A Phase I/II Clinical Trial of Pembrolizumab and Cabozantinib in Metastatic Renal Cell Carcinoma

Purpose: Immune checkpoint inhibitor and VEGFR inhibitor combinations are effective treatments for patients with metastatic renal cell carcinoma (mRCC). This phase I/II clinical trial evaluated the safety and efficacy of pembrolizumab and cabozantinib in patients with mRCC. Experimental Design: Eligible patients had mRCC with clear-cell or non-clear cell histology, adequate organ function, Eastern Cooperative Oncology Group 0–1 performance status, and no prior exposure to pembrolizumab or cabozantinib. The primary endpoint was objective response rate (ORR) at the recommended phase II dose (RP2D). Secondary endpoints included safety, disease control rate (DCR), duration of response (DoR), progression-free survival (PFS), and overall survival (OS). Results: Forty-five patients were enrolled. A total of 40 patients were treated at the RP2D of pembrolizumab 200 mg i.v. every 3 weeks and cabozantinib 60 mg orally once daily, 38 of which were evaluable for response. The ORR was 65.8% [95% confidence interval (CI), 49.9–78.8] for all evaluable patients [78.6% as first-line therapy, 58.3% as second-line therapy]. The DCR was 97.4% (95% CI, 86.5–99.9). Median DoR was 8.3 months (interquartile range, 4.6–15.1). At a median follow-up of 23.54 months, the median PFS was 10.45 months (95% CI, 6.25–14.63) and median OS was 30.81 months (95% CI, 24.2–not reached). The most common grade 1 and/or 2 treatment-related adverse events (TRAE) were diarrhea, anorexia, dysgeusia, weight loss, and nausea. The most common grade 3 and/or 4 TRAEs were hypertension, hypophosphatemia, alanine transaminase elevation, diarrhea, and fatigue. There was one grade 5 TRAE of reversible posterior encephalopathy syndrome related to cabozantinib. Conclusions: Pembrolizumab and cabozantinib treatment in patients with mRCC demonstrated encouraging preliminary efficacy and a manageable toxicity profile comparable with other available checkpoint inhibitor-tyrosine kinase inhibitor combinations. Trial Registration: ClinicalTrials.gov Identifier: NCT03149822 https://clinicaltrials.gov/ct2/show/NCT03149822 Significance: This study evaluated the safety and effectiveness of the combination of pembrolizumab and cabozantinib in patients with mRCC. The safety profile was manageable. The combination showed promising activity with an objective response rate of 65.8%, median PFS of 10.45 months, and median OS of 30.81 months.


Introduction
Kidney cancer is an important cancer worldwide with an estimated 431,000 new cases globally and 81,800 new cases in the United States with a 5-year survival of 13% for stage IV disease (1,2). The development of multiple VEGFR programmed cell death protein 1 (PD-1) cell surface membrane receptor, as a second-or subsequent-line (2 L) treatment in 2015 heralded the CPI era for mRCC (7). Multiple combination therapies are now approved in the frontline setting including nivolumab and ipilimumab, pembrolizumab and axitinib, avelumab and axitinib, nivolumab and cabozantinib, and pembrolizumab and lenvatinib (8)(9)(10)(11)(12).
In this phase I/II, open-label study, we evaluated the safety, tolerability, and antitumor activity of pembrolizumab and cabozantinib in patients with mRCC.

Patient Eligibility
Eligible patients had histologically confirmed, locally advanced, recurrent, or mRCC. Patients with clear-cell RCC (ccRCC) or non-clear cell RCC (nccRCC) were included. Key inclusion criteria were Eastern Cooperative Oncology Group 0-1 performance status, adequate organ function, and resolution of prior treatment-related toxicities to Common Terminology Criteria for Adverse Events (CTCAE) v4.0 grade 1 or less. Key exclusion criteria were prior treatment with pembrolizumab or cabozantinib; prior anticancer mAb therapy within 4 weeks and prior anticancer targeted small-molecule therapy within 2 weeks of day 1 of study treatment; active autoimmune disease requiring systemic treatment in the past 2 years, diagnosis of immunodeficiency, current systemic steroid therapy equivalent to ≥10 mg/day of prednisone; history of osteonecrosis of the jaw, reversible posterior leukoencephalopathy syndrome (RPLS), pneumonitis, or wound dehiscence within 6 months of study; or uncontrolled or significant intercurrent cardiovascular, gastrointestinal, or pulmonary disorders. There was no requirement or limit on the number of prior anticancer therapies.

Study Design and Treatment
Phase I dose escalation was done using a standard 3 + 3 design. Phase II dose expansion was done using a Simon two-stage design (19)(20). Pembrolizumab was administered at a fixed dose of 200 mg by intravenous infusion every 3 weeks each 21-day cycle for up to 35 cycles in the first course. Patients could resume a second course of up to an additional 17 cycles of pembrolizumab treatment if they had disease progression after completing the first course. Cabozantinib was taken orally once daily on a continuous basis at a dose of 40 mg orally once daily or 60 mg orally once daily in phase I, and at the recommended phase II dose (RP2D) in phase II. Cabozantinib dose reductions and interruptions were allowed for management of adverse events (AE). Pembrolizumab was held if needed for AEs. Patients continued treatment in the absence of unacceptable toxicity or clinically compelling disease progression. Treatment beyond progression was allowed if the patient met prespecified criteria and was continuing to derive clinical benefit.
The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice Guidelines. The protocol was approved by the Institutional Review Board and all patients provided written informed consent prior to study procedures.

Study Endpoints
The primary endpoint of the overall study was objective response rate (ORR) in patients treated at the RP2D of pembrolizumab and cabozantinib. Secondary endpoints included dose-limiting toxicities (DLT), MTD, disease control rate (DCR), progression-free survival (PFS), overall survival (OS), duration of treatment, time to response, and duration of response (DoR). Exploratory endpoints included PD-L1 status of archival tumor samples and serum bone turnover markers in patients with bone metastases.

Antitumor Activity Assessments
Tumor assessments were performed at baseline and every 9 weeks on study. Tumor responses were evaluated based on investigator assessment per RECIST 1.1 (21). ORR was defined as the sum of complete response (CR) and partial response (PR). DCR was defined as the sum of CR + PR + stable disease (SD).
Patients with missing or no response assessments were classified as not evaluable. PFS was defined as the time from study treatment initiation to the first occurrence of documented disease progression or death from any cause during the study. OS was defined as the time from study treatment initiation to death from any cause.

Safety Assessments
DLT was defined as any treatment-related AE (TRAE) grade 3 or higher that occurred during the 21-day DLT assessment window. Any dose-escalation patient who did not complete the DLT period for any reason other than a DLT was replaced by an additional patient at that same dose level. DLT-evaluable patients were required to have received the planned cycle 2 dose of pembrolizumab within 7 days of the planned administration date and to have taken 75% or more of the planned doses of cabozantinib in cycle 1. The MTD was defined as the highest dose level with no more than 1 DLT reported in 6 DLTevaluable subjects. The RP2D of cabozantinib was selected on the basis of the clinical data, not exceeding the MTD. If <2/6 subjects experience a DLT at 60 mg daily during dose escalation, then 60 mg daily would be considered the RP2D.
AEs were recorded throughout the study and during the follow-up period and graded according to NCI CTCAE Version 4.0 (22). AEs were characterized in terms regarding seriousness, causality, toxicity grading, and action taken regarding trial treatment. Serious AEs (SAE) were defined as any AE that was life threatening; resulted in death, persistent or significant disability/incapacity, hospitalization, or prolongation of an existing inpatient hospitalization; or other important medical events. percentage of tumor that has low (1+), moderate (2+), and high (3+) intensity.

Exploratory Assessments
MPS ≥ 1 is considered positive. Bone turnover markers including β C-terminal telopeptide (β-CTX), bone-specific alkaline phosphatase (BS-ALP), and procollagen type 1 amino-terminal propeptide (P1NP) were collected at baseline and on treatment in patients with bone metastases.

Statistical Analysis
The planned enrollment was estimated to be 6-9 evaluable patients in phase I and 20-38 evaluable patients in phase II. For the primary endpoint of ORR, a Simon two-stage design was used to test the null hypothesis that ORR ≤ 0.20 versus the alternative that ORR ≥ 0.50. After enrollment of 20 evaluable patients in the first stage, the trial would be terminated if 2 or fewer patients respond. If the trial continues to the second stage, a total of 38 patients would be studied. Patients from the phase I dose escalation who were treated at the RP2D were included in the efficacy assessments in Simon stage 1 of the phase II dose expansion.
The ORR, along with a 95% confidence interval (CI), was estimated. The PFS and OS survival curves and medians were estimated using the Kaplan-Meier method and reported with the corresponding two-sided 95% Brookmeyer-Crowley CI, from the on-treatment date to progressive disease (PD), death, or data cut-off date. All statistical analyses were performed by an independent statistician to ensure unbiased data review and conducted on R version 4.1.0, R Core Team (2021).

Data Availability Statement
Source data generated in this study are not publicly available due to compromise of patient privacy. Derived data supporting the findings of this study may be available from the corresponding author upon request.

Patients
Between October 2017 and August 2020, 45 patients were enrolled (8 patients in phase I and 37 patients in phase II) across five sites within the University of Colorado/UCHealth system. A total of 40 patients were treated at the RP2D and were evaluable for PFS and OS, and 38 patients were evaluable for response ( Fig. 1). Baseline characteristics for all patients are listed in Table 1

MTD and DLT Evaluation
Eight patients were enrolled in phase I [5 patients in the first cohort of pembrolizumab 200 mg and cabozantinib 40 mg (2 were not evaluable for DLT due to missing ≥25% of the planned cabozantinib doses in cycle 1, not related  Fig. 2A

Exploratory Studies
There were 44 samples of archival tissue available for PD-L1 testing, 40 of which

Discussion
We conducted this phase I/II study of pembrolizumab and cabozantinib beginning in 2017, prior to the FDA approvals of combination CPI-TKI therapies including pembrolizumab and axitinib in 2019, and nivolumab and cabozantinib and pembrolizumab and lenvatinib in 2021. In our study, the primary endpoint of ORR for all patients treated at the RP2D was met, with ORR 65.8%.
The ORR was 73.3% among 1 L patients and 56% among 2 L patients. These results are better than historical results from cabozantinib alone (33% in 1 L and 21% in 2 L+; refs. 7, 13) and similar to the reported ORR for 1 L CPI-TKI studies of axitinib and pembrolizumab (59.3%), cabozantinib and nivolumab (55.7%), lenvatinib and pembrolizumab (71%), as well as the 2 L cabozantinib plus atezolizumab study where ORR was 53% and 58% in the in the 40 mg and 60 mg ccRCC groups, respectively (9,11,12,24). Compared with other  [24][25][26]. While combination CPI-TKI therapy is well established in the 1 L setting, there are very few studies evaluating combination therapy in the 2 L setting. Our study shows preliminary activity of pembrolizumab and cabozantinib in both the 1 L and 2 L settings for mRCC, suggesting that this combination could be efficacious even after prior CPI and VEGFR exposure.
The safety profile of pembrolizumab plus cabozantinib was manageable and generally consistent with AEs reported with other CPI-TKI combinations. As with the phase I nivolumab and cabozantinib study (27), the AEs related to pembrolizumab and cabozantinib were manageable with early recognition and management of immune-mediated and non-immune-mediated AEs and dose reduction of cabozantinib. While the phase III nivolumab plus cabozantinib combination uses a starting cabozantinib dose of 40 mg/day (28), in our study, the RP2D of cabozantinib in combination with pembrolizumab is 60 mg/day. However, 70% of patients required dose reduction of cabozantinib. This is higher compared with 60% and 46% of cabozantinib monotherapy patients with dose reductions in the METEOR and CABOSUN studies, respectively (7,13). In general, the incidence of grade 3 or 4 TRAEs with pembrolizumab and cabozantinib in our study was similar to that reported for the cabozantinib plus nivolumab and cabozantinib plus atezolizumab studies (24,28). As with other CPI-TKI combinations, the overlapping toxicities seen with pembrolizumab plus cabozantinib include diarrhea, elevated liver enzymes, and hypothyroidism. There were no new safety signals in terms of general TRAEs and immune-mediated AEs.
Our exploratory analyses included PD-L1 tumor testing in all patients with archival tissue and bone turnover markers in patients with bone metastases at baseline. In contrast to previously seen responses to pembrolizumab in advanced RCC with positive PD-L1 expression (29), there was no correlation between MPS and ORR in our patients. The degree of MPS positivity also did not correspond with degree of response, as has been demonstrated in nonsmall cell lung cancer (30). Bone turnover markers were infrequently elevated in patients with bone metastases, indicating their limited diagnostic utility for bone metastases in RCC (31). β-CTX, a marker of bone resorption specific to collagen type 1, most accurately detects PD and PR (32). BS-ALP, a product of osteoblasts, and P1NP, a product of proliferating osteoblasts and fibroblasts, are both markers of bone formation (32). P1NP activity decreased in all patients, making any correlation to treatment possibly coincidental and uninterpretable. Our study was initially designed with historical comparisons with single-agent cabozantinib. As cabozantinib monotherapy has approval in both 1 L and ≥2 L settings for mRCC, we also included patients regardless of line of therapy. As ORR is expected to be highest with 1 L and lower with 2 L therapies, combining these cohorts could have overestimated the response in the 2 L setting and underestimated the response in the 1 L setting. What we observed was ORR of 65.8% in all response-evaluable patients, ORR 73.3% (11/15 patients) in 1 L and ORR 56% (14/25 patients) in the 2 L setting, which is similar to the reported ORR with other combinations in 1 L or 2 L settings, and higher when compared with cabozantinib alone in these settings.

Conclusions
The combination of pembrolizumab and cabozantinib treatment in patients with mRCC demonstrated encouraging preliminary efficacy that is

Role of the Funder/Sponsor
This was an investigator-initiated study coordinated by the University of Colorado Cancer Center (sponsor) and E.T. Lam (principal investigator) who were responsible for administering the study and coordinating the subsites involved in the study. The funder did not have a role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation or decision to submit the article. The funder has reviewed and approved of the article for publication.