Julien Taieb

Georges Pompidou European Hospital, Sorbonne Paris Cité, Paris Descartes University, Paris, France;
ESMO 2018 Scientific Committee Member.

Is 3 months of adjuvant therapy
sufficient for stage III colon cancer?

FEATURE

The result of a tremendous collaborative effort between medical oncologists across Europe, the United States and Asia, the largest ever randomised controlled trial has been performed in stage III colon cancer.

The focus of a Special Symposium at the 2017 ESMO Congress co-chaired by Andrés Cervantes, Chair of the ESMO Educational Committee, and Alberto Sobrero, ESMO 2017 Congress Co-chair, the study was conducted to determine whether a 3-month regimen of adjuvant chemotherapy would be non-inferior to standard 6-month treatment. In order to recruit sufficient patient numbers to demonstrate non-inferiority of the shorter regimen, the International Duration Evaluation of Adjuvant chemotherapy (IDEA) collaboration comprised six randomised phase III trials conducted across 12 countries, with almost 13,000 patients included in the pre-planned pooled analysis. The six trials were: CALGB/SWOG (Cancer and Leukemia Group B/Southwest Oncology Group) 80702, IDEA France, SCOT (Short Course Oncology Treatment), ACHIEVE (Adjuvant Chemotherapy for Colon Cancer with High Evidence), TOSCA (Three or Six Colon Adjuvant) and HORG (Hellenic Oncology Research Group). Julien Taieb led the French part of the trial, together with Thierry André (St. Antoine Hospital, Paris), and describes his experience of this ground-breaking study.


For me, this was a hugely complex, yet extremely interesting and exciting 10-year collaborative academic project. Alberto Sobrero (Ospedale San Martino, Genova, Italy) provided the initial inspiration for the trial around 15 years ago. He was interested in exploring whether the duration of adjuvant therapy for colon cancer could be shortened to reduce chemotherapy-induced toxicity; the current 6-month regimen of oxaliplatin-based adjuvant chemotherapy is associated with cumulative dose-dependent neurotoxicity that can result in significant and long-term impairment in patient quality of life. Given that up to approximately 70% of patients can be cured of stage III disease, this is a major issue. If a shorter duration of adjuvant therapy could be given without compromising clinical outcomes, it should be possible to reduce the adverse effects of treatment, and this is what the IDEA study set out to investigate.


As the trial was the first of its kind, the statistical design was experimental. There was a lot of discussion about the 95% confidence interval (CI) margin that would be considered acceptable to demonstrate non-inferiority. The upper limit for the 95% CI was finally predefined as 1.12.


Initial results were presented during the 2017 annual congress of the American Society of Clinical Oncology and were recently published in the New England Journal of Medicine (Grothey A, et al. 2018). Surprisingly, given the impressively large patient population, the data were not clear-cut. From a statistical perspective, the trial was negative. Three-year disease-free survival (DFS), the primary endpoint, indicated that 6 months of oxaliplatin-based adjuvant chemotherapy conferred very little (<1%) benefit over 3 months of therapy, but the non-inferiority of 3 months versus 6 months of treatment was not proven as the upper limit of the two-sided 95% CI interval (1.15) exceeded the predefined upper limit for non-inferiority (1.12). In other words, after switching from 6 to 3 months’ therapy, the maximum possible relative risk increase was 15%, rather than the predefined 12%. As an added complexity, subgroup analyses demonstrated that outcomes were dependent on treatment regimen (CAPOX [XELOX] versus FOLFOX) and disease risk group. Notably, treatment duration appeared to affect the efficacy of FOLFOX, but not CAPOX; while non-inferiority was clearly demonstrated with 3 versus 6 months of CAPOX, the same was not true for FOLFOX, where the 3-month regimen was found to be inferior to 6 months. When analysed by risk group, 3 months of therapy was inferior to 6 months for high-risk disease (T4, N2 or both) but equivalent to 6 months for low-risk patients (T1–3, N1).


We were able to reach two clear conclusions based on robust, statistically and clinically relevant data: (i) for patients with high-risk stage III colon cancer (approximately 40% of the study population), 6 months of adjuvant FOLFOX is considered more effective than 3 months; and (ii) for patients with low-risk disease (approximately 60% of the study population), 3 months of CAPOX is non-inferior (or even slightly better) than 6 months of therapy.

“The non-inferiority of 3 versus 6 months of adjuvant CAPOX for patients with low-risk stage III colon cancer will change the treatment of this subset of patients.”

I am intrigued by the slightly improved efficacy of 3 versus 6 months of CAPOX. The reason for this is unclear. There was no increase in treatment-related mortality in patients treated for 6 months. As compliance with therapy was not monitored, it is possible that patients assigned to 6 months of CAPOX may have discontinued oral capecitabine early, reducing the difference in duration of therapy between groups. In addition, the variable outcomes observed between treatment regimens suggests that some bias may have been introduced as a result of the study design allowing treatment to be assigned on an individual patient basis rather than an individual centre basis. This might potentially have resulted in different CAPOX and FOLFOX patient populations.


The data are not conclusive for CAPOX in patients with high-risk disease and FOLFOX in low-risk disease, and treatment approach will likely vary between countries; in France, for example, 3 months of CAPOX is preferred for all low-risk patients. Where CAPOX is not indicated, FOLFOX may be administered for 3 months, but the FOLFOX 6-month regimen will remain standard for all high-risk patients.


Based on our experience in France, the advantages of a 3- versus 6-month adjuvant therapy regimen are becoming apparent; at 4 years of follow-up, the incidence of chronic peripheral neuropathy has declined from approximately 8% to 2%. As well as the accompanying improvements in quality of life, the shorter regimen has the added benefit of greater convenience for patients and is also economically advantageous; in France, savings of €50 million per year are estimated based on 3 versus 6 months of adjuvant treatment for low-risk stage III colon cancer. Therefore, 3 months of treatment is good news for both patients and payers.


In view of the complexity of the data, there has been further expert collaboration to assimilate the study findings; in addition to the Special Symposium at the 2017 ESMO Congress, an accompanying paper was subsequently published to consider how the results of the IDEA pooled analysis might impact clinical practice and treatment guidelines for early-stage colon cancer. Many other papers are yet to be published outlining country-specific experience and recommendations based on this extensive data set. There is still much to be said about the findings of this hugely important collaborative project.

Annals of Oncology: The hard road to data interpretation: 3 or 6 months of adjuvant chemotherapy for patients with stage III colon cancer?