Abstract 5482: Validation of somatic variant interpretations from comprehensive cancer panels using QIAGEN Clinical Insight Interpret

Abstract

Abstract Introduction: Next generation sequencing (NGS) based assays have revolutionized the field of genetic testing and paved their way into routine clinical practice. Owing to the complexity of the data, bioinformatics has become a necessary component for any laboratory implementing a clinical NGS test. However, there is an unmet need for accurate, efficient and standardized variant interpretation for actionable insights. In this effort, AMP/ASCO/CAP released a joint consensus recommendation proposing a four-tiered system to categorize somatic alterations based on their clinical significance in cancer diagnosis, prognosis, or therapeutics for standardization of interpretation and reporting of results among laboratories. Thus, the aim of the study was to assess and validate QIAGEN Clinical Insight (QCI) Interpret software for clinical laboratory test interpretation. Also, the AMP/ASCO/CAP guidelines were compared to an accredited laboratory approach for variant classification. Patients and Methods: 87 [BL-Q1] common and rare solid tumor cases were analyzed using the TrueSight170 NGS assay and secondary analysis platform (Illumina, San Diego, CA). The data was uploaded and analyzed using QCI Interpret. Patient case draft reports were produced using a TST170 customizable workflow with automated variant filtering, and variant classifications. Molecular geneticists compared the previous expert variant classification against the QCI Interpret computed AMP tier classification. Accuracy and specificity assessed as true positive, true negative, false positive and false negative rates were determined. Results: Preliminary analysis on 52 cases with 496 alterations (SNVs, in-frame deletions, CNVs, fusions) resulted in 44 expertly classified alterations with strong therapeutic, prognostic, or diagnostic actionability (Tier I), 147 expertly classified alterations of potential therapeutic prognostic or diagnostic actionability (Tier II), and 304 variants of unknown clinical significance (Tier III). The QCI interpret computed classification yielded 44 Tier I (Strong clinical significance), 153 Tier II, 299 Tier III [SNS2] variants. Expert and QCI Interpret computed Tier 1 classification are in 100% agreement. The accuracy for Tier II and Tier III classification were 96% (153 vs. 147) and 98% (299 vs. 304), respectively. However, the Manual interrogation of the 6 discrepant variants revealed that these were emerging biomarkers that were previously characterized as Tier III but due to recent biological evidence are now classified as Tier II. Conclusion: Systematic evaluation of an automated classification based on AMP/ASCO/CAP using an up to date knowledge base is critical to accurately classify actionable mutations and detect emerging relevant mutations. The guidelines seem to yield results sufficiently good for clinical use and are a big step forward regarding standardization and use for classifying somatic alterations. The QCI Interpret seems to be a critical bioinformatics tool that adheres to the AMP/ASCO/CAP classification system and can be used across clinical laboratories for accurate, efficient and standardized variant interpretation for actionable insights. Citation Format: Nikhil Sahajpal, Ashis Mondal, Meenakshi Ahluwalia, Allan Njau, Ravindra Kolhe. Validation of somatic variant interpretations from comprehensive cancer panels using QIAGEN Clinical Insight Interpret [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5482.