0449 Unbiased Examination of Metabolomic Signatures in Sleep Disordered Breathing and Group 1 PAH in the (PVDOMICS) Cohort

Abstract

Abstract Introduction Metabolic alterations have been identified as a novel driver of pulmonary arterial hypertension (PAH) pathogenesis. Sleep disordered breathing (SDB), characterized by intermittent hypoxemic episodes, has also been implicated in PAH, particularly when associated with more profound degrees of hypoxia, and metabolic dysfunction. We hypothesize to observe in SDB and PAH in humans a similar pattern of metabolic alterations as previously described in animal models of PAH. Methods Pulmonary Vascular Disease Phenomics (PVDOMICS,NCT02980887), a multicenter study, was designed in part to elucidate molecular signatures in PH. We analyzed data from Group 1 PAH participants who underwent home sleep apnea testing with circulating plasma metabolomic measures (Metabolon). We conducted an unbiased analysis of SDB measures: apnea-hypopnea index (AHI, 3% oxygen desaturation hypopnea definition) and % recording time< 90% SpO2 (TRT< 90%), with metabolites (954 after qc filtering) using a flexible and robust semi-parametric proportional hazards model adjusted for age, sex, body mass index (BMI), diabetes mellitus (DM), race and study site. Metabolites with False discovery rate (FDR) adjusted p-values < 0.1 were called significant. Closed-set testing with the Simes method and Metabolon provided annotations (90 sets) was used to test metabolite sets. Results The cohort (n=175) was 72% female, 81% White and 16% DM with (1st Quartile/Median/3rd Quartile) Age: 42/54/63 years, BMI: 24.6/29.1/34.1 kg/m2, AHI: 0.9/4.3/12.6 and TRT< 90%: 2.3/37/87.4. 21 metabolites showed significant association with TRT< 90% (8 with positive direction/13 negative) and 0 with AHI. Among metabolite sets, Pregnenolone steroids had the strongest association with TRT< 90% (6 members; adjusted p=0.038; True Discovery Proportion (TDP)=0.33) followed by Fatty Acid Metabolism (Acyl Choline) (7 members; adjusted p-value=0.046; TDP=0.71) and Polyamine Metabolism (7 members; adjusted p-value p=0.054; TDP=0.14). Conclusion Three primary metabolite pathways were associated with sleep-related hypoxia in Group 1 PAH. Polyamine metabolism and fatty acid metabolism changes have previously been observed in a rat PAH model, consistent with our findings. Findings specific to metabolic biomarkers in SDB and PAH are novel and may lead to insights that broaden our understanding of PAH pathobiology and potential therapeutic targets. Support (if any)