Assisting the implementation of screening for type 1 diabetes by using artificial intelligence on publicly available data.

The type 1 diabetes community is coalescing around the benefits and advantages of early screening for disease risk. To be accepted by healthcare providers, regulatory authorities and payers, screening programmes need to show that the testing variables allow accurate risk prediction and that individualised risk-informed monitoring plans are established, as well as operational feasibility, cost-effectiveness and acceptance at population level. Artificial intelligence (AI) has the potential to contribute to solving these issues, starting with the identification and stratification of at-risk individuals. ASSET (AI for Sustainable Prevention of Autoimmunity in the Society; www.asset.healthcare ) is a public/private consortium that was established to contribute to research around screening for type 1 diabetes and particularly to how AI can drive the implementation of a precision medicine approach to disease prevention. ASSET will additionally focus on issues pertaining to operational implementation of screening. The authors of this article, researchers and clinicians active in the field of type 1 diabetes, met in an open forum to independently debate key issues around screening for type 1 diabetes and to advise ASSET. The potential use of AI in the analysis of longitudinal data from observational cohort studies to inform the design of improved, more individualised screening programmes was also discussed. A key issue was whether AI would allow the research community and industry to capitalise on large publicly available data repositories to design screening programmes that allow the early detection of individuals at high risk and enable clinical evaluation of preventive therapies. Overall, AI has the potential to revolutionise type 1 diabetes screening, in particular to help identify individuals who are at increased risk of disease and aid in the design of appropriate follow-up plans. We hope that this initiative will stimulate further research on this very timely topic.

Major cardiovascular events and subsequent risk of kidney failure with replacement therapy: a CKD Prognosis Consortium study.

AIMS: Chronic kidney disease (CKD) increases risk of cardiovascular disease (CVD). Less is known about how CVD associates with future risk of kidney failure with replacement therapy (KFRT). METHODS AND RESULTS: The study included 25 903 761 individuals from the CKD Prognosis Consortium with known baseline estimated glomerular filtration rate (eGFR) and evaluated the impact of prevalent and incident coronary heart disease (CHD), stroke, heart failure (HF), and atrial fibrillation (AF) events as time-varying exposures on KFRT outcomes. Mean age was 53 (standard deviation 17) years and mean eGFR was 89 mL/min/1.73 m2, 15% had diabetes and 8.4% had urinary albumin-to-creatinine ratio (ACR) available (median 13 mg/g); 9.5% had prevalent CHD, 3.2% prior stroke, 3.3% HF, and 4.4% prior AF. During follow-up, there were 269 142 CHD, 311 021 stroke, 712 556 HF, and 605 596 AF incident events and 101 044 (0.4%) patients experienced KFRT. Both prevalent and incident CVD were associated with subsequent KFRT with adjusted hazard ratios (HRs) of 3.1 [95% confidence interval (CI): 2.9-3.3], 2.0 (1.9-2.1), 4.5 (4.2-4.9), 2.8 (2.7-3.1) after incident CHD, stroke, HF and AF, respectively. HRs were highest in first 3 months post-CVD incidence declining to baseline after 3 years. Incident HF hospitalizations showed the strongest association with KFRT [HR 46 (95% CI: 43-50) within 3 months] after adjustment for other CVD subtype incidence. CONCLUSION: Incident CVD events strongly and independently associate with future KFRT risk, most notably after HF, then CHD, stroke, and AF. Optimal strategies for addressing the dramatic risk of KFRT following CVD events are needed.

Including measures of chronic kidney disease to improve cardiovascular risk prediction by SCORE2 and SCORE2-OP.

AIMS: The 2021 European Society of Cardiology (ESC) guideline on cardiovascular disease (CVD) prevention categorizes moderate and severe chronic kidney disease (CKD) as high and very-high CVD risk status regardless of other factors like age and does not include estimated glomerular filtration rate (eGFR) and albuminuria in its algorithms, systemic coronary risk estimation 2 (SCORE2) and systemic coronary risk estimation 2 in older persons (SCORE2-OP), to predict CVD risk. We developed and validated an 'Add-on' to incorporate CKD measures into these algorithms, using a validated approach. METHODS: In 3,054 840 participants from 34 datasets, we developed three Add-ons [eGFR only, eGFR + urinary albumin-to-creatinine ratio (ACR) (the primary Add-on), and eGFR + dipstick proteinuria] for SCORE2 and SCORE2-OP. We validated C-statistics and net reclassification improvement (NRI), accounting for competing risk of non-CVD death, in 5,997 719 participants from 34 different datasets. RESULTS: In the target population of SCORE2 and SCORE2-OP without diabetes, the CKD Add-on (eGFR only) and CKD Add-on (eGFR + ACR) improved C-statistic by 0.006 (95%CI 0.004-0.008) and 0.016 (0.010-0.023), respectively, for SCORE2 and 0.012 (0.009-0.015) and 0.024 (0.014-0.035), respectively, for SCORE2-OP. Similar results were seen when we included individuals with diabetes and tested the CKD Add-on (eGFR + dipstick). In 57 485 European participants with CKD, SCORE2 or SCORE2-OP with a CKD Add-on showed a significant NRI [e.g. 0.100 (0.062-0.138) for SCORE2] compared to the qualitative approach in the ESC guideline. CONCLUSION: Our Add-ons with CKD measures improved CVD risk prediction beyond SCORE2 and SCORE2-OP. This approach will help clinicians and patients with CKD refine risk prediction and further personalize preventive therapies for CVD.

Phase III, randomised, double-blind, placebo-controlled, multicentre trial to evaluate the efficacy and safety of rhGAD65 to preserve endogenous beta cell function in adolescents and adults with recently diagnosed type 1 diabetes, carrying the genetic HLA DR3-DQ2 haplotype: the DIAGNODE-3 study protocol.

INTRODUCTION: Type 1 diabetes (T1D) is an autoimmune disease leading to the destruction of the insulin-producing beta cells resulting in insulin deficiency and hyperglycaemic. Today, no approved therapy exists to halt this detrimental immunologic process. In a recent phase 2b study, intralymphatic administration of recombinant human glutamic acid decarboxylase 65 kDa (rhGAD65) adsorbed to Alhydrogel adjuvant to individuals recently diagnosed with T1D and carrying the HLA DR3-DQ2 haplotype showed promising results in preserving endogenous insulin secretion, confirming the results of a large meta-analysis of three randomised placebo-controlled trials of subcutaneous rhGAD65. The aim of the current precision medicine phase 3 study is to determine whether intralymphatic administration of rhGAD65 preserves insulin secretion and improves glycaemic control in presumed responder individuals with recently diagnosed T1D carrying HLA DR3-DQ2. METHODS AND ANALYSIS: Individuals ≥12 and <29 years recently diagnosed with T1D (<6 months) will be screened for the HLA DR3-DQ2 haplotype, endogenous insulin production estimated by fasting C-peptide and presence of GAD65 antibodies. 330 patients are planned to be randomised to 3 monthly intralymphatic injections of rhGAD65 or placebo (both accompanied by oral vitamin D supplementation), followed by 22 months of follow-up. The study is powered to detect a treatment effect in the two coprimary endpoints; change from baseline in AUC(0-120min) C-peptide levels during a mixed meal tolerance test, and change from baseline in glycaemic control estimated by haemoglobin A1c at 24 months. Secondary endpoints include effects on glucose patterns collected by masked continuous glucose monitoring, proportion of patients in partial remission and number of episodes of severe hypoglycaemia and/or diabetic ketoacidosis. ETHICS AND DISSEMINATION: The trial is approved by Ethics Committees in Poland (124/2021), the Netherlands (R21.089), Sweden (2021-05063), Czech Republic (EK-1144/21), Germany (2021361) and Spain (21/2021). Results will be published in international peer-reviewed scientific journals and presented at national and international conferences. TRIAL REGISTRATION NUMBER: EudraCT identifier: 2021-002731-32, NCT identifier: NCT05018585.

An online atlas of human plasma metabolite signatures of gut microbiome composition.

Human gut microbiota produce a variety of molecules, some of which enter the bloodstream and impact health. Conversely, dietary or pharmacological compounds may affect the microbiota before entering the circulation. Characterization of these interactions is an important step towards understanding the effects of the gut microbiota on health. In this cross-sectional study, we used deep metagenomic sequencing and ultra-high-performance liquid chromatography linked to mass spectrometry for a detailed characterization of the gut microbiota and plasma metabolome, respectively, of 8583 participants invited at age 50 to 64 from the population-based Swedish CArdioPulmonary bioImage Study. Here, we find that the gut microbiota explain up to 58% of the variance of individual plasma metabolites and we present 997 associations between alpha diversity and plasma metabolites and 546,819 associations between specific gut metagenomic species and plasma metabolites in an online atlas ( https://gutsyatlas.serve.scilifelab.se/ ). We exemplify the potential of this resource by presenting novel associations between dietary factors and oral medication with the gut microbiome, and microbial species strongly associated with the uremic toxin p-cresol sulfate. This resource can be used as the basis for targeted studies of perturbation of specific metabolites and for identification of candidate plasma biomarkers of gut microbiota composition.

Intralymphatic GAD-Alum (Diamyd®) Improves Glycemic Control in Type 1 Diabetes With HLA DR3-DQ2.

AIMS: Residual beta cell function in type 1 diabetes (T1D) is associated with lower risk of complications. Autoantigen therapy with GAD-alum (Diamyd) given in 3 intralymphatic injections with oral vitamin D has shown promising results in persons with T1D carrying the human leukocyte antigen (HLA) DR3-DQ2 haplotype in the phase 2b trial DIAGNODE-2. We aimed to explore the efficacy of intralymphatic GAD-alum on blood glucose recorded by continuous glucose monitoring (CGM). METHODS: DIAGNODE-2 (NCT03345004) was a multicenter, randomized, placebo-controlled, double-blind trial of 109 recent-onset T1D patients aged 12 to 24 years with GAD65 antibodies and fasting C-peptide > 0.12 nmol/L, which randomized patients to 3 intralymphatic injections of 4 µg GAD-alum and oral vitamin D, or placebo. We report results for exploratory endpoints assessed by 14-day CGM at months 0, 6, and 15. Treatment arms were compared by mixed-effects models for repeated measures adjusting for baseline values. RESULTS: We included 98 patients with CGM recordings of sufficient quality (DR3-DQ2-positive patients: 27 GAD-alum-treated and 15 placebo-treated). In DR3-DQ2-positive patients, percent of time in range (TIR, 3.9-10 mmol/L) declined less between baseline and month 15 in GAD-alum-treated compared with placebo-treated patients (-5.1% and -16.7%, respectively; P = 0.0075), with reduced time > 13.9 mmol/L (P = 0.0036), and significant benefits on the glucose management indicator (P = 0.0025). No differences were detected for hypoglycemia. GAD-alum compared to placebo lowered the increase in glycemic variability (standard deviation) observed in both groups (P = 0.0219). Change in C-peptide was correlated with the change in TIR. CONCLUSIONS: Intralymphatic GAD-alum improves glycemic control in recently diagnosed T1D patients carrying HLA DR3-DQ2.

The association between short-term, chronic localized and chronic widespread pain and risk for cardiovascular disease in the UK Biobank.

AIMS: The interplay between pain of different chronicity and cardiovascular disease (CVD) is incompletely understood. We aimed to investigate the association between different levels of chronic or nonchronic pain and risk of CVD. METHODS AND RESULTS: Participants in the UK Biobank who reported pain at baseline were divided into three groups according to pain duration and widespreadness. Participants reporting no pain were controls. Multivariable Cox regression was used to investigate the association between pain and incidence of myocardial infarction, heart failure, stroke, cardiovascular mortality, and composite CVD (defined as any of the before-mentioned cardiovascular events). Of 475 171 participants, 189 289 reported no pain, 87 830 reported short-term pain, 191 716 chronic localized pain, and 6336 chronic widespread pain (CWP). During a median of 7.0 years' follow-up, participants with chronic localized pain and CWP had, after adjustment for age, sex, established cardiovascular risk factors, physical activity, anxiety, depression, cancer, chronic inflammatory/painful disease, pain/anti-inflammatory medication, socioeconomic status, a significantly increased risk for composite CVD [hazard ratio (HR) 1.14, confidence interval (CI) 1.08-1.21, P < 0.001; and HR 1.48, CI 1.28-1.73, P < 0.001, respectively] compared with controls, with similar results when using the different specific CVDs as outcomes. Population attributable risk proportion for chronic pain as a risk factor for composite CVD was comparable with that of diabetes (8.6 vs. 7.3%, respectively). CONCLUSION: Chronic pain is associated with an increased risk for myocardial infarction, stroke, heart failure, and cardiovascular death independent of established cardiovascular risk factors, socioeconomic factors, comorbidities and medication. Our study, the largest to date, confirms and extends our understanding of chronic pain as an underestimated cardiovascular risk factor with important public health implications.

Albuminuria Testing in Hypertension and Diabetes: An Individual-Participant Data Meta-Analysis in a Global Consortium.

[Figure: see text].

Association between Cardiorespiratory Fitness and Circulating Proteins in 50-Year-Old Swedish Men and Women: a Cross-Sectional Study.

BACKGROUND AND AIMS: A strong cardiorespiratory fitness is suggested to have beneficial effects on cardiovascular risk; the exact mechanisms underlying the cardioprotective effects of fitness remain uncertain. Our aim was to investigate associations between cardiorespiratory fitness and multiple plasma proteins, in order to obtain insights about physiological pathways associated with the effects of exercise on cardiovascular health. METHODS: In the Prospective investigation of Obesity, Energy and Metabolism (POEM) study (n=444 adults aged 50 years, 50% women), cardiorespiratory fitness was measured by a maximal exercise test on bicycle ergometer with gas exchange (VO2peak) normalized for body lean mass (dual-energy X-ray absorptiometry (DXA)). We measured 82 cardiovascular proteins associated with cardiovascular pathology and inflammation in plasma samples with a proximity extension assay. RESULTS: In sex-adjusted linear regression, VO2peak was associated with 18 proteins after Bonferroni correction for multiple testing (p<0.0006). Following additional adjustment for fat mass (DXA), fasting glucose (mmol/L), low-density lipoprotein (LDL, mmol/L), smoking status, waist/hip ratio, blood pressure (mmHg), education level, and lpnr (lab sequence number), higher VO2peak was significantly associated with lower levels of 6 proteins: fatty-acid binding protein-4 (FABP4), interleukin-6 (IL-6), leptin, cystatin-B (CSTB), interleukin-1 receptor antagonist (IL-1RA), and growth differentiation factor 15 (GDF-15), and higher levels of 3 proteins: galanin, kallikrein-6 (KLK6), and heparin-binding EGF-like growth factor (HB-EGF), at nominal p-values (p<0.05). CONCLUSIONS: We identified multiple novel associations between cardiorespiratory fitness and plasma proteins involved in several atherosclerotic processes and key cellular mechanisms such as inflammation, energy homeostasis, and protease activity, which shed new light on how exercise asserts its beneficial effects on cardiovascular health. Our findings encourage additional studies in order to understand the underlying causal mechanisms for these associations.

Plasma proteomics and lung function in four community-based cohorts.

BACKGROUND: Underlying mechanism leading to impaired lung function are incompletely understood. OBJECTIVES: To investigate whether protein profiling can provide novel insights into mechanisms leading to impaired lung function. METHODS: We used four community-based studies (n = 2552) to investigate associations between 79 cardiovascular/inflammatory proteins and forced expiratory volume in 1 s percent predicted (FEV1%) assessed by spirometry. We divided the cohorts into discovery and replication samples and used risk factor-adjusted linear regression corrected for multiple comparison (false discovery rate of 5%). We performed Mendelian randomization analyses using genetic and spirometry data from the UK Biobank (n = 421,986) to assess causality. MEASUREMENTS AND MAIN RESULTS: In cross-sectional analysis, 22 proteins were associated with lower FEV1% in both the discovery and replication sample, regardless of stratification by smoking status. The combined proteomic data cumulatively explained 5% of the variation in FEV1%. In longitudinal analyses (n = 681), higher plasma levels of growth differentiation factor 15 (GDF-15) and interleukin 6 (IL-6) predicted a more rapid 5-year decline in lung function (change in FEV1% per standard deviation of protein level -1.4, (95% CI, -2.5 to -0.3) for GDF-15, and -0.8, (95% CI, -1.5 to -0.2) for IL-6. Mendelian randomization analysis in UK-biobank provided support for a causal effect of increased GDF-15 levels and reduced FEV1%. CONCLUSIONS: Our combined approach identified GDF-15 as a potential causal factor in the development of impaired lung function in the general population. These findings encourage additional studies evaluating the role of GDF-15 as a causal factor for impaired lung function.

Incorporating kidney disease measures into cardiovascular risk prediction: Development and validation in 9 million adults from 72 datasets.

BACKGROUND: Chronic kidney disease (CKD) measures (estimated glomerular filtration rate [eGFR] and albuminuria) are frequently assessed in clinical practice and improve the prediction of incident cardiovascular disease (CVD), yet most major clinical guidelines do not have a standardized approach for incorporating these measures into CVD risk prediction. "CKD Patch" is a validated method to calibrate and improve the predicted risk from established equations according to CKD measures. METHODS: Utilizing data from 4,143,535 adults from 35 datasets, we developed several "CKD Patches" incorporating eGFR and albuminuria, to enhance prediction of risk of atherosclerotic CVD (ASCVD) by the Pooled Cohort Equation (PCE) and CVD mortality by Systematic COronary Risk Evaluation (SCORE). The risk enhancement by CKD Patch was determined by the deviation between individual CKD measures and the values expected from their traditional CVD risk factors and the hazard ratios for eGFR and albuminuria. We then validated this approach among 4,932,824 adults from 37 independent datasets, comparing the original PCE and SCORE equations (recalibrated in each dataset) to those with addition of CKD Patch. FINDINGS: We confirmed the prediction improvement with the CKD Patch for CVD mortality beyond SCORE and ASCVD beyond PCE in validation datasets (Δc-statistic 0.027 [95% CI 0.018-0.036] and 0.010 [0.007-0.013] and categorical net reclassification improvement 0.080 [0.032-0.127] and 0.056 [0.044-0.067], respectively). The median (IQI) of the ratio of predicted risk for CVD mortality with CKD Patch vs. the original prediction with SCORE was 2.64 (1.89-3.40) in very high-risk CKD (e.g., eGFR 30-44 ml/min/1.73m2 with albuminuria ≥30 mg/g), 1.86 (1.48-2.44) in high-risk CKD (e.g., eGFR 45-59 ml/min/1.73m2 with albuminuria 30-299 mg/g), and 1.37 (1.14-1.69) in moderate risk CKD (e.g., eGFR 60-89 ml/min/1.73m2 with albuminuria 30-299 mg/g), indicating considerable risk underestimation in CKD with SCORE. The corresponding estimates for ASCVD with PCE were 1.55 (1.37-1.81), 1.24 (1.10-1.54), and 1.21 (0.98-1.46). INTERPRETATION: The "CKD Patch" can be used to quantitatively enhance ASCVD and CVD mortality risk prediction equations recommended in major US and European guidelines according to CKD measures, when available. FUNDING: US National Kidney Foundation and the NIDDK.

Non-targeted urine metabolomics and associations with prevalent and incident type 2 diabetes.

Better risk prediction and new molecular targets are key priorities in type 2 diabetes (T2D) research. Little is known about the role of the urine metabolome in predicting the risk of T2D. We aimed to use non-targeted urine metabolomics to discover biomarkers and improve risk prediction for T2D. Urine samples from two community cohorts of 1,424 adults were analyzed by ultra-performance liquid chromatography/mass spectrometry (UPLC-MS). In a discovery/replication design, three out of 62 annotated metabolites were associated with prevalent T2D, notably lower urine levels of 3-hydroxyundecanoyl-carnitine. In participants without diabetes at baseline, LASSO regression in the training set selected six metabolites that improved prediction of T2D beyond established risk factors risk over up to 12 years' follow-up in the test sample, from C-statistic 0.866 to 0.892. Our results in one of the largest non-targeted urinary metabolomics study to date demonstrate the role of the urine metabolome in identifying at-risk persons for T2D and suggest urine 3-hydroxyundecanoyl-carnitine as a biomarker candidate.

Kidney Disease Biomarkers Improve Heart Failure Risk Prediction in the General Population.

BACKGROUND: The kidneys play an important role in heart failure (HF), but it is unclear if renal biomarkers improve HF risk prediction beyond established risk factors. We aimed to assess whether adding biomarkers of kidney disease to conventional risk factors improved 10-year risk prediction for incident HF in a contemporary community sample. METHODS: We included 450 212 participants in the UK Biobank aged 39 to 70 years without HF who had been assessed in 2006 to 2010 with the urine albumin-to-creatinine ratio and estimated glomerular filtration rate (eGFR) based on serum creatinine and cystatin C. There were 1701 incident cases of HF during up to 10.3 years of follow-up (mean 8.2±0.7 years). We used the Atherosclerosis Risk in Communities study heart failure risk score excluding natriuretic peptides as the base model to which we added eGFR and urine albumin-to-creatinine ratio. Harrell's C-statistic of ARIC-HF was 0.845 (95% CI, 0.831-0.859). RESULTS: Each combination of added kidney measures (creat-eGFR, cysC-eGFR, and urine albumin-to-creatinine ratio) led to significant improvement in risk discrimination, calibration, and reclassification. The optimal pair of added kidney measures was cysC-eGFR and urine albumin-to-creatinine ratio (ΔC=0.019 [95% CI, 0.015-0.022]). Addition of cysC-eGFR made the largest contribution to reclassification improvement (continuous net reclassification improvement 0.323 [95% CI, 0.278-0.360]). CONCLUSIONS: In a large community sample, the addition of kidney disease markers to conventional risk factors improved prediction of 10-year HF risk. Our results support including kidney disease markers in the identification of persons at highest risk of HF and demonstrate a possible role of impaired kidney function in HF development in asymptomatic persons.

The association between plasma proteomics and incident cardiovascular disease identifies MMP-12 as a promising cardiovascular risk marker in patients with chronic kidney disease

BACKGROUND AND AIMS: Previous proteomics efforts in patients with chronic kidney disease (CKD) have predominantly evaluated urinary protein levels. Therefore, our aim was to investigate the association between plasma levels of 80 cardiovascular disease-related proteins and the risk of major adverse cardiovascular events (MACE) in patients with CKD. METHODS: Individuals with CKD stages 3-5 (eGFR below 60 ml min-1 [1.73 m]-2) from three community-based cohorts (PIVUS, ULSAM, SAVA), one diabetes cohort (CARDIPP) and one cohort with peripheral artery disease patients (PADVA) with information on 80 plasma protein biomarkers, assessed with a proximity extension assay, and follow-up data on incident MACE, were used as discovery sample. To validate findings and to asses generalizability to patients with CKD in clinical practice, an outpatient CKD-cohort (Malnutrition, Inflammation and Vascular Calcification (MIVC)) was used as replication sample. RESULTS: In the discovery sample (total n = 1316), 249 individuals experienced MACE during 7.0 ± 2.9 years (range 0.005-12.9) of follow-up, and in the replication sample, 71 MACE events in 283 individuals over a mean ± SD change of 2.9 ± 1.2 years (range 0.1-4.0) were documented. Applying Bonferroni correction, 18 proteins were significantly associated with risk of MACE in the discovery cohort, adjusting for age and sex in order of significance, GDF-15, FGF-23, REN, FABP4, IL6, TNF-R1, AGRP, MMP-12, AM, KIM-1, TRAILR2, TNFR2, CTSL1, CSF1, PlGF, CA-125, CCL20 and PAR-1 (p < 0.000625 for all). Only matrix metalloproteinase 12 (MMP-12) was significantly associated with an increased risk of MACE in the replication sample (hazard ratio (HR) per SD increase, 1.36, 95% CI (1.07-1.75), p = 0.013). CONCLUSIONS: Our proteomics analyses identified plasma MMP-12 as a promising cardiovascular risk marker in patients with CKD.