Association of Pathologic Nodal Status with Minimal Residual Disease after Neoadjuvant Treatment and Resection of Locally Advanced Rectal Cancer.

205 Background: Studies have shown that minimal residual disease (MRD) identified by detection of circulating tumor DNA (ctDNA) is associated with recurrence after surgical resection of colorectal cancer. Though multiple trials are evaluating the use of ctDNA to guide adjuvant therapy, data on the utility of ctDNA after neoadjuvant therapy is limited. In this study, we evaluated the associations between ctDNA and recurrence after neoadjuvant treatment and resection of rectal cancer (LARC). Methods: Consecutive patients with primary rectal cancer treated with neoadjuvant systemic therapy and/or chemoradiotherapy followed by resection between 12/2020-4/2023 were identified. Patients were tested for ctDNA via the MD Anderson INTERCEPT platform using a high-sensitivity patient informed assay. Patients with metastases or a first ctDNA test more than 3 months after resection were excluded. MRI-determined stage, extramural vascular invasion (mrEMVI), pelvic sidewall adenopathy (PSW), pathologic lymphovascular invasion (LVI), perineural invasion (PNI) and tumor regression grade (TRG) were collected. Associations between these factors and ctDNA status were analyzed with Fischer’s exact test. Recurrence free survival (RFS) was analyzed with the log-rank test. Results: Sixty-seven patients (3 treated with chemoradiotherapy alone, 15 with systemic therapy alone and 49 with total neoadjuvant therapy) were identified. Positive ctDNA was identified in 6 (8.9%) within 3 months after resection, all of whom received total neoadjuvant therapy, and 3/6 recurred within 12 months. Clinical T status (p=0.146), N status (p=0.842), mrEMVI (p=0.475) and PSW (p=0.318) were not associated with MRD. TRG was not associated with MRD (TRG0-1: 4% vs TRG2-3: 11.3%, p=0.547). Pathologic N status was associated with MRD (ypN0: 4.4% vs ypN+: 18%, p=0.049). MRD was associated with a median RFS of 3.2 months (HR 39, 95% CI: 4.1-380, p=0.0015). 1-year RFS was 98% for patients without MRD. Conclusions: Minimal residual disease after multi-modality treatment of LARC is associated with early metastatic recurrence. The patient informed MRD assay appears both sensitive and specific for detection of relapse. Advanced clinical stage and high-risk radiologic features were not associated with MRD, while post-treatment pathologic N status was. Given the high relapse rate for MRD+ patients, MRD detection after resection warrants additional investigation for metastases and/or enrollment on MRD treatment clinical trial. Methods to predict MRD prior to resection are urgently needed.