With the increasing availability of active agents, the importance of postprogression survival (PPS) in clinical trials has been recognized for several malignancies. However, little is known about PPS in advanced gastric cancer (AGC).Methods
A literature search identified 43 randomized trials in chemotherapy-naive patients with AGC. We partitioned overall survival (OS) into progression-free survival (PFS) and PPS, and then examined the correlation between median OS and either median PFS or median PPS. The correlation between differences in OS (ΔOS) and those in PFS (ΔPFS) between trial arms was also investigated. We examined whether any association might be affected by year of completion of trial enrollment.Results
The average median OS was significantly longer in recent (2006 and later) trials than in older (2005 and earlier) trials (10.60 versus 8.64 months, P < 0.001), as was the average median PPS (5.34 versus 3.74 months, P = 0.001). The proportion of patients who received second-line chemotherapy was highly associated with median PPS (r2 = 0.657). Median PPS was highly correlated with median OS for all trials (r = 0.732), and this correlation was more pronounced in recent trials (r = 0.850). In contrast, the correlation between median PFS and median OS was less pronounced in recent trials (r = 0.282) than in older trials (r = 0.689), as was that between ΔPFS and ΔOS.Conclusion
PPS is highly correlated with OS for first-line chemotherapy in AGC, especially in recent trials, likely because of the incremental use of second-line chemotherapy. Currently, OS is accepted as the gold standard for efficacy evaluation in phase III trials for AGC. However, as PPS increases, OS can become skewed, and a statistically significant benefit in terms of PFS will likely become masked with OS as the end point.