Dose optimization for MORAb-202, an antibody-drug conjugate (ADC) highly selective for folate receptor-alpha (FRα), using population pharmacokinetic (PPK) and exposure-response (E-R) efficacy and safety analyses.

3090 Background: MORAb-202 is an ADC consisting of farletuzumab (an antibody that binds to FRα) paired with eribulin mesylate (a microtubule dynamics inhibitor) conjugated via a cathepsin B-cleavable linker. A phase 1 dose-escalation and expansion study in patients with advanced solid tumors evaluated MORAb-202 doses ranging from 0.3 mg/kg to 1.2 mg/kg IV every 3 weeks (Shimizu 2021, CCR). The dose-expansion part included starting doses of 0.9 mg/kg and 1.2 mg/kg in an ovarian cancer (OC) cohort. Objective response rates (ORR) by investigator per RECIST v1.1 and rates of all-grade interstitial lung disease (ILD), an adverse event of interest, were lower at the 0.9 mg/kg dose vs the 1.2 mg/kg dose. To support dose optimization for clinical benefit while reducing the risk of ILD, a MORAb-202 PPK model was developed to characterize the pharmacokinetics and to obtain model-predicted exposure measures. Methods: Exposure was predicted for different dosing scenarios: flat dosing, bodyweight (BW)-based dosing with or without a dose cap, adjusted ideal BW dosing, and body surface area (BSA)-based dosing. E-R analyses for efficacy (ie, ORR) and safety (ie, ILD by expert review) were conducted using logistic-regression analysis. Simulations (N = 1000) were performed using a BW distribution from a previous phase 3 farletuzumab study in OC (Vergote 2016, JCO) to predict the probability of ORR and ILD in patients treated with MORAb-202. Results: MORAb-202 exposures were dose proportional, and the pharmacokinetics were described by a 2-compartment model with zero-order IV infusion and first-order elimination. Patients with higher BW had less-than-proportional increases in clearance (allometric exponent [AE] 0.571) and distribution volume (AE 0.524). MORAb-202 demonstrated a positive exposure (based on area under the curve [AUC]) dependence to ORR and ILD. The probability of achieving a tumor response was higher with higher AUC (odds ratio [OR] for an AUC unit change of 1000 µg•h/mL: 1.73 [95% CI 1.06–3.11]). The probability of an ILD event was higher with higher AUC (OR for an AUC unit change of 1000 µg•h/mL: 3.50 [95% CI 1.89–7.81]). Simulations across BW ranges (34.2–144 kg) indicated that BSA-based dosing (33 mg/m2), compared with BW-based dosing (0.9 mg/kg), yielded similar predicted median (90% prediction interval) rates for ORR (33.7% [19.3–62.2] vs 37.9% [20.6–67.5]) and all-grade ILD (46.8% [18.2–88.2] vs 55.1% [20.7–91.9]). However, BSA-based dosing is predicted to reduce ILD in the highest BW quartile (> 80–144 kg) by approximately 35% compared with BW-based dosing. Conclusions: Based on this assessment, BSA-based dosing is predicted to lower the exposure-dependent ILD risk in patients with higher BW and is being further evaluated in ongoing clinical studies. Clinical trial information: NCT03386942.