FDA Approval Summary: Belzutifan for Patients with Advanced Renal Cell Carcinoma

Abstract

On December 14, 2023, the United States Food and Drug Administration (FDA) approved belzutifan (Welireg, Merck & Co., Inc.) for patients with advanced renal cell carcinoma (RCC) following a programmed death receptor-1 or programmed death-ligand 1 (PD-1/PD-L1) inhibitor and a vascular endothelial growth factor tyrosine kinase inhibitor (VEGF-TKI). FDA granted traditional approval based on LITESPARK-005 (NCT04195750), an open-label, randomized, head-to-head trial of 746 patients with advanced RCC that progressed following both a PD-1/PD-L1 inhibitor and a VEGF-TKI. Patients were randomized (1:1) to receive belzutifan or everolimus. The primary endpoints were progression-free survival (PFS) assessed by blinded independent central review (BICR) and overall survival (OS). A statistically significant improvement in PFS was demonstrated for belzutifan compared with everolimus [hazard ratio (HR)=0.75 (95% CI: 0.63, 0.90); 1-sided p-value=0.0008]. Kaplan-Meier curves reflected non-proportional hazards with similar median PFS estimates of 5.6 months (95% CI: 3.9, 7.0) in the belzutifan arm and 5.6 months (95% CI: 4.8, 5.8) in the everolimus arm. While not reaching full maturity, OS results appeared to show a favorable trend in the belzutifan arm compared to everolimus [HR=0.88 (95% CI: 0.73, 1.07)]. The confirmed objective response rate by BICR was 22% and 3.6% in belzutifan and everolimus arms, respectively. Observed toxicities differed between treatment arms, but drug discontinuations and interruptions due to treatment-emergent adverse events were lower on the belzutifan arm compared to the everolimus arm, and a descriptive analysis of patient-reported symptom and functional outcomes was suggestive of favorable tolerability for belzutifan compared to everolimus.

Introduction

The current standard of care for patients with metastatic renal cell carcinoma (RCC) in the frontline setting includes programmed death receptor-1 or programmed death-ligand 1 (PD-1/PD-L1)-containing combination therapy (1-5). Vascular endothelial growth factor tyrosine kinase inhibitors (VEGF-TKI) are approved both as monotherapy or in combinations in front-line and later settings (6).

Belzutifan (Welireg, Merck & Co., Inc.) is a hypoxia inducible factor-2α inhibitor initially approved by the United States Food and Drug Administration (FDA) in August 2021 for treatment of adult patients with von Hippel-Lindau (VHL) disease who require therapy for associated RCC, central nervous system hemangioblastomas, or pancreatic neuroendocrine tumors not requiring immediate surgery (7). On December 14, 2023, the FDA approved belzutifan for patients with advanced RCC following a PD-1/PD-L1 checkpoint inhibitor and a VEGF-TKI (8). FDA granted this approval based on the results of the randomized LITESPARK-005 trial (NCT04195750) comparing belzutifan to everolimus in the indicated population. The FDA review team considered the statistically significant progression-free survival (PFS) improvement in a head-to-head trial supported by other efficacy endpoints and a favorable safety and tolerability profile compared to everolimus to be clinically meaningful. This article summarizes the FDA’s review and risk/benefit assessment in support of approval of belzutifan for this indication.

Study design

LITESPARK-005 was an open-label, randomized, controlled trial comparing the efficacy and safety of belzutifan and everolimus in patients with unresectable locally advanced or metastatic clear cell RCC with progression following both a PD-1/PD-L1 checkpoint inhibitor and VEGF-targeted therapy (TKI or monoclonal antibodies). Patients were randomized 1:1 to receive either belzutifan (120 mg, once daily) or everolimus (10 mg, once daily), with randomization stratified by the International mRCC Database Consortium (IMDC) prognostic scores (0 vs 1-2 vs 3-6) and number of prior VEGF-targeted therapies for advanced RCC (1 vs 2-3). Restaging scans were performed on Week 9 Day 1, then every 8 weeks thereafter for the first 49 weeks, and then every 12 weeks thereafter. Treatment was continued until radiographic disease progression per response evaluation criteria in solid tumors (RECIST) v1.1 or unacceptable toxicity.

The dual primary endpoints included PFS assessed by BICR per RECIST 1.1 and OS. A key secondary endpoint was confirmed overall response rate (ORR) by BICR per RECIST v1.1. Efficacy analyses were planned at three time points:

(i) interim analysis one (IA1) when the interim PFS analysis (567 events), the first interim OS analysis, and the final ORR analysis would be performed;

(ii) interim analysis two (IA2) when the final PFS analysis (626 events) and the second interim OS analysis would be performed; and (iii) the final OS analysis (483 events).

The study would be considered successful if either PFS or OS (or both) was statistically significant at any IA or final analysis. To control an overall type I error rate (α) at the 1-sided 0.025 level, the graphical method and group sequential design for multiple analyses were used. Initial α assigned to OS, PFS, and ORR was 0.019, 0.005, and 0.001, respectively. Besides the planned analyses, FDA performed sensitivity analyses to evaluate the robustness of the efficacy findings.

Patient characteristics

In LITESPARK-005, 374 patients were assigned to the belzutifan arm and 372 patients were assigned to the everolimus arm. Overall, the baseline characteristics were well-balanced between the treatment arms. Key demographics and baseline disease characteristics of patients enrolled in LITESPARK-005 are presented in Supplementary Table S1. Twenty percent of the patients were from North America. Distribution of race and ethnicity was as follows: 79% White; 12% Asian; 1% Black or African American; 11% Hispanic or Latino. Half of the patients had received only 1 prior VEGF targeted therapy. Twelve percent of patients in the belzutifan arm and 14% in the everolimus arm had 1 prior line of therapy (i.e. received anti-PD-1/PD-L1 and anti-VEGF therapies in combination rather than sequentially).

Efficacy results

A statistically significant improvement in PFS per BICR assessment was demonstrated for belzutifan compared to everolimus, with a HR of 0.75 (95% CI: 0.63 to 0.90; and 1-sided p-value:0.0008) at IA1, considered the final PFS analysis. Median PFS was 5.6 months (95% CI: 3.9 to 7.0) for belzutifan and 5.6 months (95% CI: 4.8 to 5.8) for everolimus. The Kaplan-Meier (KM) curves of PFS per BICR crossed around 6 months and then started to separate, indicating the presence of non-proportional hazards (Figure 1).

Figure 1. Kaplan-Meier Plot of Progression-Free Survival in LITESPARK-005.

(Adapted from the FDA's multi-disciplinary review; there are no restrictions on its use.)

FDA conducted an exploratory analysis using restricted mean survival time (RMST) to further quantify the magnitude of PFS improvement. The mean difference of the RMST of PFS between the two treatment arms was 2.2 months using the smallest value among the largest observed times across the two treatment arms (22.5 months).

There was a slightly higher rate of censoring at randomization prior to treatment in the everolimus arm [n=10 (2.7%)] compared to belzutifan arm (n=0). Additionally, 21 patients (5.6%) in the belzutifan arm and 68 patients (18.3%) in the everolimus arm were censored due to initiation of new anticancer therapy prior to disease progression by BICR.

To further evaluate the effect of imbalanced censoring at randomization prior to treatment on PFS analysis, FDA performed the following conservative and worst-case scenarios sensitivity analyses of PFS per BICR assessment:

1. a tipping point analysis which quantified the increase or decrease in the risk of event for the 10 censored patients on the everolimus arm, generally used to assess the point at which statistical significance would be broken (result: 95% lower risk of progression would be required in these patients compared to the remainder of the arm, which is unlikely to occur and indicates robustness of the primary PFS analysis),

2. considered each of the 10 censored patients on the everolimus arm to have a PFS time derived from a random sample of the patients in the top quintile for PFS of the everolimus arm (result: HR 0.78; 95% CI: 0.60 to 1.00), and

3. a worst case scenario that assumed that the 10 censored patients were alive and did not progress until the date of data cutoff (November 1, 2022) or death, whichever occurred first (result: HR 0.78; 95% CI: 0.60 to 1.01).

To understand the impact of new anticancer therapy on PFS per BICR assessment, the FDA performed sensitivity analysis of PFS by considering new anti-cancer therapy in the belzutifan arm to reflect progression event at the start date of that therapy (HR 0.81; 95% CI: 0.63 to 1.05).

OS results were immature with 91% of total planned OS events (59% deaths in the randomized population) at IA2. The OS at IA2 was not statistically significant [HR=0.88 (95% CI: 0.73 to 1.07), and 1-sided p-value of 0.0994 which did not cross the efficacy boundary of 0.014] with medians of 21.4 (95% CI: 18.2 to 24.3) months in belzutifan arm and 18.1 (95% CI: 15.8 to 21.7) months in the everolimus arm.

A statistically significant improvement in confirmed ORR assessed by BICR in patients with measureable disease at baseline was observed in the belzutifan arm compared to the everolimus arm (22.0% vs 3.6%, P-value <0.0001) (Table 1).

Table 1:

Efficacy Results in LITESPARK-005

Efficacy Outcome Measure	Belzutifan N=374	Everolimus N=372
Progression-Free Survival
Number of events, n (%)	257 (69%)	262 (70%)
Progressive disease	234 (63%)	222 (60%)
Death	23 (6%)	40 (11%)
Median in months (95% CI) *	5.6 (3.9, 7.0)	5.6 (4.8, 5.8)
Hazard ratio † (95% CI)	0.75 (0.63, 0.90)
p-Value ‡	0.0008
Overall Survival
Number of events, n (%)	213 (57.0%)	228 (61.3%)
Median in months (95% CI) *	21.4 (18.2, 24.3)	18.1 (15.8, 21.8)
Hazard ratio † (95% CI)	0.88 (0.73, 1.07)
p-Value ‡‡	0.0994 (not significant)
Confirmed Objective Response Rate
Number of patients with measurable disease at baseline	373	364
ORR % (n) (95% CI)	22% (82) (18, 27)	4% (13) (2, 6)
Complete response	3% (10)	0% (0)
Partial response	19% (72)	4% (13)
p-Value§	<0.0001
Duration of Response
Median in months (range)	NR (1.7+ to 23.2+)	17.2 (3.8 to 18.0+)
Time to Response
Median in months (range)	3.7 (range 1.7 to 16.6)	3.7 months (range 1.8 to 5.4)

Source: FDA analysis

^*From product-limit (Kaplan-Meier) method for censored data

^†Based on the stratified Cox proportional hazard model.

^‡One-sided p-value based on stratified log-rank test compared with the significance boundary of 0.0021.

^‡‡One-sided p-value based on stratified log-rank test compared with the significance boundary of 0.014.

^§One-sided p-value based on stratified Miettinen and Nurminen (M&N) method.

Safety results

All-grade treatment-emergent adverse events (TEAEs), Grade 3-5 TEAEs, treatment emergent serious adverse events, and dose reductions due to TEAEs were comparable between belzutifan and everolimus arms; drug discontinuations and interruptions due to TEAEs were lower on the belzutifan arm compared to the everolimus arm (Table 2). The specific observed toxicities differed between treatment arms. For example, the Grade 3-5 TEAEs that occurred in ≥5% of patients in the belzutifan arm were anemia and hypoxia compared to anemia, pneumonia, fatigue, COVID-19, hyperglycemia, and hypertriglyceridemia in the everolimus arm. Dose reductions in the belzutifan arm were most commonly due to hypoxia and anemia, compared to stomatitis, anemia, pneumonitis, fatigue, and hyperglycemia for everolimus.

Table 2:

Summary of Safety for LITESPARK-005

Safety Event	Belzutifan N=372 n (%)	Everolimus N=360 n (%)
All-Grade TEAEs	368 (99)	357 (99)
Grade 3-4 TEAEs	202 (54)	203 (56)
Grade 3-5 TEAEs	214 (58)	221 (61)
Grade 5 (Deaths due to TEAEs)	12 (3.2)	18 (5)
Serious TEAEs	143 (38)	134 (37)
Drug interrupted due to AEs	145 (39)	173 (48)
Drug dose reduced due to AEs	49 (13)	51 (14)
Drug discontinued due to AEs	21 (6)	52 (14)

Source: FDA analysis. TEAE: treatment-emergent adverse events

Patient reported outcomes

Patient-reported outcomes (PROs) were assessed for patients on LITESPARK-005 using three measures (FKSI-DRS, EORTC QLQ-C30, and EQ-5D-5L) assessed at the following timepoints: Day 1 of weeks 1, 3, 5, 9, every 4 weeks afterwards, at the treatment discontinuation visit, and at the 30-day post-treatment safety follow-up visit. PRO data were high quality, meaning a high proportion of the patients who were eligible for PRO assessment responded (e.g., compliance rate was 92% at week 17). In LITESPARK-005, a change from baseline to week 17 timepoint was selected as a primary PRO analysis timepoint of interest, however all PRO analyses were considered exploratory as they were not prespecified or multiplicity adjusted. FDA focused it’s PRO analysis on expected symptomatic toxicity and potential impact on functional outcomes to inform tolerability. Instead of time to event PRO endpoints, FDA focused on mean change from baseline results at specific timepoints and overall.

At week 17, compared to baseline, patients in both arms experienced a worsening in physical and role functioning domains from the EORTC QLQ-C30. In general, there was greater detriment in patients treated with everolimus compared to belzutifan at week 17 relative to baseline, however confidence invervals did overlap. In terms of patient-reported symptoms, most symptom scales from the EORTC QLQ-C30 also demonstrated similar or worsening symptoms for patients treated with everolimus compared to patients treated with belzutifan. Appetite loss was markedly different between arms, favoring belzutifan. Despite higher incidence of clinician-reported hypoxia for belzutifan, patients treated with belzutifan did not report increased dyspnea compared to those who received everolimus. Furthermore, patients on the everolimus arm were more likely to report any level of dyspnea (particularly at week 9, 13, and 17 assessment timepoints). A similar result was seen for FKSI-DRS item B1 (dyspnea), which supports that patient reported dyspnea did not appear to increase on the belzutifan arm compared to everolimus arm.

Regulatory Insights

Interpretation of the PFS results in LITESPARK-005 was challenging due to the presence of non-proportional hazards leading to an essentially equal median PFS in both arms, as well as imbalanced censoring that raised concern for potential bias in the estimate of treatment benefit. Sensitivity analyses of PFS based on worst case scenarios demonstrated consistent results to the primary PFS outcome, supporting their robustness, and a tipping point analysis suggested that the censored patients would need an unrealistically decreased risk of progression for statistical significance to be lost.

FDA noted several strengths of the submitted PRO data, including sustained high compliance rates and measurement of important treatment-related symptoms such as dyspnea. Limitations on interpreting the PRO results included the incomplete assessment of patient-reported side effects, sparse PRO assessment frequency, and no formal multiplicty-adjusted comparative tolerability endpoint. Although the selected PRO measures covered most of the expected treatment related symptoms with belzutifan and everolimus, use of a PRO item library to assess additional symptoms, such as edema and headache, would have provided a more comprehensive assessment of tolerability. Nonetheless, the PRO findings complemented the clinician-reported safety data and dose modification information supporting relatively better tolerability of belzutifan compared to everolimus.

LITESPARK-005 was a head-to-head comparison against an FDA approved therapy. The magnitude of PFS advantage seen with belzutifan over everolimus was incremental and would not have been considered meaningful in a placebo-controlled trial. However, in this head-to-head “replacement” design, FDA considers the superior PFS benefit to be additive to the effect of everolimus, a product that has already demonstrated substantial evidence of efficacy in this disease. Additional evidence supporting a favorable risk:benefit included secondary efficacy endpoints demonstrating an increased ORR, and PRO and clinician-reported safety data suggesting favorable tolerability of belzutifan compared to everolimus. A statistically significant OS improvement was not demonstrated at IA2. Nonetheless, the analysis was sufficiently mature and the OS trend appeared favorable, making a detriment extremely unlikely at the final analysis.

Only a small proportion of enrolled patients in LITESPARK-005 (15%) had received a PD-1/PD-L1 inhibitor in combination with VEGF-TKI and no subsequent line of therapy (i.e. patients with only 1 prior line of therapy). FDA was concerned about applicability of overall LITESPARK-005 trial results to the post-approval population, many of whom will receive the combination of PD-1/PD-L1 inhibitor and VEGF-TKI concurrently (rather than sequentially) in the 1st line for advanced RCC, as practice patterns change and as combination therapy with these drugs is now an accepted standard of care in the first line. However, exploratory subgroup analyses in LITESPARK-005 showed that the population of patients with only 1 prior line of therapy demonstrated consistency of results in terms of PFS and OS with the overall trial population.

A substantial proportion (32%) of patients treated with beluzutifan received erythropoiesis stimulating agents (ESA) for anemia. As there have been prior reports suggesting increased cancer mortality and/or tumor progression in patients with various solid tumors receiving ESAs (9-12), FDA conducted exploratory analyses on the effect of ESA use on OS and PFS. No detriment in PFS or OS was apparent in patients using ESAs. Therefore, while the FDA previously included a warning in product labeling cautioning against use of ESAs for patients with VHL disease, a similar warning was not added for patients with advanced RCC treated with belzutifan. The previous warning against use of ESAs concurrently with belzutifan in VHL disease was not modified given the much earlier disease setting and absence of randomized safety data to provide similar reassurance in the VHL disease setting.

In LITESPARK-005, hypoxia occurred in 15% of patients in the belzutifan arm (69% of whom required supplemental oxygen). Despite the higher incidence of clinician-reported hypoxia with belzutifan, review of PRO results suggested that patients did not report increased dyspnea on the belzutifan arm compared to the everolimus arm. The United States Prescribing Information (USPI) for belzutifan notes patients’ oxygen saturation should be monitored before initiation of, and periodically throughout, treatment with belzutifan, and includes dose modification recommendations for hypoxia. Additionally, hypoxia has been listed in the Warnings and Precautions section of the USPI for belzutifan.

Conclusion

The FDA review team considered the observed PFS improvement with belzutifan compared to an active control in a head-to head trial, no OS detriment, internal consistency across primary and secondary endpoints, and favorable safety and tolerability compared to everolimus to be clinically meaningful (Table 3). Approval of belzutifan for treatment of patients with advanced RCC in 2^nd or later line of treatment offers a therapy with a different mechanism of action and safety profile compared to other available systemic therapies (VEGF-TKIs and everolimus), expanding treatment options that can be individualized to a patient's preferences and comorbidities.

Table 3:

Benefit-Risk Assessment

Dimension	Evidence and Uncertainties	Conclusions and Reasons
Analysis of Condition	The 5-year overall survival (OS) of patients with advanced renal cell carcinoma (RCC) is 15%.	Advanced RCC is a serious and life-threatening condition.
Current Treatment Options	Available treatment options for patients with advanced RCC after progression on prior PD-1/PD-L1 checkpoint inhibitor and VEGF-TKI therapy include axitinib, cabozantinib, everolimus, everolimus plus lenvatinib, and tivozanib.	Although treatment options exist, none is curative and none is approved specifically for patients whose disease has progressed after prior PD-1/PD-L1 checkpoint inhibitor and prior VEGF-TKI therapy. There is an unmet medical need in this setting.
Benefit and Risk	Efficacy was evaluated in LITESPARK-005, an open-label, randomized, head-to-head trial of 746 patients with unresectable locally advanced or metastatic clear cell RCC that progressed following both a PD-1/ PD-L1 checkpoint inhibitor and a VEGF-TKI. Patients were randomized 1:1 to receive 120 mg belzutifan or 10 mg everolimus once daily. Randomization was stratified by IMDC risk category and number of prior VEGF-TKIs. A statistically significant improvement in PFS was demonstrated for belzutifan compared with everolimus, with a hazard ratio of 0.75 [(95% CI: 0.63, 0.90); 1-sided p-value=0.0008]. Kaplan-Meier curves reflected non-proportional hazards with similar median PFS estimates of 5.6 months in both arms. While OS results were immature at the current analysis, with 59% of deaths reported, no trend towards an OS detriment was observed. Belzutifan demonstrated a safety profile consistent with its previously-known safety profile with no new safety signals identified. Percentages of all-grade TEAEs, Grade 3-5 TEAEs, serious TEAEs, and dose reductions were similar between arms but patients receiving belzutifan had fewer dose interruptions and drug discontinuations due to TEAEs. Higher ORR and longer DoR, favorable safety results, and descriptive PRO data in the belzutifan arm also supported a favorable benefit-risk profile of belzutifan in this setting.	Compared to everolimus, belzutifan demonstrated a PFS improvement in a head-to head trial, with no OS detriment, a favorable safety profile, and a descriptive improvement in PROs. The FDA considered this favorable benefit-risk profile for belzutifan to represent a clinically meaningful improvement over everolimus in this setting.

Source: FDA’s multi-disciplinary review