Genetic Associations with C-peptide Levels before Type 1 Diabetes Diagnosis in At-Risk Relatives.

OBJECTIVE: We sought to determine whether the type 1 diabetes genetic risk score-2 (T1D-GRS2) and single nucleotide polymorphisms (SNPs) are associated with C-peptide preservation before type 1 diabetes diagnosis. METHODS: We conducted a retrospective analysis of 713 autoantibody-positive participants who developed type 1 diabetes in the TrialNet Pathway to Prevention Study who had T1DExomeChip data. We evaluated the relationships of 16 known SNPs and T1D-GRS2 with area under the curve (AUC) C-peptide levels during oral glucose tolerance tests conducted in the 9 months before diagnosis. RESULTS: Higher T1D-GRS2 was associated with lower C-peptide AUC in the 9 months before diagnosis in univariate (ß=-0.06, P<0.0001) and multivariate (ß=-0.03, P=0.005) analyses. Participants with the JAZF1 rs864745 T allele had lower C-peptide AUC in both univariate (ß=-0.11, P=0.002) and multivariate (ß=-0.06, P=0.018) analyses. CONCLUSIONS: The type 2 diabetes-associated JAZF1 rs864745 T allele and higher T1D-GRS2 are associated with lower C-peptide AUC prior to diagnosis of type 1 diabetes, with implications for the design of prevention trials.

A toxicology study of Csf2ra complementation and pulmonary macrophage transplantation therapy of hereditary PAP in mice.

Pulmonary macrophage transplantation (PMT) is a gene and cell transplantation approach in development as therapy for hereditary pulmonary alveolar proteinosis (hPAP), a surfactant accumulation disorder caused by mutations in CSF2RA/B (and murine homologs). We conducted a toxicology study of PMT of Csf2ra gene-corrected macrophages (mGM-Rα+Mϕs) or saline-control intervention in Csf2raKO or wild-type (WT) mice including single ascending dose and repeat ascending dose studies evaluating safety, tolerability, pharmacokinetics, and pharmacodynamics. Lentiviral-mediated Csf2ra cDNA transfer restored GM-CSF signaling in mGM-Rα+Mϕs. Following PMT, mGM-Rα+Mϕs engrafted, remained within the lungs, and did not undergo uncontrolled proliferation or result in bronchospasm, pulmonary function abnormalities, pulmonary or systemic inflammation, anti-transgene product antibodies, or pulmonary fibrosis. Aggressive male fighting caused a similarly low rate of serious adverse events in saline- and PMT-treated mice. Transient, minor pulmonary neutrophilia and exacerbation of pre-existing hPAP-related lymphocytosis were observed 14 days after PMT of the safety margin dose but not the target dose (5,000,000 or 500,000 mGM-Rα+Mϕs, respectively) and only in Csf2raKO mice but not in WT mice. PMT reduced lung disease severity in Csf2raKO mice. Results indicate PMT of mGM-Rα+Mϕs was safe, well tolerated, and therapeutically efficacious in Csf2raKO mice, and established a no adverse effect level and 10-fold safety margin.

imply: improving cell-type deconvolution accuracy using personalized reference profiles.

Using computational tools, bulk transcriptomics can be deconvoluted to estimate the abundance of constituent cell types. However, existing deconvolution methods are conditioned on the assumption that the whole study population is served by a single reference panel, ignoring person-to-person heterogeneity. Here, we present imply, a novel algorithm to deconvolute cell type proportions using personalized reference panels. Simulation studies demonstrate reduced bias compared with existing methods. Real data analyses on longitudinal consortia show disparities in cell type proportions are associated with several disease phenotypes in Type 1 diabetes and Parkinson's disease. imply is available through the R/Bioconductor package ISLET at https://bioconductor.org/packages/ISLET/ .

A discrete approximation method for modeling interval-censored multistate data.

Many longitudinal studies are designed to monitor participants for major events related to the progression of diseases. Data arising from such longitudinal studies are usually subject to interval censoring since the events are only known to occur between two monitoring visits. In this work, we propose a new method to handle interval-censored multistate data within a proportional hazards model framework where the hazard rate of events is modeled by a nonparametric function of time and the covariates affect the hazard rate proportionally. The main idea of this method is to simplify the likelihood functions of a discrete-time multistate model through an approximation and the application of data augmentation techniques, where the assumed presence of censored information facilitates a simpler parameterization. Then the expectation-maximization algorithm is used to estimate the parameters in the model. The performance of the proposed method is evaluated by numerical studies. Finally, the method is employed to analyze a dataset on tracking the advancement of coronary allograft vasculopathy following heart transplantation.

Intake of B vitamins and the risk of developing islet autoimmunity and type 1 diabetes in the TEDDY study.

PURPOSE: The aim was to study the association between dietary intake of B vitamins in childhood and the risk of islet autoimmunity (IA) and progression to type 1 diabetes (T1D) by the age of 10 years. METHODS: We followed 8500 T1D-susceptible children born in the U.S., Finland, Sweden, and Germany in 2004 -2010 from the Environmental Determinants of Diabetes in the Young (TEDDY) study, which is a prospective observational birth cohort. Dietary intake of seven B vitamins was calculated from foods and dietary supplements based on 24-h recall at 3 months and 3-day food records collected regularly from 6 months to 10 years of age. Cox proportional hazard models were adjusted for energy, HLA-genotype, first-degree relative with T1D, sex, and country. RESULTS: A total of 778 (9.2) children developed at least one autoantibody (any IA), and 335 (3.9%) developed multiple autoantibodies. 280 (3.3%) children had IAA and 319 (3.8%) GADA as the first autoantibody. 344 (44%) children with IA progressed to T1D. We observed that higher intake of niacin was associated with a decreased risk of developing multiple autoantibodies (HR 0.95; 95% CI 0.92, 0.98) per 1 mg/1000 kcal in niacin intake. Higher intake of pyridoxine (HR 0.66; 95% CI 0.46, 0.96) and vitamin B12 (HR 0.87; 95% CI 0.77, 0.97) was associated with a decreased risk of IAA-first autoimmunity. Higher intake of riboflavin (HR 1.38; 95% CI 1.05, 1.80) was associated with an increased risk of GADA-first autoimmunity. There were no associations between any of the B vitamins and the outcomes "any IA" and progression from IA to T1D.  CONCLUSION: In this multinational, prospective birth cohort of children with genetic susceptibility to T1D, we observed some direct and inverse associations between different B vitamins and risk of IA.

Detecting potential outliers in longitudinal data with time-dependent covariates.

BACKGROUND: Outliers can influence regression model parameters and change the direction of the estimated effect, over-estimating or under-estimating the strength of the association between a response variable and an exposure of interest. Identifying visit-level outliers from longitudinal data with continuous time-dependent covariates is important when the distribution of such variable is highly skewed. OBJECTIVES: The primary objective was to identify potential outliers at follow-up visits using interquartile range (IQR) statistic and assess their influence on estimated Cox regression parameters. METHODS: Study was motivated by a large TEDDY dietary longitudinal and time-to-event data with a continuous time-varying vitamin B12 intake as the exposure of interest and development of Islet Autoimmunity (IA) as the response variable. An IQR algorithm was applied to the TEDDY dataset to detect potential outliers at each visit. To assess the impact of detected outliers, data were analyzed using the extended time-dependent Cox model with robust sandwich estimator. Partial residual diagnostic plots were examined for highly influential outliers. RESULTS: Extreme vitamin B12 observations that were cases of IA had a stronger influence on the Cox regression model than non-cases. Identified outliers changed the direction of hazard ratios, standard errors, or the strength of association with the risk of developing IA. CONCLUSION: At the exploratory data analysis stage, the IQR algorithm can be used as a data quality control tool to identify potential outliers at the visit level, which can be further investigated.

Associations of dietary patterns between age 9 and 24 months with risk of celiac disease autoimmunity and celiac disease among children at increased risk.

BACKGROUND: Higher gluten intake in childhood is associated with increased incidence of celiac disease autoimmunity (CDA) and celiac disease. It remains to be studied whether different dietary patterns independent of gluten intake contribute to the incidence. OBJECTIVES: This study aimed to explore associations of dietary patterns by age 2 y with risk of CDA and celiac disease in genetically susceptible children. METHODS: Data was used from 6726 participants at genetic risk of type 1 diabetes and celiac disease enrolled in the observational cohort, The Environmental Determinants of Diabetes in the Young (TEDDY) study. Children were annually screened for tissue transglutaminase autoantibodies (tTGAs) from age 2 y. Principal component analysis extracted dietary patterns, based on intake of 27 food groups assessed by 3-d food records at age 9 to 24 mo. The primary outcome was CDA (i.e., persistently tTGA-positive in at least 2 consecutive samples), and the secondary outcome was celiac disease. During follow-up to mean age 11.0 (standard deviation 3.6) y, 1296 (19.3%) children developed CDA, and 529 (7.9%) were diagnosed with celiac disease. Associations of adherence to dietary patterns (per 5-unit increase) with the study outcomes were estimated by Cox regression models adjusted for risk factors including gluten intake. RESULTS: At age 9 mo, a dietary pattern higher in the food groups vegetable fats and milk was associated with reduced risk of CDA (hazard ratio [HR]: 0.88; 95% confidence interval [CI]: 0.79, 0.98; P = 0.02). At 24 mo, a dietary pattern higher in the food groups wheat, vegetable fats, and juices, and lower in milk, meat, and oats at age 24 mo was associated with increased risk of CDA (HR: 1.18; 95% CI: 1.05, 1.33; P < 0.001) and celiac disease (HR: 1.24; 95% CI: 1.03, 1.50; P = 0.03). CONCLUSIONS: Dietary patterns in early childhood are associated with risk of CDA and celiac disease in genetically predisposed children, independent of gluten intake.

Distinct transcriptomic profiles in children prior to the appearance of type 1 diabetes-linked islet autoantibodies and following enterovirus infection.

Although the genetic basis and pathogenesis of type 1 diabetes have been studied extensively, how host responses to environmental factors might contribute to autoantibody development remains largely unknown. Here, we use longitudinal blood transcriptome sequencing data to characterize host responses in children within 12 months prior to the appearance of type 1 diabetes-linked islet autoantibodies, as well as matched control children. We report that children who present with insulin-specific autoantibodies first have distinct transcriptional profiles from those who develop GADA autoantibodies first. In particular, gene dosage-driven expression of GSTM1 is associated with GADA autoantibody positivity. Moreover, compared with controls, we observe increased monocyte and decreased B cell proportions 9-12 months prior to autoantibody positivity, especially in children who developed antibodies against insulin first. Lastly, we show that control children present transcriptional signatures consistent with robust immune responses to enterovirus infection, whereas children who later developed islet autoimmunity do not. These findings highlight distinct immune-related transcriptomic differences between case and control children prior to case progression to islet autoimmunity and uncover deficient antiviral response in children who later develop islet autoimmunity.

imply: improving cell-type deconvolution accuracy using personalized reference profiles.

Real-world clinical samples are often admixtures of signal mosaics from multiple pure cell types. Using computational tools, bulk transcriptomics can be deconvoluted to solve for the abundance of constituent cell types. However, existing deconvolution methods are conditioned on the assumption that the whole study population is served by a single reference panel, which ignores person-to-person heterogeneity. Here we present imply, a novel algorithm to deconvolute cell type proportions using personalized reference panels. imply can borrow information across repeatedly measured samples for each subject, and obtain precise cell type proportion estimations. Simulation studies demonstrate reduced bias in cell type abundance estimation compared with existing methods. Real data analyses on large longitudinal consortia show more realistic deconvolution results that align with biological facts. Our results suggest that disparities in cell type proportions are associated with several disease phenotypes in type 1 diabetes and Parkinson's disease. Our proposed tool imply is available through the R/Bioconductor package ISLET at https://bioconductor.org/packages/ISLET/.

Type 1 Diabetes Risk Phenotypes Using Cluster Analysis.

Background: Although statistical models for predicting type 1 diabetes risk have been developed, approaches that reveal clinically meaningful clusters in the at-risk population and allow for non-linear relationships between predictors are lacking. We aimed to identify and characterize clusters of islet autoantibody-positive individuals that share similar characteristics and type 1 diabetes risk. Methods: We tested a novel outcome-guided clustering method in initially non-diabetic autoantibody-positive relatives of individuals with type 1 diabetes, using the TrialNet Pathway to Prevention (PTP) study data (n=1127). The outcome of the analysis was time to type 1 diabetes and variables in the model included demographics, genetics, metabolic factors and islet autoantibodies. An independent dataset (Diabetes Prevention Trial of Type 1 Diabetes, DPT-1 study) (n=704) was used for validation. Findings: The analysis revealed 8 clusters with varying type 1 diabetes risks, categorized into three groups. Group A had three clusters with high glucose levels and high risk. Group B included four clusters with elevated autoantibody titers. Group C had three lower-risk clusters with lower autoantibody titers and glucose levels. Within the groups, the clusters exhibit variations in characteristics such as glucose levels, C-peptide levels, age, and genetic risk. A decision rule for assigning individuals to clusters was developed. The validation dataset confirms that the clusters can identify individuals with similar characteristics. Interpretation: Demographic, metabolic, immunological, and genetic markers can be used to identify clusters of distinctive characteristics and different risks of progression to type 1 diabetes among autoantibody-positive individuals with a family history of type 1 diabetes. The results also revealed the heterogeneity in the population and complex interactions between variables.

Hydroxychloroquine in Stage 1 Type 1 Diabetes.

OBJECTIVE: Innate immune responses may be involved in the earliest phases of type 1 diabetes (T1D). RESEARCH DESIGN AND METHODS: To test whether blocking innate immaune cells modulated progression of the disease, we randomly assigned 273 individuals with stage 1 T1D to treatment with hydroxychloroquine (n = 183; 5 mg/kg per day to a maximum of 400 mg) or placebo (n = 90) and assessed whether hydroxychloroquine treatment delayed or prevented progression to stage 2 T1D (i.e., two or more islet autoantibodies with abnormal glucose tolerance). RESULTS: After a median follow-up of 23.3 months, the trial was stopped prematurely by the data safety monitoring board because of futility. There were no safety concerns in the hydroxychloroquine arm, including in annual ophthalmologic examinations. Preplanned secondary analyses showed a transient decrease in the glucose average area under the curve to oral glucose in the hydroxychloroquine-treated arm at month 6 and reduced titers of anti-GAD and anti-insulin autoantibodies and acquisition of positive autoantibodies in the hydroxychloroquine arm (P = 0.032). CONCLUSIONS: We conclude that hydroxychloroquine does not delay progression to stage 2 T1D in individuals with stage 1 disease. Drug treatment reduces the acquisition of additional autoantibodies and the titers of autoantibodies to GAD and insulin.

Dietary Intake and Body Mass Index Influence the Risk of Islet Autoimmunity in Genetically At-Risk Children: A Mediation Analysis Using the TEDDY Cohort.

Background/Objective: Growth and obesity have been associated with increased risk of islet autoimmunity (IA) and progression to type 1 diabetes. We aimed to estimate the effect of energy-yielding macronutrient intake on the development of IA through BMI. Research Design and Methods: Genetically at-risk children (n = 5,084) in Finland, Germany, Sweden, and the USA, who were autoantibody negative at 2 years of age, were followed to the age of 8 years, with anthropometric measurements and 3-day food records collected biannually. Of these, 495 (9.7%) children developed IA. Mediation analysis for time-varying covariates (BMI z-score) and exposure (energy intake) was conducted. Cox proportional hazard method was used in sensitivity analysis. Results: We found an indirect effect of total energy intake (estimates: indirect effect 0.13 [0.05, 0.21]) and energy from protein (estimates: indirect effect 0.06 [0.02, 0.11]), fat (estimates: indirect effect 0.03 [0.01, 0.05]), and carbohydrates (estimates: indirect effect 0.02 [0.00, 0.04]) (kcal/day) on the development of IA. A direct effect was found for protein, expressed both as kcal/day (estimates: direct effect 1.09 [0.35, 1.56]) and energy percentage (estimates: direct effect 72.8 [3.0, 98.0]) and the development of GAD autoantibodies (GADA). In the sensitivity analysis, energy from protein (kcal/day) was associated with increased risk for GADA, hazard ratio 1.24 (95% CI: 1.09, 1.53), p = 0.042. Conclusions: This study confirms that higher total energy intake is associated with higher BMI, which leads to higher risk of the development of IA. A diet with larger proportion of energy from protein has a direct effect on the development of GADA.

HLA Genotype and Probiotics Modify the Association Between Timing of Solid Food Introduction and Islet Autoimmunity in the TEDDY Study.

OBJECTIVE: To study the interaction among HLA genotype, early probiotic exposure, and timing of complementary foods in relation to risk of islet autoimmunity (IA). RESEARCH DESIGN AND METHODS: The Environmental Determinants of Diabetes in the Young (TEDDY) study prospectively follows 8,676 children with increased genetic risk of type 1 diabetes. We used a Cox proportional hazards regression model adjusting for potential confounders to study early feeding and the risk of IA in a sample of 7,770 children. RESULTS: Any solid food introduced early (<6 months) was associated with increased risk of IA if the child had the HLA DR3/4 genotype and no probiotic exposure during the 1st year of life. Rice introduced at 4-5.9 months compared with later in the U.S. was associated with an increased risk of IA. CONCLUSIONS: Timing of solid food introduction, including rice, may be associated with IA in children with the HLA DR3/4 genotype not exposed to probiotics. The microbiome composition under these exposure combinations requires further study.

Gastrointestinal Infections Modulate the Risk for Insulin Autoantibodies as the First-Appearing Autoantibody in the TEDDY Study.

OBJECTIVE: To investigate gastrointestinal infection episodes (GIEs) in relation to the appearance of islet autoantibodies in The Environmental Determinants of Diabetes in the Young (TEDDY) cohort. RESEARCH DESIGN AND METHODS: GIEs on risk of autoantibodies against either insulin (IAA) or GAD (GADA) as the first-appearing autoantibody were assessed in a 10-year follow-up of 7,867 children. Stool virome was characterized in a nested case-control study. RESULTS: GIE reports (odds ratio [OR] 2.17 [95% CI 1.39-3.39]) as well as Norwalk viruses found in stool (OR 5.69 [1.36-23.7]) at <1 year of age were associated with an increased IAA risk at 2-4 years of age. GIEs reported at age 1 to <2 years correlated with a lower risk of IAA up to 10 years of age (OR 0.48 [0.35-0.68]). GIE reports at any other age were associated with an increase in IAA risk (OR 2.04 for IAA when GIE was observed 12-23 months prior [1.41-2.96]). Impacts on GADA risk were limited to GIEs <6 months prior to autoantibody development in children <4 years of age (OR 2.16 [1.54-3.02]). CONCLUSIONS: Bidirectional associations were observed. GIEs were associated with increased IAA risk when reported before 1 year of age or 12-23 months prior to IAA. Norwalk virus was identified as one possible candidate factor. GIEs reported during the 2nd year of life were associated with a decreased IAA risk.

ISLET: individual-specific reference panel recovery improves cell-type-specific inference.

We propose a statistical framework ISLET to infer individual-specific and cell-type-specific transcriptome reference panels. ISLET models the repeatedly measured bulk gene expression data, to optimize the usage of shared information within each subject. ISLET is the first available method to achieve individual-specific reference estimation in repeated samples. Using simulation studies, we show outstanding performance of ISLET in the reference estimation and downstream cell-type-specific differentially expressed genes testing. We apply ISLET to longitudinal transcriptomes profiled from blood samples in a large observational study of young children and confirm the cell-type-specific gene signatures for pancreatic islet autoantibody. ISLET is available at https://bioconductor.org/packages/ISLET .

The evolution of the mitochondrial disease diagnostic odyssey.

BACKGROUND: Mitochondrial diseases often require multiple years and clinicians to diagnose. We lack knowledge of the stages of this diagnostic odyssey, and factors that affect it. Our goals are to report the results of the 2018 Odyssey2 (OD2) survey of patients with a medical diagnosis of mitochondrial disease; and to propose steps to reduce the odyssey going forward, and procedures to evaluate them. METHODS: Data are from the NIH-funded NAMDC-RDCRN-UMDF OD2 survey (N = 215). The main outcomes are Time from symptom Onset to mitochondrial disease Diagnosis (TOD) and Number of Doctors Seen during this diagnostic process (NDOCS). RESULTS: Expert recoding increased analyzable responses by 34% for final mitochondrial diagnosis and 39% for prior non-mitochondrial diagnosis. Only one of 122 patients who initially saw a primary care physician (PCP) received a mitochondrial diagnosis, compared to 26 of 86 (30%) who initially saw a specialist (p < 0.001). Mean TOD overall was 9.9 ± 13.0 years, and mean NDOCS 6.7 ± 5.2. Mitochondrial diagnosis brings extensive benefits through treatment changes and increased membership in and support of advocacy groups. CONCLUSIONS: Because TOD is long and NDOCS high, there is great potential for shortening the mitochondrial odyssey. Although prompt patient contact with primary mitochondrial disease specialists, or early implementation of appropriate tests, may shorten the diagnostic odyssey, specific proposals for improvement require testing and confirmation with adequately complete, unbiased data across all its stages, and appropriate methods. Electronic Health Record (EHRs) may help by accessing diagnostic codes early, but their reliability and diagnostic utility have not been established for this group of diseases.

Possible heterogeneity of initial pancreatic islet beta-cell autoimmunity heralding type 1 diabetes.

The etiology of type 1 diabetes (T1D) foreshadows the pancreatic islet beta-cell autoimmune pathogenesis that heralds the clinical onset of T1D. Standardized and harmonized tests of autoantibodies against insulin (IAA), glutamic acid decarboxylase (GADA), islet antigen-2 (IA-2A), and ZnT8 transporter (ZnT8A) allowed children to be followed from birth until the appearance of a first islet autoantibody. In the Environmental Determinants of Diabetes in the Young (TEDDY) study, a multicenter (Finland, Germany, Sweden, and the United States) observational study, children were identified at birth for the T1D high-risk HLA haploid genotypes DQ2/DQ8, DQ2/DQ2, DQ8/DQ8, and DQ4/DQ8. The TEDDY study was preceded by smaller studies in Finland, Germany, Colorado, Washington, and Sweden. The aims were to follow children at increased genetic risk to identify environmental factors that trigger the first-appearing autoantibody (etiology) and progress to T1D (pathogenesis). The larger TEDDY study found that the incidence rate of the first-appearing autoantibody was split into two patterns. IAA first peaked already during the first year of life and tapered off by 3-4 years of age. GADA first appeared by 2-3 years of age to reach a plateau by about 4 years. Prior to the first-appearing autoantibody, genetic variants were either common or unique to either pattern. A split was also observed in whole blood transcriptomics, metabolomics, dietary factors, and exposures such as gestational life events and early infections associated with prolonged shedding of virus. An innate immune reaction prior to the adaptive response cannot be excluded. Clarifying the mechanisms by which autoimmunity is triggered to either insulin or GAD65 is key to uncovering the etiology of autoimmune T1D.

Physical Activity and the Development of Islet Autoimmunity and Type 1 Diabetes in 5- to 15-Year-Old Children Followed in the TEDDY Study.

OBJECTIVE: This study investigated physical activity and its association with the development of islet autoimmunity and type 1 diabetes in genetically at-risk children aged 5-15 years. RESEARCH DESIGN AND METHODS: As part of the longitudinal Environmental Determinants of Diabetes in the Young (TEDDY) study, annual assessment of activity using accelerometry was conducted from age 5 years. Time-to-event analyses using Cox proportional hazard models were used to assess the association between time spent in moderate to vigorous physical activity per day and the appearance of one or several autoantibodies and progression to type 1 diabetes in three risk groups: 1) 3,869 islet autoantibody (IA)-negative children, of whom 157 became single IA positive; 2) 302 single IA-positive children, of whom 73 became multiple IA positive; and 3) 294 multiple IA-positive children, of whom 148 developed type 1 diabetes. RESULTS: No significant association was found in risk group 1 or risk group 2. A significant association was seen in risk group 3 (hazard ratio 0.920 [95% CI 0.856, 0.988] per 10-min increase; P = 0.021), particularly when glutamate decarboxylase autoantibody was the first autoantibody (hazard ratio 0.883 [95% CI 0.783, 0.996] per 10-min increase; P = 0.043). CONCLUSIONS: More daily minutes spent in moderate to vigorous physical activity was associated with a reduced risk of progression to type 1 diabetes in children aged 5-15 years who had developed multiple IAs.