No SSE-FS benefit is observed with radium-223 plus cabozantinib in renal cell carcinoma

Adding radium-223 dichloride (Ra-223) to cabozantinib (Cabometyx) did not improve symptomatic skeletal event-free survival (SSE-FS) compared with cabozantinib alone in patients with metastatic renal cell carcinoma (RCC) and bone metastases, according to results from the phase 2 RADICAL trial (Alliance A031801, NCT04071223) presented at the 2026 American Society of Clinical Oncology Annual Meeting.¹

Rana R. McKay, MD, FASCO, a genitourinary medical oncologist at UC San Diego Health in San Diego, California, presented the findings. Although the primary end point was not met, McKay highlighted a notable divergence in overall survival (OS) curves after 1 year that she described as a signal warranting further investigation.

Background and rationale

Bone metastases (BM) occur in approximately 30% of patients with RCC and are associated with worse OS and decreased quality of life.^1,2 The pathophysiology involves dysregulated osteoblast and osteoclast activity within the bone microenvironment. Ra-223 is a bone-seeking alpha-emitter with established efficacy in metastatic castration-resistant prostate cancer, and cabozantinib is a VEGF tyrosine kinase inhibitor (TKI) with known activity in patients with RCC and bone metastases. A pilot study of Ra-223 in combination with VEGF-targeted therapies (N=30) demonstrated safety and a reduction in bone turnover markers, supporting the rationale for the current trial.^1,3

The trial hypothesis was that the addition of Ra-223 to cabozantinib would improve SSE-FS compared with cabozantinib alone in patients with metastatic RCC and bone metastases.

How was the RADICAL study designed?

RADICAL was a randomized, open-label phase 2 trial enrolling patients with RCC of any histology who had at least 1 untreated bone metastasis, a Karnofsky Performance Status of at least 60%, and were receiving osteoclast-targeted therapy (OTT) unless contraindicated. Patients were randomly assigned 1:1 to cabozantinib (40–60 mg orally once daily) plus Ra-223 (55 kBq/kg intravenously every 28 days for 6 cycles; Arm A) or cabozantinib alone (60 mg orally once daily; Arm B). The starting dose of cabozantinib in the combination arm was 40 mg, with escalation to 60 mg in the absence of persistent grade 2 or 3 toxicity during the first cycle.

The primary end point was SSE-FS, defined as time to symptomatic fracture related to bone metastases, radiation to symptomatic bone metastases, surgery to symptomatic bone metastases, spinal cord compression, or death. Secondary end points included safety, SSE-FS in subgroups, time to first SSE, objective response rate (ORR), progression-free survival (PFS), time to subsequent treatment, and OS.

Stratification factors included prior/concurrent OTT use, International Metastatic RCC Database Consortium (IMDC) risk group, prior treatment, and opioid use. The trial was designed to detect an improvement in 6-month SSE-FS from 65% (cabozantinib) to 78% (combination), targeting 124 evaluable patients and 99 SSE-FS events, with 85% power using a stratified log-rank test (one-sided α=0.05). A prespecified interim futility analysis was triggered when 50 of the 99 expected SSE-FS events had been observed, using the Wieand rule (stopping recommended if HR >1.0).

Patient population

A total of 98 patients were randomly assigned: 48 to the combination arm and 50 to the cabozantinib-alone arm. Median age was 63 years (interquartile range [IQR] 56.0–71.0). Clear cell histology was present in 82.7% of patients, and 62.2% had a prior SSE before enrollment. The majority of patients (87.8%) had received prior systemic therapy, including prior PD-1/PD-L1 therapy in 60.2% and prior VEGF TKI in 22.4%. OTT use was reported in 79.6% of patients, and 49.0% were using opioids at baseline. Approximately 19.4% had liver metastases in addition to bone metastases.

Regarding treatment exposure, the median daily cabozantinib dose was 31.4 mg in the combination arm vs 40.0 mg in the monotherapy arm. In Arm A, 52.2% of patients completed all 6 cycles of Ra-223, with a median of 6 cycles received.

Efficacy outcomes

After 90 patients were enrolled, the prespecified futility analysis was triggered at 50 SSE-FS events (17 SSEs and 33 deaths). The stratified HR was 1.24 (95% CI, 0.62–2.48), crossing the Wieand futility boundary. Accordingly, the Data Safety Monitoring Board recommended closure to accrual, and the trial enrolled a total of 98 patients.

In the all-enrolled population (N=98), with a median follow-up of 13.1 months, median SSE-FS was 16.7 months (90% CI, 13.5–39.6) with Ra-223 plus cabozantinib vs 17.6 months (90% CI, 11.5–24.5) with cabozantinib alone (stratified HR, 1.46; 90% CI, 0.86–2.51; one-sided P = .12). A post hoc unstratified analysis, conducted given divergence between stratified and unstratified estimates attributable to the overparameterized model, yielded an unstratified HR of 0.97 (90% CI, 0.62–1.51; one-sided P = .46) for the primary end point.

Subgroup analyses of SSE-FS across IMDC risk group, OTT use, prior treatment status, and baseline opioid use did not identify any subgroup with a consistent benefit favoring the combination, although confidence intervals were wide given limited sample sizes within each subgroup.

Median PFS was 10.3 months (95% CI, 5.5–16.8) in Arm A vs 10.5 months (95% CI, 8.7–16.9) in Arm B (stratified HR, 1.37; 95% CI, 0.74–2.53; p=0.32). Median OS was 28.3 months (95% CI, 15.1–not estimable [NE]) with the combination vs 19.7 months (95% CI, 12.3–NE) with cabozantinib alone (stratified HR, 1.40; 95% CI, 0.70–2.79; p=0.34). The OS curves diverged after 1 year and remained separated. The post hoc unstratified HR for OS was 0.82 (95% CI, 0.47–1.45; P = .50).

"The combination did not demonstrate a benefit of SSE-free survival. However, the OS curves diverge, and I think there's a signal there to further investigate," McKay said.

No statistically significant differences were observed between arms for time to first SSE, ORR, or time to subsequent anticancer therapy. Confirmed ORR was 19.4% in Arm A vs 25.0% in Arm B (P = .78). Median time to subsequent anticancer therapy was 18.6 months (95% CI, 13.8–NE) vs 19.2 months (95% CI, 15.1–NE) (stratified HR, 0.89; 95% CI, 0.40–1.96; P = .77).

Safety

The safety profile was consistent with the known toxicities of each agent. Grade 3 or higher adverse events (AEs) occurred in 71% of patients in the combination arm vs 77% in the cabozantinib-alone arm. Hematologic toxicity was more common with the combination, including higher rates of anemia (76.0% vs 59.1%), decreased white blood cell count (60.8% vs 42.9%), and decreased lymphocyte count (50.0% vs 38.8%). Other frequently observed AEs in both arms included diarrhea, nausea, fatigue, and hypertension.

"The combination was tolerable with a manageable safety profile consistent with the known toxicities of each agent," McKay said.

Limitations and conclusions

McKay outlined 3 key study limitations. First, the trial had a prolonged accrual period that coincided with a major shift in the RCC treatment landscape, including the widespread adoption of immune checkpoint inhibitor-based combinations as first-line standard of care, resulting in this trial largely enrolling second-line patients. Second, the use of 4 stratification factors relative to the modest sample size led to an overparameterized model with unstable stratified HR estimates, producing substantial divergence from the unstratified analyses. Third, the low on-trial SSE rate—attributable in part to the fact that most patients had experienced a prior SSE and were receiving concurrent OTT at baseline—may have limited the primary end point's sensitivity to detect a treatment effect.

McKay noted the historical significance of the trial.

"I think this is a groundbreaking study because it's the first randomized trial of a radiopharmaceutical in renal cell carcinoma, and it's the first trial that's site specific in renal cell carcinoma."

DISCLOSURES: McKay reported consulting/advisory roles with Ambrx, Arcus Biosciences, Astellas Medivation, AstraZenec, AVEO, Bayer, Blue Earth Diagnostics, Bristol Myers Squibb, Calithera Biosciences, Caris Life Sciences, Daiichi Sankyo, Dendreon, Esiai, Exelixis, Janssen; Lilly, Merck, Myovant Sciences, NeoMorph, Novartis, Pfizer, Precede Bio, Sanofi, Seagen, Sorrento Therapeutics, Sumitomo Pharma Oncology, Telix Pharmaceuticals, Tempus, and Vividion Therapeutics. She also reported institutional research funding from Artera, AstraZeneca, Bayer, Bristol Myers Squibb, Exelixis, Oncternal Therapeutics, and Tempus.

REFERENCES

1. McKay RR, Atherton P, Ballman KV, et al. A phase 2 randomized trial of radium-223 dichloride and cabozantinib in patients (pts) with renal cell carcinoma (RCC) with bone metastases (BM): RADICAL (Alliance A031801). J Clin Oncol. 2026 (suppl 16; abstr 4500). doi:10.1200/JCO.2026.44.16_suppl.4500

2. McKay RR, Bossé D, Xie W, et al. The activity of cabozantinib, crizotinib, and volitinib in patients with advanced papillary renal cell carcinoma and bone metastases. Eur Urol. 2014;66(6):1064-1072. doi:10.1016/j.eururo.2014.07.021

3. McKay RR, Kroeger N, Xie W, et al. Impact of bone and liver metastases on patients with renal cell carcinoma treated with targeted therapy. Eur Urol. 2014;65(3):577-584. doi:10.1016/j.eururo.2013.08.012