PFS Valid OS Surrogate Marker In Advanced Pancreatic Cancer

Progression-free survival correlates strongly with overall survival, both at the trial and treatment arm level, in frontline chemotherapy trials in advanced pancreatic cancer

medwireNews: A literature-based analysis suggests that progression-free survival (PFS) can serve as a suitable surrogate for overall survival (OS) in trials evaluating first-line chemotherapeutic options in patients with advanced pancreatic cancer.

Researcher Hiroyuki Isayama, from The University of Tokyo in Japan, and team believe that using an intermediate endpoint such as PFS could reflect treatment response in an unbiased way, as, unlike OS, it would not be confounded by the effects of subsequent lines of therapy or competing risk factors.

Moreover, as assessing OS requires prolonged follow-up, “the use of PFS as the primary end-point in clinical trials could facilitate the early introduction of new effective chemotherapy regimens into clinical practice”, they write in the European Journal of Cancer.

The analysis included 50 phase II and III randomised controlled trials of first-line chemotherapy, including biological agents, for advanced pancreatic cancer and comprised a total of 15,906 patients.

Of the potential surrogate endpoints assessed in the study – such as PFS, response rate and disease control rate (DCR) – PFS had the strongest correlation with OS, with a Spearman rank correlation coefficient (rs) of 0.76. By contrast, DCR only correlated moderately with OS (rs=0.50) and the correlation of response rate with OS was even lower (rs=0.39).

Applying weighted linear regression models to trials in which hazard ratios for PFS and OS for an experimental versus a control arm were available, the researchers found that the between-arm difference in PFS was the greatest determinant of the variation in OS, such that 84% of the variation in OS would be explained by PFS.

And again, although the regression models for DCR and response rate were significant, only 65% and 30% of the OS variation would be explained by these surrogate endpoints, respectively.
Thus, the study confirms the superiority of PFS over other intermediate endpoints as a surrogate for OS in this patient population and setting, observe Hiroyuki Isayama et al.

The “robust” correlation between PFS and OS was maintained across various subgroups – for instance, when comparing trials that did versus did not include at least one arm with a biological agent, which they find reassuring as biological agents often result in cytostatic rather than conventional cytotoxic effects that may not be reflected by changes in tumour size.

However, in order for PFS to serve as a suitable OS surrogate, the team stresses the importance of not only standardising response criteria, as PFS is dependent on the definition of tumour progression, but also of determining which response criteria (eg, RECIST or WHO) are the most appropriate in this setting.


Hamada T, Nakai Y, Isayama H, et al. Progression-free survival as a surrogate for overall survival in first-line chemotherapy for advanced pancreatic cancer. Eur J Cancer 2016; 65: 11–20. doi: 10.1016/j.ejca.2016.05.016

medwireNews ( is an independent medical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2016