New onset diabetes (NOD) has been regarded as one of the factors to be considered before deciding to prescribe a statin [1-
3]. The recent Canadian Network for Observational Drug Effect Studies Investigators  study and an older meta-analysis 
suggest that the administration of higher potency statins is associated with an increased risk of NOD compared with lower potency
statins [1-3]. The potential increase in caloric and fat intake during statin treatment may play a role in the pathogenesis of
statin-related NOD . However, the exact mechanisms involved in the pathogenesis of NOD associated with statin treatment
remain to be defined.
Recent evidence suggests that the prevalence of NOD is significantly lower in familial hypercholesterolemia (FH) patients
(n=14,296) compared with their unaffected relatives (n=24,684) . For receptor-negative and receptor-defective low density
lipoprotein receptor (LDLr) mutations, the odds ratio was 0.35 (0.27-0.45) and 0.51 (0.42-0.62) (p for trend <0.001), respectively,
compared with unaffected relatives . FH patients may be more motivated to follow lifestyle measures (diet, exercise
and not smoking) which may contribute to the reported decrease in the risk of NOD in patients with FH treated with statins.
Statins may cause NOD, especially in predisposed individuals, through several mechanisms [1, 6]. They may affect insulin
secretion, reduce translocation of glucose transporter 4 resulting in hyperglycaemia and hyperinsulinaemia, decrease intracellular
signalling through the reduction of several downstream products, and finally they may decrease leptin and adiponectin levels
. However, the correlation between statin intensity and risk of NOD has not been fully defined.
We could not find definitive information in the literature stating if activation of sterol regulatory element-binding proteins
(SREBPs) occurs in patients with FH as part of the attempt to increase hepatic LDLr numbers. In contrast, statins increase
LDLr numbers through activation of SREBPs 1a, 1c and 2 , which are also causally related to insulin resistance . Thus,
the more potent the statin, the greater the possible increase in SREBPs and LDLr as well as the reduction in plasma LDL cholesterol
(LDL-C); however, the “cost” may be in terms of insulin resistance and a higher incidence of NOD. Thus, it is imperative
to carefully balance the risk-benefit ratio [1, 9]. However, is this the case for antibodies against proprotein convertase subtilisin
kexin-9 (PCSK9), which prolong the life of the LDLr? These drugs have no apparent adverse effect on SREBPs . In
fact, a study using a novel antibody (1B20) against PCSK9 has shown that its combination with simvastatin in dyslipidaemic
rhesus monkeys reduced LDL-C more than either agent alone, while at the same time decreasing liver mRNA levels of SREBPregulated
genes . This mRNA effect remains to be confirmed in humans.
The phase III studies using antibodies against PCSK9 (e.g. evolocumab or alirocumab) have not reported any trend for increased
NOD [11, 12]; however the issue of NOD should be carefully monitored in these on-going trials. It took some 20 years
to realise that statin-induced NOD occurs. If there is a similar problem with PCSK9 inhibitors (potentially the next blockbuster
in hypolipidaemic treatment ) we should know as soon as possible. Moreover, we need to define the risk of NOD in patients
treated with a statin and PCSK9 inhibitor combination. It may be that by lowering the statin dose required to reach LDLC
targets, there will also be a decrease in NOD. This potential benefit may help justify, at least in part, the cost of using PCSK9
A recent study reported an off-target effect of cholesterol ester transfer protein (CETP) inhibitors, a new drug category
aimed at increasing high density lipoprotein cholesterol (HDL-C) levels and reduce atherosclerosis; these drugs are still under
investigation in phase III trials . CETP inhibitors reduce the levels of mature SREBP 2, leading to attenuated transcription
of hepatic LDLr and PCSK9 . Thus, it is possible that these drugs will not have the diabetogenic effects of statins.
We also need to consider the evidence that currently available lipid lowering drugs, like ezetimibe and colesevelam, may
also have beneficial effects on insulin sensitivity. However, the evidence for ezetimibe comes from relatively small studies [15-
19] with design limitations and the results were not always consistent . This may reflect which patient category was studied.
It is also of interest that ezetimibe down-regulated liver SREBP-1c and improved hepatic insulin resistance in mice fed a
high-fat diet although there were also effects on other mediators . Colesevelam is actually licensed (in some countries) to
lower glucose levels although the mechanisms involved are not clear .
The above evidence suggests that ezetimibe and colesevelam as well as two new drug categories (i.e. PCSK9 and CETP
inhibitors) for the treatment of dyslipidaemia might decrease/avoid the risk of NOD. This will hopefully lead to combinations with statins of lesser intensity or at lower doses which may result in a reduced risk of statin-related NOD. The key issue is
whether such combinations will be as effective as statin monotherapy in terms of lowering the risk of vascular events.
Dept. of Clinical Biochemistry (Vascular Disease Prevention Clinics) Royal Free Hospital campus University College London Medical School University College London (UCL) Pond Street, London NW3 2QG, UK.