VCOM Research Day Program Book 2023

Medical Student Research Biomedical

03 A Machine Learning Model for Predicting Lung Adenocarcinoma Risk Based on Clonal Hematopoietic Mutations in Tumor Infiltrating Immune Cells

Ryan Shahidi; Ramu Anandakrishnan Corresponding author:

Edward Via College of Osteopathic Medicine – Virginia Campus

Despite considerable progress in detection and treatment, cancer remains a leading cause of death and disability worldwide. One of the largest contributors to reducing cancer mortality is early detection. Previous research has focused primarily on somatic mutations in tumor cells. Clonal hematopoiesis, a recently discovered phenomenon, results in the expansion of somatic mutations in clonally proliferating immune cells. These mutations may cause an ineffective anti-tumor immune response, thereby increasing cancer risk. We hypothesized that a machine learning model could be used to predict

increased cancer risk based on variant allele fraction in tumor infiltrating immunocyte (TII) samples. The cancer sample utilized TII from lung adenocarcinoma that was isolated and underwent whole genome sequencing to determine the variant allele fraction containing mutations that were identified by preliminary in silico analysis. A machine learning classifier will be used to estimate cancer risk based on the variant allele fraction. The goal of this work is to generate a highly predictive model that can be used to quickly classify clinical samples. With an effective model, a multigene mutation panel can be used to

screen for high-risk individuals as part of routine blood tests in an outpatient setting. Identification of patients with an ineffective anti-tumor immune response could increase early cancer detection, enabling treatment to start at an earlier and more effective stage.medical schools nationwide, promoting a future generation of knowledgeable, compassionate physicians who are prepared to tackle the opioid epidemic.


Made with FlippingBook Digital Proposal Maker