ISPAD Clinical Practice Consensus Guidelines 2024: Screening, Staging, and Strategies to Preserve Beta-Cell Function in Children and Adolescents with Type 1 Diabetes

ADA:

American Diabetes Association

AAB:

autoantibodies

BMI:

body mass index

CGM:

continuous glucose monitoring

DIPP:

diabetes prediction and prevention

DKA:

diabetic ketoacidosis

DPTRS:

Diabetes Prevention Trial-Type 1 Risk Score

DSMES:

diabetes self-management education and support

FDA:

Food and Drug Administration

FPG:

fasting plasma glucose

GADA:

glutamic acid decarboxylase autoantibody

GPPAD:

Global Platform for the Prevention of Autoimmune Diabetes

GRS:

genetic risk scores

HbA1c:

glycosylated hemoglobin A1

HLA:

human leukocyte antigen

IAA:

insulin autoantibodies

IA-2A:

insulinoma associated-2 autoantibody

IFG:

impaired fasting glucose

IGT:

impaired glucose tolerance

ISPAD:

International Society for Pediatric and Adolescent Diabetes

JDRF:

Juvenile Diabetes Research Foundation

OGTT:

oral glucose tolerance test

PLS:

progression likelihood score

SMBG:

self-monitoring fingerstick blood glucose

TEDDY:

The Environmental Determinants of Diabetes in the Young

T1D:

Type 1 Diabetes

Type 1 diabetes (T1D) is characterized by four stages based on antibody status and clinical features (Fig. 1):

Stage 1 includes multiple islet autoantibodies (AABs) confirmed in at least 2 samples (using validated assays). Individuals with Stage 1 have normoglycemia and are asymptomatic.

Stage 2 includes multiple islet AABs confirmed on at least 2 samples with elevated fasting glucose or impaired glucose tolerance (IGT) documented by oral glucose tolerance test (OGTT), glycosylated hemoglobin (HbA1c) 5.7–6.5% (39–48 mmol/mol), or ≥10% change in HbA1c. Additional sub-classifications or stages are likely to be adopted as clinicians and researchers seek to describe specific subpopulations. Stage 2a encompasses those with marginally elevated glucose levels. Stage 2b includes those with glucose levels nearing Stage 3 thresholds (see section on OGTT for glycemic thresholds defining stage). While Stages 2a and 2b are formally defined in prior literature, we propose this nomenclature to provide clinicians and researchers with additional descriptive terms for these people.

Stage 3 includes hyperglycemia, meeting American Diabetes Association (ADA) glycemic and clinical diagnostic criteria. People may be symptomatic or asymptomatic. Additional sub-classifications or stages are likely to be adopted as clinicians and researchers seek to describe specific subpopulations. Stage 3a describes those who are asymptomatic but who meet glycemic diagnostic criteria. Stage 3b describes those with classic onset with overt hyperglycemia and symptoms (e.g., polyuria, polydipsia, and unexplained weight loss) and an immediate need for insulin initiation.

Stage 4 includes long-standing T1D. The stages of T1D inform the progression of the condition. Children with a single islet AAB do not have T1D but are considered “at risk” since they carry an approximately 15% risk of developing Stage 3 T1D within 15 years [1]. In contrast, children with 2 confirmed AABs have early-stage T1D. Among children living with Stage 1 (normoglycemia), 44% will progress to Stage 3 T1D in 5 years, and 80 to >90% will progress within 15 years. In children living with Stage 2 T1D (dysglycemia), 75% will progress to Stage 3 T1D in 5 years and nearly 100% during their lifetime [1‒4].

People with a first-degree relative with T1D have up to a 15-fold increased relative lifetime risk of T1D compared to the general population. The prevalence of T1D among people with a first-degree relative is 5% by age 20 compared to 0.3% among the general population [5‒7]. Nevertheless, more than 90% of children diagnosed with T1D do not have a family history of this condition [8, 9]. Those from the general population who go on to develop T1D generally also have an increased genetic risk.

More than 70 genetic T1D variants have been identified through genome-wide association studies [10]. Human leukocyte antigen (HLA) DR and HLA DQ loci confer approximately half of the genetic risk for T1D [11‒13]. The highest-risk HLA haplotypes are DRB1*03:01-DQA1*05:01-DQB1*02:01 (also expressed as DR3-DQ2) and DRB1*04:01-DQA1*03:01-DQB1*03:02 (also expressed as DR4-DQ8). In the general population, children with the HLA DR3-DQ2/DR4-DQ8 genotype have 5% risk for islet autoimmunity and T1D [14‒16]. First-degree relatives of persons already known to have T1D who themselves carry HLA DR3-DQ2/DR4-DQ8 have a further increase in risk that reaches around 20% [15, 17]. Additional risk provided by non-HLA risk genes is roughly equivalent to that provided by HLA DR-DQ alone [16, 18].

The highest non-HLA genetic contribution arises from the INS and PTPN22 genes [19]. These, and other risk regions, are included in (poly)genetic risk scores (GRS) that combine HLA and non-HLA genes to substantially improve risk estimates for islet autoimmunity and T1D, particularly in the general population [16, 20‒22]. With ongoing refinement, GRS continue to see increasing sensitivities (70–80%) and specificities (85–90%) and can be used to identify people with increased risk for T1D [23‒25]. Notably, the risk of developing islet autoimmunity declines exponentially with increasing age. Also, genetic factors are not as predictive of this risk in older children, and there is a paucity of data in adults [26‒28]. Furthermore, once a young person develops multiple islet AABs, HLA, and GRS offer little additional predictive value for stratifying the rate of progression to diabetes [7, 22, 29, 30].

The incidence of T1D continues to increase globally. However, there has been a significant reduction in the proportion of people with the highest-risk HLA haplotypes developing T1D. This observation likely highlights the significant contribution environmental exposures play to the pathogenesis of T1D [31]. Environmental exposures are likely to interact with genes to drive islet autoimmunity and dysglycemia. The effects of nutrition, growth, and intercurrent infections, along with their interactions with biological “omic” systems (i.e., proteome, transcriptome, genome, metabolome, microbiome, virome, and lipidome) have been explored in at-risk birth cohorts [32‒34]. Putative exposures likely vary between people and interact with different gene-environment and environment-environment factors. In addition, environmental exposures may influence the development of insulin autoantibody (IAA) or glutamic decarboxylase (GADA) antibody as the first appearing AAB. Initiating AAB responses may reflect unique T1D endotypes or subtypes defined by distinct pathophysiological mechanisms [35].

Screening for T1D is gaining international momentum. While most initiatives are being performed in the context of research and implementation of science studies, screening for T1D may become standard of care in many parts of the world. Indeed, in 2023 Italy became the first country worldwide to include, by law, a Public Health National Policy that supports screening for T1D and celiac disease in the general pediatric population [36]. Furthermore, in support of screening programs, billing and diagnosis codes for presymptomatic T1D have been developed in both the USA (Effective October 1st, 2024: E10.A0 – type 1 diabetes mellitus, presymptomatic, unspecified; E10.A1 – type 1 diabetes mellitus, presymptomatic, Stage 1; E10.A2 – type 1 diabetes mellitus, presymptomatic, Stage 2) [37] and the UK (Presymptomatic diabetes mellitus type 1 – 1290118005) [38].

The long-term vision for T1D screening programs is to identify people at risk of or with early-stage T1D and offer them preventive approaches capable of delaying or preventing the condition entirely. Currently, achievable benefits driving recommendations for screening include the following:

• 1.

  Prevention of DKA and its associated short- and long-term morbidity and mortality. Rates of DKA at diagnosis of Stage 3 T1D are 15–80% worldwide in the general population [39‒44], whereas screening programs combined with long-term follow-up reduce DKA rates to less than 5% [7, 39, 40, 45, 46]. DKA prevention at diagnosis has potential lifelong benefits, including avoidance of acute morbidity (cerebral edema, shock), neurocognitive impairment, and mortality [47, 48]. There are also non-causal associations between DKA at onset and risk of future DKA episodes [42, 49], severe hypoglycemia [49], and suboptimal long-term glycemic outcomes identified in some [50‒53], but not all studies [54], which, may in turn, increase the risk of future diabetes-related complications [55].

• 2.

  Improving short-term outcomes (symptoms, weight loss, DKA, prolonged hospitalization) [40‒44].

• 3.

  Improving quality of life and reducing psychological burden at the time of T1D diagnosis. Caregiver anxiety and depressive symptoms increase in response to their child’s multiple islet AAB positive test results. Preparation for insulin therapy, education, and psychological support may help reduce caregiver anxiety and smooth the transition to Stage 3 T1D and insulin therapy [7, 47], but more research is needed in these areas.

• 4.

  Providing opportunities for people to participate in research studies. Despite the benefits associated with screening for T1D, potential harms must also be considered. For some people and families, screening leads to increased anxiety and depressive symptoms [48‒50] and many diagnosed with Stages 1, 2a, or 2b T1D have a limited understanding of its progression [50].

Optimal approaches to screening depend on several factors, including local screening objectives, background population risk, the structure of the local healthcare system, and available resources.

The two strategies currently used for T1D screening are as follows:

• 1.

  Genetic risk/family history-based islet AAB screening

• 2.

  Population-wide islet AAB screening

Until recently, most screening programs focused on those with a family history of T1D. While family history-based screening markedly increases per-test probability of identifying people with islet AABs, it fails to identify 90% of those who will ultimately develop T1D. As such, alternative approaches increasingly utilize either general population or genetic risk stratified screening. As the use of GRS continues to scale, thresholds for at-risk populations can be altered to suit the screening purpose [51‒53]. Furthermore, advancements in islet AAB assays permit ultra-low blood volume testing, including the use of capillary samples and dried blood spots, which facilitate minimally invasive collections at home or in community settings [54, 55]. Programs such as the Global Platform for the Prevention of Autoimmune Diabetes (GPPAD), Fr1da, Autoimmunity Screening for Kids (ASK), Population Level Estimate of T1D Risk Genes in Children (PLEDGE), Combined Antibody Screening for Celiac and Diabetes Evaluation (CASCADE), the Australian T1D National Screening Pilot, and the TRIAD study continue to demonstrate the feasibility of general population and genetic risk stratified screening and follow-up programs [56‒58]. Additional studies and analyses are needed to balance sensitivity, specificity, public health priorities, and cost-effectiveness when developing specific screening programs.

Optimal ages for performing AAB screening in the general population continue to be refined using growing data sets from international cohort studies. One analysis suggested that one-time AAB screening performed at 3–5 years of age provided only 35% sensitivity for diagnosing T1D by age 15 years, while sensitivity could be improved to ∼82% with testing at both 2 and 6 years [26, 59]. Alternative models derived from a compilation of prospective cohort studies suggested that the optimal time to identify T1D onset in adolescence (10–18 years of age) is either a single screen at age 10 years (sensitivity 63%) or repeated screening at both ages 10 and 14 years (sensitivity 72%) [60]. Notably, sampling after 2 years of age misses the small but important subset of children who rapidly develop T1D in the first 2 years of life and have the highest rates of DKA [51, 53, 61, 62].

Optimal islet AAB testing frequency in genetically at-risk children remains unclear. Observational studies have used varying frequencies of AAB screening in children with increased genetic risk. In The Environmental Determinants of Diabetes in the Young (TEDDY) study, screening was performed every 3 months through 2 years of life. However, other studies have employed annual AAB testing while still others have performed AAB testing just once between 1 and 5 years of age [63‒66]. More frequent AAB testing (e.g., 6 monthly) may be beneficial in children less than 3 years of age given their more rapid progression to Stage 3 T1D and increased risk of severe DKA.

Once early-stage T1D has been identified, regular glycemic surveillance is recommended to allow T1D staging, inform education and provide opportunities to participate in research or receive T1D modifying therapies [67]. OGTT is the gold standard for staging persons with two or more islet AABs (Fig. 2). However, when OGTT is not feasible, alternative approaches including HbA1c, capillary or venous glucose (2-h postprandial, random or fasting), and CGM can provide important information for people with early-stage T1D, parents, and providers. Home fingerstick glucose measurements and urine test strips can provide real-time data for early detection of hyperglycemia and DKA prevention. Surveillance frequency should depend on the risk of progression, with more frequent monitoring offered to children at high risk of progression e.g., those with dysglycemia in Stage 2, those who seroconvert at a young age, with high insulinoma-associated-2 AAB (IA-2A), or 3–4 islet AABs or other high progression risk metrics [1, 7, 68].

Children with a persistent single islet AAB who spread to multiple antibodies (Stage 1) do so most frequently within 2 years of seroconversion. This spreading is most frequently observed in children under 5 years of age [4, 69]. In single AAB-positive children under 3 years of age, the Juvenile Diabetes Research Foundation (JDRF) guidance advises monitoring AAB status every 6 months for 3 years, then annually for another 3 years [67]. Metabolic monitoring in children positive for a single antibody by annual random venous or capillary BG and HbA1c testing should be considered [67, 70, 71]. As with all screening and monitoring programs, the economic and psychological impacts of repeated screening must always be considered [1, 7].

Glycemic staging and ongoing monitoring should be offered to persons positive for multiple islet AABs [67]. Glycemic monitoring is important for identifying children suitable for early clinical interventions or those seeking to participate in prevention trials. The intensity of those efforts should depend not only on the risk of progression but also on the goals of the family and resource availability. Various monitoring tools are available (Table 1). Participation in prevention programs generally requires OGTT staging (see next section). In other children, less intensive methods may be suitable. All families should be counseled about the expected progression to Stage 3 T1D, how to cope with the diagnosis of early-stage T1D, options for glycemic monitoring, and how to identify signs and symptoms of hyperglycemia [48, 67, 72].

All families should have a team to contact and be given materials for home glucose monitoring. Additional efforts should be made to ensure follow-up of children with multiple AABs as those who do not receive ongoing monitoring and education have high rates of DKA [73].

The standard 2-h OGTT following 1.75 g/kg (75 g maximum) oral glucose administration remains the gold standard test for informing on T1D progression and staging [74‒82]. Consumption of at least 150 g of carbohydrates in the 3 days prior to OGTT is recommended. Fasting, intermediate, and 2-h glucose values defining Stages 1, 2a, 2b, and 3 T1D are provided below (Box 1). Suggested glycemic thresholds for Stage 2a and 2b should be used for descriptive purposes only. When combined with OGTT glucose data, metrics such as age, sex, C-peptide, presence of IA-2A, HbA1c, and BMI, allow for the calculation of scores providing additional information on the speed of progression to Stage 3 T1D. These tools, typically utilized in the research setting, include the 5-timepoint Diabetes Prevention Trial-Type 1 Risk Score (DPTRS) [75, 76], the 2-timepoint DPTRS60 [78], Index60 [79], the single timepoint M120 [80] and the progression likelihood score (PLS) [82]. ADA criteria should continue to be used to document the diagnosis of Stage 3 T1D. Asymptomatic and symptomatic Stage 3 T1Ds are categorized as Stage 3a and Stage 3b, respectively.

While the OGTT remains a gold standard, it is not always feasible or acceptable [83]. Alternative approaches are suggested and discussed below (Box 1).

HbA1c is widely used in clinical practice as an indicator of glycemic outcomes in people living with Stage 3/4 T1D and is generally not affected by short-term variations in food intake and physical activity. In some settings, HbA1c offers a more practical marker of glucose metabolism and T1D staging than the OGTT [75, 76]. Several studies have shown the utility of HbA1c in predicting progression to clinical T1D [84‒86]. HbA1c starts to increase approximately 2 years before a Stage 3 diagnosis, reflecting the gradual deterioration in endogenous insulin secretion and increasing fluctuation in plasma glucose levels. Data from the T1D prediction and prevention (DIPP) study indicated that a 10% rise in HbA1c values taken 3–12 months apart, an additional rise during the subsequent 6 months, and two consecutive values of ≥5.9% predicted progression to Stage 3 T1D in 1 year [84]. The TEDDY study supported these findings, showing that an increase of ≥10% in HbA1c from baseline is as informative as OGTT in predicting the likelihood of developing Stage 3 in young people with genetic risk and islet AABs [85, 86]. Notably, the 2024 ADA Standards of care include HbA1c of 5.7–6.4% (39–47 mmol/mol) or ≥10% increase in HbA1c as diagnostic of Stage 2 T1D [87]. Nonetheless, caution is needed in relying on HbA1c in young children who may progress rapidly, and may be missed before a rise in HbA1c can be observed, or in the setting of an undiagnosed hemoglobinopathy or other conditions that affect hemoglobin turnover [88]. We concur with the JDRF consensus that states that children living with Stage 1 T1D should have HbA1c measured once every 3 months when less than 3 years of age, at least every 6 months when 3–9 years old, and at least every 12 months in children over 9 years old [59].

CGM is increasingly being used as a tool to predict progression to Stage 3 T1D [89‒91], and to detect people who are asymptomatic in Stage 2 and Stage 3a [92]. CGM can provide real-time data and may be useful as it detects increased glucose variability, elevated glucose levels and reduced time in range [89]. Different cut-offs for glucose values have been posited to predict progression to Stage 3 T1D [90, 91]. In one study, a cut-off of 10% time spent above 140 mg/dL (7.8 mmol/L) indicated an 80% risk of progression to Stage 3 T1D over 1 year (88% sensitivity, 91% specificity, 67% positive predictive value, 97% negative predictive value) [90]. While CGM may be a practical alternative to OGTT, controversies still exist in the use of CGM in monitoring early-stage T1D and further evidence is needed to help understand its role, including the use in clinical trials, whether to be used masked or unmasked CGM, its acceptability in this setting, duration and frequency of sensor wear, and its use in guiding when and how to start insulin therapy. Machine learning technologies are a promising and evolving area that may also provide additional insights in the interpretation of CGM data [93].

In the DIPP study, the median time to diagnosis after random plasma glucose ≥7.8 mmol/L (140 mg/dL) was 1 year in children with Stage 1 T1D [81]. Random plasma glucose ≥7.8 mmol/L provided a relatively low sensitivity (21% [95% CI 16%, 27%]) but high specificity (94% [95% CI 91%, 96%]) [81]. Surprisingly, little evidence exists for the accuracy of capillary SMBG in predicting or monitoring Stage 1 or Stage 2 T1D in children. That said, adult data suggest that capillary glucose is a reliable comparator to venous glucose (85–>90% accuracy for diabetes or IGT) during the OGTT [94, 95]. Further evidence is needed to inform optimal frequency and appropriate glucose values for utilizing SMBG in those with early-stage T1D. However, it may be pragmatic to use levels for IFG, IGT, and frank hyperglycemia in this context. As recommended by the JDRF guidance [67] and endorsed by this ISPAD guideline, random venous, or capillary glucose should be measured at the same time as HbA1c in children and young people with early-stage T1D.

When neither venous nor capillary glucose monitoring is available, home urine glucose testing offers a non-invasive and inexpensive way to detect hyperglycemia above the renal threshold. Urine ketone testing may also be available and can be useful in ruling out ketonuria.

In children, adolescents, and young adults identified with islet AABs, ongoing structured individualized education is recommended to improve risk perception and prevent DKA at diagnosis [40‒42]. Education plays a critical role in promoting recommended self-monitoring and in helping families to appreciate opportunities for both clinical and research-based interventions. These favorable outcomes may, in part, be due to the structured and person-centered training of caregivers/families immediately after the results of the screening are communicated. In several studies prospectively following children with Stage 1 and 2 T1D, families are assigned a contact person to answer questions at any time and are provided with guidebooks specifically designed for children [96]. Education not only imparts factual knowledge but also supports caregivers/families in coping with the often-unexpected, elevated risk or diagnosis. The ADA standards for DSMES and the ISPAD guidelines [97, 98] can be used to guide education for people with Stage 1 and 2 T1D. However, age, rate of progression, and family dynamics should inform education topics and determine the intensity of educational interventions. Educational programs need to include strategies for healthy coping, symptoms awareness, plans for glycemic monitoring, consideration of research or treatment opportunities to delay progression, and introduction to insulin therapy [99]. Finally, diabetes education should be targeted, accessible in multiple settings, engaging and person-centered, and should consider the cultural, linguistic, emotional, developmental, and socioeconomic framework of each child and family.

Positive genetic and islet AAB screening results have been associated with parental symptoms of depression and diabetes-specific anxiety [7, 47‒50, 72, 100], particularly in mothers [7, 49, 100]. In programs using genetic screening, the majority of those at high-risk will never develop T1D [16, 20] and anxiety declines among caregivers whose genetically at-risk children remain negative for islet AABs [16, 20, 49]. Among caregivers whose child develops early-stage T1D, diabetes-specific anxiety is common and may remain high, particularly among caregivers of children who are multiple AAB positive [16, 20, 49, 50]. Since the latency period before progression to clinically evident T1D may last for years [74], the unpredictable timing of the child’s diagnosis with clinical T1D only adds to the psychological burden. However, the Fr1da study showed that the severity of depressive symptoms in parents of children diagnosed with Stage 1 or Stage 2 T1D was significantly lower than depressive symptoms reported by parents of children diagnosed with Stage 3 T1D without prior screening participation [7].

Since psychological distress is common among parents and children at the time of Stage 3 T1D diagnosis [101, 102], some have speculated that screening programs identifying children at Stage 1 or Stage 2 T1D, before clinical onset of T1D, might reduce the burden experienced by children and families at the time of clinical diagnosis. One published study addressing this found no differences in anxiety between caregivers of children diagnosed with T1D as part of a screening/monitoring study compared to caregivers of children diagnosed with no prior screening/monitoring experience. However, the caregivers of children identified through the screening study did report lower parenting stress and better quality of life in their children compared to caregivers whose children were diagnosed with no prior screening study experience; the impact on the children themselves was not examined [47].

The nature, degree, and duration of psychological burden in children and caregivers who continue to undergo glycemic surveillance without developing Stage 3 T1D for some years, in those who discontinue follow-up, and in those who develop Stage 3 T1D remain an important understudied area of inquiry. The available literature to date has focused solely on parents of at-risk children; studies are needed to address any possible psychological burden on the children themselves. Additionally, members of certain vulnerable groups – ethnic minorities, those with limited education, or those with a history of depression may respond with greater depressive and anxiety symptoms [48, 50, 72]. Since mental health outcomes in children with Stage 3 T1D are associated with their caregivers’ own mental health and coping with T1D [103, 104], psychosocial support needs to be available for both children diagnosed with Stage 1 or Stage 2 T1D and their caregivers. Since the need for psychological support can change over time and with the child’s developmental level, psychological support needs to be available on an ongoing basis.

A major consideration for wider expansion of screening is the total cost and the incremental cost-effectiveness of screening, education and glycemic surveillance programs [20, 58]. Cost-effectiveness analyses in the USA for islet AAB-only screening suggest it can be cost-effective with a 20% reduction in DKA at diagnosis and a 0.1% (1.1 mmol/mol) reduction in HbA1c over a lifetime [105, 106]. New AAB measurement techniques, such as multiplex electrochemiluminescence assays, needless sample volume and labor time (as compared to radio-binding assays) and thus are more cost-efficient [107].

GRS stratified screening protocols could also improve cost-effectiveness as this approach may identify the small subset of the general population from which the majority of future T1D diagnoses will come [20, 58]. Further economic modeling is required, including assessment of different screening and surveillance models of care in individual countries due to differing health systems, burden of T1D, and local costs of treatment.

In some, but not all lower-resource countries, islet autoimmunity 4 and genetic risk may be more heterogeneous, adding further complexity to screening [108‒111]. Lower-resource countries often have higher rates of DKA and associated mortality; however, the lower T1D incidences may make screening efforts less cost-effective. Priorities in such countries remain on access to and improvements in clinical care for Stage 3 T1D, coupled with correct etiological diagnosis.

The approval of preventive therapies, such as teplizumab, adds significant treatment costs to delaying T1D progression. However, such efforts may result in substantial health benefits that justify their cost-effectiveness [112]. Nevertheless, additional lower-cost options are clearly needed [113].

Efforts to prevent the development of autoimmunity have historically been referred to as primary prevention, while efforts to delay progression from Stage 1 or Stage 2 to Stage 3 T1D are referred to as secondary prevention (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000543035). While a number of immune and metabolic-based therapies have been studied, teplizumab, a monoclonal antibody targeting the T-cell surface marker CD3, is the only therapy that has, to date, been approved by a regulatory agency for use in delaying progression from Stage 2 to Stage 3 T1D [114, 115]. Trials with other drugs targeting (1) autoimmune responses; (2) antigen presentation; (3) glycemic dysregulation; and (4) beta-cell stress/dysfunction, are underway.

Stage 3 interventions or “new onset” studies seek to halt the condition, preserve residual beta-cell function, and potentially delay or prevent complications of T1D in children and adults with newly diagnosed (6–12 weeks) Stage 3 T1D. Numerous efforts have been made to intervene at this relatively late stage due to the ease in identifying people who might still receive benefit [116]. Multiple agents have demonstrated capacity to delay C-peptide decline in Stage 3 T1D, namely, cyclosporine, teplizumab, abatacept, alefacept, rituximab, golimumab, low dose anti-thymocyte globulin, verapamil, imantinib, and baricitinib [117‒122] (online suppl. Table 1).

A growing number of studies continue to focus on Stage 3, where a recent meta-analysis demonstrated a link between maintenance of residual C-peptide and clinical outcomes such as reductions in HbA1c and insulin doses [123]. These studies not only have the prospect of providing direct benefit to people with newly diagnosed T1D but also provide required safety data, particularly in children, where C-peptide decline is faster than in adults, to support moving therapies into Stage 1 or Stage 2 T1D. Based on the existing US approval of teplizumab for intervention in Stage 2 T1D, and the recently published PROTECT study demonstrating efficacy in Stage 3 T1D [124], teplizumab could become the first agent to receive regulatory approval for use in Stage 3 T1D. Moving forward, use of “induction and maintenance” combination therapies, driven by an individual’s stage, genetic risk, and response biomarkers is likely to provide more effective means of preserving beta-cell function in T1D [122].

Notably, clinical trials for Stage 3 T1D have not historically been conducted in low-resource countries. These trials have also enrolled mostly white participants in study sites primarily located in the USA, UK, Europe, and Australia. So far, neither efficacy nor risks have been shown to differ by racial/ethnic background in published Stage 3 trials; however, it is possible such differences could be missed due to the preponderance of white participants [125].

The November 2022 US approval of teplizumab to delay the development of Stage 3 T1D marked a major milestone in the T1D field. Teplizumab is a CD3-directed monoclonal antibody that preserves beta-cell function in people with Stage 3 T1D [124, 126], and delays the onset of Stage 3 T1D in those with Stage 2 T1D [127]. In a phase 2, randomized, placebo-controlled trial of 76 people who were relatives of people with established T1D and had Stage 2 T1D, the median time to onset of clinical T1D was ultimately delayed by about 2.7 years in the teplizumab group with a single 14-days intravenous infusion course compared to the placebo group [115, 126, 128, 129].

Challenges with teplizumab use include its limited indication for Stage 2 T1D (i.e., few people are currently being identified), high cost ($194,000 USD), and logistical difficulties with its 14-day infusion course. A recent Pediatric Endocrine Statement provides an overview of considerations for the use of Teplizumab in clinical practice [130].

It is important to note that the clinical trial that led to the US FDA approval of teplizumab enrolled only 76 relatives of people with T1D who had Stage 2 T1D, nearly all of whom were white, non-Hispanic. Although its effectiveness in delaying T1D clinical diagnosis in those without a family history of T1D and other racial/ethnic groups was not formally studied, the FDA-approved indication encompasses all people (8 years and older) with Stage 2 T1D. While FDA-approved, it is not widely accessible globally, restricting its standard of care status. Nonetheless, where approved, it can be offered to people with Stage 2 T1D [129]. In centers lacking access to infusion dedicated areas on weekends, leveraging home health services to provide infusion over weekends or on days 6–14 could be a potential solution. Additional studies are needed to determine clinical efficacy in a wider population and if subsequent courses of teplizumab or other therapeutics will further delay progression of T1D.

Intervention trials in early-stage T1D should ideally be inclusive for all children and young people across the globe. In addition, with numerous agents showing promise across T1D stages, requests for off-label prescribing are rising. Off-label use of disease-modifying therapy may be considered where systematic monitoring of C-peptide levels, efficacy, and adverse reactions is feasible and can be guided by experienced clinicians [131]. Data on off-label prevention and intervention therapy should ideally be recorded in registries for future analysis.

Screening for early-stage T1D is an important tool for both researchers and clinicians. As evidenced by the success of both family history targeted and general population programs, screening and staging provide important opportunities to reduce DKA, begin education before insulin is required, offer condition-modifying immunotherapies, and encourage participation in studies seeking to delay progression to Stage 3 T1D. In the coming years, general population screening programs will expand and a growing cohort of people with Stage 1 and Stage 2 T1D will be identified. As detailed throughout this guideline, children and young people with early-stage T1D should receive personalized diabetes education, scheduled metabolic assessments, and appropriate psychological support. Finally, with the approval of teplizumab in Stage 2 T1D in the USA, a growing list of agents capable of slowing beta-cell decline, and improving tools to screen and stage T1D, clinical and research programs will continue to rapidly evolve.

A literature search was conducted to gather updated evidence, using a combination of relevant medical subject headings (MeSH, Emtree) and free-text terms specific to each chapter’s focus. Studies published from 2021–2022 onward, related to children and young adults, were retrieved from MEDLINE. The Project Officer, in collaboration with chapter leads and co-authors, performed the literature searches. The resulting articles were then uploaded to COVIDENCE for screening and review. Two authors/experts involved in drafting this guideline version, independently screened the articles. Any disagreements were resolved by a third reviewer. Where relevant, further literature was included.

The draft chapter was posted on the ISPAD forum to allow feedback from the greater ISPAD membership. Modifications were made with authorship consensus, with the chapter receiving endorsement from the ISPAD editorial team.

Recommendations were graded as per the ADA evidence grading system for “Standards of Medical Care in Diabetes” [132]. This hierarchical A-E grading system sets A as having the highest level of evidence, and E having the lowest. Literature search terms are summarized in online supplementary material 1.

M.J.H. – Scientific Advisory Board: SAB BIO; consultant: Sanofi and Mannkind. C.S. – participation on an Immunology Advisory Board for Vertex Pharmaceuticals. R.E.J.B. – independent consultant/advisor to Prevention Bio. A.-G.Z. – Advisory Board: Provention Bio/Sanofi; DMC for Provention Bio/Sanofi, Sanofi, and ITB-MED. K.J.B., K.C., J.C., M.E.C., H.E.L., M.T., L.J., F.U., K.L., T.O., M.L.M., D.K.W., and M.L.M.: no conflicts of interest to declare.

The 2024 Consensus guidelines were supported by unrestricted grants from Abbott Diabetes Care, Dexcom, Medtronic, and Sanofi. These companies did not take part in any aspect of the development of these guidelines.

M.J.H. and M.L.M. co-directed the guideline development process. M.J.H., K.J.B., R.E.J.B., K.C., J.J.C., M.E.C., H.E.L., L.J., K.L., T.O., E.K.S., C.S., M.T., F.U., A.-G.Z., D.K.W., and M.L.M. contributed equally to the content of individual chapters. M.J.H. synthesized these contributions to develop the original full draft of the guideline. Subsequently, K.J.B., R.E.J.B., K.C., J.J.C., M.E.C., H.E.L., L.J., K.L., T.O., E.K.S., C.S., M.T., F.U., A.-G.Z., D.K.W., and M.L.M. participated in revising both the original manuscript and subsequent versions. M.L.M., as the editor of the guideline, supervised the overall development and revision process.