Efficacy and safety of nivolumab plus ipilimumab in patients with metastatic variant histology (non-clear cell) renal cell carcinoma

Abstract

Background

Nivolumab plus ipilimumab (nivo/ipi) is a standard of care first-line (1 L) therapy for patients with metastatic clear-cell renal cell carcinoma (ccRCC), but its role in patients with metastatic, non-ccRCC has not been fully defined. We report a single-institution experience with nivo/ipi in non-ccRCC.

Methods

Between November 2017 and February 2024, 55 patients with metastatic non-ccRCC received nivo/ipi at MD Anderson Cancer Center. The tumor response was assessed by blinded radiologists using RECIST v1.1. The overall response rate (ORR), progression-free survival (PFS), PFS milestone, duration of response (DoR), and overall survival (OS) were determined. Next-generation sequencing (NGS) was performed on available tumor specimens.

Results

Twenty-five (45.5%) patients had papillary histology (pRCC), 12 (21.8%) patients had chromophobe (chRCC), and 18 (32.7%) patients had unclassified RCC (uRCC). Fifty-two (94.5%) patients received nivo/ipi in 1 L. Sarcomatoid features (SF) were found in 20 (36.4%) cases. ORR was 48% (12/25) in pRCC, 25% (3/12) in chRCC (all 3 cases had SF), 27.8% (5/18) in uRCC, and 55% (11/20) across histologies with SF.

The median PFS was 10.6 months (95% CI: 2.8 to 22.8) in pRCC, 3.6 months (95% CI: 0.9 – NE) in chRCC, and 3 months (95% CI: 2.1 to 7) in uRCC; 6-month milestone PFS was 56% (95% CI: 36.3 to 75.7), 41.7% (95% CI: 22 to 61.3), and 38.9% (95% CI: 21.7 to 56.1) in pRCC, chRCC, and uRCC, respectively. The median DoR for the entire cohort was 8.5 months (95% CI: 8 – NE). The median OS was 36.7 months (95% CI: 11.5 to 54.8) in pRCC, 25.7 months (95% CI: 0.9 – NE) in chRCC, and 11.1 months (95% CI: 6.5 – NE) in uRCC.

Ten (18.2%) patients discontinued treatment due to treatment-related adverse events (AEs). Grade 3/4 immune-mediated AEs were noted in 17 (30.9%) patients. We performed NGS on 26 cases: TP53 (42%), PTEN (23%), and TERT (23%) alterations were most frequently found, with TERT and TP53 mutations enriched in pRCC and chRCC, respectively.

Conclusion

Nivo/ipi produced favorable outcomes in patients with pRCC supporting its use as 1 L therapy. Responses in patients with chRCC were noted exclusively with SF. Despite achieving an ORR of 27.8% with nivo/ipi, patients with uRCC had short PFS and inferior OS.

Correspondence to Dr Nizar M Tannir; [email protected]

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Nivolumab and ipilimumab (nivo/ipi) in advanced variant histology renal cell carcinoma (vhRCC) is supported by limited data, with CheckMate-920 being the only Phase 3b/4 trial reporting an objective response rate (ORR) of 19.6% in 46 patients, while additional evidence comes from small retrospective cohorts.

WHAT THIS STUDY ADDS

• Our study (n=55) confirms the results of CheckMate-920 by showing that patients with treatment-naïve metastatic papillary RCC (pRCC) benefit most from nivo/ipi with an ORR of 48%, median progression-free survival (PFS) of 10.6 months, and median overall survival (OS) of 36.7 months. The study adds original insights on the role of sarcomatoid features (SF) driving higher response rates (55%) across subtypes, with all responses in chromophobe RCC (chRCC; ORR 25%) and most in unclassified RCC (uRCC; ORR 27.8%) occurring in tumors with SF. Patients with uRCC had limited PFS and inferior OS, highlighting the variable efficacy of nivo/ipi in vhRCC subtypes. Our additional NGS analysis of 26 cases reveals TP53 (42%), PTEN (23%), and TERT (23%) alterations, with TERT mutations enriched in pRCC and TP53 mutations enriched in chRCC.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY

• Our findings support the use of nivo/ipi as the first-line standard of care in pRCC and highlight the predictive value of SF for better responses across the three subtypes. They also suggest that the immunologically cold chRCC subtype, traditionally showing low response rates with other immunotherapy combinations (eg, ORR 0% (0/7) with cabozantinib plus nivolumab and 11.1% (1/9) with cabozantinib plus atezolizumab) shows evidence of response to nivo/ipi in the presence of SF.

Introduction

Variant histology renal cell carcinomas (vhRCC), more commonly referred to as non-clear cell RCC (non-ccRCC) make up approximately a quarter of all RCC but portend a worse prognosis than clear-cell RCC (ccRCC).¹ While the heterogeneity and rarity of vhRCC historically limited clinical trial enrollment, recent studies have included more patients, addressing earlier exclusions driven by poor responses to cytokine-based therapies.² Therefore, the therapy for patients with vhRCC is often adopted from clinical trial results in ccRCC. The histological subtypes of vhRCC include papillary RCC (pRCC), chromophobe RCC (chRCC), microphthalmia transcription factor (MiTF) family/translocation RCC, renal medullary carcinoma (RMC), collecting duct carcinoma (Bellini’s tumor), fumarate hydratase-deficient RCC, and unclassified RCC (uRCC).¹ Despite the initial advent of vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitors (TKIs) and the more recent development of immune checkpoint inhibitor (ICI)-based combinations in ccRCC, substantial evidence showing improved survival with ICI-containing regimens over TKIs in vhRCC is limited.^{3 4} The distinct biology of each vhRCC subtype likely drives differential therapeutic responses, underscoring the need to refine the current histology-agnostic therapeutic strategies.⁵

The CheckMate 214 trial of frontline (1 L) therapy of nivolumab plus ipilimumab (nivo/ipi) vs sunitinib was the first randomized controlled phase 3 trial to show superior OS of an ICI-based therapy compared with a VEGFR-TKI in patients with advanced ccRCC and intermediate- and poor-risk disease, as well as the intent-to-treat (ITT) population.⁶ This led to the Food and Drug Administration (FDA) approval of nivo/ipi in patients with advanced RCC. In a post-hoc analysis of CheckMate 214, nivo/ipi was associated with longer OS, longer progression-free survival (PFS), higher objective response rate (ORR), and higher complete response (CR) rate compared with sunitinib among patients with International Metastatic RCC Database Consortium (IMDC) intermediate/poor-risk disease who had sarcomatoid features (SF) in their tumors.⁷ With a median 8 year follow-up of CheckMate 214 still showing more durable responses and improved OS with nivo/ipi compared with sunitinib in ccRCC in an ITT population and among patients with IMDC intermediate- or poor-risk, the interest in investigating this regimen remains high in different vhRCC subtypes.⁸

CheckMate 920 was the first prospective trial of nivo/ipi in a cohort of treatment-naïve patients with metastatic vhRCC (n=52), demonstrating an ORR of 19.6% (95% CI: 9.4 to 33.9) seen exclusively among patients with pRCC and uRCC, with eight of nine responding patients remaining without disease progression.⁹ The median OS was 21.2 mo (95% CI: 16.6 – NE) and median PFS was 3.7 mo (95% CI: 2.7 to 4.6); toxicity events were consistent with previous reports. With limited prospective and retrospective studies of nivo/ipi in this domain,^10–13 it is largely unknown which histological subtypes would derive the most benefit from a particular therapeutic strategy, such as treatment with nivo/ipi.¹ Herein, we report a real-world experience of using nivo/ipi in patients with vhRCC treated over a span of 6 years at our institution and provide data on response rates, survival outcomes, safety, and genomic mutational analysis in patients with metastatic vhRCC.

Materials and methods

Patients

Baseline characteristics and clinical outcomes were retrospectively collected from the electronic medical records of 55 patients treated with nivo/ipi at the University of Texas MD Anderson Cancer Center (MDACC) from November 2017 to February 2024. The study was approved by the Institutional Review Board of MDACC (Protocol# PA16-0736).

Ipilimumab was administered intravenously at the dose of 1 mg/kg and nivolumab at 3 mg/kg, every 3 weeks for up to four cycles (induction phase), followed by nivolumab monotherapy at the dose of 480 mg every 4 weeks (maintenance phase).

The eligibility criteria for inclusion in this analysis included histopathologic confirmation of one of three vhRCC subtypes (pRCC, chRCC, and uRCC), documentation of metastatic disease, and receipt of at least one cycle of nivo/ipi. Other criteria included available baseline clinical and imaging data prior to the start of the treatment. Patients whose tumors were found to harbor VHL mutations by next-generation sequencing (NGS) were considered to have ccRCC and were excluded from the study.

Endpoints and statistical analysis

Primary endpoints of interest included radiological response rates and survival outcomes. The response was assessed based on at least one scan after the initiation of ipilimumab plus nivolumab and confirmed on subsequent imaging at least 6 weeks apart. Best overall response (BOR) was defined by the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Patients who succumbed to complications of their cancer or were clinically determined to have disease progression before having imaging studies for response assessment were considered to have had progressive disease (PD). ORR was defined as the percentage of patients with confirmed BOR of CR or partial response (PR) by RECIST v1.1. The clinical benefit rate (CBR) was defined as the BOR of CR, PR, or stable disease (SD) for a duration of longer than 6 months.

The OS time was measured from cycle 1 day 1 of nivo/ipi until the date of death or date of last contact with a patient known to be alive. OS times were estimated using the Kaplan–Meier method (GraphPad Prism, v10, Boston, MA, USA), and patients were censored at the time of the last follow-up if they remained alive at the time of analysis. For PFS analysis, censoring corresponded to stopping treatment for non-progression or remaining on treatment at the last date of follow-up. The median follow-up time was calculated using the reverse Kaplan–Meier method. The time on treatment (TOT) was defined as the length of time from the initiation of nivo/ipi to the final dose administered. The time to the next treatment (TTNT) was defined as the time from nivo/ipi start to the date of the start of the next treatment after nivo/ipi. The duration of response (DoR) was defined as the time from the date of documentation of major response (CR, PR) to disease progression or death and was estimated using the Kaplan–Meier method.

Numerical values were summarized using median and range, and categorical variables were summarized using frequency and percentage. Adverse events related to nivo/ipi were abstracted and graded using Common Terminology Criteria for Adverse Events v5.0 criteria based on available documentation in the medical records.

Results

Baseline characteristics of patients

Between November 2017 and February 2024, 55 patients with vhRCC were started on nivo/ipi. Their baseline characteristics are summarized in table 1. The majority of these patients (52, 94.5%) received nivo/ipi as 1 L therapy for metastatic disease. The majority of patients were male (37, 67.3%) and Caucasian (51, 92.7%). Variant histological subtypes were comprised of pRCC (n=25, 45.5%), chRCC (12, 21.8%), and uRCC (n=18, 32.7%). SF was histologically confirmed in 20 (36.4%) patients, including six with pRCC, eight with chRCC, and six with uRCC. Eighteen SF cases were confirmed in the primary tumor, while two were identified in metastatic sites (one retroperitoneal lymph node and one bone biopsy). Twenty-four patients (43.6%) had undergone nephrectomy before receiving systemic therapy with nivo/ipi. The vast majority of patients (39/55, 70.9%) had synchronous metastatic disease. Two patients (3.6%) had IMDC favorable-risk disease, 27 (49.1%) patients had intermediate-risk disease, and 26 (47.3%) patients had poor-risk disease with 10 (18.2%) patients having more than ≥4 risk factors. Nodal metastasis at the treatment initiation was present in 45 (81.8%) patients, and metastatic disease was noted in several organs, including the lungs (30 patients, 54.5%), bone (27 patients, 49.1%), and liver (16 patients, 29.1%).

Table 1

Baseline patient characteristics (n=55)

Variables	Papillary (n=25)	Chromophobe (n=12)	Unclassified (n=18)	All Patients (n=55)
Age at metastatic disease
Median	68	62.5	63.5	65
Range	21–83	47–71	38–78	21–83
IQR	56–73	58.5–65	57–68	57–70
Sex
Male	19 (76%)	7 (58.3%)	11 (61.1%)	37 (67.3%)
Female	6 (24%)	5 (41.7%)	7 (38.9%)	18 (32.7%)
Race
White	21 (85%)	12 (100%)	18 (100%)	51 (92.7%)
Black	4 (16%)	0 (0%)	0 (0%)	4 (7.3%)
Sarcomatoid differentiation	6 (24%)	8 (66.7%)	6 (33.3%)	20 (36.4%)
Metastatic status
Synchronous	17 (68%)	6 (50%)	16 (88.9%)	39 (70.9%)
Metachronous	8 (32%)	6 (50%)	2 (11.1%)	16 (29.1%)
Nephrectomy before systemic therapy
Yes	10 (40%)	9 (75%)	4 (22.2%)	23 (41.8%)
Therapies before nivo/ipi
None	24 (96%)	11 (91.7%)	17 (89.4%)	52 (94.5%)
TKI therapy	1 (4%)*	1 (8.3%)*	1 (5.3%)**	3 (5.5%)
IMDC risk groups
Favorable	1 (4%)	0 (0%)	1 (5.6%)	2 (3.6%)
Intermediate	10 (40%)	9 (75%)	8 (44.4%)	27 (49.1%)
Poor	14 (56%)	3 (25%)	9 (50%)	26 (47.3%)
Metastatic disease sites, at nivo/ipi start
Lymph nodes	22 (88%)	5 (41.7%)	18 (100%)	45 (81.8%)
Lung	15 (60%)	6 (50%)	9 (50%)	30 (54.5%)
Liver	7 (28%)	2 (16.7%)	7 (38.9%)	16 (29.1%)
Bone	14 (56%)	3 (25%)	10 (55.6%)	27 (49.1%)
Brain	1 (4%)	0 (0%)	1 (5.6%)	2 (3.6%)

IMDC, International Metastatic Renal Cell Carcinoma Database Consortium; Nivo/ipi, nivolumab plus ipilimumab; TKI, tyrosine kinase inhibitor.

Data on tumor programmed death ligand 1 (PD-L1) expression, analyzed with the 22C3 immunohistochemistry assay, were available for five patients only: two without SF in their tumors (uRCC with tumor proportion score (TPS) 50%; uRCC with focal PD-L1 positivity) and three with SF (uRCC with a combined positive score (CPS) 50; pRCC with TPS 80%; chRCC with CPS 10, where 2% of chromophobe cells, representing 90% of the specimen, and 80% of sarcomatous cells, representing 10% of the specimen, were PD-L1 positive).

Survival and response analysis

General cohort

In the entire cohort of 55 patients, ORR was 36.4%, with the highest ORR in pRCC (48%) and the lowest in chRCC (25%) (table 2). CBR was 54.5% in the entire cohort and specifically highest in pRCC (64%).

Table 2

Treatment outcomes with nivolumab plus ipilimumab (n=55)

Outcomes	Papillary mRCC (n=25)	Chromophobe mRCC (n=12)	Unclassified mRCC (n=18)	All patients with mRCC (n=55)
ORR, n (%)	12 (48%)	3 (25%)	5 (27.8%)	20 (36.4%)
CR	1 (4%)	0 (0%)	1 (5.6%)	2 (3.6%)
PR	11 (44%)	3 (25%)	4 (22.2%)	18 (32.7%)
SD	4 (16%)	3 (25%)	3 (16.7%)	10 (18.2%)
PD*	6 (24%)	4 (33.3%)	8 (44.4%)	18 (32.7%)
NE	3 (12%)	2 (16.7%)	2 (11.1%)	7 (12.7%)
ORR among patients with SF, n (%)	4/6 (66.7%)	3/8 (37.5%)	4/6 (66.7%)	11/20 (55%)
CBR, n (%)	16 (64%)	6 (50%)	8 (44.4%)	30 (54.5%)
Median OS from nivo/ipi start, mo (95% CI)	36.7 (11.5–54.8)	25.7 (0.9 – NE)	11.1 (6.5 – NE)	19.4 (11.5–36.7)
Median follow-up, month (95% CI)	35.1 (12.3–72.6)	51.4 (21.2–58.8)	33.5 (24.4–42.9)	35.1 (30.8–52.3)
Median PFS, month (95% CI)	10.6 (2.8–22.8)	3.6 (0.9 – NE)	3 (2.1–7)	4.6 (2.8–9.6)
6-month milestone PFS (%) (95% CI)	56% (36.3–75.7)	41.7% (22–61.3)	38.9% (21.7–56.1)	47.2% (35.3–59.1)
12-month milestone PFS (%) (95% CI)	31% (11.4–54.7)	25% (0.5–49.5)	16.7% (0–33.9%)	25% (13.1–36.9)
Median DoR, month (95% CI)	20.1 (8.3 - NE)	8 (8 – NE)	8.5 (2.9 – NE)	8.5 (8 – NE)
Median TOT, month (IQR)	2.9 (0.4–10.9)	2.6 (0.9–10.2)	2.1 (1.2–5.4)	2.1 (0.7–10.3)
Median TTNT, month (IQR)	8.8 (3–23.4)	3 (2.6–13.9)	3.3 (2.6–8.4)	4.1 (2.7–11.7)

*PD defined by RECIST version 1.1 or clinical disease progression in case of no restaging imaging.

CBR, clinical benefit rate; CR, complete response; DoR, duration of response; mo, months; mRCC, metastatic renal cell carcinoma; NE, Non evaluable; ORR, objective response rate; OS, Overall survival; PD, progressive disease; PFS, progression-free survival; PR, partial response; SD, stable disease; SF, sarcomatoid features; TOT, time on treatment; TTNT, time to next treatment.

The median PFS was 10.6 months (95% CI: 2.8 to 22.8), 3.6 months (95% CI: 0.9 – NE), and 3 months (95% CI: 2.1 to 7) in pRCC, chRCC, and uRCC, respectively (figure 1). Twenty-three (43.6%) patients, mostly with pRCC, were alive without disease progression at 6 months, while 12 (21.8%) patients remained progression-free at 12 months. The 6-month milestone PFS was 56% (95% CI: 36.3 to 75.7), 41.7% (95% CI: 22 to 61.3), and 38.9% (95% CI: 21.7 to 56.1), while the 12-month milestone PFS was 31% (11.4–54.7), 25% (0.5–49.5), and 16.7% (0–33.9%) in pRCC, chRCC, and uRCC, respectively. The median DoR across the entire cohort was 8.5 months (95% CI: 8 – NE), with the following breakdown per histology: 20.1 months (95% CI: 8.3 – NE) for 12 patients with pRCC, 8 months (95% CI: 8 – NE) for three patients with chRCC and SF, and 8.5 months (95% CI: 2.9 – NE) for five patients with uRCC. The median TOT of the cohort was 2.1 months (IQR: 0.7–5.5) and generally consistent across the three subtypes: 2.9 months (IQR: 0.4–10.9), 2.6 months (IQR: 0.9–10.2), 2.1 months (IQR: 1.2–5.4) in pRCC, chRCC, and uRCC, respectively. The median TTNT of the cohort was 4.1 months (IQR: 2.7–11.7), highest in pRCC of 8.8 months (IQR: 3–23.4), 3 months (IQR: 2.6–13.9) in chRCC, and 3.3 months (IQR: 2.6–8.4), respectively.

Figure 1. Kaplan-Meier (KM) survival curves showing median progression-free survival in the entire cohort with metastatic renal cell carcinoma treated with nivolumab plus ipilimumab (nivo/ipi) and per histological subtype. Patients were censored if they stopped the treatment for non-progression or remained on treatment at the time of analysis.

With a median follow-up time of 35.1 months (95% CI: 30.8 to 52.3), the median OS for the entire cohort was 19.4 months (95% CI: 11.5 to 36.7) (figure 2). The median OS was 36.7 mo (95% CI: 11.5 to 54.8) in pRCC, with a median follow-up of 35.1 months (95% CI: 12.3 to 72.6); 25.7 months (95% CI: 0.9 – NE) in chRCC, with a median follow-up of 51.4 months (95% CI: 21.2 to 58.8), and 11.1 months (95% CI: 6.5 – NE) in uRCC, with a median follow-up of 33.5 months (95% CI: 24.4 to 42.9). Overall, eight patients have been without disease progression for over 2 years up to the date of analysis, including three patients with pRCC (two of them with SF), three patients with chRCC (two of them with SF), and two with uRCC (one with SF).

Figure 2. Kaplan-Meier (KM) survival curves showing the median overall survival in the entire cohort with metastatic renal cell carcinoma treated with nivolumab plus ipilimumab and per histological subtype. Patients were censored at the last follow-up if they remained live at the time of analysis.

There were no significant differences in the median OS or PFS between patients with and without prior nephrectomy (online supplemental figure 1). The median OS was 18.5 months (95% CI: 2.6 – NE) for patients with nephrectomy, compared with 28.3 months (95% CI: 6.5 to 28.3) for those without nephrectomy (p=0.98). Similarly, the median PFS was 4.3 months (95% CI: 1.4 – NE) for nephrectomy patients and 5.1 months (95% CI: 2.3 to 9.6) for those without nephrectomy (p=0.19).

We evaluated the primary renal tumor response to nivo/ipi in 32 patients with intact primary tumors, of whom 21 (65.6%) had sufficient data for tumor size assessment (10 uRCC, 9 pRCC, and 2 chRCC). The median primary tumor size in the 21-patient cohort was 8 cm [IQR: 4.3–11.7]. A ≥30% reduction in the primary tumor size was achieved in 7/21 (33.3%) patients treated with nivo/ipi. The overall median percentage reduction in the tumor size was 19.6% (IQR: 1.25%–32%), with a median change of −1.2 cm (IQR: −3.3 to−0.1]. For nine patients with pRCC, the median reduction was 5.9% (IQR: 1.3%–23.8%) with a median change of −0.3 cm (IQR: −1.5 to −0.1), whereas for the 10 patients with uRCC, median reduction was 29.3% (IQR: −13.4% to –35.5%) with a median change of −2.6 cm (IQR: −3.8–0.6).

Patients with SF in their tumors

Among the subset of patients with SF in their tumors (n=20), ORR was 55%. Specifically, ORR was 66.7% (4/6) in both pRCC and uRCC, and 37.5% (3/8) in chRCC (table 2). The median OS was 22.5 months (95% CI: 2.6 – NE) for all patients with SF vs 18.5 months (95% CI: 7.4 to 47.9) for those without SF. In pRCC, the median OS was 15 months (1.1 – NE) vs 36.7 months [2.3–54.8], and in chRCC, 27 months (0.6 – NE) vs 37.8 months (14.2 – NE), for patients with vs without SF, respectively (online supplemental figure 2). The median PFS was 11.6 months (95% CI: 1.1 – NE) and 7.2 months (95% CI: 1.5 to 22.8) in pRCC with and without SF; 6.2 months (95% CI: 0.6 – NE) and 3.4 months (95% CI: 1.4 – NE) in chRCC with and without SF, respectively (online supplemental figure 3).

Among patients with uRCC (n=18), the median PFS was longer and clinically meaningful in patients with SF vs those without SF (PFS: 9.2 months (95% CI: 2.3 – NE) vs 2.8 months (95% CI: 0.8 to 6.2), HR=0.4 (95% CI: 0.1 to 1), p=0.07). The median OS was 28.3 months (95% CI: 6.5 – NE) and 7.2 months (95% CI: 1 to 31.2), respectively (figure 3). Moreover, the median TOT was highest among patients with uRCC and SF (8.3 mo (IQR: 2.1–17.5]).

Figure 3. Kaplan-Meier survival curves comparing median overall survival and progression-free survival in patients with unclassified metastatic renal cell carcinoma treated with nivolumab plus ipilimumab.

Prior and subsequent lines of therapy

Most patients received nivo/ipi (52, 94.5%) in the 1 L setting (table 2). Only three (5.5%) patients received prior systemic therapy, including TKIs (two patients received 1 L sunitinib: one patient with pRCC, and one patient with chRCC; one patient with uRCC received cabozantinib). Of the three TKI-exposed patients previously treated with TKI, two had PD, and one achieved SD in response to nivo/ipi.

Thirty-five (63.6%) patients received up to four cycles of induction nivo/ipi without maintenance nivolumab (27 patients (four cycles), nine (three cycles), seven (two cycles), 12 (one cycle)), and 20 (36.4%) patients received maintenance nivolumab after the completion of all four induction cycles. Twenty-four (43.6%) patients were able to receive subsequent therapies, including cabozantinib (13/24, 54.2%), cabozantinib plus nivolumab (3/24, 12.5%), lenvatinib plus everolimus (2/24, 8.3%), axitinib plus pembrolizumab (2, 8.3%), lenvatinib plus pembrolizumab (1, 4.2%), bevacizumab (1, 4.2%), gemcitabine plus doxorubicin (1, 4.2%), and enfortumab vedotin (1, 4.2%). Forty-three (78.2%) patients experienced disease progression during or shortly after nivo/ipi induction or during maintenance nivolumab, while 12 patients did not have disease progression up to the time of analysis. Four patients (one with pRCC, one with chRCC, two with uRCC) underwent cytoreductive nephrectomy, all after maintenance nivolumab, on clinician’s decision with or without disease progression. Pathological outcomes included pRCC (ypT1a ypN0 with 70–75% fibrosis and therapy effects), chRCC (ypT3 ypN4/22 M0 R0 with necrosis noted), and uRCC (ypT2a ypN0/2 M1 R0 with 80% necrosis; ypT3 cN2 M1 with one completely necrotic tumor in the perinephric adipose tissue and a smaller, mostly viable, tumor with focal necrosis in the interpolar kidney).

We show overall patient disposition during and after induction therapy with nivo/ipi in online supplemental figure 4. Of the 19 patients who discontinued therapy without subsequent treatment, 15 did so after induction nivo/ipi and four after maintenance nivolumab. Among the former 15 patients, eight succumbed to cancer-related complications due to PD (including two chRCC, three pRCC, and three uRCC cases), three patients died from intercurrent disease, two were lost to follow-up, one progressed after discontinuing nivo/ipi due to toxicity and later passed away, and one chose not to pursue further treatment after toxicity. Among the four patients who did not receive subsequent therapies after maintenance nivolumab, one developed autoimmune colitis, one with cytoreductive nephrectomy entered active surveillance, one died from intercurrent disease, and one stopped maintenance nivolumab at the time of analysis.

Safety

Due to the retrospective nature of the study, safety data were available for analysis mostly for Grade 3 and 4 AEs. Grade 3/4 immune-mediated adverse events (IMAEs) were recorded in 17 (30.9%) patients: colitis (six patients, 10.9%), pneumonitis (three patients, 5.5%), hepatitis (three patients, 5.5%), nephritis (two patients, 3.6%), thyroiditis (one patient, 1.8%), pancreatitis (one patient, 1.8%), and thrombocytopenia (one patient, 1.8%). Grade 3/4 treatment-related adverse events (TRAEs) included pulmonary embolism (PE) (one patient, 1.8%) and anaphylaxis (one patient, 1.8%). The first patient was diagnosed with PE and right deep vein thrombosis after the third cycle, stabilized in the ICU then discharged on anticoagulation, but transitioned to hospice care. The other patient had the anaphylactic episode 10 min after the second nivolumab dose, showing facial flushing, dyspnea, hypotension, rigors, and peripheral cyanosis. The infusion was stopped, and later after effective same-day treatment, he continued with the next two nivolumab doses using a desensitization protocol. One patient developed multiple IMAEs including nephritis, thyroiditis, pneumonitis, and colitis. Another patient was hospitalized during the treatment for infectious colitis and bowel obstruction requiring surgical correction. There were no Grade 5 AEs. IMAEs were managed with high-dose glucocorticoids plus other immune suppressive agents and discontinuation of the immune checkpoint therapy.

Among the 35 patients who received induction nivo/ipi only, 18 (51.4%) discontinued treatment due to PD during induction, while 10 (28.6%) discontinued due to toxicity. Of these ten patients, four experienced PD after stopping nivo/ipi due to TRAEs and began subsequent line therapy, four patients did not experience PD after stopping the treatment despite the TRAEs, one had a change in systemic treatment due to toxicity before developing PD, and one experienced toxicity after the second dose and decided not to pursue further treatment.

Exploratory molecular and genomic characterization

Immunohistochemistry and molecular analysis of patients with uRCC (n=18) are included in table 3. Somatic alterations mostly comprised TP53, TERT, PTEN, NF2, and SETD2 in these patients. There were no germline mutations present, except one patient who had Li–Fraumeni syndrome (TP53 germline alteration).

Table 3

Characteristics of patients with unclassified histology (n=18)

Pt #	Age at dx	SF (Y/N)	Sex	IMDC score	IHC	Molecular alterations	Best ORR (RECIST v1.1)	OS (mo)	Censor
1	67	N	M	0	N/A	N/A	NE	1	1
2	60	N	M	6	(+) PanCK, PAX8, P504S, vimentin, PDL1 (-) CAIX, CK7, PAX-2, PIN dual, SOX10, SALL4, TFE3	N/A	SD	10.3	1
3	68	N	F	3	(+) PanCK, CK8, CD10 (-) PAX8, EMA, SOX-10, HMB45, MelanA, S100	TP53	PD	7.4	1
4	37	N	M	2	(+) TFE3, PAX8, vimentin, P504S, EMA (-) CK7, p63, HMWCK, CAIX, uroplakin-2, panmelanoma, cathepsin K, GATA3	SF3B1	PD	6.5	1
5	69	N	M	1	(+) CD10, vimentin, P504S, FH (-) CD117, CK7, CAIX, cathepsin K	IDH2, MSH6	SD	3.2	1
6	61	N	F	2	(+) PAX-8, CD10, CK7 (-) CK20, GATA3, CAIX, ER, TTF1	PTEN	PD	31.2	1
7	55	N	M	3	(+) Keratin, PAX-8, PAX-2, vimentin, CD10, TFE3, and p504s	ERBB3, NF2, TERT	PD	42.9	0
8	64	Y	M	3	(+) PanCK, desmin, SMA	FANCA, NF2, PTEN, TERT, TP53	PR	11.9	1
9	63	Y	M	1	(+) Cytokeratin, PAX8, SMA	TP53, SETD2	PD	31.3	0
10	66	N	M	2	(+) PanCK, PAX-8, CAM 5.2	N/A	NE	7	1
11	57	Y	F	1	(+) EMA, CK7, CD10	N/A	CR	37.3	0
12	67	Y	M	5	(+) PAX8	N/A	PD	6.5	1
13	49	Y	F	2	(+) PanCK, SMA, vimentin	N/A	PR	24.4	0
14	63	N	F	3	(+) PanCK, vimentin, pax-8	ABL1, MGMT, PALB2, PIK3C2G, ROS2, SETD2, TBC1D4	PR	33.5	0
15	77	Y	F	5	(+) PanCK, Vimentin, EMA	ARID1A, EPHA7, KMT2D, NFE2L2, TERT	PR	19.4	1
16	41	N	M	3	(+) PAX8, CAM5, CD10, vimentin	N/A	PD	0.8	1
17	70	N	F	3	(+) Cytokeratin 7, PAX8, vimentin	N/A	PD	1	1
18	68	N	M	1	(+) PanCK, PAX8	ATR, BRCA1, CUL3, KMT2C, MDC1, MET, RAD51B, RNF43, SLIT2, STAT5A, TNFAIP3	SD	27.5	0

IHC, immunohistochemistry; IMDC, International Metastatic Renal Cell Carcinoma Database Consortium; mo, months; OS, overall survival; PD, progressive disease; PFS, progression-free survival; SD, stable disease; SF, sarcomatoid features.

Clinical targeted DNA sequencing output with NGS was combined from analyses on metastatic tumor biospecimens from 26 patients (47.3% of the entire sample). Mutations with frequency of alterations below 4% were not represented in our summary oncoplot (figure 4). The most frequent somatic alterations were TP53 (42%), PTEN (23%), TERT (23%), and NF2 (15%). TP53 mutations were common in chRCC (8/9 cases) but absent in pRCC (0/7). In contrast, TERT mutations were enriched in pRCC (3/7) but absent in chRCC (0/9). Additionally, two out of four chRCC tumors with SF had both TP53 plus PTEN mutations, compared with none of those without SF.

Figure 4. Oncoplot showing the somatic alterations found in metastatic tumor biospecimens from 26 patients (47.3% of entire sample) treated with nivolumab plus ipilimumab (nivo/ipi), classified by the type of alteration, pathogenicity, vhRCC histological subset, the presence of sarcomatoid features and linked with the objective radiographic response to nivo/ipi.

Discussion

Histological subtypes of vhRCC represent diverse clinical and biological entities and should not be approached with the same treatment strategy, as the efficacy of a treatment regimen depends on many factors including the tumor microenvironment. Among multicenter or single-center retrospective studies of ICIs in vhRCC, our study represents one of the largest datasets to date, investigating nivo/ipi specifically across pRCC, chRCC, and uRCC subtypes, with and without SF.^14–18

In our study, patients with pRCC derived a higher clinical benefit from nivo/ipi compared with those with chRCC and uRCC, with the highest ORR, longest OS, longest PFS, highest 6-month milestone PFS, longest DoR, and longest TTNT. Although patients with chRCC achieved a modest benefit with nivo/ipi, their outcomes should be viewed in the context of the presence of SF in two-thirds of chRCC cases. Patients with uRCC had the worst OS, likely due to the inherent aggressive biology of uRCC leading to a higher prevalence of liver metastases and a greater nodal disease burden. Molecular alterations in our uRCC cohort, including NF2 and SETD2, are known to confer a worse prognosis in uRCC.^{19 20} At our institution, uRCC cases undergo rigorous pathological review (eg, no clear cell morphology, the absence of CAIX expression or underlying VHL mutations) and NGS, often revealing alternative histologies, leading to fewer diagnoses of pure uRCC. Some somatic alterations identified in our uRCC cohort align with a previous molecular analysis of 62 primary uRCC tumors, where NF2 and SETD2 were the most frequent, each occurring in 18% of cases.¹⁹

The presence of SF was associated with higher ORR across all three histologies (pRCC, chRCC, and uRCC). Although the median PFS was nearly doubled in pRCC and chRCC with SF in their tumors, this did not translate to OS benefit conclusively, which could be explained by the small sample sizes. Additionally, SF was associated with potential PFS benefit in uRCC (HR=0.4 [HR=0.4 [95% CI: 0.1 to 1], p=0.07), suggesting a potential use of nivo/ipi in this patient population.

The overall low tumor mutational burden (TMB), well known in RCC,²¹ could not be confidently correlated with survival outcomes due to limited numbers (figure 4). We did observe that four out of six patients with pRCC and four out of six patients with uRCC who had TERT promoter alterations in their tumors, as well as four patients who had SETD2 mutations in their tumors, had objective responses to nivo/ipi. Our oncoplot also shows distinct molecular profiles of pRCC and chRCC, with TERT mutations enriched in the former and TP53 mutations more common in the latter.

CheckMate 920 was the first phase 3b/4 trial of 1 L nivo/ipi in 52 patients with metastatic vhRCC, which demonstrated an ORR of 19.6% in 46 response-evaluable patients and durable responses in pRCC (one CR, four PR) and uRCC (one CR, three PR).⁹ Our study complements the results of CheckMate 920 regarding the clinical activity of nivo/ipi in vhRCC. It is similar to CheckMate 920 in the following aspects: the treatment-naïve metastatic status (1 L nivo/ipi); the inclusion of three variant histological subtypes, pRCC, chRCC, and uRCC (comprizing 90.4% in CheckMate 920 vs 100% in ours); and the presence of SF (28.8% in CheckMate 920 vs 36.4% in our study). This higher-than-expected SF prevalence, normally around 5% overall in vhRCC and 20% in the metastatic setting, might be explained by the preferential treatment with nivo/ipi of this subcohort of patients at two large academic centers, compared with the choice of TKI+ICI in patients without SF.⁵

Both studies show comparable median OS: 21.2 months (95% CI: 16.6 – NE) in the CheckMate 920 study and 19.4 months (95% CI: 11.5 to 36.7) in ours. However, CheckMate 920 reported a lower ORR compared with ours (19.6% [9.4%–33.9%] vs 36.4%) and a higher SD rate (37% vs 18.2%). We should note that although our study was retrospective, tumor response assessment by RECIST v1.1 was carried out by blinded radiologists. Despite the differences between CheckMate 920 and our study, two response trends to nivo/ipi are inferred: limited responses in chRCC (all responses in CheckMate 920 were exclusively in pRCC and uRCC: two CR and seven PR) and higher ORR in the presence of SF (35.7% vs 55% in our study), in keeping with the benefit of nivo/ipi over sunitinib in ccRCC with SF.^{7 22} The prevalence of Grade 3/4 IMAEs was also largely comparable in the two studies (19/52 (36.5%) in CheckMate 920 vs 17/55 (30.9%) in ours) with no reported Grade 5 events.⁹ The lower rates of grade 3 and 4 IMAEs in our study and CheckMate 920 might be explained by the frontline treatment setting, as patients might not be as prone to developing toxicity compared with heavily pretreated cohorts. Our study also showed that a ≥30% reduction in primary tumor diameter was achieved in around 33% of patients treated with nivo/ipi, aligning with the 35% primary tumor response rate reported in a post hoc exploratory analysis of the CheckMate 214 trial.²³

In CheckMate 214, 79% of patients completed all four cycles of nivo/ipi induction, as those receiving only one to three cycles were taken off protocol and ineligible for maintenance nivolumab.^{6 8} In contrast, the retrospective study of our study highlights a significantly higher proportion of patients who did not transition to maintenance therapy, with 51.4% discontinuing due to PD during induction and 28.6% due to toxicity. This observation may reflect the more aggressive disease biology in our cohort compared with CheckMate 214 and CheckMate 920.

The most recent data on the role of dual checkpoint inhibition in vhRCC came from the large, randomized phase 2 trial, SUNNIFORECAST (NCT03075423), of nivo/ipi vs the standard of care in treatment-naïve metastatic vhRCC. This study included, similarly to our study and CheckMate 920, patients with papillary, chromophobe, and unclassified RCC, and many other vhRCC subtypes such as translocation RCC, collecting duct carcinoma, mucinous tubular and spindle cell carcinoma, renal medullary carcinoma, and tubulocystic RCC. Included in this study were 14 patients who had SFs in their tumors, but no information was provided regarding the epithelial component underpinning the SFs. While the study met the unconventional 12-month OS used by the investigators as the primary endpoint in favor of nivo/ipi (86.9% for nivo/ipi vs 76.8% for SOC, p=0.0141), the results showed no significant difference in OS, which is a more appropriate, comprehensive, and clinically meaningful endpoint, as 12-month OS fails to distinguish among survival times of patients living beyond this milestone (median OS 42.4 months (95% CI: 35.2 – 55.5) vs 33.9 months (95% CI: 25.5 – NE), p=0.29). Furthermore, the median PFS was 5.52 months (95% CI: 4.3 – 8.23) vs 5.65 months (95% CI: 5.49 – 8.46) (HR=0.99 (95% CI: 0.76 – 1.18)) for the nivo/ipi arm and the SOC arm in the general vhRCC cohort. Of note, ORR with nivo/ipi were 29.2%, 25.9%, and 37.7% in papillary, chromophobe, and non-papillary subtypes, respectively.²⁴

The observed antitumor activity of nivo/ipi in patients with advanced vhRCC in our study, in CheckMate 920, and in SUNNIFORECAST, seems to be lower across most efficacy measures relative to that reported in patients with advanced ccRCC in CheckMate 214, especially with its latest 8 year median follow-up.⁸ Ten (18.2%) patients discontinued the treatment in our study due to TRAEs, consistently with the 23.6% discontinuation rate reported in the nivo/ipi arm of CheckMate 214. However, the clinical benefit shown with nivo/ipi in our study is important given the potential of treatment discontinuation, as shown in the 8 year follow-up of CheckMate 214, where 49.8% of patients with durable responses (CR or PR) to nivo/ipi in metastatic ccRCC were able to stop therapy.⁸ The ORR of 36.4% in our study was consistent with estimates in smaller retrospective series, which reported ORRs ranging from 30% to 36%,^11–13 with higher response rates and more durable outcomes with nivo/ipi in the 1 L setting rather than salvage therapy following disease progression.²⁵ A retrospective multicenter study by McKay et al also reported modest antitumor activity of anti-PD-1 and anti-PD-L1 agents in pRCC and translocation RCC, as well as ccRCC with sarcomatoid and/or rhabdoid features.²⁶

A comparison of efficacy outcomes for patients with vhRCC treated with nivo/ipi in our study with other studies of single-agent ICI or ICI/TKI combination therapy is highlighted in online supplemental table 1. An important observation to note is that a single-agent ICI strategy does not seem optimal in the 1 L treatment of patients with vhRCC, except the notable efficacy of single-agent pembrolizumab studied in KEYNOTE 427 Cohort B. Among 165 patients, ORR was 25.4% for pRCC (with two-thirds of responders having durable responses at 18 mo), 9.5% for chRCC, 34.6% for uRCC, and 44.7% among those with sarcomatoid dedifferentiation.²⁷ In CheckMate 374, single-agent nivolumab yielded a low ORR of 13.6% among 44 patients, with ORR of 8.3% (2/24) for pRCC, 28.6% (2/7) for chRCC, 18.2% (1/8) for uRCC, and 50% (2/4) among patients with SF.²⁸ We previously reported a higher ORR (44.4%) in nine patients receiving nivolumab in combination with ipilimumab or VEGF-directed agents vs 13% ORR in 31 patients receiving single-agent nivolumab.¹⁰

With no current prospective trials comparing head-to-head VEGFR-TKI/ICI combinations to dual ICI-based regimens for the treatment of patients with metastatic vhRCC, the selection of therapy is often based on patient comorbidities and performance status and taking into consideration patient preference.¹ Recent options for patients with metastatic pRCC include lenvatinib plus pembrolizumab, as shown in KEYNOTE-B61, and cabozantinib plus nivolumab (online supplemental table 1).^{27 29} At a median follow-up of 15 months in KEYNOTE-B61, ORR with lenvatinib plus pembrolizumab were 54% and 52% in pRCC and uRCC, respectively, and 47% and 50%, respectively, with cabozantinib plus nivolumab at a median follow-up of 13 months.

The therapeutic options are even more limited for chRCC, with all three patients with chRCC in our study who achieved a major response to nivo/ipi having SF. This finding contrasts with the limited efficacy of single-agent ICIs in such an immunologically cold subtype, with limited PD-L1 expression, decreased TMB, and a highly suppressive immune environment.^{1 30 31} No responses were reported with nivo/ipi in CheckMate 920 or cabozantinib plus nivolumab.^{9 29} The only modest evidence of response was reported with lenvatinib plus pembrolizumab in KEYNOTE B61, although it is plausible that these responses were predominantly driven by lenvatinib (ORR 28% and 1-year PFS milestone of 53%).^{32 33} A large retrospective study of 109 patients with metastatic chRCC reported no objective responses and a median time to treatment failure of 2.6 months in six patients treated with single-agent nivolumab, including four patients with SF.³⁴

Conclusion

The spectrum of vhRCC represents diverse histological subtypes with different biology and variable response to ICIs and divergent outcomes. In keeping with the results of CheckMate 920 and SUNNIFORECAST, our data showed effectiveness of nivo/ipi in pRCC. SF was associated with higher response rates and numerically longer PFS in all three vhRCC subtypes, but the numbers of patients were small to make meaningful inferences. Although the number of patients with chRCC and SF in our study was small (8/12), nivo/ipi produced an ORR of 37.5% (3/8), making it a therapeutic option in these patients. The results of our study underscore the limitation of nivo/ipi in patients with vhRCC and hence the pressing need to accelerate discovery of new molecular targets and expand our understanding of the tumor microenvironment in vhRCC to develop novel and more effective therapies for these patients and improve their outcomes.

Implications for practice

Nivo/ipi has demonstrated efficacy in ccRCC; however, the empiric support for its utility in variant histology subtypes remains limited due to the rarity of a number of these diverse tumor subtypes and variable biology, which has not been well-studied. Patients with treatment-naïve papillary metastatic RCC seemed to benefit most from nivo/ipi (ORR: 48%). Sarcomatoid dedifferentiation was associated with a higher objective response rate (ORR: 55%) across the three subtypes. Notably, all objective responses in chromophobe (25% ORR, consisting of three PRs) and the majority in unclassified (four out of five responses) occurred in patients whose tumors had SF.

We sincerely thank Ms. Katie Coleman, a kidney cancer patient advocate and survivor, for her unwavering support, inspiration and dedication to advancing knowledge and treatment for kidney cancer. We would like to thank the patients and their caregivers for their participation and support, which made this study possible.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study was approved by the Institutional Review Board of MDACC (protocol PA16-0736).

Footnote

X @mjmoussa_, @Oalhalabimd

MJM, JK and NRW contributed equally.

Presented at A portion of this work was concurrently presented during the Poster Session at the 2025 ASCO Genitourinary Cancers Symposium.

Contributors MJM and NMT accept full responsibility for the work, the conduct of the study, and the decision to publish. They attest that all authors meet the criteria for authorship and are accountable for all aspects of the work. NMT is the guarantor. Conceptualization and study design: NMT and OA. Methodology: NMT, OA, and PM. Data collection or curation: MJM, JK, NRW, KLM, DSS, TKB, HA, and EH. Data analysis and interpretation: MJM, NMT, OA, PM, and MTC. Supervision: NMT and OA. Resources and materials: NMT, EJ, AS, SG, KLM, DSS, TKB, KME, PR, PT, and KS. Writing - original draft preparation: MJM, JK, NRW, OA, and NMT. Writing - review and editing: NMT, OA, MTC, PM, EJ, and ACJ. Visualization: MJM, JK, NRW, and YL. Project administration: NMT. Validation: NMT.

Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Competing interests MJM, JK, NRW, KLM, TKB, YL, PR, PT, KS, HA, KME, ACJ, and SG report no disclosures. DSS reports research support from Blue Earth Therapeutics. AS reports scientific advisory board fees from Bristol Myers Squibb, Exelixis, and Pfizer; and research funding from Bristol Myers Squibb, Eisai, EMD Serono, and 4D Pharma. EH reports scientific advisory board fees from Pfizer, Eisai, and Telix Pharmaceuticals; honoraria from Targeted Oncology; and affiliations with Pelotonia Institute for Immuno-Oncology. EJ reports research funding from Arrowhead, Merck, and NiKang, as well as honoraria from Aveo, Aravive, Calithera, Eisai, Exelixis, Ipsen, Merck, NiKang, Novartis, Pfizer, and Takeda. PM reports honoraria for the service on scientific advisory boards for Mirati Therapeutics, Bristol Myers Squibb, and Exelixis; consulting for Axiom Healthcare Strategies; nonbranded educational programs supported by DAVA Oncology, Exelixis, and Pfizer; and research funding for clinical trials from Takeda, Bristol Myers Squibb, Mirati Therapeutics, Gateway for Cancer Research, and the University of Texas MD Anderson Cancer Center. MTC reports consultancy or advisory role for Astellas, AstraZeneca, AXDev, Eisai, EMD Serono, Exelixis, Genentech, Pfizer, and SeaGen; research funding from ApricityHealth, Aravive, AstraZeneca, Exelixis, Janssen, and Pfizer/EMD Serono; nonbranded educational programs from Bristol Myers Squibb, Merck, Pfizer/EMD Serono, and Roche. OA reports scientific advisory board fees from Seagen, Silverback Therapeutics, and Cardinal Health; educational program speaker from Curio Science and Aptitude Health; and research funding to the institution from AstraZeneca, Ikena Oncology, Genentech, and Arcus Biosciences. NMT reports research funding from Bristol-Myers Squibb Company (BMS), Calithera Biosciences, Nektar Therapeutics, Exelixis, Pfizer, Novartis, Arrowhead Pharmaceuticals, Mirati Therapeutics, Takeda, Epizyme, and Eisai Medical Research; consulting, advisory, travel accommodations, and expenses from BMS, Calithera Biosciences, Nektar Therapeutics, Exelixis, Pfizer, Novartis, Eisai Medical Research, Ipsen, Lilly Oncology, Neoleukin Therapeutics, Surface Oncology, ONO Pharmaceutical, and Oncorena; and honoraria from BMS, Exelixis, Nektar Therapeutics, Calithera Biosciences, Eisai Medical Research, ONO Pharmaceutical, Eli Lilly, Oncorena, Ipsen, and Surface Oncology.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ Msaouel P, Genovese G, Tannir NM. Renal Cell Carcinoma of Variant Histology: Biology and Therapies. Hematol Oncol Clin North Am 2023; 37: 977–92. doi:10.1016/j.hoc.2023.04.019

² Rini BI, McDermott DF, Hammers H, et al. Society for Immunotherapy of Cancer consensus statement on immunotherapy for the treatment of renal cell carcinoma. J Immunother Cancer 2016; 4: 81. doi:10.1186/s40425-016-0180-7

³ de Velasco G, McKay RR, Lin X, et al. Comprehensive Analysis of Survival Outcomes in Non-Clear Cell Renal Cell Carcinoma Patients Treated in Clinical Trials. Clin Genitourin Cancer 2017; 15: 652–60. doi:10.1016/j.clgc.2017.03.004

⁴ Kroeger N, Xie W, Lee J-L, et al. Metastatic non-clear cell renal cell carcinoma treated with targeted therapy agents: characterization of survival outcome and application of the International mRCC Database Consortium criteria. Cancer 2013; 119: 2999–3006. doi:10.1002/cncr.28151

⁵ Salgia NJ, Zengin ZB, Pal SK, et al. Renal Cell Carcinoma of Variant Histology: New Biologic Understanding Leads to Therapeutic Advances. Am Soc Clin Oncol Educ Book 2024; 44: e438642. doi:10.1200/EDBK_438642

⁶ Motzer RJ, Tannir NM, McDermott DF, et al. Nivolumab plus Ipilimumab versus Sunitinib in Advanced Renal-Cell Carcinoma. N Engl J Med 2018; 378: 1277–90. doi:10.1056/NEJMoa1712126

⁷ Tannir NM, Signoretti S, Choueiri TK, et al. Efficacy and Safety of Nivolumab Plus Ipilimumab versus Sunitinib in First-line Treatment of Patients with Advanced Sarcomatoid Renal Cell Carcinoma. Clin Cancer Res 2021; 27: 78–86. doi:10.1158/1078-0432.CCR-20-2063

⁸ Tannir NM, Albigès L, McDermott DF, et al. Nivolumab plus ipilimumab versus sunitinib for first-line treatment of advanced renal cell carcinoma: extended 8-year follow-up results of efficacy and safety from the phase III CheckMate 214 trial. Ann Oncol 2024; 35: 1026–38. doi:10.1016/j.annonc.2024.07.727

⁹ Tykodi SS, Gordan LN, Alter RS, et al. Safety and efficacy of nivolumab plus ipilimumab in patients with advanced non-clear cell renal cell carcinoma: results from the phase 3b/4 CheckMate 920 trial. J Immunother Cancer 2022; 10: e003844. doi:10.1136/jitc-2021-003844

¹⁰ Chahoud J, Msaouel P, Campbell MT, et al. Nivolumab for the Treatment of Patients with Metastatic Non-Clear Cell Renal Cell Carcinoma (nccRCC): A Single-Institutional Experience and Literature Meta-Analysis. Oncologist 2020; 25: 252–8. doi:10.1634/theoncologist.2019-0372

¹¹ Gupta R, Ornstein MC, Li H, et al. Clinical Activity of Ipilimumab Plus Nivolumab in Patients With Metastatic Non-Clear Cell Renal Cell Carcinoma. Clin Genitourin Cancer 2020; 18: 429–35. doi:10.1016/j.clgc.2019.11.012

¹² Bando Y, Furukawa J, Okamura Y, et al. Comparative Efficacy of Combination Therapy of Ipilimumab Plus Nivolumab for Non-clear Cell Renal Cell Carcinoma. Anticancer Res 2022; 42: 973–9. doi:10.21873/anticanres.15557

¹³ Izumi K, Inoue M, Washino S, et al. Clinical outcomes of nivolumab plus ipilimumab in patients with metastatic non-clear cell renal cell carcinoma: Real-world data from a Japanese multicenter retrospective study. Int J Urol 2023; 30: 714–21. doi:10.1111/iju.15128

¹⁴ Graham J, Wells JC, Dudani S, et al. Outcomes of patients with advanced non-clear cell renal cell carcinoma treated with first-line immune checkpoint inhibitor therapy. Eur J Cancer 2022; 171: 124–32. doi:10.1016/j.ejca.2022.05.002

¹⁵ Schwartzman W, Elias R, Patel VM, et al. Safety and efficacy of immune checkpoint inhibitors (ICI) in metastatic non-clear cell renal cell carcinoma (nccRCC): An institutional experience. JCO 2020; 38: 640. doi:10.1200/JCO.2020.38.6_suppl.640

¹⁶ Yip SM, Wells C, Moreira R, et al. Checkpoint inhibitors in patients with metastatic renal cell carcinoma: Results from the International Metastatic Renal Cell Carcinoma Database Consortium. Cancer 2018; 124: 3677–83. doi:10.1002/cncr.31595

¹⁷ Koshkin VS, Barata PC, Zhang T, et al. Clinical activity of nivolumab in patients with non-clear cell renal cell carcinoma. J Immunother Cancer 2018; 6: 9. doi:10.1186/s40425-018-0319-9

¹⁸ Ernst MS, Navani V, Wells JC, et al. Outcomes for International Metastatic Renal Cell Carcinoma Database Consortium Prognostic Groups in Contemporary First-line Combination Therapies for Metastatic Renal Cell Carcinoma. Eur Urol 2023; 84: 109–16. doi:10.1016/j.eururo.2023.01.001

¹⁹ Chen Y-B, Xu J, Skanderup AJ, et al. Molecular analysis of aggressive renal cell carcinoma with unclassified histology reveals distinct subsets. Nat Commun 2016; 7: 13131. doi:10.1038/ncomms13131

²⁰ Casuscelli J, Vano Y-A, Fridman WH, et al. Molecular Classification of Renal Cell Carcinoma and Its Implication in Future Clinical Practice. Kidney Cancer 2017; 1: 3–13. doi:10.3233/KCA-170008

²¹ Strickler JH, Hanks BA, Khasraw M. Tumor Mutational Burden as a Predictor of Immunotherapy Response: Is More Always Better? Clin Cancer Res 2021; 27: 1236–41. doi:10.1158/1078-0432.CCR-20-3054

²² Rini BI, Signoretti S, Choueiri TK, et al. Long-term outcomes with nivolumab plus ipilimumab versus sunitinib in first-line treatment of patients with advanced sarcomatoid renal cell carcinoma. J Immunother Cancer 2022; 10: e005445. doi:10.1136/jitc-2022-005445

²³ Albiges L, Tannir NM, Burotto M, et al. First-line Nivolumab plus Ipilimumab Versus Sunitinib in Patients Without Nephrectomy and With an Evaluable Primary Renal Tumor in the CheckMate 214 Trial. Eur Urol 2022; 81: 266–71. doi:10.1016/j.eururo.2021.10.001

²⁴ Randomized phase-II study of nivolumab plus ipilimumab vs. Standard of care in untreated and advanced non-clear cell RCC - full text view - clinicaltrials.gov. 2024. Available: https://classic.clinicaltrials.gov/ct2/show/ NCT03075423

²⁵ Atkins MB, Jegede OA, Haas NB, et al. Phase II Study of Nivolumab and Salvage Nivolumab/Ipilimumab in Treatment-Naive Patients With Advanced Clear Cell Renal Cell Carcinoma (HCRN GU16-260-Cohort A). J Clin Oncol 2022; 40: 2913–23. doi:10.1200/JCO.21.02938

²⁶ McKay RR, Bossé D, Xie W, et al. The Clinical Activity of PD-1/PD-L1 Inhibitors in Metastatic Non-Clear Cell Renal Cell Carcinoma. Cancer Immunol Res 2018; 6: 758–65. doi:10.1158/2326-6066.CIR-17-0475

²⁷ McDermott DF, Lee J-L, Ziobro M, et al. Open-Label, Single-Arm, Phase II Study of Pembrolizumab Monotherapy as First-Line Therapy in Patients With Advanced Non-Clear Cell Renal Cell Carcinoma. J Clin Oncol 2021; 39: 1029–39. doi:10.1200/JCO.20.02365

²⁸ Vogelzang NJ, Olsen MR, McFarlane JJ, et al. Safety and Efficacy of Nivolumab in Patients With Advanced Non-Clear Cell Renal Cell Carcinoma: Results From the Phase IIIb/IV CheckMate 374 Study. Clin Genitourin Cancer 2020; 18: 461–8. doi:10.1016/j.clgc.2020.05.006

²⁹ Lee C-H, Voss MH, Carlo MI, et al. Phase II Trial of Cabozantinib Plus Nivolumab in Patients With Non–Clear-Cell Renal Cell Carcinoma and Genomic Correlates. JCO 2022; 40: 2333–41. doi:10.1200/JCO.21.01944

³⁰ Zarrabi K, Walzer E, Zibelman M. Immune Checkpoint Inhibition in Advanced Non-Clear Cell Renal Cell Carcinoma: Leveraging Success from Clear Cell Histology into New Opportunities. Cancers (Basel) 2021; 13: 3652. doi:10.3390/cancers13153652

³¹ Zoumpourlis P, Genovese G, Tannir NM, et al. Systemic Therapies for the Management of Non-Clear Cell Renal Cell Carcinoma: What Works, What Doesn’t, and What the Future Holds. Clin Genitourin Cancer 2021; 19: 103–16. doi:10.1016/j.clgc.2020.11.005

³² Pal SK, McGregor B, Suárez C, et al. Cabozantinib in Combination With Atezolizumab for Advanced Renal Cell Carcinoma: Results From the COSMIC-021 Study. J Clin Oncol 2021; 39: 3725–36. doi:10.1200/JCO.21.00939

³³ Albiges L, Gurney H, Atduev V, et al. Pembrolizumab plus lenvatinib as first-line therapy for advanced non-clear-cell renal cell carcinoma (KEYNOTE-B61): a single-arm, multicentre, phase 2 trial. Lancet Oncol 2023; 24: 881–91. doi:10.1016/S1470-2045(23)00276-0

³⁴ Ged Y, Chen Y-B, Knezevic A, et al. Metastatic Chromophobe Renal Cell Carcinoma: Presence or Absence of Sarcomatoid Differentiation Determines Clinical Course and Treatment Outcomes. Clin Genitourin Cancer 2019; 17: e678–88. doi:10.1016/j.clgc.2019.03.018