Clinical trials in humans

Realisation that type 1 diabetes is an immune-mediated disease gave rise to the hope that immune intervention might modify its course. The landmark demonstration that ciclosporin treatment produced partial preservation of beta cell function gave rise to a major drive for more effective therapy. Interventions in high risk individuals (secondary intervention) are feasible but require major investment of time and resources; the recent focus has therefore been on tertiary intervention, and on screening and testing a range of interventions soon after clinical diagnosis. These strategies, broadly speaking, have aimed for tolerance induction by antigen-specific therapy, or for manipulation of T cell sub-populations; some therapies have been tried on an empirical basis because of known safety and utility in other clinical situations. General limitations of all approaches have included uncertainty as to the major effector mechanisms in humans, excessive reliance on the NOD mouse model, the lack of useful surrogate endpoints, and the paramount need for safety when treating a non-lethal condition with a steadily improving prognosis. The use of small numbers may give rise to both type 1 (false positive) and type 2 (false negative) errors. Strategies combining tolerance induction and beta cell regeneration hold promise for the future.

Historical aspects

A variety of clinical observations in humans indicate that type 1 diabetes is not a disease of genetic predestination, and does not inevitably lead to beta cell destruction. These include its rising incidence, suggesting non-genetic (and therefore avoidable) influences; the observation that type 1 diabetes in a monozygotic twin affects the co-twin in no more than 30–50% of instances, the identification of a prolonged latent period between the induction of islet-specific autoimmunity and clinical onset of disease (implying the existence of immunoregulatory mechanisms), and the observation that long term survival of functional beta cells is both possible and clinically relevant following the diagnosis of type 1 diabetes.

Attempted intervention at diagnosis began in the 1980s, with claimed partial preservation of beta cell function with prednisolone, the ciclosporin trials in 1985–6, and a trial of azathioprine combined with prednisone in 1988. The unwanted effects of such therapies outweighed their benefits. The 1990s saw huge interest in the possibilities of antigen-specific therapies, raising the hope of harmless 'magic bullet' manipulation of the immune system, and also saw the launch of major trials designed to delay or prevent progression to clinical diabetes in high risk relatives.[1] [2] These hopes were disappointed. The 2000s saw a more pragmatic approach which has allowed a wide range of therapies to be tested in the newly-diagnosed, and the first commercial trials of immune intervention in type 1 diabetes.[3]

Most of the ongoing clinical trials of immunotherapy in human type 1 diabetes are co-ordinated by TrialNet, and their list of publications can be accessed at www.diabetestrialnet.org/publications/publications.

Non-specific immunosuppression

Ciclosporin A seemed suitable for use in type 1 diabetes in the 1980s, based on its benefits in organ transplantation and its efficacy in the recently introduced NOD mouse model. Pioneers of its use in diabetes included J-F Bach and Cal Stiller. The French[4] and Canadian–European[5] studies showed that ciclosporin reduced insulin requirements over the first 12–18 months of therapy, with the best results in those with the highest basal C-peptide and the highest trough levels of the drug. Problems included nephrotoxicity (largely reversible in this context) and the induction of insulin resistance, and any benefit was lost as soon as the drug was stopped.

Antigen-based therapies

Studies in the NOD mouse and other animal models showed that appropriate presentation of soluble beta cell antigens could restore immune tolerance to these and 'bystander' antigens in the islets. This promising strategy is however generally ineffective in the mouse once diabetes has become established. Insulin is a safe and specific islet autoantigen, and trials of injected, oral and nasally administered insulin have been performed in high risk or newly diagnosed individuals. These were uniformly negative, except in a subgroup of high-risk relatives with very high insulin autoantibody levels in the oral insulin trial; this indication is under further investigation. An altered peptide ligand based on a modified sequence 9–23 in the insulin B chain has also been tested unsuccessfully. Other islet autoantigens tested include the p277 peptide of heat-shock protein hsp60, and vaccination with GAD alum;[6] once again, with little or no benefit.

Antibody therapies

A range of monoclonal antibodies have been developed which target receptors on immune effector cells, including CD3, CD20 and CD25 (daclizumab). The CD3 complex 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. Anti-CD3 antibodies can thus achieve targeted elimination of activated immune effector cells. OKT3, a murine anti-CD3 antibody, was 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. Humanised anti-CD3 molecules were therefore engineered to overcome these problems by removal of antigenic sequences and by use of a human Fc tail modified to prevent binding with the Fc receptor. The resulting antibodies are only weakly mitogenic, and does not produce the full cytokine release syndrome. These antibodies include otelixizumab, teplizumab and visilizumab. A commercial Phase III trial of otelixizumab was however aborted in 2011 due to lack of efficacy.

An alternative approach has been to use anti-CD20 antibodies, which selectively deplete B lymphocytes. A clinical trial with rituximab showed partial preservation of beta cell function over 12 months of therapy in the newly diagnosed.[7] This observation suggests that B lymphocytes may play a greater role in the pathogenesis of type 1 diabetes than previously believed.

Cytokines

Cytokines are protein mediators of inflammatory and immunoregulatory responses released by most types of nucleated cell under stress. They fulfil autocrine, paracrine and endocrine functions, and act as central mediators of innate and adaptive immune responses as of tissue damage, defence and repair. Interleukin-1 beta is a master pro-inflammatory cytokine, and IL-1 blockade is currently in trial in type 1 diabetes,[8] and should report shortly.

Summary

This brief overview summarises a massive research effort extending over three decades. The positive knowledge gained is that it is possible to modulate beta cell function by a range of interventions in newly diagnosed patients; the bad news is that all benefits observed are transient and superimposed upon declining beta cell function. Transient benefits of immune intervention
Transient benefits of immune intervention

With the benefits of hindsight it can be seen that chances of aborting a mature immune onslaught upon the beta cell by manipulation of isolated elements of the immune effector system were always going to be slender. Other limitations have included: 1. Ignorance of the mechanisms or environmental factors responsible for activation of immune mechanisms directed against the pancreatic islets in man. 2. Inability to monitor early changes in the human pancreas, and consequent over-reliance upon analogies with the NOD mouse. 3. The paramount requirement for safety when intervening to prevent a non-lethal condition with a steadily improving long-term prognosis. 4. The need to restrict interventions to those who already have established hyperglycaemia, whereas intervention earlier in the disease process would offer a greater potential margin of benefit. 5. Lack of surrogate measures of cellular immune activation or beta cell mass. 6. The high cost and resource implications of testing multiple potential interventions in human volunteers. 7. The risk inherent in testing relatively small numbers, including false positive observations (type 1 error), together with the risk of missing true benefits (type 2 error).

These are considerable challenges, but should be set against the enormous increase in knowledge that has been acquired in the effort to overcome them. In the words of Walt Kelly, 'we are confronted with insurmountable opportunities', but these will eventually be realised.

References

  1. ^ Staeva-Vieira T et al. Translational mini-review series on type 1 immune-based therapeutic advances for type 1 diabetes. Clin Exp Immunol 2007;148:17–31

  2. ^ Rewers M, Gottlieb P: Immunotherapy for the prevention and treatment of type 1 human trials and a look into the future. Diabetes Care 2009;32:1769–82.

  3. ^ Bach J-F, Chatenoud L. A historical view from 30 eventful years of immunotherapy in autoimmune diabetes. Semin Immunol 2011;23:174–81

  4. ^ Feutren G et al. Cyclosporin increases the rate and length of remissions in insulin-dependent diabetes of recent onset. Results of a multicentre double-blind trial. Lancet 1986;2(8499)\;119–24.

  5. ^ The Canadian-European Randomized Control Trial Group. Cyclosporin-induced remission of IDDM after early intervention. Association of 1 yr of cyclosporin treatment with enhanced insulin secretion. Diabetes 1988;37:1574–82.

  6. ^ Wherrett DK et al. Antigen-based therapy with glutamic acid decarboxylase (GAD) vaccine in patients with recent-onset type 1 a randomised double-blind trial. Lancet 2011;378(9788):319–27

  7. ^ Pescovitz MD et al. Rituximab, B lymphocyte depletion, and preservation of beta-cell function. New Engl J Med 2009;361:2143–52

  8. ^ Pickersgill LM, Mandrup-Poulsen TR. The anti-interleukin-1 in type 1 diabetes action trial – background and rationale. Diabetes Metab Res Rev 2009;25:321–4

Comments

Nobody has commented on this article

Commenting is only available for registered Diapedia users. Please log in or register first.