Control Arms Must Be Selected for Meaningful Comparison

OncologyLive, Vol. 20/No. 15, Volume 20, Issue 15

Although concerns have been raised in recent years regarding the need for randomized trials to augment the body of clinical understanding, one critical issue that has failed to generate sufficient discussion is how the choice of the control arm affects the interpretation of an individual study’s outcome and potentially undermines the ethical basis for that particular study.

Maurie Markman, MD

The modern practice of medical oncology is, perforce, based on solid evidence of the clinical utility of the regimens being administered to patients. Data generated through phase III randomized trials are considered the highest level of that evidence.

Although concerns have been raised in recent years regarding the need for randomized trials to augment the body of clinical understanding, one critical issue that has failed to generate sufficient discussion is how the choice of the control arm affects the interpretation of an individual study’s outcome and potentially undermines the ethical basis for that particular study.

Consider, for example, the highly questionable and widely discussed decision several years ago to initially not permit crossover in a phase III trial comparing the highly clinically active BRAF inhibitor vemurafenib (Zelboraf) with the widely recognized essentially inactive dacarbazine in patients with BRAF mutation—positive metastatic melanoma.1 Or one might note the use of an excessively toxic dose of pegylated liposomal doxorubicin (50 mg/m2) in the control arm of several phase III studies in platinum-resistant ovarian cancer even though solid evidence existed that the lower-dose community standard of 40 mg/m2 has equivalent efficacy and less toxicity.2

More recently, investigators published the results of a phase III trial purportedly designed to examine the utility of a weekly versus a once-every-3-weeks dose of paclitaxel as primary therapy for advanced ovarian cancer. Remarkably, the study permitted the addition of bevacizumab (Avastin) to each trial arm, including the control arm, making any conclusion regarding the main study question quite dubious at best.3 To be clear, I am not challenging the documented utility of bevacizumab as a component of primary treatment of ovarian cancer; rather, my intention is to point out how the use of a bevacizumab-containing control arm influences the answer to the primary question of the trial.

In this discussion of the appropriate control arm, it is relevant to highlight the speed at which novel therapeutic strategies are becoming standard-of-care options. It is quite unclear how the results of an ongoing or soon-to-bereported randomized phase III trial will be interpreted if the control arm treatment, once a viable option for therapy, becomes a discarded regimen. This is particularly problematic if the trial results based on this now-inferior therapy reveal substantial, rather than marginal, improvements in meaningful survival outcomes.

Finally, I turn to the apparent selection of such a study arm in a recently activated phase III trial in the ovarian cancer arena. A novel approach to cancer management that employs the delivery of low-intensity alternating electric fields to a tumor has been approved by the FDA for the treatment of malignant gliomas and was shown in a small phase II trial (n = 31 patients) in platinum-resistant ovarian cancer to produce a 25% objective response rate and result in a progression-free survival (PFS) of 8.9 months when delivered in combination with weekly paclitaxel.4 These results were interpreted as encouraging compared with findings from a historical group of patients in this clinical setting treated with weekly paclitaxel alone. The control arm of the recently activated phase III trial will be single-agent paclitaxel delivered on a weekly schedule, and the primary study endpoint is overall survival (OS), with PFS being a secondary endpoint.

One wonders how the decision to employ weekly paclitaxel as a control arm can be justified when the results of a landmark phase III randomized trial comparing weekly paclitaxel, topotecan, or pegylated liposomal doxorubicin with one of these agents plus bevacizumab in platinum-resistant ovarian cancer revealed a strikingly improved PFS (6.7 vs 3.4 months; HR, 0.48) in favor of the bevacizumab-containing strategy.5 Further, the specific combination of weekly paclitaxel plus bevacizumab revealed a median PFS of 10.4 months (versus 3.9 months in the weekly paclitaxel control arm; HR, 0.46), with an objective response rate of 53.3% for the combination compared with 30.2% for single-agent paclitaxel.6

Although there was no statistically significant difference in OS within the entire study population, this outcome is almost certainly related to posttreatment strategies administered to study participants, including bevacizumab-based regimens in women who did not receive this agent during the trial, which ultimately impacted the study’s ultimate survival result. However, in the paclitaxel-specific cohort, the median OS was 22.4 months when this agent was combined with bevacizumab compared with 13.2 months for paclitaxel alone (HR, 0.65).6

Therefore, a phase III study that employs weekly paclitaxel in platinum-resistant disease as a control arm has a very high probability of being uninterpretable, because any subsequently reported “favorable outcome” will be—and should be—seriously challenged due to the selection of an irrelevant study control.

Finally, one must ask: Will future patients with ovarian cancer, the clinical oncology community, and our society really benefit from the conduct of a study for which the results cannot be adequately interpreted?

References

  1. Chapman PB, Hauschild A, Robert C, et al; BRIM-3 Study Group. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med. 2011;364(26):2507-2516. doi: 10.1056/NEJMoa1103782.
  2. Markman M. Serious ethical dilemma of single-agent pegylated liposomal doxorubicin employed as a control arm in ovarian cancer chemotherapy trials. J Clin Oncol. 2010; 28(19):e319-e320. doi: 10.1200/JCO.2010.28.9934.
  3. Chan JK, Brady MF, Penson RT, et al. Weekly vs. every 3-week paclitaxel and carboplatin for ovarian cancer. N Engl J Med. 2016;374(8):738-748. doi: 10.1056/NEJMoa1505067.
  4. Vergote I, von Moos R, Manso L, Van Nieuwenhuysen E, Concin N, Sessa C. Tumor treating fields in combination with paclitaxel in recurrent ovarian carcinoma: results of the INNOVATE pilot study. Gynecol Oncol. 2018;150(3):471-477. doi: 10.1016/j.ygyno.2018.07.018.
  5. Pujade-Lauraine E, Hilpert F, Weber B, et al. Bevacizumab combined with chemotherapy for platinum- resistant recurrent ovarian cancer: the AURELIA open-label randomized phase III trial. J Clin Oncol. 2014;32(13):1302-1308. doi: 10.1200/JCO.2013.51.4489.
  6. Poveda AM, Selle F, Hilpert F, et al. Bevacizumab combined with weekly paclitaxel, pegylated liposomal doxorubicin, and topotecan in platinum-resistant recurrent ovarian cancer: analysis by chemotherapy cohort of the randomized phase III AURELIA trial. J Clin Oncol. 2015;33(32):3836-3838. doi: 10.1200/JCO.2015.63.1408.