Serial ctDNA monitoring to predict response in mCRC patients treated with anlotinib in combination with oxaliplatin and capecitabine (CAPEOX) as the first-line therapy.

Abstract

e15581 Background: Circulating tumour DNA (ctDNA) sequencing offers a non-invasive approach to monitor therapeutic efficacy in colorectal cancer. However, the mutational landscape of ctDNA during treatment and its impact on clinical outcomes remain unknown. This study aimed to investigate the value of ctDNA in predicting treatment response and PFS in mCRC patients treated with anlotinib plus CAPEOX as first-line therapy. Methods: We conducted a multicenter, single-arm, phase II, exploratory study of patients with RAS/BRAF wild-type unresectable metastatic colorectal cancer (mCRC) who underwent Anlotinib in combination with CAPEOX as the first-line therapy (NCT04080843). Plasma samples from 26 patients were collected at three timepoints: 1 week before treatment (baseline), close to the time of optimal remission (C1) and the time of the last blood collection or the time of disease progression (C2). The maximum somatic allele frequency (MSAF) was calculated for each case and used to provide an estimate of the ctDNA fraction in the blood. Results: 77 plasma samples from the 26 enrolled mCRC patients were sequenced by NGS based on a panel comprising 202 genes. 90 variations were detected in baseline plasma, with a positive rate of 96.2% (25/26). Adenomatous polyposis coli ( APC), was the most mutated gene (84.6%, 22/26). Other mutated genes were TP53 (80.8%, 21/26), FBXW7 (19.2%, 5/26), KRAS (11.5%, 3/26), PIK3CA (11.5%, 3/26) and ROS1 (11.5%, 3/26). The median MSAF of baseline ctDNA was significantly higher in patients who had only liver metastases. Compared with baseline, the ctDNA MSAF level at C1 timepoint was decreased among the whole 26 patients (p < 0.001). By comparing ctDNA positive group (ctDNA MSAF > 0) and ctDNA negative group (ctDNA MSAF = 0) at C1 timepoint, CR cases were mainly in ctDNA negative group, while PD cases accounted for a high proportion in the ctDNA positive group. The median progression-free survival (mPFS) of the whole patients was 11.27 months (95%CI 7.33m to 14.52m). The mPFS was 14.52 months (95% CI 14.4 to NR) of patients with ctDNA negative at C1, which was longer than 9.17 months (95% CI 7.16 to 14.06) of patients with ctDNA positive at C1 (p = 0.012). For 25 patients who had plasma samples of three timepoints, we divided the patients into Non-PD group and PD group according to the clinical outcome at C2 timepoint. Compared with C1 timepoint, the ctDNA MSAF level at C2 timepoint was increased in both Non-PD group and PD group. However, the increment of ctDNA MSAF level in PD group was higher than that in non-PD group (p = 0.000069 vs p = 0.024). Conclusions: This ctDNA profiling study revealed the utility of ctDNA in predicting clinical outcomes in mCRC patients under first-line treatment. Serial ctDNA monitoring might become useful in an active surveillance strategy for the management of mCRC. Clinical trial information: NCT04080843 .