Correlation between progression free survival and response rate in patients with metastatic colorectal carcinoma.

BACKGROUND: All endpoint measures currently used to evaluate chemotherapeutic treatment benefit in clinical trials of patients with metastatic colorectal carcinoma (MCRC) have disadvantages. Overall survival (OS) is the most objective parameter but is flawed as an efficacy criterion, partly because potentially active second-line treatments are not controlled in many Phase III studies, and measured OS may be influenced (positively or negatively) to an unknown degree by second-line therapy. Measuring progression free survival (PFS) may be a means of isolating the real impact of first-line regimens.

METHODS: To test this hypothesis, the authors analyzed pooled data from all Phase III studies on first-line treatment in patients with MCRC reported from 1990 through 2000 for correlations between PFS, response rate (RR), and OS in MCRC patients that may suggest PFS as an improved indicator of the efficacy of first-line treatment. Spearman rho correlation coefficients (r) and regression equations were used.

RESULTS: The authors analyzed data from 29 studies involving nearly 13,500 patients. Significant correlations were found between PFS and RR (r = 0.655; P < 0.0001; a 10% RR increment corresponded to a 1-month increase in PFS), between RR and OS (r = 0.408; P = 0.0009; an 11.4% RR increment corresponded to a 1-month increase in OS), and between PFS and OS (r = 0.481; P < 0.0001; a 1-month increase in PFS corresponded to a 0.68-month increase in OS).

CONCLUSIONS: PFS is the most appropriate primary endpoint for MCRC studies, because it can express the antitumor activity of a first-line chemotherapy regimen regardless of measures used after disease progression. PFS deserves further evaluation as an endpoint measure.