Modelling humoral immune response after a single RSV vaccination in RSV-seropositive adults aged 60 and older
Piotr Juszczak (1), Anna Dari (2), Arangassery Rosemary Bastian (3), Daniel Stieh (3), Juan José Pérez Ruixo (2)
(1) Janssen Research and Development, Janssen, Basel, Switzerland, (2) Janssen Research and Development, Janssen, Beerse, Belgium, (3) Janssen Vaccines and Prevention, Janssen, Leiden, The Netherlands
Introduction: Respiratory Syncytial Virus (RSV) is a cause of a considerable number of lower and upper respiratory tract infections, morbidity, and mortality in older adults [1]. This population have been pre-exposed to RSV in their lifetime as evidenced by RSV specific antibodies secreted by long-lived plasma cells (LLPC) and RSV specific memory B-cells. However, the adaptive immune memory, acquired through previous RSV infection(s), i.e., affinity maturation of antibodies, LLPC and memory B- and T-cells, is not sufficient to prevent new RSV infections with severe symptoms in this population [1].
Objective: We aim to characterize kinetics of humoral immune responses in individuals previously naturally exposed to RSV that received a single injection of an RSV candidate vaccine or placebo. The RSV vaccine candidate is a combination of an adenovirus 26 vector (Ad26.RSV.preF) with expressing a membrane bound RSV F protein stabilized in the prefusion conformation (preF) [2] and a soluble recombinant RSV preF protein administered as a single IM injection.
Methods: A 9-compartments mechanistic model has been developed that describe the dynamics of memory B-cells, LLPC and IgG antibodies secreted by LLPC for subjects receiving placebo, and Ad26.RSV.preF, preF protein, memory B-cells, LLPC, short-lived plasma cells (SLPC) and secreted IgG antibodies by LLPC and SLPC for vaccinated subjects.
Parameters of the model were estimated by fitting data from four phase 1-2b clinical trials: NCT03982199, NCT03502707, NCT04354480, NCT04453202. Measurements of binding antibodies (ELISA), neutralizing antibodies (wtVNA) to the RSV-A preF protein in the prefusion confirmation and memory B-cells (ELISpot), up to 3 years post an injection from 521 placebo and 1514 vaccinated subjects were used. A total of 13 combinations of doses of both vaccine components were included.
As binding and neutralizing antibody titers were highly correlated, the mechanistic model was used to describe the time course of binding antibodies. Neutralizing antibodies were modelled as a function of the binding antibodies. Given the data available and the complexity of the model, several parameters were fixed to published values, e.g., elimination of SLPC and elimination of antibodies [4].
Model-based simulations were used to understand the role of the vaccine components on the time course of RSV A specific humoral biomarkers (memory B-cells, SLPC, LLPC and neutralizing and binding antibodies) and to understand the duration of protection in the context of a hypothetical protective threshold in a correlate of protection (CoP) analysis.
Results: The proposed model is an extension of published models [3,4]. We extended these models by accounting for a patient population that is pre-exposed to RSV, different antigen formulation (preF protein and Ad26.RSV.preF) and dosing (through KPD approach [5]) and memory B-cell data. The mechanistic model adequately described the time course of observed binding and neutralizing antibody serum titers, and memory B-cells counts, as well as their associated variabilities up to 3 years following the injection of the vaccine or a placebo. LLPC and memory B-cells are in almost steady-state pre-vaccination and return to a higher steady-state level around day 183 post-vaccination. No association between age, gender, ethnicity, weight, and study with model parameters has been identified.
Assuming a hypothetical protective threshold of neutralizing antibody titers above 900 IC50 titers (CoP) and a percentage of subjects above the threshold at baseline of 17%, model-based simulations estimated that a percentage of subjects with neutralizing antibody titers above the CoP would be: 71, 62, 59, 55 and 53% after 1 to 5 years following a single dose of the combination 1×1011 viral particles of Ad26.RSV.preF and 150μg of RSV preF protein, and 12, 10, 9, 8, 7% in placebo recipients, assuming no RSV infections during the 5 years of follow up occurred.
Conclusions: The extended mechanistic model adequately describes the observed time course of memory B-cells, neutralizing and binding antibody titers in RSV-seropositive participants up to 3 years following a single vaccination with different doses of two antigens. Model based simulations may aid to understand the duration of the protection in the context of the CoP and the optimal time for boosting.
References:
[1] Shi T., Denouel A., Tietjen A., et al. Global disease burden estimates of respiratory syncytial virus-associated acute respiratory Infection in older adults in 2015: A systematic review and meta-analysis. The Journal of Infectious Diseases 2020; 222(7):577-583.
[2] McLellan J., Chen M., Leung S., et al. Structure of RSV fusion glycoprotein trimer bound to a prefusion-specific neutralizing antibody. Science 2013; 340: 1113–1117.
[3] Balelli I., Pasin C., Prague M., et al. A model for establishment, maintenance and reactivation of the immune response after vaccination against Ebola virus. Journal of Theoretical Biology 2020; 495: 110254.
[4] Dari A., Boulton M., Neyens M., et al. Quantifying antibody persistence after a single dose of COVID-19 vaccine Ad26.COV2.S in humans using a mechanistic modeling and simulation approach. Clinical Pharmacology & Therapeutics 2023; 113(2): 380-38
[5] Jacqmin, P., Snoeck, E., Schaick, E. et al Modelling Response Time Profiles in the Absence of Drug Concentrations: Definition and Performance Evaluation of the K-PD Model. Journal of Pharmacokinetics and Pharmacodynamics, 2007, 34(1).