Studies of clinical efficacy commonly report more than one clinical endpoint. For example, randomized controlled trials of treatments for cancer will normally report time to disease progression as well as overall survival. It is likely that disease progression will be associated with higher mortality rates. Disease progression rates will also have consequences for the societal economic burden of the disease. Economic evaluation of the cost-effectiveness of different treatment regimes therefore requires the joint estimation of both disease progression and mortality. We describe a model to combine evidence from studies reporting time to event summaries for disease progression and/or mortality, motivated by a systematic review of 1st-line treatment for advanced breast cancer to provide inputs for an economic evaluation as part of the National Institute for Health and Clinical Excellence (NICE) clinical guideline on treatment of advanced breast cancer in England and Wales. The review identified a network of treatment comparisons, which provides the basis for indirect comparison. A variety of outcomes were reported: overall survival, time to progression (overall and responders only), and the proportion of responder, stable, progressive disease, and non-assessable patients. There were only five trials, and not all trials reported all outcomes. The scarcity of the available evidence required us to make strong assumptions in order to identify model parameters. However, this evidence structure often occurs in health technology assessment (HTA) of treatments for cancer. We discuss the validity of the assumptions made, and the potential to assess their validity in other applications of HTA of cancer therapies. Copyright © 2011 John Wiley & Sons, Ltd.