A-T is a lethal, chronic degenerative disease. Due to the improved treatment options in the recent years, hitherto largely unknown disease features such as endocrine dysfunction, liver disease and cardiovascular diseases are gaining in importance (12, 15, 36, 37). The present work clearly demonstrates the high rate of type 2 diabetes (17.9%) among post-pubertal patients.
Diabetes is one of the leading causes of death worldwide (38). It leads to a high cardiovascular risk, micro-angiopathy, dyslipidemia, nephropathy, neuropathy and repressed immune system (38). HbA1c values in the upper normal range indicate a high risk for later diabetes (39). In condition of A-T, IR and diabetes have rarely been investigated in clinical settings. In view of the comorbidities (e.g. malnutrition, neurological deficit and immunodeficiency), consistent diabetes therapy is of particular importance.
In the cytoplasm of the cells, ATM causes activation of the serine/threonine-specific protein kinase Akt in response to insulin. Akt is an important protein which participates in the signaling cascade for the inhibition of apoptotic signals (40). In response to insulin, protein translation is stimulated, and glucose uptake is controlled by GLUT 4 (24). In mice with a muscle-specific deletion in the GLUT 4 gene develop IR and glucose intolerance (41). Low ATM levels will therefore contribute to the development of IR and glucose intolerance in A-T via the down-regulation of Akt activity in muscle cells (24).
In 2000, the insulin signaling induced ATM-dependent phosphorylation of 4E-BP1 was reported (45). Ever since, the deficiency in the insulin and insulin-like growth factor 1 (IGF-1) axes has been demonstrated in the absence of ATM (20, 26). Apo E knockout mice without ATM protein showed increased IR and were prone to develop a metabolic syndrome (22).
Additionally, ATM is a regulator of adipocyte differentiation. In Atm-deficient mice lack of induction of C/EBPα and PPARγ, central transcription factors for adipocyte differentiation, as well a reduced fat mass were reported (42). Of course, fat mass is of particular importance for glucose metabolism and homeostasis. There was no significant difference when comparing fat mass of A-T patients to sex and age matched healthy controls in humans (2). Apparently, the significantly decreased lean mass is a major contributor to the disturbed glycemic control in A-T patients.
Apart from that, there have been few reports on endocrine abnormalities in A-T patients (37). While poor weight gain, stunting and delayed pubertal development have been characterized as a typical findings in A-T (2, 3, 19, 20, 37), abnormalities in glucose metabolism are hardly described as clinical manifestation (30–32). We have recently reported about liver involvement in A-T and dyslipidemia (12). In synopsis of lipid metabolism disorder and IR, A-T patients suffer from an incomplete metabolic syndrome with increased risk for cardiovascular events (15, 16, 43).
Due to better care, life expectancy of A-T patients has emerged over the last decades (44). Especially in the light of new treatment options such as bone marrow transplantation (45–47), dexamethasone treatment (48–50) and gene therapy (51–53) disease facets with manifestation in the later disease course should be screened and treated. According to our data, diabetes screening is indicated starting for the age of 12 years. HbA1c is an easy to obtain, inexpensive marker that can be used to evaluate individual courses and therapy response. However, OGTT is more sensitive in diagnosing IR than HbA1c and fasting glucose. This shows that the OGTT is still of value and confirms the current recommendation of the English CF society: HbA1c reflects glycemic control over a period of time. This may have some advantages over the OGTT, but the use of HbA1c as a screening for CF is often within the normal range despite an OGTT diagnostic of diabetes in CF (54). Taken these information into account, we truly believe that both measurements, HbA1c and OGTT, should be applied in A-T patients.
First line treatment for insulin-resistant diabetes is metformin (38). However, to our clinical experience, not all A-T patients respond to treatment with metformin. As has been shown in 2011, inhibition of ATM in rat hepatoma cell lines diminished the effect of metformin by reduced phosphorylation and activation of AMP-activated protein kinase (25). Additionally, the gene variant SNP rs11212617 at a locus that includes the ATM was proved to influence the glycemic response to metformin in type 2 diabetes (55). In line with these studies, Connelly et al. reported that the absence of ATM leads to dysglycaemia and IR with lower Matsuda index when compared to controls while performing an OGTT (28). Nevertheless, they could not show altered fasting glucose levels, insulin concentrations or insulinogenic index measurements (28).
In addition to that, it is important to consider the general condition of the patient with particular attention to the neurological status, body composition and independence in the patients’ everyday life. For instance, subcutaneous injections often present an insurmountable barrier to self-administration due neurological impairment [unpublished clinical observation]. Apart from clinical experience, research on endocrine and metabolic alterations in A-T is rare (15, 22, 37, 56). No guidelines for treatment of diabetes in this challenging patient group are available.
In case a patient does not respond adequately to metformin therapy, insulin treatment is recommended. However, subcutaneous insulin injections are not feasible for older A-T patients with considerable neurological deficit. There is a dilemma between the autonomy of patients and the necessary treatment. To improve compliance, a different treatment regimen with oral antidiabetic drugs such as repaglinide may be used in special cases (57). In the Frankfurt A-T cohort, one of our A-T patients with diabetes had poorly controlled serum glucose levels under treatment with insulin glargin. We initiated a treatment with repaglinide. Hereunder, with a very favorable side-effect profile, a good therapeutic success and at the same time excellent compliance was achieved.
Nevertheless, the beneficial effects of insulin as anabolic hormone should be taken into consideration when escalating diabetes therapy (58). Especially in malnourished patients, an amelioration of the nutritional status with weight gain could be achieved with insulin injections. The insulin/IGF-1 axis increases muscle mass and bone density and improves insulin sensitivity as well as enhancement of free fatty acid oxidations in the muscles. Also, it was shown recently that the IGF-1 pathway has beneficial effect on cardiovascular and cerebrovascular disease (59). However, insulin as anabolic hormone and growth factor may possibly increase the cancer risk in A-T patients (60).
However, this study has some limitations. Due to the retrospective design, we cannot provide a complete data set for the diagnosis of type 2 diabetes, since many patients of our national cohort are admitted to our center for routine care annually or even every second year only. Still, to our best knowledge, this is the first prospective study on diabetes in 31 A-T patients and confirms our retrospective analysis of longitudinal data sets of our national cohort. Due to the large number of cases, we think we have delivered reliable data that clearly demonstrate the need for an annual diabetes screening in patients ≥ 12 years.