Early Clinical Experience with Cabozantinib for Advanced Renal Cell Carcinoma in the UK: Real-World Treatment Pathways and Clinical Outcomes

Please cite this article as: Balaji Venugopal , Manon Pillai , Thomas Powles , Philip Savage , Agnieszka Michael , Kate Fife , Bhupinder Klair , Valerie Perrot , Bernadett Szabados , Early Clinical Experience with Cabozantinib for Advanced Renal Cell Carcinoma in the UK: RealWorld Treatment Pathways and Clinical Outcomes, Clinical Genitourinary Cancer (2021), doi: https://doi.org/10.1016/j.clgc.2021.09.005


Introduction
Among patients with renal cell carcinoma (RCC), approximately one-third present with metastatic disease at the time of diagnosis, 1,2 and one third of patients undergoing radical nephrectomy will relapse. 3 One current treatment for patients with advanced RCC (aRCC) is cabozantinib, an oral inhibitor of multiple receptor tyrosine kinases, including vascular endothelial growth factor (VEGF) receptors, the hepatocyte growth factor receptor protein MET, and the GAS6 receptor AXL. 4 In the phase 3 METEOR trial (NCT01865747), cabozantinib significantly improved clinical outcomes (overall survival [OS], progression-free survival [PFS] and objective response rate [ORR]) compared with everolimus in patients with aRCC who had progressed after prior VEGF-targeted therapy. 5,6 Furthermore, the CABOSUN trial (NCT01835158) reported significant clinical benefit in PFS for treatment-naïve patients with intermediate-or poor-risk metastatic RCC treated with cabozantinib compared with sunitinib. 7,8 The findings from these trials led to the approval of cabozantinib for treatment-naïve, intermediate-or poor-risk patients with aRCC, or those who have received prior VEGF-targeted therapy by the European Medicines Agency. 9 While under assessment by the National Institute for Health and Care Excellence (NICE) and Scottish Medicines Consortium (SMC) in 2016/17 for reimbursement in patients with aRCC following prior VEGF-targeted therapy, cabozantinib was made available in the UK to patients with aRCC via a managed access programme (MAP). Cabozantinib was approved for reimbursement by NICE in early August 2017 for adults with aRCC after VEGF-targeted therapy, 10 at which point it was eligible for reimbursement in England and Wales for this indication and the MAP was discontinued.
Given the limited data available on the real-world use of cabozantinib after prior VEGF targeted therapy in the UK, 11 the present study (Clinical Experience with cabozantinib in patients with advanced REnal cell carcinoma in the UK Study [CERES], NCT03696407) aimed to describe treatment pathways and treatment-related outcomes in patients with aRCC treated with cabozantinib via the UK MAP .

Study Design and Patient Population
CERES was a multicentre, retrospective, non-interventional study involving patients with aRCC treated with cabozantinib after prior VEGF targeted therapy through the MAP, between August 2016 and July 2017, at six specialist centres across the UK (Error! Reference source not found.). The MAP closed to new patients after the approval of cabozantinib by NICE in August 2017. 10 The study period included: i) a pre-cabozantinib initiation period (evaluating patient and disease characteristics, and prior therapies); ii) a post-cabozantinib period with an 'index date' defined as the date of the first cabozantinib dose; and iii) a 24-month follow-up period immediately after cabozantinib initiation or until the patient's death (whichever occurred earlier) during which treatment was administered in accordance with institutions' standard-of-care.
Eligible patients were aged 18 years or older at cabozantinib initiation, had a diagnosis of aRCC, and initiated cabozantinib via the MAP. Patients were identified by local clinical staff through a retrospective review of hospital medical records (pharmacy records, databases or electronic prescribing systems). Patients were excluded if their medical records were unavailable.
The study was performed in accordance with the recommendations of the Declaration of Helsinki and the International Ethical Guidelines for Epidemiological Studies. All the ethical, governance and legal compliance of the study protocol and supporting documents were approved by relevant Research Ethics Committees before the study start. For patients who were alive at the time of data collection, only those who had provided written informed consent were included. For deceased patients, data were collected and anonymized by members of the direct clinical care team to preserve confidentiality, in accordance with the UK National Health Service Confidentiality Code of Practice.

Endpoints and Outcomes
Baseline characteristics and treatment patterns Baseline patient demographics and clinical characteristics were described at date of cabozantinib initiation (i.e., index date). Treatment patterns were analysed over the 24

Tolerability
Cabozantinib safety data were collected and entered into the Global Safety Database as per the requirements of the MAP, but no explicit safety objectives were included in the study design.
Patterns of dose interruptions, reductions and discontinuations do, however, provide inferential evidence of treatment tolerability.

Statistical Analysis
Descriptive statistics were used to characterize the study population and prescribing patterns (mean A minimum sample size of 100 patients was specified to ensure adequate precision in the calculation of proportions and in the median (95% CI) PFS estimates for the FAS and pre-specified subgroups.
The calculation of precision was based on the median (95% CI) PFS reported for cabozantinib in the METEOR study (7.4 [6.6-9.1] months). 6 A post hoc analysis stratified outcomes by Charlson Comorbidity Index (CCI) total score (≤ 6 and > 6) to assess the impact of baseline comorbidities on patient survival. The CCI weights each prespecified comorbidity according to its associated risk of death (scores 1-6); a higher CCI total score indicates a higher risk of death. 12 The presence of metastatic solid tumour accounts for 6 points and was chosen as a cut-off point.

Patient Characterization
In total, 106 patients were enrolled in the study, of whom 100 had initiated cabozantinib between tumours, the most common comorbidities were moderate-to-severe chronic kidney disease (investigator-defined, 52.0%) and uncomplicated diabetes mellitus (11.0%) (Error! Reference source not found.). Median (IQR) CCI total score was 8.0 (6.0-8.0) and 63.0% of patients had a total CCI score of more than 6.

Early Versus Late MAP Treatment and Cabozantinib Treatment Line
Demographic characteristics were generally balanced between the early (n = 57) and later (n = 43) MAP treatment subgroups (Error! Reference source not found.). Compared with the later subgroup, the early subgroup had a higher median number of prior lines of therapy (1.0 versus 2.0, respectively), a higher proportion of patients with moderate-to-severe kidney disease (41.9% versus 59.6%, respectively), and a higher proportion of patients with at least one dose reduction (60.5% versus 70.2%, respectively) (Supplemental Tables 4 and 5). In contrast, there was a trend towards poorer performance status and higher IMDC risk score in the later (vs early) subgroup (Supplemental Table 4). The proportion of patients with temporary treatment interruptions was lower in the early subgroup than the later subgroup (19.3% versus 27.9%, respectively) (Error! Reference source not found., Error! Reference source not found., Error! Reference source not found.). In both groups, AEs were the most common cause of dose reductions and interruptions.
Similar proportions of patients (> 80%) discontinued treatment in each subgroup, most commonly owing to disease progression and AEs.
The distribution of anticancer tyrosine kinase inhibitors (TKIs) prescribed prior to cabozantinib initiation was broadly similar between the subgroups, most commonly sunitinib or pazopanib followed by axitinib. In both groups, nivolumab was most commonly prescribed after cabozantinib (Error! Reference source not found.).

12
PFS and OS were similar for the early and later MAP treatment subgroups, and when assessed by cabozantinib treatment line (Figure 2).

CCI Subgroups
Demographic characteristics between the post hoc subgroups of patients with a CCI total score of 6 or less (n = 37) and those with a CCI total score of greater than 6 (n = 63) are shown in Error! Reference source not found.. The proportion of patients with a poor risk score (based on IMDC categorization) was higher in the subgroup with a CCI total score of greater than 6 than in the subgroup with a CCI total score of 6 or less. Assessment of the impact of baseline comorbidities on clinical outcomes suggested an association between lower CCI total score (≤ 6 versus > 6) and

Discussion
This retrospective analysis of data from the CERES Study contributes to the limited body of realworld data on the use of cabozantinib for the treatment of patients with aRCC in routine care.
Within this unselected population, cabozantinib was most commonly prescribed as 2L therapy (41.0%). The majority of patients (84%) initiated cabozantinib at the recommended dose of 60 mg/day, but approximately two-thirds of patients had at least one subsequent dose reduction.
Cabozantinib demonstrated clinical activity in the full study population, and across all treatment lines. In the post hoc CCI subgroup analyses, median (95% CI) OS and PFS were significantly longer in patients with a lower burden of comorbid disease at baseline (CCI score of 6 or lower) compared with those with a higher burden of disease (CCI score greater than 6). These findings suggest that CCI total score may hold potential as a prognostic and/or predictive indicator in patients with aRCC. [13][14][15][16] Given the lack of available evidence on the use of cabozantinib in routine practice in the UK, this multicentre study provides valuable insights into how cabozantinib was used in a broad, unselected real-world patient population, and the clinical outcomes achieved. The findings serve as a benchmark against which to measure the effectiveness of RCC treatments in routine practice, as well as the impact of future changes in UK clinical practice. Furthermore, the sample size was informed by the median PFS estimates from the phase 3 METEOR trial, 6 which contributes to the reliability and robustness of the current analysis.
CERES was limited by its retrospective nature and by the use of data that were collected for the purposes of the MAP rather than for clinical research, limiting the scope of the variables collected and the level of data validation. PFS, for example, was evaluated in accordance with local or RECIST findings (or death) and was not adjudicated by a central and independent review of radiological assessments across the participating centres. Caution must be taken when comparing median PFS for the CERES population with that reported for the METEOR randomized controlled trial (RCT) because of these differences in outcome validation.  18,19 these treatment patterns have become outdated, which may limit the ability to extrapolate these findings to current patient populations.
Finally, although the results of the post hoc CCI analysis are of clinical interest, the study sample size was powered for precision in the treatment pathway and PFS outcomes. Therefore, results of the subgroup and post hoc analyses should be considered as hypothesis-generating only.
The CERES patient population had a substantial burden of comorbid disease: 63.0% had a CCI score of greater than 6, 19% were categorized with an ECOG PS score of at least 2, and over a fifth of patients were categorized as poor risk according to IMDC risk scores. As is common when comparing real-world and RCT populations, the patients enrolled in CERES had more severe disease than those in the METEOR trial of cabozantinib in patients with RCC who had progressed after prior VEGFtargeted therapy. In METEOR, cabozantinib was used as 3L or later-line therapy in approximately 30% of patients (compared with 59% in CERES); no patients in METEOR had an ECOG PS of 2 or higher and only 16% (in the primary PFS analysis) had a MSKCC poor-risk classification. 5

Conclusion
The CERES study provides valuable insights into how cabozantinib was used in a broad, unselected  Exploratory analyses of the CERES data suggest that Charlson Comorbidity Index total score may be worth further investigation as a potential prognostic indicator in patients with aRCC.