Prevention of type 1 diabetes
Prevention may be attempted at three levels: in early childhood before there is evidence of immune activation directed against islet cells (primary prevention); in non-diabetic individuals with humoral or metabolic markers of high risk of progression to diabetes (secondary prevention); and in the attempt to prolong residual beta cell function in the newly diagnosed (tertiary prevention). Interventions may aim to avoid, limit or reverse harmful immune effector mechanisms, and could potentially be supplemented by other measures designed to enhance beta cell survival or regeneration. The first large trial of primary prevention, by exclusion of cow's milk from the early diet, is currently under way. Secondary prevention trials have been undertaken with oral, inhaled or injected insulin, and with nicotinamide, but the results have been disappointing. Many tertiary prevention studies have been undertaken. Non-specific interventions such as ciclosporin A provide proof of principle that immunotherapy can prolong beta cell function, but the benefits are short-lived and the adverse effects often unacceptable; anti-CD3 antibodies currently appear more promising and are being evaluated in clinical trials.
The pathogenesis of type 1 diabetes is conveniently summarised in terms of the postulated decline in beta cell mass as the disease process moves through its successive stages. Potential intervention strategies can be matched to each of these phases.
Three phases of intervention
Autoantibody combinations conferring a high risk of progression to diabetes typically become established within the first 3 years of life, and primary prevention should be therefore be attempted as early in life as possible; ideally soon after birth. Children born to a family affected by type 1 diabetes, especially those with high risk HLA genotypes, are most appropriate for trial interventions, and such families are highly motivated to participate. Safety is the major criterion for any form of primary prevention, since only a small percentage of those at risk this will be expected to develop diabetes.
Primary prevention might be attempted by avoidance of environmental risk factors, if these could be identified with any confidence. For example, unequivocal identification of a viral cause for type 1 diabetes might lead to vaccination against the virus. Alternatively, potential risk factors could be removed from the environment, as for example by avoidance of cow's milk in early infancy. A major multinational trial known as TRIGR (Trial to Reduce IDDM in the Genetically at Risk) is currently under way to test this hypothesis.
This is offered to individuals at increased risk of progression to diabetes, identified by a positive family history plus islet autoantibody testing, with additional measurement of first phase insulin secretion and glucose tolerance in those who have antibodies. Metabolic testing identifies those at imminent risk of progression, and helps to stage the disease process. Three large studies have reported to date.
The Diabetes Prevention Trial – Type 1 (DPT-1) reported on use of injected insulin therapy. This unblinded trial screened 84,000 first degree relatives – 250 for each entrant to the trial. Risk was estimated by measurement of ICA, followed by performance of two intravenous glucose tolerance tests, allowing identification of an estimated >50% risk of progression to diabetes within 5 years. Parenteral insulin proved ineffective in the prevention of type 1 diabetes, as did a second trial based on oral insulin administration. 
The European Nicotinamide Diabetes Intervention Trial (ENDIT) was a blinded comparison of high dose nicotinamide or placebo which required screening of >30,000 first degree relatives in 21 countries. Recruitment was based upon high titre ICA, followed by an intravenous glucose tolerance test to allow disease staging. Subsequent analysis showed that a screening strategy based upon multiple antibodies could potentially have allowed two intervention trials to be conducted within the same screening population. Nicotinamide also proved ineffective in the prevention of diabetes.
These major studies have demonstrated the feasibility of large scale controlled trials in antibody-positive first degree relatives, but the logistics are daunting, and the number of interventions that can be tested is very limited. For this reason most investigators have opted to look for efficacy in recently diagnosed patients before testing further agents in secondary intervention trials.
The functional bea cell mass may be sufficient to allow a patient to enter the 'honeymoon' period following diagnosis, and even to stop insulin altogether for a matter of weeks or months. Preservation of this functional mass offers many potential benefits, such as better glucose control with less risk of hypoglycaemia. Conversely, good glucose control from diagnosis has been shown to to prolong beta cell survival.
Trials with ciclosporin A demonstrated that beta cell survival can be improved by immunotherapy, although the effect of intervention proved transient, and not all patients derived benefit. This unfortunately includes the majority of children, in whom beta cell loss appears particularly rapid.
Proposed immune interventions fall into two main classes; immunosuppressive therapy which aims to over-ride host immune responses by blockade of selected pathways, or antigen-based therapies designed to induce tolerance. Combinations of these alternatives also appear attractive.
Given the need for safety, the most appealing approach is induction of tolerance. This is based on the inbuilt capacity of the immune system to switch off responses to antigens selectively. Thus, in theory, unwanted antigenic responses could be modified by appropriate presentation of the target antigen. Animal experiments suggest that large doses of antigen induce anergy or deletion of antigen-specific T cells, whereas low doses invoke regulatory T cells. The latter are primed against a single antigen and migrate to the tissue in which this is present. Once activated by the presence of the target antigen, however, they secrete cytokines which produce non-specific downregulation of all proinflammatory (Th1) immune processes in the immediate vicinity, including those directed against other antigenic determinants. This characteristic, known as bystander suppression, could be particularly valuable in the treatment of autoimmune disease in which responses are directed against multiple epitopes of multiple autoantigens, some of which are still unknown.
Oral tolerance is effective in mouse models of autoimmunity, in which prior exposure to an antigen abolishes or limits the harmful consequences of later exposure to the same antigen. Tolerance can also be achieved in the presence of an established autoimmune process, although with less benefit. A series of trials of oral tolerance in human autoimmune disease were launched in the 1990s, but have proved disappointing.
More promising results have, however, been reported from recent studies of immunosuppression based on use of antibodies directed against the CD3 complex, which is located on the surface of lymphocytes in stable association with the T cell receptor, and plays a central role in antigen-specific activation of T cells. OKT3, a murine anti-CD3 antibody, has been widely used to suppress T cell responses involved in organ rejection, but is highly antigenic and stimulates production of neutralising antibodies. A further problem is that the invariant Fc region of this antibody binds to the Fc receptor on macrophages and lymphocytes, causing massive cytokine release. Anti-CD3 molecules have therefore been engineered to overcome these problems by substitution of rodent sequences within the antigen-binding regions of the molecule and by use of a human Fc tail modified to prevent binding with the Fc receptor. The resulting antibody is non-mitogenic, only weakly antigenic, and does not produce the full cytokine release syndrome. Very promising results have been reported from the NOD mouse, which – in marked contrast to all other interventions – responds preferentially at the stage of overt diabetes and enters lasting remission, provided that treatment is given within 7 days of clinical onset. Trials in humans fall short of the dramatic benefit seen in the mouse, but are sufficiently promising to attract pharmaceutical investment in the area.
One encouraging development in this field has been the agreement of groups within the USA and worldwide to work together in developing and performing future intervention trials, but on current evidence there is still a long way to go.
^ TRIGR Study Group. The Trial to Reduce IDDM in the Genetically at Risk (TRIGR): recruitment, intervention and follow-up. Diabetologia 2011;54:627–33
^ Diabetes Prevention Trial – Type 1 Diabetes Study Group. Effects of insulin in relatives of patients with type 1 diabetes mellitus. New Engl J Med 2002;346(22):1685–91
^ Skyler JS et al. Effects of oral insulin in relatives of patients with type 1 The Diabetes Prevention Trial – Type 1. Diabetes Care 2005;28(5):1068–76
^ Gale EAM et al. The European Nicotinamide Diabetes Intervention Trial (ENDIT): a randomized controlled trial of intervention before the onset of type 1 diabetes mellitus. Lancet 2004;363(9413):925–31
^ Diabetes Control and Complications Trial Research Group. Ann Intern Med 1998;128(7):517–23.
^ Weiner HL. Oral tolerance for the treatment of autoimmune diseases. Annu Rev Med 1997;48:341–51
^ Herold KC et al. Anti-CD3 monoclonal antibody in new-onset type 1 diabetes mellitus. New Engl J Med 2002;346(22):1692–8
^ Skyler JS, Ricordi C. Stopping type 1 diabetes – attempts to prevent or cure type 1 diabetes in man. Diabetes 2011;60:1–8