Tim Eisen, FMedSci, FRCP, PhD
In a phase III trial (SORCE) reported in the Journal of Clinical Oncology, Tim Eisen, FMedSci, FRCP, PhD, of Cambridge University Hospitals NHS Foundation Trust, and colleagues found no disease-free survival benefit with 3 years of sorafenib vs placebo as adjuvant therapy in patients with renal cell carcinoma at intermediate or high risk of recurrence.1 A minority of sorafenib patients completed protocol therapy.
In the intergroup double-blind trial, 1,711 patients from sites in Belgium, Denmark, France, the Netherlands, Spain, the United Kingdom, and Australia undergoing excision of primary renal cell carcinoma were randomly assigned 2:3:3 between July 2007 and April 2013 to receive placebo for 3 years (n = 430), sorafenib for 1 year followed by placebo for 2 years (n = 642), or sorafenib for 3 years (n = 639). The initial starting dose of sorafenib was 400 mg twice daily (received by 13% of patients), with permitted dose reductions for toxicity to 400 mg once daily and then to 400 mg on alternate days. In January 2009, the starting dose was amended to 400 mg once daily to address a higher-than-expected discontinuation rate. After 3 weeks, the dose could be maintained or escalated to the full dose of 400 mg twice daily at clinician discretion. The primary outcome analysis was investigator-reported disease-free survival, comparing 3 years of sorafenib vs placebo in the intent-to-treat population.
Two planned interim efficacy analyses were performed after approximately 200 and 400 of the planned events had occurred using a significance level of P = .001 at each interim analysis, and the independent data monitoring committee requested an additional efficacy analysis after approximately 500 events. After each analysis, the committee recommended the trial continue as planned. The final analysis was performed without adjustment for multiple testing.
Overall, 84% of patients had clear cell histology and risk of recurrence based on a Leibovich score that was intermediate in 53% and high in 47%.
Median follow-up at the time of analysis was 6.5 years. Despite the reduction in the initial dose for the first 3 weeks, permitting flexibility in the requirement for escalation to standard dose and encouraging dose reductions to reduce toxicity, 33% and 25% of patients who received sorafenib completed 1 and 3 years of protocol treatment, respectively.Excessive toxicity and patient refusal were reported as the primary reasons for stopping treatment.
No difference in disease-free survival was observed for the 3-year sorafenib group vs the placebo group, with 245 vs 167 events being observed (hazard ratio [HR] = 1.01, 95% confidence interval [CI] = 0.83–1.23, P = .95). Median disease-free survival was not reached in either group. Since nonproportional hazards were observed (P = .042), a restricted mean survival time analysis was performed; estimated analysis over 10 years was 6.81 years in the 3-year sorafenib group vs 6.82 years in the placebo group (difference = 0.01 year, P = .988). No difference in disease-free survival was observed between the 1-year sorafenib group vs the placebo group (HR = 0.94, 95% CI = 0.77–1.14, P = .509). Nonproportional hazards were observed (P = .001). The restricted mean survival time estimate over 10 years was 6.98 years in the 1-year sorafenib group and 6.79 years in the placebo group (difference = 0.19 years, P = .422).
A total of 20% of patients in each group received additional systemic treatment after disease progression.
No differences in overall survival were observed vs placebo for the 3-year sorafenib group (HR= 1.06, 95% CI = 0.82–1.38, P = .638) or for the 1-year sorafenib group (HR = 0.92, 95% CI = 0.71–1.20, P = .541). Overall survival at 10 years was 70% in the 3-year group, 69% in the 1-year group, and 69% in the placebo group.
In preplanned subgroup analyses, no significant differences in disease-free survival for the 3-year sorafenib group (HR = 1.02, P = .875) or 1-year sorafenib group (HR = 0.92, P =.49) vs the placebo group were observed among patients at high risk of recurrence. Similarly, no differences were observed for the 3-year group (HR = 0.99, P = .93) or the 1-year group (HR = 0.94, P = .55) among the population with clear cell renal cell carcinoma.
The most common adverse events of any grade in the sorafenib groups were hand-foot syndrome (77% of 3-year group, 79% of 1-year group, 32% of placebo group), fatigue (74%, 74%, 60%), rash (71%, 70%, 30%), and hypertension (64%, 60%, 48%). Grade ≥ 3 adverse events occurred in 63.9% of the 3-year sorafenib group, 58.6% of the 1-year sorafenib group, and 29.2% of the placebo group. Serious adverse events occurred in 24%, 21.6%, and 19.1% of patients, respectively. Grade 3 hand-foot skin reaction was reported in 24% of patients in each sorafenib group. Grade 3 hypertension occurred in 26% and 24% of the sorafenib groups and 20% of the placebo group. Treatment was stopped due to excessive toxicity in 34%, 30%, and 4% of patients.
The investigators stated: “[In our analysis,] [w]e did not observe proportional hazards; we found an apparent separation between the curves, but it does not support further investigation when the overall results are so definitive.”
They concluded: “Sorafenib should not be used as adjuvant therapy for [renal cell carcinoma]. Active surveillance remains the standard of care for patients at intermediate or high risk of recurrence after nephrectomy and is the appropriate control of our current international adjuvant [renal cell carcinoma] trial, RAMPART [ClinicalTrials.gov identifier NCT03288532].”
DISCLOSURE: The study was supported by Cancer Research UK, an educational research grant from Bayer, and others. Dr. Eisen, in addition to his academic affiliation, is Franchise Head of GU Oncology at Roche.
1. Eisen T, Frangou E, Oza B, et al: Adjuvant sorafenib for renal cell carcinoma at intermediate or high risk of relapse: Results from the SORCE randomized phase III intergroup trial. J Clin Oncol. October 14, 2020 (early release online).