Abstract
Type 1 diabetes (T1D) is an autoimmune disease in which the insulin-producing β cells of the pancreas are destroyed by T lymphocytes. Recent studies have demonstrated that monitoring for pancreatic islet autoantibodies, combined with genetic risk assessment, can identify most children who will develop T1D when they still have sufficient β cell function to control glucose concentrations without the need for insulin. In addition, there has been recent success in secondary prevention using immunotherapy to delay the progression of preclinical disease, and primary prevention approaches to inhibiting the initiating autoimmune process have entered large-scale clinical trials. By changing the focus of T1D management from late diagnosis and insulin replacement to early diagnosis and β cell preservation, we can anticipate a future without the need for daily insulin injections for children with T1D.
Type 1 diabetes (T1D), which was formerly known as insulin-dependent diabetes mellitus, is an autoimmune disease that irreversibly destroys the insulin-producing β cells in the pancreatic islets of Langerhans. Although autoantibodies to islet cell components are robust markers of the disease process, evidence indicates that the β cell damage is caused by T cell cytotoxicity and cytokine release in concert with disease mechanisms within the β cell itself. Progressive loss of insulin-secretory capacity eventually leads to hyperglycemia, which can develop at any age but has a median age of diagnosis of 12 years. The discovery of insulin in 1921 and its use for the first time as a replacement therapy in 14-year-old Leonard Thompson on Wednesday, 11 January 1922, in Toronto, Canada, was a landmark in medical science. Before this, children with T1D died from insulin deficiency because of uncontrolled fatty acid mobilization and ketone body production resulting in life-threatening acidosis [diabetic ketoacidosis (DKA)]. However, the lengthened survival of people with T1D after the discovery of insulin revealed the problem of long-term complications such as blindness from retinopathy caused by raised glucose concentrations that occurred despite insulin therapy.
The Diabetes Control and Complications Trial (DCCT) in 1993 showed that improving glycemic control could delay long-term complications (1). However, despite major advances in insulin pharmacokinetics and delivery over the past two decades, only a minority of children and adults with T1D can achieve optimal levels of glycemic control in the long term (2–4). One hundred years after the discovery of insulin, an alternative approach has become possible: avoiding the need for insulin by either interrupting the autoimmune disease process at an early (preclinical) stage (secondary prevention) or preventing the onset of autoimmunity in the first place (primary prevention). In this review, we discuss how this might be achieved, particularly in children; the potential advantages that this approach might bring; and the challenges that remain to be overcome.
Delaying the need for insulin therapy
Clinically, T1D typically presents with several weeks of weight loss and polydipsia (extreme thirst) and/or polyuria (excessive urination) caused by hyperglycemia. This was traditionally considered to be the time of disease onset. However, it has since become clear that this represents a late stage of the disease when an estimated 80% or more of β cell function has already been lost. It is now possible to identify T1D at a preclinical phase in which pancreatic β cell function is still sufficient to control blood glucose concentrations without the need for insulin therapy (Fig. 1). Slowing the loss of β cell function delays the need for insulin, and even if insulin is eventually required, this approach still has multiple benefits.
Factors before birth and early life exposure combine with genetic risk, resulting in autoimmunity. Once autoimmunity is clearly established [stage 1, multiple types of islet-specific autoantibody (Ab) detected with normal blood glucose levels (euglycemia)], this represents disease onset with inevitable progression to β cell loss, ultimately affecting the ability to control glucose (stage 2, dysglycemia), and finally levels of glycemia diagnostic of T1D and the need for insulin (stage 3).
First, delaying a diagnosis of clinical T1D extends the period of a life free from the daily burden of continuous monitoring, daily dietary and exercise challenges, multiple insulin injections, and the risk of hypoglycemia, all of which are consequences of insulin therapy and thus do not exist for as long as insulin therapy is not required. This is particularly beneficial to the many individuals who struggle to achieve glycemic control with insulin therapy and are at the highest risk of long-term complications, which incur great cost to the healthcare system. Second, by definition, glucose concentrations during this insulin-free period are below the threshold that contribute to the risk of long-term complications, resulting in prolonged benefits by reducing early exposure to elevated glucose. As demonstrated in the DCCT and its long-term follow-up study [Epidemiology of Diabetes Interventions and Complications Study (EDIC)], improved metabolic control in the early years of T1D reduces complication rates even 30 years later (5). Third, if or when patients eventually progress to requiring insulin, the preservation of even limited residual β cell function has been shown to be associated with less hypoglycemia and improved glucose control [as indicated by lower amounts of glycosylated hemoglobin (HbA1c)] and reduced risk of long-term complications such as retinopathy (6, 7). Finally, cross-sectional analyses indicate that metabolic control improves after the age of 25 years (2), presumably the result of greater maturity and a more regularized lifestyle. Therefore, increasing the age at which insulin is required is likely to contribute substantially to reduced lifetime glycemic exposure.
The identification of patients at a preclinical stage of T1D in itself has benefits, resulting in reduced presentation of DKA by up to 90% in children under the age of 5 years (8, 9). DKA around the time of diagnosis has been associated with neurocognitive deficits (10). In addition, diagnosis at the preclinical stage allows time for families to adjust to the diagnosis and prepare for insulin therapy and glucose monitoring in a calm, outpatient setting in advance of its being required (9).
Staging T1D: Identifying preclinical disease
In the early phases of the autoimmune process, when β cell loss is limited, blood glucose concentrations are normal and cannot be used to diagnose the disease. Therefore, a prerequisite for delaying or preventing the need for insulin therapy is the discovery of biomarkers that can identify early disease in most if not all individuals. As the disease progresses, glucose concentrations begin to rise, but the individual may still be asymptomatic. A combination of biomarkers of the autoimmune process and glucose concentrations can then be used to stage progression of the disease toward insulin dependence.
Autoantibodies as biomarkers of early disease
The identification of strong familial and genetic associations and islet-specific autoantibodies [antibodies to insulin (IAA), glutamate decarboxylase (GADA), islet antigen 2 (IA2A), and islet specific zinc transporter (ZnT8A) (11, 12)] in people with T1D has paved the way for the study of individuals at risk of future T1D. Such natural history studies have included individuals who have a family member with T1D (first-degree relative studies) and/or those recruited from the general population without (13) or with (14) high T1D genetic risk. These studies have revealed that >90% of children with T1D have autoantibodies to at least one islet-specific autoantibody at diagnosis, and these can appear years before clinical diagnosis of T1D. The first autoantibody to appear is against insulin, with a peak incidence around the age of 12 months (14, 15). Up to 90% of children with a single type of islet-specific autoantibody do not progress to T1D, but seroconversion to the presence of two or more autoantibodies (which occurs at a median age of 2.1 years) comes with an 84% risk of clinical T1D by the age of 18 years (16). A disease model of presymptomatic autoimmune β cell destruction identified by the presence of islet-specific autoantibodies offers the possibility of intervening early, before clinical diagnosis of T1D, to maximize the preservation of β cells. The very high risk associated with two or more islet-specific autoantibodies has prompted a movement to define multiple-autoantibody–positive individuals as having preclinical T1D: Stage 1 is when glucose concentrations are normal, with progression to stage 2 when they start to rise (impaired glucose tolerance) and stage 3 when concentrations reach the standard criteria for clinical diagnosis of T1D (17) (Fig. 1).
Advances in islet autoantibody detection are making mass screening more feasible. Reliable testing for two or three autoantibodies (typically, antibodies to GAD, IA2, and ZnT8) can now be performed on as little as 4 μl of blood using advanced luciferase immunoprecipitation system technology (18) or an agglutination polymerase chain reaction (PCR)–based detection system (19). This allows reliable detection of islet autoantibodies from dried blood spots on filter paper or small capillary samples that can be obtained at home or in other community settings and sent by mail to the laboratory. Indeed, PCR-based islet autoantibody testing has recently been made available to the general public in the United States and is being introduced in Europe and Australia. Experience from the TrialNet Pathway to Prevention study, which has screened >200,000 people 2.5 to 45 years of age who are related to a child with T1D, indicated that 3.8% were single-autoantibody positive and 3% were multiple-autoantibody positive. In the general population, rates are about one-tenth of these [0.3% (8)].
Children identified as being single-autoantibody positive and those who are multiple-autoantibody positive and normoglycemic will need to join a monitoring program linked to clinical care providers. Cost-effective and acceptable arrangements for monitoring and for monitoring intervals have yet to be defined. However, recent progress has been made in defining those in whom the disease is progressing more rapidly (20), which will allow alternative therapies to be offered to children who continue to progress after the initial immunointervention, further prolonging the insulin-free period.
Combined genetic and autoantibody screening
The possibility of identifying preclinical T1D has been greatly enhanced with information derived from natural history studies. However, how this information could be applied in a whole population to reliably and cost-effectively detect most individuals before they progress to insulin dependence remains a challenge. Cross-sectional screening for islet autoantibodies in first-degree relatives only identifies 10 to 15% of all T1D cases because most incident cases have no affected first-degree relatives (21). Recent efforts to perform cross-sectional screening in 90,576 children between the ages of 1.75 and 5.99 years has highlighted that a presymptomatic diagnosis of asymptomatic T1D defined by multiple islet autoantibodies is possible (8), and was 96.4% sensitive (proportion of true cases identified as multiple autoantibody positive) and >99% specific (proportion of non-T1D identified by lack of multiple autoantibodies) for T1D presenting during the 3 years after screening. However, it is important to note that this will miss cases presenting before screening and cases that seroconvert after screening and will identify some children who are “at risk” by being single-autoantibody positive but probably will not progress. Recent data from The Environmental Determinants of Diabetes in the Young (TEDDY) study suggest that a single screen for multiple islet autoantibodies between the ages of 3 and 4 years has a nearly 40% sensitivity (with >90% specificity) for T1D presenting before the age of 12, with a risk of T1D within the next 5 years of 50 to 60% (22). A solution to increasing the sensitivity of screening is to include a second autoantibody screen at a later stage. However, this may still only achieve a sensitivity approaching 50% for childhood T1D presenting before the age of 12.
Another option is to consider whether improving characterization of genetic information from birth can make screening more efficient and enable the prediction of T1D in very early life. The identification of newborn babies considered to be at high risk based on known susceptible and protective human leukocyte antigen (HLA) class II haplotypes (the region of the genome with the strongest effect of risk) has allowed natural history studies to follow a subset of a population accounting for ~50% of childhood T1D. Recently, T1D genetic risk scores (GRS), aggregating genetic risk into a continuous risk variable, have increased the sensitivity and specificity of genetic screening. One study focused on recruiting infants with a >10% risk of islet autoimmunity in the first few years of life (~1/1000 infants) for an early life intervention trial, but this strategy misses most childhood cases of T1D (23). A recently improved GRS incorporating more information on HLA class II haplotype interactions was able to stratify nearly 80% of childhood T1D within the top 10% in a population (24). Combining this GRS with family history information and repeated autoantibody testing in this 10% could allow identification of preclinical T1D in most children (25) (Fig. 2).
Combining GRS and autoantibodies can be used to identify preclinical T1D. (A) Neonates with the top 10 to 15% of genetic risk as scored by GRS can be selected. This identifies 85 to 93% of the children diagnosed with T1D under the age of 5 years. (B) Reducing risk with age allows the population followed prospectively to have risk reduced further from 10% to 1% at age 8 to 10 years by successive rounds of autoantibody screening and recalculation of combined risk including GRS. Information is derived from (24) and (25).
We propose that screening and prevention of insulin-requiring T1D in childhood be developed and implemented in two broad phases (Fig. 3). Phase 1 comprises serological screening for islet autoantibodies only. This would first be conducted in preschool children (age 2 to 5 years). Because <15% of children with T1D in this population require insulin, this would need to be repeated at a later age to detect late seroconverters. Phase 2 would commence with GRS estimation at birth to detect early-onset cases (<3 years) and to guide the need for and frequency of islet autoantibody testing (Fig. 3). Once preclinical T1D is identified, secondary prevention (to delay T1D diagnosis) can be undertaken with immunointerventions to reduce autoimmunity while separate efforts are underway for primary prevention to stop the development of autoimmunity in the first place.
Diagnosis and intervention before requiring insulin can be sequentially improved as newer approaches (e.g., GRS calculation, combined interventions, and more specific interventions to slow β cell loss) are introduced. The result is a progressively later age of diabetes onset (secondary prevention), leading to the need for insulin to become rarer in childhood. Ultimately, primary prevention approaches are preferable to avoid the need for ongoing immune interventions. NAS, non–antigen-specific.
Recent advances in secondary prevention of T1D
Evidence that it is possible to delay the diagnosis of T1D through immunointervention has recently been presented in a landmark study. First-degree relatives between 8 and 50 years of age (median, 14 years) from families with T1D received transient T cell modulation with the monoclonal antibody teplizumab (anti-CD3) in the dysglycemic phase before disease diagnosis (stage 2; Fig. 1). This resulted in a delay in the need for insulin treatment for a median of at least 3 years (20, 26). The importance of this observation has been highlighted by teplizumab being granted “breakthrough” status by the US Food and Drug Administration (FDA) in 2019 and “prime” status by the European Medicines Agency (EMA) in 2020. Teplizumab therapy entails 12 to 14 days of intravenous infusions and causes transient T cell depletion. At least part of the mechanism of action involves engagement of the CD3-epsilon chain on the surface of cytotoxic CD8+ T cells, delivering a partially agonistic signal that leads to their nonresponsiveness and conversion to a partial exhaustion phenotype [as indicated by expression of the surface markers eomesodermin (EOMES), T-cell immunoreceptor with immunoglobulin (Ig) and ITIM domains (TIGIT), and killer cell lectin-like receptor subfamily G member 1 (KLRG-1) (27)]. Other studies also suggest that the expansion of regulatory T cell (Treg) populations explains the longer-term persistence of benefit (28, 29). Safety monitoring to date has shown no adverse effects beyond the dosing period.
Results of a prevention study with abatacept, a fusion protein that combines cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) and Ig to inhibit full T cell activation, in first-degree relatives at an earlier stage (stage 1) is enrolling subjects 1 to 45 years of age with two or more islet autoantibodies and normal glucose tolerance. Results are expected to be reported in the next year (www.clinicaltrials.gov identifier NCT01773707). Additional secondary prevention studies are planned, one using anti-thymocyte globulin (ATG) in children from age 6 years at a dose that depletes T effector cells rather than Tregs (www.clinicaltrials.gov identifier NCT04291703) and another that depletes B cells with the CD20 monoclonal antibody rituximab (to reduce antigen presentation) in combination with abatacept to reduce T cell activation in children from age 8 years (www.clinicaltrials.gov identifier NCT03929601).
To date, seven different selective but non–antigen-specific immunointerventions have shown evidence of slowing the autoimmune process, resulting in β cell preservation (improved insulin secretary capacity) in newly diagnosed (stage 3) T1D in at least one phase 2 study (Table 1). Many others are currently being tested, providing additional candidate therapies for use in prevention studies (30). Most recently, results of the tumor necrosis factor (TNF) monoclonal antibody golimumab have been reported (www.clinicaltrials.gov identifier NCT03298542). When given every 2 weeks as a subcutaneous injection to 84 individuals 6 to 21 years of age with newly diagnosed T1D, it showed impressive preservation of β cell function and was well tolerated (31). Anti–interleukin-21 (anti–IL-21) to inhibit trafficking of CD8+ T cells to pancreatic islets in combination with a glucagon like peptide (GLP-1) agonist to promote β cell survival (www.clinicaltrials.gov identifier NCT02443155) has also shown potential for benefit (32). Given the high degree of safety and known genetically validated mechanisms, and because of reports of clinical efficacy in several other autoimmune diseases, ultra-low-dose IL-2 to selectively expand Tregs is also being tested in clinical trials in children newly diagnosed with T1D (stage 3; e.g., www.clinicaltrials.gov identifier NCT03782636).
The agents listed in this table have published evidence of preservation of β cell function from clinical trials in new-onset (stage 3) T1D. See text for mechanisms of action.
An alternative to non–antigen-specific immunotherapies are immunointerventions that restore tolerance in an antigen-specific manner. No such therapies have proven efficacy at this time, but many approaches based on administering β cell–derived antigens or epitopes derived from them (e.g., from proinsulin or GAD) are being explored (33). In general, they have proven to be safe (33–35), and if an effective platform approach can be established, they might be preferred to nonspecific immunotherapies or may be used in combination with other therapies (Fig. 3).
Approaches to primary prevention of T1D
A child’s degree of risk for T1D is likely to begin in utero (Fig. 1). Given that mothers with T1D provide about half the risk of disease in children and that the fathers’ contribution to the risk increases with paternal age, it can be assumed that many of the T1D risk alleles inherited by the child begin their effects from birth onward in a complex interplay with equally numerous environmental factors (11, 12). Many researchers still seek the disease “trigger,” but it is very likely that part of the environmental contribution is a loss of protective factors related to the hygiene hypothesis and industrialization, underpinning the disruption of symbiotic health-promoting intestinal microbiota (dysbiosis) (12). If we could fully understand and reintroduce these protective factors across the general population, then we could potentially halt the increasing incidence of T1D (36).
Because the earliest autoantibody to appear is IAA, oral administration of insulin to promote immune tolerance as early as possible after birth to children with high genetic risk of T1D could be a preventative strategy. A randomized placebo-controlled trial of daily oral insulin versus placebo, the Primary Oral Insulin Trial (POInT; www.clinicaltrials.gov identifier NCT03364868), is under way. Treatment is for 3 years, with the primary endpoint being a 50% reduction in the frequency of two or more anti-islet autoantibodies or progression to T1D for up to 7 years, with results expected in 2025 (37). Pregnant mothers are being recruited, and DNA is obtained from the blood spots from newborn baby’s Guthrie filter card (used in the official neonatal screening program for conditions such as phenylketonuria) or a trial-specific custom filter card. This DNA is genotyped with a GRS based on 47 single-nucleotide polymorphisms associated with T1D risk. If the baby is in the top 25% of the GRS, and therefore at 10% risk of developing two or more autoantibodies or T1D by age 6 years, then he or she is randomized into the trial, starting treatment between 4 and 7 months of age. A total of 1040 children had been randomized by March 2021 after nearly 250,000 babies across five European countries were genotyped. The infrastructure and networks for initial recruitment and enactment of such large-scale primary prevention trials in T1D, including central biobanking of biological samples and databases, is referred to as the Global Platform for the Prevention of Autoimmune Diabetes (GPPAD). GPPAD is currently the only one of its kind but demonstrates the feasibility of such ambitious efforts across multiple countries in their general populations rather than relying on first-degree relatives of T1D patients.
There is a vast literature on the role of the intestinal microbiota in the development of islet autoimmunity and T1D. Overall, microbial dysbiosis and the consequences for immune tolerance, intestinal inflammation, and gut epithelial functions are likely causal factors in T1D, but more research is required. In addition, T1D risk variants alter gut microbial composition and antibodies to commensal bacterial antigens (38). Both breastfeeding (exclusively) and probiotic use provide some protection but only if probiotics in commercially available formulas are used in the first 27 days after birth (39). A randomized controlled trial within GPPAD will test the ability of daily probiotic supplementation versus placebo starting as early as possible, which is ~42 days because of the time required to obtain the GRS and consent from families with babies with the highest GRS (as in POInT). Treatment will be for 12 months to reduce the frequency of two or more autoantibodies or T1D up to age 6.5 years in 1144 randomized children. The trial is using a single-strain probiotic, Bifidobacterium longum subsp. infantis, because it is an efficient metabolizer of the human milk oligosaccharides (HMOs) in breast milk. Therefore, the babies should be breast fed for as long as possible to optimize the beneficial effects of HMOs. HMOs are natural prebiotics, and these and synthetic versions have also been reported to have widespread beneficial effects on gut epithelial functions and the infant immune system (38, 40). However, much more research is needed to confirm these effects on autoimmunity and inflammation.
Coxsackievirus is strongly implicated as being one of the risk cofactors in early T1D development (40). A multistrain vaccine has been developed and needs to be tested in children at risk of T1D (40). The administration of tolerogenic peptides or DNA or RNA vectors that encode T1D autoantigens is also probably part of the future for primary prevention (32), along with other antigen-specific approaches. Childhood obesity also appears to be a contributing factor in T1D (41), possibly by influencing the development of the microbiota dysbiosis. Perhaps early dietary and behavioral prevention studies can be designed to reduce T1D incidence but, again, within a randomized placebo-controlled trial approach to ensure robust findings.
Conclusions
With sufficient investment and commitment, it should be possible with the current state of knowledge to prevent almost all cases of DKA, avoid the requirement for hospital admission at diagnosis, and move the modal age of diagnosis of T1D from age 12 years to around age 15. Further developments, including repeated and sequential interventions, should push the age of diagnosis out further until β cell deficiency that requires insulin therapy becomes a rare occurrence in children under the age of 18. Beyond this, GPPAD and other trials in T1D and trials in atopic or allergic diseases such as dermatitis and peanut allergy are expected to yield insights that will lead to successful primary prevention of T1D and, ultimately, its removal from society.
Acknowledgments: The Juvenile Diabetes Research Foundation (JDRF)/Wellcome Diabetes and Inflammation Laboratory is supported by grants from JDRF (grant no. 4-SRA-2017-473-A-A), the Wellcome Trust (grant no. 107212/A/15/Z), and the Innovative Medicines Initiative (grant nos. 115797 and 945268). The Innovative Medicines Initiative 2 Joint Undertaking [grant no. 115797 (INNODIA) and grant no. 945268 (INNODIA HARVEST)] receives support from the Union’s Horizon 2020 research and innovation program, European Federation of Pharmaceutical Industries and Associations (EFPIA), JDRF, and The Leona M. and Harry B. Helmsley Charitable Trust. The Wellcome Centre for Human Genetics is supported by a Wellcome Core Award (grant no. 203141/Z/16/Z). The UK Type 1 Diabetes Immunotherapy Consortium is supported by grants from Diabetes UK and JDRF. W.H. is affiliated with the University of Washington Diabetes Research Center [funded by National Institutes of Health (NIH) grant no. DK017047] and with the TEDDY Consortium (funded by NIH grant no. DK063829).
"type" - Google News
July 30, 2021 at 12:54AM
https://ift.tt/2TJiScD
Preventing type 1 diabetes in childhood - Science Magazine
"type" - Google News
https://ift.tt/2WhN8Zg
https://ift.tt/2YrjQdq
Bagikan Berita Ini
0 Response to "Preventing type 1 diabetes in childhood - Science Magazine"
Post a Comment