Résumé
Abstract only 3063 Background: An important limitation of the combination of target therapies (TT) to radiation (RT) is its potentially high level of toxicity. The objective of this study was to describe and quantify the drug related acute adverse events presented by patients taking part in phase I and/or phase II trials on TT combined to curative RT for the treatment of glioblastoma multiforme (GBM). Methods: A meta-analysis of summary data including all phase I, I/II and II trials published between 2000 and 2011 on newly diagnosed GBM patients treated by TT with RT was performed. Pooled incidence rates (IR) of all and specific toxicities were estimated using a generalized linear mixed model with fixed and random study effects. The pooled median of progression-free (PFS) and overall survivals (OS) were also considered. These results were compared to those obtained by patients receiving standard treatment (ST). Results: Twenty-one trials (8 phases I, 3 phases I/II and 10 phases II), testing 9 different drugs, have been selected, including 915 patients. The median follow-up varied from 1.4 to 34.0 months. The estimated pooled IR of all toxicities was 68.6 [53.5-88.0] per 1000 person-months of follow-up, compared to 20.7 [17.6-24.1] on ST (p<0.001). The pooled IRs of thrombocytopenia, neutropenia or leucopenia, fatigue, gastrointestinal events and treatment-related deaths are 11.2 [7.5-16.9], 9.0 [6.1-13.4], 6.4 [4.1-10.0], 5.5 [2.9-10.2] and 3.2 [1.6-6.4], compared to 4.1 [2.8-5.8], 4.9 [3.5-6.6], 4.6 [3.2-6.4], 0.8 [0.3-1.6] and 0.1 [0.0-0.6], respectively, on ST. Median survival times are 7.7 [6.1-9.8] and 13.8 months [11.9-16.1] on TT studies, and 6.9 [5.8-8.2] and 14.6 [13.2-16.8], on ST, for PFS and OS, respectively, without statistically significant difference (p=0.39 and 0.55). The heterogeneity among specific toxicities and survival endpoints may be partially explained by the study design and/or the intake or not of temozolomide. Conclusions: The use of TT combined to RT represented a significative increase in severe adverse events in patients with GBM compared to ST, whereas no significant difference was observed in survival endpoints. Grants from French Cancer League.