A008: Using Clinical Data & Physiologically-based Pharmacokinetic Modelling to Assess the Effect of Hepatic Impairment on Avapritinib Pharmacokinetics to Guide Dose Adjustment

Statement of Purpose, Innovation or Hypothesis: Avapritinib is a tyrosine kinase inhibitor approved in the USA and Europe for adult patients with unresectable/metastatic gastrointestinal stromal tumors (GIST) harboring a PDGFRA exon 18 (USA) or D842V mutation (Europe), advanced systemic mastocytosis (AdvSM; Europe after ≥1 therapy), or indolent systemic mastocytosis (ISM). Avapritinib is extensively metabolized by the liver and eliminated through biliary excretion. A post hoc population pharmacokinetic (PK) analysis in patients with GIST indicated comparable avapritinib clearance among those with normal hepatic function and mild/moderate hepatic impairment (HI), suggesting no dose adjustments were required. Here, we report on avapritinib PK in participants with severe HI in the phase 1 open-label single-dose study, BLU-285-0107, and on a previously developed avapritinib physiologically-based pharmacokinetic (PBPK) model, which extended findings from BLU-285-0107 and predicted effects of mild, moderate, and severe HI on steady-state avapritinib PK to inform dose adjustments in intended patient populations.

Description of Methods & Materials: BLU-285-0107 assessed the effect of severe HI (Child-Pugh Class C) on avapritinib PK (total and unbound plasma) following a single 100-mg oral dose. Eight participants with severe HI and eight with normal hepatic function (matched for age, body-mass index, and sex to participants with severe HI) were enrolled. Serial blood samples for PK analysis were collected from predose up to 576 hours postdose. A PBPK model was previously developed and verified using clinical data and drug–drug interaction studies. Further modifications to the rate of absorption and fraction absorbed to recover the reduced maximum plasma concentration (C_max; severe HI) and shorter time to C_max (T_max control and severe HI), and area under the curve (AUC) observed in BLU-285-0107 were required. This model was leveraged to predict changes in avapritinib PK after repeat once-daily (QD) dosing in participants with severe HI versus healthy participants.

Data & Results: Severe HI was associated with a 20% (1.2-fold) increase in AUC from time 0 extrapolated to infinity (AUC_0-inf) and a 20% (0.8-fold) decrease in C_max for total and a 61% (1.61-fold) increase in AUC_0-inf and an 8% (1.08-fold) increase in C_max for unbound plasma avapritinib. The fraction unbound in plasma was higher (37%) in severe HI. The PBPK model predicted avapritinib PK exposures (total and unbound) following a single 100-mg oral dose in participants with severe HI that were within 2-fold (most within 1.5-fold) of observed data (BLU-285-0107). In repeat avapritinib QD dosing analyses, minimal impact on systemic exposure in participants with mild/moderate HI was predicted. In participants with severe HI, steady-state systemic exposures of unbound avapritinib were increased by 1.6- and 1.7-fold for C_max and AUC for dosing interval, respectively, and considered potentially clinically meaningful. Further simulations indicated dose reductions from 200 mg QD to 100 mg QD (AdvSM), 300 mg QD to 200 mg QD (GIST), and 25 mg QD to 25 mg every other day (ISM) would be appropriate for severe HI.

Interpretation, Conclusion or Significance: Moderate increases in unbound, but not total, avapritinib were observed in participants with severe HI. Based on clinical data and PBPK predictions, lower starting doses of avapritinib are recommended in patients with severe HI, per the US prescribing information.

Disclosures: Anitha Suram, Preethi Swamy, James Baker, Ronda Rippley, and Wendy Ankrom are employees and equity holders of Blueprint Medicines Corporation. Maria Roche is an employee and equity holder of Blueprint Medicines Corporation and also reports other from Epizyme, outside the submitted work. Thomas C. Marbury is an employee and equity owner of Orlando Clinical Research Center. Zoe Barter reports no conflicts of interest.

Citations/References:

Additional Information/Authors: N/A