An analysis of survival data from a clinical trial is presented. The trial was conducted to evaluate new chemotherapies for colorectal cancer. The centres participating in the trial are modelled via random effects, to allow for heterogeneous patient responses across centres. Patient response is measured in terms of both progression and survival times, facillitating finer inferences about the effects of treatments and centres on the response. In particular, the data suggest that patient responses are homogeneous before disease progression, but heterogeneous after progression. For one of the new therapies, some increase in efficacy is detected before, but not after, progression.