Nils Bundgaard (1), Alexandra Kropotova (2), Daniel Kaschek (1)
(1) IntiQuan AG, Basel, Switzerland, (2) SAb Biotherapeutics Inc, Sioux Falls, United states of America
Objectives:
Influenza, a significant global health concern, can causes severe illness, hospitalizations, and death predominantly in high-risk (HR) groups, including immunocompromised and immunosenescent patients. SAB Biotherapeutics has developed SAB-176, a fully human polyclonal Immunoglobulin G antibody produced in transchromosomic bovines (TcB), which has shown high-titer neutralizing activity against multiple strains of Influenza A and B, offering potential for treating HR patients. Data from a phase 2a study and published data were combined to develop a model that simulates time to symptom alleviation based on the viral load dynamics in both healthy and HR patients infected with Influenza A or Influenza B. The model was used to project doses of a Phase 2b study.
Methods:
This analysis was conducted in three steps. First a mechanistic viral load model was created to reflect influenza infection dynamics in healthy patients, using public literature data [1,2]. The model was extended with data from the placebo group of the Phase 2a study SAB-176-201. Subsequently, treated subjects were included in the analysis and the efficacy of SAB-176 was estimated. Second, a Joint Time-to-Event Model was developed to link viral load dynamics and symptom alleviation time based on public data [2,3]. The hazard function used a log-logistic distribution, and the viral load was linked to the survival function. Differences between healthy and HR patients were assumed to be due to different immune responses.
Lastly, the final model was used to simulate viral load and symptom alleviation time in HR patients with Influenza infection. Simulations were conducted for different doses of SAB-176, treatment times post infection, and infection with IAV or IBV variants.
Results:
Publicly available literature on influenza infections showed an exponential growth phase of viral load in the first two days of infection and symptoms onset, followed by a linear clearance phase. In the SAB-176-201 study, the viral kinetics of the placebo group exhibited more sustained, prolonged viral kinetics, with a reduced maximum viral load. Symptoms’ severity was reduced and the time to symptom development was delayed in the treatment group, indicating that the SAB-176 antibody reduced and delayed influenza-related symptoms.
A mechanistic viral load model was fitted to literature data and SAB-176-201 data. A difference in immune response was able to explain differences between the viral load dynamics between the studies with otherwise healthy vs. HR patients, while other parameters such as the production rate or death rate of infected cells were fixed to values based on literature. The neutralization efficacy of the SAB-176 treatment was estimated, but only the lower bound of the treatment effect could be identified.
The joint model developed in this analysis established a link between viral load and time to alleviation of symptoms. The viral load dynamics predicted for HR patients showed higher viral peak load and a more sustained viral load profile, consistent with literature on HR patients.
The joint model was used to simulate the effect of different dosing scenarios of SAB-176 treatment in HR patients. A favorable dosing scenario (top projected dose, 48 hours after symptom onset, A/California/2009) reduced the time at which only 25% of HR patients still had symptoms from approximately 200 hours in the untreated group, to 140 hours in the actively treated HR patients. Using a 2x lower dose level, the threshold of 25% was reached 12 hours later than for the projected top dose.
Conclusions:
A mechanistic viral load model, based on public literature and data from a phase 2a study, was developed, revealing that SAB-176 significantly reduced viral load. Variability in viral load profiles between studies was adequately described by differences in the immune response. A joint Time-to-event model was developed, linking viral load and symptom alleviation time. High-risk patient data exhibited higher viral loads, prolonged viral shedding, and longer symptom alleviation times. The model accurately predicted these outcomes. The TTE model simulated the effect of different SAB-176 dosing scenarios in high-risk patients. A favorable dosing scenario was simulated to reduce symptom duration from approximately 200 hours to around 140 hours, an effect size similar or better to the one observed with approved anti-influenza treatments [2].
References:
[1] Smith AM, Perelson AS. Influenza A virus infection kinetics: quantitative data and models. Wiley Interdiscip Rev Syst Biol Med. 2011 Jul-Aug;3(4):429-45. doi: 10.1002/wsbm.129. Epub 2010 Dec 31. PMID: 21197654; PMCID: PMC3256983.
[2] Hayden FG, Sugaya N, Hirotsu N, Lee N, de Jong MD, Hurt AC, Ishida T, Sekino H, Yamada K, Portsmouth S, Kawaguchi K, Shishido T, Arai M, Tsuchiya K, Uehara T, Watanabe A; Baloxavir Marboxil Investigators Group. Baloxavir Marboxil for Uncomplicated Influenza in Adults and Adolescents. N Engl J Med. 2018 Sep 6;379(10):913-923. doi: 10.1056/NEJMoa1716197. PMID: 30184455.
[3] Retout S, De Buck S, Jolivet S, Duval V, Cosson V. A Pharmacokinetics-Time to Alleviation of Symptoms Model to Support Extrapolation of Baloxavir Marboxil Clinical Efficacy in Different Ethnic Groups with Influenza A or B. Clin Pharmacol Ther. 2022 Aug;112(2):372-381. doi: 10.1002/cpt.2648. Epub 2022 Jun 10. PMID: 35585696.
Reference: PAGE 32 (2024) Abstr 10933 [www.page-meeting.org/?abstract=10933]
Poster: Drug/Disease Modelling - Infection