Antigen-Based Vaccination and Prevention of Type 1

0 downloads 0 Views 220KB Size Report
(IPEX) syndrome due to FOXP3 mutations [5]. Two subsets of Treg express ... the HOMA-R index) [12], further refine risk in autoantibody- positive individuals.
Curr Diab Rep DOI 10.1007/s11892-013-0415-7

PATHOGENESIS OF TYPE 1 DIABETES (D DABELEA, SECTION EDITOR)

Antigen-Based Vaccination and Prevention of Type 1 Diabetes Leonard C. Harrison & John M. Wentworth & Yuxia Zhang & Esther Bandala-Sanchez & Ralph M. Böhmer & Alana M. Neale & Natalie L. Stone & Gaetano Naselli & Julian J. Bosco & Priscilla Auyeung & Maryam Rashidi & Petra Augstein & Grant Morahan

# Springer Science+Business Media New York 2013

Abstract Insulin-dependent or type 1 diabetes (T1D) is a paradigm for prevention of autoimmune disease: Pancreatic β-cell autoantigens are defined, at-risk individuals can be identified before the onset of symptoms, and autoimmune diabetes is preventable in rodent models. Intervention in asymptomatic individuals before or after the onset of subclinical islet autoimmunity places a premium on safety, a requirement met only by lifestyle–dietary approaches or autoantigenbased vaccination to induce protective immune tolerance. Insulin is the key driver of autoimmune β-cell destruction in the nonobese diabetic (NOD) mouse model of T1D and is an early autoimmune target in children at risk for T1D. In the NOD mouse, mucosal administration of insulin induces regulatory T cells that protect against diabetes. The promise of autoantigen-specific vaccination in humans has yet to be realized, but recent trials of oral and nasal insulin vaccination in at-risk humans provide grounds for cautious optimism. Keywords Autoimmune . Type 1 diabetes . Islet . β cell . Preclinical . Autoantibody . T cell . Regulation . Prediction . Prevention . Vaccination . Insulin . GAD65 . Mucosa . Nasal . Oral . Immune . Suppression . Tolerance . Clinical trial . Antigen

Introduction If immune responses to autoantigens are regulated physiologically, the logical strategy to counter aberrant immune responses leading to autoimmune disease is autoantigen-based L. C. Harrison (*) : J. M. Wentworth : Y. Zhang : E. Bandala-Sanchez : R. M. Böhmer : A. M. Neale : N. L. Stone : G. Naselli : J. J. Bosco : P. Auyeung : M. Rashidi : P. Augstein : G. Morahan Walter & Eliza Hall Institute of Medical Research, 1G Royal Parade, Parkville 3052, Victoria, Australia e-mail: [email protected]

immunotherapy. This approach has a parallel in allergenspecific immunotherapy, which has been practiced for decades and shown in multiple randomized trials to be efficacious in allergic asthma, rhinitis, and other conditions. The aim of the administration of autoantigens to induce protective immune tolerance, “negative vaccination” [1], is to boost or restore autoantigen-specific immunoregulatory mechanisms and avert pathological autoimmunity. In animal models of autoimmune disease, this has been achieved in several ways: administration of autoantigen by a “tolerogenic” route (mucosal, dermal), cell type (resting dendritic cell), mode (with blockade of costimulation molecules), or form (as an “altered peptide ligand”) [2, 3]. Mechanisms of antigen-induced tolerance include deletion and/or anergy of effector T cells and induction of regulatory T cells (Treg). Treg cells stand center stage following their renaissance as CD4+CD25 + cells programmed by the transcription factor forkhead box P3 (FOXP3) and the unequivocal evidence that they are required naturally to prevent autoimmune disease [4], exemplified by the immune dysregulation, polyendocrinopathy, and X-linked (IPEX) syndrome due to FOXP3 mutations [5]. Two subsets of Treg express FOXP3: natural Treg (nTreg) and induced Treg (iTreg) [reviewed in ref. 6]. Like conventional T cells (Tconv), nTreg cells are thymus derived, whereas iTreg cells differentiate in the periphery from naïve Tconv under “tolerogenic” conditions of antigen presentation, such as in the presence of transforming growth factor (TGF)-β in the gut [7]. Of importance clinically is the ability of Treg to exert antigen-nonspecific “bystander” or “linked” suppression. Thus, Treg generated in response to epitopes within a specific antigen, presented by different human leukocyte antigen (HLA) molecules to heterogeneous T-cell receptors (TCRs), may impair (by direct cell contact and/or the release of soluble immunosuppressive factors) the ability of local antigenpresenting dendritic cells to elicit effector T-cell responses to the same or another antigen presented locally at the site of the lesion or in the draining lymph nodes. Bystander suppression

Curr Diab Rep

does not depend on the “tolerizing” autoantigen necessarily being the primary driver of pathology. It is clinically important because it obviates the specificity restrictions imposed by variation in the HLA and TCR repertoire in humans. Prevention of autoimmune disease can be defined as primary or secondary depending on whether treatment is initiated before or after onset of the disease process. Attempts to prevent autoimmune disease in otherwise healthy individuals would ideally be undertaken in those in whom risk of clinical disease can be predicted with high sensitivity and precision, using well-characterized agents that have been shown in animals and humans with the disease to be at least safe, if not efficacious. Immune tolerance induction by autoantigen-specific vaccination is likely to satisfy the prime requirement for safety in at-risk individuals, assuming that it does not accelerate the disease process. In T1D, there are good reasons why insulin itself is the most appropriate autoantigen for vaccination. In nonlymphoid tissues, insulin is restricted to the β cells of the pancreatic islets, whereas other T1D autoantigens are expressed more widely. A large body of data directly demonstrates that insulin, acting as an autoantigen, is the primary driver of β-cell destruction in the nonobese diabetic (NOD) mouse model of T1D [8–10]. In infants genetically at risk for T1D and followed from birth, circulating autoantibodies to insulin are usually the first marker of subclinical disease, preceding autoantibodies to the molecular weight 65,000 isoform of glutamic acid decarboxylase (GAD65) and to other islet antigens [for a review, see ref. 11]. In children who have a T1D first-degree relative and normal β-cell function, the 5-year risk for clinical diabetes is 50 % if they have autoantibodies to one, two, and three islet antigens, respectively. Autoantibodies to insulin are the most predictive. These predictions have been validated in the Diabetes Prevention Trial–1 (DPT–1) trials of systemic and oral insulin (see below). Measures of insulin secretion (first-phase insulin response to intravenous glucose-FPIR), as well as insulin action (estimated from fasting blood glucose and plasma insulin as the HOMA-R index) [12], further refine risk in autoantibodypositive individuals. In NOD mice, administration of insulin, proinsulin peptides, or proinsulin DNA via mucosal routes, where it is not absorbed but acts locally to induce Treg, protects against development of diabetes [reviewed in refs. 1 and 3]. As well as autoimmunity to insulin, NOD mice and humans with T1D share other features—namely, polygenic inheritance dominated by the major histocompatibility complex, disease transfer by bone marrow, and a protracted subclinical phase. Finally, because insulin has been administered systemically to humans as a hormone for over 90 years to treat diabetes, its application as a vaccine in phase 2 trials does not present major regulatory hurdles.

Randomized Trials of Antigen-Specific Vaccination in T1D These trials are summarized in Table 1, and further details of recent trials can be found at ClinicalTrials.gov.

Vaccination with Insulin Intensive systemic insulin treatment was shown to prolong the “honeymoon phase” after diagnosis of T1D [13], but whether systemic insulin acted only as a hormone to control blood glucose and “rest” β cells (making them less sensitive to immune attack) or also as an antigen to induce immune tolerance was not clear. To answer this question, the multicenter DPT–1 was launched in the U.S. in 1994. The aim of DPT–1 was to determine whether either systemic or oral insulin would delay or prevent diabetes onset in islet autoantibody-positive at-risk first-degree relatives. Low-dose systemic insulin (annual intravenous insulin infusions and daily subcutaneous injections) was given to a high-risk group of relatives (>50 % risk of diabetes over 5 years) matched with an untreated but closely monitored control group, but it had no effect on diabetes incidence [21]. Subsequently, in DPT–1 oral insulin, relatives with a 25 %–50 % 5-year risk of diabetes received 7.5 mg human insulin or placebo daily for a median of 4.3 years. There was no effect overall, but post trial hypothesis testing revealed a significant delay in the onset of diabetes in participants who had significant concentrations of circulating insulin autoantibodies at entry [23]. Recently, it was reported [32•] that the presence of insulin autoantibodies at baseline predicted a beneficial outcome of combination treatment with oral insulin and anti-CD3 antibody in NOD mice. The posttrial finding in DPT–1 oral insulin is the basis for a further, ongoing international trial of oral insulin in atrisk relatives, under the auspices of TrialNet, using the same 7.5-mg dose. Ideally, this follow-up trial will also employ a higher dose, because 7.5 mg is a very small dose, as compared with that required to induce antidiabetogenic Treg in the NOD mouse (see below). Two trials of oral insulin, up to 7.5 mg daily for 12 months, in people with recently diagnosed T1D, showed no protective effect on residual β-cell function [19, 20]. In light of DPT–1, this is not surprising, but it is worth reflecting on why, in contrast to studies in mice, these trials of oral insulin in T1D, and of other autoantigens in multiple sclerosis [33] and rheumatoid arthritis [34, 35], failed to demonstrate clinical benefit. There are several possible nonexclusive explanations (Table 1), as is discussed below. Induction of antigen-specific tolerance, without concomitant inactivation or deletion of pathogenic effector cells, is unlikely to have an effect on end-stage disease. If a balance between pathogenic and protective T cells is deterministic,

Curr Diab Rep Table 1 Randomized trials of antigen-specific vaccination in T1D Antigen

Participants Follow-up Outcome (n) (months)

Reference

Parenteral insulin (i.v. vs. s.c. 2 weeks) RD (26) Parenteral (s.c.) insulin and sulfonylurea RD (27) (glipizide) Parenteral (s.c.) insulin RD (49)

12 12

Parenteral (s.c.) insulin

RD (10)

10

Parenteral insulin (i.v. vs. s.c. 2 weeks)

RD (19)

12

Parenteral (s.c.) insulin

AR (14)

84

Oral insulin

RD (80)

12

Oral insulin

RD (131)

12

Parenteral (s.c.) insulin (Diabetes Prevention Trial Type 1 [DPT–1]) Intranasal insulin (Melbourne Intranasal Insulin Trial [INIT I]) Oral insulin (DPT–1)

AR (339)

44

AR (38)

48

Increased antibody and decreased T-cell responses to insulin

[22]

AR (372)

52

[23]

Intranasal insulin (Diabetes Prediction and Prevention Project [DIPP]) Parenteral (s.c) insulin B chain 9-23 “altered peptide ligand” NBI6024-0101 (“Neurocrine”) Parenteral (s.c.) insulin B chain in incomplete Freund’s adjuvant

AR (224)

21

No effect on diabetes development overall; Post hoc analysis revealed >4-year delay in diabetes onset in participants with insulin autoantibodies No effect to delay progression to diabetes

[24]

RD (188)

25

No effect on C-peptide response to mixed meal

[25]

RD (12)

24

[26]

Intranasal insulin (Intranasal Insulin Trial II [INIT II]) Intranasal insulin (Intranasal Insulin Trial III [INIT III])

AR (120)

60

No effect on C-peptide response to mixed meal; development of sustained insulin-specific antibody and T-cell responses Ongoing

RD (52)

24

[27••]

Parenteral (s.c.) GAD65

RD (47)

6

Parenteral (s.c.) GAD65

RD (70)

30

Parenteral (s.c.) GAD65 Parenteral (s.c.) GAD65 Parenteral (i.m.) proinsulin plasmid DNA

RD (334) RD (145) RD (80)

15 12 12

Oral insulin (TrialNet Study TN07) Parenteral (s.c.) GAD65

AR (300) AR (50)

72–96 60

No effect on metabolic parameters; suppression of T-cell responses to insulin and antibody responses to subcutaneous insulin Increase in fasting and stimulated plasma C-peptide with intermediate dose of 20 μg Delay in loss of C-peptide secretion, in those treated within 6 months of clinical diagnosis No effect on metabolic parameters No effect on C-peptide response to mixed meal Transient improvement in C-peptide response to mixed meal concomitant with decreased CD8 T-cell response to proinsulin Ongoing Ongoing

60

Higher meal-stimulated C-peptide, lower HbA1c Higher basal and glucagon-stimulated C-peptide, more remissions Higher glucagon-stimulated C-peptide and improved insulin sensitivity and glycemic control Higher C-peptide response to oral glucose, HbA1C unchanged Higher meal and glucagon-stimulated C-peptide and lower HbA1c Delay in onset of diabetes, no effect on islet antibody levels No effect on basal C-peptide, HbA1c, insulin dose, or insulin antibodies No effect on basal, glucagon- or meal-stimulated C-peptide, HbA1c, insulin dose, or islet antibody levels No effect on diabetes development

antigen-specific vaccination should be most effective early in subclinical disease or earlier. The route of delivery may be an important variable. Oral delivery may not be optimal, because proteins are degraded after ingestion and the concentration or form of peptide

[13] [14] [15] [16] [17] [18] [19] [20]

[21]

Harrison LC et al.

[28] [29] [30••] [31••] Roep BO et al. (in preparation) Skyler J et al. Elding Larsson H et al.

reaching the jejunum may be inadequate to induce mucosamediated tolerance. Even with a peptide, mucosal responses have been observed after naso-respiratory, but not oral, administration [36]. In the mouse, nasal administration of the model antigen, ovalbumin, induced ovalbumin-specific T-cell

Curr Diab Rep

responses in cervical, mediastinal, and mesenteric mucosal lymph nodes, whereas oral administration induced responses only in the mesenteric nodes [37]. Responses to antigens administered to the mucosa may not always be protective. The mucosa is normally a “tolerogenic” environment but can be a “two-edged sword,” depending on the nature of the antigen and local factors. Oral ovalbumin simultaneously induced both Treg and pathogenic cytotoxic CD8+ T cells, so that a clinical effect (in a transgenic ovalbumin model of T1D) was not seen unless induction of CD8+ T cells was suppressed by coadministration of anti-CD40 ligand monoclonal antibody [38]. A proinsulin B-C chain peptide that induces CD4+ Treg in NOD mice is a “combitope” of CD4+ (I-Ag7-restricted) and CD8+ (Kd-restricted) T-cell epitopes and is significantly more protective after intranasal delivery if the C-terminal p9 anchor residue for binding to Kd is either deleted or mutated [39]. Human insulin contains cytotoxic T-cell epitopes, but whether mucosal insulin induces cytotoxic T cells, as well as Treg, is unknown. There is a pressing need to be able to monitor immune responses to mucosal autoantigens in human trials. Dosage is an important consideration. Induction of “oral tolerance” in the NOD mouse, the rationale for the DPT–1 trial of oral insulin, required milligrams of gavaged insulin to induce CD4+ Treg and reduce the incidence of diabetes [40, 41]. However, the daily dose of oral insulin in DPT–1 and in the other human trials was just 7.5 mg, which, on a body weight basis, equates to only a few micrograms in the mouse. Ingested proteins begin to degrade in the stomach, and the bioavailability of oral insulin in the upper small intestine is unpredictable. Despite this, dose-ranging studies in humans to determine a bioavailable dose of oral insulin have not been undertaken, questioning the wisdom of undertaking and then repeating such a major trial. Although the measurement of insulin-specific T cells remains problematic, the response to oral insulin could have been assessed simply by measuring insulin antibodies, which were found to increase after aerosol insulin in NOD mice [42] and after nasal insulin in humans [22]. None of the oral autoantigen trials sought immune effects, without which evidence of bioavailability is lacking. Future trials must incorporate measures of relevant immune biomarkers. Insulin autoantibodies denote risk for T1D, and the increase in insulin antibodies in response to aerosol or nasal insulin is seemingly contrary to tolerance induction. However, the small but significant increase in insulin antibodies was associated with a simultaneous decrease in T-cell responses to insulin, an observation consistent with the early descriptions of mucosal tolerance and seminal studies in humans of mucosal immunity to the experimental antigen, keyhole limpet hemocyanin (KLH). Nasal administration of KLH to human volunteers elicited an antibody response, but on subsequent challenge with subcutaneous KLH, both antibody and T-cell

responses decreased [43]. Thus, the modest increase in antibody after nasal KLH was offset by a major decrease in the antibody response to subsequent challenge by injected KLH. In a recent randomized trial of nasal insulin in individuals with recent-onset T1D not initially requiring insulin treatment, those allocated to the nasal insulin arm had markedly blunted insulin antibody responses when eventually commenced on subcutaneous insulin [27••] (Fig. 1). This evidence for nasal insulin-induced immune tolerance to exogenous insulin cannot necessarily be extrapolated to endogenous “autoantigenic” insulin but provides a mechanistic rationale for randomized trials of nasal insulin vaccination in individuals at risk for T1D. Future trials need to determine whether nasal insulin induces insulin-specific regulatory T cells and, like nasal KLH, suppresses T-cell responses to rechallenge indicative of T-cell tolerance. In the completed T1D Prediction and Prevention Project (DIPP) in Finland [24], nasal insulin (1 U/kg daily) did not alter the rate of progression to diabetes in islet autoantibodypositive children less than 3 years of age. These children were at very high risk, and many appeared to have had borderline βcell function based on low first-phase insulin responses to i.v. glucose. As in the oral insulin trials, markers of an insulin bioeffect and evidence of immune tolerance were not reported. In the ongoing Type 1 Diabetes Prevention Trial, also known as the Intranasal Insulin Trial II (INIT II), in Australia, New Zealand, and Germany (see stopdiabetes.com.au), nasal insulin (440 IU) or nasal placebo is being administered daily for 7 days and then weekly for a year to T1D relatives 4–30 years of age with autoantibodies to at least two islet antigens (~40 % risk of diabetes over 5 years). The primary outcome is clinical diabetes; secondary outcomes are measures of metabolic and immune function. INIT II and DIPP differ in several potentially important ways: In INIT II, the insulin dose is substantially higher; the participants are older and have less advanced subclinical disease and, therefore, a lower risk.

Fig. 1 Nasal insulin vaccination suppresses the insulin antibody response to subsequent subcutaneous injection of insulin. (Adapted from: Fourlanos S, Perry C, Gellert SA et al: Evidence that nasal insulin induces immune tolerance to insulin in adults with autoimmune diabetes. Diabetes 2011, 60:1237-45) [27••]

Curr Diab Rep

As has been mentioned, antigen-specific vaccination alone is unlikely to work in end-stage disease and may be most effective before the onset of subclinical disease. To address this possibility, a trial of oral insulin vaccination (the PrePOINT trial) is underway in children genetically predisposed to T1D but without evidence of underlying islet autoimmunity [44].

diabetes. At least these trials appear to establish the safety of this agent, justifying an ongoing secondary prevention trial (DIAPREV-IT) of GAD65-alum in islet autoantibody-positive at-risk children.

Vaccination with GAD65

Common sense and experience teach a number of lessons in designing further trials of antigen-specific vaccination in T1D. Dose-finding studies based on immune response markers should first be undertaken before embarking on costly, longterm trials based on diabetes as a clinical end-point. Assuming it is safe, vaccination should occur as early as possible prior to the onset of clinical diabetes. In subclinical disease, vaccination might be boosted by concomitant administration of other agents, but these need to be safe and, at present, are limited to “natural” agents such as vitamin D, although others may become available. On the basis of NOD mouse studies and, now, human trials, vaccination after the onset of clinical diabetes is ineffective but could be tested in combination with conventional immune modulating/suppressing agents administered systemically or, in some cases, even together with antigen via the mucosa or skin. More attention needs to be given to the selection of participants for T1D prevention trials. The contemporary increase in the incidence of T1D is due to the effects of environment on individuals with lower risk HLA and, no doubt, other, susceptibility genes [57]. The number of individuals presenting with T1D who have the classic highest risk HLA DR 3,4 haplotypes has not changed in absolute terms, although their age at diagnosis is lower. How this change in environment–gene interactions might influence the response to preventative approaches, including antigen-specific vaccination, is unknown. However, insulin resistance, a well-known consequence of obesity, was shown to be an independent risk factor for progression to diabetes in at-risk children with islet autoantibodies [12]. Thus, not only insulin secretion, but also insulin resistance determines rate of progression to clinical diabetes and, possibly, the response to preventative approaches and must be considered in trial design and outcome. Risk for T1D is conferred mainly by a restricted set of HLA susceptibility genes, followed then by the insulin gene and, less clearly, by more than 50 other genetic loci [58]. As compared with inbred NOD mice, humans will obviously be heterogeneous in their responses to therapeutic interventions. The application of (pro)insulin as a therapeutic tool in T1D provides a possible insight into personalized pharmacogenomics. The second strongest genetic contribution to human T1D, the insulin gene locus (IDDM 2), maps to a variable number of tandem repeats (VNTRs) upstream of the insulin-coding region; long (class III) and short (class I) VNTR alleles are associated, respectively, with lower and higher susceptibility

Autoantibodies to GAD65 are a sensitive and specific marker of T1D but, in the absence of other islet autoantibodies, are associated with indolent islet autoimmunity, as in adult-onset T1D [45]. Autoantibodies to endogenous GAD are present in the NOD mouse [46], and T-cell reactivity to human GAD65 was reported to be a disease marker in this model [47, 48]. In contrast to the marked protection from diabetes afforded by insulin expressed transgenically in antigen-presenting cells in the NOD mouse [49], transgenic expression of GAD65 was without effect [50]. Nevertheless, systemic administration of GAD65 or plasmid DNA encoding GAD65 [51, 52] or of GAD65 peptides nasally [53] has been reported to decrease the incidence of diabetes. A Swedish company, Diamyd P/L, produced recombinant GAD65 and initiated trials of a subcutaneous GAD65-alum (aluminium hydroxide) formulation [54]. In a dose-ranging phase 2 trial in adults with T1D, no serious adverse events occurred, and at an intermediate dose of 20 μg (given twice, 4 weeks apart), but not at 4, 100, or 500 μg, fasting and stimulated plasma C-peptide increased over 24 weeks [28]. A phase 2 trial in children and adolescents with recent-onset T1D then reported that subcutaneous GAD65-alum delayed loss of C-peptide secretion, in those treated within 6 months of clinical diagnosis [29]. Persistence of an effect would be consistent with induction of disease-specific immune tolerance. In fact, the investigators reported that GAD65-alum vaccination was associated not only with a sustained increase in antibodies to GAD65, but also with the appearance of GAD65-specific Treg [55•, 56•]. Despite these promising early results, GAD65-alum vaccination had no clinical benefit in people with recent-onset T1D in a much larger phase 3 trial by the same investigators [30••] and in a multicenter phase 2 trial undertaken by TrialNet investigators [31••]. This would not be the first time that results from initial trials with relatively small numbers of participants have not been substantiated in larger, independent follow-up trials. On the basis of GAD65 antibody responses, it seems clear that the GAD65alum vaccine was bioactive. However, confirmation of the earlier T-cell findings is awaited with interest because it will be important to know whether tolerance to exogenous GAD65 was or was not induced in the absence of a metabolic effect. For the reasons discussed above, it should not be a surprise that the GAD-alum vaccine was ineffective after the onset of clinical

Conclusions

Curr Diab Rep

to T1D [59]. The length of the VNTR correlates with the level of insulin gene transcription in the thymus [60] and with an insulin splice variant in a peripheral population of myeloid cells [61]. The VNTR therefore may predetermine the extent of deletion of insulin-specific T cells during development and the frequency and avidity of such cells in the periphery. That oral insulin benefited only participants in DPT–1 with insulin autoimmunity suggests that allelism at IDDM2 could determine the immune response not only to endogenous insulin as a target autoantigen, but also to oral insulin as a potential therapeutic tool. Advances in functional genomics, including the elucidation of T1D subtypes, should facilitate the selection and assessment of participants in future trials of antigen-specific vaccination in T1D. Acknowledgments This work was supported by the National Health and Medical Research Council of Australia (Program Grant 516700; Fellowship [LCH] 637301) and was made possible through Victorian State Government Operational Infrastructure Support and Australian Government NHMRC IRIIS. Compliance with Ethics Guidelines Conflict of Interest Leonard C. Harrison, John M. Wentworth, Yuxia Zhang, Esther Bandala-Sanchez, Ralph M. Böhmer, Alana M. Neale, Natalie L. Stone, Gaetano Naselli, Julian J. Bosco, Priscilla Auyeung, Maryam Rashidi, Petra Augstein, and Grant Morahan declare that they have no conflict of interest. Human and Animal Rights and Informed Consent This article does not contain any studies with human or animal subjects performed by any of the authors.

References Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance 1. Harrison LC. Vaccination against self to prevent autoimmune disease: the type 1 diabetes model. Immunol Cell Biol. 2008;89:139–45. 2. Faria AM, Weiner HL. Oral tolerance: mechanisms and therapeutic applications. Adv Immunol. 1999;73:153–64. 3. Harrison LC, Hafler DA. Antigen-specific therapy for autoimmune disease. Curr Opin Immunol. 2000;12:704–11. 4. Wing K, Sakaguchi S. Regulatory T cells exert checks and balances on self tolerance and autoimmunity. Nat Immunol. 2010;11:7–13. 5. Bennett CL, Christie J, Ramsdell F, et al. The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat Genet. 2001;27:20–1. 6. Zhang Y, Bandala-Sanchez E, Harrison LC. Revisiting regulatory T cells in type 1 diabetes. Curr Opin Endocrinol Diabetes Obes. 2012;19:271–8. 7. Curotto de Lafaille MA, Lafaille JJ. Natural and adaptive foxp3+ regulatory T cells: more of the same or a division of labor? Immunity. 2009;30:626–35.

8. Narendran P, Mannering SI, Harrison LC. Proinsulin – a pathogenic autoantigen in type 1 diabetes. Autoimmun Rev. 2003;2:204–10. 9. Zhang L, Nakayama M, Eisenbarth GS. Insulin as an autoantigen in NOD/human diabetes. Curr Opin Immunol. 2008;20:111–8. 10. Krishnamurthy B, Dudek NL, McKenzie MD, et al. Responses against islet antigens in NOD mice are prevented by tolerance to proinsulin but not IGRP. J Clin Invest. 2006;116:3258–65. 11. Harrison LC. Risk assessment, prediction and prevention of type 1 diabetes. Pediatric Diabetes. 2001;2:71–82. 12. Fourlanos S, Narendran P, Byrnes G, Colman P, Harrison LC. Insulin resistance is a risk factor for progression to type 1 diabetes. Diabetologia. 2004;47:1661–7. 13. Shah SC, Malone JI, Simpson NE. A randomized trial of intensive insulin therapy in newly diagnosed insulin-dependent diabetes mellitus. N Engl J Med. 1989;320:550–4. 14. Selam JL, Woertz L, Lozano J, et al. The use of glipizide combined with intensive insulin treatment for the induction of remissions in new onset adult type I diabetes. Autoimmunity. 1993;16:281–8. 15. Linn T, Ortac K, Laube H, Federlin K. Intensive therapy in adult insulin-dependent diabetes mellitus is associated with improved insulin sensitivity and reserve: a randomized, controlled, prospective study over 5 years in newly diagnosed patients. Metabolism. 1996;45:1508–13. 16. Kobayashi T, Nakanishi K, Murase T, Kosaka K. Small doses of subcutaneous insulin as a strategy for preventing slowly progressive b-cell failure in islet cell antibody-positive patients with clinical features of NIDDM. Diabetes. 1996;45:622–6. 17. Schnell O, Eisfelder B, Standl E, Ziegler AG. High-dose intravenous insulin infusion versus intensive insulin treatment in newly diagnosed IDDM. Diabetes. 1997;46:1607–11. 18. Füchtenbusch M, Rabl W, Grassl B, et al. Delay of type I diabetes in high risk, first degree relatives by parenteral antigen administration: the Schwabing Insulin Prophylaxis Pilot Trial. Diabetologia. 1998;41:536–41. 19. Pozzilli P, Pitocco D, Visalli N, et al. No effect of oral insulin on residual beta-cell function in recent-onset type 1 diabetes (the IMDIAB VII). IMDIAB Group. Diabetologia. 2000;43:1000–4. 20. Chaillous L, Lefevre H, Thivolet C, et al. Oral insulin administration and residual beta-cell function in recent-onset type 1 diabetes: a multicentre randomised controlled trial. Diabete Insuline Orale group. Lancet. 2000;356:545–9. 21. Diabetes Prevention Trial-Type 1 Diabetes Study Group. Effects of insulin in relatives of patients with type 1 diabetes mellitus. N Engl J Med. 2002;346:1685–91. 22. Harrison LC, Honeyman MC, Steele CE, et al. Pancreatic beta-cell function and immune responses to insulin after administration of intranasal insulin to humans at risk for type 1 diabetes. Diabetes Care. 2004;27:2348–55. 23. Skyler JS, Krischer JP, Wolfsdorf J, et al. Effects of oral insulin in relatives of patients with type 1 diabetes: the Diabetes Prevention Trial–Type 1. Diabetes Care. 2005;28:1068–76. 24. Nanto-Salonen K, Kupila A, Simell S, et al. Nasal insulin to prevent type 1 diabetes in children with HLA genotypes and autoantibodies conferring increased risk of disease: a double-blind, randomised controlled trial. Lancet. 2008;372:1746–55. 25. Walter M, Philotheou A, Bonnici F, Ziegler AG, Jimenez R. NBI6024 Study Group: no effect of the altered peptide ligand NBI-6024 on beta-cell residual function and insulin needs in new-onset type 1 diabetes. Diabetes Care. 2009;2:2036–40. 26. Orban T, Farkas K, Jalahej H. Autoantigen-specific regulatory T cells induced in patients with type 1 diabetes mellitus by insulin B-chain immunotherapy. J Autoimmun. 2010;34:408–15. 27. •• Fourlanos S, Perry C, Gellert SA, et al. Evidence that nasal insulin induces immune tolerance to insulin in adults with autoimmune diabetes. Diabetes. 2011;60:1237–45. In a randomized controlled trial, nasal insulin markedly suppressed the antibody response to a

Curr Diab Rep

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

subsequent “challenge” by subcutaneous insulin. This is the first evidence of immune tolerance to a mucosally administered autoantigen and provides a rationale for trials of mucosal insulin vaccination in people at risk for T1D. Agardh CD, Cilio CM, Lethagen A, et al. Clinical evidence for the safety of GAD65 immunomodulation in adult-onset autoimmune diabetes. J Diabetes Complications. 2005;19:238–46. Ludvigsson J, Faresjo M, Hjorth M, et al. GAD treatment and insulin secretion in recent-onset type 1 diabetes. N Engl J Med. 2008;359:1909–20. •• Ludvigsson J, Krisky D, Casas R, et al. GAD65 antigen therapy in recently diagnosed type 1 diabetes mellitus. N Engl J Med. 2012;366:433–42. This large randomized trial of subcutaneous GAD65 with alum as adjuvant followed an earlier trial by the same investigators [ref. 29] that suggested that the vaccine reduced the decline in β-cell function after diagnosis. In the previous trial, 70 patients 10–18 years of age with T1D for less than 18 months were treated with two 20-μg doses of either GAD-alum or placebo a month apart. Changes in fasting and meal-stimulated C-peptide secretion were similar after 15 months but declined less in those treated with GAD-alum within 6 months of diagnosis. Here, 334 patients of similar age but diagnosed less than 3 months were randomized to two doses of GAD-alum or placebo. GAD65-alum induced antibodies to GAD65 and was not associated with major side effects, as previously, but did not alter the decline in β-cell function. •• Wherrett DK, Bundy B, Becker DJ, et al. Antigen-based therapy with glutamic acid decarboxylase (GAD) vaccine in patients with recent-onset type 1 diabetes: a randomised double-blind trial. Lancet. 2011;378:319–27. This well-powered randomized controlled trial in 145 people with recent-onset T1D (