Abstract 5886: Cross-test performance verification of tissue-TMB and a cfDNA somatic variant Next Generation Sequencing (NGS) panel for detection of actionable mutations

Introduction: Comprehensive tumor profiling by NGS has become widespread in oncology. However, a relatively small number of the developed panels have gained regulatory approvals and/or authorization. Given the need for higher levels of compliance for panels used for clinical laboratory testing, we sought to extend the utility of panels intended for use in blood and tissue testing. In this study we verified the performance of a 52-gene liquid biopsy hotspot panel [evaluated for single nucleotide variants (SNVs) and insertions/deletions (Indels) as low as 0.5% allele frequency (AF) and copy number amplifications (CNAs) down to 1.4x] to be used with formalin-fixed paraffin embedded (FFPE) specimens. In the same study, we evaluated the performance of a well characterized tissue TMB/TML (tumor mutation burden) assay for concordance in somatic variant calling against the hotspot panel. The targeted TMB/TML assay covers 1.7 Mb of the genome (1.2 Mb exome) and detects variants at 5% AF. The performance of the latter panel was previously reported in the harmonization study conducted by the FOCR TMB consortia. Methods: Twelve de-identified clinical FFPE specimens and a commercial (contrived) FFPE reference material were tested for SNVs and Indels using the validated 52-gene hotspot pan-cancer assay and the TMB/TML panel. Co-occurrence of SNVs/INDELs (n = 40) was observed between the two panels. The reference result for the clinical FFPE specimens was generated by the variant caller function in the TMB/TML bioinformatic workflow and included 14 SNVs. The pan-cancer panel was also used to test for 19 SNVs and 7 Indels present in the contrived FFPE reference specimen. Results: TMB scores ranged from 1.7-104 Muts/Mb. A high level of inter chip concordance was observed between duplicate runs (R2 = 0.997). Duplicate NGS libraries were generated from the reference contrived FFPE specimen, and both passed QC. Eleven of 12 clinical samples passed the pan-cancer informatics QC thresholds. Thirteen of 14 SNVs (93%) were detected in the clinical samples (AF range: 13-58%) and eighteen of 19 SNVs (95%) and 5 of 7 Indels (71%) were detected in both replicate tests of the FFPE reference material (AF range:1.9-13.9%). Although no reference was available, one ERBB2 CNA was also detected with the pan-cancer panel (at 10.45x). Since the starting material was DNA (and not RNA or cDNA), fusion variants could not be detected although they are covered by the hotspot panel. Conclusions: These studies demonstrate extended utility for a previously validated liquid biopsy hotspot pan-cancer panel for the detection of SNVs/Indels and CNAs in FFPE. We simultaneously demonstrated the potential utility of a well characterized TMB panel for detection and reporting of actionable SNVs/Indels. Additional studies are on-going to further validate the TMB panel for actionable variant reporting. Citation Format: Leisa Jackson, Ashley King-Roads, Gary A. Pestano. Cross-test performance verification of tissue-TMB and a cfDNA somatic variant Next Generation Sequencing (NGS) panel for detection of actionable mutations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5886.