Real-world efficacy and safety of pembrolizumab plus lenvatinib in patients with metastatic renal cell carcinoma: a multi-institutional retrospective study

Abstract

Despite the potential of pembrolizumab and lenvatinib combination therapy in treating metastatic renal cell carcinoma (mRCC), real-world evidence on its efficacy and safety remains limited. We retrospectively analyzed data from 118 patients with mRCC who initiated pembrolizumab plus lenvatinib as first-line therapy between August 2022 and March 2024. Patient characteristics, treatment outcomes, and adverse events were evaluated. Oncological outcomes were stratified following the International Metastatic RCC Database Consortium risk classification. The objective response and disease control rates were 69.5% and 98.5%, respectively. Median progression-free survival (PFS) was 26.8 months (95% confidence interval (CI), 23.5–27.8)), whereas the median overall survival (OS) was not reached (95%CI, 20.5– not reached). Grade ≥ 3 treatment-related adverse events occurred in 39.8% of all patients. Clinical outcomes were also compared between patients aged ≥ 75 and < 75 years. Median age was 70 years; 70.3% and 29.7% aged < 75 and ≥ 75 years, respectively. No significant differences were observed in PFS and OS between the two groups (p = 0.14 and p = 0.44, respectively). Pembrolizumab plus lenvatinib demonstrated favorable efficacy and manageable safety in a real-world cohort of patients with mRCC, including older adults.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-27578-6.

Introduction

Renal cell carcinoma (RCC) is the seventh most common cancer globally, accounting for approximately 2–3% of all adult malignancies and leading to over 170,000 deaths annually¹. At the time of initial diagnosis, approximately 30% of patients present with distant metastases, necessitating systemic therapy^2,3. Owing to the distinct hypervascular characteristics of RCC tissues, tyrosine kinase inhibitors have played a central role in metastatic RCC (mRCC) treatment over the past two decades⁴. In addition, immune checkpoint inhibitors (ICIs) such as antibodies targeting programmed cell death protein-1 (PD-1) or the PD-1 ligand 1 (PD-L1), which block immunosuppressive PD-1/PD-L1 signaling pathways, have become standard therapeutic options in the management of mRCC⁵.

ICI-based combination therapies have been recently developed, significantly improving clinical outcomes for patients with mRCC, as evidenced in several clinical trials^6–9. Among these, the combination of pembrolizumab and lenvatinib has shown particularly durable antitumor activity, with an objective response observed in 71.3% of patients^10,11. Updated data from the second interim analysis of progression-free survival (PFS), with a median follow-up of 49.8 months, revealed that pembrolizumab plus lenvatinib significantly improved PFS and nearly doubled the objective response rate (ORR) compared to sunitinib. However, real-world evidence regarding the clinical efficacy of this combination therapy remains limited.

In the present study, we aimed to evaluate the real-world efficacy and toxicity of pembrolizumab plus lenvatinib in patients with mRCC across all risk categories as defined by the International Metastatic RCC Database Consortium (IMDC). Clinical outcomes were also compared between patients aged < 75 and ≥ 75 years since individuals aged ≥ 75 years are classified as late-stage elderly in Japan.

Results

Baseline characteristics

The clinical characteristics of all patients are summarized in Table 1. The median patient age was 70 years (range: 35–86 years). The most prevalent histological subtype was clear cell carcinoma, observed in 106 patients (89.8%). The number of metastatic sites ranged from one to six, with 50 (42.4%) patients presenting with multiple metastatic sites (≥ 2 metastatic lesion) at diagnosis. The most frequent site of distant metastasis was the lungs (63.6%), followed by the lymph nodes (28.8%). Based on the IMDC risk classification, 22.9% of patients were categorized as favorable risk, 50.8% as intermediate risk, and 26.3% as poor risk. Prior to study enrollment, 76 patients (64.4%) underwent nephrectomy, and 37 patients (31.4%) underwent tissue biopsies. The most common treatment administered immediately after pembrolizumab plus lenvatinib was cabozantinib (9.3%).

Table 1.

Patient characteristics of patients at baseline.

	All patients
Number of patients	118
Sex
Male	83 (70.3)
Female	35 (29.7)
Age	Median 70 (35–86)
< 75	83 (70.3)
≥ 75	35 (29.7)
KPS
80–100	109 (92.4)
< 80	9 (7.6)
Number of metastatic sites
Single	68 (57.6)
Multiple	50 (42.4)
Metastatic site
Lung	75 (63.6)
Liver	13 (11.0)
Bone	19 (16.1)
Brain	3 (2.5)
Lymph node	34 (28.8)
Adrenal	11 (9.3)
Tissue type
Clear	106 (89.8)
Non-clear	12 (10.2)
IMDC risk
Favorable	27 (22.9)
Intermediate	60 (50.8)
Poor	31 (26.3)
Confirmation of tissue type
Nephrectomy	76 (64.4)
Biopsy	37 (31.4)
N/A	5 (4.2)
NLR
Low (< 4)	58 (49.2)
High (≥ 4)	60 (50.8)
CRP
Low (< 1.0)	73 (61.9)
High (≥ 1.0)	45 (38.1)

CRP; C-reactive protein, IMDC; International Metastatic RCC Database Consortium, KPS; Karnofsky performance status, N/A; not availavle, NLR; Neutrophil-lymphocyte ratio.

NLR was calculated as a ratio between the neutrophil and lymphocyte counts measured in peripheral blood. The cut-off values of NLR and CRP depending on previous reports^36–38.

Clinical outcomes of pembrolizumab plus lenvatinib

The median follow-up duration was 16.4 months (range: 5.3–33.8). At data cutoff, 57 (48.3%) and 66 (55.9%) patients discontinued pembrolizumab and lenvatinib therapy, respectively, mainly due to the AE (29.7%) and disease progression (14.4%). During the observation period, cancer-related mortality occurred in 15 (12.7%) patients. The median PFS was 26.8 months (95% confidence interval (CI), 23.5–27.8), whereas the median OS was not reached (NR) (95%CI, 20.5–NR, Fig. 1). Of the 118 patients, 13 achieved CR (11.0%), 69 achieved PR (58.5%), 31 achieved SD (26.3%), and 5 achieved PD (4.2%), resulting in an ORR and disease control rate (DCR) of 69.5% and 95.8%, respectively (Table 2).

Fig. 1.

Kaplan–Meier survival curve for all patients with mRCC treated with pembrolizumab plus lenvatinib. (A) Progression-free survival (PFS) and (B) overall survival (OS). NR; not reached.

Table 2.

Association of overall response rate with IMDC favorable-, intermediate-, and poor-risk patients.

	All (n = 118)	Favorable (n = 27)	Intermediate (n = 60)	Poor (n = 31)
Best response, n (%)
CR	13 (11.0)	8 (29.6)	4 (6.7)	1 (3.2)
PR	69 (58.5)	11 (40.7)	39 (66.7)	19 (58.1)
SD	31 (26.3)	6 (22.2)	16 (25.0)	9 (32.3)
PD	5 (4.2)	2 (7.4)	1 (1.7)	2 (6.5)
ORR
n (%)	82 (69.5)	19 (70.4)	43 (71.7)	20 (64.5)
DCR
n (%)	113 (95.8)	25 (92.6)	59 (98.3)	29 (93.5)

DCR; disease control rate, ORR; objective response rate.

The ORRs stratified by IMDC risk were: favorable 0 (70.4%), intermediate (71.7%), and poor (64.5%), respectively. When we evaluated the ORRs by IMDC score, the ORR with score 1, 2, 3, and 4–6, were 71.3%, 73.1%, 65.0%, and 53.8%, respectively (Fig. 2A). PFS and OS in each IMDC risk group was shown in Fig. 2B and C. While there was no statistically significant difference in ORR between the intermediate- and poor-risk groups (p = 0.226), patients with favorable/intermediate-risk showed significantly longer PFS (hazard ratio (HR), 0.344; 95% CI, 0.137–0.862, p = 0.023, Fig. 3A) and OS (HR, 0.0809; 95% CI, 0.0264–0.247, p < 0.0001, Fig. 3B) than those with poor risk. Moreover, median CRP levels were significantly lower in the IMDC favorable/intermediate-risk than in the poor- risk groups (Fig. 3C). Although patients in the favorable-risk group did not exhibit a significantly longer PFS compared with those in the intermediate- or poor-risk groups (HR, 0.671; 95% CI, 0.262–1.718; p = 0.406; Supplementary Fig. S1A), they demonstrated a significantly longer OS (HR, 0.271; 95% CI, 0.0851–0.856; p = 0.027; Supplementary Fig. S1B).

Fig. 2.

Survival analysis for patients according to the International Metastatic RCC Database Consortium (IMDC) risk classification. (A) Objective response rates stratified by the number of IMDC risk factors. (B) Progression-free survival (PFS) by IMDC risk group: favorable- (n = 27), intermediate- (n = 60), and poor-risk (n = 31) groups. (C) Overall survival (OS) by IMDC risk group. (D) Comparative analysis of baseline C-reactive protein levels between IMDC poor- and favorable/intermediate risk-groups.

Fig. 3.

Survival analysis for patients with International Metastatic RCC Database Consortium (IMDC) favorable/intermediate- and those with poor- risk. Kaplan-Meier curve of (A) progression-free survival (PFS) and (B) overall survival (OS) in patients with IMDC favorable/intermediate- (n = 87) and poor-risk (n = 31) groups. (C) Comparative analysis of baseline C-reactive protein levels between IMDC poor- and favorable/intermediate risk-groups.

In this cohort, 83 (70.3%) and 35 (29.7%) patients were aged < 75 and ≥ 75 years, respectively (Supplementary Table S1). ORRs were 71.4% in the < 75-year group and 62.7% in the ≥ 75-year group, with CR rates of 14.3% and 9.6%, respectively. However, no significant differences were observed in PFS (HR, 0.553; 95% CI, 0.228–1.45; p = 0.14) or OS (HR, 1.032; 95% CI, 0.361–2.952; p = 0.44) between the two age groups (Fig. 4).

Fig. 4.

Survival analysis by age group. (A) Progression-free survival (PFS) and (B) overall survival (OS) in patients aged < 75 (n = 83) and aged ≥ 75 (n = 35). NR; not reached.

AEs leading to pembrolizumab discontinuation occurred in 44.6% of patients aged < 75 years and 57.1% of patients aged ≥ 75 years. For lenvatinib, discontinuation occurred in 48.1% and 74.3% of patients aged < 75 and ≥ 75 years, respectively. The frequency of AEs leading to discontinuation of pembrolizumab or lenvatinib was not significantly different between the two groups (p = 0.785, Supplementary Table S1).

Influence of upfront cytoreductive nephrectomy for synchronous mRCC

Next, we assessed the efficacy of nephrectomy in patients with synchronous metastatic disease (M1) at mRCC diagnosis who subsequently received pembrolizumab plus lenvatinib. Among 61 patients (51.7%) with M1 disease, 19 patients (31.1%) underwent cytoreductive nephrectomy (CN). To align the clinical background, patients with M1 disease were divided into the IMDC intermediate- and poor-risk groups, and PFS and OS were evaluated. As a result, there was no statistically significant difference of PFS (HR, 0.764; 95% CI, 0.148–3.953; p = 0.748, HR, 1.355; 95% CI, 0.221–8.313; p = 0.743, respectively) and OS between patients with and without CN in both IMDC groups (HR, 1.871; 95% CI, 0.096–10.370; p = 0.679, HR, 1.419; 95% CI, 0.240–8.397; p = 0.700) (Supplementary Fig. S2 and S3).

Safety analysis

Treatment related AEs are summarized in Table 3. Overall, treatment-related AEs occurred in 84.7% of patients, with hypertension (39.8%), hypothyroidism (22.9%), and Palmer-plantar erythrodysesthesia syndrome (18.6%) being the most common AEs. Grade ≥ 3 AEs occurred in 39.8% of patients, with the most frequent being hypertension (8.5%), followed by proteinuria (6.8%) and diarrhea (6.8%). Eighteen (18.0%) patients required steroid treatment for immune-related AEs.

Table 3.

Summary of AEs in mRCC patients with pembrolizumab plus lenvatinib therapy.

	Number of patients (n = 118)
	Any Grades, n (%)	Grade3-4, n (%)
Any event	100 (84.7%)	47 (39.8%)
Diarrhea	13 (11.0)	6 (5.1)
Hypertension	47 (39.8)	10 (8.5)
Hypothyroidism	27 (22.9)	0 (0)
Decreased appetite	6 (5.1)	2 (1.7)
Fatigue	20 (16.9)	5 (4.2)
Dysphonia	1 (0.8)	1 (0.8)
Proteinuria	20 (16.9)	8 (6.8)
PPES	22 (18.6)	3 (2.5)
Arthralgia	4 (3.4)	0 (0)
Rash	13 (11.0)	1 (0.8)
Vomiting	1 (0.8)	0 (0)
Dysgeusia	2 (1.7)	0 (0)
Hyponatremia	3 (2.5)	2 (1.7)
Myasthenia gravis	1 (0.8)	1 (0.8)
Hepatotoxicity	10 (8.5)	3 (2.5)
Interstitial pneumonia	4 (3.4)	2 (1.7)
Nephrotoxicity	7 (5.9)	0 (0)
Thrombocytopenia	9 (7.6)	1 (0.8)
Serum amylase increased	2 (1.7)	0 (0)
Anemia	1 (0.8)	0 (0)
Adrenal failure	3 (2.5)	1 (0.8)
Type1 DM	1 (0.8)	1 (0.8)
Arthritis	1 (0.8)	0 (0)

DM; Diabetes Mellitus, PPES; Palmer-plantar erythrodysesthesia syndrome.

Discussion

The therapeutic landscape for mRCC has undergone considerable transformation with the advent of ICIs. Over the past seven years, five ICI-based combination therapies have been approved and are now preferred as first-line treatments for advanced clear cell RCC (ccRCC) by the National Comprehensive Cancer Network¹². Among them, the combination of pembrolizumab and lenvatinib showed substantial efficacy in patients with ccRCC as well as in non-ccRCC patients, showing consistent benefit across various metastatic sites, including the lungs, lymph nodes, liver, and bones^13–15. However, real-world evidence regarding this combination therapy remains limited, highlighting the need for data that can guide treatment decisions in clinical practice. Accordingly, in this study, we aimed to evaluate the real-world efficacy and safety of pembrolizumab plus lenvatinib in patients with mRCC to explore factors influencing clinical outcomes.

First, our findings confirmed favorable clinical efficacy of pembrolizumab plus lenvatinib, as reflected by a high ORR of 69.5% and prolonged PFS and OS (Fig. 1; Table 2). Lenvatinib is a multi-target TKI that selectively inhibits several receptor pathways, including vascular endothelial growth factor receptor (VEGFR), fibroblast growth factor receptor (FGFR), platelet-derived growth factor receptor α (PDGFR α), RET, and KIT molecules¹⁶. Our findings are consistent with those observed in the pivotal CLEAR trial (including a subgroup analysis of patients enrolled in Japan), although the CR rate in the present study was lower than that in previous reports (11.0% vs. 16.1–19.0%)^10,14. We speculated that this discrepancy may be attributed to a higher proportion of patients with IMDC poor-risk status (26.3%) in our cohort, who likely had higher baseline tumor burden than those in the favorable- or intermediate-risk groups¹⁷. In addition, baseline CRP levels were significantly elevated in the poor-risk group, which may reflect a more immunosuppressive tumor microenvironment and systemic inflammation that impair ICI efficacy (Fig. 2D)^18,19.

Second, to the best of our knowledge, this is the first study to demonstrate comparable efficacy of pembrolizumab plus lenvatinib in patients aged ≥ 75 and < 75 years in terms of PFS and OS (Fig. 4). Despite growing use of ICIs in mRCC, data on their safety and efficacy in elderly patients remain limited in real-world settings. Several clinical trials reported PFS and OS for the subgroup aged ≥ 65 years^7,9,10,20,21, whereas three trials performed the subgroup analysis among patients aged ≥ 75 years^7,20,22 since geriatric oncology guidelines recommend patients ≥ 75 years should routinely undergo comprehensive geriatric assessment before treatment decisions²³. Additionally, clinical trial participants tend to be younger than the general patient population by a median age of 64.9 years²⁴. Importantly, in our study, discontinuation rates due to AEs were not significantly different between the two age groups (≥ 75 and < 75 years), suggesting that this ICI-based combination therapy is feasible and tolerable in elderly patients. Nonetheless, prospective studies are warranted to better define the risk-benefit profile of pembrolizumab and lenvatinib in elderly patients with mRCC.

Third, in patients with synchronous metastatic disease at diagnosis, we observed there was no significant difference between patients with and without CN in terms of PFS and OS when patients were stratified into the IMDC intermediate- and poor-risk groups for evaluation (Supplementary Fig. S2 and 3). The relevance of upfront CN in mRCC remains controversial^25,26. Bakouny et al. reported that upfront CN was associated with significantly better OS in ICI-treated patients (HR,0.72; 95%CI, 0.67–0.78; p < 0.001) using IMDC database²⁷. Singla et al. also showed improved OS in patients receiving CN before ICI therapy compared to those who did not undergo CN²⁸. Previous reports demonstrated that primary RCC tumors secrete pro-inflammatory cytokines that drive inflammation and suppress T-cell activity, potentially limiting the effectiveness of systemic immunotherapy^29,30. On the other hand, Li et al. reported that the deferred CN was observed to be correlated with superior OS compared to the upfront CN in patients with ICI-based combination subgroup analysis³¹. However, these findings are retrospective, observational, and subject to selection bias, which underscores the need to conduct randomized controlled trials with ICI-based regimens.

Finally, we found that the percentage of grade ≥ 3 AEs was lower in our study population when compared to that in general population treated with pembrolizumab plus lenvatinib (39.8% vs. 82.4%). As an additional example, in patients with avelumab plus axtinib therapy, grade ≥ 3 AEs occurred in 17.1% of Japanese patients, whereas 71.2% of global population experienced grade ≥ 3 AEs^6,32. Differences in the incidence and severity of drug-related AEs between Asian and global (primarily Western) populations may arise from a combination of genetic, physiological, environmental, and clinical factors^33,34. On the other hand, Asian populations frequently experience higher rates of hematologic or hepatic toxicities but sometimes lower rates of gastrointestinal or fatigue-related events³³. These differences highlight the importance of intense monitoring while receiving treatment.

This study has some limitations. First, the retrospective and observational study design with a small sample size introduces potential selection bias. Second, data were limited to those available in medical records, which may have resulted in an underestimation of disease burden and treatment-related AEs. Third, this report is based on a Japanese population, and therefore these findings cannot be directly generalized to all racial or ethnic groups. Fourth, short median follow-up might underscore the incidence of AEs and clinical prognosis. Finally, due to the low number of events such as disease progression or death, we were unable to perform multivariate analyses of PFS and OS.

Despite these limitations, to the best of our knowledge, this is the first real-world study to demonstrate that pembrolizumab plus lenvatinib offers promising clinical outcomes in patients with mRCC. Our findings validate the efficacy and tolerability of this combination therapy across all age groups, including patients aged ≥ 75 years. These results support the use of pembrolizumab plus lenvatinib as a viable first-line treatment option for mRCC in routine clinical practice.

Materials and methods

Patients

We conducted a multicenter retrospective study involving 118 patients with mRCC who received pembrolizumab plus lenvatinib as first-line therapy between August 2022 and March 2024. For each patient, data were collected on age, Karnofsky Performance Status (KPS), sex, IMDC risk group, tumor histology, comorbidities, and sites of metastasis.

Tumor response was evaluated using imaging modalities such as computed tomography (CT) or magnetic resonance imaging (MRI), according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1, every 2–3 months. We evaluated tumor responses (complete response [CR], partial response [PR], stable disease [SD], or progressive disease [PD]), PFS, and overall survival (OS). PFS was defined as the time from initiation of pembrolizumab plus lenvatinib treatment to either documented disease progression or death from any cause. OS was defined as the time from treatment initiation to death from any cause or last follow-up. Adverse events (AEs) during treatment were collected and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE) version 5.0³⁵.

This multicenter study was conducted in accordance with the ethical standards of the Declaration of Helsinki and approved by the Institutional Review Board of each participating institution (approval number 018 − 0003 at Osaka University Hospital). The institutional review board of Osaka University Hospital approved that this research was properly conducted in an opt-out format. All methods were performed in accordance with the relevant guidelines and regulations by including a statement.

Informed consent

was obtained from all patients for data collection.

Treatment and assessment

All patients received at least one dose of pembrolizumab (200 mg intravenously every three weeks) and lenvatinib (20 mg orally once daily). Treatment was continued until disease progression, clinical deterioration, unacceptable toxicity, or patient withdrawal.

Patients continued to receive pembrolizumb or lenvatinib until disease progression in the present study.

Statistical analysis

Categorical variables were compared using Fisher’s exact test or the chi-square test. The Mann–Whitney U test (two-tailed) was　performed to evaluate differences in CRP between patients in the IMDC favorable/intermediate- and poor-risk groups. PFS and OS were estimated using the Kaplan–Meier method and compared using the log-rank test. Statistical significance was set at p < 0.05. All statistical analyses were conducted using the JMP software (version 17.0; SAS Institute, Cary, NC, USA).

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

T. K. was responsible for the study conception and design, conducted the study, collected and analyzed the data, and drafted the manuscript. Y. N., M. S., T. T., T. I., S. T., H. O., M. T., K. O., K. K., H. H., Y. I., T. H., Y. Y., K. H., A. K., Y. O., K. N., S. T., K. F., H. T., O. M., M. N., and N. N. conducted the study, collected and analyzed the data, drafted the study, and revised the manuscript. All authors have read and approved the final manuscript for submission.

Funding

The study is based on retrospective data analysis. The authors did not receive support from any organization for the submitted work. No funds, grants, or other support was received.

Data availability

The data that support the findings of this study are available from the corresponding author, TK, upon reasonable request.

Declarations

Competing interests

The other authors declare no competing interests.