Trial Failure Still Yields Insights for Uterine Leiomyosarcoma

A multicenter, international, randomized phase III trial (ClinicalTrials.gov identifier: NCT01533207) set out to evaluate a chemotherapy regimen in the adjuvant setting for patients with high-grade uterine leiomyosarcoma but failed to generate clinically meaningful results due to slow patient accrual and lack of statistical power. Despite this, experts told Cancer Network that insights can be gleaned from this trial. The results were published in the Journal of Clinical Oncology.

Investigators enrolled patients with resected high-grade uterine leiomyosarcoma that was confined to the uterus and randomly assigned patients to receive either an adjuvant chemotherapy regimen with gemcitabine plus docetaxel followed by doxorubicin or no treatment at all, which is considered the standard of care. The primary endpoint was overall survival.

A total of 701 sites were open to enroll patients, with the goal of enrolling 216 patients. Although the study was open for 4 years, only 38 patients enrolled, or approximately 17% of the goal, leading to the study halting patient enrollment. In all, 20 patients were assigned to the adjuvant chemotherapy regimen and 18 patients were assigned to the control group.

“I think the authors did about as good a job as you could do with this disease,” Robert Edwards, MD, Chair and Professor of Obstetrics, Gynecology and Reproductive Services at the University of Pittsburgh School of Medicine, told Cancer Network. He applauded the investigators for designing the trial as multicenter and international.

However, despite the trial including hundreds of sites internationally, accrual was slow and problematic. Edwards cited the rare nature of the disease as one reason for slow accrual. Beyond that, patients may have not been interested in joining the clinical trial because neither the intervention nor control arm were desirable, he said.

Edwards explained that trials with a no-treatment control arm rarely accrue effectively and suggested having a more progressive trial design for rare cancers such as uterine leiomyosarcoma. Rather than receive no treatment, patients in the control group could receive a targeted therapy based on molecular profiling of the tumor. “It’s not as scientifically pure as what was proposed in this trial, but when you’re dealing with such a terrible disease, it’s hard to get patients to agree to follow through with the [no-treatment] control arm,” he said.

As for the chemotherapy arm, Eloise Chapman-Davis, MD, a gynecologic oncologist at Weill Cornell Medicine and New York-Presbyterian, told Cancer Network that patients typically do not want to undergo cytotoxic chemotherapy because risk of recurrence is high, around 50% to 70%. Rather than undergo chemotherapy prophylactically, patients can just wait for their tumor to recur and then initiate chemotherapy. Furthermore, evidence suggests that adjuvant chemotherapy not only fails to improve outcomes, but may actually decrease survival.

Due to the small study population and limited number of events, a median overall survival could not be calculated for the control arm. Instead, restricted mean survival time was used to determine the mean survival in each arm.

Patients on the chemotherapy arm had worse overall survival than those on the control arm, living on average 12 fewer months (34.3 months vs 46.4 months). Recurrence free–survival was, on average, longer for patients in the chemotherapy arm compared with those in the control arm (18.1 months vs 14.6 months).

However, the study authors wrote, “Neither survival outcome comparison was considered statistically robust, due to the small sample size.”

Acknowledging the small study population, Chapman-Davis concluded that the trial adds to the existing literature that supports observation as the standard of care in the adjuvant setting for patients with high-grade uterine leiomyosarcoma. In addition, she does not foresee this trial being redone.

“This may be the only study that we have, because if this took 700 international sites and they still only got 38 patients, I don’t think that doing this study again is going to be better,” she said.