A population-based study on the effect of temozolomide in the treatment of glioblastoma multiforme

The effect of temozolomide (TMZ) and radiotherapy (RT) in the treatment of glioblastoma multiforme (GBM) has been well documented in randomized controlled trials. Here we present our findings on the effect of TMZ added to RT at a population level. The Cancer Registry of Norway was searched for patients with a GBM diagnosis from January 1, 2000 to December 31, 2007. Subsequently, the prescriptions registered to these patients were obtained from the Norwegian Prescription Database. The data were analyzed according to era (pre-TMZ introduction or post-TMZ introduction) and according to treatment received. Furthermore, a matching procedure was utilized to reduce the bias between the RT + TMZ and RT alone treatments so that the effect of TMZ could be better scrutinized. We identified 1157 GBM patients. The median overall survival (OS), in months, was 8.3 (95% confidence interval: 7.6-9.0) and 10.1 (95% confidence interval: 9.1-11.0) in the pre-TMZ and TMZ eras, respectively (P < .001). By treatment, we found median OS for the control, RT alone, and RT + TMZ groups to be 2.5, 9.0, and 16.2 months, respectively (P < .001). Two-year survival was 0%, 4%, and 25%, respectively. The effect of age on TMZ effect was insignificant. In the matched group analysis, TMZ provided a 7.6-month OS benefit. Our population data reproduce the beneficial effect of TMZ from randomized controlled trials with a median OS of 16.2 months and 25% 2-year survival.