Abstract 978: Functional characterization and a real-world clinical laboratory pilot of the Foundation for the National Institutes of Health’s (FNIH) circulating tumor DNA (ctDNA) quality control materials (QCMs)

Abstract Background: Analysis of ctDNA now permits assessment of targetable mutations (e.g. EGFR L858R/Ex19del) using noninvasive blood tests. There are several FDA-approved tests using digital PCR (ddPCR) or next-generation sequencing (NGS) and many more laboratory developed tests in use. The FNIH Biomarkers Consortium identified the absence of validated standards to qualify the performance of ctDNA assays as a critical need and set out to aid in the characterization of QCMs. Here we report on a commutability study comparing the performance of three commercial QCMs to clinical samples at or above LoD (limit of detection) and update on a real-world assessment of these QCMs by 11 clinical laboratories spanning 4 continents and 3 assay modalities. Methods: QCMs were provided by Horizon Discovery (H), Thermo Fisher (T) and SeraCare (S). Clinical samples with known EGFR L858R and Ex19del mutations were prepared by diluting specimens at known concentrations into cell-free DNA (cfDNA) from healthy donors. The functional characterization compared detection of the mutations in QCM to clinical samples and consisted of 2 parts: (1) Evaluation of EGFR L858R/Ex19del (range 0.1% to 5% allele frequency [AF]) by ddPCR, amplicon and hybrid-capture NGS; (2) Evaluation near the LoD of the amplicon NGS assay at the assay LoD95 of 0.3% AF. For the clinical pilot, participants were sent QCM pools formulated at 0.5% and 1% AF by ddPCR. Some participants also tested clinical specimens with EGFR L858R and Ex19del at 0.5% and 1% AF by ddPCR. All participants were blinded to AF and identity of the manufacturer. Results: For Part 1, QCM results from hybrid-capture NGS of EGFR L858R were similar to ddPCR; for Ex19del, QCM AF trended lower compared to ddPCR. In contrast, L858R clinical samples showed a strong positive bias compared to ddPCR. For amplicon NGS, QCMs performed similarly to clinical samples, with amplicon NGS AF closest to ddPCR AF for T QCM. For Part 2, the observed hit rates mostly approximated expected values. For H QCM, observed LoD95 (10% for EGFR L858R; 50% for Ex19del) and subsequent dilution hit rates were lower than expected. For the clinical pilot, median Ex19del AF of amplicon NGS were higher than for hybrid-capture NGS for both the 1% QCM formulations and 0.5% QCM formulations. Median QCM L858R AF were similar for amplicon and hybrid-capture NGS, with greatest contributor to variance being inter-lab differences. Conclusions: This project, unprecedented in the number of participants and analytical rigor, revealed unexpected differences in performance of both assays and QCMs. These findings merit consideration by laboratories that rely on QCMs to develop and perform assays for ctDNA testing as diagnostics. Citation Format: Cloud P. Paweletz, Thomas D. Forbes, Grace A. Heavey, Laura M. Yee, Hua-Jun He, Zhiyong He, Dana E. Connors, Daniel Stetson, Susan Keating, Kenneth D. Cole, Gary Kelloff, Robert T. McCormack, Mickey P. Williams, Chris Karlovich. Functional characterization and a real-world clinical laboratory pilot of the Foundation for the National Institutes of Health’s (FNIH) circulating tumor DNA (ctDNA) quality control materials (QCMs) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 978.