Citations

Abstract Functional magnetic resonance imaging (fMRI) data are inherently complex, characterized by high dimensionality, intricate inter-regional dependencies, and substantial individual variability across experimental paradigms. Traditional linear mixed models (LMMs) provide a principled framework that models population-level fixed effects while estimating variance components arising from subject-level random effects; however, they often fail to adequately capture nonlinear relationships inherent in neuroimaging data. To address these limitations, we introduce the nonlinear mixed model (NMM) approach, an innovative extension of the LMM framework that integrates neural networks to flexibly model complex fixed-effect relationships while preserving the random-effects structure to account for individual differences. NMM advances fMRI analysis by: (1) identifying robust functional connectivity (FC) patterns consistently observed across multiple paradigms; (2) leveraging SHapley Additive exPlanations (SHAP) analysis to provide post-hoc interpretability of the nonlinear fixed effects, quantifying how age, sex, and paradigm contribute to predicted FC and how these effects are distributed across large-scale brain networks; and (3) using subject-specific random effects as neural fingerprints that not only show systematic variability across attention and default mode systems but also predict standardized cognitive scores, demonstrating biological relevance. Applied to the Philadelphia Neurodevelopmental Cohort (PNC) across emotion, n-back, and resting-state paradigms, NMM achieved superior model fit relative to classical LMMs, as evidenced by lower mean squared error (MSE) in predicting FC. This framework offers a statistically rigorous and practically explainable approach for modeling large-scale FC from modest covariates while explicitly separating population-level effects from stable individual variability in functional brain organization.

Abstract The use of chimeric antigen receptor (CAR) T-cell therapy is increasing for adult B-cell acute lymphoblastic leukemia (B-ALL), with 3 CD19 CAR T-cell products commercially available. Several key clinical questions related to best practices for CAR T-cell administration in this population exist, and limited prospective randomized trials have been conducted to fill these knowledge gaps. Thus, to help guide clinical practice, we conducted a modified Delphi study to develop and validate consensus recommendations on the administration of commercially available CAR T-cell therapy for adults with B-ALL. Consensus panelists (n = 9) included principal investigators (PIs) from Real World Outcomes Collaborative of CAR T-Cell Therapy in Adult ALL (ROCCA) consortium sites and were selected based on expertise and CAR T-cell center volume. Final panel consensus recommendations were distributed for rating by the remaining PIs from ROCCA consortium sites, which served as the validation group (n = 27). Consensus topics included patient selection, bridging and pre–CAR T-cell leukemia staging, lymphodepletion, and CAR T-cell treatment setting, specific toxicity prevention and management, post–CAR T-cell response assessment and disease monitoring, and the role of consolidation and/or maintenance therapies after CAR T-cell therapy. Initially, 58 recommendation statements were evaluated for consensus. After 2 panel meetings, a total of 34 statements achieved consensus rating among the expert panel. After rating by the validation group, all but 1 recommendation statement continued to meet consensus, for a total of 33 consensus recommendation statements on the administration of CAR T-cell therapy in adult B-ALL.

Abstract INTRODUCTION Research into cognitive dispersion – a cognitive process score measuring the intra‐individual variability (IIV) across a single testing session – suggests utility in neurodegenerative populations. Given widespread deficits observed in sporadic early‐onset Alzheimer's disease (EOAD), however, it is unclear if examining cognitive dispersion shows benefit in this condition. METHODS A total of 309 participants (188 amyloid‐positive EOAD, 43 amyloid‐negative early‐onset dementia [EOnonAD], 78 cognitively normal [CN]) completed neuropsychological testing twice over 12 months. Dispersion‐related differences among groups were assessed, as was cognitive dispersion's capacity to predict domain‐specific cognitive trajectories, and its convergence with imaging biomarkers. RESULTS EOAD participants displayed higher cognitive dispersion than EOnonAD participants, and associations with EOAD‐specific biomarkers. Additionally, cognitive dispersion was associated with 12‐month reliable change across several cognitive domains. DISCUSSION These preliminary results examine cognitive dispersion in sporadic EOAD. When used with baseline cognitive performance, cognitive dispersion measures may enrich future clinical trials in EOAD by enhancing treatment monitoring over time.

Abstract Inflammatory bowel diseases (IBD), principally Crohn’s disease (CD) and ulcerative colitis (UC), are common chronic disorders involving inflammation and often progressive tissue damage. Genome-wide association studies have mapped many risk signals, but the causal variants, effector genes and relevant cellular contexts remain difficult to resolve, limiting mechanistic interpretation and therapeutic translation. Here we performed a multi-ancestry GWAS meta-analysis of 125,992 individuals with IBD and more than 1.2 million controls, identifying 619 independent association signals (374 novel) at 420 IBD regions that account for 77–80% of SNP-based heritability. Fine-mapping resolved 81 high-confidence variants, 41 not previously reported. Although most signals were shared between CD and UC, 39% showed IBD subtype specificity, with UC signals showing stronger enrichment in functional annotations from intestinal epithelial, secretory and enteroendocrine cells, and CD showing stronger genetic correlations with circulating inflammatory biomarkers, including C-reactive protein and glycoprotein acetylation. Latent causal modelling supported a causal effect of decreased high-density lipoprotein on CD risk. By integrating bulk and single-cell eQTL and pQTL resources using colocalisation and Mendelian randomisation, together with coding-variant evidence from exome sequencing, we prioritised 664 candidate effector genes across 341 signals, including 390 newly implicated IBD genes, revealing new biological mechanisms and candidate therapeutic targets supported by human genetics.

Abstract Introduction Interstitial lung disease-associated pulmonary hypertension (ILD-PH) is a clinically challenging entity that lies on a pathophysiologic spectrum between pulmonary arterial hypertension (PAH) and interstitial lung disease (ILD). We compared the metabolomes of patients with ILD-PH to those with PAH and ILD to characterize the PAH-ILD axis and determine whether ILD-PH patients can be phenotyped based on their metabolomic similarity to PAH. Methods Using untargeted metabolomics profiles from patients with PAH, ILD, and ILD-PH in the large, U.S.-based, NIH-funded, multi-center PVDOMICS cohort, we trained an anchor-based classifier on PAH and ILD metabolomes using supervised PLS-DA with a 75/25 train/test split. Primary and secondary PH or risk-group classifications in PVDOMICS were based on expert adjudication of clinical data including right heart catheterization hemodynamics, pulmonary function testing, chest imaging, and laboratory results. We then applied the trained model to each ILD-PH patient to assign a PAH similarity score. We classified patients as having a PAH-like or ILD-like metabolomic phenotype using an optimal PAH similarity score cutoff of 0.5 then compared PAH metabolomic similarity to expert-adjudicated primary and secondary WSPH classifications in PVDOMICS. We implemented FELLA, a network-based metabolomics enrichment method, to holistically characterize the biologic underpinnings of PAH, ILD-PH, and ILD metabolomes. Results We analyzed 244 patients with PAH, 54 with ILD, and 47 with ILD-PH. The anchor-based classifier demonstrated high train-test performance with a c-statistic of 0.81 and 78% agreement with expert-adjudicated diagnoses. Among the ILD-PH patients, 23 were classified as PAH-like and 24 as ILD-like (74% agreement with expert adjudication, Cohen’s kappa 0.49, Hedge’s g estimate of 0.86). FELLA enrichment revealed core glycolytic and bioenergetic pathways were enriched in both pure PAH and ILD-PH. Comparison of diffusion Z-scores between groups demonstrated subtle but biologically coherent differences in several reaction-level nodes, including lactate, NADPH oxidoreductase, and phosphofructokinase, which were enriched in ILD-PH. Conclusion Patients with ILD-PH demonstrate clinically relevant metabolomic phenotypes. Examination of the network biology underlying ILD-PH may point to potential therapeutic targets in this vulnerable subgroup with lagging clinical outcomes. This abstract is funded by: Pulmonary Fibrosis Foundation PFF Scholars Program

Abstract Rationale The pathophysiology and clinical impact of liver injury in pulmonary hypertension (PH) is poorly understood. Liver injury in PH is often attributed to hepatic congestion, but some studies have failed to find strong associations with hemodynamics, suggesting alternative mechanisms. We sought to define the prevalence of liver injury and patterns of liver injury across the spectrum of PH and to determine whether patterns of liver injury were associated with clinical characteristics, hemodynamics and outcomes. Methods We performed an analysis of the PVDOMICS study, a multicenter, prospective, longitudinal cohort study that enrolled incident and prevalent patients with PH and disease comparators. The prevalence of liver injury, defined as aspartate or alanine aminotransferase (AST or ALT) or total bilirubin above the upper limit of normal, and different patterns of liver injury [hepatocellular (elevated AST or ALT with normal bilirubin), cholestatic (elevated bilirubin) and no liver injury] was defined among participants with known or suspected PH. Associations between patterns of liver injury with demographics, hemodynamics, right ventricular function, exercise capacity and quality of life (QoL) were assessed. We used Cox proportional hazards models adjusted for age and sex to assess the association of liver injury patterns with transplant-free survival. Results Liver injury was present in 27.8% (288/1036) of participants (19.0% cholestatic and 8.8% hepatocellular). Patients with liver injury were less likely to be female and have prevalent disease with differences in PH classification (Table 1). Patients with hepatocellular liver injury had a higher body mass index. A cholestatic liver injury pattern was most consistently associated with worse functional class, biventricular filling pressures, pulmonary vascular resistance, right ventricular function, natriuretic peptide elevation, exercise capacity, and REVEAL risk scores. Total bilirubin levels and right atrial pressure as a marker of systemic congestion were weakly correlated (r2=0.02, p < 0.0001). Compared to no liver injury, cholestatic liver injury was associated with an increased risk of death in age and sex adjusted models (HR 1.6, 95% CI 1.3-2.1, p < 0.001) but hepatocellular liver injury was not (p > 0.05). Conclusions Liver injury is common among patients with known or suspected PH. In contrast to a hepatocellular injury pattern, a cholestatic liver injury pattern was most consistently associated with worse hemodynamics, exercise capacity, QoL and survival. However, the weak relationship between right atrial pressure and bilirubin suggests that systemic congestion alone is unlikely to completely explain cholestatic liver injury. Further study is needed to uncouple the pathophysiological basis of cholestasis in PH. This abstract is funded by: NHLBI

Abstract Rationale Pulmonary hypertension (PH) is a heterogeneous condition classified into 5 WSPH groups but lacking detailed molecular distinction. The PVDOMICS cohort includes all PH groups alongside healthy and disease comparators, is deeply clinically phenotyped and has RNA sequencing data. Pseudogenes are conserved genome elements that due to sequence-defects cannot be translated into functional proteins. Some pseudogenes never-the-less produce RNA-transcripts which can regulate the expression of other, protein coding genes. Therefore, our study was focused on identifying differentially expressed pseudogene-transcripts within the dataset from the PVDOMICS study and gaining insights into their potential function. Methods Peripheral venous whole blood was collected from PVDOMICS participants [n = 525 PH patients; n = 312 controls without PH who were comparators (n = 256) or healthy (n = 56)]. RNA was isolated, RNA sequencing was performed. After quality control, differential transcript expression between PH and control groups was analyzed using the limma-voom pipeline, rSEA with adjustment for age, sex, race, smoking, WSPH group, latent factors and false discovery rate. We focused on a) transcripts expressed at increased levels (n = 218): the first 200 DEG (differentially-expressed-genes)-transcripts based on log2 difference plus increased DEG within the first 200 most significant p-values; and b) transcripts expressed at decreased levels (n = 292): last 200 DEG based on log2 difference plus decreased DEG within the first 200 most significant p-values. Results Six pseudogene transcripts were among the increased and 35 pseudogene transcripts were among the decreased DEG in the PH group compared to controls. Some of the differentially expressed pseudogenes (DEP) belonged to gene families, like the ribosomal protein family (7 DEP), mitochondria expressed transcripts (3 DEP), zinc finger proteins (2 DEP). Associations of pseudogenes with biological function or disease were extracted via GeneCards(R) and information from GWAS analysis. The DUSP8P5 pseudogene was among the DEP and is known for its association with left heart disease and hypertension; 2 additional DEP have known association with hypertension. Further known associations were with inhibitory or inflammatory immune processes (8 DEP), bone density (6 DEP), metabolism and diabetes (6 DEP), respiratory function or disease (4 DEP), erythrocyte biology (3 DEP), cancer (5 DEP), kidney disease (1 DEP), and demographic parameters like eye or hair color (2 DEP). Ten DEP had no known association with function or disease. Conclusions We identified pseudogene RNA-transcripts associated with PH across WSPH groups. The local vs. systemic effects of these pseudogene transcripts on PH remain to be determined and regulators of pseudogene-transcript expression (e.g. environmental exposures) identified. This abstract is funded by: 1 X01 HL148611 (TopMed), PVDOMICS NHLBI, PVDOMICS Pulmonary Hypertension Association, HT9425-25-1-0533 (GG), 1R21OH012955-01 (GG)

Abstract Rationale Lymphangioleiomyomatosis (LAM) is a rare, progressive, cystic lung disease affecting women. Sirolimus has been shown to stabilize lung function in women with LAM, but real-world data on long-term safety and efficacy are limited. Methods The NIH Multicenter Safety and Durability of Sirolimus for LAM (MIDAS, NCT02432560) observational registry was launched in 2015. All adult (≥18 years) women with LAM (treated or untreated) were eligible to enroll at any of the 20 U.S. sites and followed longitudinally with serial assessments of lung function, health-related quality of life, and treatment related adverse event (TRAEs). Baseline cohort characteristics and TRAEs are summarized using descriptive analyses. Marginal structural models (MSMs) with inverse probability of treatment weighting estimated the causal effect of sirolimus on annualized forced expiratory volume in one-second (FEV1) decline, adjusting for age at LAM diagnosis, prior treatment with sirolimus, menopausal status, race, presence of angiomyolipomas, supplemental oxygen use; time-varying covariates included diffusing capacity for carbon monoxide (DLCO) and serum vascular endothelial growth factor-D (VEGF-D) levels. Results Longitudinal data was available in 350 LAM patients, most of whom (246, 70.3%) had sporadic LAM. The median age at the baseline pulmonary function test (PFT) measurement was 46 years (range: 19 - 74) and the median (IQR) number of serial PFTs obtained per patient was 3 (1 - 5). The average follow-up duration per patient was 2.8 years. Over half of the patients (196, 56%) were treated with sirolimus during the MIDAS observation period for a median (IQR) treatment duration of 4.9 (2.4 - 7.5) years. MSM analysis demonstrated that sirolimus treatment was durably effective and reduced FEV₁ decline by 74.36 mL/year (95% CI: 15.87 - 132.85) compared with no treatment (-85.31 mL/year for the untreated group vs. -10.95 mL/year for the treated group, p = 0.01). Sirolimus was generally well tolerated; treatment interruption ≥3 months was observed in 38 (19.4%) of the 196 sirolimus treated patients. Logistic regression analysis showed that higher levels of FEV1 (odds ratio: OR = 3.12 [1.45 - 7.11], p = 0.005), VEGF-D (OR = 1.0007 [1.0004 - 1.001], p < 0.001), and DLCO (OR = 1.023 [1.0053 - 1.0413, p = 0.01]), were associated with treatment interruption. Conclusions In the MIDAS cohort, sirolimus treatment resulted in durable stabilization of lung function with a favorable safety profile over a 5-year period. These real-world data support sirolimus as a chronic suppressive therapy for LAM, demonstrating durable efficacy in slowing lung disease progression and tolerability with long-term use. This abstract is funded by: National Institutes of Health

Abstract Background Regulator of G protein signaling (RGS)2 terminates bronchoconstrictive Gaq signaling, and reductions in RGS2 expression associate with a hypercontractility phenotype. RGS2 promoter variants rs2746071 and rs2746072 are hypofunctional in gene reporter assays, and individuals with severe asthma homozygous for these two RGS2 variants experience higher asthma exacerbation rates. Whether these RGS2 promoter variants directly cause a hypercontractile human airway smooth muscle (HASM) phenotype is unknown. Methods We used the hTERT immortalized HASM cell line, derived from a human non-asthmatic donor. RGS2expression was reduced with siRNA (150nM anti-RGS2 vs. control siRNA) or through CRISPR transformation (Synthego©) to express the RGS2 promoter variants of interest, confirmed by Sanger sequencing. Gene and protein expression were quantified by qPCR and Western blotting. Intracellular Ca ++ flux and oscillation was measured with Fluo-8 (Abcam) with excitation/emission at 490/525 nm. HASM contractility to histamine treatment was measured using Traction Force Microscopy implemented in a 96-well plate format. Results Relative to cell lines with reference genotypes, HASM CRISPR transformed with the two RGS2 promoter variants exhibited a 41.8% (95% CI 28.6%--62.5%; p < 0.0001) reduction in RGS2 relative expression by qPCR, and a 35% (95% CI 8.6%—85.7%; p = 0.024) reduction in protein expression by Western blot. Methacholine-triggered calcium flux was nearly 50% greater for the cell lines with the RGS2 homozygous variants vs. those with reference genotypes (p = 0.043) as was the contractile response to histamine. HASM with reference genotypes exposed to anti-RGS2 siRNA showed the same phenotype, consistent with a causal relationship between the variants and expression (decreased) and receptor function (enhanced). Similarly, HASM CRISPR-transformed with the two RGS2 promoter variants exhibited an up to 3-fold increase in contraction magnitude (represented by strain energy, reported in pJ) in response to histamine >1uM (p < 0.05) relative to cell lines with reference genotypes. Conclusion Two promoter single nucleotide polymorphisms substitutions are sufficient to reduce RGS2 expression and induce a hypercontractile HASM phenotype. This abstract is funded by: AAAAI Bridge Award, NHLBI R21