Women living with HIV continue to perform worse than their HIV-negative peers on a wide range of cognitive performance tests, even after many years on successful antiretroviral therapy, according to research presented at CROI 2017 in Seattle, Washington. However, underperformance trends differed depending on the area of cognitive function -- and some showed signs of possible convergence as women age.
The aim of the study, presented on Feb. 15 in a poster and discussion session by Leah H. Rubin, Ph.D., of the University of Illinois at Chicago's Department of Psychiatry, was to compare changes in cognitive performance over a four-year period between three groups: HIV-positive women on virologically suppressive treatment, HIV-positive women on treatment whose viral load was not suppressed and HIV-negative women. The data used in this research were obtained from the massive, ongoing Women's Interagency HIV Study (WIHS).
"Cognitive impairment is a problem despite combination antiretroviral therapy, and there is no question that cognitive impairment remains a problem despite viral suppression," Rubin said. "I think we know very little ... about the cognitive trajectories, particularly in those who are virally suppressed and also in women. Within the WIHS, we have a great opportunity to be able to focus on these things longitudinally."
Rubin added that the WIHS data also allowed researchers to avoid heterogeneity in study results, in which virally suppressed and unsuppressed patients are conflated. "What the heterogeneity does within our samples is introduce a barrier to understanding the emergent burden of disease in groups that achieve viral suppression, and [it] really causes a problem when we're trying to identify the mechanistic etiologies for patients that adhere to treatment and maintain viral suppression."
The study included 932 participants: 239 participants were categorized as HIV positive, on antiretroviral therapy and virally suppressed; 392 were categorized as HIV positive but not virally suppressed (all had been prescribed treatment, but nearly half reported inconsistent use of HIV medications over the last four years); and 301 were categorized as HIV negative.
In many ways, the women whose data were sampled for this study are highly representative of the U.S. epidemic: Average age was in the mid-40s; 61% to 69% were African American (depending on the study arm); 16% to 20% were Latinx; 7% to 16% were white; 39% to 48% had a household income of $12,000 or lower; and the average length of time on HIV treatment was 10 to 12 years.
The women were administered a battery of neuropsychological tests every two years to assess attention, executive function, fluency, learning, memory, motor skills and speed. Test performance was measured by T-score.
Test results showed that, in terms of global function, all study arms experienced a decline in T-scores over time throughout the four-year span examined in the study. However, HIV-negative women consistently outperformed HIV-positive women by roughly one standard deviation. Global performance did not differ between HIV-positive women who were virally suppressed and those who were not.
"It's not typical to see the pattern of cognitive performance to actually decline over time," Rubin said. "I would argue that, within our population at large, we're talking about midlife women that may have low cognitive reserve; they [may have] been abusing substances over time, with a lot of trauma and psychological risk factors. So, above all, we may be looking at accelerated aging, and then HIV may be causing this additional injury."
Differences in trends emerged, depending on the specific cognitive domain tested.
Attention: All study arms experienced sharp declines in T-score from baseline through year four. HIV-negative women began slightly higher than non-suppressed HIV-positive women and experienced similar rates of decline. T-scores for suppressed HIV-positive women started lower than the other two arms and appeared to drop at a slightly faster rate.
Executive function: All three study arms saw virtually no change in T-score during the study period. HIV-negative women scored very slightly higher than non-suppressed HIV-positive women, who in turn scored very slightly higher than suppressed HIV-positive women.
Fluency: HIV-negative women and non-suppressed HIV-positive women began with similar T-scores (non-suppressed was very slightly higher). Both had declined slightly by year four, with non-suppressed HIV-positive women appearing to decline faster. Suppressed HIV-positive women began the study with lower fluency scores, but they held steady through year four, effectively bringing all three groups closer together over time.
Learning: All study arms experienced sharp declines in T-score from baseline through year four. HIV-negative women began highest and ended highest. HIV-positive non-suppressed women began lowest and declined at a similar rate to HIV-negative women. HIV-positive suppressed women began between the other two groups and then declined at a slower rate, bringing them close to the HIV-negative group by year four.
Memory: All study arms experienced sharp declines in T-score from baseline through year four. HIV-negative women began considerably higher than the other groups but declined at a faster rate. HIV-positive non-suppressed women began slightly lower than suppressed women and declined at a slightly faster rate. By year four, all three groups were much closer in T-score than when they began.
Motor skills: This was the most unusual set of T-score trends. Virally suppressed HIV-positive women began the study period with slightly higher scores than HIV-negative women, but whereas the suppressed women experienced a decline, the HIV-negative women's scores increased, yielding a small lead for HIV-negative women by year four. Non-suppressed HIV-positive women began with slightly lower scores that then declined at a slightly slower rate than suppressed women, but not slow enough to catch up.
"[T]his type of work is starting to show us ... a framework for understanding detrimental cognitive change and something that would help individuals who are really trying to focus on the pathophysiology that's causing these impairments," Rubin said.
Rubin noted that, by the summer of 2017, her research team will have new data allowing them to plot T-scores for another set of time points.
Myles Helfand is the editorial director of TheBody.com and TheBodyPRO.com.
Follow Myles on Twitter: @MylesatTheBody.