Phenotypic Heterogeneity Research Articles

Background: Tirzepatide induces weight loss (WL), but its precise use among those more likely to benefit from WL and experience cardiometabolic effects can enable more scalable deployment. Phenotypic heterogeneity within RCT participants may reveal markers of response. Hypothesis: A phenomapping-informed machine learning tool derived from a phase 3 Tirzepatide RCT can define its individualized treatment effect (ITE) on weight loss and components of metabolic syndrome (MS) and generalize to another RCT population. Methods: In SURPASS-1, we calculated pairwise participant Gower’s distance similarity using 46 pre-randomized baseline characteristics and fit linear mixed models of neighborhoods with phenotypic similarity to estimate the ITE of percent WL and MS components (BMI, waist circumference, SBP, DBP, fasting glucose (FBG), and HDL change). We then trained XGBoost models with Boruta SHAP to predict the ITE of Tirzepatide on WL and each of the MS components. Then, we evaluated whether the ITE tool deployed in SURMOUNT-2 identified those with high predicted WL with mean difference ITE across responder tertiles, and the HR of time to reach WL >15% faster of high vs low responders. We then correlated predicted ITE of WL and each of the MS components in both RCTs. Results: In SURPASS-1 (n=357, 166 (46%) female, 64 (18%) obese), median WL was -8.2 v. -1.3% (-13.6, -4.6 v. -3.1, 1.0, 25, 75% IQR, T v P.) Our tool, developed in SURPASS-1, showed a significant treatment-ITE interaction (p< .001) for WL, BMI, HDL, and waist circumference, (p=.002) for DBP, and (p=0.014) for FGB and significant stratification of responders by ITE tertile compared to true WL (one-way ANOVA .p<.001) The most informative features were female sex, hypertension, age, FBG, baseline weight, and SBP. The validation RCT, SURMOUNT-2, (n=936, 476 (51%) female, 936 (100%) obese), had a significant treatment-ITE interaction estimate (p=.045). High responders had a predicted mean difference of WL of 15.7% (13.5%-17.9%, p <.001), moderate, 13.0% (10.9%-15.0%, p<.001), and low, 11.2% (9.0%-13.0%, p<.001) (one-way ANOVA p <.001, Fig. A). High responders had a faster time to WL > 15% than low with a median 36 vs 72 weeks, HR 2.1 (1.6-2.7, Fig. B). There were significant correlations (p<.001) between the predicted ITE of WL and MS components (Table 1). Conclusions: A machine learning tool can predict individualized WL and pleiotropic cardiometabolic effects by Tirzepatide in a new patient population.

Ataxia-telangiectasia (AT) is a rare neurodegenerative disorder caused by biallelic ATM gene mutations. While most patients exhibit classical features-progressive ataxia, oculocutaneous telangiectasia, and oculomotor apraxia-atypical presentations and overlapping phenotypes with AT-like disorders pose diagnostic challenges. To describe clinical and genetic findings in patients with suspected AT and assess the diagnostic utility of whole-exome sequencing (WES). We analyzed 20 patients with clinical features suggestive of AT who underwent genomic evaluation. Pathogenic or likely pathogenic ATM variants were found in 14 /19 patients with available data. Three had mutations in MRE11A or PCNA, consistent with ATLD1 and ATLD2, respectively. Two patients with classic phenotypes lacked conclusive genetic findings. Our findings highlight the phenotypic and genetic heterogeneity of AT and limitations of WES. We propose the integration of whole-genome sequencing (WGS) and RNA sequencing as complementary tools to improve diagnostic yield in AT and AT-like syndromes.

Background: Non-human primates (NHPs), particularly aged cynomolgus monkeys, share key physiologic, metabolic and genetic similarities with humans, offering valuable translational models for chronic cardiac diseases. Objective: This study aimed to establish and characterize chronic heart failure phenotypes, and investigate the presence of myocardial amyloidosis in aged NHPs using advanced multimodal imaging and biomarker profiling. Methods: Over 1,000 aged male cynomolgus monkeys underwent comprehensive cardiac evaluation. Ventricular structure and function were assessed by 2D-echocardiography (2D echo) and cardiac magnetic resonance imaging (MRI). Left ventricular myocardial proton density fat fraction (LV-PDFF) and epicardial adipose tissue (EAT) volume were determined. Myocardial amyloidosis was detected via 18F-florbetapir PET-CT. Circulating biomarkers of cardiac injury and inflammation (NT-proBNP, troponin I, hs-CRP, IL-1, IL-6, TNF-α) were measured. Healthy young NHPs served as controls. Results: Left-sided HF was present in 30% of aged NHPs with heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF) was equally represented. HFrEF models showed markedly reduced ejection fraction (EF<60%; 53 ± 4.6% vs. 73 ± 4.3% controls). HFpEF, was defined by normal/preserved EF (≥60%) and diastolic dysfunction markers (E/A<1, e'>7, and E/e'>11 for mild HFpEF; E/A>1, e'/a'<1, e'<7, and E/e'>11 for moderate HFpEF). Reduced MRI LV global longitudinal strain (GLS) and elevated extracellular volume fraction (ECV) were recorded for both HF subtypes indicating myocardial fibrosis. LV-PDFF was increased in HFpEF and HFrEF groups compared to controls. PET-CT identified myocardial amyloidosis in a subset of aged animals, independent of HF classification. Biomarkers of cardiac stress and systemic inflammation were significantly elevated in HF-affected NHPs. Conclusion: This study establishes aged cynomolgus monkeys as a translational model that recapitulates key features of human cardiac dysfunction, including heterogeneous HF phenotypes, and myocardial amyloidosis. The integration of multimodal imaging and biomarker profiling provides a robust platform for preclinical evaluation of novel therapeutics targeting HF subtypes and related cardiac pathologies.

Background: Congenital long QT syndrome (LQTS) is characterized by a prolongation of the heart rate-corrected QT interval (QTc) and an increased risk of syncope, seizure and sudden cardiac arrest/death (SCA/SCD). Genome-wide association studies have identified numerous QT interval-influencing loci of small effect sizes. Herein, we sought to investigate whether a polygenic background for the QT interval contributes to the phenotypic heterogeneity in patients with genotype-positive LQTS. Methods: We used summary statistics from a recent genome-wide meta-analysis for QT interval (n = 117,532) to derive 160 different polygenic scores for QT interval (PGS QT ) using LDpred2-grid, each comprising ~1.4 million common variants with varying assumptions of the underlying genetic architecture. Next, we applied these PGS QT to predict the QTc value in a genetically unrelated, European ancestry subset of the ‘All of Us’ cohort with available electrocardiogram and genetic data (n=6,404), where the best performing PGS QT was determined based on explained variance. Next, using 225 patients with genetically confirmed LQTS who were evaluated in the Mayo Clinic Windland Smith Rice Genetic Heart Rhythm Clinic, we tested the effect of the best performing PGS QT on QTc duration and LQTS-associated cardiac events (defined as a composite of syncope, seizures, sustained ventricular arrhythmia, appropriate ICD shocks, and SCA/SCD) using Bayesian regression models. Results: Overall, the best performing PGS QT explained 11.8% of the variance in the QTc in the ‘All of Us’ cohort. Among patients with genotype-positive LQTS (n [%]; KCNQ1 125 [55.6], KCHN2 76 [33.8], SCN5A 21 [9.3], Other 3 [1.3]), those with higher PGS QT had a higher QTc (predicted mean QTc [95% uncertainty interval], 95 th percentile vs 50 th percentile for PGS QT , 478 ms [464-491] vs 459 ms [449-468]). In comparison to the median PGS, those LQTS patients with a PGS QT >95 th percentile were more than twice as likely to have experienced a LQTS-associated cardiac event (Odds ratio [95% UI], 95 th percentile vs 50 th percentile for PGS QT ; 2.16 [1.02-5.41]), ( Figure ). Conclusions: Individuals with genotype-positive LQTS who have high polygenic susceptibility for QTc prolongation are more likely to phenotypically exhibit a higher QTc value and experience potentially life-threatening LQTS-associated cardiac events. Accounting for polygenic background may refine risk stratification strategies for patients with genotype-positive LQTS

Background: Coronary spastic angina (CSA) is a functional coronary disorder characterized by transient epicardial constriction in the absence of significant atherosclerosis. However, its clinical phenotypes remain poorly defined, and the contributions of metabolic and inflammatory factors to CSA susceptibility and vascular dysfunction are not fully understood. Hypothesis: We hypothesized that distinct CSA phenotypes can be identified through multidimensional profiling of metabolic and inflammatory stress, and that these phenotypes are associated with differences in endothelial function and spasm risk. Methods: We analyzed data from the FUJI-SPASM study, a prospective single-center registry of 1,125 patients undergoing intracoronary acetylcholine testing. Among them, 568 patients without obstructive coronary artery disease who presented ischemic symptoms were included. Unsupervised k-means clustering was applied to 19 standardized clinical and biochemical variables, including lipid and apolipoprotein profiles, glycemic indices (HOMA-IR, HbA1c), inflammatory markers (CRP, fibrinogen), and demographic factors. Principal component analysis (PCA) was then performed to extract latent axes of biological variation. Results: Clustering identified two phenotypes. Cluster 0 (metabolic phenotype) showed elevated triglycerides, non-HDL-C, apoB, apoC2, apoC3, FFA, and HOMA-IR. Cluster 1 (inflammatory phenotype) was characterized by higher CRP, fibrinogen, older age, and higher smoking. The frequency of CSA did not significantly differ between clusters (46% vs. 39%, p = 0.10). PCA revealed two principal components: PC1, primarily loaded with atherogenic lipids and insulin resistance indices, and PC2, characterized by positive loadings for CRP and fibrinogen and inverse loadings for HDL-related apolipoproteins—representing metabolic and inflammatory burdens, respectively. In a subanalysis of 159 patients with flow-mediated dilation (FMD) data, those with high PC1 and PC2 scores had significantly lower FMD values (p < 0.001) and higher CSA positivity (p < 0.05), suggesting cumulative vascular impact of metabolic and inflammatory stress. Conclusion: This study identified two clinically relevant CSA phenotypes defined by metabolic and inflammatory profiles. PCA further demonstrated that a dual burden of these biological stresses is closely linked to endothelial dysfunction and increased CSA risk. These findings support a phenotype-based, multidimensional approach to CSA evaluation.

Background: Despite recognized phenotypic heterogeneity of idiopathic non-ischemic cardiomyopathy (NICM), management remains generalized and dominantly guided by left ventricular (LV) ejection fraction (EF) and New York Heart Association class. Deep learning (DL) models can integrate complex, multimodal phenomics data to support individualized prognostication; however, are considered black box models that limit clinical adoption. We developed an explainable DL model, DeepPhenome-NICM, for patient-specific prediction of major adverse cardiac events (MACE) in NICM leveraging multi-domain phenomics data captured at time of cardiovascular MR (CMR) imaging. Methods: 1,142 patients with CMR confirmed diagnosis of NICM were identified from the Cardiovascular Imaging Registry of Calgary, defined as LV EF <50% in the absence of any identifiable ischaemic or non-ischaemic aetiology. All patients underwent baseline health questionnaires with standardized reporting at the time of CMR imaging and were followed for a minimum of 6 months for the composite outcome of all-cause mortality, survived cardiac arrest, ventricular tachycardia, or heart failure hospitalization. A total of 50 routinely captured variables were included in a final trained DL survival model (DeepPhenome-NICM), inclusive of patient-reported, CMR-derived, and electronic health record-derived variables. Data were split into training (80%) and test (20%) sets. Model performance was assessed on the test set. Shapley values, a measure of additive feature contribution to model prediction, were estimated to deliver model explainability. Results: Baseline characteristics of the study population are reported in Table 1. Over a median follow-up of 3.8 years, 210 patients (18.4%) experienced MACE. Using the hold-out test set, the DeepPhenome-NICM model achieved a mean time-dependent AUC of 0.83 (95% CI 0.75-0.89) with a 1- and 5-year AUC of 0.87 (0.79-0.93) and 0.82 (0.73-0.89), respectively. Stratification of patients by the median predicted patient-specific risk score yielded significant discrimination of event-free survival, with the high-risk group experiencing a 4.3-fold increased risk (HR; 95% CI 2.1-8.9; p<0.001; Figure 1) . Figure 2 shows the respective influence of top predictors on model prediction. Conclusions: DeepPhenome-NICM is a multimodal DL model that identifies high risk patients with NICM at time of CMR using a composite phenomics based approach. External validation of this model is planned.

Background: Pediatric gastroesophageal reflux disease (GERD) presents with heterogeneous phenotypes across ages, and diagnostic challenges persist. Multichannel intraluminal impedance and pH (MII-pH) monitoring is the current gold standard, while mean nocturnal baseline impedance (MNBI) has emerged as a marker of mucosal integrity in adults. Pediatric normative data are still lacking. We aim to characterize age-related reflux patterns and assess the association between MNBI and pathological acid exposure in children undergoing MII-pH monitoring. Methods: We retrospectively analyzed 226 children evaluated with 24 h MII-pH monitoring between 2017 and 2025. Clinical and laboratory data were reviewed. Children were stratified by age (<1 year vs. ≥1 year). Pathological reflux was defined as reflux index (RI) > 10% in infants and >7% in older children. MNBI was measured at distal channels (Z5, Z6) during nocturnal recumbency. Correlations between MNBI and RI were assessed. Diagnostic performance of MNBI for pathological acid exposure was evaluated using ROC analysis. Results: Infants had more total reflux episodes, particularly weakly acidic and liquid, whereas older children exhibited more gas reflux. MNBI correlated inversely with RI (Z6 r = −0.337; Z5 r = −0.281; both p < 0.0001). In infants, MNBI poorly discriminated pathological acid exposure. In older children, MNBI showed better performance, with optimal cut-offs of ~2525 Ω (Z6) and 3079 Ω (Z5) yielding specificities > 85%. Conclusions: MNBI is a reproducible, age-sensitive marker of reflux burden in children, best suited for older children where it complements RI in diagnosing acid-mediated GERD. Larger prospective studies are needed to establish pediatric reference values.

Abnormalities in PBX1 represent a monogenic cause of congenital anomalies of the kidney and urinary tract (CAKUT). However, their phenotypic heterogeneity poses a challenge for timely detection, particularly in the absence of overt anomalies in the kidney and urinary tract. Here, we present a 28-year-old male diagnosed with a rare PBX1 nonsense variant identified during the evaluation of early-onset chronic kidney disease. As part of the initial workup for decreased renal function and proteinuria, a kidney biopsy was performed, revealing focal segmental glomerulosclerosis (FSGS) and acute tubular necrosis without an identifiable cause. He was initially treated with renin-angiotensin system inhibitors, followed by glucocorticoid and/or cyclosporine therapy for four years. Despite these interventions, his serum creatinine levels gradually increased without any improvement in proteinuria. Genetic testing, performed seven years after the initial visit, revealed a rare de novo heterozygous PBX1 variant, p.Arg93Ter (c.277C > T), classified as likely pathogenic. Reverse phenotyping identified cryptorchidism and dysmorphic external ears, both of which are extrarenal manifestations commonly associated with PBX1 -related CAKUT. Although this variant is predicted to be deleterious, it is flagged as escaping nonsense-mediated decay, which may explain the absence of apparent structural anomalies in the kidneys. PBX1 is prominently expressed in interstitial and endothelial cells in both fetal and adult human kidneys, and its function is not directly implicated in podocyte or tubular cell biology. Therefore, the inadvertent pathological findings in this genetic disorder may be attributed to reduced nephron endowment and/or disturbance in reciprocal cellular interactions. This case broadens the phenotypic spectrum of PBX1 -related disorders and highlights its renal manifestations, further expanding the clinical heterogeneity of FSGS.

Integrated cortical and plasma proteomic analysis of Cntnap2 knockout mice and human models of autism spectrum disorder: Potential involvement of galectin-3-binding protein.

Employing nullclines to balance treatment efficacy and neurotoxicity for sustained tumor control.

A possible role of Apolipoprotein E in Idiopathic Generalized Epilepsy.

14494 Heterogeneity of Molecular Phenotypes in Uterine Fibroids: Role of MED12 Mutations in Gene Expression Profiles

Differential impact of mutant Ataxin-3 in hindbrain regions: further evidence of white matter loss as a core pathological feature.

Phenotypic heterogeneity-distinct molecular and behavioral variations within a population-significantly influences collective invasion and tumor progression. Here, we use a molecular approach to explore how the underlying metabolic heterogeneity in non-small cell lung cancer (NSCLC) influences invasion and pack patterning. Assessment of three-dimensional (3D) pack patterning revealed invasive heterogeneity across NSCLC cell lines and patient-derived samples. Flow cytometry identified IL13RA2 as a biomarker for invasive potential, enabling isolation of subpopulations with distinct invasive characteristics. By integrating a cell surface biomarker (IL13RA2±) with mitochondrial membrane potential (TMRM), we identified and isolated three distinct subpopulations. Two-dimensional (2D) analyses revealed differences in mitochondrial polarity and transcriptional programs associated with migration and oxygensensitivity. In 3D, these subpopulations invaded with distinct patterns, from contiguous circular packs to structured chains. Assessments under varied oxygen tension demonstrated that oxygen availability and subpopulation metabolism together influence collective invasion patterning. When recombined at ratios recapitulating the original population, both stochastic and opportunistic cooperative dynamics emerged, dependent on subpopulation composition and oxygen levels. Our molecular approach, integrating cell surface and metabolic characteristics, enables the isolation of unique subpopulations and demonstrates that phenotypic and metabolic heterogeneity, population composition, and oxygen availability collectively pattern invasion packs and drive collective invasion.

Exercise-induced hypoalgesia (EIH) describes an acute reduction in pain following exercise. Individuals with low back pain (LBP) demonstrate variable EIH responses, such as increased pain sensitivity (hyperalgesia). Pain phenotyping, classifying individuals with pain sensitivity and psychological characteristics, may explain this variability. LBP phenotypes were classified using pain sensitivity, psychological variables, and EIH response to dynamic resistance exercise. In this observational cohort study, 78 individuals with LBP (mean age = 21.7 ± 2.6 years, 59% female) completed clinical pain ratings (Numeric Pain Rating Scale [NPRS]), temporal summation (TS), conditioned pain modulation (CPM), psychological questionnaires (Fear-Avoidance Beliefs Questionnaire-Physical Activity subscale [FABQ-PA]), and a 45-degree Roman chair back extension exercise. Pressure pain threshold (PPT) pre- and post-exercise determined EIH. Ward's hierarchical cluster analysis was performed with five variables (NPRS, TS, CPM, FABQ-PA, EIH). One-way ANOVA examined differences in demographic, clinical, and psychosocial factors by cluster. Three distinct clusters emerged: (1) efficient pain modulation with low fear-avoidance and high EIH (n = 40); (2) impaired pain modulation with moderate fear-avoidance and hyperalgesia (n = 19); (3) moderate pain modulation with high fear-avoidance and moderate EIH (n = 19). Clusters differed significantly by age (p = 0.02, ηp 2 = 0.098), disability (p = 0.004, ηp 2 = 0.138), psychosocial burden (p = 0.03, ηp 2 = 0.086), and pain catastrophizing (p = 0.03, ηp 2 = 0.09). Individuals with LBP exhibit heterogeneous pain phenotypes. Psychological factors may diminish EIH, and impaired modulation may produce hyperalgesia post-exercise. Identifying pain phenotypes may enhance personalised rehabilitation strategies, optimising interventions for LBP. This study provides novel evidence that biopsychosocial phenotypes influence pain responses to resistance exercise, offering a potential pathway to personalised rehabilitation for LBP.

The somatic activating variant in BRAF (p.V600E) was recently described as a novel cause of macrocystic head and neck lymphatic malformations in three individuals. Other recent studies profiling the genetic causes of more complex lymphatic anomalies identified this same pathogenic BRAF variant. Our aim was to expand the phenotypic description of the somatic BRAF p.V600E variant in individuals with vascular anomalies. We searched the database of individuals with vascular anomalies at our institution for those identified as having complex lymphatic anomalies and somatic BRAF p.V600E variants. A comprehensive retrospective review of identified individuals' electronic health records was conducted. Six individuals with complex lymphatic anomalies had the BRAF p.V600E variant. All individuals had diffuse lymphatic malformations and abnormal lymphatic conduction. The conduction abnormalities were observed only after surgical interventions in five of the patients, in some cases, manifesting years later. There was immense phenotypic heterogeneity within the cohort. Complex lymphatic anomalies are an important new phenotype associated with the pathogenic BRAF p.V600E variant. Even in initially asymptomatic individuals within this patient population, longitudinal follow-up is necessary, and surgical intervention should be pursued with caution. Further investigation is needed to better understand the etiology of the conduction problems associated with this pathogenic variant to inform the multidisciplinary approach to treatment.

Microscopic nucleic acid-to-protein ratio: A label-free approach for pancreatic cancer detection.

Osteoarthritis (OA) is a predominantly degenerative disease with heterogeneous phenotypes, including subsets with low-grade inflammation. Emerging data suggest that the gut microbiota may contribute to OA biology. While noting that definitive, longitudinal links between acupuncture-induced microbiome modulation and joint outcomes are yet to be established clinically, acupuncture has been associated with symptomatic improvement of OA alongside immune and microbiome changes. Within a brain-gut-joint framework, preclinical, and mechanistic studies (<i>N</i> = 13) were assessed in this scoping review. The neuro-immune pathways, such as the vagal–adrenal/splenic, and sympatho–adrenal/splenic axes, when activated by acupuncture, can modulate gut microbial composition. These activated pathways likely involve the vagus nerve (a component of the parasympathetic nervous system) and the sympathetic nerves (which are connected to gastrointestinal-related organs, including the adrenal gland, spleen, and gut). Acupuncture may also activate the neuroendocrine system via the hypothalamic–pituitary–adrenal axis, which regulates stress, thereby releasing hormones that contribute to anti-inflammatory effects. Acupuncture may support the treatment and management of OA via the brain-gut-joint axis. Given that acupuncture is associated with fewer side effects than conventional medications, it represents a promising therapeutic strategy for OA.

Age-related hearing loss affects one-third of the population over 65 years. However, the diverse pathologies underlying these heterogeneous phenotypes complicate genetic studies. Here we show that by applying computational phenotyping approaches based on audiometrically measured hearing loss, we can overcome challenges associated with accurate phenotyping for older adults with hearing loss. Using this phenotyping strategy, we uncover differences in the associations observed between genetic variants and sensory and metabolic hearing loss. Sex-stratified analyses of these sexually dimorphic hearing loss phenotypes reveal a locus of relevance to sensory hearing loss in males, but not females. Enrichment analyses implicate genes involved in frontotemporal dementia in metabolic hearing loss, while genes relating to sensory processing of sound by hair cells are implicated in sensory hearing loss. Our study enhances our understanding of these two hearing loss phenotypes, representing the first step in the development of more precise treatments for these pathologically distinct hearing loss phenotypes.

Genotype-Phenotype Characterization of Lynch Syndrome-associated Upper Tract Urothelial Carcinoma.