A. Yin Edwards 1, Yang Zhang 2, Katrina Hui 1, Loan Tran 2, YuWei Lin 1, Kara Schmelzer 1, Peter Manley 2, Katherine Pehlivan 2, Elly Barry 2, Amy Cheung 1, Jing Wang 2
1 Certara (Radnor, United States), 2 Day One Biopharmaceuticals, Inc (Brisbane, United States)
Introduction: Genetic alterations and dysregulation of the mitogen-activated protein kinase (MAPK) pathway have been found in many different types of adult and pediatric cancers [1]. One of the most frequently altered genes in this pathway is v-Raf murine sarcoma viral oncogene homolog B (BRAF) [2]. Tovorafenib is an oral, selective, central nervous system-penetrant, type II RAF kinase inhibitor that received US FDA accelerated approval for patients ≥6 months of age with relapsed/refractory (r/r) BRAF-altered pLGG at the recommended dosage of 380 mg/m² (not to exceed 600 mg) once weekly (QW), based on results from the phase 2 FIREFLY-1 trial (NCT04775485). Evaluation of QT prolongation risk is required per regulatory guidance to support new drug applications (International Council on Harmonisation [ICH] E14 [3]).
Objectives: Here we describe tovorafenib concentration-QTc analysis in adult patients from Study C28001 and in pediatric and young adult patients from FIREFLY-1. This analysis supported the clinical development of tovorafenib as part of the new drug application (NDA) submission by characterizing the effect of tovorafenib exposure on ECG parameters, as a substitute for a thorough QT (TQT) study.
Methods: Triplicate ECG measurements and time-matched observed plasma tovorafenib concentration data from Study C28001 and FIREFLY-1 were collected. Adult and pediatric data were not pooled for the concentration–QTc analysis because age-related differences in cardiac electrophysiology, exposure distributions, disease characteristics, and ECG measurement variability violated the assumption of a common exposure–QTc relationship across populations. Separate analyses were therefore conducted to avoid biasing slope estimates and to preserve sensitivity for detecting age-specific QTc effects. Primary cardiac endpoints included QT intervals assessment (with Fredericia [QTcF] and population corrections [QTcP]). Pre-specified linear mixed-effects (LME) models [4] relating baseline-adjusted QTcF (ΔQTcF) or QTcP (ΔQTcP) to tovorafenib concentrations were tested, consisting of terms for drug-dependent effects, drug-independent effects, and residual error. The final LME models were used to predict drug-induced changes in the respective ECG parameters from baseline with associated 2-sided 90% confidence interval (CI) at the recommended phase 2 dose (RP2D): C28001, 600 mg QW; FIREFLY-1, 420 mg/m² (not to exceed 600 mg] QW. Categorical analyses were also performed to further evaluate cardiac safety based on regulatory guidance thresholds [3].
Results: There were 480 time-matched ECG-concentration pairs from 61 adult patients at doses of 20 to 280 mg every other day (Q2D) or 400 to 800 mg QW from Study C28001. A total of 597 time-matched ECG-concentrations pairs from 114 pediatric and young adult patients dosed at 420 mg/m² (not to exceed 600 mg) QW were available from Study FIREFLY-1 with a data cut-off date of 22-Dec-2022. Tovorafenib did not cause a mean QTc prolongation of ≥10 msec at doses used for the simulations (C28001: 600 mg QW; FIREFLY-1: 420 mg/m² [not to exceed 600 mg] QW). Model-based predictions at the geometric mean of Cmax for the 600 mg QW regimen based on the adult models were 1.52 msec (90% CI: -4.19, 7.23) for ΔQTcF and 3.69 msec (90% CI: -2.41, 9.79) for ΔQTcP. Model-based predictions at the geometric mean of Cmax for the 420 mg/m² (not to exceed 600 mg) QW regimen based on the pediatric models were -1.70 msec (90% CI: -3.76, 0.360) for ΔQTcF and -0.323 msec (90% CI: -2.55, 1.91) for ΔQTcP. In Study C28001, none and 1 patient had QTcF>500 msec and QTcP>500 msec (evaluable N=64), respectively; one and 2 patients had ΔQTcF>60 msec and ΔQTcP>60 msec over baseline (evaluable N=61), respectively. There were no significant outliers for QTc based on categorical analysis (i.e., QTcF or QTcP >500 msec [evaluable N=137] or ΔQTcF or ΔQTcP >60 msec over baseline [evaluable N=131]) in FIREFLY-1 patients.
Conclusions: This analysis indicates that at the simulated exposures of tovorafenib in both adults (600 mg QW) and pediatric oncology patients (420 mg/m² [not to exceed 600 mg] QW), there was no clinically significant mean increase in QTc intervals.
References:
1. Yaeger R, Corcoran RB. Targeting alterations in the RAF-MEK pathway. Cancer Discov. 2019;(3):329-341.
2. Sholl M A narrative review of BRAF alterations in human tumors: diagnostic and predictive implications. Precis Cancer Med. 2020;3:26.
3. ICH Steering Committee. ICH Harmonized Tripartite Guideline. The clinical evaluation of QT/QTc interval prolongation and proarrhythmic potential for non-antiarrhythmic drugs. 2005.
4. Garnett C, Bonate PL, Dang Q, et al. Scientific white paper on concentration-QTc modeling. J Pharmacokinet Pharmacodyn. 2018; 45(3): 383-397.
Reference: PAGE 34 (2026) Abstr 11892 [www.page-meeting.org/?abstract=11892]
Poster: Drug/Disease Modelling - Safety