Abstract 5789: Multi-omic artificial intelligence outcome modeling of ovarian cancer, phase I: Whole exome and whole transcriptome data

Abstract Background: Over a decade ago, the Cancer Genome Atlas (TCGA) provided the initial genomic characterization of ovarian cancer with targeted exome capture and sequencing. Unlike other TCGA analysis, they were unable to identify prognostic mutational profiles outside of BRCA status. There has been limited characterization of the ovarian cancer genomic profile since. Our objective presented here was to perform whole exome and whole transcriptome analysis of 241 ovarian cancer samples and compare to the TCGA dataset. This is the first phase of a muti-step ongoing analysis wherein we will input and correlate the complete genomic profile, complete patient outcome data, and immunohistochemical staining into a multi-omic artificial intelligence (AI) machine learning model with the aim of improved individualized outcome prediction. Methods: Tumor samples from 2000-2016 were obtained at time of surgery under approved University of Pittsburgh approved IRB consent. Whole exome and whole transcriptome sequencing were performed in collaboration with Helomics Corporation. Sequencing analysis was compared to TCGA data. Results: 241 patient samples underwent sequencing analysis. Similar to TCGA, most mutations were missense. I am having a difficulty interpreting the magee vs tcga comparison graphs, how to discuss the TP53 and BRCA comparison without using the lollipop charts and highlighting the rna-seq data into written form. Please add. Conclusions: We present our findings of one of the largest molecular characterizations of ovarian cancer. Similar to the TCGA, there is noted heterogeneity to the genomic profile of ovarian cancer. Multi-omic AI outcome modeling has the potential to overcome the gap defining prognostic sub-groups so that we can tailor therapies to the individual Citation Format: Brian Orr, Robert P. Edwards, Mackenzy Radolec. Multi-omic artificial intelligence outcome modeling of ovarian cancer, phase I: Whole exome and whole transcriptome data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5789.