Assoc Dir Merck & Co., Inc., West Point, Pennsylvania, United States
Disclosure(s):
lihong Du, Data Sci: No financial relationships to disclose
Background and Objectives: End stage kidney disease (ESKD) patients, specifically those on hemodialysis (HD) carry an elevated risk for thrombotic events with associated morbidity and mortality. MK-2060 is a dual-target inhibitor (FXI and FXIa) monoclonal antibody (mAb) being developed for the prevention of thrombotic cardiovascular events in ESKD patients. A Phase 2 study to assess efficacy and safety of MK-2060 in ESKD patients on HD via an arteriovenous graft (AVG) has recently been completed. The objective of this work is to characterize exposure-response (E-R) relationships for efficacy (time to first AVG thrombosis event) and safety (time to first major bleeding event) endpoints in the patients.
Methods: A popPK model was developed using data of 432 participants from 5 Phase 1 studies and 1 Phase 2 study who received IV infusion treatment. Post-hoc steady-state MK-2060 exposures (Cavg, Cmax and Ctrough) of the patients with ESKD in the Phase 2 study were derived from the individual empirical Bayes estimates (EBE) using the median actual dose for each participant. A semi-parametric Cox proportional hazards (Cox PH) model was used to estimate the relative risk of an efficacy or safety event associated with predictors (e.g., exposures) and covariates without specifying the baseline hazard function. Separate Cox PH models were developed for efficacy and safety endpoints. Exploratory Kaplan-Meier (KM) analyses for biomarkers (FXI inhibition and activated partial thromboplastin time (aPTT) prolongation) and the efficacy and safety endpoints were also conducted to examine the findings from Cox PH models.
Results: PK concentrations of MK-2060 were adequately described by a two-compartment model with first-order elimination and standard allometric scaling for body weight on all PK parameters. ESKD patients demonstrated relatively higher clearance and central volume than healthy participants, and no other covariates were included in the PK model. The E-R analyses included data from 501 patients with ESKD who received QW IV 1 hr infusion of 20 mg, 6 mg, or matching placebo in 1:1:1 ratio for approximately 1 year of median duration based on an event-driven study design. Statistics of the Cox PH model for the safety endpoint showed steady-state Cavg was a significant predictor for the time to first bleeding and suggested an increased hazard of bleeding with increasing MK-2060 exposure. Statistics of the Cox PH model for the efficacy endpoint showed steady-state Cavg was not a significant predictor for the time to first thrombosis and suggested futility of MK-0260 exposure on reducing the hazard of thrombosis. Similar findings were observed when analyzing using other exposure metrics (Cmax and Ctrough). These modeling results were also consistent with the insignificant relationships from KM analyses for the biomarkers and time to first bleeding or thrombosis event.
Conclusions: The E-R analyses of MK-2060 for primary efficacy and safety endpoints does not support its effectiveness regarding thrombosis reduction. Furthermore, the studied dose range appears to be associated with notable bleeding risks.
