Abstract LB-104: Clinical actionability and clinical trial matching for GENIE patient genotypes using My Cancer Genome, Personalized Cancer Therapy, and OncoKB

Abstract

Abstract AACR Project GENIE is an international data-sharing project with the goal of enhancing precision oncology. On January 5, 2017, data from ~19,000 somatic tumor genotype reports and selected clinical information from patients (http://www.aacr.org/RESEARCH/RESEARCH/PAGES/AACR-PROJECT-GENIE-DATA.ASPX) were released to the public. The eight member institutions contributed patient data, which were then made available on a dedicated cBioPortal website and for download by Sage Bionetworks. As part of the analysis of the first data release, member institutions began work to combine and reconcile their clinical actionability knowledgebases (KBs). In this abstract, we present efforts to combine clinical actionability assertions from My Cancer Genome (MCG; https://www.mycancergenome.org; Vanderbilt-Ingram Cancer Center), Personalized Cancer Therapy (https://pct.mdanderson.org; MD Anderson Cancer Center), and OncoKB (http://oncokb.org; Memorial Sloan Kettering Cancer Center). We also present data on matching GENIE patient genotypes to clinical actionability assertions from these KBs and to biomarker-driven clinical trials to begin defining the landscape of clinical actionability across a large cohort of patients at all stages of disease.The three KBs were combined using a modification of OncoKB’s levels of evidence for clinical actionability. Categories included standard of care therapies (e.g., those with FDA labels and in NCCN guidelines) and investigational therapies with strong clinical data, both on and off the recommended diagnosis indications. Diagnoses were mapped to the OncoTree tumor type hierarchy. Initial efforts to combine the KBs resulted in >500 therapeutic assertions. Using the combined KBs, we matched these assertions to GENIE patient diagnoses and genotypes. In our preliminary results, >33% of patient samples match at least one therapeutic assertion. Of these, ~15% match at the standard of care level, and ~8% match at the level of investigational therapies. The remaining matches were exploratory or matched by biomarker but not diagnosis. Most frequently matching diagnoses at the standard-of-care level were non-small cell lung cancer, breast cancer, and melanoma. Further details will be presented.The MCG team curates diagnosis and biomarker eligibility criteria for all recruiting cancer clinical trials reported in ClinicalTrials.gov. As of January 2017, 5,201 recruiting cancer clinical trials from ClinicalTrials.gov have been reviewed, and 1,884 trials were found to have biomarker eligibility criteria. Of these, 352 trials are testing a targeted therapy and have a known driver mutation as an inclusion criterion. Based on preliminary work, ~16% of patient samples match at least one trial in the limited set. When the trial list is expanded to include all biomarker-driven trials, including those exploring the impact of mutations along an entire cell signaling pathway, ~84% of patient samples match at least one trial; matching biomarkers are often exploratory and patient benefit from the trial intervention is not necessarily expected. For the limited trial set, patients with breast cancer, non-small cell lung cancer, glioma, and melanoma were most likely to match a trial. For the expanded trial set, patients with non-small cell lung cancer, colorectal cancer, breast cancer, and glioma were most likely to match a trial. We will show how genetic testing panel size affects trial matching and present clinical trial matching by disease, gene, alteration type, drug class, and cell signaling pathway.In conclusion, the GENIE project has resulted in more than just the shared data. It has fostered collaborations between institutions to reconcile and improve precision cancer medicine KBs and provided a resource for improving cancer genetic testing and the practice of precision cancer medicine. [C. M. and D. C. contributed equally to this work.] Citation Format: Christine M. Micheel, Debayani Chakravarty, Jiaojiong Gao, Ian Maurer, Clinton Miller, Kenna R. Shaw, Mia A. Levy, Nikolaus Schultz, on behalf of the AACR Project GENIE Consortium. Clinical actionability and clinical trial matching for GENIE patient genotypes using My Cancer Genome, Personalized Cancer Therapy, and OncoKB [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-104. doi:10.1158/1538-7445.AM2017-LB-104