Total word count (including tables and figure legends): 3927 Number of tables and figures : 4 ABSTRACT (Word count 250) Background and aims -PCSK9 is an endogenous inhibitor of LDL receptor pathway.Recently, Mendelian randomization studies have raised a doubt about the diabetogenic risk of PCSK9 inhibitors. Here, we assessed the relationship between plasma PCSK9 levels and the risk of new onset diabetes (NOD).Methods -Fasting plasma PCSK9 levels were measured at baseline by ELISA in subjects without lipid lowering treatment in IT-DIAB (n=233 patients with prediabetes, follow-up 5 years) and ELSA-Brasil (n=1,751; 27.5% with prediabetes, follow-up 4 years) prospective cohorts. The primary outcome in both studies was the incidence of NOD. The association of NOD with plasma PCSK9 levels was studied using multivariate Cox models.
Results -Plasma PCSK9 levels were not significantly associated with NOD in IT-DIAB (HR (+1SD) 0.96, CI95% [0.76; 1.21]) and ELSA-Brasil (OR (+1SD) 1.13 [0.89; 1.42]). In ELSA-Brasil, a significant positive association between PCSK9 and worsening of glucose homeostasis, including the progression from normoglycemia to prediabetes, was found (OR (+1SD) 1.17 [1.04; 1.30], P=0.0074). Plasma PCSK9 concentration was also positively associated with the change in fasting plasma glucose between the first and second visit in ELSA-Brasil (β = 0.053, CI95% [0.006; 0.10], P=0.026). Plasma PCSK9 levels positively correlated with total cholesterol in IT-DIAB and ELSA-Brasil, but not with glucose homeostasis parameters, except for a positive correlation with HOMA-IR in ELSA-Brasil.Conclusions -Plasma PCSK9 levels were not significantly associated with NOD risk in longitudinal analyses. These data suggest that inhibition of the PCSK9 extra-cellular pathway should not be deleterious for glucose homeostasis.