Phase 3 Randomized Trials: Expanding Clinically Meaningful Analysis Beyond the Stated Goals

The clinical oncology community points with justified pride to its commitment to the basic principles of “evidence-based” medicine in its efforts to optimize the health and welfare of all members of society. Although care delivery will always be substantially influenced by both unique clinical factors and individual patient choice, the foundation of physician recommendations and medical/regulatory policy and payment will be guided by the outcomes of published clinical research.

One could not imagine an oncologist today recommending a systemic antineoplastic approach without first carefully considering the available trial data regarding recognized evidence-based options in a particular setting. Of course, it is important to acknowledge that the decision-making process may also include the voluntary or increasingly mandated (by payers and health care organizations) use of clinical guidelines.

The rapidly evolving arena of precision cancer medicine and the establishment of valid molecular targets have created meaningful patient subgroups too limited in size to reasonably require novel therapeutic strategies to be evaluated for efficacy in a phase 3 randomized trial. However, such studies continue to be considered the gold standard in providing the most comprehensive data for determining the clinical utility (potential benefits vs toxicity) of any new approach to disease management in the oncology sphere.

The specific intent of this commentary is not to focus on the established role or even the limitations of phase 3 randomized trials, but rather to acknowledge and highlight the potential for clinically meaningful interpretations of these major clinical efforts to extend well beyond the stated study objectives. The 2 examples supporting such considerations provided below are from the realm of gynecologic cancer research because of the clinical and research interests of this commentator, rather than any claim for the uniqueness of this group of malignancies.

First, consider the provocative, almost 30-year-old experience associated with the paradigm-changing introduction of paclitaxel as a component (with platinum) of primary chemotherapy for patients with advanced ovarian cancer.^1,2 The initial trial conducted by the Gynecologic Oncology Group (GOG), comparing the investigative regimen of cisplatin plus paclitaxel with the standard of care at the time (cisplatin plus cyclophosphamide) revealed a stunning 14-month improvement in overall survival (OS; median 38 months vs 24 months) with the paclitaxel-containing regimen.¹

Far less frequently discussed are the results of the primary chemotherapy phase 3 trial that immediately followed, conducted by the same group of investigators with essentially the same study design. The study revealed a remarkably different survival outcome with cisplatin plus paclitaxel (median survival, 26.6 months) for women with a similar extent of disease as in the earlier research effort.²

This almost 1-year decrease in median OS is both striking and possibly disturbing until one appreciates that although the patient populations may have been alike in the measurement of the amount of cancer, they were probably substantially different in other clinically relevant features.

When the initial phase 3 ovarian cancer trial was initiated, there was great concern in the oncology community that paclitaxel could be associated with a major cardiac toxicity risk, resulting in the exclusion of individuals with known cardiac disease from participation and the likely discouragement of other patients with related risk factors. During this trial, the absence of serious cardiac adverse effects was reassuring and almost certainly resulted in a different attitude (by clinicians, prospective patients, and their families) regarding treatment risk when the subsequent study was initiated.

It is therefore reasonable to speculate this second trial included patients with potentially more serious comorbidities (eg, cardiac, pulmonary, renal, diabetic) which may have contributed to an overall shorter population-based survival outcome. It is relevant to note that this experience provides strong support for the acknowledged value of randomization when comparing survival outcomes of a novel strategy to that of a standard-of-care therapeutic option.

The second example is that of 2 phase 3 randomized ovarian cancer drug registration trials conducted more than 7 years apart, both of which included the identical chemotherapy regimen of carboplatin and gemcitabine—one as the experimental arm and the second as the control arm.^3,4

As just highlighted, when conducting cross-trial comparisons there is always the concern that the involved populations are different in clinical or other relevant factors. However, the circumstances here are somewhat unique and suggest the potential reasonableness of this effort. In the previously discussed frontline ovarian cancer trials where there was a striking observed difference in OS between the patient populations receiving identical chemotherapy, there was also a major difference in measured progression-free survival (PFS) between the first study compared with the second (median 18 months vs 14 months, respectively).^1,2 These data suggest a different direct impact of the therapy on disease outcome as measured by PFS, as well as the noted effect on OS.

By contrast, in the 2 second-line ovarian cancer carboplatin plus gemcitabine trials, even though separated by years in their conduct, the observed PFS for these treatment regimens was essentially identical (median 8.6 months in the earlier study vs 8.4 months in the later study). These data would suggest a very similar impact of the identical therapy on clinical outcomes.

Yet, when one examines OS for the 2 trial populations, very different results are observed. For the first study, median OS was 18 months vs 35.2 months for the later trial (an improvement of more than 1 year) despite the evidence this superior outcome did not result from any measured difference in efficacy of the specific chemotherapy regimen being analyzed.

What is the most likely explanation for this outcome? Although one can only provide a hypothesis in the absence of objective data, it is reasonable to speculate the improvement in survival resulted from a variety of beneficial therapeutic options after patients were removed from the study treatment in the second trial, which very possibly were not available in the earlier period.

If the hypothesis is correct, this is surely good news for patients with ovarian cancer, but also evidence that OS is a problematic end point for evaluating the clinical utility of a new anticancer strategy.

Maurie Markman, MD, is president of Medicine & Science at Cancer Treatment Centers of America, a part of City of Hope.

References

1. McGuire WP, Hoskins WJ, Brady MF, et al. Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med. 1996;334(1):1-6. doi:10.1056/ NEJM199601043340101
2. Muggia FM, Braly PS, Brady MF, et al. Phase III randomized study of cisplatin versus paclitaxel versus cisplatin and paclitaxel in patients with suboptimal stage III or IV ovarian cancer: a gynecologic oncology group study. J Clin Oncol. 2000;18(1):106-115. doi:10.1200/JCO.2000.18.1.106
3. Pfisterer J, Plante M, Vergote I, et al; AGOOVAR; 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