PII-106 - PLASMA 3-HYDROXYPROPIONIC ACID CONCENTRATION IS POSITIVELY ASSOCIATED WITH METABOLIC DECOMPENSATION EVENTS IN PATIENTS WITH PROPIONIC ACIDEMIA.

M. Liang¹, J. Calderón², V. Ivaturi², N. Carillo¹, H. Attarwala¹; ¹Moderna Therapeutics, Inc., Cambridge, MA, USA, ²Pumas-AI Inc., Baltimore, MD, USA.

Background: Propionic acidemia (PA) is a rare, severe inherited metabolic disease with high morbidity and mortality and no effective therapies. The disorder typically occurs in infants and is characterized biochemically by accumulation of propionyl-CoA and other toxic metabolites, such as 3-hydroxypropionic acid (3-HP), and clinically by recurrent metabolic decompensation events (MDEs). The objective of this study was to identify predictive biomarkers of PA disease severity and prognosis.

Methods: Data from a prospective, observational, longitudinal, natural history study of people with methylmalonic acidemia and PA (MaP study; NCT03484767) were analyzed to model the relationship between plasma 3-HP levels and MDEs in PA. A recurrent time-to-event model was used to relate 3-HP levels and MDE rates. The base model assumed that the hazard function for the start of an MDE event is influenced by 3-HP concentration, a term to describe the time course of the hazard between MDE events (or since birth), and between-subject variability. Model performance was evaluated using visual predictive checks.

Results: Data from 47 participants with PA were included in the model. A higher rate of MDEs was associated with higher 3-HP plasma concentrations. Older patients (age ≥12 years) had lower rates of MDEs compared to younger patients (age ≤2 years); age was a covariate in the final model. Visual predictive checks by 3-HP levels and age groups showed agreement between observed data and model simulations. Reducing 3-HP levels by half was associated with a 29% (95% CI, 18%-38%) decrease in the rate of MDEs.

Conclusion: This study reinforces the association between 3-HP levels and MDE risk in patients with PA, suggesting that 3-HP could potentially be used as a surrogate biomarker to predict the risk and rate of MDEs in clinical trial design for novel therapies.