Cancer Treatment Reviews 89 (2020) 102062 Contents lists available at ScienceDirect Cancer Treatment Reviews journal homepage: www.elsevier.com/locate/ctrv Anti-tumour Treatment Exposure-response modeling of cabozantinib in patients with renal cell T carcinoma: Implications for patient care ⁎ Daniel Castellanoa, , Jose Pablo Marotob, Fawzi Benzaghouc, Naila Taguievac, Linh Nguyend, Douglas O. Claryd, Eric Jonasche a Hospital Universitario 12 de Octubre, Madrid, Spain b Hospital de la Santa Creu i Sant Pau, Barcelona, Spain c Ipsen Pharma, Boulogne-Billancourt, France d Exelixis, Inc., Alameda, CA, USA e Department of Genitourinary Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA A R T I C LE I N FO A B S T R A C T Keywords: Cabozantinib is an oral tyrosine kinase inhibitor (TKI) approved for the treatment of patients with advanced Cabozantinib renal cell carcinoma (RCC) at a dose of 60 mg/day. As with other TKIs, cabozantinib is associated with high Renal cell carcinoma interpatient variability in drug clearance and exposure that can significantly impact safety and tolerability across Exposure-response a patient population. To optimize cabozantinib exposure (maintaining efficacy and tolerability) for the in- Dose dividual, patients may require treatment interruption with dose reduction (40 mg/day and then 20 mg/day). In Efficacy the pivotal Phase 3 METEOR trial, cabozantinib significantly improved overall survival, progression-free survival Tolerability and the objective response rate compared with everolimus in patients with advanced RCC who had received previous treatment with a VEGFR TKI. Dose reductions were common for patients receiving cabozantinib (60%) but effective as only 9% discontinued treatment due to adverse events (AEs). In this review, we discuss phar- macometric analyses that evaluated the impact of cabozantinib dose on efficacy and safety outcomes during the METEOR study. Exposure-response models demonstrate that the risk of experiencing adverse events and dose reduction is increased in patients with low cabozantinib clearance versus typical clearance and decreased in patients with high clearance. Dose reduction of cabozantinib to manage AEs is predicted to have minimal impact on efficacy as AEs are more likely to occur in patients with low clearance and higher exposure to cabozantinib. These analyses further support a dose modification strategy to optimize cabozantinib exposure for individual patients. Introduction become a standard of care for advanced RCC [6,7]. In these studies, cabozantinib demonstrated clinical benefit over existing standards of The vascular endothelial growth factor (VEGF) pathway is an es- care with a safety and tolerability profile that was manageable with tablished target in clear-cell renal cell carcinoma (RCC) with a number dose modification. of VEGF receptor (VEGFR) tyrosine kinase inhibitors (TKIs) approved Dose modification is a common strategy with TKI therapy to balance for the treatment of patients with advanced disease. As our under- efficacy and tolerability. TKIs, including cabozantinib, are associated standing of RCC biology has advanced, new targets have emerged. with high interpatient variability in drug clearance [8–13]. Variability Cabozantinib is an oral TKI that inhibits VEGFR as well as the novel in clearance leads to wide ranges in steady-state exposure to the drug. targets MET and AXL, which are implicated in RCC growth, metastasis, Generally, patients start TKI treatment at the recommended dose, and if and therapeutic resistance [1]. Based on positive data from the phase 3 needed, the dose is reduced to address overexposure and resolve in- METEOR study (ClinicalTrials.gov, NCT01865747) in patients pre- tolerable adverse events (AEs). However, there is some concern that viously treated with at least one antiangiogenic agent and the rando- dose reduction may impact efficacy. It is, therefore, essential to develop mized phase 2 CABOSUN study (NCT01835158) in treatment-naïve informed strategies to optimize the dose for the individual patient. patients with intermediate- or poor-risk disease [2–5], cabozantinib has Pharmacometric analyses have been used to support and guide ⁎ Corresponding author at: Medical Oncology Service, Hospital Universitario, 12 de Octubre, Avda. De Córdoba, s/n, 28041 Madrid, Spain. E-mail address: cdanicas@hotmail.com (D. Castellano). https://doi.org/10.1016/j.ctrv.2020.102062 Received 5 December 2019; Received in revised form 15 June 2020; Accepted 17 June 2020 0305-7372/ © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/). D. Castellano, et al. Cancer Treatment Reviews 89 (2020) 102062 dosing strategies with TKIs [14–16]. Pharmacometrics integrates erythrodysesthesia (PPE, 42%), and hypertension (37%); and the most pharmacology, statistics, and computational modeling [14]. Population common grade 3/4 AEs were hypertension (15%), diarrhea (11%), fa- pharmacokinetic (popPK) models characterize PK parameters across a tigue (9%), and PPE (8%) [2]. Cabozantinib dose reductions were fre- patient population and identify factors associated with variability in quently employed to manage AEs [2–5]. In METEOR, 40% of patients exposure [17], while exposure–response (ER) models predict the re- receiving cabozantinib maintained the 60-mg/day dose, while 60% lationship of drug exposure with efficacy, safety, and tolerability and required at least 1 dose reduction to 40 mg/day and 20% required a the potential clinical implications of dose modification [15]. second dose reduction to 20 mg/day [2,26]. AEs that led to cabo- To better inform dosing strategies with cabozantinib, popPK and ER zantinib dose reduction included diarrhea (16%), PPE (11%), fatigue models were developed to evaluate the relationship of cabozantinib (10%), and hypertension (8%). While dose reductions were more fre- dose, clearance, and exposure with efficacy, safety, and tolerability quent with cabozantinib than with everolimus (60% vs 25%), the rates during the METEOR study [18–21]. These analyses support dose of discontinuation due to AEs were similar (9% vs 10%). modification as the principal strategy to manage tolerability of cabo- Results from the METEOR and CABOSUN studies supported ap- zantinib while maintaining efficacy [22,23]. In this review, we present proval of cabozantinib by the United States Food and Drug an overview of previously published pharmacometric analyses of ca- Administration for the treatment of patients with advanced RCC, and bozantinib in patients with advanced RCC and discuss the implications approval by the European Medicines Agency for the treatment of pa- of these analyses in terms of patient care and appropriate dose mod- tients with advanced RCC previously treated with a VEGF-pathway ification. inhibitor or treatment-naïve patients with intermediate- or poor-risk disease [6,7]. Clinical outcomes of cabozantinib in RCC Pharmacokinetics of cabozantinib in patients with RCC For both the METEOR and CABOSUN studies, cabozantinib was Cabozantinib PK has been characterized in healthy volunteers and administered orally at a dose of 60 mg/day, with dose interruptions and in patients with various solid tumors, including RCC. In healthy vo- reductions to 40 and 20 mg/day recommended to manage AEs [2,4]. lunteers receiving a single 20-, 40-, or 60-mg dose of cabozantinib, The 60-mg/day dose was selected based on the activity, safety, and maximum plasma concentrations were reached in 3 to 4 h with a mean tolerability of cabozantinib in a phase 1 trial of patients with previously maximum plasma concentration of 343 ng/mL at the 60-mg dose treated metastatic RCC and studies of cabozantinib in other solid tu- [8,27]. With daily oral dosing, the median time to steady-state cabo- mors [24,25]. Supplementary Table 1 summarizes outcomes from key zantinib concentration in patients with solid tumors is approximately clinical studies of cabozantinib in RCC. In the pivotal phase 3 METEOR 15 days with a 4- to 5-fold mean cabozantinib accumulation (based on study, cabozantinib was compared with the mammalian target of ra- area under the curve) compared with single-dose administration [28]. pamycin (mTOR) inhibitor everolimus in patients with advanced RCC Cabozantinib clearance is variable. Cabozantinib is metabolized who had received prior therapy with a VEGFR TKI (N = 658) [2,3]. primarily by the cytochrome P450 (CYP) 3A4 pathway and to a far Cabozantinib improved progression-free survival (PFS) as assessed by lesser extent by CYP2C9, with a relatively long terminal half-life independent review committee (IRC) relative to everolimus (median (99–120 h). Food intake, hepatic impairment, and concomitant use of 7.4 vs 3.9 months; hazard ratio [HR] 0.51; 95% confidence interval [CI] medications that inhibit or induce the CYP3A4 pathway are known to 0.41–0.62; p < 0.0001), as well as overall survival (OS) (median 21.4 affect cabozantinib plasma concentrations [29–32]. In healthy volun- vs 16.5 months; HR 0.66; 95% CI 0.53–0.83; p = 0.00026) and the teers administered 60 mg, the average clearance was 2.35 L/h with a objective response rate (ORR) (17% vs 3%; p < 0.0001) [3]. In the coefficient of variability (CV) of 67% [27]. phase 2 CABOSUN study, cabozantinib was compared with sunitinib as Generally, the PK characteristics of cabozantinib in patients with a first-line treatment in intermediate and poor-risk patients (per Inter- advanced RCC are consistent with those of healthy volunteers (Table 1). national Metastatic Renal Cell Carcinoma Database Consortium [IMDC] As part of the METEOR study, plasma samples were obtained for PK criteria) with metastatic RCC (N = 157) [4,5]. PFS assessed by IRC was assessments from patients assigned to the cabozantinib treatment arm, significantly longer with cabozantinib than with sunitinib at 8.6 versus and a popPK model was developed using data from METEOR patients 5.3 months (HR 0.48; 95% CI 0.31–0.74; p = 0.0008), with a corre- along with healthy volunteers [18,33]. The estimated terminal half-life sponding ORR of 20% versus 9% [5]. for patients with RCC was 99 h. Cabozantinib clearance was variable, The safety and tolerability profile of cabozantinib was consistent with a predicted range of 0.51 to 7.24 L/h (Fig. 1A) [21] and an esti- across the METEOR and CABOSUN studies. In the METEOR study, the mated CV of 46% [26]. In a multiple covariate model, baseline patient most commonly reported AEs of any grade were diarrhea (74%), fatigue characteristics such as age, body mass index, and baseline laboratory (56%), nausea (50%), decreased appetite (46%), palmar-plantar parameters including hemoglobin, bilirubin, alanine aminotransferase, serum albumin, and creatinine clearance did not have a statistically Table 1 significant effect on clearance. Asian race and female gender were as- Cabozantinib PK parameters estimated from popPK base model for patients sociated with lower cabozantinib clearance compared with their re- with RCC and single dose PK data from healthy volunteers [18,26,27] spective counterparts (reductions of − 27% and − 21%, respectively); Parameter RCC PopPK* Healthy Volunteers however, these were deemed not clinically significant given the varia- 60-mg/day dose 60-mg dose (single-dose study) bility in clearance for the overall population [18,26]. t1/2 99 h 111 h The relationship of cabozantinib clearance with dosing was ex- CL/F 2.23 L/h 2.35 L/h plored using data from METEOR [21]. From the popPK model, clear- Coefficient of variability 46% 67% ance for a typical white male patient was estimated at 2.23 L/h [18]. Vc/F 319 L 363 L Additional clearance values of 1.3 and 3.3 L/h were chosen to represent patients with low and high clearance characteristics, respectively CL/F, plasma clearance; h, hour; popPK, population pharmacokinetics; RCC, (Fig. 1A) [21]. At the 60-mg/day dose, the average plasma concentra- renal cell carcinoma; t1/2, terminal elimination half-life; Vc/F, apparent volume of distribution of central compartment. tion of cabozantinib at steady state was estimated to be 1122 ng/mL for * Based on pooled data from healthy volunteers (N = 63) and patients with patients with typical cabozantinib clearance (Fig. 1B), with the con- RCC enrolled in the cabozantinib arm of METEOR with 2 measurable phar- centration increasing by + 71% (1923 ng/mL) for patients with low macokinetic samples (N = 282). PK parameters were estimated from the PopPK clearance but decreasing by − 32% (758 ng/ml) for patients with high Model. clearance [21]. A stepwise dose reduction to 40 or 20 mg/day for a 2 D. Castellano, et al. Cancer Treatment Reviews 89 (2020) 102062 Fig. 1. Cabozantinib in RCC – relationship of clearance with dosing. (A) Distribution of calculated cabozantinib clearance values in patients from the METEOR study (N = 282), with values modeled for low, typical, and high clearance indicated [21]; (B) Predicted cabozantinib concentrations at varying doses with different population PK-derived clearance values (N = 282) [21]; (C) Predicted fractions of patients without dose modification for selected values of cabozantinib clearance at a dose of 60 mg/day (N = 317) (Lacy et al. [20], reprinted under the terms of the Creative Commons Attribution 4.0 International License, http://creativecommons.org/ licenses/by/4.0/; the figure has been modified from the original); (D) Distribution of predicted steady-state cabozantinib Cavg at 60-mg/day dosing for patients in METEOR based on their history of dose reduction (median and quartiles shown) (data on file); (E) Predicted steady-state cabozantinib Cavg for all patients at the 60- mg/day dose and at the final dose level experienced (median and quartiles shown) (data on file). *Based on popPK analyses. Cavg, average plasma concentration; CL/ F, plasma clearance; PK, pharmacokinetics; popPK, population PK; RCC, renal cell carcinoma. patient with lower clearance would bring exposure to a more typical initial 60-mg/day dose to a median of 957 ng/mL when estimated at the level. final dose level of each patient (Fig. 1E), with a corresponding reduc- The probability of dose modification during METEOR increased as tion in the standard deviation from 566 ng/mL to 414 ng/mL (data on clearance decreased; patients with low clearance had a high rate of dose file). modifications during the initial weeks of cabozantinib treatment (Fig. 1C) [20]. Furthermore, the subgroups of patients who dose-re- Cabozantinib Exposure-Response modeling duced to 40- or 20-mg/day dose levels showed higher initial exposure compared with those who did not (data on file) (Fig. 1D). With dose Initial ER models were developed to characterize the relationship of reductions, predicted median cabozantinib exposure across the entire cabozantinib dose with outcomes during METEOR for patients with treated population was lowered from a median of 1340 ng/mL at the typical baseline characteristics and clearance [20]. Given the scope of 3 D. Castellano, et al. Cancer Treatment Reviews 89 (2020) 102062 Fig. 2. Kaplan–Meier estimates of progression-free survival for simulated doses of cabozantinib for a typical white male patient with cabozantinib clear- ance of 2.23 L/h. Lacy et al. [20], reprinted under the terms of the Creative Commons Attribution 4.0 Inter- national License, http://creativecommons.org/licenses/ by/4.0/; the figure has been modified from the original. Predicted PFS for typical individual predicted steady-state average cabozantinib concentration at 20-mg/day (green), 40-mg/day (orange), and 60- mg/day (blue) doses with 95% CIs; based on 172 events in 315 RCC patients from the METEOR study with at least 1 measurable cabozantinib concentra- tion. Time-varying average cabozantinib con- centration was estimated over the 3 weeks prior to a PFS event (Cavg3w); the relationship between Cavg3w and PFS was evaluated using nonlinear models over a range of EC50 values (concentration that gives half-maximal response) for Cavg3w. An EC50 of 100 ng/mL resulted in the best model fit and was used to generate the predicted PFS curves and is considerably lower than the predicted steady-state average cabozantinib concentrations associated with doses of 20 mg, 40 mg, and 60 mg (375, 750, and 1125 ng/mL, respectively). aCabozantinib con- centrations correspond to model-predicted typical individual steady-state average concentrations for the 20-, 40-, and 60-mg/day once-daily dosing regimens. bHR relative to a 1125-ng/mL cabozantinib concentration calculated over the 3 weeks prior to time of event. Estimated PFS EC50 = 100 ng/mL based on best nonlinear ER model. Cavg, average plasma concentration; CI, confidence interval; EC50, concentration of the drug that gives the half-maximal response; HR, hazard ratio; PFS, progression-free survival; RCC, renal cell carcinoma. this review, we have limited our discussion on ER methodology, but have a modest impact on PFS (HR = 1.06) for patients with lower detailed methods are described by Lacy et al. [20]. Cox proportional clearance, while substantially improving AE tolerability (Fig. 3). The hazard models estimated hazards for PFS and select AEs across the 60-, hazard rate of PPE, for example, was predicted to decrease by −49% 40-, and 20-mg/day dosing options. Given the variability in cabo- with a dose of 40 mg/day versus 60 mg/day for patients with low ca- zantinib clearance and its relationship with exposure and dose mod- bozantinib clearance. Conversely, for patients with higher clearance, ification, ER models were then extended to evaluate the relationship of use of the 40-mg/day dose has a more noticeable impact on PFS com- cabozantinib dose with outcomes across the spectrum of patients with pared with 60 mg/day (HR = 1.13), with a corresponding decrease varying clearance values. The models compared 60- and 40-mg/day of −24% in the hazard rate of PPE [21]. doses in patients with low (1.3 L/h), typical (2.23 L/h), and high It is important to note the limitations inherent to ER analyses, for clearance (3.3 L/h) [20,21]. example, the effects of unrecognized confounders and imbalances For the typical patient, starting cabozantinib at a lower dose of 20 or within subgroups [34]. Best practice necessitates the use of data from 40 mg/day was predicted to increase the hazard of disease progression controlled, randomized trials for ER analysis; however, clinical trial or death compared with a 60-mg/day dose (Fig. 2) [20]. A dose of populations may differ considerably from those seen in the clinic, who 20 mg/day was predicted to have notable impact, with a +39% hazard are often older and have more comorbidities and interpatient variation. increase for PFS compared with a dose of 60 mg/day, while the pre- Nevertheless, ER modeling is a powerful tool for benefit-risk evaluation dicted hazard increase for a dose of 40 mg/day versus 60 mg/day was and can help to maximize the utility of clinical trial data, including in a limited at +10%. Results from tumor response models were consistent regulatory setting. ER models from METEOR formed part of the clinical with PFS estimates. The predicted ORR for patients with typical clear- development of cabozantinib and were considered by regulatory bodies ance was 8.7% at a 20-mg/day dose, 15.6% at 40 mg/day, and 19.1% at when formulating label recommendations [26,35]. Taken together, ER 60 mg/day; and in a tumor growth model, the predicted median change models support a dose modification strategy to improve tolerability in tumor size from baseline was −4.5%, −9.1%, and −11.9%, re- with cabozantinib while maintaining efficacy. Significant AEs likely spectively [20]. reflect lower clearance and overexposure to cabozantinib; dose reduc- Exposure-response models for safety focused on select AEs asso- tion is predicted to significantly improve AE tolerability in these pa- ciated with dose modification, including PPE (grade ≥ 1), fatigue/ tients, while having a minimal effect on efficacy. Box 1 summarizes key asthenia (grade ≥ 3), diarrhea (grade ≥ 3), and hypertension (systolic points of the benefit-to-risk profile of cabozantinib described in the ER blood pressure > 160 mmHg or diastolic blood models. pressure > 100 mmHg). For the typical patient, a higher cabozantinib dose was predicted to increase the risk of AEs [19–21]. Based on a 60- Optimizing the cabozantinib dose for an individual Patient: mg/day versus a 40-mg/day dose, the HR was 1.49 (95% CI 1.27–1.75) Practical strategies for dosing for PPE, 1.42 (95% CI 1.11–1.82) for fatigue/asthenia, 1.36 (95% CI 1.15–1.60) for hypertension, and 1.33 (95% CI 1.04–1.70) for diarrhea; Clinicians are faced with a number of challenges for treating in- corresponding HRs for a 60-mg/day versus a 20-mg/day dose were 2.21 dividual patients with advanced RCC. Patients are often older, present (95% CI 1.60–3.06), 2.01 (95% CI 1.22–3.31), 1.85 (95% CI with comorbidities, and are receiving multiple concomitant medica- 1.33–2.57), and 1.78 (95% CI 1.08–2.91). There was no apparent ER tions [36–38]. The cabozantinib label provides guidance on dosing and relationship for nausea/vomiting (grade ≥ 3) or stomatitis (grade ≥ 3). dose modification based on adverse events, food intake, concomitant When ER modeling is extended to include both lower and higher medications, and hepatic impairment (Box 2). Prior to starting patients cabozantinib clearance values, the impact of dose modification on on cabozantinib, it is important for clinicians to conduct a thorough outcomes becomes more well defined. Reducing the dose is predicted to medical examination, including an assessment of medical history, 4 D. Castellano, et al. Cancer Treatment Reviews 89 (2020) 102062 Fig. 3. Estimated hazard of PFS and selected adverse events and corresponding Kaplan–Meier estimates for doses of cabozantinib (40 mg/day vs 60 mg/day) as assessed by predicted clearance values. Survival curves reproduced with permission from Jonasch et al. [21]. Based on: 172 events in 315 patients for PFS; 137 events for PPE (grade ≥ 1), 42 events for fatigue/asthenia (grade ≥ 3), 103 events for hypertension (systolic blood pressure > 160 mmHg or diastolic blood pressure > 100 mmHg), and 38 events for diarrhea (grade ≥ 3) in 318 patients. Clearance values: Low, 1.3 L/h; typical, 2.23 L/h; and high, 3.3 L/h. AE, adverse event; CI, confidence interval; CL/F, plasma clearance; HR, hazard ratio; PFS, progression-free survival; PPE, palmar-plantar erythrodysesthesia. current medications, and any prior experience with TKI treatment. CYP3A4 interactions). Clinicians should identify concomitant medications that induce or in- Because of the interpatient variability in clearance (CV of 46%) and hibit the CYP3A4 pathway [6,7]. Strong CYP3A4 modifiers should be exposure, clinicians should anticipate that many patients receiving ca- avoided, and patients should be switched to alternate medications that bozantinib will require dose modifications to manage exposure and have no or minimal CYP3A4 modification whenever possible. If con- tolerability. The most common AEs associated with cabozantinib gen- comitant use with a strong CYP3A4 modifier cannot be avoided, then erally emerge within 3–5 weeks of treatment initiation and are likely a the cabozantinib dose should be adjusted as indicated per the US label marker of low cabozantinib clearance and high exposure [26,39]. As (note the European Summary of Product Characteristics does not in- shown in the Kaplan–Meier curves (Fig. 3), PPE emerges rapidly in clude recommendations for dose adjustments based on potential patients with low cabozantinib clearance. During METEOR, median 5 D. Castellano, et al. Cancer Treatment Reviews 89 (2020) 102062 Box 1 Cabozantinib Exposure-Response (ER) Summary and Recommendations. • ability. ER modeling provides further support for treating patients with the highest tolerable dose of cabozantinib, balancing efficacy and toler- • There is considerable interpatient variability in cabozantinib clearance and exposure. Because cabozantinib is administered at a fixed dose, clinicians should consider variability in cabozantinib clearance to optimize the dose for each patient during therapy. • ER models predict improved efficacy with higher doses of cabozantinib but also increased risk of adverse events (AEs). • treatment. Patients with high cabozantinib clearance are less likely to experience an AE that requires dose modification during the early months of These patients should be maintained at the 60-mg/day dose as tolerated. • Patients with typical cabozantinib clearance are likely to experience an AE that requires dose modification during the early months of treatment. The dose should be reduced to 40 mg/day when needed to improve tolerability. ER modeling predicts that dose reduction to 40 mg/day will not have a clinically significant impact on efficacy. Further dose reduction to 20 mg/day may be necessary to improve long- term tolerability for a minority of treated patients. • Patients with low cabozantinib clearance are at high risk of experiencing an AE that requires dose modification during the early months of treatment. These patients may require a second-level dose reduction to 20 mg/day. At 20 mg/day, cabozantinib exposure for a patient with low clearance is comparable to exposure for a patient with typical clearance receiving 40 mg/day or for a patient with high clearance receiving 60 mg/day. • Dose reduction to 20 mg/day may negatively impact efficacy for patients with high or typical cabozantinib clearance. The 20-mg/day dose should be used only in patients for whom 40 mg/day is intolerable; this intolerability is likely associated with low clearance and dose reduction should have a minimal effect on efficacy. time to first occurrence was 3.0 weeks (interquartile range [IQR], A number of prophylactic and supportive care measures can be 2.0–6.1) for hypertension, 3.4 weeks (IQR, 2.3–6.1) for PPE, and implemented to mitigate the risk and severity of some of the more 4.9 weeks (IQR, 2.7–8.1) for diarrhea [26]. Corresponding to the early common side effects, including PPE, fatigue, gastrointestinal events, onset of AEs, the median time to first dose reduction was 7.9 weeks and hypertension [39,40]. Clinicians should identify any potential (range, 1.4–50.7), and median time to second dose reduction was drug–drug interactions with concomitant medications prior to treat- 13.3 weeks (range, 4.1–45.3) [26]. Given that the time to steady-state ment initiation [39,41,42]. Cabozantinib does not have a clinically concentration for cabozantinib is ~ 15 days with a relatively long half- significant effect on QTc interval; and therefore, coadministration with life of ~ 99 h, dose interruption until resolution or improvement is QTc-prolonging agents such as antiemetic therapy with serotonin in- appropriate prior to dose reduction. Clinicians should be mindful that hibitors is not contraindicated. Clinicians should consider monitoring some AEs can be more readily managed with supportive care than patients with a history of QT interval prolongation or at-risk patients others. The ratio of patients requiring a dose reduction for an AE to the (eg, patients with relevant cardiac disease) during cabozantinib treat- total number of patients experiencing the AE (any grade) was 38/139 ment [6,7,29,41,43]. (27%) for PPE, 54/245 (22%) for diarrhea, 25/122 (20%) for hy- Currently, there is no strong evidence to support dose adjustments pertension, and 33/186 (18%) for fatigue [26]. based on age, sex, or race [18,33]. However, in a subgroup analysis of It is important for clinicians to discuss with their patients the ob- METEOR outcomes by age, AEs emerged more rapidly in a small sub- jectives of treatment and the balance between efficacy (prolonged group of elderly patients (n = 27) receiving cabozantinib compared survival, tumor control, and potential symptom relief) and treatment with younger patients; and elderly patients required more frequent dose side effects [2–5]. Patients need to be aware that AEs can impact quality reductions. The median time to a grade ≥ 3 AE was 3.4 weeks for of life and to understand how to recognize the signs of the more patients ≥ 75 years of age versus 16.1 weeks for patients 65–74 years common events. Patients should understand that adverse effects need to of age, and the corresponding rates of dose reduction were 85% versus be addressed promptly and may require dose interruptions, reductions, 61% [44]. These data suggest reducing the initial dose of cabozantinib or possibly treatment discontinuation. Patients need reassurance that based on age may be warranted, but additional data are needed to dose reductions are unlikely to have a clinically significant impact on support a clinical recommendation. Regardless, hepatic function should efficacy but should improve tolerability so that they can remain on be assessed as elderly patients can have reduced liver mass and hepatic treatment. function [45], and older patients should be closely monitored for Box 2 Cabozantinib Dosing Recommendations [6,7] • The approved dose for cabozantinib is 60 mg/day, which should be maintained during the course of treatment as tolerated. • Cabozantinib should be administered at least 1 h before or at least 2 h after eating. • Grade 1/2 adverse events (AEs) can often be managed with supportive care. Cabozantinib should be held if a patient experiences an intolerable grade 2 AE, a grade 3/4 AE, or osteonecrosis of the jaw (a rare [ < 1%] but serious event). Once the AE has resolved to grade 1 or to baseline level, cabozantinib should be restarted at a dose reduced by 20 mg/day. If the previous dose was 20 mg/day, patients should be restarted at 20 mg/day or cabozantinib should be discontinued. • Clinicians should be aware of factors that can modify cabozantinib exposure. Cabozantinib is metabolized by the CYP3A4 pathway; therefore, patients receiving concomitant strong CYP3A4 inhibitors should have their dose lowered by 20 mg/day, while those receiving concomitant strong CYP3A4 inducers should have the dose increased by 20 mg/day. If the concomitant medication is discontinued, the cabozantinib dose should be adjusted appropriately after a 2–3-day washout period. • Per the US FDA prescribing information, patients with moderate hepatic impairment (Child-Pugh B) should have their dose reduced to 40 mg/day given the potential for increased exposure, and these patients should be closely monitored. (Note that the European Summary of Product Characteristics does not recommend dose adjustments for moderate hepatic impairment due to limited data). Cabozantinib should be avoided in patients with severe hepatic impairment (Child-Pugh C). • Dose adjustments are not recommended for patients with mild hepatic impairment or mild to moderate renal impairment. Cabozantinib use has not been established in patients with severe renal impairment. 6 D. Castellano, et al. Cancer Treatment Reviews 89 (2020) 102062 emerging AEs. A number of descriptive studies have reported outcomes Role of the funding source: in patients treated with cabozantinib using modified dosing strategies [46–48]. However, more retrospective and prospective data are needed Medical writing and editorial assistance was funded by Exelixis, Inc. to support definitive randomized controlled trials to evaluate novel (Alameda, CA) dosing strategies with cabozantinib, such as a dose-escalation approach. As the clinical data set for cabozantinib in RCC grows, future studies Acknowledgments should continue to assess the impact of comorbidities and concomitant medications (eg, CYP3A4 modifiers) on dosing and clinical outcomes. Editorial and medical writing assistance were provided by Karen O’Leary, PhD, and Michael Raffin (Fishawack Communications Inc., Conshohocken, PA). Conclusions Appendix A. Supplementary data Clinical data and ER modeling support cabozantinib at a dose of 60 mg/day for the treatment of patients with advanced RCC, with a Supplementary data to this article can be found online at https:// dose modification strategy to manage AEs. Because of interpatient doi.org/10.1016/j.ctrv.2020.102062. variability in cabozantinib clearance, clinicians should anticipate that many patients may experience AEs during the early months of treat- References ment. Grade 1 or 2 AEs can often be managed with supportive care alone, but intolerable grade 2 or grade 3/4 AEs require prompt dose [1] Yakes FM, Chen J, Tan J, Yamaguchi K, Shi Y, Yu P, et al. Cabozantinib (XL184), a interruption and subsequent reduction. Judicious use of dose mod- novel MET and VEGFR2 inhibitor, simultaneously suppresses metastasis, angio- genesis, and tumor growth. Mol Cancer Ther. 2011;10:2298–308. ification in patients experiencing adverse events is expected to improve [2] Choueiri TK, Escudier B, Powles T, Mainwaring PN, Rini BI, Donskov F, et al. overall tolerability with a minimal impact on efficacy outcomes. The Cabozantinib versus everolimus in advanced renal-cell carcinoma. N Engl J Med. dose should be adjusted as indicated for patients receiving concomitant 2015;373:1814–23. [3] Choueiri TK, Escudier B, Powles T, Tannir NM, Mainwaring PN, Rini BI, et al. medication with strong CYP3A4 inducers or inhibitors, or for patients Cabozantinib versus everolimus in advanced renal cell carcinoma (METEOR): final with moderate hepatic impairment (per the US Food and Drug results from a randomised, open-label, phase 3 trial. Lancet Oncol. 2016;17:917–27. Administration prescribing information [6]). Clinicians may also con- [4] Choueiri TK, Halabi S, Sanford BL, Hahn O, Michaelson MD, Walsh MK, et al. Cabozantinib versus sunitinib as initial targeted therapy for patients with metastatic sider patient age prior to starting cabozantinib. Adverse events may renal cell carcinoma of poor or intermediate risk: The Alliance A031203 CABOSUN occur earlier in older patients, and these patients should be monitored Trial. J Clin Oncol. 2017;35:591–7. closely and dose reductions implemented expeditiously. [5] Choueiri TK, Hessel C, Halabi S, Sanford B, Michaelson MD, Hahn O, et al. Cabozantinib versus sunitinib as initial therapy for metastatic renal cell carcinoma Several ongoing clinical trials will evaluate cabozantinib in combi- of intermediate or poor risk (Alliance A031203 CABOSUN randomised trial): nation with checkpoint inhibitors including nivolumab ± ipilimumab Progression-free survival by independent review and overall survival update. Eur J (NCT03635892, NCT03937219), pembrolizumab (NCT03149822), and Cancer. 2018;94:115–25. atezolizumab (NCT03170960), in patients with advanced RCC. Dose [6] Exelixis, Inc. Cabometyx Prescribing Information. 2019. Accessed February 22, 2019. https://cabometyx.com/downloads/CABOMETYXUSPI.pdf. escalation studies will define the optimal dosing regimen for each [7] Ipsen Pharma. Cabometyx Summary of Product Characteristics. 2019. Accessed combination, and future analyses will seek to maximize the benefit/risk February 22, 2019. https://www.ema.europa.eu/documents/product-information/ profile of cabozantinib within the evolving treatment paradigm for cabometyx-epar-product-information_en.pdf. [8] Lacy SA, Miles DR, Nguyen LT. Clinical pharmacokinetics and pharmacodynamics advanced RCC. of cabozantinib. Clin Pharmacokinet. 2017;56:477–91. [9] Miles D, Jumbe NL, Lacy S, Nguyen L. Population pharmacokinetic model of ca- bozantinib in patients with medullary thyroid carcinoma and its application to an exposure-response analysis. Clin Pharmacokinet. 2016;55:93–105. Declaration of interest [10] Terada T, Noda S, Inui K. Management of dose variability and side effects for in- dividualized cancer pharmacotherapy with tyrosine kinase inhibitors. Pharmacol DC has received fees for consulting and/or advisory roles with Ther. 2015;152:125–34. [11] Klumpen HJ, Samer CF, Mathijssen RH, Schellens JH, Gurney H. Moving towards Astellas Pharma, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol- dose individualization of tyrosine kinase inhibitors. Cancer Treat Rev. Myers Squibb, Ipsen, Janssen Oncology, Lilly, MSD Oncology, Novartis, 2011;37:251–60. Pfizer, Pierre Fabre, Roche/Genentech, and Sanofi; institutional re- [12] Josephs DH, Fisher DS, Spicer J, Flanagan RJ. Clinical pharmacokinetics of tyrosine kinase inhibitors: implications for therapeutic drug monitoring. Ther Drug Monit. search funding from Janssen Oncology; and travel/accommodations 2013;35:562–87. and/or expenses from Bristol-Myers Squibb, Pfizer, and Roche. JPM has [13] Widmer N, Bardin C, Chatelut E, Paci A, Beijnen J, Leveque D, et al. Review of received fees for consulting and/or advisory roles with Astellas Korea, therapeutic drug monitoring of anticancer drugs part two–targeted therapies. Eur J Bayer, Bristol Myers-Squibb, Janssen China R&D, Novartis, Pfizer, and Cancer. 2014;50:2020–36. [14] Buil-Bruna N, Lopez-Picazo JM, Martin-Algarra S, Troconiz IF. Bringing model- Roche; and research funding from Roche. FB is an Ipsen Pharma em- based prediction to oncology clinical practice: a review of pharmacometrics prin- ployee (Medical R&D Department). NT is an Ipsen employee. LN is an ciples and applications. Oncologist. 2016;21:220–32. employee of Exelixis, Inc., and also owns Exelixis stocks. DOC is an [15] US Food and Drug Administration. Guidance for Industry Exposure-Response Relationships — Study Design, Data Analysis, and Regulatory Applications. employee of Exelixis, Inc., owns Exelixis stocks, and receives shared 2003;2018. Accessed April 10, 2018. https://www.fda.gov/AboutFDA/ royalties from a UCSF patent. EJ has received fees for consulting and/or CentersOffices/OfficeofMedicalProductsandTobacco/CDER/ucm225044.htm. advisory roles with Eisai, Exelixis, Genentech/Roche, Novartis, and [16] US Food and Drug Administration. Guidance for Industry Postmarketing Studies and Clinical Trials — Implementation of Section 505(o)(3) of the Federal Food, Pfizer; research funding from Exelixis, Novartis, Peloton Therapeutics, Drug, and Cosmetic Act. 2011;April 10. Accessed 2018, April 10. https://www.fda. Pfizer; and travel/accommodations and/or expenses from Pfizer. gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ UCM172001.pdf. [17] US Food and Drug Administration. Guidance for Industry Population Pharmacokinetics. 1999;2018. Accessed April 10, 2018. https://www.fda.gov/ Author Contributions downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ UCM072137.pdf. [18] Lacy S, Hutmacher M, Yang B, Motzer RJ, Escudier B, Powles T, et al. Population All authors contributed to the conception of the manuscript and pharmacokinetic and exposure-response modeling of cabozantinib in patients with intrepretation of data. DC, DOC, and EJ contributed to the drafting of renal cell carcinoma in the Phase 3 METEOR study. Abstract (2565) and poster the manuscript. All authors provided critical review and revisions, and (P265) presented at the American Society of Clinical Oncology (ASCO) Annual all authors approved the final version of the manuscript for submission Meeting 2016, June 3–7, Chicago, IL, USA. 2016. [19] Lacy S, Hutmacher M, Yang B, Motzer RJ, Escudier B, Powles T, et al. Cabozantinib and publication. 7 D. Castellano, et al. Cancer Treatment Reviews 89 (2020) 102062 exposure-response modeling of safety endpoints in patients with renal cell carci- Exposure-Response Analysis of Clinical Endpoints in Drug Development. CPT noma in the Phase 3 METEOR study. Abstract (D10) and poster (P447) presented at Pharmacometrics Syst Pharmacol. 2015;4:565–75. the Genitourinary Cancers Symposium 2017, February 16–18, Orlando, FL, USA. [35] Center for Drug Evaluation and Research (CDER). Clinical Pharmacology and 2017. Biopharmaceutics Review - Cabozantinib. Accessed December 21, 2018. https:// [20] Lacy S, Nielsen J, Yang B, Miles D, Nguyen L, Hutmacher M. Population exposure- www.accessdata.fda.gov/drugsatfda_docs/nda/2016/ response analysis of cabozantinib efficacy and safety endpoints in patients with 208692Orig1s000ClinPharmR.pdf. renal cell carcinoma. Cancer Chemother Pharmacol. 2018;81:1061–70. [36] Miller LA, Stemkowski S, Saverno K, Lane DC, Tao Z, Hackshaw MD, et al. Patterns [21] Jonasch E, Motzer RJ, Escudier BJ, Powles T, Markby DW, Nielsen J, et al. of care in patients with metastatic renal cell carcinoma among a U.S. payer popu- Cabozantinib exposure-response modeling of efficacy and safety endpoints as a lation with commercial or medicare advantage membership. J Manag Care Spec function of clearance in patients with renal cell carcinoma. Presented at the ASCO Pharm. 2016;22:219–26. 2018 Genitourinary Cancers Symposium, February 8–10, San Francisco, CA. 2018. [37] MacLean E, Cisar L, Mehle K, Eremina D, Quigley JM. Real-world axitinib use in the [22] Guida FM, Santoni M, Conti A, Burattini L, Savini A, Zeppola T, et al. Alternative united states: a retrospective study using linked datasets. J Manag Care Spec Pharm. dosing schedules for sunitinib as a treatment of patients with metastatic renal cell 2016;22:723–32. carcinoma. Crit Rev Oncol Hematol. 2014;92:208–17. [38] Procopio G, Bellmunt J, Dutcher J, Bracarda S, Knox J, Brueckner A, et al. Sorafenib [23] Schmidinger M, Danesi R, Jones R, McDermott R, Pyle L, Rini B, et al. tolerability in elderly patients with advanced renal cell carcinoma: results from a Individualized dosing with axitinib: rationale and practical guidance. Future Oncol. large pooled analysis. Br J Cancer. 2013;108:311–8. 2018;14:861–75. [39] Gerendash BS, Creel PA. Practical management of adverse events associated with [24] Choueiri TK, Pal SK, McDermott DF, Morrissey S, Ferguson KC, Holland J, et al. A cabozantinib treatment in patients with renal-cell carcinoma. Onco Targets Ther. phase I study of cabozantinib (XL184) in patients with renal cell cancer. Ann Oncol. 2017;10:5053–64. 2014;25:1603–8. [40] Schmidinger M, Danesi R. Management of adverse events associated with cabo- [25] Smith DC, Smith MR, Sweeney C, Elfiky AA, Logothetis C, Corn PG, et al. zantinib therapy in renal cell carcinoma. Oncologist. 2018;23:306–15. Cabozantinib in patients with advanced prostate cancer: results of a phase II ran- [41] Hussaarts K, Veerman GDM, Jansman FGA, van Gelder T, Mathijssen RHJ, van domized discontinuation trial. J Clin Oncol. 2013;31:412–9. Leeuwen RWF. Clinically relevant drug interactions with multikinase inhibitors: a [26] European Medicines Agency. CHMP assessment report, CABOMETYX. 2016;2017. review. Ther Adv Med Oncol. 2019;11. 1758835918818347. Accessed January 20, 2017. http://www.ema.europa.eu/docs/en_GB/document_ [42] ESMO. Drug-drug interactions with kinase inhibitors. Accessed April 12, 2019. library/EPAR_-_Public_assessment_report/human/004163/WC500214070.pdf. https://oncologypro.esmo.org/Oncology-in-Practice/Anti-Cancer-Agents-and- [27] Nguyen L, Benrimoh N, Xie Y, Offman E, Lacy S. Pharmacokinetics of cabozantinib Biological-Therapy/Drug-Drug-Interactions-with-Kinase-Inhibitors. tablet and capsule formulations in healthy adults. Anticancer Drugs. [43] Miles DR, Lacy SA, Wada DR, Milwee S, Yaron Y, Nguyen LT. Assessment of ca- 2016;27:669–78. bozantinib treatment on QT interval in a phase 3 study in medullary thyroid cancer: [28] Kurzrock R, Sherman SI, Ball DW, Forastiere AA, Cohen RB, Mehra R, et al. Activity evaluation of indirect QT effects mediated through treatment-induced changes in of XL184 (Cabozantinib), an oral tyrosine kinase inhibitor, in patients with me- serum electrolytes. Cancer Chemother Pharmacol. 2017;80:295–306. dullary thyroid cancer. J Clin Oncol. 2011;29:2660–6. [44] Donskov F, Motzer RJ, Voog E, Hovey EJ, Grüllich C, Nott LM, et al. Outcomes [29] Nguyen L, Holland J, Mamelok R, Laberge MK, Grenier J, Swearingen D, et al. based on age in the phase 3 METEOR trial of cabozantinib (cabo) versus everolimus Evaluation of the effect of food and gastric pH on the single-dose pharmacokinetics (eve) in patients with advanced renal cell carcinoma (RCC). Abstract (517) and of cabozantinib in healthy adult subjects. J Clin Pharmacol. 2015;55:1293–302. poster (G8) presented at: Annual Genitourinary Cancers Symposium of the [30] Nguyen L, Holland J, Miles D, Engel C, Benrimoh N, O'Reilly T, et al. American Society of Clinical Oncology; February 16-18, 2017; Orlando, FL. Pharmacokinetic (PK) drug interaction studies of cabozantinib: Effect of CYP3A [45] Pajares B, Torres E, Trigo JM, Saez MI, Ribelles N, Jimenez B, et al. Tyrosine kinase inducer rifampin and inhibitor ketoconazole on cabozantinib plasma PK and effect inhibitors and drug interactions: a review with practical recommendations. Clin of cabozantinib on CYP2C8 probe substrate rosiglitazone plasma PK. J Clin Transl Oncol. 2012;14:94–101. Pharmacol. 2015;55:1012–23. [46] Procopio G, Prisciandaro M, Iacovelli R, Cortesi E, Fornarini G, Facchini G, et al. [31] Nguyen L, Holland J, Ramies D, Mamelok R, Benrimoh N, Ciric S, et al. Effect of Safety and efficacy of cabozantinib in metastatic renal-cell carcinoma: real-world renal and hepatic impairment on the pharmacokinetics of cabozantinib. J Clin data from an Italian managed access program. Clin Genitourin Cancer. Pharmacol. 2016;56:1130–40. 2018;16:e945–51. [32] Nguyen L, Chapel S, Tran BD, Lacy S. Updated population pharmacokinetic model [47] Lemke E, Shah AY, Chandramohan A, Campbell MT, VanAlstine M, Jonasch E, of cabozantinib integrating various cancer types including hepatocellular carci- Tannir NM. Real-world experience with cabozantinib in metastatic clear cell renal noma. J Clin Pharmacol. 2019. https://doi.org/10.1002/jcph.467. cell carcinoma (mccRCC). 684–684 JCO 2018;36(6_suppl). https://doi.org/10. [33] Lacy S, Yang B, Nielsen J, Miles D, Nguyen L, Hutmacher M. A population phar- 1200/JCO.2018.36.6_suppl.684. macokinetic model of cabozantinib in healthy volunteers and patients with various [48] Boegemann M, Krabbe L-M, Schrader AJ, Semjonow A, Klaile Y, Schlack K. Dose cancer types. Cancer Chemother Pharmacol. 2018;81:1071–82. escalation of cabozantinib as a viable option for the treatment of mRCC. J Clin [34] Overgaard RV, Ingwersen SH, Tornoe CW. Establishing Good Practices for Oncol. 2018;36:;Abstract:636. 8