Kapil Gadkar, Siddharth Sukumaran, Tracy Staton, Manoj Rodrigo, Cynthia Stokes, Heleen Scheerens, Saroja Ramanujan
Genentech, USA
Background: Asthma is a chronic inflammatory disease of the airways involving numerous underlying immunological and stromal pathways. Various treatments in development target activities or proteins in these pathways, and show differential impact on clinical outcomes and pathway biomarkers. Although specific molecular pathways are being characterized more thoroughly, the understanding of the linkage between the underlying mechanisms with the functional clinical outcomes is still very limited. Given this limited mechanistic understanding and to expedite clinical research, a human experimental model of rhinovirus (RV) infection challenge is being developed to support a proof-of-activity study to evaluate drug candidates as a means to prioritize and expedite clinical research.
Methods: We have developed a mechanism-based systems model representing different cellular and soluble contributors to asthma, including (1) innate immune, adaptive immune, and airway resident cells (2) soluble proteins such as IL5, IL13, IL4, and IgE and (3) clinical markers and endpoints such as FeNO and FEV1. The model has been developed utilizing preclinical in vitro and in vivo data, and clinical data for multiple drugs including anti-IgE, anti-IL5, anti-IL13, and anti-IL-4Rα. The model was then utilized for making predictions for novel mono and combination therapies to support clinical decision making. To further expedite the clinical development of drug candidates the model includes the RV infection challenge representation and predictions were made for responses of mild asthmatics on the RV challenge to treatment.
Results: The model was calibrated and verified to successfully describe the clinical measurements for different patient severities and for a range of interventions. The model is used for making predictions for existing mono therapies (including anti-IL13, anti-IL-4Rα) and novel combination therapies and identifying patient sub-populations that respond favorably to the different interventions. The model is also utilized to design the proof-of-activity RV clinical study.
Conclusions: The model is useful to elucidate biological pathways underlying observed effects of the different interventions and allows us to explore and predict the impact of additional interventional strategies for which little to no clinical data is available.
Reference: PAGE 25 (2016) Abstr 5796 [www.page-meeting.org/?abstract=5796]
Poster: Drug/Disease modeling - Other topics