirmed 
COVID -19 through the entire study 
follow -up period in participants who 
received BNT162b2 at initial 
randomization or subsequentlyIn participants who received 
BNT162b2 (at initial randomization or 
subsequently):
Incidence per 1000 pe rson-years of 
follow -up COVID -19 incidence per 1000 
person -years of follow -up based 
on central laboratory or locally 
confirmed NAAT
To evaluate the immune response over 
time to prophylactic BNT162b2 and 
persistence of immune response in 
participants with and without 
serological or virological evidence of 
SARS -CoV -2 infection before 
vaccinationGMC/GMT andGMFR at baseline and 
1, 6, 12, and 24 months after 
completion of vaccination Full-length S -binding or 
S1-binding IgG levels 
 SARS -CoV -2 neutralizing tit ers
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221098
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 17
TMF Doc ID: 98.03Table 3. List of Primary, Secondary, and Tertiary/Exploratory Objectives, 
Estimands, and Endpoints for Phase 2/3
ObjectivesaEstimands Endpoints
To describe the incidence of non-S 
seroconversion to SARS -CoV -2
through the entire study follow-up 
period in participants who received 
BNT162b2 at initial randomizationIn participants who received 
BNT162b2 at initial randomization:
Incidence per 1000 person -years of 
follow -upIncidence of asymptomatic 
SARS -CoV -2 infection per 1000 
person -years of follow -up based on 
N-binding antibody seroconversion in 
participants with no serological or 
virological evidence of past 
SARS -CoV -2 infection or confirmed 
COVID -19 
To describe the efficac yof prophylactic 
BNT162b 2against asymptomatic 
SARS -CoV -2 infection in participants 
with evidence of infection up to the 
start of the asymptomatic surveillance 
periodIn participants complying with the key 
protocol criteria (evaluable 
participants):
100 × (1 –IRR) [ratio of active vaccine 
to placebo]Incidence of asymptomatic 
SARS -CoV -2 infection per 1000 
person -years of follow -up based on 
central laboratory –confirmed NAAT in 
participants with serological or 
virological evidence (up to the start of 
the asymptomatic surveillance period) 
of past SARS -CoV -2 infection
To describe the serological responses to 
the BNT vaccine candidate and 
characterize the SARS -CoV -2 isolate
in cases of:
 Confirmed COVID-19
 Confirmed severe COVID -19
 SARS -CoV -2 infection without 
confirmed COVID -19 Full-length S -binding or 
S1-binding IgG levels 
 SARS -CoV -2 neutralizing titers
 Identification of SARS -CoV -2 
variant(s)
To describe the safety, 
immunogenicity, and efficacy of 
prophylactic BNT162b2 in individuals 
with confirmed stable HIV disease All safety, immunogenicity, and 
efficacy endpoints described 
above
To describe the safety and 
immunogenicity of prophylactic 
BNT162b2 in individuals 16 to 55 
years of age vaccinated with study 
intervention produced by 
manufacturing “Process 1” or “Process 
2”b AEs
 SAEs
 SARS -CoV -2 neutralizing titers
To describe the immune response to 
any VOCs not already specifiedGeometric mean NT for any VOCs not 
already specified, after any dose of 
BNT162b2 SAor BNT162b2 SARS -CoV -2NTs for any VOCs 
not already specified
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221099
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 18
TMF Doc ID: 98.03Table 3. List of Primary, Secondary, and Tertiary/Exploratory Objectives, 
Estimands, and Endpoints for Phase 2/3
ObjectivesaEstimands Endpoints
To describe thecell-mediated immune 
response, and additional humoral 
immune response parameters, to the 
reference strain and SA in a subset of 
participants:
7 Days and 1 and 6 months after
BNT162 b2SAgiven as 1 or 2 doses 
to BNT162b2-experienced
participants
7 Days and 1 and 6 months after 
BNT162b2 SAgiven as 2 doses to 
BNT162b2 -naïve participants
7 Days and 1 and 6 months after 
BNT162b2 given as a third dose to 
BNT162b2 -experienced 
participants
a.HIV-positive participants in Phase 3 will not be included in analyses of the objectives, with the exception 
of the specific exploratory objective.
b.See the protocol, Section 6.1.1, for description of the manufacturing process.
2.2.Study Design
2.2.1. Overall Design
This is a multicenter, multinational, Phase 1/2/3, randomized, placebo -controlled, 
observer -blind, dose -finding, vaccine candidate– selection, and efficacy  study  in healthy  
individuals.
The study  consists of 2 parts. Phase 1: to identify  preferred vaccine candidate(s) and dose 
level(s); Phase 2/3: an expanded cohort and efficacy part.  These parts , and the progression 
between them, are detailed in the schema ( see p rotocol, Section 1.2).
The study  will evaluate the safet y, tolerability , and immunogenicit y of 2 different 
SARS -CoV-2 R NAvaccine candidate s against C OVID -19and the efficacy  of 1 candida te:
As a 2 -dose (separated by  21 day s) schedule ;
At various different dose levels in Phase 1;
As a booster;
In 3 age groups ( Phase 1: 18 to 55 y ears of age, 65 to 85 y ears of age ; 
Phase 2/3: ≥12years of age [stratified as 12 -15, 16 -55, or >55 years of ag e]).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221100
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 19
TMF Doc ID: 98.03Dependent upon safet y and/or immunogenicit y data generated during the course of this 
study , or the BioNTech study  conducted in Germany  (BNT162 -01), it is possible that groups 
in Phase 1 may  be started at the next highest dose , groups may  not be started , groups may  be 
terminated early , and/or groups may  be added with dose levels below the lowest stated dose 
or intermediate between the lowest and highest stated doses.
The study  is observer -blinded, as the ph ysical appearance of the investigational vaccine 
candidates and the placebo may  differ. The participant, investigator, study  coordinator, and 
other site staff will be blinded.  At the study  site, only  the dispenser(s)/administrator(s) are 
unblinded.
To facilitate rapid review of data in real tim e, sponsor staff will be unblinded to vaccine 
allocation for the participants in Phase 1.
In order to describe the boostability  of BNT162, an additional dose of BNT162b2 at 30 µg 
will be given to Phase 1 participants approximately 6 to 12 months after their second dose of 
BNT162 b1 or BNT162b2. This will provide an early  assessment of the safety  of a third dose 
of BNT162, as well as its immunogenicit y.The assessment of boostability  will be further 
expanded in a subset of Phase 3 participants who will receive a third dose of BNT162b2 or a 
third and potentially  a fourth dose of protot ype BNT162b2 VOC(based upon the South African 
variant and hereafter referred to as BNT162b2 SA).  To further describe potential homologous 
and heterologous protection against emerging SARS -CoV -2 VOCs, a new cohort of 
participants will be enrolled who are COVID -19 vaccine– naïve ( ie, BNT162b2 -naïve) and 
have not experienced COVID-19. They  will receive BNT162b 2SAgiven as a 2-dose series, 
separated b y 21 days .
2.2.2. Phase 1
Each group (vaccine candidate/dose level/age group) will comprise 15 participant s; 
12participants will be randomized to receive active vaccine and 3 to receive placebo.
For each vaccine candidate/dose level/age group, the following apply:
Additional safet yassessme nts (see protocol, Section 8.2).
Controlled enrollment (required only  for the first candidate and/or dose level studied):
No more than 5 participants (4 active, 1 placebo) can be vaccinated on the first 
day.
Thefirst 5 participants must be observed by  blinded site staff for at least 4 hours 
after vaccination for an y acute reactions.
Vaccination of the remaining participants will commence no sooner than 24hours 
after the fifth participant received his or her vaccination.
Application of stopping rules .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221101
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 20
TMF Doc ID: 98.03IRC review of safet y data to determine escalation to the next dose level in the 18 -to 
55-year age cohort:
Escalation between dose levels will be based on IRC review of at least 7 -day 
post–Dose 1 safet y data in this study  and/or the BioNTech study  conducted in 
German y (BNT162 -01).
Note that, since both candidates are based upon the same RNA platform, dose 
escalation for the second candidate studied may  be based upon the safet y profile 
of the first candidate studied being deemed acceptable at the same, or a hi gher, 
dose level b y the IRC .
Groups of participants 65 to 85 y ears of age will not be started until safet y data for the RNA 
platform have been deemed acceptable at the same, or a higher, dose level in the 18 -to 
55-year age cohort by the I RC.
In this phase, 13groups will be studied, corresponding to a total of 195participants.
The I RC will select 1 vaccine candidate that, in Phase 1, has an established dose level per age 
group based on induction of a post–Dose 2 immune response, including neutralizing 
antibodies, which is expected to be associated with protection against COVID -19, for
progress ion into Phase 2/ 3.
Participants who originally  received placebo and become eligible for receipt of BNT162b2 or 
another COVID -19 vaccine according to recommendations detailed separately , and available 
in the electronic stud y reference portal ,will have the opportunity  to receive BNT162b2 in a 
phased manner as part of the study . The investigator will ensure the participant meets at least 
1 of the recommendation criter ia. 
Any Phase 1 placebo recipient who has not already  been offered the opportunity  to receive 
BNT162b2 will be given this opportunity  no later than at the approximate time participants in 
Phase 2/3 reach Visit 4.
Any participant who originally  received pl acebo but then goes on to receive BNT162b2 will 
move to a new visit schedule (p rotocol ,Section 1.3.3).
In order to describe the boostability  of BNT162, and potential heterologous protection 
against emerging SARS -CoV -2 VOCs, an additional dose of BNT162b2 at 30 µg will be 
given to Phase 1 participants approximately  6to 12 months after their second dose of 
BNT162 b1 or BNT162b2.
Phase 1 participants who originally received BNT162b1 or BNT162b2 at dose levels of 10, 
20, or 30 µg at Doses 1 and 2 will be offer ed an additional dose of BNT162b2 at 30 µg
approximately  6to 12 months after their second dose of BNT162. 
Participants are expected to participate for up to a maximum of approximately  26 months.  
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221102
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 21
TMF Doc ID: 98.032.2.3. Phase 2/3
On the basis of safet y and/or immunogenicit y data generated during the course of this study , 
and/or the BioNTech study  conducted in Germany  (BNT162 -01), 1 vaccine candidate was 
selected to proceed into Phase 2/3.  Participants in this phase will be ≥12 y ears of age, 
stratified as follows: 12 to 15 y ears, 16 to 55 years, or >55 years. The 12- to 15 -year stratum 
will comprise up to approximately  2000 participants enrolled at selected investigational sites. 
It is intended that a minimum of 40% of participants will be in the >55 -year stratum.  
Commencement of each age stratum will be based upon satisfactory post –Dose 2 safety  and 
immunogenicit y data from the 18 -to 55 -year and 65 -to 85 -year age groups in Phase 1, 
respectivel y.  The vaccine candidate selected for Phase 2/3 evaluation is BNT162b2 at a dose 
of 30µg.
Phase 2/ 3 is event -driven. Under the assumption of a true VE rate of ≥60%, after the second
dose of study  intervention , a target of 164 primary -endpoint cases of confirmed COVID -19 
due to SARS -CoV -2 occurring at least 7days following the second dose of the primary  series 
of the candidate vaccine will be sufficient to provide 90% power to conclude true VE >30% 
with high probability . The total number of participants enrolled in Phase 2/ 3 may  vary  
depending on the incidence of COVID -19 at the time of the enrollment, the true underl ying 
VE, and a potential earl y stop for efficacy  or futility .
Assuming a COVID -19 attack rate of 1.3% per y ear in the placebo group, accrual of 164 first 
primary -endpoint cases within 6 months, an estimated 20% nonevaluab le rate, and 1:1 
randomization, the BNT162b2 vaccine candidate selected for Phase 2/3 is expected to 
comprise approximately 21,999 vaccine recipients.   This is the number of participants 
initially  targeted for Phase 2/3 and may  be adjusted based on advice from DMC analy ses of 
case accumulation and the percentage of participants who are seropositive at baseline.  
Dependent upon the evolution of the pandemic, it is possible that the COVID -19 attack rate 
may be much higher, in which case accrual would be expec ted to be more rapid, enabling the 
study ’s primary  endpoint to be evaluated much sooner.
The first 360 participants enrolled (180 to active vaccine and 180 to placebo, stratified 
equally  between 18 to 55 years and >55 to 85 years) will comprise the “Phase 2” portion.  
Safety  data through 7 day s after Dose 2 and immunogenicity  data through 1 month after 
Dose 2 from these 360 participants will be anal yzed by  the unblinded statistical team, 
reviewed b y the DMC, and submitted to appropriate regulatory  authorities for review. 
Enrollment may  continue during this period and these participants would be included in the 
efficacy  evaluation in the “Phase 3” portion of the study .
In Phase 3, up to approximately  2000 participants, enrolled at selected sites, are anticip ated 
to be 12 to 15 y ears of age.  Noninferiority  of immune response to proph ylactic BNT162b2 
in participants 12to 15 years of age to response inparticipants 16 to 25 years of age will be 
assessed based on the GM Rof SARS -CoV -2 neutralizing titers using a 1.5 -fold margin. A 
sample size of 225 evaluable participants (or 280vaccine recipients) per age group will 
provide a power of 90.4 % to declare the noninferiority in terms of GM R(lower limit of 95%
CI forGMR>0.67) . A random sample of 280 participant s from each of the 2 age groups 
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221103
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 22
TMF Doc ID: 98.03(12to 15 y ears and 16 to 25 y ears) will be selected as an immunogenicity  subset for the 
noninferiority  assessment.
The initial BNT162b2 was manufactured using “Process 1”; however, “Process 2” was 
developed to support an in creased scale of manufacture. In the stud y, each lot of “Process 
2”-manufactured BNT162b2 will be administered to approximately  250 participants 16 to 55 
years of age. The safety  and immunogenicity of prophylactic BNT162b2 in individuals 16 to 
55 years o f age vaccinated with “Process 1” and each lot of “Process 2” study  intervention 
will be described. A random sample of 250 participants from those vaccinated with study  
intervention produced b y manufacturing “Process 1” will be selected for this descripti ve 
analysis.
For evaluation of boostability  and protection against emerging VOCs, 600 existing Phase 3 
participants 18 to 55 y ears of age will be rerandomized in a 1:1 ratio to receive either a third 
dose of BNT162b2 or a third dose of BNT162b2 SA.
An addit ional group of 30 existing Phase 3 participants 18 to 55 y ears of age will be enrolled 
to receive a third and fourth dose of BNT162b2 SA.For these 30 participants, through 1 
month after their first dose of BNT162b2 SAthe participant swill be blinded to th eir vaccine 
allocation ,but the investigator and sponsor will not be. Serum samples from these 
participants may  be used for assay  development purposes and, except for objectives relating 
to response to a fourth dose, their results will be analy zed separat ely from the main 
immunogenicit y anal yses.
Three hundred participants 18 to 55 y ears of age who are COVID -19 vaccine –naïve 
(ie, BNT162b2-naïve) and have not experienced COVID-19 will be enrolled as a new cohort 
of participants toreceive BNT162b2 SAgiven as a 2- dose series .
Participants are expected to participate for up to a maximum of approximately 26 months. 
The duration of stud y follow -up may  be shorter among participants enrolled in Phase 1 
dosing arms that are not evaluated in Phase 2/ 3. 
Participants ≥16 y ears of age who originall y received placebo and become eligible for receipt 
of BNT162b2 according to recommendations detailed separately , and available in the 
electronic stud y reference portal ,will have the opportunity  to receive BNT162b2 i n a phased 
manner as part of the study . The investigator will ensure the participant meets at least 1 of 
the recommendation criteria.
Any Phase 2/3 placebo recipient ≥16 y ears of age who has not already  been offered the 
opportunity  to receive BNT162b2 wil l be given this opportunity  no later than 6 months after 
Vaccination 2 (at the time of the originall y planned Visit 4).
Any participant who originally  received placebo but then goes on to receive BNT162b2 will 
move to a new visit schedule (p rotocol , Sectio n 1.3.3).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221104
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 23
TMF Doc ID: 98.03The changes to the protocol as part of protocol amendment 14 to assess boostability  and 
homologous/heterologous protection against emerging VOCs allow the evaluation of safet y 
and immunogenicit y of BNT162b2 SA:
When given as a third dose to C4591001 Phase 3 participants who received a second dose 
of BNT162b2 approximately  6 months previousl y (ie, BNT162b2 -experienced) and have 
not experienced COVID -19. 
In a small separate group of individuals who previously  received 2 doses of BNT162b2 
followed b y1 dose of BNT162b2 SA, a second BNT162b2 SAdose will also be given 1 
month after Dose 1 of BNT162b2 SA.
When given as a 2 -dose series, separated b y 21 days, in newl y recruited participants who 
are COVID -19 vaccine– naïve (ie, BNT162b2 -naïve) and have not experienced 
COVID -19.
In addition, a group of C4591001 Phase 3 participants who received a second dose of 
BNT162b2 approximately  6 months previousl y will receive a third dose of BNT162b2.
This approach will allow an evaluation of immunogenicity against the re ference ancestral 
SARS -CoV -2 strain (Wuhan -Hu-1/USA -WA1) and the selected South African VOC, using a 
noninferiority  approach based on neutralizing antibody  titers in prior BNT162b2 vaccinees 
who receive either a homologous boost (with BNT162b2) or a hetero logous boost (with 
BNT162b2 SA), as well as new vaccinees receiving 2 doses of BNT162b2 SA.
An intensive period of surveillance to evaluate the efficacy  of BNT162b2 against 
asymptomatic SARS -CoV -2 infection may  be conducted at selected sites among Phase 2/3 
participants following approval of protocol amendment 11. After an initial in -person visit 
where a blood sample will be collected and a nasal (midturbinate) swab obtained, nasal 
(midturbinate) swabs will be obtained from consented participants every  2 weeks until 
Visit 4, or a sufficient number of cases of SARS -CoV -2 infection have accrued to evaluate 
this objective, whichever is sooner, per the SoA in the protocol , Section 1.3. 6. The swabs 
will be tested at a central laboratory  using NAAT to detect SARS -CoV -2. Participants who 
are unblinded because they  become potentially  eligible for receipt of BNT162b2 according to 
recommendations detailed separately , and available in the electronic study  reference portal, 
will not participate in surveillance for as ymptomatic SARS -CoV -2 infection. However, 
participants who provided additional consent to conduct biweekl y swabbing for surveillance 
of as ymptomatic infection should continue to swab even after unblinding if they  originall y 
received BNT162b2.
Surveillance for as ymptomatic SARS -CoV -2 infection (swabbing) should ceas e in 
participants enrolled into the subset of participants who will receive an additional dose of 
BNT162b2 or BNT162b2 SA.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221105
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 24
TMF Doc ID: 98.033.ENDPOINTS AND BASELINE VARIAB LES: DEFINITIONS AND 
CONVENTIONS
3.1.Primary Endpoints
3.1.1. Safety Endpoints
For all participants in Phase 1, a subset of at least 6000 participants randomized in Phase 2/3,
receiving at least 1 dose of study  intervention, BNT162b2- experienced participants receiving 
1 or 2 doses of BNT162b2 SA, BNT162b2 -naïve participants receiving 2 doses of 
BNT162b2 SA, and BNT162 b2-experienced participants receiving the third dose of 
BNT162b2 in Phase 3, below are the primary  safety  endpoints for local reactions and
systemic events :
Local reactions (pain at the injection site , redness ,and swelling ) within 7 day s after 
each dose in each vaccine group .
Systemic events ( fever, fatigue, headache, chills, vomiting, diarrhea, new or worsened 
muscle pain, and new or worsened joint pain ) within 7 day s after each dose in each 
vaccine group .
For all participants randomized in Phase 1 and P hase 2/3, receiving at least 1 dose of study  
intervention, below are the primary  safet y endpoints for AEs and SAEs (the last dose in 
Phase 1 is the second dose):
AEs from Dose 1 to 1 month after the second dose.
SAEs from Dose 1 to 6 months after the secon ddose.
In addition, f or the first 360 participants randomized in Phase 2 /3 (Phase 2 portion) , receiving 
at least 1 dose of stud y intervention, below are the primary  safety  endpoints for AEs and 
SAEs :
AEs from Dose 1 to 7 day safter the second dose.
SAEs from Dose 1 to 7 day safter the second dose.
Last, f or the BNT162b2 -experienced participants receiving 1 or 2 doses of BNT162b2 SA, 
BNT162b2 -naïve participants receiving 2 doses of BNT162b2 SA, and BNT162b2-
experienced participants receiving the third d ose of BNT162b2 in Phase 3, below are the 
primary  safety  endpoints for AEs and SAEs:
AEs from Dose 1 to 1 month after the last dose.
SAEs from Dose 1 to 5 or 6 months after the last dose.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221106
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 25
TMF Doc ID: 98.033.1.1.1. Local Reactions
The local reactions assessed and reported in the e -diary are redness, swelling, and pain at the 
injection site, from Day 1 through Day 7 after each dose, where Day  1 is the day  of each 
dose. This section describes derivations with details for the assessment of local reactions: 
presence, severity  level, dur ation, and onset day .
Presence or Absence
For the data summary  of the presence (y es or no) of a local reaction during the interval from 
Day 1 through Day  7 for each dose , where Day  1 is the day  of each dose , the following 
variables are required in order to compute the proportions:
Presence (y es or no) of each severe/ Grade 4 local reaction on each day  and any  day 
(Day  1 through Day  7);
Presence (y es or no) of each local reaction by maximum se verity on an y day (Day 1 
through Day  7).
For each local reaction an d any  local reaction on any  day,Table 4explains the algorithm to 
derive the presence of a reaction (y es or no) during the interval from Day  1 through Day  7, 
where Day 1 is the day  of each dose .
Table 4. D erived Variables for Presence o f Each and Any Local Reaction Within 
7Days for Each Dose
VariableaYes (1) No (0) Missing (.)
Presence of each local 
reaction .Participant reports the
reaction as “yes” onany
day ( Day 1 through
Day7).Participant reports the 
reaction as “no” on all 
7 days ( Day 1 t hrough
Day 7) or as a 
combination of “no” and 
missing on all 7 days 
(Day 1 through Day 7).Participant does not report 
anydata on all 7 days (Day 1 
through Day 7) for the 
reaction.
Presence of any local
reaction .Participant reports any 
local reaction as “yes” on 
any day (Day 1 t hrough
Day 7).For all 3 local reactions, 
participant reports “no” 
on all 7 days(Day 1 
through Day 7) or as a 
combination of “no” and 
missing on all 7 days 
(Day 1 through Day 7).Participant does not report any 
data for all 3 local reactions on 
all 7 days ( Day 1 t hrough 
Day 7).
a.The variables will be derived for each and any of the local reactions (redness, swelling, and pain at the 
injection site) and for each and any of the severe local reactions within the interval from Day 1 through 
Day 7 after each dose .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221107
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 26
TMF Doc ID: 98.03Severity and Maximum Severity
Redness and swelling will be measured and recorded in measuring device units 
(range: 1to21) and then categorized during anal ysis as absent, mild, moderate, or severe 
based on the grading scale in Table 5. Measuring device units can be converted to 
centimeters according to the following formula: 1 measuring device unit = 0.5 cm.  Pain at 
the injection site will be assessed b y the participant as absent, mild, moderate, or severe 
according the grading scale in Table 5.
Table 5.Local Reaction Grading Scale
Mild
(Grade 1)Moderate
(Grade 2)Severe
(Grade 3)Potentially Life 
Threatening
(Grade 4)
Pain at the 
injection siteDoes not interfere 
with activity .Interferes with 
activity .Prevents daily 
activity .Emergency room 
visit or 
hospitalization for 
severe pain.
Redness >2.0cm to 5.0 cm
(5 to 10 measuring 
device units) .>5.0 cm to 10.0 cm 
(11 to 20 measuring 
device units) .>10cm 
(≥21measuring 
device units) .Necrosis or 
exfoliative 
dermatitis .
Swelling >2.0cm to 5.0 cm 
(5 to 10 measuring 
device units) .>5.0 cm to 10.0 cm 
(11 to 20 measuring 
device units) .>10cm 
(≥21measuring 
device units) .Necrosis .
For each local reaction reported for each dose , the maximum severity  grade will be derived 
for the e -diary  collection period (Day  1 through Day  7, where Day  1 is the day  of each dose ) 
as follows:
maximum severity  grade = highest grade (maximum severity ) within 7days after 
vaccination (Day  1 through Day  7) among severity  grades where the answers are 
neither “no” nor missing for at least 1 day  during the interval from Day  1 through 
Day 7.
Duration (First to Last Day Reported)
For participants experiencing any  local reactions (or those with a derived reaction as 
described in Table 5), the maximum duration (last day  of reaction –first day  of reaction + 1) 
will be derived for each study  vaccination.   Resolution of the reaction is the last day  on 
which the reaction is recorded in the e -diary  or the date the reaction ends if it is unresolved 
during the participant e-diary recording period (end date collected on the CRF), unless 
chronicity is established.  If there is no known end date, the duration will be considered 
unknown and set to missing. However, if a reaction is ongoing at the time of a subsequent 
vaccination, the end date/day for the ongoing reaction would be the date/day that the next 
vaccine is administered ,which will be used for the duration computation. Participants with 
no reported reaction have no dur ation.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221108
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 27
TMF Doc ID: 98.03Onset Day
The onset day  of each local reaction will be derived. Onset day  is defined as the first day  of 
reporting an y severit y.
For the onset day  of each local reaction, if participant s report change in severity  of the local 
reaction, onl y the fi rst day  of reporting that specific local reaction will be counted.
3.1.1.2. Systemic Events ( Systemic Event Symptoms and Fever) 
The sy stemic events assessed and recorded in the e-diary  are vomiting, diarrhea, headache, 
fatigue, chills, new or worsened muscle pain, and new or worsened joint pain from Day 1 
through Day 7, where Day  1 is the day of each dose .  The derivations for s ystemic events will 
be handled in a way  similar to the way  local reactions are handled for presence of event, 
severit y level, duration, and onset day .
The variables associated with the sy stemic events will be computed in a way  similar to the 
way local reactions are computed ( seeSection 3.1.1.1 ). Maximum temperature range over 
the period from Day  1 through Day  7 will be mapped into the ranges described in Table 7for 
summary  of maximum temperature.
The symptoms will be assessed by  the participant as absent, mild, moderate, or severe 
according to the grading scale in Table 6.
Table 6.Systemic Event Grading Scale
Mild
(Grade 1)Moderate
(Grade 2)Severe
(Grade 3)Potentially Life 
Threatening
(Grade 4)
Vom iting 1-2 times in 
24hours .>2 times in 
24hours .Requires IV 
hydration .Emergency room visit 
or hospitalization for 
hypotensive shock .
Diarrhea 2 to 3 loose stools 
in24 hours .4 to 5 loose stools 
in 24 hours .6 or more loose 
stools in 24 hours .Emergency room visit 
or hospitalization for 
severe diarrhea .
Headache Does not interfere 
with activity .Some interference 
with activity .Prevents daily 
routine activity .Emergency room visit 
or hospitalization for 
severe headache .
Fatigue Does not interfere 
with activity .Some interference 
with activity .Prevents daily 
routine activity .Emergency room visit 
or hospitalization for 
severe fatigue .
Chills Does not interfere 
with activity .Some interference 
with activity .Prevents daily 
routine activity .Emergency room visit 
or hospitalization for 
severe chills .
New or worsened 
muscle painDoes not interfere 
with activity .Some interference 
with activity .Prevents daily 
routine activity .Emergency room visit 
or hospitalization for 
severe ne w or worsened 
muscle pain .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221109
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 28
TMF Doc ID: 98.03Table 6.Systemic Event Grading Scale
Mild
(Grade 1)Moderate
(Grade 2)Severe
(Grade 3)Potentially Life 
Threatening
(Grade 4)
New or worsened 
joint painDoes not interfere 
with activity .Some interference 
with activity .Prevents daily 
routine activity .Emergency room visit 
or hospitalization for 
severe ne w or worsened 
joint pain .
Abbreviation: IV = intravenous.
Oral temperature will be collected in the evening, daily ,for 7 day s following each dose
(Day s 1 through 7, where Day  1 is the day  of each dose ) and at an y time during the 7 days 
that fever is suspected. Fever is defined as an oral temperature of ≥38.0°C (100.4 °F).  The 
highest temperature for each day  will be recorded in the e -diary .
Temperature will be measured and recorded to 1 decimal place. Temperatures recorded in 
degrees Fahrenheit will be programmatically  converted to degrees Celsius for reporting.
Temperatures <35.0 °C and >42.0°C will be excluded from the anal ysis.  Fever will be 
grouped into ranges for the anal ysis according to Table 7below.
Table 7.Scale for Fever
≥38.0°C to 38 .4°C(100.4 °F to 101.1 °F)
>38.4°C to 38.9 °C(101.2 °F to 102.0 °F)
>38.9 °C to 40.0 °C(102.1 °F to 104.0 °F)
>40.0 °C(>104.0 °F)
Note: Fever is defined as temperature ≥38.0 °C (≥100.4° F).
3.1.1.3. Use of Antipyretic Medication
The use of antip yretic medication is also recorded in the e -diary  from Day  1 through Day 7, 
where Day  1 is the day  ofeach dose .  For the use of antipy retic medication from Day 1 
through Day  7after each dose , the following endpoints and variables will be derived for 
analysis following the sa me rules as for local reactions ( see Section 3.1.1.1 ),where 
applicable.
Presence (y es or no) of use of antipy retic medication on ea ch day  (Day  1 through Day  7);
Presence (y es or no) of use of antipy retic medication on any  day (Day  1 through Day  7);
Duration (first to last day reported) of use of antipyretic medication; 
Onset day  of use of antipy retic medication.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221110
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 29
TMF Doc ID: 98.03The use of antip yretic medication will be summarized and included in the sy stemic event 
summary  tables but will not be considered a s ystemic event.
3.1.1.4. Adverse Events
AEs will be assessed from the time of informed consent through 1 month after the second 
dose or 1 month after the last dose for thesubset for evaluation ofboostability andprotection
against emerging VOCs .
The primary  endpoint s “AEs from Dose 1 to1 month after the second dose”and“AEs from 
Dose 1 to 1 month after the last dose,” forevaluatio n of boostability  and protection against 
emerging VOCs ,and other AE endpoints will be summarized by  SOC and PT at the 
participant level. For the subset for evaluation ofboostability andprotection against
emerging VOCs, Dose 1 refers to the first dose of BNT162b2 SAor first dose of BNT162b2 
booster .
Theseprimary  endpoint swill be supported by  summaries and listings of related AEs, severe 
AEs, and immediate AEs (within the first 30 minutes after each dose ).
AE reporting will be based on the specific rep orting period. Standard algorithms for 
handling missing AE dates will be applied as described in the Pfizer Vaccine data standard 
rules.
For Phase 2/
3 only, a 3-tier approach will be used to summarize AEs.  Under this approach, 
AEs are classified into 1 o f 3 tiers . Different analy ses will be performed for different tiers :
Tier 1 events: These are prespecified events of clinical importance and are identified 
in a list in the product’s Safety Review Plan .
Tier 2 events: These are events that are not Tier 1 but are considered “relatively  
common .”A MedDRA PTis defined as a Tier 2 event if there are at least 1%
participant s with the AE term in at least 1vaccine group.
Tier 3 events: These are events that are neither Tier 1 nor Tier 2.
3.1.1.5. Serious Adverse Events 
SAEs will be collected from the time the participant provides informed consent to 
approximately  6 months after the second dose of study  intervention (Visit 8 for Phase 1 
participants and Visit 4 for Phase 2/3 participants).
For BNT162b2 -experienced participants in the subset for evaluation of boostability  and 
protection against emerging VOCs , SAEs will be collected from the time the participant 
provides informed consent (for participation in the subset) through and including Visit 306 
(5 or 6 months after the last dose, depending upon group).
For BNT162b2 -naïve participants inthe subset for evaluation of protection against emerging 
VOCs, SAEs will be collected from the time the participant provides informed consent 
through and including Visit 405 (6 months after the second dose).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221111
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 30
TMF Doc ID: 98.03The safet yendpoint s“SAEs from Dose 1 to 6 months after the second dose”and“SAEs from 
Dose 1 to 5 or 6 months after the last dose” for evaluation of boostability  and protection 
against emerging VOCs will be summarized by SOC and PT at the participant level. For the 
subset for evaluation of boostability  and protection against emerging VOCs, Dose 1 refers to 
the first dose of BNT162b2 SA or first dose of BNT162b2 booster.
3.1.1.6. Hematology and Chemistry Laboratory Param eters(for Phase 1 O nly)
For participants in Phase 1, below are the additional primary  safet y endpoints:
Abnormal hematology  and chemistry  laboratory  values 1 and 7 day s after Dose 1; and 
7 day s after Dose 2.
Grading shifts in hematology  and chemistry  laboratory  assessments between baseline 
and 1 and 7 days after Dose 1; and before Dose 2 and 7 day s after Dose 2.
The following safet y laboratory  tests will be performed at the times defined in the protocol,
Section 1.3 (schedule of activities) . Additional laboratory  results may  be reported on these 
samples as a result of the method of anal ysis or the ty pe of analy zer used by  the clinical 
laboratory , or as derived from calculated values. These additional tests would not require 
additional collection of blood.  Unschedule d clinical laboratory  measurements may  be 
obtained at an y time during the stud y to assess any perceived safety issues.
Hem atology Chemistry
Hem oglobin
Hem atocrit
RBC count
MCV
MCH
MCHC
Platelet count
WBC count
Total neutrophils (Abs)
Eosinophils (Abs)
Monocytes (Abs)
Basophils (Abs)
Lymphocytes (Abs)BUN and creatinine
AST, ALT
Total bilirubin
Alkaline phosphatase
Clinically  significant abnormal laboratory  findings should be recorded in the AE CRF in 
accordance with the following grading scale ( Table 8).Additionally , the primary  criterion 
for abnormalit y will follow the Pfizer safet y rule book.
Table 8.Laboratory Abnormality Grading Scale
Hem atology Mild
(Grade 1) Moderate 
(Grade 2) Severe
(Grade 3) Potentially Life 
Threatening 
(Grade 4) 
Hem oglobin 
(Fem ale) -g/dL 11.0 –12.0 9.5 –10.9 8.0 –9.4 <8.0 
Hem oglobin 
(Male) -g/dL 12.5 –13.5 10.5 –12.4 8.5 –10.4 <8.5 
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221112
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 31
TMF Doc ID: 98.03Table 8.Laboratory Abnormality Grading Scale
WBC increase -
cells/mm310,800 –15,000 15,001 –20,000 20,001 –25,000 >25,000 
WBC decrease -
cells/mm32500 –3500 1500 –2499 1000 –1499 <1000 
Lymphocytes 
decrease -cells/mm3750 –1000 500 –749 250 –499 <250 
Neutrophils decrease 
-cells/mm31500 –2000 1000 –1499 500 –999 <500 
Eosinophils -
cells/mm3650 –1500 1501 –5000 >5000 Hypereosinophilic 
Platelets decreased -
cells/mm3125,000 –140,000 100,000 –124,000 25,000 –99,000 <25,000 
Chemistry Mild (Grade 1) Moderate 
(Grade 2) Severe 
(Grade 3) Potentially Life 
Threatening 
(Grade 4) 
BUN - mg/dL 23 –26 27 –31 >31 Requires dialysis 
Creatinine -mg/dL 1.5 –1.7 1.8 –2.0 2.1 –2.5 >2.5 or requires 
dialysis 
Alkaline phosphate -
increase by factor 1.1 –2.0 ×ULN 2.1 –3.0 ×ULN 3.1 –10 ×ULN >10 ×ULN 
Liver function tests -
ALT, AST 
increase by factor 1.1 –2.5 ×ULN 2.6 –5.0 ×ULN 5.1 –10 ×ULN >10 ×ULN 
Bilirubin -when 
accompanied 
by any increase in 
liver function test -
increase by factor 1.1 –1.25 ×ULN 1.26 –1.5 ×ULN 1.51 –1.75 ×ULN >1.75 ×ULN 
Bilirubin -when 
liver function test is 
norm al -increase by 
factor 1.1 –1.5 ×ULN 1.6 –2.0 ×ULN 2.0 –3.0 ×ULN >3.0 ×ULN 
Abbreviations: ALT = alanine aminotransferase; AST = aspartate aminotransferase; BUN = blood urea 
nitrogen; ULN = upper limit of normal; WBC = white blood cell.
3.1.2. Immunogenicity Endpoints (for the Phase 2/3 Subset for E valuation of 
Boostability and P rotection Against E merging VOCs Only)
SARS -CoV -2 reference strain NTs.
SARS -CoV -2 SA NTs.
In order to allow direct comparability with the reference strain, the anti -SA NTs may  be 
adjusted to account for intrinsic variant or assay  characteristics.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221113
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 32
TMF Doc ID: 98.03Titers (and IgG concentrations , secondary  and exploratory  endpoints) above the LLOQ are 
considered accurate and their quantitated values will be reported. Values below the LLOQ, 
denoted as BLQ, will be set to 0.5 × LLOQ for analysis.  However, this calculation may  be 
adjusted based upon additional data from the assay .  LLOQ results will be included in the 
analysis specification once they  are available .
3.1.3. Vaccine Efficacy Endpoints (for Phase 2/ 3 Only)
COVID -19 incidence per 1000 person- yearsof follow -up based on central laboratory  
or locally  confirmed NAAT in participants with no serological or virological 
evidence ( prior to 7 day s after receipt of the second dose)of past SARS -CoV -2 
infection (counting cases from 7 day s after the second dose).
COVID -19 incidence per 1000 person- years of follow -up based on central laboratory  
or locally  confirmed NAAT (counting cases from 7 day s after the second dose ).
3.2. Secondary Endpoints
3.2.1. Immunogenicity Endpoints
Phase 1
In participants compl ying with the key protocol criteria (evaluable participants) at the 
following time points after receipt of stud y intervention:
7 and 21 day s after Dose 1; 7 and 14 day s and 1, 6, 12, and 24 months after Dose 2.
Below are the secondary  immunogenicit y endpoints for Phase 1:
SARS -CoV -2 neutralizing titers .
S1-binding IgG levels .
RBD- binding IgG levels .
Phase 2/3
Participants 12 to 15 y ears of age and 16 to 25 years of age:
SARS -CoV -2 neutralizing titers.
Participants in the subset for evaluation of boostability  and protection against emerging
VOCs :
SARS -CoV -2 reference strain NTs.
SARS -CoV -2 SA NTs. 
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221114
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 33
TMF Doc ID: 98.03In order to allow direct comparability with the reference strain, the anti -SA NTs may  be 
adjusted to account for intrinsic variant or assay  characteristics .
3.2.2. Vaccine Ef ficacy Endpoints (for Phase 2/3 Only)
COVID -19 incidence per 1000 person- years of follow -up based on central laboratory  
or locally  confirmed NAAT in participants with no serological or virological 
evidence ( prior to 14days after receipt of the second dose ) of past SARS -CoV -2 
infection (counting cases from 14 day s after the second dose) .
COVID -19 incidence per 1000 person- years of follow -up based on central laboratory  
or locally  confirmed NAAT (counting cases from 14 day s after the second dose) .
Confirmed s evere COVID -19 incidence per 1000 person -years of follow -up in 
participants with no serological or virological evidence (prior to 7 day s and prior to 
14 day s after receipt of the second dose) of past SARS -CoV -2 infection (counting 
cases from 7 day s and 14 day s after the second dose ).
Confirmed severe COVID- 19 incidence per 1000 person -years of follow -up(counting 
cases from 7 day s and 14 day s after the second dose) .
According to the CDC -defined s ymptoms, COVID -19 incidence per 1000
person -years of follow -up based on central laboratory  or locall y confirmed NAAT in 
participants with no serological or virological evidence ( prior to 7 day s and prior to 
14 day s after receipt of the second dose) of past SARS -CoV -2 infect ion(counting 
cases from 7 day s and 14 day s after the second dose) .
According to the CDC -defined s ymptoms, COVID -19 incidence per 1000
person -years of follow -up based on central laboratory  or locall y confirmed NAAT
(counting cases from 7 day s and 14 day s after the second dose) .
Incidence of as ymptomatic SARS -CoV -2 infection per 1000 person -years of 
follow -up based on N -binding antibod y seroconversion in participants with no 
serological or virological evidence of past SARS -CoV -2 infection or confirmed 
COVID -19.
Incidence of as ymptomatic SARS -CoV -2 infection per 1000 person -years of 
follow -up based on central laboratory –confirmed NAAT in participants with no 
serological or virological evidence (up to the start of the as ymptomatic surveillance 
period) of past SARS -CoV -2 infection.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221115
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 34
TMF Doc ID: 98.033.3.Exploratory Endpoints 
3.3.1. Safety Endpoints (for Phase 1 Boostability Assessment Only)
Local reactions (pain at the injection site, redness, and swelling) for up to 7 day s after 
Dose 3.
Systemic events (fever, fatigue, headache, chills, vomiting, diarrhea, new or worsened 
muscle pain, and new or worsened joint pain) for up to 7 day s after Dose 3 .
AEs from Dose 3 to 1 month after Dose 3.
SAEs from Dose 3 to 1 month after Dose 3 .
3.3.2. Vaccine Efficacy Endpoints (for Phase 2/3 Only)
COVID -19 inci dence per 1000 person- years of blinded follow -up based on central 
laboratory  or locally  confirmed NAAT in participants without, and with and without, 
evidence of infection (counting cases from 7 day s after the second dose) .
COVID -19 incidence per 1000 pers on-years of follow -up based on central laboratory  
or locally  confirmed NAAT in participants who received BNT162b2 at initial 
randomization or subsequently  (counting cases from 7 day s after the second 
BNT162b2 vaccination ).
Incidence of as ymptomatic SARS -CoV-2 infection per 1000 person -years of 
follow -upbased on N -binding antibod y seroconversion in participants who received 
BNT162b2 and who have no serological or virological evidence of past SARS -CoV -2 
infection or confirmed COVID -19.
Incidence of as ymptom atic SARS -CoV -2 infection per 1000 person -years of 
follow -up based on central laboratory –confirmed NAAT in participants with 
serological or virological evidence (up to the start of theasymptomatic surveillance 
period) of past SARS- CoV -2 infection.
3.3.3. Immunogenicity En dpoints
InPhase 1 participants participat inginboostability  assessment at the following time points 
after receipt of a third dose of BNT162b2:
At the time of Dose 3 and 7 day s and 1 month after Dose 3.
Below are the exploratory immunoge nicity endpoints for Phase 1:
SARS -CoV -2 reference -strain neutralizing titers .
SARS -CoV -2 SA -variant neutralizing titers .
Full-length S- binding or S1-binding IgGlevels .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221116
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 35
TMF Doc ID: 98.03InPhase 2 /3participants at the following time points after receipt of study  intervention:
Baseline and1, 6, 12, and 24 months after completion of vaccination.
Below are the exploratory immunogenicit y endpoints for P hase 2/3:
SARS -CoV -2 neutralizing titers .
Full-length S- binding or S1-binding IgGlevels .
3.3.4. Additional Endpoints (for Phase 2/3 Only)
All safet y, immunogenicity ,and efficacy  endpoints described above will be 
summarized separatel y for participants with confirmed stable HIV . 
AEs, SAEs , and SARS -CoV -2 neutralizing titers will be summarized separatel y for 
participants 16 to 55 of age vaccinated with study  intervention produced b y 
manufacturing “Process 1” and each lot of “Process 2 .”All participants who receive d
“Process 2” vaccine and a random sample of 250 participants 16 to 55 y ears of age 
selected from those who received “ Process 1” vaccine will be included for the 
side-by-side descriptive summary  of “Process 1” and each lot of “Process 2.”
Identification of SARS -CoV -2 variant(s) .
SARS -CoV -2 NTs for any VOCs not alread y specified.
Cell-mediated immune response endpoints .
3.4.Baseline and Other Variables
Measurements or samples collected prior to Dose 1 are considered the baseline data for the 
assessments.
3.4.1. Demographics, Medical History, and Physical Examination
The demographic variables are ag e at Dose 1 (in years ), sex (male or female), race 
(black/African American, American Indian or Alaskan native, Asian, Native Hawaiian or 
other Pacific Islander, white), and ethnicit y (Hispanic/Latino, non -Hispanic/non -Latino, not 
reported).  In cases where more than 1 category  is selected for race, the participant would be 
counted under the category  “multiracial” for analysis. For Phase 2/3, BM I will also be 
included in the demographic variables .
Age at the time of vaccination (in y ears) will be derived based on the participant ’s birthday . 
For example, if the vaccination day  is 1 day  before the participant ’s 19th birthday , the 
participant is considered to be 18 y ears old. For participant s who wer e randomized but not 
vaccinated, the randomization date will be used in place of the date of vaccination at Dose 1
for the age calculation.  If the randomization date is also missing, then the informed consent 
date will be used for the age calculation.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221117
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 36
TMF Doc ID: 98.03Medical history  will be categorized according to MedDRA. Comorbidities that increase the
risk for severe COVID -19 illness will be categorized based on medical history  terms .
For Phase 1, a physical examination will be performed. It will evaluate an y clinically 
significant abnormalities within the following body s ystems: general appearance; skin; head, 
eyes, ears, nose, and throat; heart; lungs; abdomen; musculoskeletal; extremities; 
neurological; and l ymph nodes. Clinically  signi ficant abnormal results will be recorded in 
the CRF .
For Phase 2/3, If the clinical assessment indicates that a ph ysical examination is necessary  to 
comprehensivel y evaluate the participant, ph ysical examination will be performed and 
recorded an y findings in the source documents and, if clinically  significant, it will be 
recorded on the medical history  CRF.
3.4.2. E- Diary Completion
For all participants in Phase 1, a subset of at least 6000 in Phase 2/3, and participants in the 
subset for evaluation of boostability  and protection against emerging VOCs, an e-diary  will 
be considered transmitted if any  data for local reactions, sy stemic events, or use of 
antipy retic medication are present for any  day.  If all data are missing for all items on the 
e-diary  for all 7 day s after vaccination, then the e -diary  will be considered not transmitted. 
An e -diary  will be considered completed if all expected data for all 7 day s are available 
(ie,not missing). Otherwise, the e -diary  will be considered incomplete. For any given day , 
an e-diary  will be considered complete if all expected data are available.
3.4.3. Prior/Concomitant Vaccines and Concomitant Medications
The following concomitant medications and vaccinations will be recorded in the CRF:
All vaccinations received f rom 28 day s prior to study  enrollment until the 6- month 
follow -up visit (Visit 8 for Phase 1 participants, and Visit 4 for Phase 2/3 participants).
In addition, for Phase 1 participants who go on to receive a third dose of BNT162, 
concomitant vaccinations will be collected from the time the participant provides 
informed consent (for receipt of Vaccination 3) through and including Visit 8c (1 month 
after the third dose).  For BNT162 -experienced participants in the subset for evaluation of 
boostability  and p rotection against emerging VOCs, all vaccinations received will be 
recorded from 28 day s prior to the time the participant provides informed consent (for 
participation in the subset) through and including Visit 306.  For BNT162b2- naïve 
participants inthe subset for evaluation of protection against emerging VOCs, all 
vaccinations received will be recorded from 28 days prior to stud y enrollment through 
and including Visit 405.
Prohibited medications listed in the protocol , Section 6.5.1 ,will be recorde d, to include 
start and stop dates, name of the medication, dose, unit, route, and frequency .
In addition, for participants enrolled in Phase 1, all current medication at baseline will be 
recorded, to include start date, name of the medication, dose, unit, route, and frequency .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221118
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 37
TMF Doc ID: 98.033.5.Safety Endpoints
Local reactions, s ystemic events, AEs, and SAEs have been described above in the primary  
safet y endpoints.
4.ANALYSIS SET S (POPULATIONS FOR ANAL YSIS )
Data for all participants will be assessed to determine if participa ntsmeet the criteria for 
inclusion in each anal ysis population prior to unblinding and releasing the database and 
classifications will be documented per SOPs.
Population Description
Enrolled All participants who have a signed ICD.
Randomized All participants who are assigned a randomization number in the 
IWR sy stem.
Dose 1 evaluable 
immunogenicit yFor Phase 1 onl y, all eligible randomized participants who receive 
the vaccine to which they are randomly  assigned at the first dose, 
have at least 1 v alid and determinate immunogenicity  result from 
theblood collection within an appropriate window after Dose 1
(same as visit window, ie ,within 19-23 day s after Dose 1) , and 
have no other important protocol deviations as determined by  the 
clinician.
Dose 2 evaluable 
immunogenicit yAll eligible randomized participants who receive 2 doses of the 
vaccine to which they  are randomly  assigned ,with Dose 2 
received within the predefined window (within 19-42 day s after 
Dose 1), have at least 1 valid and determina te immunogenicit y 
result after Dose 2from the blood collection within an 
appropriate window after Dose 2 ( within 6 -8 day s after Dose 2 
for Phase 1 and within 28- 42 day s after Dose 2 for P hase 2/3), 
and have no other important protocol deviations as determined b y 
the clinician.
Dose 3 booster 
evaluable 
immunogenicit yAll eligible randomized participants who receive 2 doses of 
BNT162b2 as initially  randomized, with Dose 2 received within 
the predefined window , receive a third dose of BNT162b2 or 
BNT16 2b2 SA as rerandomized, have at least 1 valid and 
determinate immunogenicity  result after Dose 3 from a blood 
collection within an appropriate window, and have no other 
important protocol deviations as determined b y the clinician.
Dose 4 booster 
evaluable 
immunogenicit yAll eligible randomized participants who receive 2 doses of 
BNT162b2 as initially  randomized, with Dose 2 received within 
the predefined window , receive 2 booster doses of BNT162b2 SA 
as rerandomized, have at least 1 valid and determinate 
immunogenicit y result after Dose 4 from a blood collection 
within an appropriate window, and have no other important 
protocol deviations as determined by  the clinician.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221119
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 38
TMF Doc ID: 98.03Population Description
Dose 1 all -available 
immunogenicit yFor Phase 1 onl y: all randomized participants who rec eive at least 
1 dose of the stud y intervention with at least 1 valid and 
determinate immunogenicity  result after Dose 1 but before Dose 
2.
Dose 2 all -available 
immunogenicit yAll randomized participants who receive at least 1 dose of the 
study  intervention with at least 1 valid and determinate 
immunogenicit y result after Dose 2.
Dose 3 booster all -
available 
immunogenicit yAll randomized participants who receive 2 doses of BNT162b2 at 
initial randomization, receive a third dose of BNT162b2 or 
BNT162b2 SAat rerandomization, and have at least 1 valid and 
determinate immunogenicity  result after Dose 3.
Dose 4 booster all -
available 
immunogenicit yAll randomized participants who receive 2 doses of BNT162b2 at 
initial randomization, receive 2 booster doses of BNT162b2 SAat 
rerandomization, and have at least 1 valid and determinate 
immunogenicit y result after Dose 4 .
Evaluable eff icacy
(7 day s)All eligible randomized participants who receive all 
vaccination(s) as randomized , with Dose 2 received within the 
predefined window (within 19 -42 day s after Dose 1) and have no 
other important protocol deviations as determined by  the clinician
on or before 7 day s after Dose 2.
Evaluable efficacy
(14 day s)All eligible randomized participants who receive all 
vaccination(s) as randomized, with Dose 2 received within the 
predefined window (within 19 -42 day s after Dose 1) and have no 
other important protocol deviations as determined by  the clinician 
on or befor e 14days after Dose 2.
Evaluable efficacy
(seroconversion)All eligible randomized participants who receive all 
vaccination(s) as randomized, with Dose 2 received within the 
predefined window (within 19 -42 day s after Dose 1) , have at least 
1 N-binding an tibody  test result available at a post–Dose 2 visit,
and have no other important protocol deviations as determined b y 
the clinician prior to Dose 2 .
Evaluable efficacy
(asymptomatic 
surveillance )All eligible randomized participants who receive all 
vaccination(s) as randomized, with Dose 2 received within the 
predefined window (within 19 -42 day s after Dose 1) , consented 
to participate in the asymptomatic surveillance, and have no other 
important protocol deviations as determined b y the clinician on o r 
before the start of the asymptomatic surveillance period .
All-available efficacy Dose 1 all -available efficacy : All randomized participants who 
receive at least 1 vaccination.
Dose 2 all -available efficacy : All randomized participants who 
complete 2 vaccination doses.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221120
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 39
TMF Doc ID: 98.03Population Description
Safety All randomized participants who receive at least 1 dose of the 
study  intervention.
Analy ses of reactogenicity  endpoints will be based on a subset of 
the safet y population that includes participants with any  e-diary  
data reported after vaccination.
The important protocol deviations will be determined by  themedical monitor . An important
protocol deviation is a protocol deviation that, in the opinion of the sponsor’s clinician, 
would materially  affect assessment of immunogenicity /efficacy , eg, participant receipt of a 
prohibited vaccine or medication that might affect immune response or a medication error 
with suspected decrease in potency  of the vaccine. The sponsor’s clinician will identify  
those participant s with important protocol deviations that result in exclusion from anal ysis 
populations before an y unblinded anal ysis in Phase 2/3 is carried out.
5.GENERAL METHODOLOGY AND CONVENTIONS
To facilitate rapid review of data in real time, sponsor staff will be unblinded to study  
intervention allocation for the participants in Phase 1 .The majorit y of sponsor staff will be 
blinded to study  intervention allocation in Phase 2/3.  All laboratory  testing personnel 
performing serology  assay s will remain blinded to study  intervention assigned/received 
throughout the stud y. Further details can be found in the protocol, Section 6.3. The timing 
for statistical anal yses is specified in Section 7.
5.1. Hypotheses and Decision Rules
5.1.1. Vaccine Efficacy Hypoth esis 
Phase 2/3 of the study  has 2 primary  efficacy  endpoints evaluating VE, which is defined as 
VE= 100 × (1–IRR). IRR is calculated as the ratio of first confirmed COVID- 19 illness 
rate in the active vaccine group to the corresponding illness rate in the placebo group
(see Appendix 3fordetails on the calculation of IRR and VE) . The assessment of VE will be 
based on posterior probabilities of VE 1>30% and VE 2>30% using beta-binomial models.  
VE 1represents VE for prophy lactic BNT162b 2against confirmed COVID -19 in participan ts 
without evidence of infection before vaccination, and VE 2represents VE for prophy lactic 
BNT162b 2against confirmed COVID -19 in all participants after vaccination.
For participants with multiple confirmed cases, only  the first case will contribute to the VE 
calculation for each h ypothesis. VE 1and VE 2 will be evaluated sequentially to control the 
overall ty pe I error to the desired level of 2.5%.  VEis demonstrated if there is sufficient 
evidence ( high posterior probability ) that either VE 1>30% or b oth VE 1and VE 2are >30%. 
The assessment for the primary  anal ysis will be based on posterior probability  using a 
beta-binomial model (see Appendix 2for details) .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221121
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 40
TMF Doc ID: 98.03The secondary  objectives regarding VE against asy mptomatic SARS -CoV -2 infection 
(determined b y asymptomatic seroconversion of N -binding antibody  and/or asymptomatic 
SARS -CoV -2 infection based on central laboratory–confirmed NAAT) will be evaluated 
based on t he lower bound of the 95% CI calculated using the Clopper -Pearson method . VE 
will be demonstrated if the lower bound of the 2 -sided 95% CI  for VE is >20%.
5.1.2. Immunogenicity Hypothesis 
5.1.2.1. Hypothesis for Immunogenicity Bridging of 12 to 15 Years to 16 to 25 Years
One of the secondary  objectives in the Phase 3 part of the study  is to evaluate noninferiorit y 
of the immune response to prophy lactic BNT162b2 in participants 12 to 15 y ears of age 
compared to the response in participants 16 to 25 years of age at 1 month after Dose 2. The 
(Dose 2) evaluable immunogenicity  population will be used for the following hy pothesis 
testing:
H0: ln(µ 2) –ln(µ 1) ≤ln(0.67)
where ln (0.67) corresponds to a 1.5- fold margin for noninferiorit y, ln(µ2) and ln(µ1) are the 
natural log of the geometric mean of SARS -CoV -2 neutralizing titers from BNT162b2 
recipients 12 to 15 years of age and 16 to 25 y ears of age , respectivel y, measured 1 month 
after Dose 2 .  If the lower limit of the 95% CI  for the GMR (12- 15 years of age to 
16-25years of age) is >0.67, the noninferiorit y objective is met.
5.1.2.2. Hypothesis for Boostability and Protection Against Emerging SARS -CoV -2
VOCs
The primary  and secondary  objectives for boostability  and protection against emerging 
VOCs for BNT162b2-experienced participants and BNT162b2- naïve participants will be 
assessed based on :
GMRs of SARS -CoV -2 SA and/or reference strain neutralizing titers using a 2-fold 
noninferiority  margin . Noninferiority  is met if the lower limit of the alpha -adjusted CI  
for the GMR is >0.5.
The difference in percentages of participants with seroresponse to the SA strain and/or 
the reference strain using a 10% noninferiority  margin. Noninferiorit y is met if the lower 
limit of the alpha -adjusted CI  for the difference in percentages of participants with 
seroresponse is > -10%.
Seroresponse is defined as achieving ≥4-fold rise from baseline (before Dose 1). If the 
baseline measurement is below LLOQ, the postvaccination measure of ≥4 × LLOQ is 
considered seroresponse.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221122
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 41
TMF Doc ID: 98.035.1.3. Sample Size
5.1.3.1. Phase 1
Phase 1 comprises 15 participants (randomization ratio of 4:1 so that 12 receive active 
vaccine and 3 receive placebo) per group; 13 vaccine groups are studied, corresponding to a 
total of 195 participants.
5.1.3.2. Efficacy Against COVID -19
For Phase 2/3, with ass umptions of a true VE of 60% after the second dose of study  
intervention , a total of approximately  164 first confirmed COVID -19 illness cases will 
provide approximately  90% power.  This would be achieved with 1 7,600evaluable 
participants per group or 21,999vaccine recipients randomized in a 1:1 ratio with placebo, 
for a total sample size of 43,998 , based on the assumption of a 1.3 % illness rate per year in 
the placebo group, accrual of 164 first primary -endpoint cases within 6 months, and 20% of 
the participants being nonevaluable or having serological evidence of prior infection with 
SARS -CoV -2, potentially making them immune to further infection.  Dependent upon the 
evolution of the pandemic, it is possible that the COVID -19 attack ra te may be much higher, 
in which case accrual would be expected to be more rapid, enabling the study ’s primary  
endpoint to be evaluated much sooner.  The total number of participants enrolled in Phase 2/ 3 
may vary  depending on the incidence of COVID -19 at t he time of the enrollment, the true 
underly ing VE, and a potential earl y stop for efficacy  or futility .
5.1.3.3. Efficacy Against Asymptomatic Infection
The secondary  objectives regarding VE against asy mptomatic SARS -CoV -2 infection will be 
assessed in Phase 2/3 pa rticipants (determined b y asymptomatic seroconversion of N -binding 
antibody  and/or asymptomatic SARS -CoV -2 infection based on central laboratory –confirmed 
NAAT) . Assuming a true VE of 70%, a total of 53 asymptomatic cases will provide 
approximately  90% power to conclude true VE>20%. A total of 206 cases is needed to have 
90% power if the true VE is 50%. The h ypothesis forasymptomatic seroconversion of 
N-binding antibody  will be tested if at least 206 cases are accrued. The h ypothesis for
asymptomatic i nfection based on central laboratory –confirmed NAAT in participants who 
are consented to participate in the intensive surveillance phase will be tested if at least 53 
cases are accrued.
5.1.3.4. Immunogenicity Bridging of 12 to 15 Years to 16 to 25 Years
In Phase 3 , approximately  2000 participants are anticipated to be 12 to 15 y ears of age. A 
random sample of 280participant s will be selected foreach of the 2 age groups (12 to 15 
years and 16 to 25 y ears) as an immunogenicity  subset for the noninferiority  assessment.
With the standard deviation and observed GMT difference assumed in the power analysis 
below, a sample size of 225 evaluable participants (or 280 vaccine recipients) per age group
will provide a power of 90.4% to declare the noninferiorit y of adolescent sto 16-to 
25-year-olds in terms of neutralizing antibody  GMR , 1 month after the second dose
(see Table 9).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221123
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 42
TMF Doc ID: 98.03Table 9.Power Analysis fo r Noninferiority Assessment
Criteria Standard Deviation 
(Log Value)aAssu med Observed 
GMT Difference 
(Log Scale)Number of 
Evaluable 
Participants per 
Age GroupPowerb
Lower limit of 95% 
CI for GMR 
(12-15/16 -25) >0.670.65 -0.2 225 90.4%
Abbreviation: GMR = geometric mean ratio; GMT = geometric mean titer.
a. R eference: 1 month after Dose 2, BNT162b2 (30µg), 18-to 55- year age group (C4591001 Phase 2).  
b.At 0.05 alpha level (2 -sided).  
5.1.3.5. Boostability and Protection Against Emerging SARS -CoV -2 VOCs
To assess boostability  and protection against emerging SARS -CoV -2 VOCs, approximately  
300 participants will be enrolled in each of the 3 groups (BNT162b2 -experienced participants 
to receive either a third dose of BNT162b2 [Group 1] or a third dose of BNT162b2 SA
[Group 2], BNT162b2- naïve participants to receive 2 doses of BNT162b2 SA[Group 3]) to 
provide an acceptable safety  database. 
Assuming a 20% nonevaluable rate, approximately 240 evaluable participants in each group 
will contribute to immunogenicity  evaluation. This will provide sufficient power for 
noninferiority  evaluations with appropriate multiplicity  adjustment for t ype I  error control.
For comparisons based on GMR, the assay  standard deviation in log scale is assumed to be 
0.74 b ased on results from Phase 2 of the study  and adjusted for assay  variability . A GMR of 
1 is assumed for each comparison.
For comparisons based on seroresponse, a 90% response rate is assumed for each 
comparative group or at each comparative time point. 
Within-Group Comparison for BNT162b2- Experienced Participants
For each randomized group of BNT162b2 -experienced participants (Group 1: received a 
third dose of BNT162b2 , and Group 2: received a third dose of BNT162b2 SA), with 240 
evaluable participants and the stated assumptions for the GMR and standard deviation, the 
study  has >99.9% power to demonstrate noninferiority based on GMR for the objectives in 
vaccine -experienced individuals using a 2 -fold margin. 
Assuming a true response rate of 90% in each grou p, the study  has 89.7% power to show 
noninferiority based on seroresponse rate for the objectives in vaccine -experienced 
individuals using a 10% margin. 
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221124
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 43
TMF Doc ID: 98.03Between -Group Comparison of BNT162b2 -Naïve Participants to Selected Existing Phase 3 
Participants Who Received 2 Doses of BNT162b2 
Approximately  300 participants will be selected from the existing Phase 3 participants who 
received 2 doses of BNT162b2 to form the control group for the BNT162b2 -naïve 
participants. The selection will ensure comparable distribution of age, sex, and other 
demographic factors in the control group and BNT162b2- naïve group. With 240 evaluable 
BNT162b2 -naïve participants and 240 evaluable participants in the control group and the 
above -stated assumptions for the GMR, standard deviation, and seroresponse rate , the study  
has >99.9% power to declare noninferiority based on GMR for the objectives in 
vaccine -naïve individuals using a 2- fold margin and 89.7% power to declare noninferiority
based on seroresponse rate using a 10% margin.
5.1.3.6. Safety
For safet y outcomes, Table 10shows the probability  of observing at least 1 AE for a given 
true event rate of a particular AE, for various sample sizes.  For example, if the true AE rate 
is 10%, with 12 participants in a v accine group, there is 72% probability  of observing at least 
1AE.
Table 10.Probability of Observing at Least 1 AE by Assumed True Event Rates 
With Different Sample Sizes
Assu med True 
Event Rate of an AEN=12 N=45 N=180 N=300 N=3000 N=6000 N=9000 N=15000
0.01% 0.00 0.00 0.02 0.03 0.26 0.45 0.59 0.78
0.02% 0.00 0.01 0.04 0.06 0.45 0.70 0.83 0.95
0.04% 0.00 0.02 0.07 0.11 0.70 0.91 0.97 >0.99
0.06% 0.01 0.03 0.10 0.16 0.83 0.97 0.99 >0.99
0.08% 0.01 0.04 0.13 0.21 0.91 0.99 0.99 >0.99
0.10% 0.01 0.04 0.16 0.26 0.95 0.99 0.99 >0.99
0.15% 0.02 0.07 0.24 0.36 0.99 0.99 >0.99 >0.99
0.20% 0.02 0.09 0.30 0.45 >0.99 >0.99 >0.99 >0.99
0.25% 0.03 0.11 0.36 0.53 >0.99 >0.99 >0.99 >0.99
0.30% 0.04 0.13 0.42 0.59 >0.99 >0.99 >0.99 >0.99
0.35% 0.04 0.15 0.47 0.65 >0.99 >0.99 >0.99 >0.99
0.50% 0.06 0.20 0.59 0.78 >0.99 >0.99 >0.99 >0.99
1.00% 0.11 0.36 0.84 0.95 >0.99 >0.99 >0.99 >0.99
2.00% 0.22 0.60 0.97 >0.99 >0.99 >0.99 >0.99 >0.99
3.00% 0.31 0.75 >0.99 >0.99 >0.99 >0.99 >0.99 >0.99
5.00% 0.46 0.90 >0.99 >0.99 >0.99 >0.99 >0.99 >0.99
7.00% 0.58 0.96 >0.99 >0.99 >0.99 >0.99 >0.99 >0.99
10.00% 0.72 0.99 >0.99 >0.99 >0.99 >0.99 >0.99 >0.99
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221125
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 44
TMF Doc ID: 98.035.1.4. Multiplicity Considerations
5.1.4.1. Phase 1
For Phase 1, there is no hy pothesis testing. 
5.1.4.2. Phase 2/3 Vaccine Efficacy
For Phase 2/3, a Bay esian approach will be applied for the firstprimary  efficacy  endpoint at 
the interim and final analy ses. The boundaries for declaring efficacy  at interim analy ses and 
success criteria for the final anal ysis are adjusted appropriatel y to control the type I error at 
0.025 ( Table 13).
5.1.4.3. Phase 2/3 Immunogenicity
Figure 1outlines the t ype I  error control strategy  for multiple objectives across different 
populations (BNT162b2 -experienced or BNT162b2 -naïve) and estimands (GMR or 
seroresponse). 
Theobjecti ves for BNT162b2 -experienced participants and BNT162b2 -naïve participants
will be evaluated independent ly. The vaccine -experienced and vaccine -naïve individuals are 
different populations with different objectives. The 2 populations are included in the sa me 
study  to improve operational efficiency .Therefore, no type I error adjustments will be 
applied to the assessments of the 2 populations.
For each population, the objectives will be evaluated separatel y for each estimand . To 
control the overall type I error, the 1 -sided alpha of 0.025 will be split and allocated equall y 
to each estimand. Specifically , for each estimand, the hypotheses will be tested in sequential 
order (as listed in the objectives in Section 3) using a 1-sided alpha of 0.0125 (Figure 1, 
where E and N represent vaccine -experienced and vaccine -naïve, respectively , and a and b 
represent GMR and seroresponse estimands, respectively ).  
Figure 1.Multiplicity Schema
Overall (alpha 
0.025)
BNT162b2 -
experienced 
(0.025)
GMR
(0.0125)
Sequential test E1a, 
E2a, E3a, E4a at 
0.0125
Seroresponse
(0.0125)
Sequential test 
E1b, E2b, E3b, E4b 
at 0.0125
BNT162b2 -naïve 
(0.025)
GMR
(0.0125)
Sequential test N1a 
and N2a at 0.0125
Seroresponse
(0.0125)
Sequential test N1b 
and N2b at 0.0125
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221126
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 45
TMF Doc ID: 98.035.2.General Methods
Time points for local reactions and systemic events refer to data within 7 day s after each 
dose. CIs for all endpoints in the statistical analy sis will be presented as 2- sided at the 95% 
level un less specified otherwise.
5.2.1. Analyses for Binary Data
Descriptive statistics for categorical variables (eg, proportions) are the percentage (%), the 
numerator (n) ,and the denominator (N) used in the percentage calculation, and the 95% CI s
where applicable.
The exact 95% CI for binary  endpoints for each group will be computed using the 
Fdistribution (Clopper -Pearson ).1The 95% CI for the between- group difference for binary  
endpoints will be calculated using the Miettinen and Nurminen method .2
For Phase 2/3 onl y, the 3-tier approach will be used to summarize AEs. For both Tier 1 
(ifany are identified during the stud y) and Tier 2 events, a 95% CI for the between- group 
difference in proportions will be c alculated based on the Miettinen and Nurminen2method. 
In addition, for Tier 1 events (if any ), the as ymptotic p -values will also be presented for t he 
difference in proportions, based on the same test statistic and under the assumption that the 
test statistic is asy mptotically  normally  distributed. For Tier 3 events, counts and 
percentages for each vaccine group will be provided.
A Bayesian beta -binomial model with a minimally  informative prior will be also used for VE 
primary  endpoints (see Appendix 2).
5.2.2. Analyses for Count Data
The number of occurrences of a certain event is count data and thus could be modeled using 
Poisson distribution. The incidence rate is estimated as the number of events observed 
divided by  the total person -yearsof follow -up. 
Assuming an observed event is from Poisson distribution with parameter λT, where λ is the 
incidence rate and T is the total person -yearsof follow -up, based on the relationship between 
the Poisson and chi-square distribution ,3theexact lower and upper α- percent 2 -sided 
confidence limits for λT can be estimated by :
=,/
  and =( ),/
, respectivel y, where Y is the number of events observed.
The exact lower and upper confidence limit for incidence rate λ can then be obtained as /
and /, respectivel y. 
5.2.3. Analyses for Continuous Data
Unless otherwise stated, descriptive statistics for continuous variables are n, mean, median, 
standard deviation, minimum, and maximum.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221127
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 46
TMF Doc ID: 98.035.2.3.1. Geometric Means
For immunogenicity  results of SARS -CoV -2 neutralizing titers, the GMTs will be computed 
along with associated 95% CIs. The GMTs will be calculated as the mean of the assay  
results after making the logarithm transformation and then exponentiating the mean to 
express results on the original scale .Two-sided 95% CI s will be obtained by  taking log 
transforms of titers , calculating the 95% CI with reference to Student’s t- distribution, and 
then exponent iating the confidence limits . Similarly , GMCs and 95% CI swill be calculated 
for S1-binding IgGlevels and RBD- binding IgG levels .
5.2.3.2. Geometric Mean Fold Rises
GMFRs will be defined as the result after vaccination divided by  the result before 
vaccination . GMFRs are limited to participant s with nonmissing values at both time points.
GMFRs will be calculated as the mean of the difference of logarithmicall y transformed 
neutralization titers or antibody  levels ( later result minus earlier result ) and exponentiat ing 
the mean. The associated 2- sided 95% CIs are obtained by constructing CIs using Student’s 
t-distribution for the mean difference on the natural log scale and exponentiating the 
confidence limits.
5.2.3.3. Geometric Mean Ratio s
For SARS -CoV -2 neutralizing titer s and S1 -binding IgGlevels and RBD -binding IgG levels , 
the GMRs will be provided along with associated 95% CI s.GMRs will be limited to 
participants with nonmissing values for both SARS- CoV -2 neutralizing titers and S1-binding 
IgGlevel s/RBD -binding IgG level sat each time point. The GMR will be calculated as the 
mean of the difference of logarithmicall y transformed assay  results (eg, SARS -CoV -2 
neutralizing titers minus S1- binding IgG level for each participant) and exponentiating the 
mean.  Two- sided C Is will be obtained by  calculating CIs using Student’s t -distribution for 
the mean difference of the logarithmically  transformed assay  results and exponentiating the 
confidence limits.
For SARS -CoV -2 neutralizing titers in participants 12 to 15 y ears of a ge and 16 to 25 y ears 
of age, the GMRs will be provided along with associated 95% CI. The GMR and its 2- sided 
95% CI  will be derived by  calculating differences in means and CIs on the natural log scale 
of the titers based on the Student’s t -distribution and then exponentiating the results.  The 
difference in means on the natural log scale will be 12 to 15 y ears minus 16 to 25 y ears.  
Noninferiorit y will be declared if the lower bound of the 2 -sided 95% CI for the GMR is 
greater than 0.67. 
For assessment of boostability  and protection against emerging VOCs, t he comparisons of 
different NTs (anti -SA or anti –reference strain) or the same NTs at different time points 
within the same group will be limited to participants with nonmissing values at both time 
points or both NT measurements. GMRs will be calculated as the mean of the difference of 
logarithmicall y transformed titers for each participant (eg, later time point minus earlier time 
point) and exponentiating the mean. The associated 2- sided CIs will be obtained by  
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221128
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 47
TMF Doc ID: 98.03constructing CIs using Student’s t distribution for the mean difference on the logarithm scale 
and exponentiating the confidence limits .
For the between -group comparison, GMRs will be calculated as the mean o f the difference of 
logarithmicall y transformed assay  results between 2 groups and exponentiating the mean.  
The associated 2 -sided 97.5% CIs will be obtained by  calculating CIs using Student’s 
t-distribution for the mean difference of the logarithmically  transformed titers and 
exponentiating the confidence limits.
5.2.3.4. Geometric Mean Fold Rise Ratio s
The ratios of GMFR A to GMFR B and GMFR A to GMFR C may  be explored, where 
GMFR A is the GMof the ratio of the SARS -CoV -2 neutralizing titer at the time point after 
vaccination to the corresponding titer at the time point before vaccination, G MFR B is the 
GMof the ratio of the S1-binding IgGlevel at the time point after vaccination to the 
corresponding antibody  level at the time point before vaccination, and GMFR C is the GMof 
the ratio of the RBD -binding IgG level at the time point after vaccination to the 
corresponding antibody  level at the time point before vaccination .
5.2.3.5. Reverse Cumulative Distribution Curves
Empirical RCDCs will plot proportions of participant s with values equal to or exceeding a 
specified assay  value versus the indicated assay value, for all observed assay  values. Data 
points will be joined by  a step function with data points on the left side of the step.
5.3. Methods to Manage Missing Data
For endpoints, the missing data handling rules are described in the corresponding endpoint 
sections.
For the missing dates, the sponsor data standard rules for imputation will be applied 
(eg,partial dates for AEs will be imputed according to Pfizer standard algorithms).
Missing COVID -19 test data in Phase 2/3 for computing VEwill be imputed in the 
sensitivity  anal ysis.Details are included in Section 6.1.3.1.2 .
6.ANALYSES AND SUMMARIES
6.1.Primary Endpoint (s)
6.1.1. Safety Endpoints
The safet y analyses are based on the safet y population. Analy ses of reactogenicity  endpoints 
are based on a subset of the safet y population that includes participants with any e-diary  data 
reported after vaccination. Participants will be summarized by  vaccine group according to 
the study  interventions they actuall y received. Missing e -diary  data will not be imputed; 
missing AE dates will be handled according to the Pfizer safet y rules.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221129
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 48
TMF Doc ID: 98.036.1.1.1. Local Reactions
6.1.1.1.1. Main Analysis
Estimand: The percentage of participants reporting local reactions (redness, swelling ,and 
pain at the injection site) within 7 days after each dose (Section 2.1).
Analy sis set:Safety  population (Section 4).
Analy sis time point: Within 7 day s after each dose .
Analy sis methodology : Descriptive statistics (Section 5.2.1 ).
Intercurrent events and missing data: The participants without any e-diary data 
throughout the 7 day s after vaccination will be excluded from the anal ysis at that 
particular vaccination ; missing values will not be imputed.
Report ing results: Descriptive statistics for each and any  local reaction after each dose in 
each vaccine group will be presented b y maxi mum severity and cumulatively  across 
severit y levels. Confirmed e -diary  errors will be excluded from the anal ysis. Descriptive 
summary  statistics will include counts and percentages of participants with the indicated 
endpoint and the associated 2-sided Clopper -Pearson 95% CI s.
6.1.1.1.2. Supplementary Analyses
To suppo rt the assessment of local reactions, the following endpoints (as defined in
Section 3.1.1.1 ) will be summarized with the same anal ysis time point and analysis 
population , analy sis me thodology ,and appropriate reporting results. Confirmed e -diary  
errors will be excluded from these anal yses.
Duration (day s) of each local reaction after each dose .
Onset day  of each local reaction after each dose .
These continuous endpoints will be summarized by display ing n, mean, median, standard 
deviation, minimum, and maximum for each vaccine group.
Figures:
Bar charts with the proportions of participants for each local reaction throughout 7 day s will
be plotted for each vaccine group. The bars will be divided into severit y categories to 
highlight the proportions of participants b y maximum severity.
6.1.1.2. Systemic Events
6.1.1.2.1. Main Analysis
Estimand: The percentage of participants reporting s ystemic events ( fever, fatigue, 
headache, chills, vomiting, diarrhea, new or worsened muscle pain, and new or worsened 
jointpain) within 7 day s after each dose (Section 2.1 ).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221130
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 49
TMF Doc ID: 98.03Analy sis set: S afety  population (Section 4).
Analy sis time point : Within 7 day s after each dose .
Analy sis methodology : Descriptive statistics (Section 5.2.1 ).
Intercurrent events and missing data: The participants without any  e-diary data 
throughout the 7 day s after vaccination will be excluded from the anal ysis at that 
particular vaccination ; missing values will not be imputed.
Reporting result s: Descriptive statistics for each systemic event after each dose in each 
vaccine group will be presented by  maximum severity  and cumulatively  across severity  
levels. Descriptive summary  statistics will include counts and percentages of participants 
with the indicated endpoint and the associated 2-sided Clopper -Pearson 95% C Is.
6.1.1.2.2. Supplementary Analyses
The following endpoints for assessment of s ystemic events will be summ arized similarly  to 
the assessment of local reactions:
Duration of each s ystemic event after each dose .
Onset day  of each s ystemic event after each dose .
These continuous endpoints will be summarized by display ing n, mean, median, standard 
deviation, minim um, and maximum for each vaccine group.
The use of antip yretic medication (see Section 3.1.1.3 ) will be summarized similarly  to 
systemic events, except that there is no severity  level associated with the use of antipy retic 
medication.
Figures: 
Bar charts with the proportions of participant s reporting each s ystemic event throughout 
7days after each dose will be plotted for each vaccine group.  The bars will be divided into 
severit y categories to highlight the proportions of participant s by severity .
6.1.1.3. Adverse Events
6.1.1.3.1. Main Analysis
Estimand: The percentage of participants reporting AEs from Dose 1 to 1 month after the 
second dose for all phases, from Dose 1 to 7 day s after the second dose for the first 360 
participants randomized in Phase 2 ,and from Dose 1 (of booster BNT162b2 or 
BNT162b2 SA) to 1 month after the last dose for participants in the Phase 3 subse t for 
evaluation of boostability and protection against emerging VOCs (Section 2.1 ).
Analy sis set: Safety  population (Section 4).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221131
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 50
TMF Doc ID: 98.03Analy sis time point : Dose 1 to 1 month after the second dose for all phases, Dose 1 to 7 
days after the second dose for the first 360 participants randomized in Phase 2, Dose 1 (of 
booster BNT162b2 or BNT162b2 SA) to 1 month after the last dose for participants in the 
Phase 3 subset for evaluation of boostability  and protection against emerging VOCs .
Analy sis methodology : Descriptive statistics ( Section 5.2.1 ) for all phases and additional
3-tiered approach for Phase 2/3 (Section 3.1.1.4 ).
Intercurrent events and missing data: Partial AE dates will be imputed using the Pfizer 
standard algorithm.
Reporting result s: AEs will be categorized according to MedDRA terms. A 3-tier
approach will be used to summarize AEs for Phase 2/3 only . Under this approach AEs 
are classified into 1 of 3 tiers (Section 3.1.1 .4). For both Tier 1 and Tier 2 events, 2 -sided 
95% CI s for the difference between the active vaccine and placebo groups in the 
percentage of participants reporting the events based on the Miettinen and Nurminen2
method will be provided. In addition, for Tier 1 events, the as ymptotic p -values will also 
be presented for the difference between groups in the percentage of participants reporting 
the events, based on the same test statistic and under the assumption that the test statistic 
is asy mptoticall y normally distributed. AE display s will be sorted in descending order of 
point estimates of risk difference within SOC . Descriptive summary  statistics (co unts, 
percentages, and associated Clopper- Pearson 95% CI s) will be provided for any  AEs for 
each vaccine group.
6.1.1.3.2. Supplementary Analyses
Immediate AEs (within the first 30 minutes after each dose) will also be summarized for each 
vaccine group . All AEs after informed consent and prior to the first vaccination will not be 
included in the anal yses but will be listed.
6.1.1.4. Serious Adverse Events
6.1.1.4.1. Main Analyses
Estimand:  The percentage of participants reporting SAEs from Dose 1 to 6 month s after
the second dose for all phases, from Dose 1 to 7 day s after the second dose for the first 
360 participants randomized in Phase 2 , and from Dose 1 (of booster BNT162b2 or 
BNT162b2 SA) to 5 or 6 month safter the last dose for participants in the Phase 3 subset 
for evaluatio n of boostability  and protection against emerging VOCs (Section 2.1).
Analy sis set: Safety  population (Section 4).
Analy sis time point: Dose 1 to 6 month safter the second dose for all phases, Dose 1 to 7 
days after the second dose for the first 360 participants randomized in Phase 2, Dose 1 (of 
booster BNT162b2 or BNT162b2 SA) to 5 or 6 months after the last dose for participants 
in the Phase 3 subset for evaluation of boostability and protection against emerging
VOCs .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221132
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 51
TMF Doc ID: 98.03Analy sis methodology : Descriptive s tatistics ( Section 5.2.1 ).
Intercurrent events and missing data: Partial SAE dates will be imputed using the Pfizer 
standard algorithm.
Reportin g results: SAEs will be categorized according to MedDRA terms. Counts, 
percentages, and the associated Clopper -Pearson 95% CI s of SAEs will be provided for 
each vaccine group.
6.1.1.5. Hematology and Chemistry Parameters (for Phase 1 Only)
6.1.1.5.1. Main Analyses
Estimand s: The percentage of participants with abnormal hematology and chemistry 
laboratory  values 1 and 7 day s after Dose 1; and 7 day s after Dose 2 ( Section 2.1).
The percentage of participants with grading shifts in hematology  and chemistry  
laboratory  assessments between baseline and 1 and 7 day s after Dose 1; and before 
Dose 2 and 7 day s after Dose 2 ( Section 2.1).
Analy sis set: Safety  population ( Section 4).
Analy sis time point: 1 and 7 day s after Dose 1; and 7 day s after Dose 2.
Analy sis methodology : Descriptive statistics including counts and percentage 
(Section 5.2.1 ).
Intercurrent events and missing data: Missing values will not be imputed.
Reporting results: Descriptive summary  statistics will be provided including counts and 
percentages of participants with the indicated endpoint and the associated 
Clopper -Pearson 2 -sided 95% CI s.
6.1.2. Immunogenicity Endpoints (for the Phase 2/3 Subset for Evaluation of 
Boostability and Protection Against Emerging VOCs Only)
6.1.2.1. SARS -CoV -2 Reference Strain NT and SA NT at 1 Month After Dose 3 vs
Reference Strain NT at 1 Month After Dose 2 in BNT162b2 -Experienced Participants
6.1.2.1.1. Main Analyses
Estimands:
E1a: GMR of reference strain NT 1 month after the third dose of BNT162b2 to 1 
month after the second dose of BNT162b2 in the same individuals (Section 2.1).
E2a: GMR of SA NT 1 month after 1 dose of BNT162b2 SAto the reference strain NT 
1 month after the second dose of BNT162b2 in the same individuals (Section 2.1).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221133
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 52
TMF Doc ID: 98.03Analy sis set: Dose 3booster evaluable and all -available immunogenicit ypopulations 
(Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2, 1 month after the 
third dose of BNT162b2, and 1 month after 1 dose of BNT162b2 SA.
Analy sis methodology : The comparisons of different NTs (anti- SA or anti –reference 
strain) or the same NTs at different time points within the same group will be limited to 
participants with nonmissing values at both time points or both NT measurements. GMRs 
will be calculated as the mean of the difference of logarithmically transformed titers for 
each participant (eg, later time point minus earlier time point) and exponentiating the 
mean (Section 5.2.3.3 ). The associated 2 -sided 9 7.5% CI s will be obtained by  
constructing CIs using Student’s t-distribution for the mean difference on the logarithm 
scale and exponentiating the confidence limits .Noninferiority  of E1a and E2a will be 
assessed sequentially . Noninferiority  will be declared if the lower bound of the 2- sided 
97.5% CI  for the GMR is gre ater than 0.5.
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The GMTs at each time point, GMRs ,and the associated 2 -sided 
97.5% CI s will be provided.
Figures:
Empirical RCDCs will be provided for SARS -CoV -2 SA and r eference strain NTs at each 
time point.
6.1.2.2. Seroresponse to the Reference Strain and SA Strain at 1 Month After Dose 3 vs
Seroresponse to the Reference Strain at 1 Month After Dose 2 in BNT162b2 -
Experienced Participants
6.1.2.2.1. Main Analyses
Estimands:
E1b: The difference in percentages of participants with seroresponse to the reference 
strain at 1 month after the third dose of BNT162b2 and 1 month after the second dose 
of BNT162b2 in the same individuals ( Section 2.1).
E2b: The difference in percentages of participants with seroresponse to the SA strain 
at 1 month after 1 dose of BNT162b2 SAand seroresponse to the reference strain at 
1 month after the second dose of BNT162b2 in the same individuals (Section 2.1).
Analy sis set: Dose 3booster evaluable and all -available immunogenicit ypopulations 
(Section 4).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221134
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 53
TMF Doc ID: 98.03Analy sis time points: 1 month after the second dose of BNT162b2, 1 month after the 
third dose of BNT162b2, and 1 month after 1 dose of BNT162b2 SA.
Analy sis met hodology : The percentages of participants with seroresponse at each time 
point and the difference in percentages will be provided. The 2 -sided 97.5% CI s for the 
difference in percentages of participants with seroresponse will be calculated using the 
Miett inen and Nurminen method (Section 5.2.1 ). Noninferiorit y of E1b and E2b will be 
assessed sequentially . Noninferiority  will be declared if the lower bound of the 2- sided 
97.5% CI  for the difference in percentages of participants with seroresponse is greater 
than - 10%.
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The c ounts, percentages of participants with seroresponse at each time 
point, the differe nce in percentages ,and the associated 2 -sided 97.5% CIs will be 
provided.
6.1.2.3. SARS -CoV -2 SA NT at 1 Month After Dose 2vsReference Strain NT at 1 
Month After Dose 2 in BNT162b2 -Naïve Participants
6.1.2.3.1. Main Analyses
Estimands:
N1a: GMR of SA NT 1 month after the second dose of BNT162b2 SAto the reference 
strain NT 1 month after the second dose of BNT162b2 ( Section 2.1).
Analy sis set: Dose 2evaluable and all -available immunogenicit ypopulations ( Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2 and1 month after the 
second dose of BNT162b2 SA.
Analy sis methodology : For the between -group comparison, GMR swill be calculated as 
the mean of the difference of logarithmicall y transformed assay results between 2 groups 
and exponentiating the mean ( Section 5.2.3.3 ).  The associated 2-sided 97.5% CIs will be 
obtained by calculating CIs using Student’s t -distribution for the mean difference of the 
logarithmically transformed titers and exponentiating the confidence limits .  
Noninferiorit y will be declared if the lower bound of the 2- sided 97.5% CI for the GMR 
is greater than 0.5. 
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based up on 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The GMTs at each time point, GMRs ,and the associated 2 -sided 
97.5% CI s will be provided.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221135
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 54
TMF Doc ID: 98.03Figures:
Empirical RCDCs will be provided for SARS -CoV -2 SA and r eference strain NTs at each 
time point for each vaccine group .
6.1.2.4. Seroresponse to the SA Strain at 1 Month After Dose 2 vs Seroresponse to the 
Reference Strain at 1 Month After Dose 2 in BNT162b2 -Naïve Pa rticipants
6.1.2.4.1. Main Analyses
Estimands:
N1b: The difference in percen tages of participants with seroresponse to the SA strain 
at 1 month after the second dose of BNT162b2 SAand seroresponse to the reference 
strain at 1 month after the second dose of BNT162b2 (Section 2.1).
Analy sis set: Dose 2evaluable and all -available immunogenicit ypopulations ( Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2 and1 month after the 
second dose of BNT162b2 SA.
Analy sis methodology : Thedifference in percentages of participants with seroresponse 
and associated 2 -sided 97.5% CI s will be calculated in the same way  as for primary  
endpoints E1b and E2b. Noninferiorit y will be declared if the lower bound of the 2- sided 
97.5%CI for the diffe rence in percentages of participants with seroresponse is greater 
than -10%.
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based upon 
additio nal data from the assay .  Missing data will not be imputed.
Reporting results: The counts, percentages of participants with seroresponse at each time 
point, the difference in percentages ,and the associated 2 -sided 97.5% CIs will be 
provided.
6.1.3. Vaccin e Effic acy Endpoints (for Phase 2/3 Only)
6.1.3.1. COVID -19 Incidence per 1000 Person -Years of Follow -up
6.1.3.1.1. Main Analyses
Estimand s: 
100 × (1 – IRR) [ratio of confirmed COVID -19 illness from 7 day s after the second 
dose per 1000 person -years of follow -up in participants without evidence of infection 
(prior to 7 day s after receipt of the second dose) for the active vaccine group to the 
placebo group ( Section 2.1)].
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221136
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 55
TMF Doc ID: 98.03100 × (1 – IRR) [ ratio of confirmed COVID -19 illness from 7 day s after the second 
dose per 1000 person -years of follow -up in participants with and without evidence of 
infection (prior to 7 day s after receipt of the second dose) for the active vaccine group 
to the placebo group ( Section 2.1)].
Analy sis set: Evaluable efficacy  (7 day s) and all -available efficacy  populations 
(Section 4).
Analy sis time point: At interim anal yses and final analysis when the surveillance period 
ends.
Analy sis methodology : Assessment of VE will be performed for confirmed COVID-19 
from 7days after the receipt of the second dose of study  intervention onwards, and will 
be estimated b y 100 × (1 –IRR), where IRR is the calculated ratio of COVID -19 illness 
rate per 1000 person -years of follow -up in the active vaccine group to the corresponding 
illness rate in the placebo group after the second dose (see Appendix 3for details on the 
derivation of IRR and VE).   The posterior probability  (ie, P[VE >30%|data]) at each 
interim analy sis and final analysis will be computed using a beta -binomial model and a 
specified minimally  informative beta distribution as prior (details can be found in 
Appendix 2).
Intercurrent events and missing data: Missing efficacy  data (symptom is present without 
laboratory testing data) will not be imputed in the main anal yses.
Reporting results: The point estimate of VE, 95% credible intervals using the 2.5 th
percentile and the 97.5 thpercentile, and Bay esian posterior probability  of VE greater 
than 30% will be provided (details can be found in Appendix 2).
6.1.3.1.2. Sensitivity and Supplemental Analyses
With MAR assumption, a missing efficacy  endpoint (laboratory -confirmed COVID -19 
results ) may be imputed based on predicted probability  using the full y conditional 
specification method.4  The imputation will run multiple times (up to 1000) and summary  
statistics similar to those used in the main anal ysiswill be tabulated across the imputations. 
Other imputation methods without the MAR assumption may be explored, eg , a tipping point 
analysis.
All COVI D-19 cases after Dose 1 may  be anal yzed using the Dose 1 all- available efficacy  
population. COVI D-19 disease -related informatio n may  be summarized or listed.
After the final efficacy  analy ses at 164 first primary  cases , updated efficacy  analy ses will be 
performed with additional data accrued. The point estimate of VE in the blinded follow -up 
period and associated 2 -sided 95% CI will be derived using the Clopper Pearson method 
adjusted for surveillance time , and t he posterior probability  (ie, P[VE >30%|data]) will be 
provided. VE at different follow -uptime interval sand against different variant strains may  
be assessed.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221137
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 56
TMF Doc ID: 98.03Efficacy could also be assessed over a longer time period using time -to-event data anal ysis 
methods to account for censoring ( participants censored when they  receive other vaccines or 
withdraw ) as well as potentially  confounding factors. A Kaplan -Meier curve show ing the 
cumulative incidence of COVID-19 cases over time may  also be informative to understand 
the sustainability  of VE. 
For the assessment of efficacy  in the presence of potential crossover, the established 
adjusting methods may  be considered . For example, a rank-preserving structural failure time 
model may  be appropriate to attempt to reconstruct data for the control arm as if crossover 
had not occurred, with the aim of reducing bias an d allowing the vaccine effect to be 
assessed more accuratel y.
6.2. Second aryEndpoints
6.2.1. Immunogenicity Endpoints
Phase 1
The statistical anal ysis of immunogenicit y results for Phase 1 will be primarily  based on the 
Dose 1 and Dose 2 evaluable immunogenicity  populations. Serology  data after a 
postbaseline positive SARS -CoV -2 test result will not be included in the analy sis based on 
the evaluable immunogenicity  populations.  An additional analy sis will be performed based 
on the all -available populations if there is a large enough difference in sample size between 
the all -available immunogenicity  population and the evaluable immunogenicity  population.  
Participants will be summarized according to the vaccine group to which they  were 
randomized.  Missing serology  data will not be imputed.
Phase 2/3
The statistical anal ysis of immunogenicit y results for Phase 2/3 will be based on Dose 2 
evaluable immunogenicity  population.  Serology  data after a postbaseline positive 
SARS -CoV -2 test result will not be included in the anal ysis based on the evaluable 
immunogenicit y population .  An additional anal ysis may  be performed based on the Dose 2 
all-available immunogenicity  population if needed.  Participants will be summarized 
according to the vaccine group to which they  were randomized. Missing serology data will 
not be imputed .
6.2.1.1. SARS -CoV -2 Neutralizing Titers (Phase 1 )
6.2.1.1.1. Main Analyses
Estimand s:
GMTs (Section 2.1).
GMFR from before vaccination to each subsequent time point after vaccination
(Section 2.1 ).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221138
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 57
TMF Doc ID: 98.03Proportion of participants achieving ≥4-fold rise from before vaccination to each 
subsequent time point after vaccination ( Section 2.1 ).
Analy sis set: Dose 1 and Dose 2 evaluable and all -available immunogenicity population s
(Section 4).
Analy sis time point s: 7 and 21 day s after Dose 1; 7 and 14 days and 1, 6, 12 and 
24months after Dose 2 .
Analy sis methodology : GMs and the associated 2 -sided CI s will be derived by  
calculating means and CIs on the natural log scale based on Student’s t- distribution, and 
then exponentiating the results (Section 5.2.3.1 ).GMFRs will be limited to participants 
with nonmissing values prior to the first dose and at the postvaccination time point. The 
GMFR will be calculated as the mean of the difference of logarithmically  transformed 
assay  results (later time point – earlier time point) and exponentiated to transform results 
back to the original scale.  Two- sided CIs will be obtained by  calculating CIs using 
Student’s t -distribution for the mean difference of the logarithmically  transformed assay  
results and exponentiating the confidence limits ( Section 5.2.3.2 ).Percent ages of 
partic ipants with ≥4 -fold rise will be calculated with the associated 2- sided 95% C Is
(Clopper -Pearson method ).
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.However, this calculation may  be adjusted based upon 
additional data from the assay .Missing data will not be imputed.
Reporting results: the GMTs at each time point , GMFRs from before vaccination to each 
subsequent time point a fter vaccination , and the percent ages of partic ipants achieving 
≥4-fold rise and the associated 2-sided 95% CIs from before vaccination toeach time 
point after vaccination .
Figures:
Empirical RCDCs will be provided for SARS -CoV -2 neutralizing titers afte r Dose 1 and 
after Dose 2 (Section 5.2.3.5 ).
6.2.1.2. S1-Binding IgG Levels and RBD -Binding IgG L evels (Phase 1)
6.2.1.2.1. Main Analyses
Estimand s:
GMCs (Section 2.1).
GMFR from before vaccination to each subsequent time point after vaccination
(Section 2.1 ).
Proportion of participants achieving ≥4-fold rise from before vaccination to each 
subsequent time point after vaccination ( Section 2.1).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221139
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 58
TMF Doc ID: 98.03Analy sis set: Dose 1 and Dose 2 evaluable and all-available immunogenicity  population s
(Section 4).
Analy sis time point s: 7 and 21 day s after Dose 1; 7 and 14 days and 1, 6, 12, and 
24months after Dose 2 .
Analy sis methodology : GMs and the associated 2 -sided CI s will be derived by  
calculating means and CIs on the natural log scale based on Student’s t- distribution, and 
then exponentiating the results ( Section 5.2.3.1 ). GMFRs will be limited to participants 
with nonmissing values prior to the first dose and at the postvaccination time point. The 
GMFR will be cal culated by exponentiating the mean of the difference of logarithmically  
transformed assay  results (later time point – earlier time point).  Two -sided CI s will be 
obtained by  calculating CI s using Student’s t -distribution for the mean difference of the 
logarithmicall y transformed assay  results and exponentiating the confidence limits 
(Section 5.2.3.2 ). Percenta ges of partic ipants with ≥4-fold rise will be calculated with the 
associated 2- sided 95% CI s (Clopper -Pearson method).
Intercurrent events and missing data: Concentrations below the LLOQ or denoted as BLQ 
will be set to 0.5 × LLOQ for analy sis.However, this calculation may  be adjusted based 
upon additional data from the assay .Missing data will not be imputed.
Reporting results: the GMC s, GMFRs, and percent ages of partic ipants with ≥4-fold rise 
and the associated 2 -sided 95% CI s will be provided for each study  intervention
(active/placebo) within each group before vaccination and at each time point.
Figures:
Empirical RCDCs will be provided for S1-binding IgGlevels and RBD -binding IgG levels
after Dose 1 and after Dose 2 (Section 5.2.3.5 ).
6.2.1.3. SARS -CoV -2 Neutralizing Titers vsSARS -CoV -2S1-Binding IgG Levels and 
RBD -Binding IgG L evels (Phase 1)
6.2.1.3.1. Main Analyses
Estimand s:
GMRof SARS -CoV -2 neutralizing titer sto S1-binding IgGlevels (Section 2.1).
GMR of SARS -CoV -2 neutralizing titer sto RBD- binding IgG levels (Section 2.1).
Analy sis set: Dose 1 and Dose 2 evaluable and all -available immunogenicity  population s
(Section 4).
Analy sis time point s: 7 and 21 day s after Dose 1; 7 and 14 days and 1, 6, 12, and 
24months after Dose 2 .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221140
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 59
TMF Doc ID: 98.03Analy sis methodology : GMRs will be limited to participan ts with nonmissing values for 
both SARS -CoV -2 neutralizing titers and S1-binding IgGlevel or RBD -binding IgG 
level at each time point. The GMR will be calculated as the mean of the difference of 
logarithmicall y transformed assay  results (eg, SARS -CoV -2 neutralizing titers minus 
S1-binding IgGlevels for each participant) and exponentiat ing the mean
(Section 5.2.3.3 ).Two-sided CI s will be obtained by  calculating CIs using Student’s 
t-distribution for the mean difference of the logarithmically  transformed assay  results and 
exponentiating the confidence limits (Section 5.2.3.3 ).
Intercurrent events and missing data: Concentrations below the LLOQ or denoted as BLQ 
will be set to 0.5 × LLO Q for analy sis.However, this calculation may  be adjusted based 
upon additional data from the assay .Missing data will not be imputed.
Reporting results: The GMRs and the associated 2 -sided 95% CIs will be provided for 
each study  intervention within e ach group before vaccination and at each time point.
6.2.1.4. SARS -CoV -2 Neutralizing Titer sin Participants 12 to 15 Years of Age vs Those 
16 to 25 Years of Age (Phase 2/3 )
6.2.1.4.1. Main Analyses
Estimands: GMR ,estimated by the ratio of the geometric mean of SARS -CoV -2 
neutralizing titers in the 2 age groups (12 -15 years of age to 16- 25 years of age) 1 month 
after completion of vaccination (Section 2.1).
Analy sis set: Dose 2 evaluable and all -available immunogenicit y populations ( Section 4).
Analy sis time points: 1 month after Dose 2 .
Analy sis methodology : The GMR and its 2- sided 95% CI  will be derived by  calculating 
differences in means and CI s on the natural log scale of the titers based on the Student’s 
t-distribution and then exponentiating the results.  The difference in means on the natural 
log scale will be 12 to 15 y ears minus 16 to 25 y ears.  Noninferiorit y will be declared if 
the lower bound of the 2 -sided 95% CI  for the GMR is greater than 0.67 
(Section 5.2.3.3 ).
Intercurrent events and missing data: Concentrations below the LLOQ or denoted as BLQ 
will be set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based 
upon additional data from the assay .  Missing data will not be imputed.
Reporting results: The GMRs and the associated 2 -sided 95% CIs will be provided.
6.2.1.4.2. Supplemental Analyses
The counts, percentages of participants with seroresponse (achieving ≥4-fold rise from 
baseline ,as defined in Section 5.1.2.2 ), the difference in percentages between the 2 age 
groups (12- 15 years of age minus 16-25 years of age) ,and the associated 2 -sided 95% CIs 
will be provided .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221141
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 60
TMF Doc ID: 98.036.2.1.5. SARS -CoV -2 SANT and Reference Strain NT at 1 Month After Dose 3 vs
Reference Strain NT at 1 Month After Dose 2 in BNT162b2 -Experienced Participants
6.2.1.5.1. Main Analyses
Estimands:
E3a: GMR of SA NT 1 month a fter the third dose of BNT162b2 to the reference 
strain NT 1 month after the second dose of BNT162b2 in the same individuals
(Section 2.1).
E4a: GMR of reference strain NT 1 month after 1 dose of BNT162b2 SAto 1 month 
after the second dose of BNT162b2 in the same individuals ( Section 2.1).
Analy sis set: Dose 3booster evaluable and all -available immunogenicit ypopulations 
(Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2, 1 month after the 
third dose of BNT162b2, and 1 month after 1 dose of BNT162b2 SA.
Analy sis methodology : GMRs and t he associated 2 -sided 97.5% CIs will be calculated in 
the same way  as for the primary  endpoints E1a and E2a ( Section 6.1.2.1.1 ). If 
noninferiority  is established for both E1a and E2a, E3a and E4a will be assessed 
sequentially  using the same criterion ( lower bound of the 2- sided 97.5% CI for the GMR 
is greater than 0.5).
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ fo r analy sis.  However, this calculation may  be adjusted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The GMTs at each time point, GMRs ,and the associated 2 -sided 
97.5% CI s will be provided.
Figures:
Empirical RCDCs will be provided for SARS -CoV -2 SA and r eference strain NTs at each 
time point.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221142
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 61
TMF Doc ID: 98.036.2.1.6. Seroresponse to the SA Strain and Reference Strain at 1 Month After Dose 3 vs 
Seroresponse to the Reference Strain at 1 Month After Dose 2 in BNT162b2-
Experienc ed Participants
6.2.1.6.1. Main Analyses
Estimands:
E3b: The difference in percentages of participants with seroresponse to the SA strain 
at 1 month after the third dose of BNT162b2 and seroresponse to the reference strain 
at 1 month after the second dose of BNT162b2 in the same individuals (Section 2.1).
E4b: The difference in percentages of participants with seroresponse to the reference
strain at 1 month after 1 dose of BNT162b2 SAand 1 month after the second dose of 
BNT162b2 in the same individuals (Section 2.1).
Analysis set: Dose 3booster evaluable and all -available immunogenicit ypopulations 
(Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2, 1 month after the 
third dose of BNT162b2, and 1 month after 1 dose of BNT162b2 SA.
Analy sis methodology : The difference in percentages of participants with seroresponse 
and the associated 2 -sided 97.5% CI s will be calculated in the same way  as for the 
primary  endpoints E1b and E2b ( Section 6.1.2.2.1 ). If noninferiority is established for 
both E1b and E2b, E3b and E4b will be assessed sequentially  using the same criterion 
(lower bound of the 2- sided 9 7.5% CI for the difference in percentages is greater 
than -10% ).
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The counts, percentages of participants with seroresponse at each time 
point, the difference in percentages ,and the associated 2 -sided 97.5% CIs will be 
provided.
6.2.1.7. SARS -CoV -2 SA NT After Dose 3 (BNT162b2- Experienced Participants)
6.2.1.7.1. Main Analyses
Estimands:
GMR of SA NT 1 month after 1dose of BNT162b2 SAto 1 month after the third dose 
of BNT162b2 (Section 2.1).
The difference in percentage s of participants with seroresponse to the SA strain at 
1 month after 1 dose of BNT162b2 SAand 1 month after the third dose of BNT162b2 
(Section 2.1).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221143
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 62
TMF Doc ID: 98.03Analy sis set: Dose 3booster evaluable and all -available immunogenicit ypopulations 
(Section 4).
Analy sis time points: 1 month after the third dose of BNT162b2 and 1 month after 1 dose 
of BNT162b2 SA.
Analy sis methodology : GMR and the associated 2 -sided 95% CI will be calculated in the 
same way as for the primary  endpoint N1a ( Section 6.1.2.3.1 ).The difference in 
percentages of participants with seroresponse and the associated 2 -sided 95% CI s will be 
calculated in the same way  as for the primary  endpoints N1b ( Section 6.1.2.4.1 ).
Intercurrent events and missing data: Titers below t he LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The GMTs at each time point, GMRs ,and the associa ted 2 -sided 95% 
CIs will be provided. The counts, percentages of participants with seroresponse at each 
time point, the difference in percentages ,and the associated 2 -sided 95% CIs will be 
provided.
Figures:
Empirical RCDCs will be provided for SARS -CoV -2 SA NTs at each time point for each 
vaccine group.
6.2.1.8. SARS -CoV -2 SA NT at 1 Month After Dose 4 vs Reference Strain NT at 1 
Month After Dose 2 in BNT162b2 -Experienced Participants
6.2.1.8.1. Main Analyses
Estimands:
GMR of SA NT 1 month after the second dose of BNT162b2 SAto the reference strain 
NT 1 month after the second dose of BNT162b2 in the same individuals 
(Section 2.1).
The difference in percentage s of participants with seroresponse to the SA strain at 1 
month after the second dose of BNT162b2 SAand seroresponse to the reference strain 
at 1 month after the second dose of BNT162b2 in the same individuals ( Section 2.1).
Analy sis set: Dose 4booster evaluable and all -available immunogenicit ypopulations 
(Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2 and 1 month after the 
second dose of BNT162b2 SA.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221144
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 63
TMF Doc ID: 98.03Analy sis methodology : GMR and the associated 2 -sided 95% CI will be calculated in the 
same way as for the primary  endpoint s E1a and E2a (Section 6.1.2.1 ).The difference in 
percentages of part icipants with seroresponse and the associated 2 -sided 95% CI s will be 
calculated in the same way  as for the primary  endpoints E1b and E2b ( Section 6.1.2.2.1 ).
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The GMTs at each time point, GMRs ,and the associated 2 -sided 95% 
CIs will be provided. The counts, percentages of participants with seroresponse at each 
time point, the difference in percentages ,and the associated 2 -sided 95% CIs will be 
provided.
Figures:
Empirical RCDCs will be provided for SARS -CoV -2 SA and r eference strain NTs at each 
time point.
6.2.1.9. SARS -CoV -2 SA NT at 1 Month After Dose 2 (BNT162b2 -Naïve Participants)
6.2.1.9.1. Main Analyses
Estimands:
N2a: GMR of SA NT 1 month after the second dose of BNT162b2 SAto 1 month after 
the second dose of BNT162b2 ( Section 2.1).
Analy sis set: Dose 2evaluable and all -available immunogenicit ypopulations ( Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2 and 1 month after the 
second dose of BNT162b2 SA.
Analy sis methodology : GMR and the associated 2 -sided 97.5% CI will be calculated in 
the same way  as for the primary  endpoint N1a ( Section 6.1.2.3.1 ).  Statistical superiority  
of N2a will be assessed if noninferiorit y of N1a is established. Superiorit y of N2a will be 
declared if the lower bound of the 2- sided 97.5% CI  for the GMR is greater than 1.
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adju sted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The GMTs at each time point, GMRs ,and the associated 2 -sided 
97.5% CI s will be provided.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221145
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 64
TMF Doc ID: 98.03Figures:
Empirical RCDCs will be provided for SARS -CoV -2 SA at e ach time point for each vaccine 
group .
6.2.1.10. Seroresponse to the SA Strain at 1 Month After Dose 2 (BNT162b2- Naïve 
Participants)
6.2.1.10.1. Main Analyses
Estimands:
N2b: The difference in percentages of participants with seroresponse to the SA strain 
at 1 month after the second dose of BNT162b2 SAand 1 month after the second dose 
of BNT162b2 (Section 2.1).
Analy sis set: Dose 2evaluable and all -available immunogenicit ypopulations ( Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2 and 1 month after the 
second dose of BNT162b2 SA.
Analy sis methodology : The difference in percentages of participants with seroresponse 
and the associated 2 -sided 97.5% CI s will be calculated in the same way as for the 
primary  endpoints E1b and E2b ( Section 6.1.2.2.1 ). Statistical superiority  of N2b will be 
assessed if noninferiority of N1b is established. Superiority  of N2b will be declared if the 
lower bound of the 2 -sided 97.5% CI for the difference in percentages of participants 
with seroresponse is greater than 0%.
Intercurrent events and missing data: Titers below the LLOQ or denot ed as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The counts, percentages of participants with seroresponse at each time 
point, the difference in percentages ,and the associated 2 -sided 97.5% CIs will be 
provided.
6.2.1.11. Reference Strain NT at 1 Month After Dose 2 (BNT162b2 -Naïve Participants)
6.2.1.11.1. Main Analyses
Estimands:
GMR of reference strain NT 1 month after the second dose of BNT162b2 SAto 1 
month after the second dose of BNT162b2 ( Section 2.1).
The difference in percenta ges of participants with seroresponse to the reference strain 
at 1 month after the second dose of BNT162b2 SAand 1 month after the second dose 
of BNT162b2 ( Section 2.1).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221146
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 65
TMF Doc ID: 98.03Analy sis set: Dose 2evaluable and all -available immunogenicit ypopulations ( Section 4).
Analy sis time points: 1 month after the second dose of BNT162b2 and 1 month after the 
second dose of BNT162b2 SA.
Analy sis methodology : GMR and the associated 2 -sided 95% CI will be calculate d in the 
same way as for the primary  endpoint N1a ( Section 6.1.2.3.1 ).  The difference in 
percentages of participants with seroresponse and the associated 2 -sided 95% CI s will be 
calculated in the same way  as for the primary  endpoints N1b ( Section 6.1.2.4.1 ).
Intercurrent events and missing data: Titers below the LLOQ or denoted as BLQ will be 
set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based upon 
additional data from the assay .  Missing data will not be imputed.
Reporting results: The GMTs at each time point, GMRs ,and the associated 2 -sided 95% 
CIs will be provided. The counts, percentages of participants with seroresponse at each 
time point, the difference in percentages ,and the associated 2 -sided 95% CIs will be 
provided.
Figures:
Empirical RCDCs will be provided for SARS -CoV -2 reference s train NTs at each time point 
for each vaccine group.
6.2.2. Vaccine Efficacy Endpoints (for Phase 2/3 Only)
6.2.2.1. COVID -19 Incidence per 1000 Person -Years of Follow -up
6.2.2.1.1. Main Analyses
Estimands:
100 × (1 – IRR) [ratio of confirmed COVID -19 illness from 14 day s after the second 
dose per 1000 person -years of follow -up in participants without evidence of infection 
(prior to 14 day s after receipt of the second dose) for the active vaccine group to the
placebo group ( Section 2.1)].
100 × (1 – IRR) [ratio of confirmed COVID -19 illness from 14 day s after the second 
dose per 1000 person -years of follow -up in participants with and without evidence of 
infection (prior to 14 days after receipt of the second dose) for the active vaccine 
group to the placebo group ( Section 2.1)].
Analy sis set: Evaluable efficacy  (14 day s) and all -available efficacy  populations 
(Section 4).
Analy sis time point: End of the surveillance period or at IAs if requested.
Analy sis methodology : the same method used for primary  VE endpoints will be applied 
(Section 6.1.3.1.1 ).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221147
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 66
TMF Doc ID: 98.03Intercurrent events and missing data: Missing efficacy data will not be imputed in the 
main anal yses.
Reporting results: the same output generated for primary  VE endpoints will be provided 
(Section 6.1.3.1.1 ).
6.2.2.2. Confirmed Severe COVID -19 Incidence per 1000 Person -Years of Follow -up
6.2.2.2.1. Main Analyses
Estimand s:
100 × (1 – IRR) [ ratio of confirmed severe COVID -19 illness from 7 day s and from 
14 day s after the second dose per 1000 person -years of follow -up in participants 
without evidence of infection (prior to 7 day s and14 day s after receipt of the second 
dose) for the active vaccine group to the placebo group ( Section 2.1)].
100 × (1 – IRR) [ ratio of confirmed severe COVID -19 illness from 7 day s and from 
14 day s after the second dose per 1000 person -years of follow -up in participants with 
and without evidence of infection (prior to 7 day s and 14 day s after receipt of the 
second dose) for the active vaccine group to the placebo group ( Section 2.1)].
Analy sis set: Evaluable efficacy  (7 day s and 14 day s) and all -available efficacy  
populations ( Section 4).
Analy sis time point: End of the surveillance period or at IAs if requested .
Analy sis methodology : the same method used for primary  VE endpoints will be applie d
(Section 6.1.3.1.1 ).
Intercurrent events and missing data: Missing efficacy data will not be imputed in the 
main anal yses.
Reporting results: the same output generated for primary  VE endpoints will be provided
(Sectio n 6.1.3.1.1 ).
6.2.2.2.2. Supplemental Analyses
All severe COVID -19 cases occurring after Dose 1 will be summarized descriptively .
After the final efficacy  analy ses at 164 first primary  cases, updated efficacy analyses will be 
performed for severe COVID -19incidence from 7 day s after the second dose with additional 
data accrued. The point estimate of VE in the blinded follow-up period and associated 
2-sided 95% CI will be derive d using the Clopper Pearson method adjusted for surveillance 
time, and the posterior probability  (ie, P[VE >30%|data]) will be provided.
In addition to the protocol defin ition of severe COVID -19, supportive analy ses using the 
CDC definition of severe COVID -19 will be performed .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221148
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 67
TMF Doc ID: 98.036.2.2.3. Confirmed COVID -19 Incidence per 1000 Person -Years of Follow -up
(According to the CDC -Defined Symptoms)
6.2.2.3.1. Main Analyses
Estimand s: 
100 × (1 – IRR) [ratio of confirmed COVID -19 illness according to the CDC- defined 
symptoms from 7 day sand from 14 day s after the second dose per 1000 person- years 
of follow -up in participants without evidence of infection (prior to 7 day s and 14 day s 
after receipt of the second dose) for the active vaccine group to the placebo group 
(Section 2.1)].
100 × (1 – IRR) [ratio of confirmed COVID -19 illness according to the CDC- defined 
symptoms from 7 day s and from 14 day s after the second dose per 1000 person- years 
of follow -up in participants with and without evidence of infection (prior to 7 days 
and 14 day s after receipt of the second dose) for the active vaccine group to the 
placebo group ( Section 2.1)].
Analy sis set: Evaluable efficacy  (7 day s and 14 day s)and all -available efficacy  
populations ( Section 4).
Analy sis time point: End of the surveillance period.
Analy sis methodology : Assessment of VE will be performed for centrally confirmed 
COVID -19 according to the CDC -defined s ymptoms from 7 day s and from 14 day s after 
the receipt of the second dose of study  intervention onwards, and will be estimated by  
100 × (1 – IRR), where IRR is the calculated ratio of COVID -19 illness rate according to 
the CDC -defined s ymptoms per 1000 person -years of follow -up in the active vaccine 
group to the corresponding illness rate in the placebo group after the second dose.  
The 2 -sided 95% CI for VE will be derived using the Clopper -Pearson method adjusted 
for surveil lance time .
Intercurrent events and missing data: Missing efficacy data will not be imputed in the 
main anal yses.
Reporting results: VE and the associated 2 -sided 95% CI s derived using the 
Clopper -Pearson method adjusted for surveillance time will be provi ded.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221149
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 68
TMF Doc ID: 98.036.2.2.4. Incidence of Asymptomatic SARS -CoV -2 Infection per 1000 Person -Years of 
Follow -up (According to the N-B inding Antibody S eroconversion )
6.2.2.4.1. Main Analyses
Estimands: 
100 × (1 – IRR) [ratio of asymptomatic SARS -CoV -2 infection per 1000 
person -years of follow- up based on N -binding antibody  seroconversion in 
participants wit h no serological or virological evidence of past SARS -CoV -2 
infection or confirmed COVID -19for the active vaccine group to the placebo group 
(Section 2.1)].
Analy sis set: Evaluable efficacy  (seroconversion) and all -available efficacy populations 
(Section 4).
Analy sis time point: End of the surveillance period .
Analy sis methodology : An as ymptomatic case ( Appendix 4) is defined as positive 
N-binding antibody  at a post–Dose 2 visit in participants without serological evidence of 
infection (determined b y negative N -binding antibody ) at Visit 1 or virological evidence 
of infection ( determined by  negative NAAT at Visit 1 and Visit 2 and at the time of a 
potential COVID -19 illness ). A secondary  definition will be applied without the 
requirement for a negative NAAT at Visit 2. VE will be estimated by  100 × (1 -IRR), 
where IRR is the calculated rat io of as ymptomatic infection per 1000 person -years of 
follow -up in the active vaccine group to the corresponding infection in the placebo group. 
The 2 -sided 95% CI for VE will be derived using the Clopper -Pearson method adjusted 
for surveillance time. The VEis demonstrated if the lower bound of the 2- sided 95% CI  
for VE is greater than 20%. The analy sis of the primary  definition of as ymptomatic cases 
will be based on the evaluable efficacy  (seroconversion) population and the Dose 2 
all-available effica cy population. The anal ysis of the secondary  definition of 
asymptomatic cases will be based on the Dose 1 all -available efficacy  population .
Intercurrent events and missing data: Missing efficacy data will not be imputed in the 
main anal yses.
Reporting re sults: VE and the associated 2- sided 95% CI s derived using the 
Clopper -Pearson method adjusted for surveillance time will be provided.
6.2.2.4.2. Supplemental Analyses
Descriptive summary  of VE against as ymptomatic infection over different time intervals 
(ie, prior to 1 month after Dose 2, from 1 month after Dose 2 onward), along with the 
associated 2- sided 95% CI , will be calculated using the same method as above .
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221150
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 69
TMF Doc ID: 98.036.2.2.5. Incidence of Asymptomatic SARS -CoV -2 Infection per 1000 Person -Years of 
Follow -up (According to the C entral Laboratory–Confirmed NAAT)
6.2.2.5.1. Main Analyses
Estimands: 
100 × (1 – IRR) [ratio of asymptomatic SARS -CoV -2 infection per 1000 
person -years of follow -up based on central laboratory –confirmed NAAT in 
participants without serological or virological evidence of infection (up to the start of 
the asy mptomatic surveillance period ) for the active vaccine group to the placebo
group ( Section 2.1)].
Analy sis set: Evaluable efficacy  (asymptomatic surveillance ) and all -available efficacy  
populations ( Section 4)and only participants who consented to participate in the 
asymptomatic surveillance .
Analy sis time point: End of the surveillance period.
Analy sis methodology : An as ymptomatic case definition based on central laboratory –
confirmed NAAT can be found in Appendix 5.VE will be estimated by  100 × (1 - IRR), 
where IRR is the calculated ratio of as ymptomatic infection per 1000 person- years of 
follow -up in the active vaccine group to the corresponding infection in the placebo group. 
The 2 -sided 95% CI for VE will be derived using the Clopper -Pearson method adjusted 
for surveillance time. The success criterion is met if lower bound of the 2- sided 95% CI 
for VE is greater than 20%.
Intercurrent events and missing data: Missing efficacy data will not be imputed in the 
main anal yses.
Reporting results: VE and the associated 2 -sided 95% CI s der ived using the 
Clopper -Pearson method adjusted for surveillance time will be provided.
6.3.Exploratory Endpoints
6.3.1. Safety Endpoints (for Phase 1 Boostability Assessment Only)
6.3.1.1. Local Reactions 
6.3.1.1.1. Main Analysis
Estimand: The percentage of participants reporting local reactions (redness, swelling, and 
pain at the injection site) within 7 day s after Dose 3 (Section 2.1).
Analy sis set: Phase 1 partic ipants who received a third dose of BNT162b2 6 to 12 months 
after the second dose of either BNT162b1 or BNT162b2.
Analy sis time point: Within 7 day s after Dose 3 .
Analy sis methodology : Descriptive statistics (Section 5.2.1 ).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221151
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 70
TMF Doc ID: 98.03Intercurrent events and missing data: The participants without any  e-diary data 
throughout the 7 day s after vaccination will be excluded from the anal ysis at that 
particular vaccination; missing values will not be imputed.
Reporting results: Descriptive statistics for each and any  local reaction after Dose 3 by 
initial vaccine and age group will be presented b y maximum severity  and cumulatively  
across severit y levels.  Confirmed e -diary  errors will be excluded from the anal ysis.  
Descriptive summary  statistics will include counts and percentages of participants with 
the indicated endpoint and the associated 2 -sided Clopper -Pearson 95% CIs.
6.3.1.1.2. Supplementary Analyses
To support the assessment of local reactions, the following endpoints (as defined in
Section 3.1.1.1 ) will be summarized with the same anal ysis time point andanalysis 
population , analy sis methodology , and appropriate reporting results.  Confirmed e- diary  
errors will be excluded from these an alyses.
Duration (day s) of each local reaction after each dose.
Onset day  of each local reaction after each dose.
These continuous endpoints will be summarized by display ing n, mean, median, standard 
deviation, minimum, and maximum for each initial vaccine and age group .
Figures:
Bar charts with the proportions of participants for each local reaction throughout 7 day s will 
be plotted for each initial vaccine and age group. The bars will be divided into severity  
categories to highlight the proportions of pa rticipants by  maximum severity.
6.3.1.2. Systemic Events
6.3.1.2.1. Main Analysis
Estimand: The percentage of participants reporting s ystemic events (fever, fatigue, 
headache, chills, vomiting, diarrhea, new or worsened muscle pain, and new or worsened 
joint pain) within 7 da ys after Dose 3 (Section 2.1).
Analy sis set: Phase 1 participants who received a third dose of BNT162b2 6 to 12 months 
after the second dose of either BNT162b1 or BNT162b2.
Analy sis time point: Within 7 day s after Dose 3 .
Analy sis methodology : Descriptive statistics ( Section 5.2.1 ).
Intercurrent events and missing data: The participants without any  e-diary data 
throughout the 7 day s after vaccination will be excluded from the anal ysis at that 
particular vaccination; missing values will not be imputed.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221152
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 71
TMF Doc ID: 98.03Reporting r esults: Descriptive statistics for each sy stemic event after Dose 3 in each 
initial vaccine and age group will be presented b y maximum severity  and cumulatively  
across severit y levels.  Descriptive summary statistics will include counts and 
percentages of participants with the indicated endpoint and the associated 2 -sided 
Clopper -Pearson 95% CIs.
6.3.1.2.2. Supplementary Analyses
The following endpoints for assessment of s ystemic events will be summarized similarly  to 
the assessment of local reactions:
Duration of eac h systemic event after each dose.
Onset day  of each s ystemic event after each dose.
These continuous endpoints will be summarized by display ing n, mean, median, standard 
deviation, minimum, and maximum for each initial vaccine and age group.
The use of ant ipyretic medication (see Section 3.1.1.3 ) will be summarized similarly  to 
systemic events, except that there is no severity  level associ ated with the use of antipy retic 
medication.
Figures: 
Bar charts with the proportions of participant s reporting each s ystemic event throughout 
7days after Dose 3 will be plotted for each initial vaccine and age group.  The bars will be 
divided into sever ity categories to highlight the proportions of participant s by severit y.
6.3.1.3. Adverse Events
6.3.1.3.1. Main Analysis
Estimand: The percentage of participants reporting AEs from Dose 3to 1 month after 
Dose 3 (Section 2.1).
Analy sis set: Phase 1 participants who received a third dose of BNT162b2 6 to 12 months 
after the second dose of either BNT162b1 or BNT162b2.
Analy sis time poin t: Dose 3 to 1 month after Dose 3 .
Analy sis methodology : Descriptive statistics ( Section 5.2.1 ). 
Intercurrent events and missing data: Partial AE dates will be imputed using the Pfizer 
standard algorithm.
Reporting results: AEs will be categorized according to MedDRA terms. Descriptive 
summary  statistics (counts, percentages, and associated Clopper -Pearson 95% CI s) will 
be provided fo r any AEs for each initial vaccine and age group.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221153
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 72
TMF Doc ID: 98.036.3.1.3.2. Supplementary Analyses
Immediate AEs (within the first 30 minutes after Dose 3 ) will also be summarized for each 
initial vaccine and age group.  
6.3.1.4. Serious Adverse Events
6.3.1.4.1. Main Analyses
Estimand:  The percentage of participants reporting SAEs from Dose 3to 1month after 
Dose 3 (Section 2.1).
Analy sis set: Phase 1 part icipants who received a third dose of BNT162b2 6 to 12 months 
after the second dose of either BNT162b1 or BNT162b2.
Analy sis time point: Dose 3 to 1 month after Dose 3.
Analy sis methodology : Descriptive statistics ( Section 5.2.1 ).
Intercurrent events and missing data: Partial SAE dates will be imputed using the Pfizer 
standard algorithm.
Reporting results: SAEs will be categorized according to MedDRA terms.  Counts, 
percentages, and the associated Clopper -Pearson 95% CI s of SAEs from Dose 3to 
1month after Dose 3 will be provided for each initial vaccine and age group.
6.3.2. Vaccine Efficacy En dpoints (for Phase 2/3 Only)
6.3.2.1. COVID -19 Incidence per 1000 Person -Years of Blinded Follow -up
6.3.2.1.1. Main Analyses
Estimands:
100 × (1 – IRR) [ratio of confirmed COVID -19 illness based on central laboratory  or 
locally  confirmed NAAT from 7 days after the second dose through the blinded 
follow -up period per 1000 person -years of follow -up in participants without evidence 
of infection (prior to 7days after receipt of the second dose) for the active vaccine 
group to the placebo group ( Section 2.1)].
100 × (1 – IRR) [ratio of confirmed COVID -19 illness based on central laboratory  or 
locally  confirmed NAAT from 7 days after the second dose through the blinded 
follow -up period per 1000 person -years of follow -up in participants with and without 
evidence of infection for the active vaccine group to the placebo group ( Section 2.1)].
Analy sis set: Evaluable efficacy  (7days) and all -available efficacy  populations 
(Section 4).
Analy sis time point: End of the surveillance period (blinded follow -up).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221154
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 73
TMF Doc ID: 98.03Analy sis methodology :After the primary  objectives are met at the final analy sis of at 
least 164 first primary  cases, the study  will continue with blinded follow -up until the 
participant is unblinded at the time of being eligible for receipt of BNT162b2 according 
to recommendations detailed separately , and available in the electronic study  reference 
portal, or no later than at approximately  Visit 4. A descriptive update of VE will be 
provided with additional follow- up data. VE=100 × (1 – IRR) will be estimated with 
confirmed COVID -19 illness from 7 day s after the second dose through the blinded 
follow -up period.  The 2 -sided 95% CI  for VE will be deri ved using the Clopper- Pearson 
method adjusted for surveillance time.
Intercurrent events and missing data: Missing efficacy data will not be imputed in the 
main anal yses.
Reporting results: VE and the associated 2 -sided 95% CI s derived using the 
Clopper -Pearson method adjusted for surveillance time .
6.3.2.1.2. Supportive Analyses
Supportive analy sis of time to confirmed COVID- 19 illness will be performed using 
Kaplan -Meier cumulative incidence curves. Participants who were randomized to placebo 
will be censored at t he time of receipt of BNT162b2. An RPSFT model may be explored to 
reconstruct data for the control arm.
VE at different follow -up time intervals and against different variant strains may  be assessed.
6.3.2.2. COVID -19 Incidence per 1000 Person -Years of Follow -up
6.3.2.2.1. M ain Analyses
Estimands:
COVID -19 incidence based on central laboratory  or locally  confirmed NAAT from 7 
days after the second dose per 1000 person -years of follow -up in participants without 
evidence of infection (prior to 7 day s after receipt of the second BNT162b2 
vaccination) who received BNT162b2 at initial randomization or subsequently  
(Section 2.1).
COVID -19 incidence based on central laboratory  or locally  confirmed NAAT from 7 
days after the second dose per 1000 person -years of follow -up in participants with 
and without evidence of infection who received BNT162b2 at initia l randomization or 
subsequently  (Section 2.1).
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221155
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 74
TMF Doc ID: 98.03Analy sis set: Evaluable efficacy  (7 day s) and all -available efficacy  populations 
(Section 4).For participants who were randomized to placebo and su bsequentl yreceived 
BNT162b2 after being eligible according to recommendations detailed separatel y, and 
available in the electronic study  reference portal, or no later than at approximately Visit 
4, the time of receipt of BNT162b2 will be reconsidered as baseline . All rules for 
determining evaluable efficacy  and all-available efficacy  population swill be similarly  
applied.
Analy sis time point: End of the surveillance period.
Analy sis methodology : Incidence rate (per 1000 person -years of follow -up) and exact 
2-sided 95% CI based on Poisson distribution (Section 5.2.2 ) for confirmed COVID-19 
illness from 7 day s after the second BNT162b2 vaccinatio n will be provided for 
participants who received BNT162b2 at initial randomization and subsequently .
Kaplan -Meier cumulative incidence of COVID- 19 cases over time will be plotted.
Intercurrent events and missing data: Missing efficacy  data will not be im puted in the 
main anal yses.
Reporting results: Incidence rate and the associated 2 -sided 95% CIs, and Kaplan -Meier 
cumulative incidence curve will be provided .
6.3.2.3. Incidence of Asymptomatic SARS -CoV -2 Infection per 1000 Person -Years of 
Follow -up (According to theN-Binding Antibody S eroconversion )
6.3.2.3.1. Main Analyses
Estimands: 
Incidence of as ymptomatic SARS -CoV -2 infection through the entire study of 
follow -up period per 1000 person- years of follow -up based on N -binding antibody  
seroconversion in participants who received BNT162b2 and who have no serological 
or virological evidence of past SARS- CoV -2 infection or confirmed COVID -19.
Analy sis set: Evaluable efficacy  and all -available efficacy  populations 
(Section 4).
Analy sis time point: End of the surveillance period .
Analy sis methodology : Incidence rate (per 1000 person -years of follow -up) and exact 
2-sided 95% CI based on Poisson distribution (Section 5.2.2 ) for asymptomatic infection
will be provided for participants who received BNT162b2.
Intercurrent events and missing data: Missing efficacy data will not be imputed in the 
main anal yses.
Repor ting results: I ncidence rate and the associated 2- sided 95% C Iswill be provided.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221156
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 75
TMF Doc ID: 98.036.3.2.4. Incidence of Asymptomatic SARS -CoV -2 Infection per 1000 Person -Years of 
Follow -up (According to the C entral L aboratory–Confirmed NAAT)
6.3.2.4.1. Main Analyses
Estimands: 
100 × (1 – IRR) [ratio of asymptomatic SARS -CoV -2 infection per 1000 
person -years of follow -up based on central laboratory –confirmed NAAT in 
participants with serological or virological evidence of past infection (up to the start 
of the as ymptomatic surveillance perio d) for the active vaccine group to the placebo
group ( Section 2.1)].
Analy sis set: Evaluable efficacy  (asymptomatic surveillance) and all -available efficacy  
populations ( Section 4)and only  participants who are consented to participate in the 
asymptomatic surveillance .
Analy sis time point: End of the surveillance period.
Analy sis methodology : VE will be estimated by  100 × (1 - IRR), where IRR is the 
calculated ratio of as ymptomatic infection per 1000 person -years of follow -up in the 
active vaccine group to the corresponding infection in the placebo group. The 2 -sided 
95% CI  for VE will be derived using the Clopper -Pearson method adjusted f or 
surveillance time. 
Intercurrent events and missing data: Missing efficacy data will not be imputed in the 
main anal yses.
Reporting results: VE and the associated 2 -sided 95% CI s derived using the 
Clopper -Pearson method adjusted for surveillance time wi ll be provided.
6.3.3. Immunogenicity Endpoints
6.3.3.1. SARS -CoV -2 Reference -Strain Neutralizing Titers, SARS -CoV -2 SA -Variant 
Neutralizing T iters ,and Full-Length S -Binding or S1 -Binding IgG Levels (Phase 1)
6.3.3.1.1. Main Analyses
Estimands:
GMTs /GMCs (Section 2.1).
GMFR from before Dose 3 to each subsequent time point , ie,7 day s and 1 month 
after Dose 3 (Section 2.1 ).
Analy sis set: Phase 1 participants who received a third dose of BNT162b2 6 to 12 months 
after the second dose of either BNT162b 1 or BNT162b2.
Analy sis time points: At Dose 3 and 7 day s and 1 mont h after Dose 3.
090177e196897024\Approved\Approved On: 17-Mar-2021 16:05 (GMT) 
FDA-CBER-2021-5683-0221157
Protocol C4591001 (PF-07302048 ) Statistical Analysis Plan
 CONFIDENTIAL
Page 76
TMF Doc ID: 98.03Analy sis methodology : GMs and the associated 2 -sided CI s will be derived by  
calculating means and CIs on the natural log scale based on Student’s t- distribution, and 
then exponentiating the results ( Section 5.2.3.1 ).  GMFRs will be limited to participants 
with nonmissing values prior to Dose 3 and the subsequent time point .  The GMFR will 
be calculated by exponentiating the mean of the difference of logarithmically  transformed 
assay  results (later time point – earlier time point).  Two -sided CIs will be obtained by  
calculating CIs using Student’s t -distribution for the mean difference of the 
logarithmicall y transformed a ssay results and exponentiating the confidence limits 
(Section 5.2.3.2 ).  
Intercurrent events and missing data: Concentrations below the LLOQ or denoted as BLQ 
will be set to 0.5 × LLOQ for analy sis.  However, this calculation may  be adjusted based 
upon additional data from the assay .  Missing data will not be imputed.
Reporting results: the GMTs /GMCs, GMFRs ,and the associated 2 -sided 95% CI s will be 
provided at each time point by initial vaccine and age group.
6.3.3.2. SARS -CoV -2 Reference -Strain Neutralizing Titer sand SARS -CoV -2 SA-
Variant Neutralizing Titers at 1 Month After Dose 3 vsSARS -CoV -2 Reference -Strain 
Neutralizing Titers at 1 Mon th After Dose 2 (Phase 1)
6.3.3.2.1. Main Analyses
Estimands:
GMR of SARS -CoV -2 reference -strain neutralizing titers at 1 month after Dose 3 to 
SARS -CoV -2 reference -strain neutralizing titers at 1 month after Dose 2
(Section 2.1).
GMR of SARS -CoV -2 SA-variant neutralizing titers at 1 month after Dose 3 to 
SARS -CoV -2 reference -strain neutralizing titers at 1 month after Dose 2 
(Section 2.1).
Analy sis set: Phase 1 participants who received a third dose of BNT162b2 6 to 12 months 
after the second dose of either BNT162b1 o r BNT162b2 .
Analy sis time points: 1 month after Dose 3 .
Analy sis methodology : GMRs will be limited to participants with nonmissing values at 
both time points and provided by  initial vaccine and age group .  The GMR will be 
calculated as the mean of the difference of logarithmically  transformed assay  results 
(eg,SARS -CoV -2 SA-variant neutralizing titers at 1 month after Dose 3 minus 
reference -strain titers at 1 month after Dose 2 for each participant) and exponentiating the 
mean (Section 5.2.3.3).  Two- sided CIs will be obtained by  calculating CIs using 
Student’s t -distribution for the mean difference of the logarithmically  transformed assay  
results and exponentiating the confidence limits ( 
…[truncated]