Statistical Methods to Associate Intra-Tumor Heterogeneity with Clinical Outcomes

• Activity Number: 535 - Contributed Poster Presentations: Section on Statistics in Genomics and Genetics
• Type: Contributed
• Date/Time: Wednesday, August 1, 2018 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Paul Little* and Danyu Lin and Wei Sun
• Companies: UNC Chapel Hill and University of North Carolina and Fred Hutchinson Cancer Research Center
• Keywords: intratumor; heterogeneity; subclones; frequentist; survival; recurrence

Abstract

Somatic mutations are vital for tumor initiation and progression. Most somatic mutations are rare across cancer patients and thus it is very challenging to study their associations with clinical outcomes. We studied associations between somatic mutations and survival time for 14 cancer types using omic data and clinical data from The Cancer Genome Atlas (TCGA) project. We summarized somatic mutation data by tumor mutation burden (TMB), somatic copy number aberration (SCNA) burden, as well as the degree of intra-tumor heterogeneity. TMB is the total number of single nucleotide variants (SNVs). SCNA burden is calculated as the summation of the genomic regions affected by SCNA, weighted by the deviation of copy number from diploid. The most challenging feature is the degree of intra-tumor heterogeneity (ITH). With only one tumor sample per patient, there is often an identifiability issue to precisely determine the phylogenetic tree of tumor subclones. We have developed a statistical method, SMASH (Subclone Multiplicity Allocation and Somatic Heterogeneity) to estimate the likelihood of each phylogenetic tree and to incorporate the uncertainty of ITH inference into association studies.