Investigators Hope Glioma Trial of Eflornithine Will Clinch Approval

Emmanuel S. Antonarakis, MD, associate professor of oncology and urology at Johns Hopkins Medicine

Victor A. Levin, MD

Eflornithine (α-Difluoromethylornithine) has been investigated for over 30 years as a potential neuro-oncology agent and shown activity against recurrent gliomas, but it has not gained approval as a therapy option for patients with cancer. Investigators are looking to change that with the phase III STELLAR trial (NCT02796261), which, with updated enrollment criteria, is recruiting patients with anaplastic astrocytoma (AA) that has recurred or progressed following radiation and temozolomide chemotherapy.¹ The trial is comparing the efficacy and safety of combination eflornithine and lomustine with that of lomustine monotherapy.

“[Eflornithine] has a unique history, and the drug does have activity in a variety of tumors and theoretically could be used in combination in some of those tumor types,” said Victor A. Levin, MD, emeritus professor of neuro-oncology at The University of Texas MD Anderson Cancer Center in Houston and a clinical professor in the Department of Neurosurgery at the University of California, San Francisco, School of Medicine. Levin has worked with eflornithine in the glioblastoma (GBM) setting since the 1980s.

Eflornithine inhibits ornithine decarboxylase in an irreversible fashion. It has marketing approval in the United States for topical use to prevent excessive hair growth (Vaniqa). Outside this country, eflornithine is used to treat African trypanosomiasis, a sleeping sickness.

A History of Progress

Between 1989 and 2007, more than 390 patients with high-grade gliomas, including AA, were treated in clinical trials that demonstrated activity with eflornithine alone and in combination with other agents.²

A phase I/II study compared different dose schedules of eflornithine in combination with mitoguazone (MGBG) for the treatment of patients with recurrent malignant gliomas. There were 2 patient cohorts including 7 patients with astrocytoma, 13 with AA and anaplastic oligoastrocytoma (AOA), 6 with GBM, and 1 with brainstem glioma.

In the second cohort, median time to tumor progression was 16 weeks in 11 patients with AA and AOA and 51 weeks in 6 patients with astrocytoma. These results led to an orphan drug, grant-sponsored randomized phase II study to evaluate single-agent eflornithine against eflornithine plus MGBG for recurrent malignant gliomas.

Although the combination arm was discontinued owing to unanticipated fulminant and fatal hepatotoxicity in 2 patients, 80 patients (44 with anaplastic glioma, 36 with GBM) were evaluable for response in the singleagent arm. In the anaplastic glioma subset, 45% of patients had antitumor activity, with 4 partial responses (PRs), 9 minor responses (MRs), and 7 patients with stable disease (SD). The median progression-free survival (PFS) was 49 weeks. In the evaluable GBM cohort (36), just 18% benefited from eflornithine, with 2 PRs, 2 MDs, and 2 SD responses; these 6 patients had a median PFS of 32 weeks.²

The clinical activity in anaplastic gliomas led to the phase III randomized trial of postirradiation therapy with procarbazine, lomustine, and vincristine (PCV) versus eflornithine in combination with PCV in patients with anaplastic glioma.³

In the combination arm (n = 114), 78.1% of patients had AA, 3.5% had AOA, 14.0% had anaplastic oligodendroglioma (AO), and 4.4% had other malignant gliomas. In the PCV arm (n = 114), 69.3% of patients had AA, 7.0% had AOA, 21.1% had AO, and 2.6% had other malignant gliomas. In the 2 treatment groups, PFS was 71.1 versus 37.5 months (P = .13), respectively, and median overall survival (OS) was 75.8 versus 61.1 months. For patients with AA, the median OS was 71.2 months for the combination and 46 months (P = .12) in the PCV arm.³

The above-mentioned study was published in 2003, and the patient database was updated in 2009 and 2012. “What we found when we studied [eflornithine’s] activity as a single agent against anaplastic gliomas, and when we studied it in combination with PCV, was that it expanded the duration of survival,” Levin said.

Investigators included the updated data in a 2018 paper that showed a median OS of 85 months versus 52.4 months (HR, 0.75; 95% CI, 0.52-1.07; P = .12) for patients with AA or AOA in the combination and PCV arms, respectively.²

Seeking Approval in STELLAR Trial

Figure. Phase III STELLAR Trial: Eflornithine with Lomustine Versus Lomustine¹

From these trials comes the phase III STELLAR trial evaluating the safety and efficacy of eflornithine in combination with lomustine versus lomustine monotherapy for patients whose AA has recurred/progressed after radiation and temozolomide chemotherapy. The primary endpoint is OS, and secondary outcome measures are PFS and objective response rate (Figure).¹ The trial is designed to last up to 50 months and include study periods of up to 24 months for the experimental arm versus 12 months for the comparator arm.¹ “If we can treat patients with [AA] with eflornithine longer, we might be able to push the needle further,” Levin said.

“Eflornithine is not going to work in every patient, but there is going to be a subset of patients in whom this [drug] works really well, and those patients will have a disproportionate benefit. We’re excited about the study’s design and the changes we [have] made to it because we think it will have an impact on all anaplastic gliomas, [in part because of] the new classification schema that includes the lower-grade gliomas,” Levin noted. The expanded schema includes patients who have WHO grade 2 astrocytoma or oligoastrocytoma that progresses to WHO grade 3 AA.

Eflornithine is an oral preparation, is not metabolized to a significant extent, and is excreted through urine unchanged. It is historically well tolerated by patients. The drug can cause loose stools and diarrhea, which can be treated with OTC medication, according to Levin. “Since the drug is not metabolized, it does not have an effect on [the] liver,” he added.

One notable adverse effect is hearing loss. “We do produce a dose-reduction schedule to minimize the effects on hearing,” Levin said. “But in most cases, it is reversible once the drug is stopped.”

Recruitment Barriers

The incidence of AA is roughly 0.48 per 100,000 person-years, and the 5- and 10-year relative survival rates of patients with AA are 23.6% and 15.1%, respectively.⁴ Fewer than half of tumors will recur as AA without overt transformation to GBM, according to Levin. “Larger institutions, with respect to accrual for primary brain tumor studies… maybe see 3 to 4 new patients per year for this study.”

To help increase accrual for the study, Orbus Therapeutics, the developer of eflornithine, is recruiting patients from over 90 sites worldwide. Most of them are in the United States, but they extend to Canada, Belgium, France, Germany, Italy, the Netherlands, and the United Kingdom.

“[The trial] is very hard to accrue, since [AA] is a rare tumor when it comes to recurrence,” Levin said. “Right now, I know we’ve passed the halfway mark, which is exciting, but it is very slow going.” Levin has confidence that the trial will lead to approval for eflornithine. “Patients are going to be happy because they are going to be living longer.”

Orbus Therapeutics is based in Palo Alto, California.

References

1. Study to Evaluate Eflornithine + Lomustine vs Lomustine in Recurrent Anaplastic Astrocytoma (AA) Patients (STELLAR). clinicaltrials. gov/ct2/show/NCT02796261. Updated March 15, 2019. Accessed May 7, 2019.
2. Levin VA, Ictech SE, Hess KR. Clinical importance of eflornithine (α-difluoromethylornithine) for the treatment of malignant gliomas. CNS Oncol. 2018;7(2):CNS16. doi: 10.2217/cns-2017-0031.
3. Levin VA, Hess KR, Choucair A, et al. Phase III randomized study of postradiotherapy chemotherapy with combination alpha-difluoromethylornithine- PCV versus PCV for anaplastic gliomas. Clin Cancer Res. 2003;9(3):981-990.
4. Smoll NR, Hamilton B. Incidence and relative survival of anaplastic astrocytomas. Neuro Oncol. 2014;16(10):1400-1407. doi: 10.1093/neuonc/nou053.