Models for Precision Medicine on Vital Capacity Using Longitudinal Information After Start of Pirfenidone Treatment for Patients with Idiopathic Pulmonary Fibrosis

• Activity Number: 490 - Topics in Personalized/Precision Medicine - I
• Type: Contributed
• Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Hiroki Sakaguchi* and Keiko Kawaguchi and Yukio Tada and Hideaki Watanabe and Takahiro Hasegawa
• Companies: Shionogi & Co., Ltd. and Shionogi & Co., Ltd. and Shionogi & Co., Ltd. and Shionogi & Co., Ltd. and Shionogi & Co., Ltd.
• Keywords: Precision medicine; Functional data regression; Longitudinal data

Abstract

Models for precision medicine are designed to assist with healthcare professionals and patients’ decision-making in starting, continuing, stopping or changing treatment by predicting personalized future clinical outcomes. Functional data regression model (FDRM), which is one of the methods for functional responses and/or covariates, can be applied for precision medicine. The FDRM can deal with longitudinal data to regression model by applying the functional data estimated from discrete longitudinal data as an explanatory variable. In our presentation, we will focus on the FDRM to predict the change from baseline in vital capacity at 24 weeks after the start of treatment of pirfenidone, an approved medicine for idiopathic pulmonary fibrosis (IPF), using therapeutic process data until 16 weeks. It can be useful for patients with the poor prognostic disease and healthcare professionals to make a decision on courses of treatment based on the prediction information on a future clinical outcome. Furthermore, we will evaluate the performance of the constructed FDRM for precision medicine via bootstrap method and compare it with other models for longitudinal data.