Endothelial progenitor cell (EPC) level represents a surrogate marker of cardiovascular risk and an indicator of the ongoing vascular damage. Moreover, EPCs are involved in the pathogenesis of virtually all diabetic complications. Therefore, ways to modulate EPCs are currently considered of utmost importance, especially in high-risk subjects. While many drugs with pleiotropic vasculoprotective effects have shown ability to positively modulate EPCs, there is no data on the effects of specific insulin formulations.

This is a human randomised cross-over comparison trial. The purpose is to compare the effects of two basal insulin analogues (detemir and glargine) added to oral antidiabetic therapy in poorly-controlled type 2 patients with cardiovascular disease on endothelial function and EPC levels.

The aim is to test whether optimized glycemic control with add-on basal insulin analogues improves endothelial damage and regeneration in type 2 diabetes with macroangiopathy and to compare the effects of glargine vs detemir on markers of endothelial damage and regeneration.

EPC level is the most innovative outcome measure of this study and represents the primary endpoint. Endothelial dysfunction/damage, evaluated using soluble markers, will be the secondary outcome. Given the supposed inverse correlation between EPC and endothelial damage, it is expected that EPC increase reflects amelioration in endothelial biology, a result that may have significant clinical implications in this cohort of high-risk patients.

Diabetes mellitus is associated with a strikingly high incidence of cardiovascular events. This is likely attributable to the widespread vascular damage due to uncontrolled hyperglycaemia and associated risk factors. Endothelial dysfunction is universally recognised as the first step in the natural history of atherosclerosis. Therefore, efforts are deserved to the understanding of its pathogenesis and to new strategies to prevent or limit its development. High glucose concentration is one of the most detrimental factors that negatively affect endothelial cell function. Through the classical damage pathways (including oxidative stress, PKC and MAP kinase activation and AGE accumulation), hyperglycaemia hampers the homeostasis of the vascular endothelial layer. Besides favouring endothelial cell damage, hyperglycaemia also prevents endothelial regeneration by negatively modulating endothelial progenitor cells (EPCs) [1]. EPCs are bone marrow-derived cells capable of differentiating into mature endothelial cells and repopulating the damaged endothelial layer, thus contributing to maintain endothelial homeostasis [2]. Consistently, the EPC level is directly correlated to measured of endothelial function [3]. The pool of circulating EPCs is reduced in the presence of cardiovascular risk factors, and EPC level is considered a biomarker of vascular health, which is negatively associated with measures of cardiovascular risk and with the extent of vascular damage [4]. Most importantly, EPC depletion predicts cardiovascular events independently of risk factors and of other relevant hard measures, such as left ventricular function [5]. In fact, a number of clinical and experimental studies indicate that alterations in EPCs contribute to the pathogenesis of cardiovascular disease and of virtually all diabetic complications [6]. In this light, ways to expand the EPC pool are actively pursued: for instance, modulation of EPCs has been studied to demonstrate the pleiotropic effects of vasculoprotective drugs, such as statins, glitazones and ACE inhibitors (reviewed in [7]. Diabetes is associated with a profound depletion of circulating EPCs, which is related to the metabolic control (both HbA1c and fasting glucose), adiposity and concomitant risk factors [8, 9]. Short and long-term glycaemic control is probably one of the most important determinant of the EPC pool in diabetes. Actually, advanced glycation end-products (AGEs) have been shown to exert detrimental effects on EPCs in vitro [10]. We have also demonstrated that number and function of EPCs provide a valuable marker of vascular damage in both healthy subjects and in type 2 diabetic patients [11, 12]. Interestingly, a mild EPC depletion is also present in pre-diabetic states [13], an observation that strengthens the link between glucose metabolism and EPC biology. Correction of hyperglycaemia with insulin has the potential to restore the circulating EPC pool, likely through mobilisation from bone marrow. In fact, we have shown that experimental hyperglycaemia (streptozotocin-induced diabetes) impairs ischemia-induced EPC mobilisation, which is corrected by basal insulin therapy [13]. The favourable effects of antidiabetic therapy intensification has been substantiated in a short-term study in humans [14], but the mechanisms responsible for this effect are incompletely understood. One putative mechanism is stimulation of nitric oxide production inside the bone marrow microenvironment. This is in compliance with one recent observation of ours that insulin indeed stimulates production of NO [15]. In the bone marrow, this is followed by activation of MMP9, cleavage of membrane-bound c-kit ligand, which, in turn, weakens cell-matrix interactions and triggers EPC mobilisation [16]. It is currently unknown whether this phenomenon is mediated by insulin per se or by reduction of plasma glucose, and whether there exist differences among insulin formulations. Likely, to act on the central EPC compartment (bone marrow), insulin needs to reach the marrow microenvironment at sufficiently high concentrations, a goal which could be achieved preferentially with specific insulin analogues, such as insulin detemir.

It has been suggested that weight gain, which follows initiation of insulin therapy, may exert negative effects on major cardiovascular outcomes. Related to this notion, not only the EPC pool reflects the global cardiometabolic profile, but it is also negatively related to visceral adiposity [4]. In a mouse model of obesity and diabetes, generation and function of EPC are profoundly altered [17], while it has been suggested that leptin and other adipokines may mediate this effect [18]. Hence, the net modulation of EPCs by insulin therapy may result from the balance between marrow mobilisation and the negative impact of increased adiposity. Therefore, the use of a weight-neutral insulin, such as insulin detemir, may provide additional benefits in increasing the EPC pool.

Study design Randomised cross-over trial comparing insulin detemir and insulin glargine added to current oral antidiabetic regimen in poorly-controlled (HbA1c≥8.5%) type 2 diabetic patients with cardiovascular disease.

The crossover design will allow to control for previous treatment and provide a more thorough comparison between the two treatments under investigation.

Recruitment and randomization will be performed during 6 months.

1. Basal measurements (at time zero)
2. Randomization to receive insulin detemir or glargine (at time zero)
3. First 3 months of treatment
4. Ad interim measurements (at 3 months)
5. Shift to the other insulin for 3 months
6. Final measurements (at 6 months)Initial results will be available between 6 and 12 months from the beginning of the study, depending upon time needed for recruitment.

Drop-out in case of acute illnesses or infection, acute cardiovascular events, or hospitalisation during the study period. Expected drop-out rate <10%.

Treatment protocol A protocol similar to that described by [19] will be implemented. Once-daily subcutaneous insulin detemir or insulin glargine will be added to current oral glucose lowering drugs. Doses of oral agents will remain unchanged during the study period. Based on self-measured fasting plasma glucose levels (average records from 3 consecutive days), insulin doses will be titrated, aiming at fasting concentrations of <110 mg/dl. Starting daily dose will be 10 U and then titrated individually by clinic or telephone contacts on a weekly basis, using the algorithm described in [19]. HbA1c will be measured at the end of the 3-month treatment period. Shifting from the one to the other insulin regimen, current daily insulin units will be maintained and then re-titrated as necessary

Inclusion Criteria:

• Type 2 diabetes
• Macroangiopathy (coronary, or peripheral, or cerebrovascular)
• On oral antidiabetic therapy
• HbA1c > 7.0%

Exclusion Criteria:

• Type 1 diabetes
• Acute diabetic decompensation
• Use of glitazones
• Cancer
• Acute disease or infection
• Chronic renal failure (serum creatinin > 2.0 mg/dl)
• Advanced liver disease (Child B-C)
• Immune disease, organ transplantation, immunosuppression
• Recent surgery (within 3 months)
• Recent cardiovascular events (within 3 months)
• Inability to provide informed consent
• Pregnancy and lactation