Cabozantinib (C) exposure-response (ER) modeling of safety endpoints in patients (pts) with renal cell carcinoma (RCC) in the phase III METEOR study.
447 Background: ER models were previously developed to characterize the relationship between C exposure and efficacy endpoints in RCC pts in the phase III METEOR study (J Clin Oncol 34, 2016 [suppl; abstr 2565]). Higher C exposure correlated with decreased tumor size and improved progression-free survival and objective response rate. Model-based predictions showed that C would be effective at the 60 mg starting dose evaluated in METEOR as well as dose levels of 40 and 20 mg resulting from dose reduction. In the current study, ER models were developed to characterize the relationship between C exposure and safety endpoints in RCC pts. Methods: The ER analysis included 318 RCC pts who had received at least one C dose and had at least one measurable C concentration. Time-to-event Cox proportional hazard ER models were developed to characterize the relationship between various individual predicted C exposure measures and the likelihood of dose modification and 6 specific adverse events (AEs): fatigue/asthenia, palmar-plantar erythrodysesthesia (PPE), nausea/vomiting, diarrhea, hypertension, and stomatitis. Results: A statistically significant relationship was identified between individual predicted C clearance (CL/F) and the rate of dose modification (p <0.0001), with the risk of dose modification decreasing with increasing CL/F. An increase in average C concentration was associated with increased risk of fatigue/asthenia (Grade ≥3), PPE (Grade ≥1), hypertension (systolic blood pressure [BP] >160 mmHg or diastolic BP >100 mmHg), and diarrhea (Grade ≥3). The predicted hazard ratios for these AEs were 2.01, 2.21, 1.85, and 1.78, respectively, based on the predicted steady-stage average C concentration for a 60 mg dose relative to a 20 mg dose. Statistically significant ER relationships were not identified for nausea/vomiting (Grade ≥3) or stomatitis (Grade ≥3). Conclusions: Based on the ER analysis, higher C exposures resulting from lower C CL/F are predicted to increase the dose modification rate. Reduced C exposures resulting from dose reduction are predicted to decrease the risk of fatigue/asthenia, PPE, hypertension, and diarrhea while maintaining clinical benefit. Clinical trial information: NCT01865747.