Clinical Quality in Cancer Research: Strategy to Assess Data Integrity of Germline Variants Inferred from Tumor-Only Testing Sequencing Data

Abstract

In the majority of cancers, pathogenic variants are only found at the level of the tumor; however, an unusual number of cancers and/or diagnoses at an early age in a single family may suggest a genetic predisposition. Predisposition plays a major role in about 5 to 10% of adult cancers and in certain childhood tumors. As access to genomic testing for cancer patients continues to expand, the identi cation of Potential Germline Pathogenic Variants (PGPV) through tumor-DNA sequencing is also increasing. Statistical methods have been developed to infer the presence of a PGPV without the need of a matched normal sample. These methods are mainly used for exploratory research, for example in realworld Clinico-Genomic Databases/platforms (CGDB). These databases are being developed to support many applications such as targeted drug development, clinical trial optimization and post marketing studies. To ensure the integrity of data used for research, a quality management system should be established, and quality oversight activities should be conducted to assess and mitigate clinical quality risks (for patient safety and data integrity). As opposed to well-de ned "good practice" quality guidelines (GxP) areas such as good clinical practice, there are no comprehensive instructions on how to assess the clinical quality of statistically derived variables from sequencing data such as PGPV. In this report, we aim to share our strategy and propose a possible set of tactics to assess the PGPV quality and to ensure data integrity in exploratory research.

Inferences of Mutational Status for Variants Detected From Tumor-Only Sequencing—A Quality Assurance Strategy

“…We provided several examples in our previous study on how these checks can be designed. 2 The authors suggested that integration of clinical, familial and ancestry information may improve the probability of PGV identification. 1 This could be achieved by leveraging clinicogenomic databases, which contain longitudinal data (eg, histology and laboratory data) of patients who undergo tumor genomic testing.…”

“…1 This could be achieved by leveraging clinicogenomic databases, which contain longitudinal data (eg, histology and laboratory data) of patients who undergo tumor genomic testing. 2 A last suggestion for an efficient quality check would be to cross-check inferred PGV against the Cancer Genes Census (CGC) list provided in the Catalogue of Somatic Mutations In Cancer. 7 Cancer genes are provided in a table and can be filtered for germline—it should be easy to build in the pipeline an automated check against the CGC table, to discard potential false-positive PGVs.…”

“…The study is very relevant to the field as the identification of pathogenic germline variants (PGVs) through tumor DNA sequencing continues to increase, and PGV can be detected in various cancers including those not typically associated with hereditary predisposition. 1,2 In a previous article, we designed a quality assurance (QA) strategy to assess the integrity of PGV data inferred from tumor-only sequencing data. 2 Of note, our study reviewed the LOHGIC algorithm, 3 which had been developed by the authors and was used again for their validation study, combined with All-FIT—a model to estimate tumor purity.…”

“…1,2 In a previous article, we designed a quality assurance (QA) strategy to assess the integrity of PGV data inferred from tumor-only sequencing data. 2 Of note, our study reviewed the LOHGIC algorithm, 3 which had been developed by the authors and was used again for their validation study, combined with All-FIT—a model to estimate tumor purity. 4 Our approach differed significantly in that we looked at how researchers can get assurance on PGV data integrity, without running a direct comparison against germline confirmatory testing data.…”

“…Our QA strategy consisted in a framework to evaluate the algorithms (designed to infer PGV) and to develop a set of fit-for-purpose, rule-based quality checks, in the absence of access to source data for validation and verification. 2 As the authors concluded that “independent validation of this information-theoretic methodology […] is required,” 1 we would like to provide suggestions on how a QA strategy can complement and validate further the methodology developed by the authors.…”