For patients with unresectable metastatic colorectal cancer, maintenance treatment with capecitabine (Xeloda) and bevacizumab (Avastin) significantly delayed disease progression and improved overall survival in the phase III CAIRO3 study by the Dutch Colorectal Cancer Group. Miriam Koopman, MD, PhD, of University Medical Center Utrecht in The Netherlands, presented the findings at the 2013 ASCO Annual Meeting.1
“The optimal duration of chemotherapy and bevacizumab in metastatic colorectal cancer is not well established, and the value of chemotherapy-free intervals tested in several studies is still a matter of debate. What we do know is that drug holidays are preferred by many patients,” Dr. Koopman said.
“Given this, the CAIRO3 study was designed to investigate the efficacy of observation vs maintenance treatment with capecitabine plus bevacizumab after induction treatment with six cycles of capecitabine/oxaliplatin plus bevacizumab (CAPOX-B) in patients not progressing during induction,” she said.
The study included patients with previously untreated, unresectable metastatic colorectal cancer who had stable disease or better after six cycles of standard CAPOX-B. Patients were randomly assigned between observation or maintenance with capecitabine at 625 mg/m2 twice daily and bevacizumab at 7.5 mg/kg every 3 weeks.
Upon the first episode of disease progression, patients in both arms were treated with CAPOX-B until second progression, which was the study’s primary endpoint. For patients not able to receive CAPOX-B upon first progression, the second episode of disease progression was considered equal to first progression, per protocol.
Secondary endpoints were overall survival and time to second progression, which was defined as the time to progression or death on any treatment (including CAPOX-B) following the first episode of disease progression. All endpoints were calculated from the time of randomization, which occurred after induction therapy.
A total of 558 patients from 74 Dutch centers were randomized and followed for a median of 40 months.
Progression Delayed in Maintenance Arm
“Maintenance treatment with capecitabine plus bevacizumab after six cycles of CAPOX-B was feasible, and it significantly prolonged [time to first and second episodes of disease progression],” Dr. Koopman reported.
The median first episode of progression in the observation arm occurred at 4.1 months, compared to 8.5 months in the maintenance arm, for an adjusted hazard ratio (HR) of 0.41 (P < .001), Dr. Koopman reported.
In the observation arm, 76% were reintroduced to CAPOX-B. In the maintenance arm, 47% were reintroduced to CAPOX-B.
Patients were followed until their second disease progression, and the median time to second progression at that time was 10.5 months with observation and 11.8 months with maintenance, a 23% reduction in risk of second progression (adjusted HR = .77, P = .007), the primary endpoint.
The median time to second progression—ie, time from randomization to disease progression on any treatment given after first progression—was 15 months for observation and 19.8 for maintenance (adjusted HR = .63, P < .001).
Overall Survival Also Improved
Maintenance also afforded an overall survival benefit in CAIRO3. Although the difference was not statistically significant in the unadjusted analysis, it reached significance when adjusted for covariates, including disease stage, interval between diagnosis and randomization, and other factors.
In the adjusted analysis, median overall survival was 18.2 months with observation and 21.7 months with maintenance (adjusted HR = .80, P = .035). Dr. Koopman emphasized that this overall survival time does not include the 4 months of induction treatment.
Almost 50% of each arm received four effective agents during the course of their metastatic disease, and about 11% received five. “The percentage of patients receiving four to five drugs during their metastatic disease is comparable between the arms; therefore, time on treatment appears to be an additional relevant factor for overall survival in this study,” she suggested.
Grade 3/4 toxicity was manageable. Differences between the arms included hand-foot syndrome (0% with observation vs 22% with maintenance) and neurotoxicity (5% vs 10%, respectively).
Dr. Koopman acknowledged the need to weigh the modest, though statistically significant, benefit with the potential for lesser quality of life due to continued chemotherapy. Quality-of-life data from the study are currently being analyzed.
“The primary endpoint, [second progression-free survival rate], and also time to second progression [on any treatment given after first progression], were highly significantly different between the arms. I think the time where the patients were not on treatment in the observation arm accounts for these differences. For patients who have stable disease or better after six cycles of chemotherapy, I believe maintenance is the treatment to choose,” she said.
In conclusion, Dr. Koopman maintained, “Our data support the use of bevacizumab plus capecitabine until progression or unacceptable toxicity.… When any treatment after [first disease progression] was considered, maintenance therapy also significantly prolonged the time to second progression, and this is probably the endpoint that is most relevant for clinical practice.” ■
Disclosure: Dr. Koopman reported no potential conflicts of interest.
1. Koopman M, Simkens LH, Ten Tije AJ et al. Maintenance treatment with capecitabine and bevacizumab versus observation after induction treatment with chemotherapy and bevacizumab in metastatic colorectal cancer: The phase III CAIRO3 study of the Dutch Colorectal Cancer Group. 2013 ASCO Annual Meeting. Abstract 3502. Presented June 1, 2013.