INTRODUCTION — The increase in understanding of the pathogenesis of type 1 diabetes mellitus (formerly known as insulin-dependent diabetes mellitus) has made it possible to consider intervention to slow the autoimmune disease process in an attempt to delay or even prevent the onset of hyperglycemia. This topic will review current and planned efforts to prevent type 1 diabetes. Subjects who are at high risk for type 1 diabetes can be identified using a combination of immune, genetic, and metabolic markers. (See "Prediction of type 1 diabetes mellitus".)
In the current classification of diabetes, immune-mediated type 1 diabetes is called type 1A to distinguish it from less common cases in which an autoimmune etiology cannot be determined (type 1B); the latter are said to be idiopathic [1]. The term type 1 diabetes used here refers to type 1A, or autoimmune diabetes. (See "Classification of diabetes mellitus and genetic diabetic syndromes".)
PREVENTION AND REVERSAL STRATEGIES — Several immunosuppressive and immunomodulatory agents and other drugs have been given either alone or in combination to decrease the immune-mediated destruction of beta cells that occurs in type 1 diabetes [2]. Most of the studies have been performed in recent-onset diabetes, where the majority of beta-cell function has already been lost, and the anticipated outcome is preservation of remaining beta-cell function, usually measured as area under the curve (AUC) insulin secretion in response to stimulation. Many of the studies involve small numbers of patients and are uncontrolled.
Immunomodulators
Azathioprine — Azathioprine is an immunosuppressive drug that inhibits or prevents T cell responses to antigens. In one randomized, double-blind study of 46 patients treated with azathioprine and glucocorticoids, insulin could be discontinued in 10 of 20 treated patients as compared with 2 of 20 patients in the placebo group [3]. Endogenous insulin secretion (measured as the plasma C-peptide response to a liquid meal) also improved. However, only three treated patients remained in remission at one year. Equally discouraging results were noted in a second study [4].
Mycophenolate mofetil — Mycophenolate mofetil (MMF) inhibits proliferation of both T- and B-lymphocytes. In a multicenter, randomized trial, 126 patients with type 1 diabetes for less than three months were randomly assigned to MMF, MMF plus daclizumab (an anti-interleukin [IL]-2 receptor monoclonal antibody that selectively binds the IL-2 receptor, inhibiting IL-2-mediated T-lymphocyte proliferation), or placebo [5]. After two years, there was no significant difference in the mean AUC for C-peptide levels during a mixed-meal tolerance test. Thus, neither MMF alone nor in combination with daclizumab slowed progression of beta-cell destruction in recently diagnosed type 1 diabetes.
Cyclosporine — Large-scale trials in patients with recently diagnosed type 1 diabetes in Canada and France showed that remissions were twice as common in the cyclosporine-treated patients as compared with placebo [6]. Although the remissions also lasted longer, almost all patients required insulin again within three years. An analysis of subjects recruited into the Canadian-European cyclosporine trial showed that subjects who were insulinoma-associated 2 molecule (IA-2) negative at entry to the trial were more likely to respond to cyclosporine (with a decrease in insulin requirements and an increase in C-peptide secretion) than IA-2-positive subjects [7].
Anti-CD3 antibodies — Treatment of mice with an anti-CD3 monoclonal antibody (OKT3) reverses diabetes in nonobese diabetic (NOD) mice (a model in which spontaneous autoimmunity and pancreatic islet destruction occur) [8]. However, OKT3 use is problematic in humans because of significant cytokine-mediated (tumor necrosis factor-alpha [TNFa]) side effects. Humanized monoclonal antibodies have been developed, which appear to have fewer major adverse effects (fever, headache, hypotension). These antibodies have been successfully used for the treatment of acute renal allograft rejection [9] and psoriatic arthritis [10].
The mechanism of action of this drug is not completely clear. Alterations in the number and function of regulatory T cells may contribute to the generation of an autoimmune state in type 1 diabetes. Dysfunction and/or loss of CD1-restricted T cells, T cells with g/d receptors, CD4+/CD25+ T cells, and NKT cells may all theoretically contribute to disease pathogenesis through inefficient suppression of pathogenic autoreactive T cells. The antigens that activate regulatory T cells are unknown, and the mechanisms by which these cells exert their effect on immune responses remain unclear. It has been postulated that hOKT3g may affect the dynamics of regulatory T cell populations, such as selective deletion of activated Th1 cells and/or activation of Th2 cells and their protective cytokines.
Teplizumab — A monoclonal antibody, termed hOKT3gl (Ala-Ala) (teplizumab), contains the binding region of OKT3 in which the CH2 region has been modified by site-directed mutagenesis to alter FcR-binding activity. Teplizumab has been studied in patients with recently diagnosed type 1 diabetes as well as in individuals at high risk for developing type 1 diabetes. It has been shown to delay clinical progression of type 1 diabetes, but with recently diagnosed type 1 diabetes, the improvement is modest and appears to be transient. Adverse effects include transient lymphopenia, rash, anemia, and fever.
●Recently diagnosed type 1 diabetes – In a two-year trial, 516 patients (aged 8 to 35 years) with type 1 diabetes duration ≤12 weeks were randomly assigned to receive one of three regimens of teplizumab (14-day full dose, 14-day low dose, or 6-day full dose) or placebo at baseline and at six months [11]. After one year, there was no difference in the primary composite outcome (percentage of patients with insulin use <0.5 units/kg/day and A1C <6.5 percent) among the groups (19.8, 13.7, 20.8, and 20.4 percent, respectively). Changes in AUC for C-peptide from baseline (a secondary outcome) were similar among the four groups. After two years, the 14-day, full dose regimen significantly reduced the loss of C-peptide (mean AUC) compared with placebo [12], but A1C was no different and the composite primary outcome was not significantly different between the active and placebo treatment groups. In predefined and post hoc subgroup analyses, C-peptide preservation was better in actively treated patients who were 8 to 17 years of age; residing in the United States; who were randomized within six weeks of diagnosis; and who had C-peptide mean AUC >0.2 nmol/L, A1C <7.5 percent, and insulin dose <0.4 units/kg/day. These findings suggest that teplizumab is most effective in preserving C-peptide secretion when administered early in the course of the disease.
In an unblinded, randomized trial targeting this patient population, 83 patients (mean age 12.5 years) diagnosed with type 1 diabetes within eight weeks of enrollment were randomly assigned to teplizumab (14-day course at baseline and one year) or no therapy [13]. After two years, preservation of C-peptide was significantly better in the teplizumab group. There was no significant difference in A1C levels, but teplizumab-treated patients required less insulin to achieve similar glycemic control. Response to teplizumab varied, and patients with better metabolic control at baseline (lower A1C levels and insulin dose at baseline) had the best response, suggesting that metabolic features may also help identify a subgroup of patients likely to respond to teplizumab.
●High risk for developing type 1 diabetes – In a trial with median follow-up of two years, a single 14-day course of teplizumab was compared with placebo in 76 individuals (median age 13 to 14 years) who did not have diabetes but were at high risk for developing type 1 diabetes (relatives of patient with type 1 diabetes, positive for at least two diabetes autoantibodies, and with evidence of impaired fasting glucose or impaired glucose tolerance) [14]. Teplizumab delayed the median time to diagnosis of type 1 diabetes (48.4 versus 24.4 months). In addition, type 1 diabetes was diagnosed in fewer individuals assigned to teplizumab (43 versus 72 percent, hazard ratio [HR] 0.41, 95% CI 0.22-0.78) (figure 1).
Otelixizumab — Otelixizumab is another humanized monoclonal anti-CD3 antibody (ChAglyCD3). A multicenter trial included 80 patients with new-onset type 1 diabetes who were randomly assigned to otelixizumab for six consecutive days or placebo with the following results [15]:
●At 6, 12, and 18 months, residual beta-cell function, as assessed by glucose-clamp-induced C-peptide release before and after the administration of glucagon, was better maintained in the anti-CD3 antibody group.
●Insulin dose requirements increased in the placebo but not the treatment group.
●In the subgroup of patients with initial residual beta-cell function at or above the 50th percentile of the 80 patients, mean insulin dose at 18 months was 0.22 and 0.61 international units/kg/day in the treatment and placebo groups, respectively.
●Administration of anti-CD3 antibody was associated with significant but transient side effects, including fever after the start of infusions and rash and acute mononucleosis-like syndrome after the end of treatment.
In a follow-up report on 64 patients followed for a mean of 48 months, treatment for six days with otelixizumab appeared to have a durable effect [16]. In the active treatment group, there was a delay in the rise in insulin requirements (+0.09 compared with +0.32 units/kg/day in the placebo group).
Rituximab — Rituximab, an anti-CD20 monoclonal antibody, has been used for the treatment of B-cell neoplasia and antibody-mediated autoimmune diseases, such as rheumatoid arthritis. (See "Rituximab: Principles of use and adverse effects in rheumatoid arthritis".)
It is not well studied in autoimmune diseases that have a cell-mediated pathogenesis. However, rituximab perhaps marks the beginning of new therapeutic strategies for type 1 diabetes, particularly if used in combination with insulin secretogogues and/or other immunomodulatory agents. In a preliminary trial to determine its safety and efficacy in preserving C-peptide levels in 87 patients with new-onset type 1 diabetes, the mean AUC for C-peptide levels during a mixed-meal tolerance test was significantly higher in the rituximab versus placebo group (0.56 versus 0.47 pmol/mL) [17]. The clinical significance of this modest difference in the AUC is unclear. More patients in the rituximab compared with the placebo group developed side effects (eg, fever, rash, pruritus, nausea, vomiting) after the first infusion. The reactions appeared to be minimal with subsequent infusions. There was no increase in infections or neutropenia with rituximab.
The actual mechanism of the effect of rituximab in type 1 diabetes remains to be established. This drug may reduce the production of proinflammatory cytokines that enhance the immune response locally within the pancreas or the pancreatic lymph nodes. It is also possible that antigen presentation exerted by B lymphocytes, which is required for T-lymphocyte activation, is somehow affected by rituximab. Additional anti-B-lymphocyte agents, such as humanized anti-CD20 antibodies, should be tested in newly diagnosed type 1 diabetic patients.
Interleukin-1 inhibition — IL-1 is an innate immune mediator that contributes to the pathogenesis of type 1 diabetes. It was hypothesized that IL-1 inhibition would therefore prevent ongoing beta-cell destruction in recent-onset type 1 diabetes. However, randomized trials assessing monotherapy with anakinra (human IL-1 receptor antagonist) or canakinumab (a human monoclonal anti-IL-1 antibody) were not effective in maintaining beta-cell function, as measured by standard, mixed-meal-stimulated C-peptide secretion [18].
Thymoglobulin — Thymoglobulin or antithymocyte globulin (ATG) is used in organ transplantation, but little is known about its potential beneficial effects in type 1 diabetes. Preclinical studies have shown that antilymphocyte serum treatment in NOD mice with recent-onset diabetes can induce disease remission.
In patients with newly diagnosed type 1 diabetes, equine ATG appeared to prolong the honeymoon phase of the disease. A phase II, placebo-controlled trial is in progress; patients with newly diagnosed type 1 diabetes will be randomly assigned to rabbit polyclonal ATG (Thymoglobulin 6.5 mg/kg administered over four days) or placebo. The primary endpoint will be the presence of residual endogenous insulin secretion at 12-month follow-up. The anticipated outcome is that the ATG group will have sustained endogenous insulin secretion and will exhibit a lower daily exogenous insulin requirement when compared with the control group. This study is supported by both The Immune Tolerance Network and TrialNet.
Insulin — When given by injection in the prediabetic period to NOD mice and BB rats, insulin delays or prevents diabetes [19,20]. Insulin may act as an immunomodulator in these animals [19,21]. Alternatively, it may allow the beta cells to rest and therefore become less vulnerable to further autoimmune attack because they express fewer autoantigens [22].
Two observations in NOD mice provide support for an active immunomodulatory response:
●Immunization of mice with intact insulin or its A or B chains does not reduce islet-cell infiltration; however, immunization with insulin or its B chain is associated with both prevention of diabetes and a cellular infiltrate in which the content of interferon-gamma mRNA was reduced [19]. These observations suggest that insulin converts the insulitis from a destructive to a protective response [23].
●Diabetes in mice can also be prevented by immunization with a metabolically inactive insulin analog [24]; this finding would not be expected if beta-cell rest were an important protective mechanism.
The possible rationale for insulin as an immunomodulator has come from studies in NOD mice demonstrating that an epitope on the insulin B chain is recognized by the diabetogenic, infiltrating CD8+ T cells [25]. (See "Pathogenesis of type 1 diabetes mellitus".)
It had been hoped that these findings would be applicable to humans. Two small studies suggested that insulin therapy might slow progression in patients with newly diagnosed type 1 diabetes [26] or prevent diabetes in those at high risk (islet-cell antibody-positive siblings of a diabetic proband), respectively [27]. (See "Prediction of type 1 diabetes mellitus".)
However, subsequent trials have not shown a benefit [28-32]. As examples:
●In the Diabetes Prevention Trial (DPT), 339 patients at high risk for type 1 diabetes (siblings of diabetic patients with high serum islet-cell antibody concentrations and low acute insulin responses to glucose) were randomly assigned to receive close observation or low-dose subcutaneous insulin (ultralente insulin administered twice daily; total daily dose 0.25 units/kg) plus annual, four-day continuous insulin infusions [28]. After median follow-up of 3.7 years, the cumulative incidence of diabetes was similar in the intervention and observation groups (42 versus 41 percent, respectively, HR 0.96, 95% CI 0.69-1.34) (figure 2).
●In the same trial, 372 subjects at intermediate (rather than high) risk for type 1 diabetes were randomly assigned to oral insulin (7.5 mg daily) or placebo [29]. Oral insulin was similarly ineffective in delaying or preventing type 1 diabetes (HR 0.76, 95% CI 0.51-1.14). However, in a subgroup analysis, a small beneficial effect of oral insulin administration in preventing type 1 diabetes was seen in participants with confirmed high levels of insulin autoantibodies (IAAs) (≥80 nU/mL) [29] and a greater effect in subjects with IAA levels greater than 300 nU/mL [33].
●A subsequent trial was designed to evaluate the role of oral insulin in relatives of patients with type 1 diabetes who were similar to those in the subgroup of the DPT who had experienced apparent benefit [30]. In this trial, 560 insulin autoantibody-positive relatives of patients with type 1 diabetes were randomly assigned to oral insulin (7.5 mg daily) or placebo. After a median follow-up of 2.7 years, diabetes was subsequently diagnosed in 28.5 and 33 percent of participants, respectively. There was no significant difference in time to development of type 1 diabetes (HR 0.87, 95% CI 0-1.2).
●In a Finnish trial of nasally inhaled insulin in children with human leukocyte antigen (HLA)-conferred risk for type 1 diabetes and high levels of at least two types of autoantibodies (islet cell, insulin, glutamic acid decarboxylase, protein tyrosine phosphatase-like protein), there was no difference in the onset of type 1 diabetes in those randomly assigned to intranasal insulin versus placebo (HR 1.14, 95% CI 0.73-1.77) [31].
Thus, administration of insulin at doses used in these trials did not delay or prevent the onset of type 1 diabetes in high-risk individuals.
In a subsequent multinational trial of escalating doses of oral insulin versus placebo in 25 autoantibody-negative, genetically high-risk children aged 2 to 7 years, the highest dose (67.5 mg) resulted in an immune response, suggesting a protective regulatory T cell response when higher doses of oral insulin are administered [34]. No children in the study developed diabetes. This concept requires further investigation.
Immunotherapy DAB486-IL-2 — Future treatments may use specific immunotherapy targeted at the initiation of the autoimmune process (figure 3) [2]. One such study has begun in France [35]. Prediabetic subjects are initially treated with seven daily infusions of DAB-IL2, a diphtheria toxin conjugated to part of the IL-2 molecule designed to target activated T cells; the infusions are followed by low doses of cyclosporine. It is hoped that the initial therapy with DAB-IL2 will slow the ongoing insulitis by destroying the activated T cells invading the islets and that the subsequent therapy with cyclosporine will maintain the remission.
Another possible approach is aimed at the mechanisms by which cytotoxic T cells injure beta cells. Studies in NOD mice suggest that cytokines secreted by islet-infiltrating mononuclear cells induce the expression of intercellular adhesion molecules (such as ICAM-1) on the beta cells, which accelerates their destruction by cytotoxic T cells [36]. Interruption of this process by monoclonal antibodies against ICAM-1 [36] or perhaps by insulin [23] could prevent the development of diabetes. Alternatively, the activated T cells involved in the destructive infiltrate could be specifically targeted for destruction.
GAD65 immunotherapy — Glutamic acid decarboxylase, 65 kilodalton isoform (GAD65), is a major autoantigen in patients with type 1 diabetes (see "Pathogenesis of type 1 diabetes mellitus"). When administered by injection in the prediabetic period to NOD mice, disease progression was slowed [37]. However, in a randomized trial of a vaccine formulation of recombinant human GAD65 (GAD-alum) or subcutaneous placebo (alum alone) in newly diagnosed (within 18 months) children and adolescents with type 1 diabetes, there was no difference in fasting C-peptide concentrations or insulin requirements after 15 months, despite the fact that vaccinated children demonstrated a GAD-specific immune response [38].
In a similar trial, 145 patients (aged 3 to 45 years) diagnosed with type 1 diabetes within the past three months were randomly assigned to receive three doses of GAD-alum, two doses with a third placebo dose (alum alone), or three placebo doses [39]. After one year, there was no difference in mean AUC of serum C-peptide concentration during a mixed-meal tolerance test. In a second trial in which GAD-alum or placebo was administered to 334 patients within three months after diagnosis, stimulated C-peptide concentrations declined to a similar degree in both study groups at 15-month follow-up [40]. Thus, earlier administration of alum-formulated GAD65 (3 versus 18 months after diagnosis) did not affect the primary outcome. Before autoantigen treatment is used, the dose, route, and regimen that could block antigen-specific pathogenic responses need to be better understood.
Costimulation modulation — T cell activation by antigen-presenting cells requires two distinct signals. The first is binding of the T cell receptor-peptide-MHC II complex. The second signal is binding of cell-surface costimulatory molecules (CD80 and CD86 on antigen-presenting cells and CD28 on T cells) that allow full activation of T cells.
Abatacept (also called CTLA4-Ig) is a soluble fusion protein comprising cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) and the Fc portion of IgG1. It prevents CD28 from binding to its counter-receptor, CD80/CD86 due to its higher affinity for CD80/CD86. Administration of CTLA4-Ig or successful transfer of its gene prevents full T cell activation. It ameliorates collagen-induced arthritis in mice and is beneficial in transplantation models. It is used for the treatment of rheumatoid arthritis. (See "Transplantation immunobiology", section on 'T cell costimulation' and "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Methotrexate plus abatacept' and "Treatment of rheumatoid arthritis in adults resistant to initial biologic DMARD therapy", section on 'Abatacept'.)
Costimulation modulation with abatacept may block the production of autoaggressive T cells and thereby slow beta-cell destruction, as illustrated by the findings of a randomized trial in 112 patients (aged 6 to 45 years) with recently diagnosed type 1 diabetes [41]. Patients were randomly assigned to receive abatacept (10 mg/kg) or placebo infusions (27 infusions over two years) within three months of diagnosis. After two years, the mean AUC for serum C-peptide concentration after a mixed meal was significantly higher in the abatacept group (0.378 versus 0.238 nmol/L). Abatacept delayed loss of beta-cell function for nine months, after which time the slope of the decrease in beta-cell function became similar to that in the placebo group. These findings suggest that T cell activation is still occurring at the time of clinical diagnosis, when abatacept was effective. However, continuing T cell activation subsides as the clinical course progresses, and abatacept is no longer effective.
Longer-term follow-up is required to determine whether the improvement in beta-cell function is sustained after discontinuation of abatacept. Until additional data are available, abatacept should not be used in clinical practice for the prevention of type 1 diabetes.
Bacillus Calmette-Guerin (BCG) — In view of the success of immunomodulation with BCG in NOD mice, BCG has been given in a pilot study to 17 newly diagnosed patients with type 1 diabetes [42]. Eleven (65 percent) went into remission for up to 10 months. This success rate was significantly higher than in historical controls, and there were no side effects.
A larger trial randomly assigned 72 patients with new-onset (less than four weeks) type 1 diabetes to therapy with nicotinamide alone or in combination with BCG [43]. The number of patients in remission in each group was three at three months and none at 12 months. Nicotinamide-only therapy was subsequently reported to be an ineffective prevention therapy [44]. (See 'Nicotinamide' below.)
DiaPep277 — In a study of 35 patients with type 1 diabetes and basal serum C-peptide concentrations >0.1 nmol/L, DiaPep277, an immunomodulatory peptide derived from 60kDa heat shock protein, or placebo were given at zero, one, and six months. At 10 months, mean serum C-peptide concentrations were unchanged in the treatment group, but decreased in the placebo group, suggesting that endogenous insulin production was preserved in the treatment group [45].
Donor splenocytes — Adult NOD mice, characterized by spontaneous autoimmunity that causes diabetes via destruction of pancreatic islets, appear to contain endogenous precursor cells that can give rise to new islets if the underlying autoimmune disease is eliminated [46,47]. This was demonstrated in a series of experiments in which injection of live donor splenocytes with Freund's adjuvant into NOD mice eliminated autoimmunity and resulted in the reappearance of pancreatic beta cells, endogenous insulin secretion, and normoglycemia [47]. The new beta cells had a Y chromosome, suggesting that they originated from the male donor splenocytes, rather than from the female recipient. In other experiments, injection of irradiated (not live) splenocytes also resulted in the appearance of new islets, suggesting that the new pancreatic islet cells can arise from endogenous precursor cells. In contrast, a second study in mice reported that preexisting beta cells rather than endogenous precursor cells are the major source of new beta cells during adult life and after pancreatectomy [48].
Antiinflammatory
TNF-alpha inhibitors — Tumor necrosis factor-alpha (TNFa) is a cytokine involved in the acute inflammatory process. It may play a role in the pathogenesis of beta-cell destruction in type 1 diabetes. (See "Acute phase reactants", section on 'The acute phase response'.)
Etanercept is a recombinant TNFa receptor that binds TNFa and blocks its biologic activity. In a 24-week trial of etanercept versus placebo in 18 patients with newly diagnosed type 1 diabetes, patients assigned to etanercept had lower A1C values (5.9 versus 7.0 percent) and greater increases in C-peptide AUC (39 percent increase versus 20 percent decrease) [49]. Larger trials are required to assess the benefits and risks of etanercept for the prevention of type 1 diabetes.
Interferon alfa — Interferons are multifunctional, immunomodulatory cytokines that may have effects upon the cytokine cascade, including several antiinflammatory properties. In a 12-month trial of oral human recombinant interferon alfa-2a (IF-a) versus placebo in 128 patients with newly diagnosed type 1 diabetes, patients randomly assigned to IF-a (5000 units daily) had a smaller percentage loss of mixed-meal stimulated C-peptide (28 versus 56 percent) [50]. However, there were no differences in A1C or insulin requirements.
Supplements
Nicotinamide — In animals, nicotinamide has been found to protect beta cells from chemical or autoimmune injury [51,52]. It is thought to act as a free radical scavenger or by replenishing nicotinamide adenine dinucleotide, an essential coenzyme, within damaged cells, thereby promoting DNA repair and limiting DNA damage.
Initial reports suggested clinical benefit [53,54]. A 1996 meta-analysis of 10 randomized trials of nicotinamide in a total of 211 patients with newly diagnosed type 1 diabetes found that endogenous insulin production (measured by plasma C-peptide responses) was prolonged with nicotinamide administration, although no patient had a remission [53].
However, nicotinamide was ineffective in a large prevention trial. The Nicotinamide Diabetes Intervention Trial (ENDIT) studied 552 individuals ages 3 to 40 years with a first-degree relative with type 1 diabetes, positive islet-cell antibodies (≥20 Juvenile Diabetes Federation units), and nondiabetic glucose tolerance test; the subjects were randomly assigned to receive nicotinamide (1.2 g/m2 per day) or placebo [44]. After a mean follow-up of 3.3 years, the following results were seen:
●The risk of developing diabetes was similar in the two groups: 82 of 274 (30 percent) and 77 of 275 (28 percent) in the treatment and placebo groups, respectively (HR 1.07).
●The results were the same after adjustment for age, oral glucose tolerance status, antibody status, and first-phase insulin response on an intravenous glucose tolerance test.
Vitamin D supplements — Although cow's milk may be associated with an increased risk for type 1 diabetes, one component, vitamin D, may be protective. Support for this hypothesis comes from a case-control study in seven European countries that suggested that supplementation with vitamin D in early infancy can protect against development of type 1 diabetes [55]. A similar protective effect was noted in a birth-cohort study of over 10,000 children [56]. The children who regularly took vitamin D (2000 international units daily) had a reduced risk of type 1 diabetes compared with children whose vitamin D intake was less (relative risk [RR] 0.22). However, in an observational study of fasting C-peptide levels in children recently diagnosed with type 1 diabetes, vitamin D exposure was unexpectedly inversely associated with fasting C-peptide levels [57].
Omega-3 polyunsaturated fatty acids — Omega-3 polyunsaturated fatty acids may have a modulating effect on the inflammatory response [58,59]. A relative deficiency of omega-3 fatty acids, a characteristic of many Western diets, may predispose to enhanced inflammatory reactions and therefore increase the risk of autoimmune diseases, such as type 1 diabetes. In one study, increased dietary intake of omega-3 fatty acids was associated with reduced risk of islet autoantibody responses in children at increased genetic risk for type 1 diabetes (see "Pathogenesis of type 1 diabetes mellitus", section on 'Omega-3 fatty acids'). A primary prevention trial of the effect of docosahexaenoic acid (an omega-3 fatty acid) on the natural history of type 1 diabetes is in progress [60].
Other
Avoidance of cow's milk — Although the complex dietary proteins found in cow's milk were initially thought to be an immunogenic trigger, a subsequent international, double-blind trial did not support avoidance of cow's milk for diabetes prevention. (See "Pathogenesis of type 1 diabetes mellitus", section on 'Cow's milk'.)
In an early trial, 230 breastfed infants with HLA-conferred susceptibility to type 1 diabetes and at least one family member with type 1 diabetes were randomly assigned to supplementation with a casein hydrolysate formula or a conventional cow's milk-based formula [61]. After 10 years of follow-up, at least one autoantibody developed in 17 and 33 children in the casein hydrolysate and cow's milk groups, respectively (17 versus 30 percent, adjusted HR 0.51, 95% CI 0.28-0.91). The cumulative incidence of islet cell and IA-2 autoantibodies was significantly lower in the casein hydrolysate group. The development of type 1 diabetes, which was recorded but was not a primary endpoint of the trial, was similar between the two groups (6 versus 8 percent).
In a subsequent trial, over 2000 infants with HLA-conferred susceptibility to type 1 diabetes and at least one first-degree relative with type 1 diabetes were randomly assigned to be weaned to a casein hydrolysate or a conventional adapted cow's milk formula supplemented with 20 percent of the casein hydrolysate [62,63]. After a median follow-up of 11.5 years, there was no significant difference in the cumulative incidence of type 1 diabetes (8.4 and 7.6 percent, respectively) [63].
Hematopoietic stem cell transplant — The proposed mechanism of action of hematopoietic stem cell transplantation is that it provides a period free from memory T cell influence, during which maturation of new lymphocyte progenitors can occur without recruitment to anti-self-activity. (See "The adaptive cellular immune response: T cells and cytokines", section on 'Memory T cells'.)
Allogeneic stem cell transplant (SCT) is used to treat malignant diseases. There is increasing interest in the use of SCT to treat autoimmune diseases, such as systemic sclerosis, although experience is limited. (See "Immunomodulatory and antifibrotic approaches to the treatment of systemic sclerosis (scleroderma)", section on 'Autologous stem cell transplantation'.)
One study evaluated autologous nonmyeloablative stem cell transplantation in 15 patients (mean age 19.2 years) with recently diagnosed type 1 diabetes [64]. Conditioning was achieved with cyclophosphamide and ATG. Thirteen patients did not require exogenous insulin for 1 to 35 months (mean 14.8 months) in this uncontrolled trial. Peak stimulated C-peptide levels increased in 11 of 13 patients studied at six months and four of four studied at 24 months. In a follow-up evaluation (mean 29.8 months) that included eight additional patients, the improvement in C-peptide levels was sustained at 24 and 36 months after transplantation in patients in whom insulin therapy was either continuously (n = 12) or transiently (n = 8) discontinued [65].
Although beta-cell function appears to have improved, the study has several limitations. The major limitation of this study is the lack of a randomized control group that either received no intervention or received only immunosuppression or immunomodulation. Since newly diagnosed type 1 diabetes can be associated with a honeymoon period characterized by a reduction in insulin requirements, a control group is necessary. In addition, it is not clear what role the treatment with cyclophosphamide and/or ATG may have played independent of the complete SCT procedure.
Data on insulin and IA-2 autoantibodies, two major markers currently used to confirm the diagnosis of type 1A diabetes and to predict type 1 diabetes, are not provided. The presence of at least three autoantibody markers (GAD65, IA-2, and IAAs) should be required for enrollment in a SCT trial. The study is also limited by small size, short duration of follow-up, and side effects of the procedure, including febrile neutropenia, nausea, vomiting, and alopecia. Pneumonia occurred in two individuals and premature ovarian failure in one. Thus, it is unclear if SCT has a role in the treatment/reversal of newly diagnosed type 1 diabetes.
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Type 1 diabetes (The Basics)")
●Beyond the Basics topics (see "Patient education: Type 1 diabetes: Overview (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS — Although the prevention of type 1 diabetes is still at the stage of research trials, the trials are often mentioned in the lay press. As a result, many patients with type 1 diabetes (or their parents) ask their doctors about screening of other family members (particularly children) and what could be done if the family member has a high risk for the development of type 1 diabetes. The approach to screening individuals at risk for type 1 diabetes is discussed elsewhere. (See "Prediction of type 1 diabetes mellitus".)
In a research setting, the following evaluation may be considered:
●Test individuals at high risk for type 1 diabetes development on the basis of family history for GAD65 and insulinoma-associated 2 molecule (IA-2) autoantibodies.
●If they are present and confirmed in a subsequent sample, tests for insulin, zinc transporter-8 (ZnT8), and islet-cell antibodies can be done and the first-phase insulin response to glucose determined.
The occurrence of multiple antibodies against islet autoantigens serves as a surrogate marker of disease in primary or secondary intervention strategies aimed at halting the disease process.
Genetic typing for susceptibility or protective human leukocyte antigen (HLA) alleles can also be performed. In the aggregate, this information can be used to ascertain if a high-risk subject is eligible to be entered into an ongoing prevention trial but is not recommended in routine clinical practice.
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges David McCulloch, MD, who contributed to earlier versions of this topic review.
The UpToDate editorial staff also acknowledges Massimo Pietropaolo, MD (deceased), who contributed to earlier versions of this topic.
