Abstract A2-38: Interpretation and classification system for somatic variants identified in solid tumor molecular profiling

Abstract Variant classification schemes for clinical laboratory reporting of inherited variants from molecular diagnostic tests for germline conditions have been widely published. These group variants by pathogenicity, distinguishing benign variants from those known or likely to be pathogenic. In contrast, there are no published schemes for somatic variant classification in acquired cancer. Factors such as histology, cancer type and actionability must be considered to determine the variant's clinical significance. We present a somatic variant classification scheme based on our experience in solid tumor molecular profiling using next-generation sequencing (NGS). Our protocol for somatic variant assessment from solid tumor NGS molecular profiling is comprised of: a) Determination of frequency of the variant in population databases, b) Information gathering on the variant from publicly available databases, c) Functional prediction using in silico tools for missense variants, d) Literature searches for publications relevant to variant function and actionability in the context of tumor type. Grading of Recommendations Assessment, Development and Evaluation (GRADE) principles are applied to determine whether evidence is sufficient to classify a given variant based on actionability. We applied this protocol to classify a pilot set of 258 variants in 158 consecutive patients tested using NGS. We present a classification system to interpret significance of genetic variants in molecular analysis of cancer, utilizing key factors: a) known or predicted pathogenicity of the variant; b) primary site and tumor histology in which the variant is found; c) whether the variant is recurrent in the specific gene; and, d) evidence of clinical actionability for patient management including targeted therapies. We used these factors to develop a 5-category Somatic Variant Classification scheme, for simplified reporting of variant interpretations to treating oncologists. Using this system, we classified 258 variants identified in 158 patients tested using NGS, and evaluated factors impacting the classification. In addition to the subset of findings with known clinical significance (37% of variants), a majority of the findings were potentially clinically actionable by extrapolating from evidence in other tumour types and recurrent variants of the same gene (49%). Classification depended on: Definition of “actionability”; primary tumor site and histology; level and type of evidence available; and, variant frequency. The pathogenicity of a specific gene/variant was distinct from its actionability; although both were indicative of biological relevance, only the latter informed patient management. By focusing on actionability, the SVC attempts to gauge the impact of genomic findings on patient management and care, bringing the most clinically relevant findings to the forefront of a list identified by NGS. Our Somatic Variant Classification scheme uses objective criteria to provide a structured stratification of the clinical significance of a somatic variant in a given histopathology, for a given patient, and for guiding laboratory procedures with respect to reporting. The distinction between “actionability” and “pathogenicity,” and the relevance of the former to the oncology setting, distinguishes our proposed categorization system from previously published classifications. The SVC can be applied to genomic datasets using various detection platforms, to track over time how advances in the field and new knowledge are affecting clinical care. This classification system enables an objective assessment over time of the relationship between available genomic information and the number of actionable findings which may impact patient care. Citation Format: Mahadeo A. Sukhai, Mariam Thomas, Kenneth J. Craddock, Tong Zhang, Tracy L. Stockley, Suzanne Kamel-Reid. Interpretation and classification system for somatic variants identified in solid tumor molecular profiling. [abstract]. In: Proceedings of the AACR Special Conference on Translation of the Cancer Genome; Feb 7-9, 2015; San Francisco, CA. Philadelphia (PA): AACR; Cancer Res 2015;75(22 Suppl 1):Abstract nr A2-38.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.