Rumours of beta cell death in type 1 diabetes are exaggerated

18 December 2013 (09:08 AM) - Editor's Corner

Type 1 diabetes has long been considered, indeed defined, as a condition of absolute beta cell deficiency, but evidence has accumulated over many years to suggest that this is not entirely true. One line of evidence came from measurement of residual C-peptide secretion, and the studies of Faber and Binder, which date back to the 1970s, showed that 15% of patients with type 1 diabetes of 15-35 years in duration had measurable levels of endogenous insulin secretion[1]. More sensitive assays for C-peptide confirmed this, and went on to identify measurable insulin secretion in a much greater proportion of patients. A study from Exeter in the UK using state-of-the-art assays now shows that 49/74 patients (66%) with a diabetes duration >5 years had detectable fasting levels of C-peptide, and responded to stimulation: 54 (73%) had enhanced (n=43) or stable (n=11) C-peptide secretion following a mixed meal. Furthermore, although insulin secretion tended to diminish with increasing duration of diabetes, 25/37 patients with a diabetes duration >30 years were still producing insulin [2].

A second line of evidence comes from histological examination of pancreata from people with long term type 1 diabetes. Early studies suggested persistence of normal-looking islets, but definite proof awaited histochemical demonstration of insulin staining in the islets and greater clinical awareness of the distinction between type 1 and type 2 diabetes. By the 1980’s, however, it became increasingly clear that apparently intact insulin-containing beta cells were present in a proportion of people with long-standing diabetes. These lines of evidence came together in physiological and post-mortem studies of Joslin medal winners, showing ongoing insulin secretion and histological evidence of insulin-containing beta cells after 50 years of diabetes [3].

What does this mean? In an accompanying Commentary, Denise Faustman muses about the long-term reluctance of diabetes investigators to come to terms with the implications of these studies [4]. She suggests that insensitive functional assays and the misleading example of the NOD mouse (which rapidly loses the capacity to secrete insulin) may have been responsible, to which we might add the concept of an ongoing immune-mediated onslaught on beta cell antigens – why should this stop short of complete destruction? Indeed, as she points out, persistence of beta cells helps to explain the persistence of autoimmunity in long-term diabetes. Does the presence of beta cells reflect ongoing formation of new beta cells by the pancreas? If so – and the case seems clear - the therapeutic implications are stunning. The received wisdom has always been that beta cells cannot regenerate: but if this is mistaken, it suggests that even people with long-standing diabetes have the potential to grow new beta cells of their own. The observation that the majority of people with type 1 diabetes continue to make measurable, albeit tiny, amounts of insulin has the potential to change the way we think about the disease, and even its cure.


  1. ^ Madsbad S et al. Prevalence of residual beta-cell function in insulin-dependent diabetics in relation to age at onset and duration of diabetes. Diabetes 1978;27 Suppl 1:262-4

  2. ^ Oram RA et al. The majority of patients with long-duration type 1 diabetes are insulin microsecretors and have functioning beta cells. Diabetologia 2014;57:187-191

  3. ^ Keenan HA et al. Residual insulin production and pancreatic β-cell turnover after 50 years fo diabetes: Joslin Medalist Study. Diabetes 2010;59:2846-53

  4. ^ Faustman D. Why were we wrong for so long? The pancreas of type 1 diabetic patients commonly functions for decades. Diabetologia 2014;17:1-3