Shorter Adjuvant Oxaliplatin-Based Therapy Less Effective in High-Risk Stage II CRC

Tim J. Iveson, MD, FRCP

Three months of adjuvant oxaliplatin-based therapy in patients with high-risk, stage II colorectal cancer (CRC) was found to be inferior to 6 months of treatment, according to results from a prospective pooled analysis of 4 randomized trials that were presented at the 2019 ASCO Annual Meeting.

The 4 phase III trials from which the data were pooled were the TOSCA study from Italy; the SCOT study, which recruited patients from the United Kingdom, Denmark, Spain, Sweden, and Australia; the ACHIEVE-2 study from Japan; and the HORG study from Greece. All trials were a part of the International Duration Evaluation of Adjuvant (IDEA) Therapy Collaboration.

Investigators performed the prospective, preplanned pooled analysis from the 4 concurrently conducted trials in order to assess the noninferiority of 3 months of adjuvant FOLFOX/CAPOX versus 6 months. Disease-free survival (DFS) served as the primary endpoint of analysis, which included 3273 patients.

“All patients [were randomized to receive] either CAPOX or FOLFOX chemotherapy treatment,” explained lead author Tim J. Iveson, MD, FRCP. [Which] chemotherapy [to use] was at the choice of the clinician rather than [based on randomization]. Once we knew that patients were eligible and which chemotherapy [regimen] they would receive, they were randomized to receive either 3 months of treatment or 6 months.”

Of the 3273 patients, 1254 received FOLFOX, while 2019 were given CAPOX. At a median follow-up of 60.2 months, the 5-year DFS rate for 3 months of treatment was 80.7%, while the DFS was 84.0% with 6 months of treatment (HR, 1.18; 80% CI, 1.05-1.31; P for noninferiority = .404). The estimated hazard ratio in those who received the CAPOX regimen was 1.02 (80% CI, 0.88-1.17; P for noninferiority = .087), while it was 1.42 (80% CI, 1.19-1.70; P for noninferiority = .894) for those given the FOLFOX regimen.

Although noninferiority was not shown in the overall population with 3 months of treatment versus 6 months, the choice of the adjuvant regimen appears to be important, said Iveson. Investigators noted a small difference in 5-year DFS between 3 months and 6 months of treatment with CAPOX (81.7% vs 82.0%, respectively), although this was not determined to be statistically significant. The 5-year DFS observed with FOLFOX was 79.2% at 3 months versus 86.5% at 6 months.

In an interview with OncLive, Iveson, a medical oncologist at University Hospital Southampton NHS Foundation Trust, further discussed the results of the pooled analysis as well as how these data can be applied to practice.

OncLive: Could you provide an overview of the abstract?

Iveson: The IDEA collaboration has looked at the duration of adjuvant chemotherapy in CRC. At the 2019 ASCO Annual Meeting, I presented the results of the individual patient data from the 4 trials [conducted as a part of] the IDEA collaboration, in which we looked at this in patients with high-risk, stage II CRC.

If you look across the 4 studies, the patients were of similar age and the majority of them were male. All patients had an ECOG performance status of 0 or 1. The percentage of patients with T4 disease varied across studies, as did the percentage of patients with moderately differentiated tumors.

The results of this study showed that there were more adverse events (AEs) in patients who had received 6 months of treatment compared with 3 months of treatment; [this was true for all AEs], most notably diarrhea and peripheral neuropathy. With 6 months of treatment, there were more patients with peripheral neuropathy grade 2 or above; this was found to be clinically statistically significant. With 3 months of treatment, [the rate of this event] was 13% compared with 36% [in those who received] 6 months of treatment, so almost triple.

In terms of the results, the primary endpoint was DFS. Five-year DFS with 3 months of treatment was 80.7%, while DFS with 6 months of treatment was 83.9%, an absolute difference of 3.2%. The hazard ratio was 1.17 and the upper limit of the confidence interval was 1.31, which crossed the noninferiority margin of 1.2. Overall, this meant we were unable to show noninferiority for 3 months of treatment compared with 6 months of treatment.

What do these data mean for clinical practice?

The results of this analysis of high-risk, stage II CRC also has to be looked at with the IDEA analysis of stage III patients that was presented 2 years ago. What we have seen in both these analyses is that there is a clear effect of regimen. When you look at duration of treatment, the results for CAPOX are different from the results of FOLFOX. If we take the results together of both the high-risk patients with stage II and stage III disease, we can say that these data strongly suggest that giving 3 months of CAPOX chemotherapy is noninferior to giving 6 months of CAPOX chemotherapy. [In contrast], if you look at FOLFOX, 3 months of treatment is inferior to 6 months of treatment.

Additionally, we know there is more toxicity if you give 6 months of treatment. Therefore, when you have a patient in front of you, you have to consider which chemotherapy regimen that they would benefit from, the efficacy of that treatment, as well as the potential toxicity. Based on [those factors], you can make a recommendation both in terms of which chemotherapy regimen to use as well as the duration of that treatment.

What are some of the key remaining challenges in treatment of patients with high-risk, stage II disease?

With high-risk, stage II disease, the challenges are identifying which, if any, risk factors [are present]. To be eligible for the study, a patient had to have 1 or more of the following factors: T-4 disease, a poorly differentiated tumor, bowel obstruction, bowel perforation, or inadequate nodal harvest. We don't know if those factors are equal; they probably aren't. Also, we don't know if a combination of more than 1 of these factors gives [patients] a greater risk of recurrence. If so, [these patients may] benefit more from chemotherapy than patients with just 1 risk factor.

The reason these high-risk, stage II patients were included in the IDEA collaboration is that the MOSAIC study showed an improvement in DFS. Now, with longer follow-up, we know that overall survival in high-risk stage II disease is not improved by the addition of oxaliplatin. There is more work to do to actually define which patients within this group have a greater risk and which ones have a lesser risk.

Iveson T, Sobrero AF, Yoshino T, et al. Prospective pooled analysis of four randomized trials investigating duration of adjuvant (adj) oxaliplatin-based therapy (3 vs 6 months {m}) for patients (pts) with high-risk stage II colorectal cancer (CC). J Clin Oncol. 2019;37(15 suppl; abstr 3501). doi: 10.1200/JCO.2019.37.15_suppl.3501.