|Authors: Wagner GA, Chaillon A, Franklin DR, Caballero G, Kosakovsky Pond SL, Heaton RK, Richman DD, Smith D, and the CHARTER Group|
HIV-1 dual infection (DI) has been associated with decreased CD4 T-cell counts and increased viral loads. The same markers are also associated with the development of HIV-associated neurocognitive disorder (HAND), which continues to be a prevalent and debilitating problem among HIV-infected individuals in the era of antiretroviral therapy (ART). Despite this, the neurological sequelae of DI are poorly characterized. We used next-generation sequencing (NGS) to measure the prevalence of DI among ART-suppressed CHARTER cohort participants, and examine the relationship between HAND and DI.
CHARTER cohort participants were selected who had > 4 years of follow-up and were suppressed on ART (N=35). NGS (454 FLX Titanium, Roche) was performed on PBMC-derived HIV-1 DNA populations using four PCR-amplified coding regions (env C2-V3, gag p24, pol RT, and pol PR). Samples were classified as DI when (i) maximal genetic diversity exceeded previously optimized thresholds, (ii) phylogenetic reconstruction showed two divergent populations supported with bootstrap values >95%, and (iii) local BLAST ruled out contamination with in-house strains. All study participants underwent neurocognitive, substance use, and neuromedical measurements at each study visit. A global deficit score indicative of HAND was derived from a comprehensive battery of neuropsychological tests adjusted for demographics and corrected for practice effect over time. Participants with comorbidities confounding HAND were excluded.
Of 34 participants, 21 (62%) remained neurocognitively normal throughout the study period and 14 (38%) showed varied degrees of impairment. In univariate analyses, timepoints with HAND were associated with lower CD4 T-cell counts and with a shorter estimated duration of infection (p < 0.05). Despite ART, 27 participants had detectable HIV viral loads, but these levels were not significantly different between the normal and impaired groups (p = 0.866). Nine study participants had evidence of DI in at least one HIV coding region [prevalence of 26.5%, 95%CI 13.8%-43.1%]. Among the 13 participants with HAND, 7 (54%) had DI in at least one timepoint, while 2 out of 21 (9.5%) neurocognitively normal participants had evidence of DI (p < 0.05).
Using NGS, we identified a high (26%) prevalence of DI in a population of virally suppressed HIV-infected individuals. The presence of neurocognitive impairment was associated with DI, lower CD4 counts and with shorter duration of infection. These data suggest that DI may play an important role in the development of HAND among at-risk infected individuals, perhaps due to increased viral diversity, although the elucidation of proximate causes will require further investigation.|