Circulating tumor DNA as a predictor of treatment response to neoadjuvant chemotherapy in stage 2/3 rectal cancer without high risks: Subset analysis from the COPEC trial.

e15623 Background: Circulating tumor DNA (ctDNA) serves as a valuable tool for minimal residual disease (MRD) assessment. It has been reported as a promising noninvasive biomarker for the dynamic monitoring of treatment response of metastatic colorectal cancers to chemotherapy and locally advanced rectal cancers (LARC) to neoadjuvant chemoradiotherapy. However, the potential role of ctDNA in neoadjuvant chemotherapy (NCT) in LARC remains unexplored. The present study aims to investigate the value of ctDNA in NCT based on a prospective, multicentered, non-inferior, randomized, controlled trial (ClinicalTrials.gov ID, NCT04922853). The COPEC trial enrolls stage 2/3 LARC with low or intermediate and is designed to determine the optimal time (after 2 or 4 cycles of CAPOX) to evaluate treatment response based on pathological regression grades. Methods: Colonoscopy biopsy was taken before neoadjuvant therapy. Peripheral blood samples were collected before neoadjuvant (baseline, BL), one cycle (C1) and two cycles (4-cycle group, C2) after NCT, before surgery (BS), and within one month after surgery (postoperative, OP). All tissue biopsy and plasma specimens were analyzed by NGS assays using the MinerVA platform (Genecast Biotechnology Co., Ltd). MRI tumor regression grade (mrTRG) was assessed after 2 and 4 cycles (4-cycle group) of NCT. The primary outcome was the accuracy of ctDNA in the prediction of pathological tumor regression grade (pTRG). Results: A total of 151 LARC patients were enrolled, including 82 in the 2-cycle group and 69 in the 4-cycle group. The ctDNA was detectable in 87.4% (132/151), 39.6% (55/139), 29.0% (18/62), 38.4% (58/151), and 4.4% (6/138) at BL, C1, C2, BS and OP, respectively. The ctDNApositivity rates were significantly lower in good responders determined by mrTRG (mrTRG 1-3) compared to mrTRG 4-5 group at C1 (29% vs. 63.89%, p < 0.001), C2 (20% vs.56.25%, p = 0.011), and BS time points (23.85% vs. 74.36%, p < 0.001). Furthermore, the ctDNA-positivity rates at C1 in the good response group determined by TRG (pTRG 0-1) were 6.25%, significantly lower than that in the pTRG 2-3 group (49.53%). At neither C2 nor BS time point, patients in the pTRG 0-1 subgroup were detected as ctDNA positive while the ctDNA positivity rates for patients in pTRG 2-3 subgroup were 39.13% and 50%, respectively. Additionally, the mean of hGE/ml was 0.045 in pTRG 0-1 group at C1, significantly lower comparing to 1.337 in pTRG 2-3 group (p < 0.001). Conclusions: We find that dynamic monitoring of ctDNA status was a useful tool in early predicting the effect of (treatment response to) NCT in LARC patients with low/intermediate risks. The monitoring of ctDNA status can supplement MRI in determining the NCT response and could potentially help identify poor responders to discontinue the treatment with inadequate efficacy. Clinical trial information: NCT04922853 .