Despite undetectable HIV in blood, HIV-derived peptides persist in the cerebrospinal fluid of people with cognitive impairment, suggesting the brain harbors a viral reservoir that may drive neurological damage.
Brain reservoir persistsHIV RNA and viral peptides were found in cerebrospinal fluid of HAND patients despite undetectable plasma virus and ART levels exceeding therapeutic thresholds
What the researchers found
Despite effective antiretroviral therapy (ART) that suppressed HIV to undetectable levels in blood, HIV RNA and HIV-derived peptides (from Env and Pol proteins) were still detectable in the cerebrospinal fluid (CSF) of people with HIV-associated neurocognitive disorders (HAND). This was found even though ART drug concentrations in the CSF exceeded the levels needed to suppress active HIV replication.
Critically, although HIV RNA and peptides were present in the CSF, the virus could not establish productive infection when tested in permissive immune cells — suggesting the brain harbors latently infected cells that express viral proteins without producing infectious virus. This non-replicative viral expression, rather than active infection, may drive the neuroinflammation and cognitive decline seen in HAND patients on ART.
Why it matters
Up to 50% of people with HIV develop some form of neurocognitive impairment even when their treatment successfully suppresses the virus in their blood. This study helps explain why: the brain acts as a viral reservoir where HIV can persist and produce inflammatory peptide fragments despite adequate drug levels. Understanding this mechanism is essential for developing therapies that can protect the brain — a challenge peptide-based and targeted drug delivery approaches may eventually help solve.
The numbers in context
n=24 · 10 cognitively normal + 14 with HAND · HIV RNA undetectable in plasma but present in CSF · HIV-derived peptides (Env, Pol) found in HAND samples only · ART concentrations exceeded IC50 in CSF · No productive infection from CSF virus
How the study worked
Cross-sectional study of 24 ART-treated people with HIV, stratified as cognitively normal (n=10) or having HAND (n=14, including ANI, MND, and HAD subtypes). Paired plasma and CSF samples were analyzed for HIV RNA (by RT-ddPCR), ART drug levels (by LC-MS/MS), and peptide profiles (by mass spectrometry peptidomics). Viral infectivity was assessed using viral outgrowth assays on permissive immune cells.
Who was studied
24 people with HIV on effective antiretroviral therapy, including 10 cognitively normal and 14 with HIV-associated neurocognitive disorders
What this study cannot tell us
Very small sample size (24 total, only 14 with HAND) limits statistical power and generalizability. Cross-sectional design cannot determine whether CSF viral persistence causes cognitive decline or is simply associated with it. The HAND subtypes (ANI=3, MND=9, HAD=2) had very small numbers for subgroup analysis. Peptidomic profiling identified HIV-derived peptides but could not quantify their levels precisely or determine their biological activity.
How to read the evidence
This is a well-designed observational study using advanced analytical methods (RT-ddPCR, mass spectrometry peptidomics, viral outgrowth assays), but the very small sample size (n=24) and cross-sectional design limit the strength of causal conclusions.
When this study was published
Published in 2026. This is very current research on one of the most active areas of HIV science — understanding and eliminating viral reservoirs in the central nervous system.
The bigger picture
The HIV reservoir problem — pockets of latent virus that persist despite treatment — is the central obstacle to an HIV cure. The brain is one of the most challenging reservoirs because it's protected by the blood-brain barrier, which limits drug penetration. This study shows that even when enough drug gets through, latent HIV in the brain can still produce viral peptides that may cause neuroinflammation. Addressing this will likely require novel peptide-based or nanoparticle delivery systems that can cross the blood-brain barrier and target latently infected cells.
Questions still open
- Can the HIV-derived peptides found in CSF directly cause neuroinflammation, or are they just markers of latent infection?
- Would intensified ART regimens or latency-reversing agents reduce HIV peptide expression in the brain?
- Could peptide-based drug delivery systems that cross the blood-brain barrier help clear this CNS viral reservoir?
Common questions
If HIV treatment is working, why do people still get brain problems?
Can current HIV drugs reach the brain?
Read the original research
HIV expression persists in the cerebrospinal fluid of HIV-associated neurocognitive disorders despite effective ART.
Emerging microbes & infections, 15(1), 2616945
Citation
Prates, Gabriela S; Li, Xiaoyi; Folgosi, Victor; Shabangu, Ciniso S; Souza, George G; Apoliano, Carlos; Gascon, Maria R; Monteiro, Mariana A; Gualqui, Carolina; Santos Eichler, Rosangela; Gomes, Helio; Katuwal, Nikesh; Gilbert, Cassandra; Tang, Yuyang; Casseb, Jorge; Jiang, Guochun. (2026). HIV expression persists in the cerebrospinal fluid of HIV-associated neurocognitive disorders despite effective ART.. Emerging microbes & infections, 15(1), 2616945. https://doi.org/10.1080/22221751.2026.2616945