Diabetes reflects the lifestyle and environment of the societies it afflicts, and the numbers affected by type 1 and type 2 diabetes increased dramatically over this period. Diabetes epidemiology came of age and helped to define the condition and to guide public policy towards it. The revolution in molecular biology allowed the genetic basis of both types of diabetes to be described, emphasising their complex and multifactorial nature, and monogenic diabetes was clearly characterized. The immunological background to type 1 diabetes was identified. Understanding of the pancreatic islets was transformed, although many mysteries remain, and biochemical understanding of the metabolic disorders associated with hyperglycaemia reached previously undreamed-of sophistication. Medical practice shifted from symptomatic to preventive management of diabetes, aided by availability of oral medication. Glucose control was shown to be important in the prevention of microvascular complications, as was treatment of associated cardiovascular risk factors in preventing heart disease. A patient-based approach to care based on self-empowerment and multidisciplinary team management emerged, and the technology of treatment greatly improved. The first diabetes prevention trials were undertaken, and pancreatic transplantation became feasible. Notwithstanding so much progress, the basic frustrations remained.
A Gallup poll conducted in the 1950s reputedly found that most Americans expected a cure for cancer by the end of the century, but few imagined that a man would walk on the moon. In somewhat parallel fashion, scientific and technical progress in biology has exceeded the imagination of most scientists at the mid-century, whereas our practical ability to defeat diabetes and other complex disease has lagged behind expectation.
Although Elliot Joslin (1870-1962) pioneered use of insurance tables to study the prognosis of diabetes, the epidemiology of diabetes did not really exist as an academic discipline in its own right until the 1960s, when it was founded by Kelly West (1925-1980) and others including Harry Keen in the UK. Epidemiology charted the effects of age, adiposity, ethnicity, social class and social environment upon the risk of diabetes.
It was assisted in this by pioneer studies such as those of Peter Bennett and colleagues with the Pima Indians of Arizona, the Whitehall Study in the UK, and many others around the world. Prospective studies in several populations showed that an approximate glycaemic threshold for the risk of microvascular complications could be established using the OGTT.
These changes were specific for diabetes and varied little between populations. In contrast, the risk of macrovascular disease showed a linear relationship with glycaemia extending well into the non-diabetic range, was non-specific for diabetes, and showed no clear threshold for risk. In consequence the definition of diabetes has endured, despite its limitations, whereas the threshold for intervention to prevent arterial disease remains controversial.
The heterogeneity of diabetes
Clinicians had long observed that diabetes pursued a more aggressive course in the young and thin than in older overweight individuals, who could often manage without insulin. Sir Harold Himsworth MD, FRS (1905-1993) distinguished between insulin sensitive and insensitive diabetes in the 1930s, but it was not until the 1970s that the terminology type 1 and type 2 diabetes was reintroduced, and it was not until the 1990s before if was formally adopted. See the Discovery of type 1 diabetes.
Joslin had appreciated that a strong family history of diabetes in young people was associated with a better prognosis, but it was the work of Robert Tattersall and Stephen Fajans in the 1970s that clarified the existence of the first form of monogenic diabetes, inelegantly titled "maturity onset diabetes of the young" or MODY.
Genetics of diabetes
The first half of the twentieth century saw many unsuccessful attempt to fit the inheritance of diabetes into the framework of Mendelian genetics. One suggestion was that people with juvenile diabetes carried two copies of the putative diabetes gene whereas those with maturity onset disease were heterozygous.
The HLA system associations of type 1 diabetes were the first genetic markers to be identified by Jørn Nerup and coworkers in 1974, and formed a key element in the paradigm shift that saw type 1 diabetes recognised as an autoimmune disorder. See Autoimmunity and diabetes
The quest for further diabetes genes was pursued via the candidate gene approach. The Insulin gene was the obvious one to try. This proved not to vary between those with diabetes and those without, but an upstream allele proved to be associated with type 1 diabetes. Linkage of type 2 diabetes and MODY to the glucokinase gene was established in 1992. Other genes followed slowly, but the real period of gene discovery followed the introduction of genome wide association studies (GWAS) in the following century.
The primary structure (amino acid sequence) of bovine insulin was established by Frederick Sanger in 1951, and the tertiary structure (three dimensional configuration) by Dorothy Hodgkin in 1969. This followed the recognition by Don Steiner in 1967 that insulin is stored in the form of proinsulin in the beta cell granules, and that C-peptide (connecting peptide) is generated in equimolar quantities when proinsulin is cleaved to produce insulin.
The way in which insulin produced its effect upon cells had been one of the great mysteries of diabetes. In 1966 it was demonstrated that its action was linked to a membrane bound fraction, a discovery which led to the elucidation of the structure of the receptor in 1985. Further work showed that the interaction of insulin with its receptor on insulin responsive tissues such as fat and muscle causes migration of the glucose transporter (GLUT4) to the cell surface and rapid influx of glucose.
Another great achievement was elucidation of the way in which islet beta cells transduced the stimulus provided by ambient glucose to release of stored insulin. Glucose enters the cell via the GLUT2 transporter, generating ATP, which closes the ATP-sensitive potassium channel (sulfonylurea receptor) in the cell membrane. Potassium accumulates within the cell, leading to membrane depolarisation, calcium entry and insulin release.
The insulin gene was synthesised in 1980, and the biosynthesis and marketing of human insulin soon followed. The first insulin analogues, designer molecules which released insulin more rapidly or over longer periods following injection, followed in the 1990s.
It had long been surmised that glucagon originated in the alpha cell, and this was finally confirmed by immunohistological staining in the 1960s. The structure of glucagon was deduced in 1956, and it was synthesised in the laboratory in 1968. It was marketed from the 1950s on as a treatment for hypoglycaemia.
Roger Unger (who also coined the term enteroinsular axis) noted the close juxtaposition of alpha and beta cells in healthy islets, the relative abundance of pancreatic alpha cells in type 2 diabetes together with inappropriate hyperglucagonaemia, and proposed a two-hormone model of diabetes. The importance of this observation was not well-recognised at the time, but glucagon inhibition became an important target of therapy by the end of the century.
Re-enter the incretins
As noted (History 1900 to 1950) the existence of gut factors promoting insulin secretion was recognised by the 1930s. By the 1970s it was recognised that the gut was a major endocrine organ, and Creutzfeldt revived the incretin concept in 1979. Glucose-dependent insulinotropic peptide (GIP)was the first incretin to be characterized, followed by glucagon-like peptide-1 (GLP-1).
GLP-1 based therapies were prefigured by observations showing that continuous infusion of this short-lived peptide enhanced insulin secretion and glucose control, prompting a search for analogues resistant to enzyme breakdown and for inhibitors of the DPP4 enzymes which are responsible. The incretins became an important addition to the treatment of diabetes in the following century.
Control and Complications
The debate about control and complications dates back to the 1930s but was greatly limited by the difficulty of quantifying glucose control, a problem largely solved by the introduction of HbA1c measurement and glucose self-monitoring in the late 1970s. When these became available the actual difference between patients treated by the advocates of tight vs looser control turned out to be less than previously suspected.
The argument centred for some years around the basement membrane controversy. This arose because Marvin Siperstein was able to argue (incorrectly) that the characteristic histological lesions of diabetic microangiopathy were absent in secondary diabetes, and that the lesions were therefore unrelated to glycaemia. The argument reached its climax in a famous trio of Editorials in the New England Journal in 1976, and was only resolved by publication of results from the Diabetes Control and Complications Trial (DCCT) in type 1 diabetes in 1993  and the United Kingdom Prospective Diabetes Study  in type 2 diabetes in 1998.
Not surprisingly, microvascular disease, which is glucose specific, proved to respond well to glucose-lowering therapy, whereas large vessels disease, which has a multifactorial aetiology, requires multifactorial therapy.
Glucose control plays a major role in the primary prevention of diabetic eye and kidney disease, but blood pressure is also important, particularly in accelerating the progression of established complications. This proved of major importance in slowing the progression of kidney disease, aided by the discovery that small amounts of protein in the urine (microalbuminuria) gave warning of incipient nephropathy at a time when this is still treatable.
Meanwhile a number of technical developments revolutionised the management of end-stage complications of diabetes. Laser treatment transformed the prognosis of proliferative retinopathy and lens implants greatly simplified the ease and improved the quality of cataract surgery. Dialysis and transplantation revolutionised management of end-stage renal disease, previously considered untreatable.
Although many of the pioneers had emphasised the need for people with diabetes to become their own doctors, paternalism prevailed until late into the century. Self-monitoring of blood glucose then empowered people to take charge of their own treatment, supported by the development of diabetes education programmes, introduction of a multidisciplinary team approach and a new focus upon diabetes centres. This centripetal trend was balanced by public health policies encouraging wider involvement in diabetes management at a primary care level.
Sharper needles became available - to the great relief of insulin users - and pen and pump devices became available for insulin delivery. Blood glucose strips and meters made glucose control easily visible. Implantable insulin pumps became available and but the reliable glucose sensors required to "close the loop" remained elusive.
The relationship between overnutrition and type 2 diabetes was made evident by observation of populations exposed to periods of wartime starvation, and was formally demonstrated in a series of trials of lifestyle or pharmacological intervention in individuals with impaired glucose tolerance. The problem was then to translate the results of these intensive interventions into effective public health strategies.
Insight into the immune basis of type 1 diabetes naturally resulted in an interest in the possibilities of immune intervention. This was greatly stimulated by trials demonstrating preservation of insulin secretion following diagnosis in newly diagnosed patients treated with cyclosporine. Adverse effects limited this approach to therapy, but other interventions were tested and the first large-scale trials of primary or secondary prevention were under way by the end of the century.
Progress with transplantation of kidneys and other organs prepared the way for transplantation of the pancreas. One limiting factor to whole pancreas transplantation was the need to implant the exocrine pancreas and arrange drainage for its secretions; another limitation was the need for immunosuppression. By the end of the century, after years of endeavour, this became a costly but practical solution for a few patients, usually attempted in those already requiring immunosuppression for a kidney graft.
Meanwhile, others pursued the goal of freeing the islets from the exocrine pancreas and implanting them into the body. This technique was pioneered successfully in people with some forms of pancreatic tumour, using islets from the unaffected part of the gland, and enabled them to live without the need for insulin injections following extirpation of the pancreas. By the end of the century islet transplantation became a viable treatment for diabetes.
^ Pastan I et al. Binding of hormone to tissue: the first step in polypeptide hormone action. PNAS 1966;56(6):1802-9
^ Unger RH. The essential role of glucagon in the pathogenesis of diabetes. Lancet 1975;1:1036-42
^ Creutzfeldt W. The incretin concept today. Diabetologia 1979;16:75-85.
^ Holst JJ. Glucagon-like peptide 1: from extract to agent. The Claude Bernard Lecture 2005. Diabetologia 2006;49:253-260
^ Cahill GF Jr et al. "Control" and diabetes. New Engl J Med 1976;294:1004-5
^ DCCT Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes. New Engl J Med 1993;329:977-86
^ UKPDS Group. Effect of intensive blood glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352(9131):837-53