Interpretable Machine Learning for Molecular Discovery in Lung Cancer
Abstract
Scientific Objective and Rationale: Non-small-cell lung cancer (NSCLC) is the leading cause of cancer-related death in the United States and worldwide. Each year, more people die of lung cancer than of colon, breast, and prostate cancers combined, and more than 130,000 people will die from the disease in the United States during 2022. NSCLC is the most common type of lung cancer, accounting for more than 80% of the cases. Currently, stratification for treatment of NSCLC patients depends on a set of predictive molecular markers that are associated with better outcome. For example, patients with EGFR mutations are candidates for targeted therapy, while patients with high PD-L1 expression or high tumor mutation burden (TMB) are candidates for immunotherapy in the case of lacking known markers of targeted therapy. Despite the advances in identifying actionable alterations, these predictive markers are not perfect, and most NSCLC patients do not respond to treatment and have their tumors progress to the aggressive stage. We hypothesize that no single molecular feature is enough to predict response in NSCLC patients; rather, a coordinated and interacting set of features that may have better predictive utility is needed. In this research, we will develop a novel machine learning model that is guided by known cancer biology to identify a set of features, genes, and known and novel pathways that explain the different responses and manifestations of NSCLC disease. This will help understand mechanisms of resistance in unselected populations and identify novel therapeutic targets opening the door for developing new treatments. Our proposed research directly addresses the FY22 Lung Cancer Research Program Areas of Emphasis of (1) understand the molecular mechanisms of initiation and progression to lung cancer and (2) understand mechanisms of resistance to treatment. Principal Career Goals in Lung Cancer Research: My goal is to become a leader in the field of intelligent cancer informatics with focus on lung cancer. My research program combines data-driven approaches with intelligent computational modeling to understand mechanism of resistance and progression. This award will give me the resources and protected time to gain skills and training needed to achieve my goals. I have identified a set of training opportunities including formal courses, professional workshops, and international and national meetings that will help me enhance my scientific profile and connect with the professional community. My mentor, Dr. Eliezer Van Allen, is a pioneer in the field of clinical computational oncology and has a unique record of using genomic data to understand treatment resistance in multiple cancer types. In addition, Dr. Kenneth Kehl is an expert in lung cancer with significant contribution to the field of clinical natural language processing. My mentorship team provide valuable training to achieve my career goals. Ultimate Applicability of the Research: Our goal is to understand the mechanisms of resistance to immunotherapy, targeted therapy, and aggressive disease. Understanding mechanisms of resistance and aggressiveness will help identify novel therapeutic targets opening the door for developing new therapeutic intervention. It will also help in identifying patients with a high risk of recurrence or progression and understanding their unique molecular composition. This can lead to better risk stratification of early-stage NSCLC patients who may have high-risk molecular features. The discovered mechanisms of resistance and aggressiveness will guide therapeutic development, patients’ stratification, and clinical decision-making in NSCLC patients. This applies to affected duty Service Members, Veterans, military beneficiaries, and the American public, leading to better clinical outcomes especially in resistant and aggressive NSCLC patients.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 04, 2024
- Source ID
- HT94252310023
Entities
People
- Haitham Elmarakeby
Organizations
- Dana–Farber Cancer Institute
- United States Army