Abstract 3927: Analytical performance of a novel tissue-informed non-bespoke whole genome MRD detection assay.

Abstract Background: Detection of circulating tumor DNA (ctDNA) in early-stage cancer offers significant prognostic value and has potential for guiding clinical care decisions. However, achieving sufficient sensitivity for reliable detection at this stage is challenging. Personalized panels can increase turnaround time and operational complexity.To overcome these limitations, we developed a tissue-informed molecular residual disease (TI-MRD) laboratory developed test that does not use custom panels. Methods: The TI-MRD assay defines a personalized tumor signature from whole genome sequencing (WGS) of tissue FFPE, and scans for this signature in cell-free DNA (cfDNA) from plasma, also sequenced using WGS. TI-MRD employs a novel algorithm using aneuploidy and somatic SNVs to define a tumor signature without requiring a matched normal sample. Samples from patients with early stage lung and colon cancer were used in a study for determining the lower limit of detection (LLoD) of the assay. The precision study also included samples from patients with breast, bladder, melanoma, ovarian, and pancreatic cancers. Analytical specificity of the assay was assessed using cfDNA from donors with no known history of cancer and a tumor signature derived from a panel of cancer FFPE samples. Results: The LLoD for the TI-MRD assay was established using sample titrations ranging from > 44,000 PPM (> 4.4%) to 5 PPM and was based on a ≥ 95% probability of detecting tumor fraction (TF). LLoD for the TI-MRD assay was also evaluated as a function of tumor mutational burden (TMB), where the lowest TMB value tested was < 1 mut/Mb and the highest TMB value tested was > 9 mut/Mb. The titration data for estimating the analytical sensitivity for the TI-MRD assay demonstrated that the LLoD can be < 10 PPM. Reproducibility was determined by testing 6 replicates from 10 patients with TF values ranging from 380,000 PPM (38%) to 32 PPM. Precision was also evaluated as a function of cfDNA input mass (3.5 - 20ng) and FFPE input mass (10 - 220ng). Detection of ctDNA was highly reproducible across the range of cfDNA and FFPE input masses and was maintained across a 4-log range of TF values. The TI-MRD assay also achieved an empirical specificity of 100% from donors with no known history of cancer. Conclusion: This TI-MRD assay demonstrated robust analytical performance with high sensitivity, reproducibility, and specificity for detecting ctDNA, highlighting the potential of this assay to guide therapeutic decisions in the early setting across many different cancer types. Citation Format: David Delfosse, Alexander Fine, Daokun Sun, Akshay Kakumanu, Tristen Ross, Devika Singh, Ravin Poudel, Maryam Zand, Brian Reilly, Farzana Ahmed, Liv Parsons, Tuan Nguyen, Ena Shinnishi, Noel Vega, Hanna Tukachinsky, Chang Xu, Alex Robertson, Brett Wallden.. Analytical performance of a novel tissue-informed non-bespoke whole genome MRD detection assay [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 3927.