Economic evaluation of temozolomide in the treatment of recurrent glioblastoma multiforme.

BACKGROUND Temozolomide (TMZ) is an oral alkylating agent with demonstrated efficacy as therapy for glioblastoma multiforme (GBM) and anaplastic astrocytoma. TMZ has widely replaced the procarbazine, lomustine plus vincristine (PCV) combination for the treatment of malignant brain tumours as a result of its oral administration and favourable toxicity profile. OBJECTIVES This study had three related aims. First, the cost effectiveness of TMZ (from the Finnish healthcare payer perspective) was compared with PCV in patients with GBM that had relapsed after primary treatment with surgery and radiotherapy. Second, the probability that TMZ is cost effective, compared with PCV, was estimated at different societal willingness-to-pay levels. Third, the value of new information for reducing the uncertainty related to the choice of treatment between TMZ and PCV was evaluated. METHODS The cost effectiveness of TMZ and PCV was evaluated using a decision-modelling approach. Incremental cost-effectiveness ratios (ICERs) for cost per gained life-month, progression-free life-month and QALY were calculated. Various information sources were used to acquire parameter values for the model. The efficacy information of both treatments was derived from the medical literature, quality-of-life (QOL) estimates were gathered from Finnish neuro-oncologists using visual analogue scale methods, and data on the use of healthcare resources were collected from hospital databases. The exact prices for resource use were gained from the list of Finnish health service unit costs (year 2001 prices). The model was analysed using second-order Monte Carlo simulation. The value of new information on reducing uncertainty was analysed using the expected value of perfect information (EVPI) approach. RESULTS According to the derived ICERs, 1 extra life-month gained with TMZ costs euro2367, 1 extra progression-free life-month costs euro2165, and 1 extra QALY costs euro32 471, compared with PCV, in the treatment of GBM. The probability of TMZ being the most cost-effective choice of treatment was >60% for all levels of willingness to pay >euro5000 per gained life-month. The respective probabilities were >75% for all levels of willingness to pay >euro10 000 per gained progression-free life-month and about 85% for all levels of willingness to pay >euro20 000 per gained quality-adjusted life-month. According to EVPI analysis, future research would potentially be cost effective if the costs of research were euro4.1 million (maximum). CONCLUSIONS On the basis of this Finnish analysis, TMZ has a high probability of being more cost effective than PCV for patients with GBM. The addition of QOL aspects to the prolonging of survival increases the probability further.