Improving minimal residual disease detection: A tumor-informed whole-genome approach.

Abstract

e15089 Background: Minimal Residual Disease (MRD) detection is rapidly becoming an effective tool for cancer management, enabling early relapse identification and guiding treatment decisions. Current MRD assays typically rely on targeted sequencing of a limited set of somatic mutations, with limits of detection (LOD) in the range of 100–1000 parts-per-million (PPM). Methods: This study presents the analytical validation of a novel approach that combines target-enhanced whole-genome sequencing, CancerVision (Inocras), with ppmSeq (Ultima Genomics). CancerVision’s comprehensive genomic footprint powers whole-genome circulating cell-free DNA (cfDNA) sequencing, capturing a broad spectrum of cancer-specific mutations. Leveraging ultra-sensitive sequencing, this approach achieves a markedly enhanced LOD of 2–15 PPM (2 × 10⁻⁶ to 15 × 10⁻⁶). Analytical validation was performed using three tumor and matched-normal cell line pairs (HCC2218, HCC1395, and NCI-H2126) from the American Type Culture Collection. Tumor DNA was diluted into matched-normal DNA at concentrations ranging from 10⁻² to 10⁻⁷ to simulate circulating tumor DNA (ctDNA) fractions, with each cell line tested across 15 dilution levels, resulting in a total of 28 samples per cell line. Results: Samples were sequenced with ppmSeq at a depth of 29–55x, achieving a mixed (duplex) rate of 25–39% and an absolute sequencing error rate of 6.3 × 10⁻⁷. Integration of 51,350–169,079 tumor-specific somatic mutations identified by CancerVision, combined with advanced bioinformatics for artifact removal, enabled detection of ctDNA at dilutions as low as 2 PPM (2 × 10⁻⁶). This represents a significant improvement over conventional targeted sequencing methods for MRD detection. Conclusions: This tumor-informed whole-genome approach for MRD detection provides a next-generation, non-invasive solution for cancer monitoring. Whole-genome ppmSeq™ offers a rapid, streamlined workflow that eliminates the need for custom panel design and storage. By enabling earlier relapse detection, real-time tumor dynamics tracking, and more precise therapeutic decision-making, this approach holds the potential to transform patient outcomes in oncology.