Cabozantinib Plus Nivolumab in Adult Patients with Advanced or Metastatic Renal Cell Carcinoma: A Retrospective, Non-Interventional Study in a Real-World Cohort/GUARDIANS Project

Simple Summary ICI-based combinations have led to a significant change in mRCC medical treatment. However, data from real-world (RW) cohorts are rare. In this multicenter study, we evaluated the safety and effectiveness of cabozantinib plus nivolumab in real-world cohorts from centers in Germany, Switzerland, and Austria. The median PFS in the overall cohort was 18.6 months. We also analyzed subgroups in relation to the IMDC, histology, and with regard to the presence of bone metastasis. In summary, our real-word data support the promising efficacy data of the pivotal trials, particularly in patients with non-ccRCC and those without bone metastases at the start of the treatment. Abstract Introduction: Combinations of immune-checkpoint inhibitors (ICIs) are the standard of care (SOC) for treatment-naive metastatic renal cell carcinoma (mRCC) patients. In this multicenter study, we evaluated the RW safety and efficacy of cabozantinib plus nivolumab in mRCC patients. Methods: Data were retrospectively collected from twelve cancer centers in Germany, Switzerland, and Austria. Patients with advanced or mRCC were eligible. The investigator-based objective response rate (ORR) and progression free survival (PFS) were calculated from the start of the treatment to progression or death. Descriptive statistics and Kaplan–Meier (KM) plots were utilized where appropriate. Results: In total, 96 eligible patients (66.6% male) with a median age of 66.0 years were included. The most common histology was clear-cell RCC (ccRCC) in 63.4% (n = 61). A prior nephrectomy was performed in 60.4% (n = 58). ECOG 0-1 was 68.8% (n = 66). A partial response was documented in 43.8% of patients (n = 42), a stable disease in 32.3% (n = 31), and a progressive disease in 8.3% (n = 8) as the best overall response. Response data were not evaluable in 13.5% (n = 13). The median follow-up time was 12.7 months (95% CI, 10.0–15.3). The PFS rate at 6 months was 89.8% in the overall population (86.8% for ccRCC; 90.0% for non-ccRCC). Adverse events (AEs) were reported in 82.3% (n = 79) for all grades and 41.7% (n = 40) for grades 3–5. Elevated liver enzymes (34.4%), diarrhea (31.3%), and hand–foot syndrome (29.2%) were the three most frequent AEs of any grade and causality. Discussion/Conclusions: In this real-world cohort of mRCC patients, the application of cabozantinib plus nivolumab was shown to be safe and feasible. Our data support the use of cabozantinib plus nivolumab as a first-line standard therapy in mRCC patients. Major limitations were the retrospective data capture and short follow-up time of our study.


Introduction
Renal cell carcinoma (RCC) accounts for 2% of tumor diseases worldwide, and its incidence has been increasing over the past decades [1,2].Histologically, clear-cell renal cell carcinoma (ccRCC) is the dominant subtype, accounting for 75-80% of renal cancers [3].Antisystemic cancer treatment is the main approach in patients with advanced, metastatic, or recurrent RCC who are not candidates for local therapies.ccRCC in particular is resistant to conventional chemotherapy [4].RCC is characterized by increased angiogenesis, caused by loss of the VHL gene [5].So, tyrosine kinase inhibitors (TKIs), targeting the vascular endothelial growth factor receptor (VEGFR), and ICIs have become pivotal in the systemic treatment of advanced or metastatic renal cell carcinoma (mRCC).
Recently, ICI-based combinations have been developed, leading to a significant change in mRCC medical treatment [14][15][16].Currently, there are five approved first-line combinations, including the combination of cabozantinib plus nivolumab.
The combination of ICIs and TKIs is established in the first-line treatment of mRCC and includes the possibility of combining pembrolizumab with lenvatinib [17], pembrolizumab with axitinib [18], and avelumab with axitinib [19].The current long-term data of the respective approval study confirm the clinical benefit [18,20,21].Furthermore, the CheckMate 9ER trial investigated the combination of cabozantinib plus nivolumab as a first-line therapy in mRCC patients.The progression free survival (PFS) of this combination was significantly longer compared to sunitinib at 16.6 months versus 8.3 months [22].With a median followup of 55 months, the most recent analysis presented mature survival data.The median overall survival was 46.5 versus 36.0 months in favor of the ICI combination [HR 0.77 (95% CI 0.63-0.95)][23].This effect was also shown in patients with papillary, unclassified, or translocation-associated RCC.In the phase II trial of cabozantinib plus nivolumab, the median PFS of patients with non-clear-cell RCC was increased to 12.5 months (95% CI 6.3-16.4)[24].These results of the aforementioned studies continue to support cabozantinib plus nivolumab as a standard of care for previously untreated mRCC.
However, the wide range of medical treatment options allows for individualized therapy selection to identify the most appropriate treatment option for a patient.The absence of a direct comparison between contemporary ICI combinations renders the interpretation of clinical data important.The complete response (CR) and deep response after systemic treatment are associated with the best long-term results in patients with mRCC.With the current follow-up of 55 months, the CR rate for cabozantinib plus nivolumab was 13.6% (sunitinib 4.6%) [23].ICI-TKI combinations offer a broad clinical activity with a high re-sponse rate (55-71%).The potential of these combinations becomes more evident when the low primary progression rate (5-10%) is taken into account.This broad activity has led to the development of the most effective treatments of mRCC with a median PFS in the range of 15-22 months.These strengths of the ICI-TKI combinations are of particular interest in treating symptomatic patients.Combinations with 3rd-generation TKIs (cabozantinib) have a low progression rate (6.5%) and therefore offer advantages for particularly vulnerable patients.Nevertheless, ICI-TKI combinations also have the highest rates of grade ≥ 3 AEs.However, there are differences between the ICI-TKI combinations in the incidence rate of specific AEs.
Data from real-world cohorts are rare.However, real-world patients are characterized by dismal characteristics compared to patients in clinical trials.In particular, data on relevant subgroups such as the response for patients with intermediate or poor IMDC or the presence of bone metastasis are still missing.In this multicenter study, we evaluated the safety and effectiveness of cabozantinib plus nivolumab in real-world cohorts from centers in Germany, Switzerland, and Austria.

Patient Cohort and Treatment
Clinical data were collected from twelve oncological centers (in alphabetical order: Cologne, Düsseldorf, Essen, Hannover, Heidelberg, Herne, Jena, Munich, Münster, Regensburg, St. Gallen, and Vienna).Data were retrospectively retrieved from medical records.Patients with advanced or mRCC were eligible.Patients receiving cabozantinib and nivolumab as a standard first-line treatment were included.Treatment was applied according to local standards.Cabozantinib 40 mg orally plus nivolumab 240 or 480 mg i.v. was mandatory and administered according to routine care.Dose adjustments were made at the discretion of the investigators.In total, 21.9% (n = 21) required dose reductions or interruptions.
Throughout the duration of this study, the ethical standards of the institutional and/or national research committee and the 1964 Declaration of Helsinki, as subsequently amended, were adhered to.This study was approved by the local Ethics Committee of the Medical Faculty of the University of Duisburg-Essen (22-10567-BO).

Statistical Analysis
Pseudonymized data sets were analyzed using SPSS-Statistics 29.0 (Armonk, NY, USA).The primary outcome measures were the best overall objective response rate (ORR) and PFS.PFS was defined as the time from the start of therapy to the date of radiologic or clinical progression or death.ORR was evaluated by investigators according to the Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST 1.1).Adverse events (AEs) were defined according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0.For survival analyses, the Kaplan-Meier method was used.The duration of the follow-up time was calculated from the date of treatment initiation to either the date of death or censored at the last-known follow-up.Overall, two-sided p-values ≤ 0.05 were considered statistically significant.

No. (%)
All Patients (N = 96) Treatment-related adverse events, any grade 79 (82.3) Grade ≥ 3 TRAEs 40 (41.7)Treatment-related adverse events resulting in any treatment discontinuation 24 (25) Treatment-related adverse events leading to death 1 (1.0) AEs of any cause led to discontinuation in 25% of patients.Overall, one death was considered to be treatment related (pneumonia) by investigators.

No. (%) All Patients (N = 96)
Treatment-related adverse events, any grade 79 (82.3)Grade ≥ 3 TRAEs 40 (41.7)Treatment-related adverse events resulting in any treatment discontinuation 24 (25) Treatment-related adverse events leading to death 1 (1.0) AEs of any cause led to discontinuation in 25% of patients.Overall, one death was considered to be treatment related (pneumonia) by investigators.

Discussion
To our knowledge, this is the first retrospective multicenter study exploring the outcomes of real-world patients treated with the ICI combination of cabozantinib plus nivolumab in a first-line therapy setting.
We report the data of 96 patients from 12 academic hospitals in Germany, Switzerland, and Austria.
With a median age of 66.0 years at the start of treatment, our cohort is older than those in the CheckMate 9ER trial.Furthermore, our cohort differs from the CheckMate 9ER trial in different aspects.Firstly, we have included patients with non-ccRCC.Secondly, we identified only 40 patients (41.7%) with ECOG 0. Thus, a larger proportion of this cohort had a worse performance status compared to 80% of patients with a Karnofsky performance status of 90-100% (corresponds to ECOG 0) in the CheckMate 9ER trial at therapy initiation.However, the median PFS in the overall cohort was 18.6 months, which was higher than reported in 55-month follow-up of the CheckMate 9ER trial [23].However, our results are consistent with previous data that suggests cabozantinib may enhance immune-checkpoint inhibition [10][11][12][13]25].
Interestingly, the mPFS was similar between patients with a favorable IMDC risk (mPFS 17.2 months) and intermediate/poor IMDC risk (mPFS 16.3 months).The current update of the CM9 trial shows significant differences between these two groups in favor of favorable IMDC [26].The proportion of patients with an unknown IMDC status (mPFS 30.6 months) may have influenced our result.In our cohort, the mPFS was significantly different between the subgroups with ccRCC (16.8 months) and non-ccRCC (30.6 months).The ccRCC subgroup included ccRCC with sarcomatoid differentiation (n = 8; 8.3%), which is known to exert explicit sensitivity for ICI therapies.However, the small number of patients with sarcomatoid differentiation (mPFS 6.9 months) may have had an impact on our results.Our RW results were comparable to the findings of a phase II trial, which tested cabozantinib plus nivolumab in patients with non-ccRCC subtypes [24].
The presence of bone metastasis in RCC is a well-known risk factor according to a lower life expectancy and loss of quality of life [27][28][29].In this analysis, we have demonstrated that patients with bone metastases had a shorter PFS than patients without metastases (14.4 vs. 25.8 months).
The reported toxicity was lower in our cohort when compared to the prospective CheckMate 9ER trial (82.3% for all grades and 41.7% for grade 3-5 treatment-related adverse events versus 99.7% and 75.3%).This can be explained by the retrospective character of the study and the focus on clinically relevant AEs in routine practice.All 12 sites are experienced clinical trial units which regularly participate in clinical trials and are familiar with CTCAE reporting.However, in daily clinical practice, the documentation of AEs seems to be limited to those that most severely affect patient quality of life.
While some AEs, such as hypertension, were noted in at least 34.7% of patients in the prospective CheckMate 9ERtrial (resulting in it being the third most common adverse event), only 11 patients (11.5%) had documented hypertension in our cohort.In contrast, elevated liver enzymes were reported in a higher fraction of patients compared to the aforementioned prospective trial [34.4% vs. 28.1% (ALT level)/25.3%(AST level)].Due to differences in reporting, these numbers are difficult to compare.
Our RW data also showed that patients should be informed about diarrhea as one of the most common side effects, which is comparable to the CheckMate 9ER trial (diarrhea grade 1-2 in 53% of patients and grade 3 in 6% of patients).PPE was a relevant finding and affected 29.2% of our patients, which was grade 3-4 in nine of our patients.In addition, we observed one fatal case of pneumonia.
The limitations of our study are its retrospective character with selection and recall bias as well as the short follow-up time.We did not utilize a radiologic blinded independent central review in our analysis.Moreover, the study population and subgroups are too small to perform an informative multivariable analysis.

Table 4 .
Summary of treatment-related adverse events.