Everolimus Shows Robust Activity in Lung/GI NETs

Everolimus improved progression-free survival by 7.1 months compared with placebo in patients with lung/gastrointestinal (GI) neuroendocrine tumors, representing a 52% reduction in the risk of progression or death.

James C. Yao, MD

Everolimus (Afinitor) improved progression-free survival by 7.1 months compared with placebo in patients with lung/gastrointestinal (GI) neuroendocrine tumors (NETs), representing a 52% reduction in the risk of progression or death, according to findings from the RADIANT-4 trial presented by James Yao, MD, at the 2015 NANETS symposium.

"RADIANT-4 showed everolimus is the first targeted agent to show robust anti-tumor activity, with acceptable tolerability, against a broad spectrum of NETs, including those arising from the lung and GI tract," said Yao, professor of medical oncology at the University of Texas MD Anderson Cancer Center. “This is an area of a lot of unmet need. At this time, there is no drug that has ever demonstrated activity or been approved for lung neuroendocrine tumors, and that is roughly 25% to 30% of all neuroendocrine tumors out there. In progressive GI tumors, there are also limited options.”

The RADIANT-4 trial was the first large, placebo controlled, phase III study of advanced, progressive, nonfunctional lung/GI NET, according to Yao. In the study, 302 patients with progressive, well-differentiated, nonfunctional lung/GI NETs were randomized 2:1 to receive best supportive care plus either everolimus at 10 mg per day (n = 205) or placebo (n = 97). Tumors were located in the GI tract (n = 175), lung (n = 90), or were of unknown origin (n = 36).

By central review, everolimus demonstrated a median progression-free survival (PFS) of 11 months compared with 3.9 months for placebo (HR, 0.48; 95% CI, 0.35-0.67; P <.00001). By investigator assessment, the median PFS was 14 months with everolimus compared with 5.5 months for placebo (HR, 0.39; 95% CI, 0.28-0.54; P <.00001). This benefit was consistent across subgroup analyses.

"The study results showed a significant improvement in progression-free survival," Yao said. "The median PFS was improved in the central blinded analysis by a 2.8 fold improvement, which really shows a significant benefit in this group."

There was a 50% improvement in PFS seen for those with lung tumors and a 44% benefit in those with GI NETs, which were located in the stomach, colon, rectum, appendix, cecum, ileum, duodenum, and jejunum. Those with tumors of unknown origin saw a 40% improvement in PFS with everolimus.

Tumor shrinkage of any degree occurred in 64% of patients in the everolimus group and 26% of the placebo group, Yao said. The objective response rate (ORR) for those in the everolimus group was 2% compared with 1% in the placebo arm. The disease control rate (stable disease plus ORR) was 82.4% with everolimus versus 64.9% with placebo.

Benefits appeared early and were durable at 18 months, which was an interim analysis of the study. At this point, a 36% reduction in deaths with everolimus was seen; however, it did not reach statistical significance (P = .037). Death rates in the everolimus and placebo group were similar, although the everolimus group was treated longer, Yao said. He added that the next interim analysis would occur in 2016.

Adverse events were consistent with the known safety profile for everolimus. Drug-related adverse events were mostly grades 1 and 2 stomatitis, diarrhea, fatigue, infections, rash and peripheral edema. The most common grades 3 and 4 adverse events among everolimus-treated patients were stomatitis in 9%, diarrhea in 7%, and infections in 7%, Yao said. Grade 3 and 4 adverse events were uncommon in the placebo arm.

Patients received everolimus for a median of 40.4 weeks. The main reasons for treatment discontinuations were disease progression and adverse events. Disease progression occurred in 37% of patients treated with everolimus versus 72% of those in the placebo arm. Adverse events accounted for discontinuations in 29% and 7% of the groups, for everolimus and placebo, respectively, Yao said.

Speaking after Yao's presentation at the NANETS Symposium, Kjell Öberg, MD, PhD, said the RADIANT-4 trial was the first randomized study to show that everolimus could produce significant results for patients with lung tumors. Öberg, an endocrine oncology researcher at Uppsala (Sweden) University, noted that an earlier European study supported RADIANT-4's lung-related results.

Given the consistent results from all of the RADIANT studies (1, 2, 3 and 4), Öberg supported using everolimus in grades 1 and 2 disseminated and progressive NETs, regardless of primary tumor origin. In fact, at this time, the manufacturer of everolimus, Novartis, has noted that worldwide regulatory submission is currently underway for the medication as a treatment for patients with advanced, progressive, nonfunctional GI and lung NET.

Others at the symposium also shared their support for everolimus in patients with GI/Lung NETs. In an interview with OncLive, Jennifer Eads, MD, assistant professor of Medicine, senior clinical instructor of Medicine, Case Western University, labeled the advance as a much-needed medicine for a hard-to-treat population that would quickly be utilized, given past experience with the mTOR inhibitor following its approval for pancreatic NETs in 2011.

"RADIANT-4 is a very positive study. We don't currently have really any medical therapies available beyond somatostatin analogs for patients with carcinoid—to have another option is great," She told OncLive. "The results were very positive and we have a lot of experience with the drug having used it in pancreatic neuroendocrine tumors. We know what the experience will be like, giving it to patients, and will feel very comfortable giving it."

Yao JC, Fazio N, Singh S, et al. Safety and Efficacy of Everolimus in Advanced Nonfunctional Neuroendocrine Tumors (NET) of Lung or Gastrointestinal (GI) Origin: Findings of the Randomized, Placebo-Controlled, Double-blind, Multicenter, Phase 3 RADIANT-4 Study. Presented at: 2015 NANETS Symposium; October 15-17; Austin, TX. Abstract C51.

<<<

View more from the 2015 NANETS Symposium