The Critical Nature of Decision-Making in the Development of Phase 3 Clinical Trials

Maurie Markman, MD

Randomized trials are widely recognized to play a pivotal role in the approval process of new antineoplastic agents, the development of novel combination regimens, and the decision-making associated with the establishment of clinical guidelines. And few would challenge the past, current, and future utility of such studies in defining optimal cancer management.

Despite this high-level consensus conclusion, it is relevant to note the publication of numerous solid, data-based critiques that have challenged individual standard elements of randomized trials in oncology. These include the common exclusion of individuals 65 years and older and patients with relevant comorbid medical conditions, prior therapies, or a history of other cancers; the limited accrual of individuals from certain ethnic groups and those with lower socioeconomic status; the validity of primary end points (overall survival vs other measures of clinical benefit); and the increasing cost of conducting these studies^.1-4

This commentary highlights another aspect of the randomized trial process that may cloud subsequent interpretation and specifically relates to decisions made in the critical development of the studies themselves. The examples given during this discussion will be from the realm of gynecologic oncology due to the clinical background of this author, but it is certain the issues noted will be relevant to other areas of oncologic medicine.

To begin, consider past decisions by the Gynecologic Oncology Group (GOG) to employ a particular dose of a chemotherapy agent, such as cisplatin, as a single agent or as a component of a combination regimen in a clinical investigative effort. Justification for a recommendation to deliver the so-called maximally tolerated dose is understandable because, realistically, it would not be considered safe to subsequently administer a higher dose in routine clinical practice.

Thus, it was quite reasonably elected to administer cisplatin at a dose of 75 mg/m² (every 3 weeks for 6 cycles) in combination with paclitaxel in the paradigm-changing trial that compared this novel approach with a standard-of-care regimen (cisplatin plus cyclophosphamide) as primary chemotherapy for advanced ovarian cancer.⁵

It is interesting to note that a subsequent phase 3 trial conducted by the GOG that compared single-agent cisplatin delivered at 100 mg/m² (every 3 weeks for 6 cycles) with the lower cisplatin dose (75 mg/m²) combined with paclitaxel resulted in the same objective response rate, progression-free survival, and overall survival for both regimens.⁶ Because crossover from single-agent cisplatin to a third study arm (single-agent paclitaxel) was common in this trial, one might reasonably inquire how certain investigators and clinicians are certain of the ideal systemically administered dose of cisplatin and the optimal use of the agent when employed as a component of initial therapy for ovarian cancer.

Realistically, any existing questions related to systemic cisplatin are unlikely to be explored in ovarian cancer management in the future.

However, the issue of the optimal dose of intraperitoneal cisplatin employed in primary therapy for ovarian cancer based on clinical trial experience is, in the opinion of this commentator, far more problematic. Several randomized, phase 3 trials on small-volume residual stage III ovarian cancer, all of which employed intraperitoneal cisplatin at a dose of 100 mg/m² (vs an intravenous cisplatin-based regimen), revealed improvements in both progression-free survival and overall survival with the regional approach.⁷

In a more recently conducted GOG study, intraperitoneal cisplatin-based chemotherapy was compared with an all-systemic carboplatin-plus-paclitaxel approach and failed to reveal a difference in survival outcomes between the regional and systemic strategies.⁸ Unfortunately, in this study, the intraperitoneal cisplatin dose was reduced to 75 mg/ m²—despite the fact that the previous studies (as noted earlier) had employed a higher dose of this agent.

Although this dose reduction was perhaps a rational decision in regard to the issue of systemic toxicity, the failure to include—as a control arm—the regimen previously demonstrated in several phase 3, randomized trials to improve survival is problematic with respect to an objective and unbiased analysis of the study outcome.

Study interpretation is further clouded by the fact that the investigators elected to include bevacizumab (Avastin) in all arms of the study (systemic as well as regional treatment). Although one may legitimately wish to conclude the administration of bevacizumab enhanced the delivery of systemic cisplatin (and possibly paclitaxel) such that any advantage associated with the higher local intraperitoneal drug concentrations achieved with regional administration was negated, in the absence of a non-bevacizumab control arm, this remains little more than a provocative hypothesis requiring further evaluation.

Add to this the previously noted absence of a 100 mg/m² intraperitoneal cisplatin control arm, and one might reasonably conclude that the design of this study makes the results quite difficult to interpret.

Finally, for completeness, one additional point should be highlighted. Let us accept for a moment that based on the results of this trial, an oncologist elects to discard the results of 3 previously reported phase 3, randomized trials that revealed a survival advantage for intraperitoneal cisplatin when deliv- ered at a dose of 100 mg/m² every 3 weeks for 6 cycles as treatment for small-volume residual advanced ovarian cancer. Assuming one accepts the mantra that care should be based whenever possible on the results of phase 3, randomized trials, would it be most appropriate to assume the standard-of-care antineoplastic drug regimen
in the setting of small-volume residual ovarian cancer following surgical cytoreduction should be systemic carboplatin plus paclitaxel with bevacizumab?

If the answer to this question is yes, one might then inquire as to whether all patients with ovarian cancer today in this precise clinical situation (and in the absence of medical contraindications) are offered treatment with this 3-drug regimen? And if the answer is no, then perhaps the next question should be, why not?

Reference

1. Murthy VH, Krumholz HM, Gross GP. Participation in cancer clinical trials: race-, sex-, and age-based disparities. JAMA. 2004;291(22):2720-2726. doi:10.1001/ jama.291.22.2720

2. Freedman RA, Ruddy KJ. Who are the patients in our clinical trials for cancer? J Clin Oncol. 2019;37(18):1519-1523. doi:10.1200/JCO.19.00382

3. Perez M, Murphy CC, Pruitt SL, Rashdan S, Rahimi A, Gerber DE. Potential impact of revised NCI eligibility criteria guidance: prior malignancy exclusion in breast cancer clinical trials. J Natl Compr Canc Netw. 2022;20(7):792-799.e4. doi:10.6004/ jnccn.2022.7017

4. Abi Jaoude J, Kouzy R, Mainwaring W, et al. Performance status restriction in phase III cancer clinical trials. J Natl Compr Canc Netw. 2020;18(10):1322-1326. doi:10.6004/jnccn.2020.7578

5. 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

6. 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

7. Wright AA, Cronin A, Milne DE, et al. Use and effectiveness of intraperitoneal chemotherapy for treatment of ovarian cancer. J Clin Oncol. 2015;33(26):2841- 2847. doi:10.1200/JCO.2015.61.4776

8. Walker JL, Brady MF, Wenzel L, et al. Randomized trial of intravenous
versus intraperitoneal chemotherapy plus bevacizumab in advanced ovarian carcinoma: an NRG Oncology/Gynecologic Oncology Group study. J Clin Oncol. 2019;37(16):1380-1390. doi:10.1200/ JCO.18.01568