Grant Number:	5R01CA269398-04 Interpret this number
Primary Investigator:	Li, Yi
Organization:	University Of Michigan At Ann Arbor
Project Title:	Causal Machine Learning in Cancer Survival By Integrating Multiple High-Dimensional Observational Studies
Fiscal Year:	2025

Despite overall improvements in cancer research, observational studies often face limitations in representativeness of broader patient populations and comparability across clinical settings. For example, treatment practices may vary across cohorts, and differences in patient characteristics can complicate efforts to draw generalizable conclusions about treatment outcomes. This proposal is motivated by the Boston Lung Cancer Survival Cohort (BLCSC), one of the largest lung cancer cohorts globally, consisting of cases registered since 1992 at the Dana-Farber Cancer Institute (DFCI) and the Massachusetts General Hospital (MGH), with subsequent expansion to the MD Anderson Cancer Center (MDACC) and Mayo Clinic. We also have access to the International Lung Cancer Consortium (ILCCO), an international cohort established in 2004 with a data structure like BLCSC. These rich databases provide unique opportunities for advancing modern statistical methods in cancer survival analysis. They pose methodological challenges such as unbalanced patient covariates across sites, heterogeneous data structures, and variability in treatment assignment. Leveraging BLCSC and ILCCO, we aim to develop integrative causal machine learning methods for analyzing multiple high-dimensional observational studies, with broad applications such as robust treatment comparisons and survival inference. Such methods will enable us to generate findings that are generalizable and applicable to wide-ranging patient populations.





None