Duration of oxaliplatin-based adjuvant chemotherapy in patients with Stage III or high-risk Stage II resected colon cancer.

Abstract

The duration of the adjuvant oxaliplatin-based treatment has been a hot topic in the field of colorectal cancer for the last 2 years, since the first presentation of results of the international IDEA collaboration, including more than 12,000 patients.1 At the last ASCO Annual Meeting, a new chapter of this story was added, with the presentation of results reported in the high-risk Stage II group, where the noninferiority of 3 months versus 6 months of oxaliplatin-based adjuvant therapy was not proven2 (Table 1). These results challenge the reasonable concept that a shorter treatment could be enough in the presence of a lower risk of relapse, underlying the interpretation of results in Stage III according to the risk group. In fact, though in the absence of a significant p for interaction, a heterogeneous effect of treatment shortening between low- and high-risk groups was evidenced. In particular, in the low-risk group (T1–3 and N1), 3 months of capecitabine and oxaliplatin (CAPOX) was noninferior to 6 months, whereas the noninferiority of 3 months of 5-fluorouracil and oxaliplatin (FOLFOX) respect to 6 months was not proven. In the high-risk group (T4 and/or N2), the noninferiority of 3 months of CAPOX compared to 6 months was not proven, whereas 3 months of FOLFOX was inferior to 6 months (Table 1).1 Findings from the Stage II high-risk group confirm an interaction effect between the fluoropyrimidine backbone and the treatment duration, showing a detrimental effect of 3 versus 6 months of FOLFOX while noninferiority of 3 months of CAPOX was shown.2 The interaction between the fluoropyrimidine backbone and the treatment effect is difficult to explain, since a clear biological rationale is lacking, and in previous adjuvant trials capecitabine was equal to 5-fluorouracil,3 as well as in the metastatic setting where similar efficacy was reported between CAPOX and FOLFOX.4 In addition, since the administration of 5-fluorouracil or capecitabine was at investigators' choice, there was a wide heterogeneity among the different studies included in the pooled analysis, resulting in a potential selection bias. Indeed, the choice of 5-fluoruracil versus capecitabine was probably driven by clinical factors also weighting on patients' prognosis and by treating physicians' preferences and attitudes. Acknowledging this intrinsic limitation of each subgroup analysis, these findings should be regarded as pure hypothesis-generating. Moreover, several guidelines5 recommend performing both the intention-to-treat and the per-protocol analysis in noninferiority trials, where the intention-to-treat analysis may be biased by the inclusion of patients who did not sufficiently comply with the protocol, especially when treatment adherence significantly differs between the two arms. In the per-protocol population analysis of the IDEA trial1 among Stage III patients, noninferiority of 3 months of therapy compared to 6 months was not proven for any regimen (FOLFOX or CAPOX) or risk group (high- or low-risk). A per-protocol subgroup analysis combining regimen and risk group in Stage III and a per-protocol analysis in the overall population of Stage II patients was not shown. While waiting for overall survival data, translating these data into proper recommendations is not easy. Indeed, NCCN guidelines6 currently recommend 3–6 months of FOLFOX or 3 months of CAPOX as preferred options in low-risk Stage III patients and 3–6 months of CAPOX or 6 months of FOLFOX as preferred options in high-risk Stage III patients. More cautiously, ASCO guidelines7 suggest adjuvant chemotherapy for a duration of 6 months in high-risk patients (T4 and/or N2), and 3 or 6 months of adjuvant treatment for patients at a low risk (T1–3 and N1) involving the patient in the decision-making process. In a recent update, ESMO guidelines8, 9 recommend 3 months of CAPOX as preferred options in low-risk Stage III patients and 6 months of CAPOX or FOLFOX as preferred options in high-risk Stage III patients. In addition, 3–6 months of CAPOX or 6 months of FOLFOX are suggested as preferred options in high-risk Stage II patients when an oxaliplatin-based therapy is chosen. Based on these considerations, in our opinion, the initial planned standard duration of therapy should be 6 months both in Stage III and high-risk Stage II when clinicians opt for an adjuvant oxaliplatin-based treatment. Nevertheless, a larger flexibility should be taken into account because the vast majority of the benefit from the adjuvant treatment seems provided by the first 3 months of therapy while the dose-dependent risk of long-term neurotoxicity clearly increases with longer treatment durations. Therefore, it would be more reasonable to defer a decision on how long adjuvant chemotherapy should continue after the first 3 months of therapy. The decision should be based on each individual patient's attitude, tolerance, neurotoxicity and risk of relapse, and the option of continuing the fluoropyrimidine alone should also be considered, even if no specific data are available to this regard. Yours sincerely, Roberto Moretto Alfredo Falcone Chiara Cremolini All other authors have declared no conflicts of interest.