Following improvements in therapy over recent decades, some advanced cancers can be treated as chronic, rather than acute, conditions: that is, although cures are rare, patients will often live with their disease for many years. In these cases, maintaining a patient's quality of life becomes particularly important. These patients often spend years on almost continuous therapy, with their quality of life diminished by the inevitable toxicities associated with that therapy. There is therefore considerable interest in whether introducing breaks in courses of chemotherapy can relieve side effects without decreasing progression-free or overall survival.

The large Phase III COIN trial was set up in the UK by the Medical Research Council (MRC) to test the effect of both intermittent treatment and the use of an anti-EGFR antibody compared to continuous standard chemotherapy in patients with advanced colorectal cancer. This second of two companion papers¹ compares patients stratified to receive intermittent courses of one of two standard chemotherapy combinations (study arm C) to those randomised to continuous treatment with the same chemotherapy drugs (study arm A); the first paper, comparing arm A to the same treatment with the addition of cetuximab², has been summarised earlier.

The complete protocol for the COIN trial is available on the MRC clinical trials website³. A total of 2445 patients with advanced colorectal cancer who had not received prior chemotherapy were assigned at random, without blinding, to the three equal arms. Patients on the control arm A received a combination of oxaliplatin with either capecitabine or fluorouracil, with treatment continuing until disease progression or unacceptable toxicity or until the patient chose to stop. Patients on the intermittent therapy arm C firstly received a twelve-week course of the same combination of drugs. Treatment was then stopped and the patients monitored regularly; a second 12-week period of treatment was commenced when disease progression was observed, and this cycle of on-treatment and off-treatment periods repeated until a patient chose to stop. In both arms, the choice of the second drug was made according to local policy or patient preference.

815 patients were assigned to each of the treatment arms, with overall survival taken as the primary endpoint. Both intention to treat analysis of all randomized patients and per-protocol analysis of only those patients who remained on the assigned protocol after the first 12 weeks' treatment were carried out. Time to strategy-failure-free survival was recorded as a secondary endpoint: in arm A this was defined as traditional progression-free survival, while in arm C it was defined as the time to progressive disease either during a treatment period or within eight weeks of starting a treatment break. Quality of life data was recorded at the baseline and after 12 and 24 weeks.

The median overall survival of all randomised patients was 15.8 months in the continuous and 14.4 months in the intermittent treatment arm; overall survival was slightly higher when only patients remaining on protocol at 12 weeks were taken into account. These results did not indicate that intermittent chemotherapy had been proven to be non-inferior to continuous therapy in terms of overall survival in this patient population. Subgroup analysis showed that patients with raised platelet counts did particularly badly under the intermittent protocol; the hazard ratio for intermittent versus continuous treatment was 1.54 (CI 1.17-2.03) for these patients, compared to 0.96 (CI 0.80-1.15) for those with normal platelet counts. As expected, many side effects including haematological toxicities, peripheral neuropathy and hand-foot syndrome occurred more often in patients on arm A, whereas nausea and vomiting were more frequent in those on the intermittent arm.

The investigators concluded that the trial had not proved that intermittent chemotherapy was generally non-inferior to continuous chemotherapy as a first-line treatment for advanced colorectal cancer. However, treatment breaks can benefit patients with normal platelet levels, who are likely to experience fewer side effects and improved quality of life. Patients with raised platelet levels seem to do significantly worse on intermittent treatment and should not be offered such breaks.

References

1. Adams, R.A., Meade, A.M., Seymour, M.T. and 17 others, on behalf of the MRC COIN Trial Investigators (2011) Intermittent versus continuous oxaliplatin and fluoropyrimidine combination chemotherapy for first-line treatment of advanced colorectal cancer: results of the randomised phase 3 MRC COIN trial, The Lancet Oncology, published online ahead of print 4 June 2011

2. Maughan, T.S., Adams, R.A., Smith, C.G. and 18 others, on behalf of the MRC COIN Trial Investigators (2011) Addition of cetuximab to oxaliplatin-based first-line combination chemotherapy for treatment of advanced colorectal cancer: results of the randomised phase 3 MRC COIN trial, The Lancet, published online ahead of print 4 June 2011

3. http://www.ctu.mrc.ac.uk/plugins/StudyDisplay/protocols/COIN%20Protocol%20v5.1%20October%202009.pdf