Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a circulating serine protease widely expressed in the liver that is involved in the regulation of blood cholesterol hemostasis and low-density lipoprotein (LDL) receptor (LDLR) degradation on hepatocytes, thus consequently inducing increased circulating LDL levels (1). This increase may represent an important risk factor for hypercholesterolemia, cardiovascular diseases (CVD), atherosclerosis, coronary artery disease (CAD) and stroke (2).

Interest in the role of PCSK9 in the regulation of LDL metabolism and in the pathogenesis of related diseases is increasing over the years, and previous studies have already demonstrated that PCSK9 gain-of-function mutations may be associated with autosomal dominant hypercholesterolemia and premature atherosclerosis (3), whereas loss-of-function mutations may lead to LDL level reduction and therefore may be protective against CV events (4). Although the role of PCSK9 in the liver is well defined, extrahepatic action was also observed; in fact, high levels of PCSK9 were detected in the gastrointestinal tract and in the kidney, macrophages, endothelial cells (ECs) and vascular smooth muscle cells (VSMCs). Preclinical studies have shown that increased PCSK9 levels may induce proinflammatory gene expression and apoptosis, thus promoting endothelial dysfunction far beyond LDL metabolism regulation. As a result, an increase in PCSK9 may play a direct role in the progression of atherosclerotic lesions, whereas its inhibition may be protective, with additional pleiotropic effects (5, 6). In patients with ACS and CAD, elevated plasma PCSK9 levels were found to be independently linked to inflammatory indicators such fibrinogen levels, high sensitivity C-reactive protein (hs-CRP) levels, and white blood cell count (WBCC) (7). Furthermore, it has been discovered that PCSK9 increases the synthesis of proinflammatory cytokines; for example, the PCSK9-induced enhancement of the inflammatory response may be mediated via the stimulation of the TLR4/NF-κB signaling pathway (8). Moreover, PCSK9 might seem to have an antithrombotic effect through platelet function and blood coagulation modulation (9). Data on the physiological role of PCSK9 in glucose metabolism and renal function are controversial, but a significant variable association between plasma levels of PCSK9 and dysregulation of glycemic control or worsening of kidney impairment cannot be excluded (5). Furthermore, PCSK9 concentration may be influenced by estrogens: high levels may decrease PCSK9 concentration with the consequent increase of LDLRs expression in liver (10). In fact, a significant increase in PCSK9 was observed in post-menopausal women as a consequence of the decrease in estrogen levels (11).

Circulating PCSK9 appears to be produced mainly by the liver, and its expression is regulated by numerous factors, such as thyroid hormone and thyroid replace therapy (12), diet (13), endogenous insulin and therapeutic exogenous insulin (14), resistin (15), the diurnal rhythm (16), various cholesterol-lowering drugs (17) and exercise (18). It seems that sex could modify the effects of extrinsic and intrinsic factors on the PCSK9 concentration.

The relationship between traditional CV risk factors and CV biomarkers has been well studied, but the impact of PCSK9 levels on CV risk has yet to be implemented, especially in the context of primary prevention. In particular, the evaluation of surrogate and prognostic markers of CV risk pointed out a possible correlation between arterial stiffness measured by pulse wave velocity (PWV) and atherosclerosis (19) considering the analogous underlying mechanisms in plaque formation and arterial stiffening, PCSK9 accumulation in atherosclerotic plaques could also be associated with arterial wall remodeling (20). In this context, pulse pressure (PP) may also be considered a risk factor for arterial stiffness and assumes a predictive role in CVD mortality (21, 22). A negative association between PCSK9 and PP was described in normotensive female patients (23), but recently, a correlation with high levels of PCSK9 values and PP was detected in diabetic patients (24). Additionally, the augmentation index (AI), which is used as a measure of wave reflection and arterial stiffness, showed a significant linear positive correlation with PCSK9 levels both in obese patients and in patients with familial dyslipidemias (25, 26).

Different studies have already evaluated the role of PCSK9 in CV endpoints only according to sex or a specific disease in secondary prevention (24, 27). However, primary prevention in the context of CVD is an important tool for healthcare to modify potential risk factors as soon as possible and reduce the risk of CV events, especially in post-menopausal women. Therefore, the aim of the present study was to observe the correlation between PCSK9 levels and CV parameters in post-menopausal diabetic women in primary prevention. Factors associated with PCSK9 levels in the same cohort of patients were also investigated as secondary objective.

Proprotein convertase subtilisin kexin type 9 (PCSK9) is a proprotein convertase that increases plasma low-density lipoprotein cholesterol (LDL-C) levels by triggering the degradation of LDL receptors (LDLRs). PCSK9 is linked to coronary plaque inflammation and has direct atherosclerotic effects on the vascular wall in addition to controlling the amount of LDL-C in the blood.

The results of our study show that diabetic women with late-onset menopause have a reduction of PCSK9 levels; furthermore, PCSK9 concentration was inversely correlated with aortic stiffness, indirectly measured with PWV, thus letting us to hypothesize that PCSK9 might represent a predictive marker of arterial stiffness and CV risk in this specific subset of patients.

Predictive factors of PCSK9 levels and PCSK9 influence on CV outcomes were, for the first time, assessed in post-menopausal diabetic women in primary prevention. The discovery of PCSK9 and the consequent approval of PCSK9 inhibitors has radically changed the management of patients at CV risk (31). In fact, PCSK9 has been recognized as a biomarker of CV risk in primary and secondary prevention by recent epidemiological studies (24, 27, 32), although its role has to be fully investigated in different cohorts of patients at CV risk, such as post-menopausal diabetic women.

In the present study, median serum PCSK9 levels were slightly higher than those observed in previous studies (24, 33) probably due to the characteristics of the enrolled patients. Indeed, several studies confirmed higher PCSK9 levels in patients with type 2 diabetes mellitus (T2DM) than the general population (34, 35) as well as in women than in men (11, 33). Moreover, higher baseline PCSK9 levels were observed in females with new diagnosis of T2DM than in those early affected by T2DM (36). Differences in PCSK9 concentrations were also detected between post-menopausal and pre-menopausal women (11, 37), as a consequence of the estrogen decrease related to menopause (8). Indeed, high levels of estrogens significantly reduce PCSK9 levels with the consequent increase of LDLRs in liver (38). In addition, the increased levels of PCSK9 may reflect the reduction of PCSK9 clearance mediated by LDLRs reduction consequent to the decreased estradiol levels (39) following menopause, as previously mentioned. The relationship between estradiol and PCSK9 is affected by transcriptional and post-transcriptional mechanisms through an estrogen receptor α-mediated pathway (40) as well as through the G-protein estrogen receptor activation (41) and LDLR mRNA expression (42).

In our study, women in late-onset menopause stage showed significantly reduced PCSK9 levels compared to women in normal-onset menopause. Moreover, late-onset menopause stage was inversely related to PCSK9 concentration. This data could be explained by the prolonged exposure to physiological estrogens that have protective and beneficial effects (11). No significant differences were observed between women in early- and normal-onset menopause stage and this was in contrast to a previous study showed that PCSK9 levels were not affected by estrogen replacement therapy in postmenopausal women (p = 0.6). This data support the hypothesis that menopause affects PCSK9 levels independently of estrogen therapy (39), PCSK9 clearance and LDLRs (43). Nonetheless, sex hormones might influence PCSK9 levels in post-menopausal women affected by diabetes and an increased prevalence of diabetes was observed in early-onset menopause patients compared to women in normal-onset menopause (44).

In contrast with different studies (24, 36), LDL-C values were inversely related to PCSK9 concentration in our cohort of patients. Although a correlation of PCSK9 plasma concentrations and LDL-C was previously described, LDL-C changes do not necessarily result in a respective modification of PCSK9 levels (39). The mutual correlation between PCSK9 and LDL-C is the result of a complex series of events that influence their plasmatic levels (45). In fact, when LDL-C is high, LDLR number would increase, also due to PCSK9 reduction, thus augmenting, plasma LDL-C clearance. On the contrary, when LDL-C levels are low, more PCSK9 would be active, thus stimulating liver LDLR degradation and limiting LDL-C clearance (46). In a previous study, no association was observed between PCSK9 and LDL-C in diabetic patients (47), but an effect of estradiol on the inverse correlation between LDL and PCSK9 cannot be excluded. Indeed, estradiol plasma levels may influence the correlation between PCSK9 and LDL in women (43); a negative correlation between estradiol and LDL-C adjusted for PCSK9 confirmed estradiol effects on LDL-C independently of PCSK9 (8). Indeed, the link between low PCSK9 levels or its inhibition and LDL-C levels with diabetes suggests a more complex interaction as Mendelian randomization analyses were almost concordant in showing an increased risk of new-onset diabetes in patients treated in the FOURIER and ODYSSEY trials (48). Therefore, our results confirm the possible correlation between PCSK9 and LDL-C levels also in a population of post-menopausal diabetic women.

The altered LDLR pathway clearance of plasma LDL-C may be one of the main causes of hypercholesterolemia that may also occur for mutations of LDLR and APO-B100; also, PCSK9 gain-of-function mutations may be associated with hypercholesterolemia whereas loss-of-function mutations are responsible for lowered plasma LDL-C levels, thus decreasing CV risk. In this study, we found that ApoB, which is the LDL protein component that binds LDLR, was an independent predictor of PCSK9 concentration, and as ApoB increases, an increase of PCSK9 may be detected in post-menopausal women with diabetes. Post-menopausal diabetic women show an increase of PWV values, in fact both menopause and T2DM were significantly associated with increased PWV (49). The relationship between CV risk factors and endpoints showed that PCSK9 levels influence PWV in post-menopausal diabetic women, as already observed in previous studies that demonstrated this correlation in different cohorts of patients (24, 26, 34). PCSK9 inhibitors effectiveness in reducing PWV confirmed the obtained data which also suggest the possible role of PCSK9 as a promising biomarker of CV risk and as a predictor of atherosclerosis independently of menopause (50) and lipid profile (51).

FPG is directly related to PWV as reported in other studies (52). Post-menopausal women with T2DM show increased PWV values, and both menopause status and T2DM are associated with PWV (53). This condition reflects the decreased arterial elasticity observed in diabetic patients because of oxidative stress, inflammation and advanced glycation end products that affect vessel wall (49). The passive stretching of collagen fibers not only may be responsible for PWV but also for PP changes and data obtained from other studies showed that SBP is positively associated with PWV and PP (34), as confirmed by our results; on the contrary, DBP is indirectly associated with PWV and PP.

Among the other parameters associated with CV risk, BMI was inversely related to PWV although the relation between BMI, obesity, and PWV is still controversial. In a study conducted on healthy subjects, BMI was negatively associated with PWV when adjusted for age, blood pressure, and additional CV risk factors. These data could suggest a possible benefit of obesity to arterial stiffness explained by the obesity paradox mechanism (54) and the smaller aortic diameter observed in lean individuals with the consequent increase of PWV values could further confirm this weak negative association (55). In addition, an inversely association was found between HDL and PWV values: the relationship between the lipid profile and arterial stiffness was evaluated by different studies although the results are discordant. High HDL values were related to increased arterial stiffness in post-menopausal women (56), but PWV was inversely related to LDL values in a population-based studies (57) as well as an independent, inverse relationship between HDL levels and PWV was observed in a cohort of healthy post-menopausal women (58).

In conclusion, our study demonstrated that PCSK9 levels increase in post-menopausal women and, specifically, the late onset stage menopause might influence both PCSK9 levels and consequently CV risk. The reported results further support the important correlation between PCSK9 and CV risk related to diabetes and menopause, thus highlighting its potential role as an additional biomarker of CV risk in post-menopausal diabetic women. Moreover, the obtained results might allow us to hypothesize that PCSK9 might be considered as a therapeutic target in a future clinical scenario. However additional validation studies are needed to validate our results.