Effect of screening for type 1 diabetes on early metabolic control: the DiPiS study

Predicting the Risk of Developing Type 1 Diabetes Using a One-Week Continuous Glucose Monitoring Home Test With Classification Enhanced by Machine Learning: An Exploratory Study

BACKGROUND

Detection of two or more autoantibodies (Ab) in the blood might describe those individuals at increased risk of developing type 1 diabetes (T1D) during the following years. The aim of this exploratory study is to propose a high versus low T1D risk classifier using machine learning technology based on continuous glucose monitoring (CGM) home data.

METHODS

Forty-two healthy relatives of people with T1D with mean ± SD age of 23.8 ± 10.5 years, HbA1c (glycated hemoglobin) of 5.3% ± 0.3%, and BMI (body mass index) of 23.2 ± 5.2 kg/m2 with zero (low risk; N = 21), and ≥2 (high risk; N = 21) Ab, were enrolled in an NIH (National Institutes of Health)-funded TrialNet ancillary study. Participants wore a CGM for a week and consumed three standardized liquid mixed meals (SLMM) instead of three breakfasts. Glycemic features were extracted from two-hour post-SLMM CGM traces, compared across groups, and used in four supervised machine learning Ab risk status classifiers. Recursive Feature Elimination (RFE) algorithm was used for feature selection; classifiers were evaluated through 10-fold cross-validation, using the receiver operating characteristic area under the curve (AUC-ROC) to select the best classification model.

RESULTS

The percent time of glucose >180 mg/dL (T180), glucose range, and glucose CV (coefficient of variation) were the only significant differences between the glycemic features in the two groups with P values of .040, .035, and .028 respectively. The linear SVM (Support Vector Machine) model with RFE features achieved the best performance of classifying low-risk versus high-risk individuals with AUC-ROC = 0.88.

CONCLUSIONS

A machine learning technology, combining a potentially self-administered one-week CGM home test, has the potential to reliably assess the T1D risk.

Approaches to Measuring Beta Cell Reserve and Defining Partial Clinical Remission in Paediatric Type 1 Diabetes

CONTEXT

Type 1 diabetes (T1D) results from the autoimmune T-cell mediated destruction of pancreatic beta cells leading to insufficient insulin secretion. At the time of diagnosis of T1D, there is residual beta cell function that declines over the subsequent months to years. Recent interventions have been approved to preserve beta cell function in evolving T1D.

OBJECTIVE

The aim of this review is to summarise the approaches used to assess residual beta cell function in evolving T1D, and to highlight potential future directions.

METHODS

Studies including subjects aged 0 to 18 years were included in this review. The following search terms were used; "(type 1 diabetes) and (partial remission)" and "(type 1 diabetes) and (honeymoon)". References of included studies were reviewed to determine if additional relevant studies were eligible.

RESULTS

There are numerous approaches to quantifying beta cell reserve in evolving T1D. These include c-peptide measurement after a mixed meal or glucagon stimuli, fasting c-peptide, the urinary c-peptide/creatinine ratio, insulin dose-adjusted haemoglobin A1c, and other clinical models to estimate beta cell function. Other biomarkers may have a role, including the proinsulin/c-peptide ratio, cytokines, and microRNA. Studies using thresholds to determine if residual beta cell function is present often differ in values used to define remission.

CONCLUSIONS

As interventions are approved to preserve beta cell function, it will become increasingly necessary to quantify residual beta cell function in research and clinical contexts. In this report, we have highlighted the strengths and limitations of the current approaches.

Teplizumab in Type 1 Diabetes Mellitus: An Updated Review

Type 1 diabetes mellitus (T1DM) is a chronic autoimmune condition characterized by the irreversible destruction of the β cells of the pancreas, which leads to a lifelong dependency on exogenous insulin. Despite the advancements in insulin delivery methods, the suboptimal outcomes of these methods have triggered the search for therapies that may prevent or reverse the disease. Given the autoimmune aetiology of T1DM, therapies counteracting the immune-mediated destruction of the β-cells are the obvious target. Although several treatment strategies have been attempted to target cellular, humoral and innate immunity, very few have had a clinically meaningful impact. Of all the available immunomodulatory agents, cluster of differentiation (CD) 3 antibodies have exhibited the most promising preclinical and clinical results. Muromonab-CD3, which also happened to be a murine CD3 antibody, was the first monoclonal antibody approved for clinical use and was primarily indicated for graft rejection. The adverse effects associated with muromonab-CD3 led to its withdrawal. Teplizumab, a newer CD3 antibody, has a better side-effect profile because of its humanized nature and non-Fc-receptor-binding domain. In November 2022, teplizumab became the first immunomodulatory agent to be licensed by the US Food and Drug Administration for delaying the onset of T1DM in high-risk adults and children over 8 years old. The mechanism seems to be enhancing regulatory T-cell activity and promoting immune tolerance. This article reviews the mechanism of action and the clinical trials of teplizumab in individuals with T1DM or at risk of developing the disease.

General population screening for paediatric type 1 diabetes-A qualitative study of UK professional stakeholders

AIMS

Identifying children at risk of type 1 diabetes allows education for symptom recognition and monitoring to reduce the risk of diabetic ketoacidosis at presentation. We aimed to explore stakeholder views towards paediatric general population screening for type 1 diabetes in the United Kingdom (UK).

METHODS

Qualitative interviews were undertaken with 25 stakeholders, including diabetes specialists, policymakers and community stakeholders who could be involved in a future type 1 diabetes screening programme in the UK. A thematic framework analysis was performed using the National Screening Committee's evaluative criteria as the overarching framework.

RESULTS

Diabetic ketoacidosis prevention was felt to be a priority and proposed benefits of screening included education, monitoring and helping the family to better prepare for a future with type 1 diabetes. However, diabetes specialists were cautious about general population screening because of lack of evidence for public acceptability. Concerns were raised about the harms of living with risk, provoking health anxiety and threatening the child's right to an 'open future'. Support systems that met the clinical and psychological needs of the family living with risk were considered essential. Stakeholders were supportive of research into general population screening and acknowledged this would be a priority if an immunoprevention agent were licensed in the UK.

CONCLUSIONS

Although stakeholders suggested the harms of UK paediatric general population screening currently outweigh the benefits, this view would potentially be altered if prevention therapies were licensed. In this case, an evidence-based screening strategy would need to be formulated and public acceptability explored.

Predicting Immunological Risk for Stage 1 and Stage 2 Diabetes Using a 1-Week CGM Home Test, Nocturnal Glucose Increments, and Standardized Liquid Mixed Meal Breakfasts, with Classification Enhanced by Machine Learning

Background: Predicting the risk for type 1 diabetes (T1D) is a significant challenge. We use a 1-week continuous glucose monitoring (CGM) home test to characterize differences in glycemia in at-risk healthy individuals based on autoantibody presence and develop a machine-learning technology for CGM-based islet autoantibody classification. Methods: Sixty healthy relatives of people with T1D with mean ± standard deviation age of 23.7 ± 10.7 years, HbA1c of 5.3% ± 0.3%, and body mass index of 23.8 ± 5.6 kg/m2 with zero (n = 21), one (n = 18), and ≥2 (n = 21) autoantibodies were enrolled in an National Institutes of Health TrialNet ancillary study. Participants wore a CGM for a week and consumed three standardized liquid mixed meals (SLMM) instead of three breakfasts. Glycemic outcomes were computed from weekly, overnight (12:00-06:00), and post-SLMM CGM traces, compared across groups, and used in four supervised machine-learning autoantibody status classifiers. Classifiers were evaluated through 10-fold cross-validation using the receiver operating characteristic area under the curve (AUC-ROC) to select the best classification model. Results: Among all computed glycemia metrics, only three were different across the autoantibodies groups: percent time >180 mg/dL (T180) weekly (P = 0.04), overnight CGM incremental AUC (P = 0.005), and T180 for 75 min post-SLMM CGM traces (P = 0.004). Once overnight and post-SLMM features are incorporated in machine-learning classifiers, a linear support vector machine model achieved the best performance of classifying autoantibody positive versus autoantibody negative participants with AUC-ROC ≥0.81. Conclusion: A new technology combining machine learning with a potentially self-administered 1-week CGM home test can help improve T1D risk detection without the need to visit a hospital or use a medical laboratory. Trial registration: ClinicalTrials.gov registration no. NCT02663661.

Decreased occurrence of ketoacidosis and preservation of beta cell function in relatives screened and monitored for type 1 diabetes in Australia and New Zealand

AIMS

Islet autoantibody screening of infants and young children in the Northern Hemisphere, together with semi-annual metabolic monitoring, is associated with a lower risk of ketoacidosis (DKA) and improved glucose control after diagnosis of clinical (stage 3) type 1 diabetes (T1D). We aimed to determine if similar benefits applied to older Australians and New Zealanders monitored less rigorously.

METHODS

DKA occurrence and metabolic control were compared between T1D relatives screened and monitored for T1D and unscreened individuals diagnosed in the general population, ascertained from the Australasian Diabetes Data Network.

RESULTS

Between 2005 and 2019, 17,105 relatives (mean (SD) age 15.7 (10.8) years; 52% female) were screened for autoantibodies against insulin, glutamic acid decarboxylase, and insulinoma-associated protein 2. Of these, 652 screened positive to a single and 306 to multiple autoantibody specificities, of whom 201 and 215, respectively, underwent metabolic monitoring. Of 178 relatives diagnosed with stage 3 T1D, 9 (5%) had DKA, 7 of whom had not undertaken metabolic monitoring. The frequency of DKA in the general population was 31%. After correction for age, sex and T1D family history, the frequency of DKA in screened relatives was >80% lower than in the general population. HbA1c and insulin requirements following diagnosis were also lower in screened relatives, consistent with greater beta cell reserve.

CONCLUSIONS

T1D autoantibody screening and metabolic monitoring of older children and young adults in Australia and New Zealand, by enabling pre-clinical diagnosis when beta cell reserve is greater, confers protection from DKA. These clinical benefits support ongoing efforts to increase screening activity in the region and should facilitate the application of emerging immunotherapies.

General population screening for childhood type 1 diabetes: is it time for a UK strategy?

Type 1 diabetes (T1D) is a chronic autoimmune disease of childhood affecting 1:500 children aged under 15 years, with around 25% presenting with life-threatening diabetic ketoacidosis (DKA). While first-degree relatives have the highest risk of T1D, more than 85% of children who develop T1D do not have a family history. Despite public health awareness campaigns, DKA rates have not fallen over the last decade. T1D has a long prodrome, and it is now possible to identify children who go on to develop T1D with a high degree of certainty. The reasons for identifying children presymptomatically include prevention of DKA and related morbidities and mortality, reducing the need for hospitalisation, time to provide emotional support and education to ensure a smooth transition to insulin treatment, and opportunities for new treatments to prevent or delay progression. Research studies of population-based screening strategies include using islet autoantibodies alone or in combination with genetic risk factors, both of which can be measured from a capillary sample. If found during screening, the presence of two or more islet autoantibodies has a high positive predictive value for future T1D in childhood (under 18 years), offering an opportunity for DKA prevention. However, a single time-point test will not identify all children who go on to develop T1D, and so combining with genetic risk factors for T1D may be an alternative approach. Here we discuss the pros and cons of T1D screening in the UK, the different strategies available, the knowledge gaps and why a T1D screening strategy is needed.

Competing endogenous RNA network analysis explores the key lncRNAs, miRNAs, and mRNAs in type 1 diabetes

BACKGROUND

Type 1 diabetes (T1D, named insulin-dependent diabetes) has a relatively rapid onset and significantly decreases life expectancy. This study is conducted to reveal the long non-coding RNA (lncRNA)-microRNA (miRNA)-mRNA regulatory axises implicated in T1D.

METHODS

The gene expression profile under GSE55100 (GPL570 and GPL8786 datasets; including 12 T1D samples and 10 normal samples for each dataset) was extracted from Gene Expression Omnibus database. Using limma package, the differentially expressed mRNAs (DE-mRNAs), miRNAs (DE-miRNAs), and lncRNAs (DE-lncRNAs) between T1D and normal samples were analyzed. For the DE-mRNAs, the functional terms were enriched by DAVID tool, and the significant pathways were enriched using gene set enrichment analysis. The interactions among DE-lncRNAs, DE-miRNAs and DE-mRNAs were predicted using mirwalk and starbase. The lncRNA-miRNA-mRNA interaction network analysis was visualized by Cytoscape. The key genes in the interaction network were verified by quantitatively real-time PCR.

RESULTS

In comparison to normal samples, 236 DE-mRNAs, 184 DE-lncRNAs, and 45 DE-miRNAs in T1D samples were identified. For the 236 DE-mRNAs, 16 Gene Ontology (GO)_biological process (BP) terms, four GO_cellular component (CC) terms, and 57 significant pathways were enriched. A network involving 36 DE-mRNAs, 8 DE- lncRNAs, and 15 DE-miRNAs was built, such as TRG-AS1-miR-23b/miR-423-PPM1L and GAS5-miR-320a/miR-23b/miR-423-SERPINA1 regulatory axises. Quantitatively real-time PCR successfully validated the expression levels of TRG-AS1- miR-23b -PPM1L and GAS5-miR-320a- SERPINA1.

CONCLUSION

TRG-AS1-miR-23b-PPM1L and GAS5-miR-320a-SERPINA1 regulatory axises might impact the pathogenesis of T1D.

Au@Ag nanoparticle sensor for sensitive and rapid detection of glucose

A sensitive SERS sensor based on Au@Ag nanoparticles for rapid glucose detection (5 min) via tuning of the plasmonic properties.