PFS Deserves More Than Surrogate Status as a Clinical Trial End Point

Oncology Live®Vol. 22/No. 01
Volume 22
Issue 01

In Partnership With:

Although recent benefits do not pertain to all cancers and “cure” remains a relatively uncommon event, oncologists have an increasing number of molecularly targeted and immunotherapeutic strategies to employ based on objectively meaningful clinical trial outcomes.

Maurie Markman, MD

During this past decade, we have witnessed substantial achievements in the treatment of patients with advanced malignant disease. Although these benefits do not pertain to all cancers and “cure” remains a relatively uncommon event, oncologists have an increasing number of molecularly targeted and immunotherapeutic strategies to employ based on objectively meaningful clinical trial outcomes.

Unfortunately, despite the acceleration in regulatory approval of effective strategies, we continue to see reports in the peer-reviewed literature lamenting the lack of robust overall survival (OS) benefits documented in phase 3 randomized trials. For example, a recent report examined FDA approval of anticancer therapeutics between 2000 and 2016 and concluded that despite the fact that these agents caused “substantial tumor responses,” they were “associated with...prolonging median overall survival by only 2.4 months.”1

Even in settings in which progression-free survival (PFS) has been accepted as a valid end point for regulatory approval, such as antineoplastic therapy for advanced ovarian cancer, some continue to inquire whether this outcome can serve as a reliable surrogate for OS.2 This discussion fails to acknowledge the increasing relevance of the chronic disease nature of many cancers and the potentially substantial impact of multiple therapeutic options once an individual patient ceases treatment on a given clinical trial.

Several examples emphasize this critical point. Findings from 2 second-line ovarian cancer chemotherapy trials conducted for regulatory purposes and initiated 8 years apart (1999 and 2007) had the same eligibility (and ineligibility) criteria and included an identical carboplatin/gemcitabine chemotherapy treatment arm.3,4 Although cross-trial comparisons must always be conducted with caution, it is notable that these studies were carefully monitored (as required by regulatory agencies) and resulted in essentially the same median PFS outcomes (median 8.6 and 8.4 months, respectively).3,4 Although similar outcomes would have been anticipated for similar or identical treatment regimens, OS findings observed for these 2 study populations were strikingly different. For the carboplatin/gemcitabine trial regimen initiated in 1999, the median OS was 18.0 months,3 whereas it was almost twice as long—35.2 months—for the same regimen employed in a study started 8 years later.4

Although a definitive reason for why survival was so much longer in the later study is not possible to discern, the most likely is the availability of an increasing number of potentially beneficial therapeutic agents employed either on or off label following removal of research participants from the phase 3 trial.

A second excellent example of the increasing utility of later lines of therapy and their impact on an OS end point in a phase 3 randomized trial is provided by data involving therapy for metastatic melanoma. In an analysis of 9 randomized trials that employed a dacarbazine control arm with no crossover to the experimental regimen, there was a very strong correlation between PFS and OS (correlation coefficient, 0.96). However, when trials were included that permitted crossover from dacarbazine to the effective experimental regimen, the correlation coefficient between PFS and OS decreased to 0.55.5

And in a more recent report of 5-year outcomes observed in patients with wild-type BRAF advanced melanoma treated with nivolumab (Opdivo) versus a control arm of dacarbazine, investigators found strikingly improved 5-year PFS for patients treated with the checkpoint inhibitor (28% vs 3%). This translated into improved 5-year OS for the nivolumab-treated population (39% vs 17%). The overall response rate was, not surprisingly, also superior in the nivolumab study arm (42% vs 14%).6

However, the 5-year OS rate of patients initially treated with dacarbazine in this study but who subsequently received additional therapy, including nivolumab, was 38%, essentially no different from that of the group who received the checkpoint inhibitor as primary treatment. This outcome occurred despite the substantially inferior PFS observed with dacarbazine during the trial.

Of course, the appropriate conclusion here is not that nivolumab failed to improve OS despite its influence on time to disease progression, but rather that despite the objectively poor primary trial outcome (PFS) associated with the almost inactive chemotherapeutic (dacarbazine), effective therapy delivered after completion of the study regimen was able to have an impressively favorable influence on individual participants’ ultimate survival.


  1. Ladanie A, Schmitt AM, Speich B, et al. Clinical trial evidence supporting US Food and Drug Administration approval of novel cancer therapies between 2000 and 2016. JAMA Netw Open. 2020;3(11):e2024406. doi:10.1001/jamaetneworkopen.2020.24406
  2. Paoletti X, Lewsley LA, Daniele, et al; Gynecologic Cancer InterGroup (GCIG) Meta-analysis Committee. Assessment of progression-free survival as a surrogate end point of overall survival in first-line treatment of ovarian cancer: a systemic review and meta-analysis. JAMA Netw Open. 2020;3(1):e1918939. doi:10.1001/jamanetworkopen.2019.18939
  3. Pfisterer J, Plante M, Vergote I, et al; AGO-OVAR; NCIC CTG; EORTC GCG. Gemcitabine plus carboplatin compared with carboplatin in patients with platinum-sensitive recurrent ovarian cancer: an intergroup trial of the AGO-OVAR, the NCIC CTG, and the EORTC GCG. J Clin Oncol. 2006;24(29):4699-4707. doi:10.1200/JCO.2006.06.0913
  4. Aghajanian C, Blank SV, Goff BA, et al. OCEANS: a randomized, double-blind, placebo-controlled phase III trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer. J Clin Oncol. 2012;30(17):2039-2045. doi:10.1200/JCO.2012.42.0505
  5. Flaherty KT, Henning M, Lee SJ, et al. Surrogate endpoints for overall survival in metastatic melanoma: a meta-analysis of randomized controlled trials. Lancet Oncol. 2014;15(3):297-304. doi:10.1016/ S1470-2045(14)70007-5
  6. Robert C, Long GV, Brady B, et al. Five-year outcomes with nivolumab in patients with wild-type BRAF advanced melanoma. J Clin Oncol. 2020;38(33):3937-3946. doi:10.1200/JCO.20.00995
Related Videos
Paul D. Nathan, MBBS, PhD, FRCP
Paul D. Nathan, MBBS, PhD, FRCP
Jeffrey S. Weber, MD, PhD
In this fifth episode of OncChats: Assessing the Promise of AI in Oncology, Toufic A. Kachaamy, MD, and Douglas Flora, MD, LSSBB, FACCC, discuss the need for evidence to support the utilization of different artificial intelligence tools in healthcare.
Maurie Markman, MD
Omid Hamid, MD
Elizabeth Buchbinder, MD