Clinical, behavioral and biomarker predictors of PCSK9 levels in HIV-infected patients naïve of statin therapy: A cross-sectional analysis from the Swiss HIV cohort

Abstract Background and aims Better characterization of Proprotein Convertase Subtilisin/Kexin 9 (PCSK9) profile is currently needed to tailor appropriate lipid-lowering strategies in HIV patients. Methods HIV-infected individuals aged ≥ 40 years old and naive of statin therapy included in the Swiss HIV cohort study were screened for PCSK9 levels with a routine blood sample collection in 2014 at the Geneva University Hospitals. An exploratory linear regression model was built including clinical [age, sex, ethnicity, cardiovascular risk factors, body mass index, low CD4 defined as ≤200 cells/μl, leucocytes, lymphocytes, platelet, antiretroviral therapy], behavioral (tobacco and marijuana smoking, alcohol use and physical activity) and biomarker (CRP, TNF-α, IL-8, Il-10 and MCP-1) to investigate association with continuous PCSK9 levels. Results We studied 239 HIV-infected individuals who met inclusion criteria and available PCSK9 levels with a mean age of 49 years. 35 subjects (14.6%) reported marijuana consumption, of whom 20 (57.1%) reported daily consumption and 15 (6.3%) occasional use. PCSK9 levels were correlated with low-density lipoprotein-cholesterol (LDL-C). Our exploratory model identified marijuana consumption ( p= 0.023) and low CD4 values ( p= 0.020) as significantly associated factors with higher PCSK9 levels. No association was found with Framingham risk score. Patients with marijuana consumption had significantly higher levels of PCSK9 with a dose-response effect ( p 0.001); the association persisted after adjustment for the calculated Framingham risk score ( p= 0.003) and additional adjustment for clinical variables ( p= 0.027). Conclusions In HIV-infected individuals naive of statin treatment, marijuana consumption and low CD4 values are associated with higher PCSK9 levels independently of clinically relevant confounding factors.