Exposure-safety analysis of Tovorafenib (DAY101), an investigational oral pan-RAF kinase inhibitor, in oncology patients with solid tumors
A. Yin Edwards1, Anna Largajolli1, YuWei-Lin1, Katrina Hui1, John Maringwa1, James Smith1, Elly Barry2, Eleni Venetsanakos2, Samuel C Blackman2, S.Y. Amy Cheung1
(1) Certara, Integrated Drug Development, Princeton, NJ, USA, (2) Day One Biopharmaceuticals, Inc, USA
Objectives: The MAPK pathway, regulates key cell functions, and is altered in many types of cancer. BRAF, a frequently altered gene, in which 90% of the mutations consist of the BRAF-V600 mutation. Tovorafenib (DAY101) is an orally administered, CNS-penetrant, highly selective small molecule. A Type II pan RAF kinase inhibitor, it inhibits BRAF and CRAF/RAF-1 kinases and hyperactivated signaling from B/CRAF fusions. It is being developed for treatment of pediatric low-grade glioma and solid tumors in adults harboring RAF alterations. Here, we evaluated the relationship between systemic exposure to tovorafenib and the probability of adverse events (AE) associated with MAPK inhibition and performed model simulations to confirm the recommended Ph2 dose (RP2D) and support future studies.
Methods: A developed Pop PK model using a Ph1/1b data in adult oncology patients was used to derive the individual exposure metrics as time-averaged exposure, TAE: TAE of the 1st week, TAE of the 1st cycle, steady-state area under the concentration-time curve (AUC), AUC weekly, steady‑state maximum concentration (Cmax), and steady-state minimum concentration (Cmin). AE were classified according to NCI-CTCAE grades (Common Terminology Criteria for AE v4.03). AEs of interest included skin rash (SR), anemia, ↑creatine phosphokinase (CPK), gastrointestinal (GI) disorders, diarrhea, hyperbilirubinemia, ↑ liver function test (LFT), and ocular disorders. Logistic regression analysis was used to assess the relationship between tovorafenib exposure and the probability of having Grade (G) ≥2 or ≥G3 AEs of interest. Potential covariate effects were evaluated for any significant exposure-AE relationships (based on the likelihood ratio test at P value < 0.05). Clinical trial simulations of significant exposure-AE models were performed using a virtual population mimicking the covariate distributions from the PopPK analysis population. A total of 40 simulation scenarios comprised of various dosing regimens (every 2-7 days, 3 days on/4 days off, 5 days on/2 days off) and dose levels (200, 300, 400, 600, 800mg) were evaluated for the predicted rate of AE events.
Results: AE data from a Ph1 study which included 125 patients treated with tovorafenib (20-280mg every other day, Q2D, n=86; 400-800mg once weekly, QW, n=39) was used in the analysis [1]. The median number of cycles received were 2 (range 1->49) cycles and 1 (range 1-10) cycle in the Q2D and QW cohorts, respectively. The most common AEs (based on ≥G2 incidence rate) were GI disorders (34.4%), anemia (29.6%), and SR (25.6%). ↑ CPK and LFT occurred in 7.2% of the patients. Thus, these AE endpoints were chosen for further logistic regression analysis. Of the 7 exposure metrics evaluated, Cmin was identified as the best predictor of AE. Specifically, the probability of developing either ≥G2 or ≥G3 SR, ↑ LFT, or ↑ CPK (≥G2) were significantly correlated with the exposure metric. In addition, higher Cmax was positively correlated with higher incidence of SR (≥G2 or ≥G3). No clear trend between any exposure metrics and the probability of anemia or GI disorders were detected. Subsequent covariate analysis identified dosing regimen as a significant covariate on the intercept of the relationship between Cmax and ≥G2 (Odds Ratio [OR]: 0.116 [95% CI: 0.0317, 0.424] for QW dosing) or ≥G3 SR (OR: 0.133 [95% CI: 0.0259, 0.684] for QW dosing). Also, age was found to be a significant covariate on the intercept of the relationship between Cmin and ≥G3 ↑ LFT (OR: 0.893 [95% CI: 0.822, 0.969] for 1 yr increase in age). Simulations of the final exposure-AE models showed general increase in the probability of AE events with increasing exposures. This was more evident at higher doses with more frequent dosing (ie Q2D and Q3D) and the alternate dosing scenarios in comparison to weekly dosing regimens.
Conclusions: Exposure-AE models were successfully developed to describe the relationships between tovorafenib exposure and ≥G2 or ≥G3 AE. Less frequent dosing, specifically weekly regimen, was associated with improvement in safety profile. Combined with sustained exposure and anticipated tumor target inhibition due to its long half-life, 63 hr, as well as marginal differences in efficacy between QW and Q2D dosing [2] and along with the findings from the PopPK-longitudinal safety analysis [3], 600mg QW has been chosen as the RP2D for further clinical evaluation.
References:
[1] Clinical Study Report C28001: An Open-Label, Phase 1, Dose Escalation Study of MLN2480 in Patients with Relapsed or Refractory Solid Tumors Followed by a Dose Expansion Phase in Patients With Metastatic Melanoma. Millennium Pharmaceuticals, Inc.31 May 2018.
[2] John Maringwa, Anna Largajolli, YuWei-Lin, et.al. Exposure-Response Analysis of Tovorafenib (DAY101) Relating Objective Response to Exposure, Patient- and Disease Characteristics. 2022 PAGE 2022 Abstract.
[3] Hui K, Lin, YW, Largajolli A et. al Utilizing model informed drug development to support study design for Tovorafenib (DAY101), a novel, type II, pan RAF kinase inhibitor: a focus on longitudinal safety markers. 2022 PAGE 2022 Abstract.