Comparative effectiveness of mRNA-1273 and BNT162b2 COVID-19 vaccines in immunocompromised individuals: a systematic review and meta-analysis using the GRADE framework

Introduction Despite representing only 3% of the US population, immunocompromised (IC) individuals account for nearly half of the COVID-19 breakthrough hospitalizations. IC individuals generate a lower immune response after vaccination in general, and the US CDC recommended a third dose of either mRNA-1273 or BNT162b2 COVID-19 vaccines as part of their primary series. Influenza vaccine trials have shown that increasing dosage could improve effectiveness in IC populations. The objective of this systematic literature review and pairwise meta-analysis was to evaluate the clinical effectiveness of mRNA-1273 (50 or 100 mcg/dose) vs BNT162b2 (30 mcg/dose) in IC populations using the GRADE framework. Methods The systematic literature search was conducted in the World Health Organization COVID-19 Research Database. Studies were included in the pairwise meta-analysis if they reported comparisons of mRNA-1273 and BNT162b2 in IC individuals ≥18 years of age; outcomes of interest were symptomatic, laboratory-confirmed SARS-CoV-2 infection, SARS-CoV-2 infection, severe SARS-CoV-2 infection, hospitalization due to COVID-19, and mortality due to COVID-19. Risk ratios (RR) were pooled across studies using random-effects meta-analysis models. Outcomes were also analyzed in subgroups of patients with cancer, autoimmune disease, and solid organ transplant. Risk of bias was assessed using the Newcastle-Ottawa Scale for observational studies. Evidence was evaluated using the GRADE framework. Results Overall, 17 studies were included in the pairwise meta-analysis. Compared with BNT162b2, mRNA-1273 was associated with significantly reduced risk of SARS-CoV-2 infection (RR, 0.85 [95% CI, 0.75–0.97]; P=0.0151; I2 = 67.7%), severe SARS-CoV-2 infection (RR, 0.85 [95% CI, 0.77–0.93]; P=0.0009; I2 = 0%), COVID-19–associated hospitalization (RR, 0.88 [95% CI, 0.79–0.97]; P<0.0001; I2 = 0%), and COVID-19–associated mortality (RR, 0.63 [95% CI, 0.44–0.90]; P=0.0119; I2 = 0%) in IC populations. Results were consistent across subgroups. Because of sample size limitations, relative effectiveness of COVID-19 mRNA vaccines in IC populations cannot be studied in randomized trials. Based on nonrandomized studies, evidence certainty among comparisons was type 3 (low) and 4 (very low), reflecting potential biases in observational studies. Conclusion This GRADE meta-analysis based on a large number of consistent observational studies showed that the mRNA-1273 COVID-19 vaccine is associated with improved clinical effectiveness in IC populations compared with BNT162b2.

Although immunocompromised (IC) individuals comprise only approximately 3% of the US population (7), they account for nearly half of the breakthrough COVID-19 hospitalizations (8).While there is a range of severity across conditions at the population level, adults considered immunodeficient had 2.68-fold greater adjusted odds of being hospitalized with COVID-19 compared with immunocompetent individuals due both to the underlying IC condition and therapies used for treatment (9,10).In 1 study, use of immunosuppression in patients with autoimmune disease resulted in 1. 35-fold (95% CI, 1.29-1.40)greater odds of developing life-threatening COVID-19 (11).
Influenza vaccine trials demonstrated that high-dose vaccines led to improved immune responses in IC individuals compared with standard dose vaccines and suggested that a high-dose vaccine offers greater effectiveness for IC populations (36)(37)(38)(39)(40)(41).Although both mRNA-1273 and BNT162b2 employ the mRNA mode of action, the composition of each vaccine is different.For instance, the mRNA dosage and type of lipid nanoparticles used in the delivery system differs between vaccines.The mRNA-1273 primary series contains 100 mcg of mRNA and 50 mcg for the booster (2,42), whereas BNT162b2 contains 30 mcg of mRNA for each primary and booster dose (3,43).Observational studies have consistently shown differences between the 2 mRNA COVID-19 vaccines, both in terms of immune response (15) and clinical effectiveness (44)(45)(46) in IC populations.
As SARS-CoV-2 transitions from a pandemic to an endemic state, countries are transferring vaccination programs from central government purchasing to their respective national healthcare systems, which is triggering in-depth health technology assessments to recommend the best use of available vaccines in specific populations.Several national immunization technical advisory groups, including the Advisory Committee on Immunization Practices in the United States, use the GRADE (Grading of Recommendations, Assessment, Development and Evaluations) framework for identifying questions relevant to healthcare, selecting outcomes of interest and assessing their importance, evaluating the available evidence, and synthesizing evidence to develop recommendations consistent with considerations of values and preferences of patients and the society in which they live (47,48).
This present analysis follows the GRADE framework to address the following healthcare question: Is the mRNA-1273 COVID-19 vaccine (50 or 100 mcg/dose) more clinically effective in IC populations compared with the BNT162b2 COVID-19 vaccine (30 mcg/dose)?Accordingly, we performed a systematic literature review and pairwise meta-analysis to compare COVID-19 vaccine effectiveness outcomes among IC individuals given either mRNA-1273 or BNT162b2.

Search strategy and study selection
We performed a systematic literature review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 framework (49).S1.
Clinical trials, observational studies, or any real-world evidence published as manuscripts, letters, commentaries, abstracts, or posters were included if they reported efficacy or clinical effectiveness outcomes in IC individuals ≥18 years of age vaccinated with mRNA-1273 or BNT162b2 within the same study.IC individuals were defined as people with immunocompromising conditions considered to be in the clinically extremely vulnerable (CEV) groups 1 or 2, which includes solid organ transplant, solid and hematologic cancers, hemodialysis, poorly controlled HIV infection, and autoimmune conditions requiring immunosuppressive therapy (50).Outcomes of interest were vaccine efficacy or effectiveness against symptomatic, laboratory-confirmed SARS-CoV-2 infection, SARS-CoV-2 infection, severe SARS-CoV-2 infection (defined as patients with infection and hospitalization or death, or as reported by each study), COVID-19associated hospitalization, and COVID-19-associated death.Recently published systematic literature reviews on the same topic were crosschecked to ensure relevant articles were included.Studies reporting outcomes in pregnant women, current or former smokers, physically inactive people, or those with a heterologous vaccination schedule (ie, mix of mRNA-1273 and BNT162b2), only safety data, or study protocols or economic models were excluded.The population, exposure, comparison, and outcomes used in the systematic literature review are summarized in Table S2.Two independent reviewers selected studies following a 2-level approach; a third reviewer arbitrated conflicts.Titles and abstracts were screened against inclusion criteria in level 1, followed by full-text appraisal of articles not excluded at level 1 against selection criteria in level 2.

Data extraction and quality assessment
Publication details, study and participant characteristics, vaccine type and vaccination status, at-risk condition, and clinical outcomes were extracted.Risk of bias (RoB) was assessed in accordance with Cochrane review guidelines (51) using the Newcastle-Ottawa Scale (52) for observational studies.Evidence was evaluated based on GRADE criteria (47,48).

Statistical analysis
Random-effects meta-analysis models were used to pool risk ratios (RR) and calculate absolute effects as risk difference (RD) per 100,000 individuals across studies.Inverse variance weights were calculated for individual studies with the DerSimonian-Laird method (53).Chisquare testing to evaluate heterogeneity across studies was performed (54).The I 2 statistic was estimated (0%-100%, 0% meaning no evidence of heterogeneity).Subgroup analysis was performed for patients with cancer, autoimmune disease, and solid organ transplant to account for differences between the underlying health conditions in the overall IC population.
Risk-of-bias assessment found no serious RoB for 13 studies and serious RoB in 4 nonrandomized studies, primarily because of the lack of description of comparability between cohorts or adjustment for confounding factors (Table S3).

Symptomatic, laboratory-confirmed SARS-CoV-2 infection
Of the 4 studies reporting symptomatic, laboratory-confirmed SARS-CoV-2 infection, mRNA COVID-19 vaccine type was not associated with a statistically significant risk of infection (RR, 0.88 [95% CI, 0.65-1.19];Table 2).The RD (95% CI) of mRNA-1273 compared with BNT162b2 was estimated to be 633 fewer infections (792 fewer to 473 fewer).Heterogeneity between studies analyzed for this outcome may be considered substantial (I 2 = 60.3%).Evidence certainty was rated as type 3 (low) due to the inclusion of nonrandomized studies and because of imprecision (Table 2).
Subgroup analysis was performed in 2 studies of solid organ transplant recipients (55,67) and in 1 study of patients with autoimmune disease (66).No association between mRNA COVID-19 vaccine type and risk of symptomatic, laboratoryconfirmed SARS-CoV-2 infection was observed in either the solid organ transplant recipient subgroup (RR, 1.  3).

Hospitalization due to COVID-19
mRNA-1273 was associated with a significantly lower risk of COVID-19-associated hospitalization versus BNT162b2 in the 9 studies included in the overall analysis (RR, 0.88 [95% CI, 0.79-0.97];P<0.0001).The RD (95% CI) of mRNA-1273 compared with BNT162b2 was estimated to be 14 fewer hospitalizations due to COVID-19 (from 78 fewer to 51 more).No evidence of heterogeneity was observed between studies (I 2 = 0%).The certainty of evidence for this outcome was type 3 (low) because of inclusion of nonrandomized studies and imprecision (Table 2).
x In Aslam 2021, 1 event occurred in the mRNA-1273 and BNT162b2 arms.The small number of events resulted in a wide 95% CI. y Lower grading due to imprecision and limited evidence.z I 2 =0%, X 2 =0.02, p(Q)=0.89,no issues of heterogeneity and inconsistency.aa Lower grading due to limited evidence, higher grading due to no issues with imprecision and strong association.bb I 2 =0%, X 2 =0.06, p(Q)=1, no issues of heterogeneity and inconsistency.cc In Holroyd 2022, a continuity correction of 0.5 was applied because there were 0 events in the mRNA-1273 and BNT162b2 arms.This resulted in a wide 95% CI. dd Higher grading due to no issues with indirectness, lower grading due to no strong association.ee I 2 =0%, X 2 =0.67, p(Q)=0.88,no issues of heterogeneity and inconsistency.ff In Aslam 2021, Malinis 2021, and Yetmar 2022, a continuity correction of 0.5 was applied because there were 0 events in the mRNA-1273 and BNT162b2 arms.This resulted in a wide 95% CI. gg Lower grading due to imprecision, higher grading due to no issues with indirectness and strong association.hh I 2 =0%, X 2 =0.05, p(Q)=0.82,no issues of heterogeneity and inconsistency.
ii In Holroyd 2022, a continuity correction of 0.5 was applied because there were 0 events in the mRNA-1273 and BNT162b2 arms.This resulted in a wide 95% CI. jj Lower grading due to limited evidence, imprecision, and no strong association.significantly reduced risk of death compared with BNT162b2 (RR, 0.63 [95% CI, 0.44-0.90];P=0.0119).Vaccination with mRNA-1273 was estimated to result in 15 fewer deaths associated with COVID-19 (from 307 fewer to 227 more) compared with BNT162b2.No evidence of heterogeneity was observed between the studies (I 2 = 0%).The certainty of evidence was rated as type 3 (low) due to inclusion of nonrandomized studies (Table 2).Grading was reduced for imprecision.

Discussion
In this systematic review and pairwise meta-analysis of 17 studies, we found that mRNA-1273 was associated with a significantly lower risk of SARS-CoV-2 infection, severe SARS-CoV-2 infection, hospitalization due to COVID-19, and COVID-19-associated mortality compared with BNT162b2 in adults with a broad spectrum of severe IC conditions.A summary of the RR of each outcome for mRNA-1273 versus BNT162b2 is presented in Figure 2. The certainty of this evidence was type 4 (very low) for the SARS-CoV-2 infection and severe SARS-CoV-2 infection outcomes and type 3 (low) for the COVID-19-associated hospitalization and death outcomes (Table 4).Because all included studies were pairwise comparisons between mRNA-1273 and BNT162b2, the research question was not biased by differences in period assessed, population, or viral variants within each study.When outcomes were assessed in subgroups, mRNA-1273 was associated with significantly lower risk of SARS-CoV-2 infection, severe SARS-CoV-2 infection, and COVID-19-associated hospitalization versus BNT162b2 in patients with cancer.Compared with BNT162b2, mRNA-1273 was also associated with a significantly reduced risk of SARS-CoV-2 infection in patients with autoimmune diseases and COVID-19-associated death in solid organ transplant recipients.Immunocompromised individuals have a high burden of COVID-19 due to characteristics of their underlying disease or immunosuppressive treatments that affect their ability to mount productive immune responses and increased susceptibility to severe COVID-19 (30).Physicians may seek to optimize COVID-19 vaccine type, timing, and number of doses to improve outcomes in IC patients (32).RCTs are ranked highly in the hierarchy of evidence; however, studying comparative efficacy with adequate power would require enrolling a prohibitive number of IC patients.Given the rapid spread of SARS-CoV-2 globally, there was also insufficient time following the development of mRNA COVID-19 vaccines to conduct a larger number of RCTs in IC populations.Therefore, the research question can only be assessed using large real-world databases where individual medical and pharmacy information is available and real-world studies evaluating vaccine effectiveness in IC patients.Given the risk of severe COVID-19 in IC patients, our approach may be justified given the absence of RCT data in this vulnerable population.
Limitations of this systematic literature review were that non-English studies were excluded, and publication bias was not assessed in the meta-analysis.Inherent to the GRADE framework, evidence certainty is initially set to either high if the included studies are randomized studies or low if they are observational studies.Because all 17 studies included in the pairwise meta-analysis were nonrandomized, the maximum certainty of evidence achievable in this meta-analysis was low despite the high number of observational studies and consistency of results.The pairwise meta-analysis was also limited by inconsistent outcome definitions across studies and differences in covariates between studies.For example, the vaccination scheme (2 vs 3 doses; booster) differed between studies, with a mix of primary series (100 mcg vs 30 mcg) and booster (50 mcg vs 30 mcg) pairwise comparisons included in the meta-analysis.Variants of concern changed over time, with risks of hospitalization and death (68) and vaccine effectiveness differing by variant (69).Vaccine effectiveness of 2-dose regimens could only be shown for the ancestral strain through the delta variants, whereas the omicron variant required a 3-dose schedule.Other sources of bias, such as prescribing differences by risk of severe COVID-19, ability of patients to choose the mRNA vaccine type, and differing risks of infection or COVID-19-associated morbidity and mortality among patients with different IC conditions, could not be accounted for in this meta-analysis.In addition to differences in mRNA dosage between mRNA-1273 and BNT162b2, other differences such as the lipid nanoparticle delivery system and mRNA translation efficiency may also have impacted clinical effectiveness between vaccines.
Our meta-analysis of observational studies showed that mRNA-1273 (50 or 100 mcg/dose) was associated with a significantly reduced risk of SARS-CoV-2 infection, severe SARS-CoV-2 infection, COVID-19-associated hospitalization, and death due to COVID-19 when compared with BNT162b2 (30 mcg/dose) in IC populations.Based on these findings, vaccinating IC individuals in the United States with mRNA-1273 instead of BNT162b2 would prevent an additional 14 and 15 hospitalizations and deaths per 100,000 individuals, respectively.Although the numerical differences in estimated prevented COVID-19-associated hospitalizations and deaths are small, the implications for care are considerable in IC patients, who are at increased risk of severe COVID-19.Considering the limited availability of data from RCTs and to provide needed clinical decision-making guidance, our results showed that mRNA-1273 offers better clinical outcomes compared with BNT162b2 in vulnerable IC populations.A video summary linked to this article can be found at Video S1.
This study was funded by Moderna, Inc. Authors employed by Moderna, Inc. were involved in the study design, analysis and interpretation of data, the writing of the manuscript, and the decision to submit the manuscript for publication.

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers.Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

TABLE 1
Characteristics of studies included in the meta-analysis.
AQP4-NMOSD, aquaporin-4-antibody neuromyelitis optica spectrum disorder; CLIMB, Comprehensive Longitudinal Investigation of Multiple Sclerosis at Brigham and Women's Hospital; GRUCINI, Infectious Complications Subcommittee of the Spanish Hematopoietic Stem Cell Transplantation and Cell Therapy Group; IC, immunocompromised; MGB, Mass General Brigham; MOGAD, myelin oligodendrocyte glycoprotein-antibody−associated disease; MS, multiple sclerosis; NR, not reported; SEHH, Spanish Society of Hematology and Hemotherapy; VHA, Veterans Health Administration.Vaccine dosing was abbreviated as MM or MMM for 2 or 3 doses of mRNA-1273, respectively, and as PP or PPP for 2 or 3 doses of BNT162b2, respectively.

TABLE 2
GRADE summary of findings overall.
bIn Aslam 2021, only 2 events each occurred in the mRNA-1273 and the BNT162b2 arms, therefore wide 95% CI. c Lower grading due to imprecision.Type 3 due to inclusion of nonrandomized studies.d I 2 =67.7%,X 2 =43.40, p(Q)<0.0001,substantial heterogeneity.e Outcome definitions rather heterogeneous (test-positive cases and symptomatic cases).f k Lower grading because of imprecision and indirectness due to varying outcome definitions (severe SARS-CoV-2 infection and hospitalization or death following SARS-CoV-2 infection).l I 2 =0%, X 2 =5.05, p(Q)=0.75,no issues of heterogeneity and inconsistency.m No indirect comparisons, outcome definitions in line.n In Holroyd 2022 and Yeo 2022, only 0 and 1 event in the mRNA-1273 arm, respectively; this resulted in a wider 95% CI. °I2 =0%, X 2 =2.12, p(Q)=0.91,no issues of heterogeneity and inconsistency.p In Yetmar 2022, the sample size was rather low and 0 events occurred in both arms; therefore, a continuity correction of 0.5 was necessary.A continuity correction was also necessary in Aslam 2021 and Holroyd 2022 because 0 events occurred in both arms.This resulted in a wide 95% CI.

TABLE 3
GRADE summary of findings by population subgroup.

TABLE 4
Summary of the Clinical Effectiveness Meta-Analysis.A graphical summary of the relative risks of clinical effectiveness outcomes analyzed for mRNA-1273 vs BNT162b2 COVID-19 vaccines in IC populations is shown.COVID-19, coronavirus disease 2019; IC, immunocompromised; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2. Summary of evidence for outcomes of interest.
COVID-19, coronavirus disease 2019; GRADE, Grading of Recommendations, Assessment, Development, and Evaluations; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2. a Relative importance of outcomes assessed and ranked per the GRADE framework.