Background
Heavily treatment-experienced (HTE) persons with HIV (PWH) are characterized by long infection and treatment history, resistant viruses, and a significant burden of concomitant diseases. Since these features may be associated with HIV replication in the Central Nervous System (CNS), this study aimed to assess plasma CNS tissue markers in HTE versus controls.
Methods
HTE cases were selected from the PRESTIGIO registry (Italian multicenter cohort enrolling PWH with documented four-class drug resistance) and stratified according to HIV RNA below (VS) or above 50 c/mL (VF) in their first sample in 2021-2023. Controls were consecutively treated > 50-year-old PWH with VS. PWH with ongoing CNS disorders were excluded. Plasma samples were analyzed by SIngle MOlecule Array (SIMOA SR-X, Quanterix Corp., Boston, MA, USA) for neuronal damage markers (neurofilament-light chain, NFL and total Tau protein), astrocyte activation (Glial Fibrillary Acidic Protein, GFAP) and ubiquitin-proteasome involvement (Ubiquitin C-terminal Hydrolase, UCH-L1).
Results
We included 84 controls, 106 HTE/VS and 32 HTE/VF (Table). NFL, tau and GFAP levels were significantly lower in HTE than in controls, with significant correlations between the four markers (all p<.001). Higher age correlated to higher NFL (p<.001), GFAP (p<.001) and UCH-L1 (p=.040); male sex with lower GFAP (p=.018) and tau (p<.001); and HIV acquired at birth with lower GFAP (p=.012). In HTE, lower tau was associated with fostemsavir (p=.049) and doravirine use (p=.037) and lower GFAP with darunavir (p=.009). In HTE/VF plasma HIV RNA inversely correlated to NFL (p=.048). Multivariate linear regression, corrected for age, sex, study group, body mass index (BMI) and significant variables at bivariate analyses, was performed to identify predictors of high levels of each marker. BMI (p=.040, 95%CI 0.045-1.814) and GFAP (p<.001; 95%CI 0.057-0.181) were independent predictors of NFL; CD4 nadir (p=.002; 95%CI 0.001-0.003), current CD4 (p=.009; 95%CI -0.002-0.000) and male sex (p=0.003; -1.199-0.255) of tau; age (p<.001; 95%CI 3.634-6.015), BMI (p=.009; 95%CI -4.904-0.704), male sex (p=.011; 95%CI -40.852-5.336) and HIV acquisition at birth (p<.001; 95%CI 36.643-135.900) of GFAP; and higher GFAP (p=.014, 95%CI 0.014-0.122) of UCH-L1.
Conclusions
Plasma CNS tissue marker levels do not support higher CNS injury risk in HTE-PWH. Several other variables seem independently and variably associated with higher levels of each marker.