Factors Associated with Persistently Increased IL-6 Levels in People with HIV.

Abstract

Individuals with HIV remain at higher risk of non-communicable diseases associated with interleukin-6-specific (IL-6) inflammation compared to the background population. Despite antiretroviral therapy, some individuals exhibit persistent hyperinflammation. We characterise these understudied longitudinal profiles and distinguish the predictors of chronic versus acute inflammation. A subset from the Strategic Timing of Antiretroviral Treatment (START) trial with up to seven years of biomarker follow-up was included. Inflammatory states were defined based on plasma IL-6 levels, measured at randomisation, months eight, 12, and annually until 84. Hyperinflammation was defined as IL-6>1.8 pg/mL; ≥2 consecutive elevated measurements defined persistent hyperinflammation and a single elevated bracketed by non-elevated measurements defined IL-6 blips. Variables associated with these outcomes were analysed by generalized estimating equations. Among 2,102 individuals with a median of 5 (IQR: 4-6) measurements, 861 (41%) had persistent hyperinflammation whereas 575 (27%) had blips. Participants with BMI≥40 versus 18.5 to <25 (OR: 5.49, 95%CI: 4.18-7.20) and females with black ethnicity versus white males (2.58, 2.17-3.06) had increased risk of persistent hyperinflammation but not blips. Other persistent hyperinflammation predictors included higher white blood cell (WBC) counts, HIV-1 RNA, total cholesterol to high-density lipoprotein ratio >5, older age, smoking and diabetes. Of these only higher WBCs were also associated with blips. Persistent hyperinflammation is strongly associated with demographic characteristics, comorbidities, and HIV-1 RNA, whereas blips were associated with markers of immune activation. Longitudinal IL-6 measurements are needed to distinguish persistent hyperinflammation from blips, which can lead to establishing a "high-risk phenotype" to target for clinical intervention.