Increased Efficacy but Added Toxicity with Bevacizumab plus Everolimus in pNETs

Increased Efficacy but Added Toxicity with Bevacizumab plus Everolimus in pNETs

The addition of bevacizumab to everolimus is associated with a significant improvement in efficacy but a high rate of toxicity in patients with locally advanced or metastatic pancreatic neuroendocrine tumors (pNETs), according to results of the randomized, phase II CALGB 80701 trial presented by Matthew H. Kulke, MD, of Dana-Farber Cancer Institute, during the Gastrointestinal (Noncolorectal) Cancer Oral Abstract Session held on Sunday, May 31 (Abstract 4005).

Dr. Matthew H. Kulke

In the study, 150 patients with advanced pNETs were randomly assigned to receive 10 mg daily of everolimus with or without 10 mg/kg of bevacizumab every 2 weeks, in addition to depot octreotide acetate. The combination of everolimus and bevacizumab was associated with a significant improvement in median progression-free survival (PFS) compared with everolimus alone (16.7 vs. 14.0 months; hazard ratio 0.80, 95% CI [0.55, 1.17]; p = 0.12). This difference was considered statistically significant based on a predefined threshold of a one-sided a equal to .15 using a stratified log-rank test with 90% power.

The combination was also associated with a significant improvement in response rate, with responses observed in 31% of patients receiving everolimus plus bevacizumab compared with 12% of patients receiving everolimus alone (p = 0.005). “That’s a paradigm shift for this disease, in which targeted agents are usually on the order of single digits for response rates,” Diane Reidy-Lagunes, MD, of Memorial Sloan Kettering Cancer Center, said during her discussion of the study results.

There was no significant improvement with everolimus plus bevacizumab versus everolimus alone for median overall survival (36.7 vs. 35.0 months) or time to treatment failure (12.6 vs. 12.2 months). Moreover, enthusiasm for the regimen was greatly tempered by the substantial increase in toxicity observed with the combination.

Grade 3/4 adverse events were reported in 81% of patients in the combination arm compared with 49% in the everolimus arm. “[81%] is an extremely high number,” commented Dr. Reidy-Lagunes. Key grade 3/4 adverse events occurring at higher rates in the combination arm versus the control arm included hypertension (38% vs. 8%), proteinuria (16% vs. 1%), infection (6% vs. 2%), diarrhea (11% vs. 1%), and electrolyte imbalances. Adverse events led to higher rates of dose modifications and dose delays of both study drugs in the combination arm, leading more patients to discontinue the study because of adverse events or patient/provider discretion.

The combination of the two targeted agents was pursued based on the rationale that administering agents that target different steps of the receptor tyrosine kinase/AKT/mTOR pathway could yield a synergistic effect, “potentially blocking escape mechanisms upstream and downstream,” Dr. Reidy-Lagunes explained. Previous studies had laid the groundwork for this approach, demonstrating the activity of bevacizumab with an mTOR inhibitor (temsirolimus or everolimus) in pNETs.

“Where do we go with this promising yet toxic data?” Dr. Reidy-Lagunes asked the audience. A variety of potential strategies were discussed, including the use of induction therapy followed by a less intense maintenance therapy, the use of biomarkers to identify patients most likely to benefit, and the use of sequential rather than concomitant therapy.  

Watch the session: Visit the ASCO Virtual Meeting website.