Adjuvant Everolimus Yields RFS Benefit Over Placebo in Very High-Risk Renal Cell Carcinoma

mTOR inhibitor everolimus (Afinitor) yielded a reduction in risk of recurrence or death of 21% vs placebo in very high-risk renal cell carcinoma, according to data from the phase 3 EVEREST trial (NCT01120249) presented at the 2022 American Society of Clinical Oncology Annual Meeting.1

Across all patients in the study (N = 1499), there was a 15% reduction in the risk of recurrence or death (HR, 0.85; 95% CI, 0.72-1.00; P = .025); however, this narrowly missed the bar for statistical significance (defined as P = .022). The only statistically significant improvement in recurrence-free survival (RFS) was observed in a very high-risk subgroup, which was defined as a range from pT3a grade 3 to 4 to pT4 of any grade with no nodal involvement or those with any pT stage and grade with node positivity (HR, 0.79; 95% 0.65-0.97; P = .011).

“Adjuvant everolimus improved RFS in RCC patients after nephrectomy but the nominal significance level was narrowly missed. The effect of everolimus was especially pronounced in those with very high-risk disease, a patient population for whom adjuvant therapy is currently recommended,” Christopher W. Ryan, MD, Oregon Health & Science University, Knight Cancer Institute, said during a presentation of the results. “These compelling results warrant further investigation into the role of everolimus in the current adjuvant landscape and subsets who might benefit the most.”

In the EVEREST study, 1499 patients with RCC were randomized and treated with everolimus (n = 755) or placebo (n = 744). Forty-five percent of patients had intermediate high-risk RCC, defined as a range from pT1b grade 3 to 4 with no metastases to pT3a grade 1 to 2 with no metastases. The remaining 55% of patients had very high-risk disease. Oral everolimus was administered at 10 mg daily for 54 weeks and treatment was started within 12 weeks of radical or partial nephrectomy.

The median age of patients was approximately 58 years and most had clear cell RCC (83%), with 17% having non-clear cell histology. Surgical intervention was primarily a radical nephrectomy for 90% of patients compared with 10% receiving a partial nephrectomy. The ECOG performance score was 0 for 80% of patients and 1 for the rest.

Across all patients, the 5-year estimated RFS rate was 67% with everolimus compared with 63% with placebo. In the very high-risk group, the 5-year RFS rate was 57% with everolimus vs 51% with placebo. The median was not yet reached. In those with intermediate high-risk RCC, the 5-year RFS rate was 80% vs 78%, for everolimus and placebo, respectively. Although the subgroup was small (n = 61), an RFS benefit with a hazard ratio around 0.50 favoring everolimus was nearly significant in African American patients (P = .09).

“There's a general consistent effect favoring everolimus in most subgroups tested. Of particular note, is the robust hazard ratio favoring everolimus in African Americans, but this is a very small group,” said Ryan.

There was a 10% reduction in the risk of death observed at the analysis, but this did not pass the bar for significance (HR, 0.90; 95% CI, 0.71-1.13; P = .178). Ryan added that overall survival (OS) has not yet been achieved for any adjuvant therapy in RCC. The 5-year estimated OS was 87% with everolimus compared with 85% for placebo.

At a median follow-up of 76 months, 55% of patients in the everolimus arm and 31% of patients in the placebo arm had discontinued therapy. The median time on treatment was 9.3 months for everolimus vs 12.6 months for placebo. Given this high discontinuation rate and consistent RFS increase, Ryan suggested the optimal duration of adjuvant therapy should be re-examined.

The primary cause for discontinuation was adverse events (AE), with 37% of patients in the everolimus arm discontinuing due to an AE compared with 5% in the placebo group. In the placebo arm, 14% discontinued due to disease recurrence compared with 8% in the everolimus group. Dose reductions were required for 37% of patients in the everolimus arm and for 7% in the placebo group.

Adverse events occurred in 96% of those treated with everolimus and in 81% of those in the placebo group, with grade 3 or greater AEs in 46% and 11% of patients, respectively. There were no grade 5 AEs. “Adverse events more common on the everolimus arm were those well known to be associated with this agent and were primarily gastrointestinal or cutaneous in nature,” said Ryan.

The most frequent all-grade AEs with everolimus and placebo, respectively, were oral mucositis (64% vs 19%), fatigue (56% vs 41%), diarrhea (33% vs 15%), nausea (24% vs 17%), maculopapular rash (31% vs 8%), acneiform rash (29% vs 5%), pruritus (18% vs 8%), headache (18% vs 11%), dry skin (17% vs 8%), hypertension (16% vs 13%), anorexia (16% vs 5%), dyspnea (15% vs 6%), limb edema (15% vs 5%), and pneumonitis (13% vs 0%). The most common laboratory abnormalities of grade 3 or greater were hypertriglyceridemia (11% vs 2%), hypoglycemia (5% vs 0%), anemia (2% vs 0%), ALT increase (1% vs 0%), and lymphocyte count decrease (1% vs 0%).

EVEREST in Context

The EVEREST trial represents the first results for an adjuvant mTOR inhibitor for RCC. There have been numerous studies of other agents in this setting, including several for VEGF TKIs. At this point, many of these studies have failed to show meaningful results, with the exception being the S-TRAC study of sunitinib (Sutent) and the KEYNOTE-564 trial of pembrolizumab (Keytruda), Ryan said.

“To date, 6 of these new generation placebo-controlled adjuvant trials have been published, including 5 trials of adjuvant tyrosine kinase inhibitor therapy, only 1 of which, the S-TRAC trial of sunitinib, demonstrated disease-free survival benefit,” he said. “As of yet, only 1 published study of checkpoint inhibitor therapy with pembrolizumab also showed a disease-free survival benefit. Of note, overall survival has been a secondary endpoint of all these studies and none, as of yet, has reported a significant survival benefit."

Several factors impacted the statistical model of the EVEREST study. The original sample size was planned to be 1170 patients; however, a disproportionate number of discontinuations in the everolimus arm warranted an increase to 1464 patients. With this change, the study was updated to have an 80% power to detect an 18% difference in the hazard ratio for RFS, with a 1-side alpha of 2.5%. This equates to a median RFS of 6 years with everolimus vs 4.9 years with placebo.

In addition to changes to the number of study participants, RFS events occurred less frequently than initially anticipated, resulting in an early review of the results. A final analysis of the study was planned when 804 events had occurred; however, this had not taken place at a fourth interim analysis in March 2021. At this point, investigators set the cutoff as March 2022, which were the data presented at ASCO. For this analysis, there had been 556 RFS events.

“This revision was undertaken by 2 statisticians independent of the study with no knowledge of the end point data and working in conjunction with the medical input of the study chair, which was myself, and with the data monitoring committee,” said Ryan.

In a Tweet commenting on the results, Sumanta (Monty) Kumar Pal, MD, a medical oncologist from City of Hope, called the findings thought provoking but unlikely to alter the standard of care (SOC). “[The] outcome is thought provoking,” he tweeted. “[It] will not change SOC but intriguing to think that some subsets could benefit⁠—could mTOR pathway altered patients drive benefit too?”

Additional analyses of the study remain ongoing.

Reference

Ryan CW, Tangen C, Heath EI, et al. EVEREST: Everolimus for renal cancer ensuing surgical therapy—A phase III study (SWOG S0931, NCT01120249). J Clin Oncol. 2022;40 (suppl 17; abstr LBA4500). Doi: 10.1200/JCO.2022.40.17_suppl.LBA4500.