Computational Reasoning for the N-of-1 Use of Molecularly Targeted Cancer Therapies

C.E. Credits: P.A.C.E. CE Florida CE
Speaker

Abstract

The personalized treatment of each cancer patient with targeted therapies selected based on our understanding of the molecular biology of cancer has been the long-standing goal of precision oncology. The Human Project, completed 20 years ago, provided the blueprint for the following cancer genome projects. Today, we can identify the genetic (driver) alteration in 95% of cancer cases based on whole genome sequencing (WGS), Pan-Cancer Analysis of Whole Genomes (PCAWG) Nature 578, 82–93 (2020). According to this study, each cancer harbors an average combination of four to five genetic driver alterations out of thousands of possible mutations of hundreds of cancer driver genes. The early successes of personalizing treatment decisions based on matching one genetic alteration as a predictive biomarker to one targeted therapy are not reproducible in most patients. To solve this problem, we have developed a computational (AI) reasoning method to select N-of-1 treatment options based on the totality of molecular information available for each case instead of providing information about the actionability of genetic alterations one by one.

Learning Objectives: 

1. Review the concept of precision oncology.

2. Explain the dimensionality problem of precision oncology. 

3. Summarize the use of different types of AI in precision oncology.