The effects of oxaliplatin-based adjuvant chemotherapy in high-risk stage II colon cancer with mismatch repair-deficient: a retrospective study

This retrospective and consecutive cohort study was conducted at Sun Yat-sen University Cancer Center, enrolling patients with stage II dMMR CCs between June 2011 and May 2018. All patients underwent curative resection with or without subsequent adjuvant chemotherapy. The screening strategy and treatment approach were depicted in Fig. S1.

High-risk factors were used to stratify patients into high-risk and low-risk groups. According to NCCN and ESMO guidelines, high-risk factors were defined as pT4 tumor stage, < 12 LNs, poorly differentiated histology, PNI, LVI, or preoperative carcinoembryonic antigen (CEA) > 5.0 ng/ml. Information on tumor obstruction, tumor perforation, or positive margins were rare in this cohort and hence were not included in the analysis. Preoperative CEA levels were assessed within one week before the operation, while postoperative CEA levels were evaluated within four weeks after the operation and prior to the initiation of adjuvant chemotherapy. Normalized postoperative CEA was elevated preoperative CEA (> 5.0 ng/mL) but normal postoperative CEA. Elevated postoperative CEA meaned that preoperative and postoperative CEA levels were both elevated (> 5.0 ng/mL).

MMR protein (MLH1, MSH2, MSH6, and PMS2) expression was analyzed in formalin-fixed paraffin-embedded (FFPE) tumor sections. Protein expression was considered negative when nuclear staining was completely absent in neoplastic cells. The absence of nuclear staining in tumor cells and normal staining in the surrounding normal tissue was considered as dMMR. Immunohistochemical (IHC) staining was scored by an experienced gastrointestinal pathologist.

All the categorical variables were presented with frequencies and proportions. Continuous variables were compared using the Mann–Whitney U test. Categorical variables were compared using the chi-squared test. Disease-free survival (DFS) was the primary endpoint for survival analyses. DFS was calculated from the date of surgery until the evidence of progression or the last date of follow-up and was estimated using the Kaplan–Meier method with the log-rank test. Survival data was analyzed using Cox proportional hazard regression. Only variables with P values ≤ 0.10 in univariate analysis were eligible for the multivariate Cox regression model. Statistical analyses were performed using the R software (version 4.0.2; http://​www.​r-project.​org/​) and SPSS (Version 20.0; IBM Corp, Armonk, NY, USA). All tests were two-sided and statistical significance was set at P < 0.05.

Patient and disease characteristics were shown in Table 1. Between 2011 and 2018, 262 consecutive dMMR patients were identified with stage II CCs, of whom 179 patients had at least one high-risk factor; 88 (49.2%) had only one, and 91 (50.8%) had two or more high-risk factors. The median age of the patients was 54 years (range 22–85 years). Among all the patients, 61.8% were male and 38.2% were female. The common high-risk factors included poor differentiation (37.4%), elevated preoperative CEA (29.4%), pT4 tumor stage (17.6%), < 12 LNs (15.7%), and LVI (11.1%). The presence of PNI was the least common risk factor (8.0%).

With a median follow-up of 50.1 months, the low-risk group tended to have a better 3-year DFS than the high-risk group (96.4% vs 89.4%; P = 0.056; Fig. 1). On univariate analysis, pT4 tumor stage (HR 2.93; 95% CI 1.23–6.98; P = 0.015) and elevated preoperative CEA (HR 3.08; 95% CI 1.33–7.12; P = 0.009) were associated with poor DFS (Table 1). In the multivariate Cox regression model, pT4 tumor stage (HR 2.58; 95% CI 1.06–6.25; P = 0.037) and elevated preoperative CEA (HR 2.93; 95% CI 1.26–6.82; P = 0.013) still were the independent prognostic factors of DFS, respectively. The 3-year DFS was 84.4% (95% CI, 74.7%-90.1%) in the patients with high preoperative CEA compared with 94.6% (95% CI, 90.3%-97.0%) in the patients with normal preoperative CEA. Among the 77 patients with elevated preoperative CEA, 65 (84.4%) had postoperative CEA data available; 41 (63%) of these patients had normalized postoperative CEA levels and 24 (37%) had elevated postoperative CEA levels. The overall 3-year DFS was 87.8% (95% CI, 74.4%-94.7%) in the 41 patients with normalized postoperative CEA, which was not statistically distinguishable from the 94.6% (95% CI, 90.3%-97.0%) in the patients who had normal preoperative CEA levels (HR 2.37; 95% CI, 0.81–6.94; P = 0.115). However, the 3-year DFS was significantly lower in the elevated postoperative CEA group than in the normal preoperative CEA groups (79.2% vs 94.6%; HR 4.25; 95%CI, 1.44–12.43; P = 0.008) (Fig. 2).

In total, 203 (77.5%) patients underwent surgery alone and 59 (22.5%) patients received surgery and oxaliplatin-based adjuvant chemotherapy. There were 49 (27.4%) patients and 10 (12.0%) patients who received treatment in the high-risk and low-risk groups, respectively. Patients were more likely to receive oxaliplatin-based adjuvant chemotherapy because of high-risk factors (P = 0.006), such as pT4 tumor stage (40.7% vs 10.8%; P < 0.001), PNI (22.0% vs 3.9%; P < 0.001) and LVI (20.3% vs 8.4%; P < 0.01). Additional characteristics were described in Fig. S2.

The 3-year DFS was 92.6% and 88.1% in the surgery alone and adjuvant chemotherapy group (HR 1.64; 95% CI 0.67–4.02; P = 0.280; Fig. 3a), respectively. No significant treatment benefit was observed in stage II CCs patients with dMMR in either high-risk group (HR 1.57; 95% CI, 0.62 to 3.99; P = 0.343; Fig. 3b) or low-risk group (HR 0.41; 95% CI, 0.00 to 15.25; P = 0.681; Fig. 3c). Given that ESMO guideline suggesting pT4 tumor stage and < 12 LNs were major high-risk factors in stage II CCs, so subgroup analyses were conducted. The two subgroups showed similar demographics and clinicopathologic characteristics (Tables 2 and 3). However, no potential benefit from oxaliplatin-based chemotherapy was observed in either subgroup (Fig. S3).