Pfizer COVID-19 Vaccine 78% Effective in Pregnant Women

Initial clinical trials of the emergency use authorized COVID-19 vaccines – like most clinical trials – excluded pregnant or lactating women. There are several reasons why this is standard procedure, including consideration for the safety of the mother and child and logistical challenges associated with study follow-up.

In the context of the current global pandemic, where pregnant women are at an increased risk of developing severe illness from SARS-CoV-2 infection, the lack of data surrounding the efficacy and safety of COVID-19 vaccines in this population has been problematic. Should pregnant women choose to be vaccinated with a potentially life-saving preventative with limited data, or risk infection with SARS-CoV-2?

The rollout of COVID-19 vaccines across the globe, coupled with the decision of many pregnant women to be immunized, has enabled researchers to collect and analyze real-world data.

In a new retrospective observational study, researchers from Tel Aviv and the US matched 15,060 pregnant women in Israel according to age, gestational age, residential area, population subgroup, parity, and influenza immunization status, into vaccinated/unvaccinated pairs. The vaccinated women received two doses of the Pfizer–BioNTech COVID-19 vaccine, BNT162b2. The findings – published this week in JAMA – show that vaccination with BNT162b2 in pregnant women significantly lowered the risk of SARS-CoV-2 infection, with an efficacy rate of 78%.¹  

Technology Networks spoke with Professor Gabriel Chodick, head of epidemiology and database analysis at the Maccabi Institute for Research, Tel Aviv, and co-author of the study, to learn more about the efficacy data. Chodick also discusses the challenges associated with conducting observational studies and what the data means for pregnant women that want to be immunized.

Molly Campbell (MC): Can you discuss the challenges associated with conducting vaccine research in pregnant women, during a global pandemic?

Gabriel Chodick (GC): The main challenge in any study is to ensure that the design of the study, the analysis of the data, and the interpretation of the results are done with the utmost scientific rigour to provide valid findings. This was especially important in this situation, where concerns and fears about the use of the vaccines during pregnancy were propelled by the paucity of available data on their effectiveness and safety during gestation. Given the global pandemic, an additional challenge was to execute this analysis in a timely manner to provide much needed information to help direct immunization policies and provide clinical recommendations to expectant women and their medical care team.

MC: This study adopted an observational, retrospective methodology. For our readers that may be unfamiliar, can you describe why this was the most appropriate approach for this work?

GC: By “observational” we mean that we did not conduct any medical interventions (so we did not randomly create an intervention group that received the vaccine and a control group that did not, like was done in the initial randomized clinical trials), but rather relied on “real-world” data on pregnant women who were or who were not vaccinated to make our inference. By retrospective we mean that we used data that were already collected as part of the national vaccination campaign in Israel, rather than collecting new data prospectively.

The previously conducted clinical trials for all of the four main vaccines excluded pregnant women. Thus, very little data were available from these studies concerning the use in this population. Our approach allowed us to evaluate the effectiveness and safety of the vaccines in real-world settings while providing timely clinical recommendations.

MC: Can you describe how the participants were identified and recruited for the study? How representative is the sample?

GC: Israel has a single-payer, universal healthcare system, where care is provided by four competing health funds. This study was conducted by KSM (Kahn-Sagol-Maccabi), the Research and Innovation Centre of Maccabi Healthcare Services, which is Israel’s second largest integrated health fund. The fund provides medical services to approximately 2.5 million members that constitute a representative sample (~30%) of the general Israeli population. Maccabi maintains a computerized database of anonymized electronic health records which contain extensive longitudinal clinical and demographic information. For this study, these data were used to ascertain all female members of Maccabi who were pregnant at any time from the initiation of the national vaccination campaign in Israel in late December 2020, through to the end of the study recruitment period in late February 2021.

MC: What are your key findings, and how can the results of the research be applied to the general public? 

GC: The analysis unequivocally indicated that women who received the Pfizer vaccine had a significantly lower risk of SARS-CoV-2 infection. Additionally, no elevated risks for any pregnancy complications or adverse developmental outcomes were observed in these women compared to women who were not vaccinated during pregnancy. The results suggest that vaccination is an effective strategy for minimizing the risk of SARS-CoV-2 infection among pregnant women, and that use of the vaccines is not associated with significant adverse side effects.

MC: Can you discuss the potential sources and impact of bias in this study?

GC: Perhaps the biggest threat to the validity of any observational study is a lack of exchangeability between the study groups. In other words, we must always ask ourselves whether the results we observe are indeed solely due to the difference in exposure between the groups (in this case, whether the expectant woman received the vaccine or not), or whether there are other factors that differ between the groups that drive the observed findings.

Since a random allocation of treatment, as is often done in clinical trials, helps ensure that the groups are comparable, we attempted to mimic this approach by using a matched design. This essentially means that for each woman who was vaccinated, we found another woman who was not vaccinated and who also shared similar demographic and clinical characteristics with the vaccinated woman to serve as her control. This allowed us to create two study groups that were comparable, apart from their exposure to the vaccine. Still, we normally only have data on factors we know or have the ability to measure, and thus it is always possible that there are still some meaningful differences between the groups that can bias the findings. To ensure that the two study groups were indeed exchangeable, we relied on the fact that previous data suggested that no immunity develops during approximately the 10 days immediately preceding the first dose of the vaccine. Thus, if the groups were indeed comparable, the risk of SARS-CoV-2 infection in the two groups should be similar during the first 10 days after the first dose, regardless of vaccination status. This is indeed what we observed. Of course, like with any other study, it is always possible that the results are still susceptible to some residual confounding or selection effects, and thus we always want to see that the findings are replicable in other cohorts.

MC: What are your next research steps?

GC: Similar to the case in the non-pregnant population, we need to continue follow-up on pregnant women who received the vaccine to determine how long the immunity lasts, and whether additional booster shots are needed. This also becomes important now with the emergence of new variants for which the existing vaccines are possibly less effective. In addition, many women in our study received their vaccine in early pregnancy, and their pregnancy was still ongoing by the time we ended the analysis. While we did not observe any indication that use of the vaccine was associated with any adverse effects to the pregnant woman or her foetus, extending follow-up would allow us to verify these observations using additional data. It would also be of interest to determine if, and to what extent, immunity is passed from the pregnant mother to her new-born.

Gabriel Chodick was speaking to Molly Campbell, Science Writer for Technology Networks.

Reference: Goldshtein I, Nevo D, Steinberg DM, et al. Association Between BNT162b2 Vaccination and Incidence of SARS-CoV-2 Infection in Pregnant Women. JAMA. 2021. doi: 10.1001/jama.2021.11035.