Risk of preterm birth, small for gestational age at birth, and stillbirth after covid-19 vaccination during pregnancy: population based retrospective cohort study

Dear Editor

Covid-19 vaccine safety in pregnancy was one of the many unknowns at the time of conditional marketing authorization in late 2020 because the phase III pivotal trials of mRNA vaccines did not include pregnant women.[1]

Fell et al. report a post-authorisation observational study using the provincial birth registry of Ontario (BORN) to estimate the risk of mRNA covid-19 vaccination on adverse pregnancy outcomes.[2] Of note, only liveborn or stillborn infants 20+ weeks of gestation or with birthweight at least 500 grams are systematically recorded in BORN such that adverse events after vaccination occurring before gestational week 20 are missed in this study.

Each pregnancy contributed gestational time in days between May 1, 2021 and December 31, 2021. Risk estimates of exposure are based on a total of 43,099 women who received at least 1 dose during pregnancy.

Exposure-time to full mRNA vaccination (two doses) during pregnancy was limited to only 29,650 women with median gestational age of 29.1 weeks at dose 1 with only 301 women (1%) receiving dose 1 during their first trimester. The remaining 13,416 women were not fully exposed during pregnancy but received their first dose at a median gestational age of already 34 weeks, and further doses after delivery. In other words, the contribution of each vaccinated woman to the exposure time with two vaccine doses is not identical and generally of short duration, in particular for the outcome very preterm birth (<32 weeks), as well as preterm birth (<37 weeks), while 42,063 women in the unvaccinated comparison group are continuously contributing pregnancy days at risk of adverse pregnancy outcomes due to important other factors such as maternal age, comorbidity, smoking, substance abuse, low socio-economic status, and gestational age.

Marked imbalances between the vaccinated and unvaccinated study groups with respect to characteristics not only linked to the outcome but also to the probability of being vaccinated introduce the potential for strong bias (i.e. healthy user bias). When looking at the subgroup of 422 women who received only 1 dose in total, a statistically significant 9% increase in risk of adverse pregnancy outcomes (95% CI 1.01 to 1.16) is observed. With observational data that raises more questions than provides answers, it is worth reflecting on what may be gained from randomized trial data. While pregnant women were excluded from the pre-authorization trials, a recent independent re-analysis of the phase III Pfizer and Moderna trials found serious adverse events of special interest elevated at 1 event per 800 vaccinated adults.[3] This raises the obvious question of whether such harms are limited to non-pregnant adults, or would also apply to pregnant women. The risk management plan of the Pfizer/BioNTech mRNA vaccine comprises one randomized study (NCT04754594 or C4591015) on safety and immunogenicity in pregnant women.[4][5]

Starting February 2021, the trial (NCT04754594 or C4591015) originally planned to enroll 4000 women between the 24th and 34th week of their pregnancy. However on November 16, 2021, the manufacturer stopped enrollment at only 348 women because pregnant women in the study countries U.S., Spain, Brazil, and South Africa had allegedly already been vaccinated and were no longer eligible for enrollment.[5] Fell et al. provide evidence that this can hardly be the case [2][6]. In our eyes, with a study size of only 10% of the originally planned participants, the aims of study NCT04754594 or C4591015 as specified in the risk management plan cannot be achieved. Regulators are responsible for oversight and must inform the public about what went wrong in study NCT04754594 or C4591015. Without access to individual participant data for independent scrutiny of randomized controlled trials we do not have reliable evidence to recommend Covid-19 vaccines for pregnant women.

Angela Spelsberg MD, SM
Comprehensive Cancer Center Aachen, Germany

Ulrich Keil MD, PhD , FRCP
Prof. emeritus, Institute of Epidemiology and Social Medicine, University of Muenster

References:
[1] Prugger C, Spelsberg A. et al. Evaluating covid-19 vaccine efficacy and safety in the post-authorisation phase. BMJ 2021; 375 doi: https://doi.org/10.1136/bmj-2021
[2] Fell DB, Dimanling-Cruz S, Regan AK et al. Risk of preterm birth, small for gestational age at birth, and stillbirth after covid-19 vaccination during pregnancy: population-based retrospective cohort study. BMJ 2022;378:e071416 http://dx.doi.org/10.1136/bmj-2022-071416
[3] Fraiman J, Erviti J, et al. Serious adverse events of special interest following mRNA vaccination in randomized trials. https://ssrn.com/abstract=4125239
[4] https://clinicaltrials.gov/ct2/show/NCT04754594
[5] COMIRNATY (COVID-19 mRNA VACCINE) RISK MANAGEMENT PLAN
RMP Version number: 5 https://www.ema.europa.eu/en/documents/rmp-summary/comirnaty-epar-risk-m...
[6] Razzaghi H, Meghani M et al. Covid-19 vaccination coverage among pregnant women during pregnancy. – Eight integrated health care organizations, United States, December 14, 2020 - May 8,2021. MMWR Morb Mortal Wkly Rep 2021;70:895-9. Doi:10.15585/mmwr.mm7024e2