Original Article
Is overall survival still the primary endpoint in maintenance non-small cell lung cancer studies? An analysis of phase III randomised trials
Abstract
Background: In advanced non-small cell lung cancer (NSCLC), the increasing number of active compounds available in second line settings makes overall survival (OS) with maintenance treatment a not frequently observed endpoint. A literature review was conducted to examine whether post-progression survival (PPS) correlates with OS in maintenance trials.
Materials and methods: PPS was defined as median OS - median PFS. The authors used the coefficient of correlation (r) to assess the correlation between them. They also evaluated the surrogacy of PFS and OS.
Results: Ten trials with 11 arms were identified. Overall, a stronger correlation was observed between OS and PPS [Spearman rank correlation coefficient (r) =0.75] than OS and PFS time (r=0.37). The correlation of differences in median PFS (ΔPFS) and median OS (ΔOS) is 0.64 (P=0.0326). The slope of the regression line is 0.76, indicating that a maintenance therapy producing a one-month gain in PFS will yield an estimated three weeks prolongation in OS.
Conclusions: PPS is strongly associated with OS because of intensive post-study further lines of treatment. However, improvements in PFS are moderately associated with improvements in OS. Hence, in this setting, PFS is a possible surrogate for OS.
Materials and methods: PPS was defined as median OS - median PFS. The authors used the coefficient of correlation (r) to assess the correlation between them. They also evaluated the surrogacy of PFS and OS.
Results: Ten trials with 11 arms were identified. Overall, a stronger correlation was observed between OS and PPS [Spearman rank correlation coefficient (r) =0.75] than OS and PFS time (r=0.37). The correlation of differences in median PFS (ΔPFS) and median OS (ΔOS) is 0.64 (P=0.0326). The slope of the regression line is 0.76, indicating that a maintenance therapy producing a one-month gain in PFS will yield an estimated three weeks prolongation in OS.
Conclusions: PPS is strongly associated with OS because of intensive post-study further lines of treatment. However, improvements in PFS are moderately associated with improvements in OS. Hence, in this setting, PFS is a possible surrogate for OS.
