Relative Efficacy of Granulocyte-Macrophage Colony-Stimulating Factor, Dacarbazine, and Glycoprotein 100 in Metastatic Melanoma: An Indirect Treatment Comparison

INTRODUCTION: Advances in the treatment of metastatic melanoma have been achieved in recent years: immunotherapies and targeted therapies have demonstrated survival benefits over older agents such as granulocyte-macrophage colony-stimulating factor (GM-CSF), dacarbazine, and glycoprotein peptide vaccine (gp100) in pivotal phase 3 trials. It is important to compare therapies to guide the treatment decision-making process, and establishing the relationship between older agents can strengthen the networks of evidence for newer therapies. We report the outcome of an indirect comparison of GM-CSF, dacarbazine, and gp100 in metastatic melanoma through meta-analysis of absolute treatment effect. METHODS: A systematic literature review identified trials for inclusion in the meta-analysis. A valid network meta-analysis was not feasible: treatment-specific meta-analysis was conducted. A published algorithm was used to adjust overall survival estimates from trials of GM-CSF, dacarbazine, and gp100 for heterogeneity in baseline prognostic factors. Survival estimates were compared in three patient groups: stage IIIB–IV M1c, stage IIIB–IV M1a, and stage IV M1b/c. RESULTS: One trial of GM-CSF, four of dacarbazine, and one of gp100 were included in the analysis. After adjusting for differences in baseline prognostic factors, median overall survival (OS) in all patient groups was longer for those receiving GM-CSF than for those receiving dacarbazine or gp100. The observed survival over time for GM-CSF was similar to the adjusted survival for dacarbazine and greater than for gp100 in all patient groups. CONCLUSION: The relative treatment effect of GM-CSF, dacarbazine, and gp100 has been reliably estimated by adjusting for differences in baseline prognostic factors. Results suggest that OS with GM-CSF is at least as good as with dacarbazine and greater than with gp100. Given the role of these agents as controls in phase 3 trials of new immunotherapies and targeted agents, these results can be used to contextualize the efficacy of newer therapies. FUNDING: Amgen Inc.