Independent validation of progression-free survival rate at 9 and 18 weeks (PFSR-18, PFSR-9) as predictor for overall survival (OS) in patients with malignant pleural mesothelioma (MPM): EORTC 08052 study.

Abstract

7584 Background: The increasing incidence and the plateau of current treatment outcomes in MPM necessitates development of new therapeutic approaches. Reliable and meaningful response assessment is difficult in phase II trials thus a primary endpoint based on progression free survival (PFS) rate at certain time point has been proposed (Greillier et al, 2011). EORTC 08052 was a phase II study in MPM where independent validation of PFSR-18 weeks was foreseen. We extend this to PFSR-9. Methods: Association of PFS and response with OS was assessed at two distinct time points; 9 and 18 weeks after registration. Landmarks method was used, PFS status (CR/PR/SD vs. PD) and response status (CR/PR vs. PD/SD) were determined at those time points. Cox regression and logrank test were performed and the corresponding c-indexes were calculated. Results: Of 82 registered patients, 28.4% achieved CR/PR and 77.8% had disease control (CR/PR/SD) as their best overall response. PFSR-18 and PFSR-9 were both strongly correlated with OS. Patients with no progression at 18 weeks had median OS of 16.9 months compared to 11.9 months in those who progressed at 18 weeks. Hazard ratio [HR] (95 confidence interval [CI]) was 0.46 (0.32-0.67), logrank test was 0.007 and C-index = 0.60. Adjusting for 3 important baseline prognosis factors, histology, performance status and disease stage resulted in PFS-18 as the only significant factor. When 9 weeks landmark was chosen, patients with no progression had median OS of 16.9 months vs. 6.8 months in those who progressed with HR (CI), logrank test and C-index 0.35(0.25-0.49), < 0.0001 and 0.66 respectively. When adjusted by 3 important baseline prognosis factors PFS-9 remained the only significant factor. The results also confirmed that response at 18 weeks and 9 weeks was correlated with OS. Conclusions: PFSR-18 was strongly correlated and discriminated patients with better OS from the poorer prognosis patients. An earlier endpoint, PFSR-9 was also strongly correlated to OS and had a better discriminating capacity. Previous results on correlation between PFSR-9 and PFSR-18 and OS were independently validated.