Abstract Background: Minimal residual disease (MRD) detection by ctDNA analysis enables identifying patients with high recurrence risk and refining risk stratification to guide treatment selection precisely, which may improve patient survival or avoid overtreatment. We previously developed a 1021-gene fixed panel MRD detection assay, MNavigator V1, demonstrating high prognostic evaluation value in various solid tumors, including lung cancer, colorectal cancer, and esophageal carcinoma, and relevant research results have been published. Recently, the assay upgraded to MNavigator V2, a customized patient-specific panel strategy to ultra-deep sequencing, with higher clinical performance and more economical. Here, we head-to-head compared the performance of MNavigator V1 with V2, using the same collection of blood samples from TNBC patients. Methods: The retrospective study recruited 48 patients diagnosed with stage II~III TNBC with adequate samples (NCT04501523/NCT04803539). Five patients in the cohort received neoadjuvant therapy (Neoad-group), and the standard treatment regimen of the remaining 43 patients was adjuvant therapy (ad-group) after surgery. The surgical tumor tissue of all these patients was collected. Peripheral blood samples were collected at various time points, baseline samples before neoadjuvant therapy, 7-day after neoadjuvant therapy, 1-month postoperative, 7-day after adjuvant therapy, and subsequent every 3~6 months follow-up. The personalized, tumor-informed MRD detection approach MNavigator V2 started by identifying somatic mutation of the resected tumor with a 1021-gene panel covering 1.6Mb of the human genome. Then, a customized panel consisting of up to 20 top-ranked patient-specific somatic variants was designed. After that, unique molecular identifiers (UMI) based 100,000X ultra-deep next-generation sequencing on plasma samples was conducted by adopting the mix panel of the patient-specific panel and a 5kb breast cancer universal core panel. MNavigator V1 assay, a fixed panel assay also applied the same off-the-shelf 1021-genes panel with a sequencing depth of 10,000X to detect MRD of blood samples. Results: Until May 2023, the median follow-up time of 660 days, and 8 patients were recurrence confirmed by CT imaging. The median tracked tissue-informed mutations was 7 (2~18). In this study, the screening time point was defined as the combined MRD status of the blood samples collected before and after neoadjuvant therapy, and 1-month post-surgery for the Neoad-group, as for the ad-group, the time point referred to 1-month post-surgery and post-adjuvant therapy. If any one of these time nodes' MRD was positive, the screening period MRD was positive. During the screening period, MNavigator V1 and V2 accurately predicted 6/8 (75.0%) recurrence. MRD-positive status was associated with a worse prognosis, compared to the MRD-negative population, with hazard ratios of 9.0 and 19.9 for V1 and V2 assay, respectively. The longitudinal MRD evaluation yield a higher sensitivity of 7/8 (87.5%) for recurrence prediction with MNavigator V2 at a median lead time of 104 days, whereas the V1 assay predicted 6/8 (75.0%) at 94 days, and the hazard ratio was 40.4 vs 8.5. In one brain-only metastasis patient, MRD was undetectable, which may be due to the limitation of the blood-brain barrier. For recurrence-free patients with 1 year of follow-up, 92.5% (37/40) were MRD-negative with MNavigator V2 assay. These three patients who detected MRD-positive after surgery, ctDNA cleared after adjuvant therapy, and persistently negative during follow-up, may benefit from treatment. Conclusion: Based on the patient-specific tumor mutation profile, the personalized MRD detection assay achieved better clinical performance than the fixed panel strategy. Citation Format: Shunying Li, Wei Gao, Ning Fu, Wuqiang Cao, Xiaoling Zeng, Xinhua Du, Qiang Liu. Tissue-informed personalized MRD detection assay may outperform tumor-informed fixed panel strategy in Triple Negative Breast Cancer (TNBC) [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-16-04.
Read full abstract