Abstract Body

Background

Signs of CNS injury in PWH can be detected as early as primary HIV infection (PHI = lab-confirmed HIV acquisition <12 months) prior to ART.  It is unknown whether neuronal injury is ongoing during early untreated HIV, and/or declines post-ART. In a longitudinal CNS study of participants enrolled in PHI, we modeled longitudinal trajectories of CSF and plasma neurofilament light chain (NfL), a sensitive biomarker of active neuronal damage, and examined associations with lab and neuropsychological (NP) parameters in early infection.

Methods

NfL concentrations were measured using single molecule array (SIMOA) assay in longitudinal CSF and plasma samples from the observational PISCES PHI study (2004-2014, San Francisco), in which some participants chose to delay ART. Linear-mixed models were employed to determine CSF and plasma NfL trajectories pre- and post-ART, adjusting for longitudinal HIV RNA levels and baseline age. Associations were assessed in pre- and post-ART longitudinal NfL levels between CSF and blood, and parameters including age, CSF viral load (VL), and NPZ-11, a total summed z-score from an 11 test NP battery done at each visit.

Results

At baseline PHI participants (n=42, 100% male) were median 39 years of age (IQR:31-46) and had estimated duration of infection 126 (75-184) days; 97.6% were ART-naïve. The 166 CSF and 208 plasma samples assayed reflected data from >2 (range:1-8) visits pre- and post-ART with >6 months follow-up intervals. 98% of participants had suppressed plasma VL post-ART. Overall CSF NfL increased with age (p=0.003) while plasma NfL increased with both age and plasma VL (p=0.004 and p=0.019). Plasma NfL strongly correlated with CSF NfL (p=0.002). CSF NfL tended to increase pre-ART, but this was not statistically significant (slope=0.02, p=0.44). There was no significant change in CSF NfL post-ART, nor difference between these pre- and post-ART slopes (see Figure). No significant changes in NfL levels were observed in plasma. CSF and plasma NfL levels did not associate with NPZ-11 scores pre-ART, while plasma NfL associated with a 5.1 decrease in NPZ-11 score (p<0.001) post-ART. 

Conclusions

In this longitudinal data from a PHI cohort, CSF NfL did not decline over time either pre- or post- ART. These findings suggest the presence of ongoing neuronal injury throughout early infection. Further studies with longer follow-up are needed to understand the clinical significance of neuronal injury observed in PWH in early infection.