PII-121 - IMPLEMENTATION OF QUANTITATIVE SYSTEMS PHARMACOLOGY MODELING IN TRANSLATION AND DOSE SELECTION FOR FIRST-IN-HUMAN PHASE 1/2 CLINICAL STUDY OF NTLA-2002 IN THE TREATMENT OF HEREDITARY ANGIOEDEMA

A. Abdelhady¹, A. Amaral², J. Nosbisch³, R. Wada⁴, J. Phillips², S. Sarkar³, K. Subramanian³, Y. Xu², M. Shah², J. Butler², D. Maag², M. Stroh²; ¹Intellia Therapeutics, Cambridge, MA, United States, ²Intellia Therapeutics, Cambridge, MA, , ³Applied BioMath, Concord, MA, , ⁴QuanTx Consulting.

Background: NTLA-2002 is an investigational in vivo CRISPR-based therapy designed to disrupt the KLKB1 gene, leading to a durable reduction in plasma kallikrein (KK), thereby controlling bradykinin production to prevent hereditary angioedema (HAE) attacks. NTLA-2002 employs a lipid nanoparticle (LNP) delivery system to direct CRISPR/Cas9 components to the liver, where KLKB1 is expressed. This work provides insights on NTLA-2002 pharmacokinetics/pharmacodynamics (PK/PD) and the translational analyses to support the first-in-human Ph1/Ph2 study (NCT05120830).

Methods: A semi-mechanistic quantitative systems pharmacology (QSP) model was modified to capture key determinants of NTLA-2002 PK/PD in cynomolgus (cyno) monkey and human. The model describes the disposition of NTLA-2002 LNP component (LP01) between plasma, liver, and peripheral compartments, KLKB1 editing and KK protein reduction. The model was calibrated to cyno PK/PD data then scaled to human. The relationships between dose, safety and PD were explored by nonlinear mixed effect (NLME) modeling using data obtained from the patients enrolled in NTLA-2002 Ph1 dose escalation study (n=10), who received single 25, 50 or 75mg, to support dose selection for expansion in Ph2.

Results: The cyno QSP model adequately captured the PK and total KK protein reduction in monkeys following administration of 0.25-9.0mg/kg doses. Human model-based simulations were run to project PK/PD following ascending doses of 10-100mg NTLA-2002 to predict doses achieving mean KK protein reduction from baseline by 60%. Model parameters were recalibrated to data from the 25 and 75mg cohorts and validated against data from the 50mg cohort. The final model adequately and simultaneously predicted PK and longitudinal PD following administration of 50mg. NLME analysis suggested that 25mg may reduce plasma KK by ≥60% in the mean patient by week 4 while 50mg may result in ≥60% reduction in >90% of patients. Dose-safety analyses suggested that the typical patient is not predicted to experience clinically significant elevation of transaminases at any of the studied doses.

Conclusion: The presented translational modeling and simulations informed dose escalation in Ph1 and supported the recommendation of 25 and 50mg doses for expansion in Ph2.