Predicting RSV prophylaxis in the infant target population – assessing monoclonal antibody versus maternal vaccination efficacy using MBMA
Nele Plock (1), Xiaowei Zang (2), Jos Lommerse (1), Kalpit A. Vora (2), Andrew Lee (2), S.Y. Amy Cheung (1), Brian M. Maas (2), Jeffrey R. Sachs (2)
(1) Certara, Princeton, NJ, USA, (2) Merck & Co., Inc., Rahway, NJ, USA
Introduction: Respiratory syncytial virus (RSV) is a leading cause of respiratory tract infections in infants [1]. It can cause severe illness, including bronchiolitis and pneumonia, and can lead to hospitalization, particularly in preterm infants and those with underlying medical conditions. Passive immunization with monoclonal antibodies (mAb) is a promising approach for the prevention of RSV lower respiratory tract infection (LRTI) in infants. Active maternal vaccination, which provides immunity to the newborn against RSV through the transfer of maternal antibodies, is another approach currently under development. A direct comparison of maternal RSV vaccination efficacy for infants against efficacy obtained via passive immunization is not available. This work qualified as an existing model-based meta-analysis (MBMA) [1] for the prediction of efficacy against RSV in infants following maternal vaccination. This model was used to estimate the fold-increase in serum neutralizing (SN) titer after maternal vaccination that would achieve protection similar to that of mAb given to infants born just before and during the RSV season over different study observation periods. Given that preterm infants only constitute approximately 4% of births in the United States, efficacies reported for the overall infant population are largely based on full-term infants. Therefore, the simulation framework was also used to assess potential levels of protection for preterm and full-term infants separately.
Methods: A published MBMA framework [1] was adapted to predict SN titer profiles in infants born to mothers who were previously vaccinated against RSV. The model combined information on fold-increase in SN titer in infants at birth relative to infants born to unvaccinated mothers (geometric mean ratio [GMR]) with a longitudinal SN titer baseline model developed for full-term and preterm infants. Published efficacy data from two Phase 3 trials [2,3] evaluating RSV maternal vaccination were used to qualify the model’s ability to predict maternal vaccination efficacy in infants. Next, a virtual clinical trial was designed wherein mothers expected to deliver just before or during the RSV season were administered either a hypothetical vaccine (GMR of 15, 30, 45, or 60) or placebo. Given its promising Phase 2a efficacy results, which aligned well with model-based predictions [4], MK-1654, a novel RSV F glycoprotein neutralizing mAb currently under development, was used as an mAb comparator. Clinical trial simulation was used to predict efficacy for the prevention of RSV-LRTI/RSV hospitalization 3 and 6 months after birth, comparing efficacy between maternal vaccination and MK-1654. Efficacy was assessed for the broad infant population, as well as separately for preterm and full-term infants.
Results: Model predictions captured the efficacy for published Phase 3 maternal vaccine trials, with low efficacy predicted for RSV F vaccine with adjuvant [2] and substantially higher efficacy for RSV preF [3]. The model also accurately predicted that efficacy would wane for longer observation periods following maternal vaccination. This model indicates that infants born a few months prior to the RSV season will be left vulnerable to RSV disease after maternal vaccination. Maternal vaccination resulting in a GMR of 30 and 60 for infant titers at birth was predicted to provide comparable efficacy to the clinical dose of MK-1654 over 3 and 6 months, respectively. In the preterm population, efficacy after maternal vaccination was predicted to be substantially lower, with simulated 6-months efficacy dropping below 50% for a GMR of 15. In all maternal vaccination scenarios of preterm neonates with 6 months of observation, there was no or only marginal overlap in efficacy confidence intervals versus passive immunization with MK-1654.
Conclusion: To provide comparable efficacy to MK-1654 over 6 months, these predictions suggest a maternal vaccine would need to increase titers by 60-fold, more than increases reported to date [2]. Simulation results also predict that passive immunization with MK-1654 is likely to provide better protection for preterm neonates and for infants born outside of the RSV season. In contrast, maternal vaccination may protect durably and well only for full-term infants born just prior to or during the RSV season.
References:
[1] Maas BM et al. EBioMedicine (2021) 73, 103651.
[2] Madhi SA et al. N Engl J Med (2020), 383, 426-439.
[3] NCT04424316. Press release 01 Nov 2022.
[4] Maas BM et al. RSV (2022) ARNI0268 [https://isirv.org/site/images/conferences/RSV/RSV2022/Oral%20Presentations%20at%20RSV2022_abstract%20reference%20order.pdf].