Abstract PD1-03: Clinically actionable pathogenic mutations that may be missed by conventional NGS-based testing: An analysis of 80,000 patients

• Published in: Cancer research (Chicago, Ill.) Vol. 78; no. 4_Supplement; p. PD1-03
• Main Authors: Lincoln, SE, Truty, R, Zook, J, Aradhya, S, Salit, ML, Nussbaum, RL
• Format: Journal Article
• Language: English
• Published: 15-02-2018
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.SABCS17-PD1-03

Abstract
Background: In appropriately tested patients with a personal or family history of cancer, the comprehensive assessment of inherited mutations in cancer susceptibility genes is crucial to clinical decision making. Conventional laboratory methods based on NGS (next-generation sequencing) focus on the detection of certain mutation types, notably single nucleotide variants (SNVs) and small insertions/deletions (indels), in readily accessible protein coding regions of the patients' DNA. These methods may be augmented to also detect mid-sized copy number changes (del/dup events). However, other types of clinically significant alterations can be invisible to these approaches, and for this reason their impact has been less well studied. We investigated the prevalence of such mutations in a large patient population focusing on 19 genes included in the NCCN guidelines for hereditary breast and ovarian cancer.
Methods: A diverse set of technical methods beyond conventional NGS sequencing were implemented to detect and confirm the presence of DNA alterations in over 80,000 patients. These patients were physician-referred for genetic testing and almost all had a directly relevant personal and/or family history. Pathogenicity was rigorously assessed using a method based on the ACMG 2015 guidelines. Among the technical methods used were long-read single molecule sequencing, breakpoint detection, MLPA, microarrays, and de novo methods for homopolymer associated mutations.
Results: Over 8.6% of patients with a pathogenic, potentially actionable germline mutation in one (or more) of the 19 NCCN genes harbored a mutation not easily detected by conventional sequencing or del/dup methods. In the subset of 11 NCCN genes specifically associated with breast cancer, the prevalence was slightly lower (6.8%) yet still substantial. No single class of mutation was responsible for this; rather a diversity of laboratory technical challenges were present. These included both challenging mutation types and also alterations that lie within challenging regions of these 19 genes. Data from the Precision FDA effort are consistent with these findings, showing >10-fold higher error rates for many such events when using standard NGS approaches alone.
Conclusions: Our study explored the boundaries of conventional laboratory methods and found that a significant fraction of pathogenic mutations in patients are of types that require specialized biochemical or bioinformatics methods to be used. Accurate reporting of these technically “hard” mutations is crucial to the correct diagnosis and management of cancer patients and their families, and omission of these methods can lead to high false negative rates. Based on our review of published studies, as well as a separate interlaboratory study (see associated abstract by Lincoln et al.), these methods are not yet uniformly implemented. Moreover, as some of these same genes are becoming relevant to therapeutic selection based on somatic mutation profiles, these same considerations will soon arise in tumor testing, we suspect.
Citation Format: Lincoln SE, Truty R, Zook J, Aradhya S, Salit ML, Nussbaum RL. Clinically actionable pathogenic mutations that may be missed by conventional NGS-based testing: An analysis of 80,000 patients [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr PD1-03.