INTRODUCTION Treatment of malignant pleural mesothelioma (MPM) remains a clinical challenge. The aim of this study was to identify selection factorsfor allocation of MPM patients to multimodal therapy based on survival data from 12 years of experience. METHODS Eligible patients had MPM of all histological subtypes with clinical stage T1-3 N0-2 M0. Induction chemotherapy consisted of cisplatin/gemcitabine (cis/gem) orcisplatin/pemetrexed (cis/pem), followed by extrapleural pneumonectomy (EPP).Multivariate analysis was performed to assess independent prognosticators for overall survival (OS). A Multimodality Prognostic Scorewasdeveloped based on clinical variablesavailable before surgery. RESULTS From May 1999 to August 2011, 186 MPM patients were intended to be treated with induction chemotherapy followed by EPP. Hematologic toxicity was significantly less frequent after cis/pemcompared to cis/gem, but no difference in response or OS between the regimens.128 patients underwent EPP with a30-day mortalityof4.7%. 52% percent of the patients received adjuvant radiotherapy. The median OSof patients undergoing EPP was significantly longer with 22months (95%CI:20-24) as compared to 11 months (9-12) for patients treated without EPP.Aprognostic score was defined considering tumor volume,histology, CRP,andresponsetochemotherapythat identified patient groupsnot benefittingfrom multimodality treatmentwhich was confirmed in an independent cohort. CONCLUSION Patients receiving induction chemotherapy followed by EPP for MPM of all histological subtypes and irrespective of nodal statusshowed a median survival of 22months. A prognostic score is proposed to help patient allocation for surgery after validation in an independent cohort.