P-148 Recurrence free survival as a putative surrogate for overall survival in phase III Trials of Curative – Intent Treatment of Colorectal Liver Metastases: Systematic Review

Abstract

Introduction: Analysis of overall survival (OS) in phase III trials with low progressive malignancies is an issue hardly addressed since it demands large samples, great cost and longer follow-up. Gains in recurrence free survival (RFS) for initially resectable colorectal liver metastases (CRLM) have been demonstrated by phase III trials, but this has not been associated with improvements in OS in phase III trials, although, already suggested in meta-analyses. The objective of this systematic review was to verify whether RFS surrogates OS in phase III trials for resectable CRLM. Methods: Medline, Embase, and Scopus databases were searched using the mesh terms: («colorectal liver metastases» or «colorectal liver metastasis») and (surgery or surgical or chemotherapy or «drug therapy» or «Antineoplastic Agents»[Mesh]) and (Clinical Trial or Comparative Study or Randomized Controlled Trial) from January 1990 to February 2015. Eligible studies were phase III trials testing any type of systemic therapy (neoadjuvant, adjuvant or perioperative) added to surgery in patients with resectable CRLM. A linear regression model based on hazard ratios (HR) of OS and RFS was run to estimate an association and a linear function between both outcomes.We considered the overall (unadjusted) HR for analysis. Our null hypothesis was that there was no association between RFS and OS. A two-sided p value < 0.05 was considered significant. Results: Of 3,059 studies, N = 5 phase III trials (N = 1,162 patients) looking for resectable CRLM and curative-intent treatment with surgery and systemic therapy were eligible and included for analyses. A linear regression weighted by each trial was used to estimate the association between each HR and RFS. The originated formula was: OS HR= (0.93 × RFS HR) + 0.14; with RFS CI 95% (0.48 - 1.38), with p = 0.007. Comparisons of original and assumed HR for OS based on this formula are, respectively: Langer (0.77 and 0.87); Portier (0.73 and 0.75); Nordlinger (0.87 and 0.87); Ychou (1.09 and 0.97); and Pimrose (1.49 and 1.52). A graph showing the linear association of RFS and OS is depicted in a figure. Conclusion: This study suggests a linear prediction of OS based on RFS for trials of patients with CRLM who were managed by curative-intent surgery and perioperative cancer-directed therapies. This association suggests that RFS could work as a putative surrogate of OS in this population avoiding bigger, longer or more resource-consuming trials. The OS could be assumed based on RFS and our model could be useful to better estimate sample size calculations of phase III trials of CRLM aiming for OS.