In the era of combination antiretroviral therapy (cART), people living with HIV are living near-normal lifespans [1]. Despite this, cognitive impairment is still prevalent in people living with HIV, even in those with well-controlled viremia [2]. Current adjunctive therapies to cART, such as cognitive training or other medications designed to treat comorbid conditions (ie, anticholinergics and lithium), have not demonstrated significant benefit in improving cognition and quality of life in people living with HIV. Consequently, there is a critical need for the development of adjunctive therapies that improve cognition in people living with HIV.

The effects of aerobic and resistance exercise (EXS) training interventions on physical and brain health have been well documented in people without HIV. These interventions are especially important in older adults, as older individuals tend to exhibit more sedentary lives than younger individuals [3]. Improved physical functioning and physical fitness have been reported in older people without HIV who completed exercise intervention programs [4-7]. Older adults have also demonstrated improvements in cardiorespiratory health (ie, VO₂ maximum [VO_{2 max}], the maximum rate of oxygen consumption during exercise) after completing an exercise program [8]. Reviews of the extant literature reveal that exercise intervention programs improve cognitive performance or slow cognitive decline, specifically in domains such as attention, processing speed, executive function, and memory [9-12] in both clinical populations (mild cognitive impairment and Alzheimer’s disease) and otherwise healthy older adults. Neuroimaging indices of brain integrity also improve in response to engagement in structured exercise interventions in older adults. Specifically, increased brain volumes [13,14], changes in cerebral blood flow (CBF) [15-17], and improved brain network connectivity [16,18,19] have been observed in individuals completing exercise interventions. These results are particularly evident in the frontal and temporal regions (including the hippocampus), each of which is critical for executive function, learning, and memory.

Research into the efficacy of an exercise intervention in improving physical and brain health in people living with HIV is less established. Longitudinal intervention studies have primarily focused on physical health outcomes, with evidence of improved blood pressures (BPs), cardiorespiratory fitness (VO_{2 max}), inflammatory markers, physical strength, and body composition in people living with HIV following completion of an EXS training program or other activity programs, such as yoga [20-25]. The relationship between exercise and cognitive outcomes in people living with HIV has primarily been assessed using association studies and self-reported levels of physical activity. The effects of self-reported exercise or physical activity on cognitive function in the setting of HIV have produced conflicting results, with both positive [26-30] and null findings [31] reported. Because these studies did not include an intervention component, a causal effect of exercise on cognition was not able to be established.

Few studies have used neuroimaging to examine the effects of exercise on structural brain integrity in people living with HIV. While one cross-sectional study reported larger putamen volumes in a group of physically active people living with HIV compared to sedentary people living with HIV [30], to our knowledge, there have been no prospective studies examining neuroimaging markers of brain integrity before and after completion of an EXS training intervention in older people living with HIV. Therefore, the purpose of this clinical trial is to examine the effects of a 6-month EXS training intervention (vs a stretching control group) on cardiorespiratory and physical function, plasma markers of cardiometabolic risk and neurotrophic peptides, cognitive performance, and neuroimaging metrics in a sample of sedentary older people living with HIV who are on a stable cART regimen and virologically well controlled.

The primary objective of this study is to compare the effects of supervised EXS training versus a social-interaction stretching (SIS) group on brain health in people living with HIV aged ≥40 years.

All study visits are conducted at Washington University in St. Louis. Participants are recruited from local clinic referrals, flyers posted in pharmacies, food pantries, and advocacy groups in addition to word of mouth. Our team has also met with caseworkers and physicians to introduce them to this study.

Inclusion criteria are as follows: age 40-80 years, sedentary lifestyle (defined as structured exercise for less than 2 hours per week for the last 6 months), known HIV infection, on a stable cART regimen with an undetectable plasma HIV RNA (<50 copies/mL) for at least 12 months prior to enrollment, able to receive a magnetic resonance imaging (MRI) scan, and able to attend 3 EXS or SIS sessions a week over 6 months. It should be noted that the study inclusion criteria originally only enrolled older people living with HIV older than 50 years. However, due to difficulty in enrolling this population, the study opened enrollment to people living with HIV older than 40 years during the final year of enrollment.

Exclusion criteria are as follows: history of serious cardiovascular or cerebrovascular disorder (ie, stroke and uncontrolled hypertension), pulmonary disease, or significant neurological disorders; substance use disorder within the last 6 months; pregnant or breastfeeding; currently prescribed anticoagulants or medications known to affect glucose and lipid metabolism (other than statin medications); or any other medical condition that, in the opinion of the principal investigator, makes the patient unsuitable for inclusion in the study (eg, severe psychiatric illness).

Once the participant has completed 3 months (36-40 gym visits), a midpoint visit is scheduled. At the midpoint visit, an abbreviated NP battery is administered (see Textbox 1). This battery includes tests that were administered during the baseline NP testing, with alternative versions of NP tests used when available. Additionally, the participant is again provided an Actigraph to be worn for 1 week in order to assess daily activity during participation in the study.

Once the intervention portion (78 sessions) is completed, the end point (6-month) visits are scheduled. End point visits are repeats of the baseline visits 1 and 2, with the cardiovascular testing visit occurring first. The DXA scan and VO_{2 max} test are completed, fasting blood samples are collected, and the participant is given an Actigraph to wear on the nondominant wrist for 1 week. After 1 week, the participant returns for the follow-up MRI and NP testing visit. Upon completion of the study, participants are given a 6-month membership to a local fitness facility.

Midpoint and end point visits are completed for all participants, regardless of assigned intervention.

This study aims to also identify key implementation factors needed for scale-up of an EXS program in older people living with HIV from a single site to HIV clinics at multiple institutions. In order to gain a sound understanding of contextual factors that affect implementation of an EXS program for older, sedentary people living with HIV, we are conducting interviews and focus groups with staff and participants at several time periods during the intervention. This allows us to (1) tailor aspects of the program to coincide with reported barriers to implementation, (2) develop strategies to enhance adherence, and (3) make recommendations for future intervention strategies based on modifiable psychosocial characteristics of people living with HIV as they progress through the intervention.

The recruitment goal for this study is 150 older participants living with HIV, with 100 randomized to the EXS intervention and 50 participants randomized to the SIS control intervention. With this sample size, we calculated 80% power to detect an effect size (Cohen d=0.49) in the primary analyses (aim 1 hypotheses a and b; aim 2 hypothesis a). Anticipating a 20% dropout rate at 26 weeks (final EXS n=80; SIS n=40), we would have 80% power to detect an effect size (Cohen d=0.55). For aim 2 hypothesis b, and anticipating a 20% dropout rate, we would have 80% power to detect a predictor explaining R²=0.10 or 10% of the variation in the outcome.

For aim 1, planned statistical analyses include t tests examining change in cognition (domain and global cognition z scores), cardiorespiratory fitness, physical function, and questionnaire data from the baseline to end point visits between the EXS and SIS groups. This analysis uses all randomized participants under the intent-to-treat paradigm. Additionally, linear models are used to determine the mediating effect of the cardiorespiratory fitness and physical function markers on the change from baseline in domain and global cognition z scores by including these markers, together with group assignment, as predictors in the model.

For aim 2, similar analyses are conducted to those as described for aim 1, with brain volumetrics and CBF values used in place of cognitive performance. Brain volumes are adjusted for intracranial volume. Linear regression analyses are conducted to examine relationships between changes in brain volumetrics or CBF and changes in cognitive performance from baseline to end point visits.

For the exploratory hypothesis, linear regression models are used to examine relationships between changes in inflammatory or metabolic markers and changes in brain volumetrics, CBF, or cognitive performance. Group assignment is included as a predictor in the model.

Before completing any study procedures, all participants provided written informed consent that was approved by the Institutional Review Board at Washington University in St. Louis (IRB #201508002). All study procedures were in accordance with the ethical standards described by the Helsinki Declaration. Study participants are assigned an identification number, allowing all stored records to be deidentified. Deidentified data are kept in a locked file room or stored in a password-protected database on a password-protected computer. All participants are monetarily compensated for their participation, prorated to each study visit or gym visit completed, up to US $815. Participants can withdraw from the study at any point.