Document text
Update: Rapid Cycle Analysis of RSV
Vaccines in Older Adults
Jim Donahue, PhD DVM
Marshfield Clinic Research Institute
Presented to the ACIP
June 26, 2024
1
RSV RCA Interim Report – June 26, 2024
•Description of Vaccine Safety Datalink and RCA
•Review of surveillance methods
•Descriptive analysis
•Sequential analysis
̶Immune thrombocytopenia (ITP)
̶Guillain -Barré syndrome (GBS)
̶Atrial fibrillation
2
Vaccine Safety Datalink 2024
*
3•Collaborative project between CDC and 13 integrated healthcare organizations
•Data on ~13.5 million persons per year
•Conducts rigorous vaccine safety studies and near -real-time monitoring
VSD Rapid Cycle Analyses (RCA)
4•Permits rapid assessment of vaccine safety
-Near real -time data for weekly, biweekly, or monthly analyses
•Outcome incidence in vaccinated persons compared to outcomes
incidence in a comparator group
-Outcomes are pre -specified
•Sequential analytic methods used to detect ‘statistical signals’
while maintaining a pre -defined type I error rate
-Type I error = mistakenly reject the null hypothesis (“false positive”)
•Statistical signals are interpreted as potential associations
Objectives for RSV RCA in Older Adults
5•Monitor RSV vaccine uptake
•Monitor the occurrence of pre -specified outcomes following RSV
vaccination
•Conduct near real -time surveillance of pre -specified outcomes
using RCA methods
Study Design and Population
6•Near real -time surveillance among a prospective cohort of
persons ≥60 years old who received an RSV vaccine
•Member of a participating VSD infrastructure site
•Data are extracted every week and analyses are biweekly
•Surveillance period: 8/1/2023 through 5/31/2025
-Sequential analyses started in March 2024
•Project ends September 2025
Sequential Analysis Methods
7•Biweekly analysis includes a sequential test of the one -sided null
hypothesis that the vaccine does not increase the risk in the risk interval
•A ‘statistical signal’ occurs when the analysis produces a one -sided P value
that is less than a pre -specified threshold
•The signal threshold is determined from an alpha -spending plan that keeps
the overall chance of a Type 1 error <0.05 during the surveillance period
•Formal sequential analysis stops after a signal, but surveillance continues
•Design and analysis is analogous to that used in the VSD RCA of COVID -19
vaccines (Klein N, et al. JAMA 2021; 326:1390 –9)
Risk and Comparison Intervals
8•Primary risk interval for all outcomes will be 1 -21 days following
RSV vaccination except anaphylaxis and CIDP*
-Primary comparison interval of 43 -63 days
-Secondary comparison interval of 22 -42 days
•Secondary risk interval of 1 -42 days
-Comparison interval of 43 -84 days
•Anaphylaxis and CIDP will be descriptively monitored only
-Anaphylaxis (0 -1 days after RSV vaccination)
-CIDP (1 -84 days after RSV vaccination)
*chronic inflammatory demyelinating polyneuropathy
Analysis with Vaccinated Concurrent Comparators
9•Comparators are RSV vaccinees, who on the same day as the exposed case
in a risk interval, were in the same stratum (e.g., age, sex, race/ethnicity,
VSD site), but in a comparison interval
•Outcome incidence is calculated during the risk interval and compared
with incidence in the comparison interval
-Relative risk estimates are computed with nominal 95% confidence intervals
-Adjusted for calendar day, age group, sex, race/ethnicity, VSD site
•Advantages of vaccinated concurrent comparators compared to
unvaccinated or historical comparators
-Permits adjustment for potential biases due to calendar time, site, and demographic
factors
-Less confounding by indication (e.g., persons with chronic illness more likely to seek
RSV vaccination and may be at increased risk of atrial fibrillation)
10RSV vaccinated
May 30th
RSV vaccinated
April 19th June 3rdJune 3rd1 21
43 63On each calendar day that an outcome occurs in a
vaccinee (e.g., June 3rd ), we compare vaccinees in
their risk interval (1 -21 days) with similar vaccinees
in their comparison interval (43 -63 days).
‘Similar’ means that people were in the same age
group, sex, race/ethnicity, and VSD site.
Inherently adjusts for calendar time.Risk interval 1 -21 days post -vaccination
Comparison interval 43 -63 days post -vaccination
Ref: Klein N, et al. Kaiser Permanente Northern CaliforniaVaccinated Concurrent Comparator Design
Four Exposure Groups
11GSK with simultaneous vaccination of another vaccine*
GSK without simultaneous vaccination
Pfizer with simultaneous vaccination of another vaccine*
Pfizer without simultaneous vaccination
*Non -RSV vaccines typically include routine, age -appropriate vaccines such as
COVID -19, influenza, RZV, PCV20/15, PPSV23, Td/Tdap
Pre-specified Outcomes (n=14)
12Outcome Setting1Primary risk
interval (days)2
Acute disseminated encephalomyelitis (ADEM)3E, I 1-21
Acute myocardial infarction (AMI) E, I 1-21
Anaphylaxis3 E, I 0-1
Atrial fibrillation E, I, O, T 1-21
Bell’s palsy E, I, O, T 1-21
Chronic inflammatory demyelinating polyneuropathy (CIDP)3 E, I, O, T 1-84
Deep vein thrombosis (DVT) E, I, O, T 1-21
Encephalitis / myelitis / encephalomyelitis (not ADEM or TM) E, I 1-21
Guillain -Barré syndrome (GBS)3 E, I 1-21
Immune thrombocytopenia (ITP) E, I, O, T 1-21
Myocarditis / pericarditis E, I 1-21
Pulmonary embolism (PE) E, I 1-21
Stroke E, I 1-21
Transverse myelitis (TM)3 E, I 1-21
1E=Emergency department; I=Inpatient; O=Outpatient; T=Telehealth
2All outcomes also have a secondary risk interval of 1 -42 days after vaccination, except anaphylaxis and CIDP , which are descript ively monitored only .
3Chart review regardless of whether there is a statistical signal; sequential analyses will use only chart -confirmed cases (ADEM, GBS, and TM).
RSV Vaccines Administered in VSD,
8/1/2023 –5/25/2024
13•NOTE: All subsequent analyses and slides exclude RSV vaccines in the
‘Unspecified’ manufacturer categoryGSK Pfizer UnspecTotal
N % N % N %
338,290 87.7 47,287 12.3 152 0.0 385,729
14050001000015000200002500030000
07-30-23
08-06-23
08-13-23
08-20-23
08-27-23
09-03-23
09-10-23
09-17-23
09-24-23
10-01-23
10-08-23
10-15-23
10-22-23
10-29-23
11-05-23
11-12-23
11-19-23
11-26-23
12-03-23
12-10-23
12-17-23
12-24-23
12-31-23
01-07-24
01-14-24
01-21-24
01-28-24
02-04-24
02-11-24
02-18-24
02-25-24
03-03-24
03-10-24
03-17-24
03-24-24
03-31-24
04-07-24
04-14-24
04-21-24
04-28-24
05-05-24
05-12-24
05-19-24Vaccines RSV Vaccinations by Manufacturer and Week of
Administration
GSK
Pfizer
Total
WeekStart of
sequential
analysis
Results of Sequential Analysis
Six runs —data through 5/25/2024
15
VSD Outcomes Setting1Signal
(Y/N)
Acute disseminated encephalomyelitis (ADEM) E, I N
Acute myocardial infarction (AMI) E, I N
Atrial fibrillation (AF) E, I, O, T N
Bell’s palsy (BP) E, I, O, T N
Deep vein thrombosis (DVT) E, I N
Encephalitis / myelitis / encephalomyelitis (ENCEPH) E, I N
Guillain -Barré syndrome (GBS) E, I N
Immune thrombocytopenia (ITP) E, I, O, T Y
Myocarditis / pericarditis ( MYOC ) E, I N
Stroke (STK) E, I N
Transverse myelitis (TM) E, I N
Pulmonary embolism (PE) E, I N16RSV RCA Vaccine Statistical Signals in VSD (5/25/2024)
1E = ED, I = Inpatient, O = Outpatient, T=Telehealth
Analysis Parameters Signal Information and Informative Counts Nominal AnalysisSequential
Test
Outcome
EventRisk
Interval
DaysComp
Interval
DaysVaccine
TypeSignaled
in a
Prior
Run1New
Sequential
Analysis
Signal2Events
in Risk
IntervalEvents in
Comp
IntervalAdjusted
Expected
Events in
Risk
IntervalAdjusted
Rate
Ratio395%
Confidence
Interval42-sided
P Value1-sided
P Value5
ITP 1-21 43-63 GSK w simul n/a No 2 1 0.6 3.08 0.23-92.44 0.408 0.347
GSK wo simul n/a No 19 8 8.1 2.35 0.99-5.97 0.054 0.040
Pfizer w simul n/a No 0 1 0.2 0.00 0.00-81.18 0.810 0.810
Pfizer wo simul n/a No 2 2 1.8 1.10 0.11-11.27 0.931 0.659
22-42 GSK w simul n/a No 3 2 1.4 2.21 0.31-19.61 0.432 0.340
GSK wo simul n/a Yes 19 6 6.3 3.04 1.22-8.47 0.016 0.012
Pfizer w simul n/a No 0 1 1.1 0.00 0.00-16.74 0.468 0.468
Pfizer wo simul n/a No 2 1 0.7 2.67 0.20-78.89 0.472 0.394
1-42 43-84 GSK w simul n/a No 4 7 4.3 0.93 0.22-3.42 0.931 0.662
GSK wo simul n/a No 25 14 15.6 1.60 0.80-3.28 0.186 0.121
Pfizer w simul n/a No 1 3 3.5 0.29 0.01-4.25 0.428 0.957
Pfizer wo simul n/a No 3 3 3.4 0.88 0.14-5.47 0.888 0.718VSD RCA of RSV Vaccine in Older Adults
Surveillance Initiated on 01AUG2023, Analyses Based on Data Through 16MAR2024 (run #1, week #1836)
Concurrent Comparator Sequential Analysis Signal Assessment for Immune Thrombocytopenia
---Age:60+ yrs---
17
Analysis Parameters Signal Information and Informative Counts Nominal AnalysisSequential
Test
Outcome
EventRisk
Interval
DaysComp
Interval
DaysVaccine
TypeSignaled
in a
Prior
Run1New
Sequential
Analysis
Signal2Events
in Risk
IntervalEvents in
Comp
IntervalAdjusted
Expected
Events in
Risk
IntervalAdjusted
Rate
Ratio395%
Confidence
Interval42-sided
P Value1-sided
P Value5
ITP 1-21 43-63 GSK w simul n/a No 2 1 0.6 3.08 0.23-92.44 0.408 0.347
GSK wo simul n/a No 19 8 8.1 2.35 0.99-5.97 0.054 0.040
Pfizer w simul n/a No 0 1 0.2 0.00 0.00-81.18 0.810 0.810
Pfizer wo simul n/a No 2 2 1.8 1.10 0.11-11.27 0.931 0.659
22-42 GSK w simul n/a No 3 2 1.4 2.21 0.31-19.61 0.432 0.340
GSK wo simul n/a Yes 19 6 6.3 3.04 1.22-8.47 0.016 0.012
Pfizer w simul n/a No 0 1 1.1 0.00 0.00-16.74 0.468 0.468
Pfizer wo simul n/a No 2 1 0.7 2.67 0.20-78.89 0.472 0.394
1-42 43-84 GSK w simul n/a No 4 7 4.3 0.93 0.22-3.42 0.931 0.662
GSK wo simul n/a No 25 14 15.6 1.60 0.80-3.28 0.186 0.121
Pfizer w simul n/a No 1 3 3.5 0.29 0.01-4.25 0.428 0.957
Pfizer wo simul n/a No 3 3 3.4 0.88 0.14-5.47 0.888 0.718VSD RCA of RSV Vaccine in Older Adults
Surveillance Initiated on 01AUG2023, Analyses Based on Data Through 16MAR2024 (run #1, week #1836)
Concurrent Comparator Sequential Analysis Signal Assessment for Immune Thrombocytopenia
---Age:60+ yrs---
18
•ITP signal for GSK vaccine without simultaneous vaccination in
1-21 day risk interval versus 22 -42 day comparison interval
•Quick medical record reviews
-Of the 19 cases in the risk interval, 4 were incident ITP*
-Of the 14 cases in comparison intervals, 2 were incident ITP , 1 in the
22-42 day interval and 1 in the 43 -63 day interval
•After quick review: 4 cases of ITP in the primary risk interval, 1
case each in the 22 -42 and 43 -63 day comparison intervals
•Plan to do more detailed chart review of ITP cases going forwardITP Statistical Signal and Rapid Review
19*New cases of ITP relative to RSV vaccination.
GBS Cases Following RSV Vaccination
--No Statistical Signal --
20
GBS Cases 1 -84 Days after RSV Vaccination
•GBS cases identified electronically, then receive medical record review,
and presumptive cases are adjudicated by two reviewers
•Cases defined using Brighton Level (BL) criteria*
•7 cases of GBS (BL 1 -3) following any RSV vaccination
̶5 following GSK – onset days 6, 10, 31, 54, 76
̶2 following Pfizer – onset days 9, 46
•2 BL 4 cases following GSKAutomated
CasesCompleted
Chart ReviewChart Confirmed
CasesNot
GBSCases Pending
Review
12 11 9 2 1
21*Sejvar J, et al. Vaccine. 2011;29(3):599 -612.
22Analysis Parameters Signal Information and Informative Counts Nominal AnalysesSequential
Test
Outcome
EventRisk
Interval
DaysComp
Interval
Days Vaccine TypeSignaled
in a Prior
Run2New
Sequential
Analysis
Signal3Events
in Risk
IntervalEvents in
Comparison
IntervalAdjusted
Expected
Events in Risk
IntervalAdjusted
Rate
Ratio495%
Confidence
Interval52-sided
P value1-sided
P value6
GBS11-21 43-63 GSK wo simul No No 2 1 0.9 2.33 0.10-93.95 0.603 0.515
Pfizer wo simul No No 1 0 0.0 ne 0.05-ne 0.496 0.496
22-42 GSK wo simul No No 2 1 1.3 1.55 0.11-47.77 0.775 0.600
Pfizer wo simul No No 1 0 0.0 ne 0.07-ne 0.439 0.439
1-42 43-84 GSK wo simul No No 3 1 1.4 2.17 0.16-74.98 0.605 0.501
Pfizer wo simul No No 1 0 0.0 ne 0.04-ne 0.557 0.557---Age:60+ yrs---VSD RCA of RSV Vaccine in Older Adults
Surveillance Initiated on 01AUG2023, Analyses Based on Data Through 25MAY2024 (run #6, week #1846)
Concurrent Comparator Sequential Analysis Signal Assessment for Guillain -Barre Syndrome
(1) Chart -confirmed cases only, Brighton level 1 -3
(2) n/a = not applicable
(3) No prior signal, at least 2 events in the risk interval, 1 -sided P value < 0.014
(4) Adjusted for calendar date, VSD site, age category, sex, and race/ethnicity
(5) ne = not estimable
(6) Red: new sequential analysis signal, Yellow: 1 -sided P value < 0.014 but already signaled in a prior run
Crude GBS Incidence Rates after RSV Vaccination1
23Vaccine Risk interval# of
Cases# of
Doses2Rate Per Million
(95% CI)Rate per 100,000 PY
(95% CI)
GSK 1-21 days 2 323929 6.2 (0.7 – 22.3) 10.7 (1.3 – 38.8)
GSK 1-42 days 3 323929 9.3 (1.9 – 27.1) 8.1 (1.7 – 23.5)
Pfizer 1-21 days 1 45162 22.1 (0.6 – 123.4) 38.5 (1.0 – 214.6)
Pfizer 1-42 days 1 45162 22.1 (0.6 – 123.4) 19.3 (0.5 – 107.3)
VaccineComparison
interval# of
Cases# of
Doses3Rate Per Million
(95% CI)Rate per 100,000 PY
(95% CI)
GSK 43-84 days 2 301547 6.6 (0.8 – 24.0) 5.8 (0.7 -20.9)
Pfizer 43-84 days 1 41031 24.4 (0.6 – 135.8) 21.2 (0.5 – 118.1)
1Cases restricted to chart confirmed, Brighton Level 1 -3
2Vaccines administered 8 -1-23 through 4 -13-24
3Vaccines administered 8 -1-23 through 3 -2-24Risk interval estimates
Comparison interval estimates
•Background rate for GBS in persons 60+ years old: 1.4 -3.7 per 100,000 PY (Sejvar , J., et al. Neuroepidemiol , 2011)
Atrial Fibrillation Following RSV Vaccination
--No Statistical Signal --
24
25Analysis Parameters Signal Information and Informative Counts Nominal AnalysisSequential
Test
Outcome
EventRisk
Interval
DaysComp
Interval
Days Vaccine TypeSignaled
in a
Prior
Run1New
Sequential
Analysis
Signal2Events
in Risk
IntervalEvents in
Comp
IntervalAdjusted
Expected
Events in
Risk
IntervalAdjusted
Rate
Ratio395%
Confidence
Interval42-sided
P value1-sided P
value5
AFIB 1-21 43-63 GSK w simul No No 81 75 76.7 1.06 0.75-1.49 0.756 0.411
GSK wo simul No No 198 232 222.3 0.89 0.72-1.10 0.282 0.871
Pfizer w simul No No 7 15 11.9 0.59 0.21-1.51 0.282 0.910
Pfizer wo simul No No 23 33 38.7 0.59 0.32-1.09 0.093 0.968
22-42 GSK w simul No No 91 109 106.9 0.85 0.63-1.14 0.282 0.876
GSK wo simul No No 211 222 209.8 1.01 0.83-1.22 0.954 0.497
Pfizer w simul No No 8 17 13.8 0.58 0.22-1.41 0.238 0.924
Pfizer wo simul No No 25 30 29.9 0.84 0.48-1.46 0.530 0.781
1-42 43-84 GSK w simul No No 190 155 166.7 1.14 0.90-1.44 0.267 0.146
GSK wo simul No No 422 441 444.2 0.95 0.82-1.10 0.496 0.764
Pfizer w simul No No 23 27 23.3 0.99 0.54-1.79 0.966 0.577
Pfizer wo simul No No 52 60 71.3 0.73 0.48-1.10 0.137 0.945VSD RCA of RSV Vaccine in Older Adults
Surveillance Initiated on 01AUG2023, Analyses Based on Data Through 25MAY2024 (run #6, week #1846)
Concurrent Comparator Sequential Analysis Signal Assessment for Atrial Fibrillation
---Age:60+ yrs---
•VSD initiated surveillance in older adults in January 2024
•385,729 doses of RSV vaccines have been administered to older adults
(88% GSK)
•Statistical signal for ITP in persons 60+ years old who received the GSK
RSV vaccine without simultaneous vaccination
-Most were not incident ITP cases with onset after RSV vaccination
-Plan for more detailed chart review of ITP in the fall
•No GBS signal, but few observed cases
•No other statistical signals observed to date
•Surveillance will continue for all outcomes, including GBS, through May
2025VSD RSV RCA Summary and Next Steps
26
RCA Study Team and Acknowledgments
27•Marshfield Clinic Research Institute: Ed Belongia, Hannah Berger, Kayla
Hanson, Burney Kieke, Dave McClure, Erica Scotty, Maria Sundaram, Jim
Donahue
•Centers for Disease Control and Prevention: Eric Weintraub, Tat’Yana
Kenigsberg, Amelia Jazwa , Tanya Myers, Mike McNeil, and Lily Wang
•Thank you to our colleagues at the other VSD sites: Kaiser Permanente
(Northern CA, Southern CA, Washington, Northwest, Colorado, Mid -
Atlantic), Denver Health, HealthPartners, and Harvard Pilgrim
•Many thanks to our colleagues at Kaiser Permanente Northern California
for their assistance with this project and their innovative analytic methods:
Bruce Fireman, Joan Bartlett, Kristin Goddard, Ned Lewis, and Nicky Klein
28The findings and conclusions in this presentation are those
of the authors and do not necessarily represent the official
position of the Centers for Disease Control and Prevention