A pediatric vaccine to protect infants against RSV-linked LTRD is urgently needed to cover the large burden of disease in infants. Vector-based RSV vaccines, such as the ChAd155-RSV candidate vaccine based on a chimpanzee-derived Ad-vector, elicit immune profiles that have not been associated with ERD, and may be a suitable solution for pediatric use. We have evaluated the immunogenicity, safety (in terms of ERD) and vaccine efficacy of the Chad155 RSV pediatric vaccine in a calf model, and studied different settings and regimens to mimic, as closely as possible, the human infant vaccination scenarios. We demonstrate that ChAd155-RSV was immunogenic in naïve young calves, inducing low to modest RSV nAb responses following the first dose which were further increased following the second dose. The two-dose vaccination regimen protected the calves from bRSV-induced clinical disease and lung pathology, and significantly reduced viral loads in the nasopharynx and lung after a DOI of either 4 or 16 weeks. The protection conferred by a single vaccination resembled the protection following the two-dose regimen when applying the short DOI, and was near-complete against general illness, fever and CLA after the long DOI. However, in the latter case, the effects of the single dose on respiratory rates and lung histopathology were less profound than with two doses and a long DOI. Finally, protection after two vaccinations and a short DOI was similar in calves with versus without pre-existing maternal RSV Abs.
Multiple studies demonstrated that replication-incompetent ChAd vectors against several viruses, including RSV and SARS-CoV2, are capable of inducing humoral and cell-mediated responses in humans with acceptable safety profiles16,19,24,25,42. Activation of both arms of adaptive immunity is considered critical for protection against severe RSV disease, especially in infants and older adults11,43−46. Previously, ChAd155-RSV was demonstrated to induce RSV nAbs and T-cell responses in seropositive adult humans22. Here we show that two doses of ChAd155-RSV induced robust nAb responses that reached the estimated protective titer in human infants47 (i.e., 256) and just exceeded the threshold of 6 log2 associated with a reduced risk of hospitalization in infants48. The immune responses observed here in calves, which were sustained during the 4-week study period, almost completely protected the animals from bRSV clinical disease and lung pathology, as did the two-dose regimen after the 16-week period. The latter is remarkable, considering that at challenge these nAb responses had decreased to a level (geometric mean of 101) well below the above mentioned protection threshold47. In animals receiving a single vaccine dose, the general illness, fever and lung consolidation readouts, and the virus concentrations in the lung and nasopharynx, were all comparable after long- and short-DOI. However, among these single-dose groups, respiratory rates and lung histopathological scores with a long DOI (Study 2), while still being lower than the controls, were less reduced than with the short DOI (Study 1). In the absence of a two-dose long-DOI group in the same study, it remains inconclusive whether this slightly compromised protection was due to the more severe challenge condition, or to a lower efficacy of this single-dose regime. Nevertheless, the overall results suggest that a single-dose regime also merits consideration in clinical development (NCT03636906).
Though not measured here, several lines of evidence suggest the presence of vaccine-induced immunity beyond the detected nAb responses. Indeed, in line with the intended pediatric use of ChAd155-RSV, calves received their first injection at a young age (3 to 12 months old). At this age, the bovine immune system is considered not fully mature, resulting in weak nAb responses due to the induction of low-frequency and short-living plasma-cell responses49–51. In human infants, differentiation of B cells into long-lived plasma cells after antigenic exposure is known to be reduced due to lack of B-cell survival signals in the bone marrow52–54. While a similar effect is expected to have occurred in the vaccinated calves, they nevertheless exhibited the typical kinetics of a B-cell response to vaccination, in which nAb titers elicited by the first dose were boosted by the second vaccination, and further boosted by the challenge. This suggests that the vaccination had successfully induced plasma-cell and memory B-cell responses targeting the live human challenge strain (RSV A Long). Moreover, the fact that all calves in the single-dose groups were strongly protected with reduced viral titers, even though their nAb titers were low, indicated that other arms of immunity, such as T cells, may also have contributed to this level of protection. Indeed, the M2-1 and N proteins are included in the vaccine construct to elicit T-cell immunity, and preliminary analysis in a different calf study showed that ChAd155-RSV induced both CD4+ and CD8+ T-cell responses to RSV proteins following a single injection (unpublished data). In addition, T-cell responses were detected in the vaccinated adult humans22.
A large proportion of the vaccine target population of young infants will have maternal-derived RSV Abs due to natural infection of the mother, and possibly increased levels due to future maternal RSV vaccination4,55. Moreover, many infants may receive one of the forthcoming prophylactic anti-RSV mAbs56,57. Importantly, although our results suggest that the presence of maternal Ab may have had some negative impact on the levels of the nAb response, as observed after the second dose, this did not appear to have affected the protection against the bRSV challenge conferred to these animals. The collective data thus suggest that the presence of pre-existing RSV Abs might have at most minimal effects on the efficacy of ChAd155-RSV in human infants. Nevertheless, future research could help determining which RSV F epitopes are the main targets of the humoral immune response, and whether the observed negative effect of pre-existing Ab has any impact on vaccine immunogenicity in humans.
Considering the (~3.2 times) higher dose of challenge virus in Study 2 versus Study 1, the clinical data post-challenge indicated a clear dose-response effect in the onset and severity of the symptoms. The higher CLA scores of controls in Study 2 may also be related to the more severe challenge condition. However, this difference between studies may have been exacerbated by the fact that the mean CLA score at necropsy was calculated on data from all animals, including those that were terminated 5 or 6 days earlier, when they most likely have been at their disease peak. Indeed, the overall higher CLA scores in the pre-terminated versus surviving animals is consistent with data showing a trend of an association between disease severity scores and CLA scores30. Similarly, histopathology scores were also higher in Study 2, though for these scores no difference was found between pre-terminated and recovered animals. This may be because the recovery from such microscopic symptoms takes generally longer. Nonetheless, the two-dose (short-DOI) regimens in groups without pre-existing responses displayed similar efficacies across the two studies. This indicated that the efficacy data from Study 2 can be used to support the overall conclusion on ChAd155-RSV vaccine.
Beyond its utility in RSV vaccine immunogenicity/efficacy studies, the calf model is also suitable for evaluations of vaccine safety, including the risk of ERD, a major focus in pediatric RSV vaccine development. In calves, this phenomenon has been characterized by advanced and exacerbated clinical symptoms upon bRSV challenge relative to control bRSV-challenged animals and by Th2-skewed cytokine profiles, IgE Ab production, pulmonary eosinophilia, and a high ratio of non-functional Abs to nAbs30,37,38,58−60. The current research work was not designed as a safety study to extensively investigate all ERD signs or characteristics, thus did not include a positive control vaccine (FI-RSV) inducing ERD. Nonetheless, we observed that all ChAd155-RSV-vaccinated, bRSV-challenged young calves were either free of clinical/pathological symptoms, or exhibited symptoms that were much less severe than in the PBS-injected and bRSV-challenged control groups, while in case of an ERD response the symptoms in vaccinated animals would have been more severe than in the controls. In conclusion, our observations suggest that the risk that ChAd155-RSV will induce ERD in human infants is extremely low, supporting the further clinical development of this vaccine.