Multi-omics machine learning to study host-microbiome interactions in early-onset colorectal cancer.

The incidence of early-onset colorectal cancer (eoCRC) is rising, and its pathogenesis is not completely understood. We hypothesized that machine learning utilizing paired tissue microbiome and plasma metabolome features could uncover distinct host-microbiome associations between eoCRC and average-onset CRC (aoCRC). Individuals with stages I-IV CRC (n = 64) were categorized as eoCRC (age ≤ 50, n = 20) or aoCRC (age ≥ 60, n = 44). Untargeted plasma metabolomics and 16S rRNA amplicon sequencing (microbiome analysis) of tumor tissue were performed. We fit DIABLO (Data Integration Analysis for Biomarker Discovery using Latent variable approaches for Omics studies) to construct a supervised machine-learning classifier using paired multi-omics (microbiome and metabolomics) data and identify associations unique to eoCRC. A differential association network analysis was also performed. Distinct clustering patterns emerged in multi-omic dimension reduction analysis. The metabolomics classifier achieved an AUC of 0.98, compared to AUC 0.61 for microbiome-based classifier. Circular correlation technique highlighted several key associations. Metabolites glycerol and pseudouridine (higher abundance in individuals with aoCRC) had negative correlations with Parasutterella, and Ruminococcaceae (higher abundance in individuals with eoCRC). Cholesterol and xylitol correlated negatively with Erysipelatoclostridium and Eubacterium, and showed a positive correlation with Acidovorax with higher abundance in individuals with eoCRC. Network analysis revealed different clustering patterns and associations for several metabolites e.g.: urea cycle metabolites and microbes such as Akkermansia. We show that multi-omics analysis can be utilized to study host-microbiome correlations in eoCRC and demonstrates promising biomarker potential of a metabolomics classifier. The distinct host-microbiome correlations for urea cycle in eoCRC may offer opportunities for therapeutic interventions.

Polygenic subtype identified in ACCORD trial displays a favorable type 2 diabetes phenotype in the UKBiobank population.

INTRODUCTION: We previously identified a genetic subtype (C4) of type 2 diabetes (T2D), benefitting from intensive glycemia treatment in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial. Here, we characterized the population of patients that met the C4 criteria in the UKBiobank cohort. RESEARCH DESIGN AND METHODS: Using our polygenic score (PS), we identified C4 individuals in the UKBiobank and tested C4 status with risk of developing T2D, cardiovascular disease (CVD) outcomes, and differences in T2D medications. RESULTS: C4 individuals were less likely to develop T2D, were slightly older at T2D diagnosis, had lower HbA1c values, and were less likely to be prescribed T2D medications (P < .05). Genetic variants in MAS1 and IGF2R, major components of the C4 PS, were associated with fewer overall T2D prescriptions. CONCLUSION: We have confirmed C4 individuals are a lower risk subpopulation of patients with T2D.

An interpretable predictive deep learning platform for pediatric metabolic diseases.

OBJECTIVES: Metabolic disease in children is increasing worldwide and predisposes a wide array of chronic comorbid conditions with severe impacts on quality of life. Tools for early detection are needed to promptly intervene to prevent or slow the development of these long-term complications. MATERIALS AND METHODS: No clinically available tools are currently in widespread use that can predict the onset of metabolic diseases in pediatric patients. Here, we use interpretable deep learning, leveraging longitudinal clinical measurements, demographical data, and diagnosis codes from electronic health record data from a large integrated health system to predict the onset of prediabetes, type 2 diabetes (T2D), and metabolic syndrome in pediatric cohorts. RESULTS: The cohort included 49 517 children with overweight or obesity aged 2-18 (54.9% male, 73% Caucasian), with a median follow-up time of 7.5 years and mean body mass index (BMI) percentile of 88.6%. Our model demonstrated area under receiver operating characteristic curve (AUC) accuracies up to 0.87, 0.79, and 0.79 for predicting T2D, metabolic syndrome, and prediabetes, respectively. Whereas most risk calculators use only recently available data, incorporating longitudinal data improved AUCs by 13.04%, 11.48%, and 11.67% for T2D, syndrome, and prediabetes, respectively, versus models using the most recent BMI (P < 2.2 × 10-16). DISCUSSION: Despite most risk calculators using only the most recent data, incorporating longitudinal data improved the model accuracies because utilizing trajectories provides a more comprehensive characterization of the patient's health history. Our interpretable model indicated that BMI trajectories were consistently identified as one of the most influential features for prediction, highlighting the advantages of incorporating longitudinal data when available.

Plasma metabolomic differences in early-onset compared to average-onset colorectal cancer.

Deleterious effects of environmental exposures may contribute to the rising incidence of early-onset colorectal cancer (eoCRC). We assessed the metabolomic differences between patients with eoCRC, average-onset CRC (aoCRC), and non-CRC controls, to understand pathogenic mechanisms. Patients with stage I-IV CRC and non-CRC controls were categorized based on age ≤ 50 years (eoCRC or young non-CRC controls) or  ≥ 60 years (aoCRC or older non-CRC controls). Differential metabolite abundance and metabolic pathway analyses were performed on plasma samples. Multivariate Cox proportional hazards modeling was used for survival analyses. All P values were adjusted for multiple testing (false discovery rate, FDR P < 0.15 considered significant). The study population comprised 170 patients with CRC (66 eoCRC and 104 aoCRC) and 49 non-CRC controls (34 young and 15 older). Citrate was differentially abundant in aoCRC vs. eoCRC in adjusted analysis (Odds Ratio = 21.8, FDR P = 0.04). Metabolic pathways altered in patients with aoCRC versus eoCRC included arginine biosynthesis, FDR P = 0.02; glyoxylate and dicarboxylate metabolism, FDR P = 0.005; citrate cycle, FDR P = 0.04; alanine, aspartate, and glutamate metabolism, FDR P = 0.01; glycine, serine, and threonine metabolism, FDR P = 0.14; and amino-acid t-RNA biosynthesis, FDR P = 0.01. 4-hydroxyhippuric acid was significantly associated with overall survival in all patients with CRC (Hazards ratio, HR = 0.4, 95% CI 0.3-0.7, FDR P = 0.05). We identified several unique metabolic alterations, particularly the significant differential abundance of citrate in aoCRC versus eoCRC. Arginine biosynthesis was the most enriched by the differentially altered metabolites. The findings hold promise in developing strategies for early detection and novel therapies.

Astrocyte interferon-gamma signaling dampens inflammation during chronic central nervous system autoimmunity via PD-L1.

Multiple sclerosis (MS) is an inflammatory and neurodegenerative disease of the central nervous system (CNS). Infiltrating inflammatory immune cells perpetuate demyelination and axonal damage in the CNS and significantly contribute to pathology and clinical deficits. While the cytokine interferon (IFN)γ is classically described as deleterious in acute CNS autoimmunity, we and others have shown astrocytic IFNγ signaling also has a neuroprotective role. Here, we performed RNA sequencing and ingenuity pathway analysis on IFNγ-treated astrocytes and found that PD-L1 was prominently expressed. Interestingly, PD-1/PD-L1 antagonism reduced apoptosis in leukocytes exposed to IFNγ-treated astrocytes in vitro. To further elucidate the role of astrocytic IFNγ signaling on the PD-1/PD-L1 axis in vivo, we induced the experimental autoimmune encephalomyelitis (EAE) model of MS in Aldh1l1-CreERT2, Ifngr1fl/fl mice. Mice with conditional astrocytic deletion of IFNγ receptor exhibited a reduction in PD-L1 expression which corresponded to increased infiltrating leukocytes, particularly from the myeloid lineage, and exacerbated clinical disease. PD-1 agonism reduced EAE severity and CNS-infiltrating leukocytes. Importantly, PD-1 is expressed by myeloid cells surrounding MS lesions. These data support that IFNγ signaling in astrocytes diminishes inflammation during chronic autoimmunity via upregulation of PD-L1, suggesting potential therapeutic benefit for MS patients.

Characterization of Salivary and Plasma Metabolites as Biomarkers for HCC: A Pilot Study.

(1) Background: The incidence of hepatocellular carcinoma (HCC) is rising, and current screening methods lack sensitivity. This study aimed to identify distinct and overlapping metabolites in saliva and plasma that are significantly associated with HCC. (2) Methods: Saliva samples were collected from 42 individuals (HCC = 16, cirrhosis = 12, healthy = 14), with plasma samples from 22 (HCC = 14, cirrhosis = 2, healthy = 6). We performed untargeted mass spectrometry on blood and plasma, tested metabolites for associations with HCC or cirrhosis using a logistic regression, and identified enriched pathways with Metaboanalyst. Pearson's correlation was employed to test for correlations between salivary and plasma metabolites. (3) Results: Six salivary metabolites (1-hexadecanol, isooctanol, malonic acid, N-acetyl-valine, octadecanol, and succinic acid) and ten plasma metabolites (glycine, 3-(4-hydroxyphenyl)propionic acid, aconitic acid, isocitric acid, tagatose, cellobiose, fucose, glyceric acid, isocitric acid, isothreonic acid, and phenylacetic acid) were associated with HCC. Malonic acid was correlated between the paired saliva and plasma samples. Pathway analysis highlighted deregulation of the 'The Citric Acid Cycle' in both biospecimens. (4) Conclusions: Our study suggests that salivary and plasma metabolites may serve as independent sources for HCC detection. Despite the lack of correlation between individual metabolites, they converge on 'The Citric Acid Cycle' pathway, implicated in HCC pathogenesis.

Biomarkers for Response to Anti-PD-1/Anti-PD-L1 Immune Checkpoint Inhibitors: A Large Meta-Analysis.

BACKGROUND: Immune checkpoint inhibitors (ICIs) that block PD-1/PD-L1 have consistently demonstrated durable clinical activity across multiple histologies but have low overall response rates for many cancers-indicating that too few patients benefit from ICIs. Many studies have explored potential predictive biomarkers (eg, PD-1/PD-L1 expression, tumor mutational burden [TMB]), no consensus biomarker has been identified. METHODS: This meta-analysis combined predictive accuracy metrics for various biomarkers, across multiple cancer types, to determine which biomarkers are most accurate for predicting ICI response. Data from 18,792 patients from 100 peer-reviewed studies that evaluated putative biomarkers for response to anti-PD-1/anti- PD-L1 treatment were meta-analyzed using bivariate linear mixed models. Biomarker performance was assessed based on the global area under the receiver operating characteristic curve (AUC) and 95% bootstrap confidence intervals. RESULTS: PD-L1 immunohistochemistry, TMB, and multimodal biomarkers discriminated responders and nonresponders better than random assignment (AUCs >.50). Excluding multimodal biomarkers, these biomarkers correctly classified at least 50% of the responders (sensitivity 95% CIs, >.50). Notably, variation in biomarker performance was observed across cancer types. CONCLUSIONS: Although some biomarkers consistently performed better, heterogeneity in performance was observed across cancer types, and additional research is needed to identify highly accurate and precise biomarkers for widespread clinical use.

Impacts of COVID-19 on Glycemia and Risk of Diabetic Ketoacidosis.

Reports indicate that coronavirus disease 2019 (COVID-19) may impact pancreatic function and increase type 2 diabetes (T2D) risk, although real-world COVID-19 impacts on HbA1c and T2D are unknown. We tested whether COVID-19 increased HbA1c, risk of T2D, or diabetic ketoacidosis (DKA). We compared pre- and post-COVID-19 HbA1c and T2D risk in a large real-world clinical cohort of 8,755 COVID-19(+) patients and 11,998 COVID-19(-) matched control subjects. We investigated whether DKA risk was modified in COVID-19(+) patients with type 1 diabetes (T1D) (N = 701) or T2D (N = 21,830), or by race and sex. We observed a statistically significant, albeit clinically insignificant, HbA1c increase post-COVID-19 (all patients ΔHbA1c = 0.06%; with T2D ΔHbA1c = 0.1%) and no increase among COVID-19(-) patients. COVID-19(+) patients were 40% more likely to be diagnosed with T2D compared with COVID-19(-) patients and 28% more likely for the same HbA1c change as COVID-19(-) patients, indicating that COVID-19-attributed T2D risk may be due to increased recognition during COVID-19 management. DKA in COVID-19(+) patients with T1D was not increased. COVID-19(+) Black patients with T2D displayed disproportionately increased DKA risk (hazard ratio 2.46 [95% CI 1.48-6.09], P = 0.004) compared with White patients, suggesting a need for further clinical awareness and investigation.

Diet quality indices and changes in cognition during chemotherapy.

PURPOSE: No evidence-based prevention strategies currently exist for cancer-related cognitive decline (CRCD). Although patients are often advised to engage in healthy lifestyle activities (e.g., nutritious diet), little is known about the impact of diet on preventing CRCD. This secondary analysis evaluated the association of pre-treatment diet quality indices on change in self-reported cognition during chemotherapy. METHODS: Study participants (n = 96) completed the Block Brief Food Frequency Questionnaire (FFQ) before receiving their first infusion and the PROMIS cognitive function and cognitive abilities questionnaires before infusion and again 5 days later (i.e., when symptoms were expected to be their worst). Diet quality indices included the Dietary Approaches to Stop Hypertension (DASH), Alternate Mediterranean Diet (aMED), and a low carbohydrate diet index and their components. Descriptive statistics were generated for demographic and clinical variables and diet indices. Residualized change models were computed to examine whether diet was associated with change in cognitive function and cognitive abilities, controlling for age, sex, cancer type, treatment type, depression, and fatigue. RESULTS: Study participants had a mean age of 59 ± 10.8 years and 69% were female. Although total diet index scores did not predict change in cognitive function or cognitive abilities, higher pre-treatment ratio of aMED monounsaturated/saturated fat was associated with less decline in cognitive function and cognitive abilities at 5-day post-infusion (P ≤ .001). CONCLUSIONS: Higher pre-treatment ratio of monounsaturated/saturated fat intake was associated with less CRCD early in chemotherapy. Results suggest greater monounsaturated fat and less saturated fat intake could be protective against CRCD during chemotherapy.

Identification of robust deep neural network models of longitudinal clinical measurements.

Deep learning (DL) from electronic health records holds promise for disease prediction, but systematic methods for learning from simulated longitudinal clinical measurements have yet to be reported. We compared nine DL frameworks using simulated body mass index (BMI), glucose, and systolic blood pressure trajectories, independently isolated shape and magnitude changes, and evaluated model performance across various parameters (e.g., irregularity, missingness). Overall, discrimination based on variation in shape was more challenging than magnitude. Time-series forest-convolutional neural networks (TSF-CNN) and Gramian angular field(GAF)-CNN outperformed other approaches (P < 0.05) with overall area-under-the-curve (AUCs) of 0.93 for both models, and 0.92 and 0.89 for variation in magnitude and shape with up to 50% missing data. Furthermore, in a real-world assessment, the TSF-CNN model predicted T2D with AUCs reaching 0.72 using only BMI trajectories. In conclusion, we performed an extensive evaluation of DL approaches and identified robust modeling frameworks for disease prediction based on longitudinal clinical measurements.

Salivary miRNAs as non-invasive biomarkers of hepatocellular carcinoma: a pilot study.

BACKGROUND: Improved detection of hepatocellular carcinoma (HCC) is needed, as current detection methods, such as alpha fetoprotein (AFP) and ultrasound, suffer from poor sensitivity. MicroRNAs (miRNAs) are small, non-coding RNAs that regulate many cellular functions and impact cancer development and progression. Notably, miRNAs are detectable in saliva and have shown potential as non-invasive biomarkers for a number of cancers including breast, oral, and lung cancers. Here, we present, to our knowledge, the first report of salivary miRNAs in HCC and compare these findings to patients with cirrhosis, a high-risk cohort for HCC. METHODS: We performed small RNA sequencing in 20 patients with HCC and 19 with cirrhosis. Eleven patients with HCC had chronic liver disease, and analyses were performed with these samples combined and stratified by the presence of chronic liver disease. P values were adjusted for multiple comparisons using a false discovery rate (FDR) approach and miRNA with FDR P < 0.05 were considered statistically significant. Differential expression of salivary miRNAs was compared to a previously published report of miRNAs in liver tissue of patients with HCC vs cirrhosis. Support vector machines and leave-one-out cross-validation were performed to determine if salivary miRNAs have predictive potential for detecting HCC. RESULTS: A total of 4,565 precursor and mature miRNAs were detected in saliva and 365 were significantly different between those with HCC compared to cirrhosis (FDR P < 0.05). Interestingly, 283 of these miRNAs were significantly downregulated in patients with HCC. Machine-learning identified a combination of 10 miRNAs and covariates that accurately classified patients with HCC (AUC = 0.87). In addition, we identified three miRNAs that were differentially expressed in HCC saliva samples and in a previously published study of miRNAs in HCC tissue compared to cirrhotic liver tissue. CONCLUSIONS: This study demonstrates, for the first time, that miRNAs relevant to HCC are detectable in saliva, that salivary miRNA signatures show potential to be highly sensitive and specific non-invasive biomarkers of HCC, and that additional studies utilizing larger cohorts are needed.

Metagenomics and chemotherapy-induced nausea: A roadmap for future research.

Uncontrolled chemotherapy-induced nausea and vomiting can reduce patients' quality of life and may result in premature discontinuation of chemotherapy. Although nausea and vomiting are commonly grouped together, research has shown that antiemetics are clinically effective against chemotherapy-induced vomiting (CIV) but less so against chemotherapy-induced nausea (CIN). Nausea remains a problem for up to 68% of patients who are prescribed guideline-consistent antiemetics. Despite the high prevalence of CIN, relatively little is known regarding its etiology independent of CIV. This review summarizes a metagenomics approach to the study and treatment of CIN with the goal of encouraging future research. Metagenomics focuses on genetic risk factors and encompasses both human (ie, host) and gut microbial genetic variation. Little work to date has focused on metagenomics as a putative biological mechanism of CIN. Metagenomics has the potential to be a powerful tool in advancing scientific understanding of CIN by identifying new biological pathways and intervention targets. The investigation of metagenomics in the context of well-established demographic, clinical, and patient-reported risk factors may help to identify patients at risk and facilitate the prevention and management of CIN.

Associations of weight loss with obesity-related comorbidities in a large integrated health system.

AIMS: To determine the health outcomes associated with weight loss in individuals with obesity, and to better understand the relationship between disease burden (disease burden; ie, prior comorbidities, healthcare utilization) and weight loss in individuals with obesity by analysing electronic health records (EHRs). MATERIALS AND METHODS: We conducted a case-control study using deidentified EHR-derived information from 204 921 patients seen at the Cleveland Clinic between 2000 and 2018. Patients were aged ≥20 years with body mass index ≥30 kg/m2 and had ≥7 weight measurements, over ≥3 years. Thirty outcomes were investigated, including chronic and acute diseases, as well as psychological and metabolic disorders. Weight change was investigated 3, 5 and 10 years prior to an event. RESULTS: Weight loss was associated with reduced incidence of many outcomes (eg, type 2 diabetes, nonalcoholic steatohepatitis/nonalcoholic fatty liver disease, obstructive sleep apnoea, hypertension; P < 0.05). Weight loss >10% was associated with increased incidence of certain outcomes including stroke and substance abuse. However, many outcomes that increased with weight loss were attenuated by disease burden adjustments. CONCLUSIONS: This study provides the most comprehensive real-world evaluation of the health impacts of weight change to date. After comorbidity burden and healthcare utilization adjustments, weight loss was associated with an overall reduction in risk of many adverse outcomes.

A Type 2 Diabetes Subtype Responsive to ACCORD Intensive Glycemia Treatment.

OBJECTIVE: Current type 2 diabetes (T2D) management contraindicates intensive glycemia treatment in patients with high cardiovascular disease (CVD) risk and is partially motivated by evidence of harms in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial. Heterogeneity in response to intensive glycemia treatment has been observed, suggesting potential benefit for some individuals. RESEARCH DESIGN AND METHODS: ACCORD was a randomized controlled trial that investigated whether intensively treating glycemia in individuals with T2D would reduce CVD outcomes. Using a novel approach to cluster HbA1c trajectories, we identified groups in the intensive glycemia arm with modified CVD risk. Genome-wide analysis and polygenic score (PS) were developed to predict group membership. Mendelian randomization was performed to infer causality. RESULTS: We identified four clinical groupings in the intensive glycemia arm, and clinical group 4 (C4) displayed fewer CVD (hazard ratio [HR] 0.34; P = 2.01 × 10-3) and microvascular outcomes (HR 0.86; P = 0.015) than those receiving standard treatment. A single-nucleotide polymorphism, rs220721, in MAS1 reached suggestive significance in C4 (P = 4.34 × 10-7). PS predicted C4 with high accuracy (area under the receiver operating characteristic curve 0.98), and this predicted C4 displayed reduced CVD risk with intensive versus standard glycemia treatment (HR 0.53; P = 4.02 × 10-6), but not reduced risk of microvascular outcomes (P < 0.05). Mendelian randomization indicated causality between PS, on-trial HbA1c, and reduction in CVD outcomes (P < 0.05). CONCLUSIONS: We found evidence of a T2D clinical group in ACCORD that benefited from intensive glycemia treatment, and membership in this group could be predicted using genetic variants. This study generates new hypotheses with implications for precision medicine in T2D and represents an important development in this landmark clinical trial warranting further investigation.