Effectiveness of Maternal Respiratory Syncytial Virus Vaccination in Conferring Infant Immunity: Review and Future Perspectives

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This narrative review addresses a timely topic: RSVpreF vaccination and passive immunization strategies for infant protection. The manuscript provides a broad overview of RSV epidemiology, pathophysiology, maternal immunization, monoclonal antibodies, real-world effectiveness, safety, policy considerations, and health economics.

The topic is appropriate for Vaccines.

However, several issues limit the robustness of the review in its current form:

The manuscript would benefit from a brief description of the literature identification process. Even for a narrative review, specifying databases searched and general inclusion criteria would strengthen methodological transparency.

Several sections summarize available data without critically appraising study design limitations, heterogeneity across populations, or potential biases. A stronger analytical perspective would enhance the scientific contribution of the review.

The manuscript suggests superiority of vaccination at 28-32 or 32-36 weeks versus 24-28 weeks. However, the interpretation of gestational timing subgroups may overstate the strength of evidence. It would be important to clarify whether interaction tests were prespecified and statistically significant, and to emphasize the exploratory nature of subgroup analyses.

The manuscript mentions immortal time bias in the context of hypertensive disorders. The discussion of immortal time bias in relation to hypertensive disorders requires clarification. It is unclear whether this bias pertains to post hoc observational analyses rather than the randomized MATISSE trial itself.

The paragraph discussing the French comparative study raises concerns that remain largely generic. First, the statement regarding baseline imbalances does not acknowledge that the published analysis explicitly implemented a two-step confounding control strategy combining exact matching (on discharge date, sex, gestational age, and region) and inverse probability of treatment weighting including an extensive set of maternal, infant, socioeconomic, and healthcare access variables. Post-weighting diagnostics demonstrated excellent covariate balance, with standardized mean differences close to zero across measured characteristics. Simply invoking “residual confounding” without engaging with the balance diagnostics or the robustness checks performed does not constitute a meaningful methodological critique. Second, the suggestion of “confounding by indication” appears overstated in the French 2024 context. Both maternal RSVpreF vaccination and nirsevimab were recommended universally during the campaign, rather than targeted to distinct clinical risk groups. Third, the concern regarding time-varying effects related to antibody kinetics overlooks the fact that the study explicitly evaluated time since cohort entry, reported hazard ratios across follow-up intervals. Finally, the concluding statement that comparative estimates “should be interpreted as associations rather than definitive causal effects” is broadly applicable to any observational study.

The real-world effectiveness section would benefit from a comparative synthesis of study designs and potential sources of bias across countries, rather than a descriptive listing of results.

Given increasing global use of long-acting monoclonal antibodies, the discussion of potential selection pressure and resistance at antigenic site could be expanded and supported by emerging surveillance data.

The economic section compiles country-specific threshold values but lacks a synthesis identifying the principal drivers of cost-effectiveness variability across settings.

The contrasting recommendations regarding revaccination in subsequent pregnancies present an important policy divergence that could be explored more analytically.

Author Response

 

Response to Reviewer 1 Comments
1. Summary
We sincerely appreciate your insightful and constructive comments. Your feedback has substantially strengthened the clarity, analytical rigor, and scientific contribution of our work. In response, we have undertaken a comprehensive revision of the manuscript, addressing each comment in detail and refining the structure, interpretation, and contextualization of the evidence. Please find the detailed responses below and the corresponding revisions in the re-submitted files.
2. Point-by-point response to Comments and Suggestions for Authors
Comments 1: The manuscript would benefit from a brief description of the literature identification process. Even for a narrative review, specifying databases searched and general inclusion criteria would strengthen methodological transparency.
Response 1: Thank you for this important suggestion. We agree that even in a narrative review, describing the literature identification process enhances methodological transparency. In response, we added a concise description of the databases searched, key terms used, and general inclusion criteria. This addition clarifies how the evidence base was assembled and improves reproducibility. Changes in the Manuscript: Page 2, Paragraph 56–63: “To enhance methodological transparency, we briefly describe our literature identification approach. We searched PubMed, Embase, and major public health agency repositories using terms such as ‘RSV,’ ‘maternal immunization,’ ‘RSVpreF,’ ‘nirsevimab,’ and ‘monoclonal antibodies.’”
Comments 2: Several sections summarize available data without critically appraising study design limitations, heterogeneity across populations, or potential biases. A stronger analytical perspective would enhance the scientific contribution of the review.
Response 2: We appreciate this comment and agree that a stronger analytical perspective improves the scientific value of the review. We revised multiple sections to explicitly discuss study design limitations, heterogeneity, and potential biases, including confounding, selection bias, and issues relevant to observational effectiveness studies. Changes in the Manuscript: Page 2, Paragraph 64–70: “This review incorporates a critical appraisal of study design limitations, heterogeneity across populations, and potential sources of bias…” Page 14, Paragraph 640–642: “Variability in VE estimates likely reflects contextual factors rather than true biological differences…”
Comments 3: The manuscript suggests superiority of vaccination at 28-32 or 32-36 weeks versus 24-28 weeks. However, the interpretation of gestational timing subgroups may overstate the strength of evidence. It would be important to clarify whether interaction tests were prespecified and statistically significant, and to emphasize the exploratory nature of subgroup analyses.
Response 3: We agree and have clarified that gestational timing findings are derived from exploratory subgroup analyses that were not powered for formal interaction testing. We now explicitly state that these results should be interpreted cautiously. Changes in the Manuscript: Page 10, Paragraph 450–454: “However, these gestational timing subgroup analyses were exploratory and not prespecified in the MATISSE trial. Interaction tests were not statistically significant, and the study was not powered to detect differences between gestational windows. Therefore, the apparent superiority of vaccination at 28–32 or 32–36 weeks should be interpreted with caution.”
Comments 4: The manuscript mentions immortal time bias in the context of hypertensive disorders. The discussion of immortal time bias in relation to hypertensive disorders requires clarification. It is unclear whether this bias pertains to post hoc observational analyses rather than the randomized MATISSE trial itself.
Response 4: Thank you for pointing this out. We revised the text to clarify that immortal time bias pertains to observational analyses and does not apply to the randomized MATISSE trial. The revised text now distinguishes between RCT methodology and post hoc observational analyses. Changes in the Manuscript: Page 9, Paragraph 423– Page 10, Paragraph 433: “Importantly, the MATISSE trial was a randomized controlled trial, and therefore concerns such as immortal time bias do not apply to its primary analyses…”
Comments 5: The paragraph discussing the French comparative study raises concerns that remain largely generic. First, the statement regarding baseline imbalances does not acknowledge that the published analysis explicitly implemented a two-step confounding control strategy combining exact matching (on discharge date, sex, gestational age, and region) and inverse probability of treatment weighting including an extensive set of maternal, infant, socioeconomic, and healthcare access variables. Post-weighting diagnostics demonstrated excellent covariate balance, with standardized mean differences close to zero across measured characteristics. Simply invoking “residual confounding” without engaging with the balance diagnostics or the robustness checks performed does not constitute a meaningful methodological critique. Second, the suggestion of “confounding by indication” appears overstated in the French 2024 context. Both maternal RSVpreF vaccination and nirsevimab were recommended universally during the campaign, rather than targeted to distinct clinical risk groups. Third, the concern regarding time-varying effects related to antibody kinetics overlooks the fact that the study explicitly evaluated time since cohort entry, reported hazard ratios across follow-up intervals. Finally, the concluding statement that comparative estimates “should be interpreted as associations rather than definitive causal effects” is broadly applicable to any observational study.
Response 5: We appreciate this detailed critique. We revised the section to acknowledge the two-step confounding control strategy (exact matching + IPTW), the excellent covariate balance achieved, and the robustness checks performed. We also clarified that confounding by indication was limited due to universal recommendations in France during the study period. Additionally, we incorporated the study’s explicit evaluation of time-varying effects. Changes in the Manuscript: Page 14, Paragraph 669–Page 15, Paragraph 689: “The French study implemented exact matching and inverse probability weighting… Post-weighting diagnostics demonstrated excellent covariate balance… Accordingly, policy decisions should consider the differing mechanisms, implementation pathways, and population‑level goals of each intervention rather than relying on observational comparative estimates to infer superiority.”
Comments 6: The real-world effectiveness section would benefit from a comparative synthesis of study designs and potential sources of bias across countries, rather than a descriptive listing of results.
Response 6: We agree and revised the section to synthesize methodological differences across countries, including case definitions, healthcare-seeking behavior, and surveillance intensity. Changes in the Manuscript: Page 14, Paragraph 643–647: “Taken together, real‑world effectiveness studies across Argentina, the United Kingdom, and Scotland demonstrate consistently high protection against RSV‑associated hospitalization, despite differences in study design, population characteristics, epidemic timing, and healthcare utilization. Variability in VE estimates likely reflect contextual factors rather than true differences in biological effectiveness”
Comments 7: Given increasing global use of long-acting monoclonal antibodies, the discussion of potential selection pressure and resistance at antigenic site could be expanded and supported by emerging surveillance data.
Response 7: We expanded the section to incorporate emerging genomic surveillance data and to discuss the potential for selective pressure from widespread monoclonal antibody use. Changes in the Manuscript: Page 16 (Section 6.1), Paragraph 745–754: “Although the F protein has remained antigenically stable… widespread use of monoclonal antibodies may introduce selective pressures…”
Comments 8: The economic section compiles country-specific threshold values but lacks a synthesis identifying the principal drivers of cost-effectiveness variability across settings.
Response 8: We added a summary identifying key drivers such as hospitalization costs, vaccine pricing, RSV seasonality, and coverage rates. Changes in the Manuscript: Page 18 (Section 6.5), Paragraph 864–871: “Because these analyses rely on heterogeneous assumptions—including RSV incidence, hospitalization costs, vaccine uptake, and willingness‑to‑pay thresholds—results are not directly comparable across countries. Local economic evaluations remain essential for policy decisions. Across settings, key drivers of cost‑effectiveness include RSV seasonality, hospitalization costs, vaccine or monoclonal antibody pricing, coverage rates, and health‑system capacity. These contextual factors largely explain the variability in cost‑effectiveness thresholds across countries, rather than differences in biological efficacy.”
Comments 9: The contrasting recommendations regarding revaccination in subsequent pregnancies present an important policy divergence that could be explored more analytically.
Response 9: We expanded the section to discuss durability of antibody responses, potential immune interference, and the limited evidence base informing repeat dosing. Page 17 (Section 6.3), Paragraph 798–800, 807-814: “However, these early‑phase data were not designed to evaluate pregnancy‑specific safety outcomes, and evidence on repeated dosing during pregnancy remains limited.” “This approach reflects a precautionary stance in the context of limited evidence rather than a biological rationale for or against repeated vaccination. Given the divergent recommendations across countries, additional research is needed to clarify the durability of maternal immunity, the optimal interval between doses, and the safety and effectiveness of repeated vaccination during pregnancy. Longitudinal studies evaluating antibody kinetics across multiple pregnancies, as well as postmarketing safety surveillance, will be essential to inform evidence‑based revaccination policies”

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript provides a comprehensive and timely review of maternal RSV vaccination, covering virology, immunology, clinical trial data, and post-marketing surveillance. The topic is of high public health importance given the recent licensure of maternal RSV vaccines and ongoing policy decisions globally. However, several areas require revision to improve clarity and critical appraisal of the evidence.

 

Major Comments

 

1. The presentation in Table 4 and accompanying text lists studies with disparate conclusions—some showing no association, others suggesting a possible elevated risk—without providing a balanced synthesis or explanation for the discrepancies. The authors should offer a critical appraisal of methodological differences between studies that may explain the heterogeneity, and provide a clearer overall interpretation of the current evidence base for clinicians and policymakers.

 

2. The manuscript appropriately notes that maternal vaccination and nirsevimab are complementary strategies but does not delve into the practical trade-offs that health systems face when choosing between them. A more detailed discussion is warranted, addressing factors such as: timing of protection (birth vs. post-administration), durability of protection etc.

 

3. While the description of antigenic sites Ø, I-V on the F protein is scientifically accurate, the clinical implications for vaccine effectiveness and durability are not fully developed. The authors note that waning F-specific neutralizing antibody titers contribute to reinfection susceptibility, but do not discuss how this informs booster vaccination strategies or the potential need for updated vaccines if antigenic drift occurs. A more integrated discussion linking structural virology to long-term vaccine policy would strengthen the manuscript.

 

4. Lines 562-575 describe differing international recommendations on revaccination but provide limited immunological rationale or data to inform this debate. The authors should expand this section to discuss what is known about the durability of vaccine-induced antibody responses following maternal immunization, the potential for immune interference in subsequent pregnancies, and the evidence (or lack thereof) supporting repeat dosing. This is a key policy question that deserves more thorough treatment.

 

Minor Comments

 

5. "waning F-specific neutralizing antibody titers contribute to susceptibility to reinfection" – Consider specifying that this waning occurs over months to years, and clarify that reinfection typically results in milder illness in older children and adults.
6. The discussion of age structure shifts should note whether these changes reflect true epidemiological shifts or simply increased testing and awareness in older age groups.
7. Table 3: This table is very useful for international comparison. Consider adding a column indicating whether countries recommend vaccination during every pregnancy or only the first.
8. The discussion of confounding in comparative effectiveness studies is well-reasoned. Consider adding a brief mention of target trial emulation as a methodological approach to address such biases in future research.
9. The mention of "financing mechanisms" in the conclusions is vague. Consider specifying examples such as vaccine purchase funds, Gavi support for LMICs, or reimbursement models.

Author Response

1. Summary
We sincerely appreciate your insightful and constructive comments. Your feedback has substantially strengthened the clarity, analytical rigor, and scientific contribution of our work. In response, we have undertaken a comprehensive revision of the manuscript, addressing each comment in detail and refining the structure, interpretation, and contextualization of the evidence.

2. Point-by-point response to Comments and Suggestions for Authors

Comments 1: The presentation in Table 4 and accompanying text lists studies with disparate conclusions—some showing no association, others suggesting a possible elevated risk—without providing a balanced synthesis or explanation for the discrepancies. The authors should offer a critical appraisal of methodological differences between studies that may explain the heterogeneity, and provide a clearer overall interpretation of the current evidence base for clinicians and policymakers.

Response 1: Thank you for this important observation. We agree that presenting heterogeneous findings without synthesizing methodological differences may limit interpretability. We revised the section to explicitly compare study designs, including differences in case definitions, surveillance intensity, healthcare-seeking behavior, and analytic approaches. We also added a concluding synthesis to clarify how these methodological factors explain much of the observed heterogeneity.

Changes in the Manuscript:

• Page 12, Paragraph 559–564, 570-571, Page 12, Paragraph 573-Page 13, Paragraph 597:
“However, the potential inclusion of early-onset cases (20–33 weeks’ gestation) raises concerns about immortal time bias. Immortal time bias occurs when outcomes that arise before an individual becomes eligible for vaccination are incorrectly attributed to the vaccinated group.…”

Comments 2: The manuscript appropriately notes that maternal vaccination and nirsevimab are complementary strategies but does not delve into the practical trade-offs that health systems face when choosing between them. A more detailed discussion is warranted, addressing factors such as: timing of protection (birth vs. post-administration), durability of protection etc.

Response 2: We appreciate this suggestion. We expanded the discussion to compare practical considerations relevant to policy decisions, including timing of protection (at birth vs. post-administration), predictability of antibody levels, durability of protection, and programmatic feasibility. This provides a clearer framework for policymakers evaluating the two strategies.

Changes in the Manuscript:
• Page 14, (section 5.7.) Paragraph 654–657, Page 14, Paragraph 669-Page 15, Paragraph 689

• We added section 5.8.

Page 19, Paragraph 898-899, 904-906, 915-919

Comments 3: While the description of antigenic sites Ø, I-V on the F protein is scientifically accurate, the clinical implications for vaccine effectiveness and durability are not fully developed. The authors note that waning F-specific neutralizing antibody titers contribute to reinfection susceptibility, but do not discuss how this informs booster vaccination strategies or the potential need for updated vaccines if antigenic drift occurs. A more integrated discussion linking structural virology to long-term vaccine policy would strengthen the manuscript.

Response 3: Thank you for this insightful comment. We expanded the section to link structural virology with long-term vaccine policy, including the implications of waning neutralizing antibodies, potential need for booster doses, and the possibility of updated vaccine formulations should antigenic drift occur.
Changes in the Manuscript:

• Page 16 (Section 6.1), Paragraph745–754:
“Although current………. Should meaningful antigenic drift occur, updated vaccine formulations or periodic booster doses may be required to maintain protection…”

Comments 4: Lines 562-575 describe differing international recommendations on revaccination but provide limited immunological rationale or data to inform this debate. The authors should expand this section to discuss what is known about the durability of vaccine-induced antibody responses following maternal immunization, the potential for immune interference in subsequent pregnancies, and the evidence (or lack thereof) supporting repeat dosing. This is a key policy question that deserves more thorough treatment.

Response 4: We agree and expanded this section to discuss durability of vaccine-induced antibody responses, the limited evidence on immune interference in subsequent pregnancies, and the absence of robust data on repeated dosing. This provides a more comprehensive scientific basis for understanding divergent national recommendations.

Changes in the Manuscript:
• Page 11, Paragraph 500–506:
“International differences in recommendations reflect uncertainties in durability, immune interference, and the limited evidence base for repeated maternal immunization…”

Comments 5: "waning F-specific neutralizing antibody titers contribute to susceptibility to reinfection" – Consider specifying that this waning occurs over months to years, and clarify that reinfection typically results in milder illness in older children and adults.

Response 5: We added the requested clarification and noted that reinfections are typically milder in older children and adults.

Changes in the Manuscript:
• Page 6, Paragraph 262–268:
“Waning of F-specific neutralizing antibodies occurs over months to years, contributing to susceptibility to reinfection; however, reinfections in older children and adults are generally milder…”

Comments 6: The discussion of age structure shifts should note whether these changes reflect true epidemiological shifts or simply increased testing and awareness in older age groups.

Response 6: We revised the section to explicitly acknowledge this possibility.

Changes in the Manuscript:
• Page 4, Paragraph 150–154:
“These shifts may reflect altered population immunity following the COVID‑19 pandemic, but the relative contributions of viral evolution, behavioral changes, and surveillance artifacts remain uncertain. In addition, increased RSV testing and heightened clinical awareness in older children and adults may partially contribute to the apparent rise in reported cases.”

Comments 7: Table 3: This table is very useful for international comparison. Consider adding a column indicating whether countries recommend vaccination during every pregnancy or only the first.

Response 7: We added the requested column to Table 3.

Changes in the Manuscript:
• Table 3: Added column: “Revaccination in Each Pregnancy.”

Comments 8: The discussion of confounding in comparative effectiveness studies is well-reasoned. Consider adding a brief mention of target trial emulation as a methodological approach to address such biases in future research.

Response 8: We agree and added a concise explanation of target trial emulation as a methodological framework for future observational studies.

Changes in the Manuscript:
• Page 2, Paragraph 69–70:
“Target trial emulation provides a structured framework to reduce bias in observational comparative effectiveness studies…”

Comments 9: The mention of "financing mechanisms" in the conclusions is vague. Consider specifying examples such as vaccine purchase funds, Gavi support for LMICs, or reimbursement models.

Response 9: We revised the conclusion to specify concrete examples, including national vaccine purchase programs, reimbursement models, and Gavi support for LMICs.

Changes in the Manuscript:
• Page 20, Paragraph 963–968:
“Sustainable implementation will require robust financing mechanisms, including national vaccine purchase programs, reimbursement frameworks, and—where applicable—support from international partners such as Gavi…”

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revision process has clearly strengthened the manuscript, and the authors have addressed the comments in a rigorous manner.

However there is two major points still to be addressed :

1. The manuscript published in JAMA on RSVpreF vs nirsevimab effectiveness comparison appropriately addresses potential time-dependent biases. In particular, the study design aligns eligibility assessment, exposure assignment, and follow-up initiation at maternity discharge, which ensures comparability between groups from time zero. In addition, the authors conducted lagged sensitivity analyses to account for incubation and symptom onset, further strengthening the temporal validity of the comparison. Taken together, these design features substantially minimize the risk of immortal time bias.

I encourage the authors to emphasize these methodological strengths more clearly page 19 line 896-900 and page 15 line 676-682, as they directly address concerns regarding time-related bias instead of generically stating that in observational studies there is time bias.

2. Taken together, the two studies published in JAMA and JAMA Pediatrics evaluating the effectiveness of RSVpreF and nirsevimab using different study designs consistently suggest greater protection associated with nirsevimab compared with RSVpreF. It would have been valuable to contextualize the findings within the broader body of robust real-world evidence providing precise effect estimates, and to discuss whether monoclonal antibody administration to infants may confer greater protection than maternal vaccination. Available data also suggest that protection with nirsevimab may be more sustained over time. Simply listing the results and potential biases of individual studies is insufficient. At some point, it would have been interesting to synthesize and interpret the accumulated evidence to help inform clinical decision-making.

Author Response

Reviewer 1 – Comment 1

The manuscript published in JAMA on RSVpreF vs nirsevimab effectiveness comparison appropriately addresses potential time-dependent biases. In particular, the study design aligns eligibility assessment, exposure assignment, and follow-up initiation at maternity discharge, which ensures comparability between groups from time zero. In addition, the authors conducted lagged sensitivity analyses to account for incubation and symptom onset, further strengthening the temporal validity of the comparison. Taken together, these design features substantially minimize the risk of immortal time bias. I encourage the authors to emphasize these methodological strengths more clearly (page 19 line 896–900 and page 15 line 676–682), as they directly address concerns regarding time-related bias instead of generically stating that in observational studies there is time bias.

Response

We thank the reviewer for this important comment. We agree that the methodological strengths of the JAMA comparative effectiveness study warrant clearer emphasis, as they directly address concerns regarding time‑dependent bias. We have revised the manuscript to explicitly highlight the study’s rigorous alignment of eligibility assessment, exposure assignment, and follow‑up initiation at maternity discharge, which ensures comparability between groups from a clearly defined time zero. We also added a description of the lagged sensitivity analyses conducted to account for incubation periods and symptom‑onset windows, which further minimize the risk of immortal time bias and strengthen the temporal validity of the comparison.

Reviewer 1 – Comment 2

Taken together, the two studies published in JAMA and JAMA Pediatrics evaluating the effectiveness of RSVpreF and nirsevimab using different study designs consistently suggest greater protection associated with nirsevimab compared with RSVpreF. It would have been valuable to contextualize the findings within the broader body of robust real-world evidence providing precise effect estimates, and to discuss whether monoclonal antibody administration to infants may confer greater protection than maternal vaccination. Available data also suggest that protection with nirsevimab may be more sustained over time. Simply listing the results and potential biases of individual studies is insufficient. At some point, it would have been interesting to synthesize and interpret the accumulated evidence to help inform clinical decision-making.

Response

We appreciate this insightful suggestion. We agree that synthesizing the accumulated real‑world evidence—rather than listing individual study results—is essential for informing clinical and policy decision‑making. We have therefore added a new integrative paragraph summarizing the consistent findings from JAMA and JAMA Pediatrics, highlighting the possibility of greater and more sustained protection with nirsevimab while acknowledging the limitations of observational comparisons. This addition provides a clearer interpretation of the comparative evidence and its implications for clinical decision‑making.

Author Response File: Author Response.pdf