Back HIV-Related Conditions Neurocognitive CROI 2016: Antidepressant Improves HIV-Related Cognitive Impairment

CROI 2016: Antidepressant Improves HIV-Related Cognitive Impairment


The SSRI antidepressant paroxetine (Paxil) was associated with a modest improvement in cognitive function and reduced central nervous system inflammation among people with HIV-related neurocognitive disorder, but the antifungal drug fluconazole showed no apparent benefit even though it reduced oxidative stress, according to a study presented at the Conference on Retroviruses and Opportunistic Infections (CROI 2016) last month in Boston.

While advanced HIV dementia is no longer commonly seen among people receiving effective antiretroviral therapy (ART), more subtle changes in cognitive function, known as HIV-related neurocognitive disorder or HAND, is more prevalent (15% to 50%, depending on how it's assessed). However, the precise causes of neurocognitive problems -- e.g., HIV in the brain, resulting inflammation, antiretroviral toxicities -- and how best to manage them is not fully understood.

Ned Sacktor from Johns Hopkins University School of Medicine and colleagues studied the safety and efficacy of paroxetine and fluconazole, taken alone or together, for the treatment of HAND.

HAND is associated with persistent central nervous system (CNS) inflammation, macrophage activation, and oxidative stress, and adjunctive therapies that affect these processes may play a role in its management, the researchers noted as background.

Members of Sacktor's team previously screened 2000 compounds for neuroprotective effects in an in vitro model of oxidative stress in rat nerve cells exposed to neurotoxins including the HIV tat protein. After finding that paroxetine and fluconazole appeared protective in the laboratory, they wereevaluation in this small clinical trial.

The study enrolled 45 HIV-positive people on stable ART for at least 3 months who demonstrated evidence of impairment on at least 2 neuropsychological tests. A subset of 24 patients with better than 90% adherence were included in an as-treated analysis.

Three-quarters of the participants were women, most were black, the median age was approximately 50 years, and they had a median education level of 12 years. Most had undetectable HIV viral load, the median CD4 count was approximately 500 cells/mm3, and roughly half had hepatitis C virus coinfection. They had not taken selective serotonin reuptake inhibitors (SSRIs) within the previous month.

Participants in this double-blind trial were randomly assigned to receive 20 mg once-daily oral paroxetine, 100 mg twice-daily fluconazole, the same doses of both drugs, or placebo for 24 weeks.

The researchers assessed changes in neuropsychological and motor performance using the "NPZ8" summary measure of 8 tests (including trail-making, symbol-digit, reaction time, and timed gait) and the "CalCAP" computerized test of executive function. Depression was evaluated using the Beck Depression Inventory. They measured biomarkers of neuronal injury, oxidative stress (including ceramides), macrophage/monocyte activation (CD163), and inflammation in blood serum and cerebrospinal fluid (CSF). 


  • Participants in the paroxetine arms -- either alone or in combination with fluconazole -- showed a small but significant improvement in their NPZ8 score, while those taking fluconazole alone or placebo experienced a decline after adjusting for depression (mean change +0.16 vs -0.33, respectively).
  • Participants taking paroxetine also showed a significant improvement on the CalCAP sequential reaction time test (mean change 0.41 vs 0.06, respectively).
  • People taking fluconazole showed no improvement, and some evidence of worsening, on neuropsychological performance tests.
  • There was no significant difference in depression symptom changes in the arms taking or not taking paroxetine.
  • People taking paroxetine showed a decrease in CD163, indicating reduced inflammation and macrophage activation.
  • Those taking fluconazole, either alone or with paroxetine, had changes in CSF lipid markers indicating reduced oxidative stress.
  • Paroxetine and fluconazole, alone or together, were generally safe and well-tolerated, with similar overall frequency of adverse events across the arms.

"Paroxetine treatment may be associated with cognitive improvement, even after adjusting for depression symptomatology," the researchers summarized. "Paroxetine may also be associated with less systematic macrophage activation. However, the neurocognitive improvement with paroxetine was not associated with decreases in CSF lipid markers of oxidative stress."

"Paroxetine is the first adjunctive agent to demonstrate neurocognitive improvement for a summary measure of neurocognitive performance in a double-blind, placebo controlled study for the treatment of HAND and warrants further study," they concluded. "Fluconazole was not associated with neurocognitive improvement."

"Over a period of 20 years and after 10 clinical trials, this is the first time we’ve been able to clearly demonstrate benefit in a summary measure of cognitive performance for patients with HIV-associated neurocognitive disorders," Sacktor stated in a press release issued by Johns Hopkins.



NSacktor, RL Skolasky, N Haughey, et al. Paroxetine and Fluconazole Therapy for HAND: A Double-Blind, Placebo-Controlled Trial. Conference on Retroviruses and Opportunistic Infections. Boston, February 22-25, 2016. Abstract 146.

Other Source

Johns Hopkins. Antidepressant May Improve Cognitive Symptoms in People with HIV. Press release. February 25, 2016.