Dual Trajectories of Serum Brain-Derived Neurotrophic Factor and Cognitive Function in People Living with HIV

Dual Trajectories of Serum Brain-Derived Neurotrophic Factor and Cognitive Function in People Living with HIV | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Dual Trajectories of Serum Brain-Derived Neurotrophic Factor and Cognitive Function in People Living with HIV HENRY MICHAEL, Antony Rapulana, Theresa Smit, Njabulo Xulu, Sivapragashini Danaviah, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4307577/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 03 May, 2025 Read the published version in Scientific Reports → Version 1 posted 8 You are reading this latest preprint version Abstract This study aimed to identify the interrelationships between mature BDNF (mBDNF), precursor BDNF (proBDNF) trajectories, and cognitive performance in individuals with HIV from sub-Saharan Africa over 96 weeks following antiretroviral therapy (ART) initiation. Using data from 154 participants in the ACTG 5199 study (ClinicalTrials.gov NCT00096824, 2005-06-23) in Johannesburg and Harare (2006–2009), we measured serum mBDNF and proBDNF levels via ELISA and assessed cognitive performance with neuropsychological tests. Group-based trajectory modelling indicated two mBDNF trajectories—"Stable Ascent" (83.9%) and "Peak with Gradual Decline" (16.1%)—and two proBDNF trajectories—"Gradual Increase" (85.7%) and "Gradual Decline" (14.3%). These were linked to three cognitive trajectories: "Low Baseline-Slow Improvement," "Gradual Improvement," and "Late Surge." The "Stable Ascent" mBDNF group showed a significant probability of "Gradual Improvement" (68%) in cognitive performance and a "Late Surge" (9.5%). In contrast, the "Peak with Gradual Decline" mBDNF trajectory saw no "Late Surge." A "Gradual Increase" in proBDNF corresponded to a 67.7% chance of "Gradual Improvement" in cognition. Findings suggest BDNF isoforms as potential biomarkers for cognitive interventions in HIV, emphasizing that stable or increasing BDNF levels post-ART are linked to favourable cognitive outcomes. Further research is needed to develop BDNF-based cognitive health strategies to improve outcomes for people with HIV. Biological sciences/Neuroscience Health sciences/Biomarkers Health sciences/Neurology HIV/AIDS Cognition Brain-derived neurotrophic factor Group-based trajectory modelling sub-Saharan Africa Figures Figure 1 Figure 2 Figure 3 INTRODUCTION Despite a 69% drop in HIV-related deaths since 2004 due to wider antiretroviral therapy (ART) access[ 1 ], individuals with HIV still face significant risks of comorbidities, notably HIV-associated neurocognitive impairment[ 2 ]. Affecting 43% of the global HIV community[ 3 ] and 45% in sub-Saharan Africa[ 4 ], this condition stems from various factors such as chronic immune activation, inflammation, ART side effects, opportunistic infections, social determinants, and stigma[ 5 – 7 ]. HIV-associated neurocognitive impairment compromises treatment adherence and quality of life and increases healthcare demands and costs[ 8 ]. This underscores the critical need for the discovery and targeting of new biological pathways, and the development of biomarkers for early detection, prevention, tailored interventions, and monitoring treatment effectiveness. Brain-derived neurotrophic factor (BDNF) is increasingly recognized as a potential biomarker for cognitive changes in various neuropathologic conditions due to its critical roles in synaptic plasticity, neurogenesis, and neuronal differentiation[ 9 ]. Extensive evidence from the general population highlights the benefits of mature BDNF (mBDNF) on cognitive and motor learning via mechanisms like long-term potentiation[ 10 ]. Conversely, mBDNF’s precursor, proBDNF, is associated with long-term depression and apoptotic pathways, suggesting a nuanced role in brain function[ 10 , 11 ]. Throughout this article, the term "BDNF" without a specific prefix is used to refer either to the overall category of BDNF-related proteins or in reference to instances where distinctions among the protein's forms are not made clear in cited studies. Although longitudinal research on BDNF and cognitive function in HIV is limited, various cross-sectional studies have examined the relationship between BDNF levels and cognition among people living with HIV, with inconsistent results. For example, an Italian study reported that higher BDNF levels in serum were linked to slower performance on the Grooved Pegboard Test[ 12 ]. Conversely, a U.S. study found that increased serum BDNF levels improved learning and delayed recall in older African-American individuals living with HIV[ 13 ]. Another study highlighted a nuanced relationship between plasma BDNF levels and cognitive performance in individuals with HIV, suggesting that moderate, not high, BDNF levels might be protective[ 14 ]. The variability in the impact of BDNF on cognitive function across individuals living with HIV may stem from previous studies' lack of distinction between the effects of mBDNF and its precursor, proBDNF. This oversight complicates comparisons with the broader body of literature. Additionally, inadequate consideration of temporal dynamics, potential confounders, cognitive domains assessed, and variability in sample sites and types has likely contributed to inconsistent findings, both within HIV research and more generally[ 15 ]. While mixed-effect models are adept at analyzing longitudinal data, they presuppose a uniform trajectory across individuals, which may simplify the intricate dynamics between BDNF and cognitive performance [ 16 ]. Group-based trajectory modeling (GBTM) addresses this limitation by allowing for the identification of distinct subgroups within the population that share similar patterns over time. This method is particularly advantageous for capturing heterogeneity and uncovering co-occurring or divergent trajectories of biomarkers and cognitive outcomes, offering deeper insights into underlying processes. By leveraging GBTM, this study aims to unravel the intertwined trajectories of mBDNF, proBDNF, and cognition, providing a more nuanced understanding of their relationships. Towards this end, our study adopts Group-based Dual Trajectory Modeling (GBDTM) to explore the intertwined trajectories of mBDNF, proBDNF, and cognitive performance in adults living with HIV in sub-Saharan Africa. GBDTM enhances traditional univariate group-based trajectory modelling by simultaneously examining two distinct outcomes and their evolution over time, thereby capturing their potential comorbidity or heterotypic continuity[ 17 ]. This method, rooted in finite growth mixture modelling, identifies groups of individuals sharing similar patterns over time, offering a nuanced understanding of the interplay between biological markers and cognitive performance[ 18 ]. Although previous research has shed light on BDNF's impact on patient outcomes[ 19 , 20 ] and complex cognitive phenotypes in HIV[ 21 – 24 ], it has not linked these elements. Our study is pioneering in applying GBDTM to unravel the concurrent development trajectories of mBDNF, proBDNF, and cognition among individuals with HIV, marking a significant step forward in understanding their dynamic relationship. In this study, we posed four research questions: (1) What are the distinct trajectories of mBDNF, proBDNF, and cognitive performance in a cohort of people living with HIV over 96 weeks post-ART initiation? (2) What proportion of the cohort follows each identified trajectory? (3) What sociodemographic and clinical factors influence the trajectories of mBDNF and proBDNF? (4) How are the trajectories of mBDNF and proBDNF linked to changes in cognitive performance when analyzed together? We hypothesize that an upward trajectory in cognitive performance will correlate with rising levels of mBDNF and a decrease in proBDNF levels. METHODS This study was a secondary analysis of serum specimens and data from the Johannesburg and Harare sites of the ACTG A5199, the International Neurological Study (INS) from 2006–2009. The ACTG 5199 study evaluated the neurological and neuropsychological effects of three randomly assigned antiretroviral regimens in resource-limited settings and the study methods have been detailed in previous reports[ 25 – 27 ]. Cohort description This analysis focused on ART-naïve individuals with CD4 counts under 300, who were part of the ACTG A5175 multinational trial assessing the efficacy and safety of antiretroviral treatments (ClinicalTrials.gov, NCT00084136) [ 28 ]. Participants were required to have a Karnofsky performance score of at least 70 and no more than a week’s experience with antiretroviral therapy prior to the study's commencement. Those with recent substance or alcohol abuse, severe psychiatric illnesses, hospitalization within the fortnight preceding study entry, or any condition likely to affect study participation or results interpretation were excluded. Follow-up included neurocognitive and neurological evaluations every 24 weeks over a period of up to 192 weeks, with this study reporting on data and specimens collected within the first 96 weeks. Measurements Quantitative ELISA Analysis of Serum mBDNF and proBDNF Levels Quantitative analysis of serum mBDNF and proBDNF levels was conducted using a specific enzyme-linked immunoassay (ELISA) kit, adhering to the standardized protocols provided by Biosensis Pty Ltd., Thebarton, SA, Australia. To maintain consistency and reduce variability between assays, samples, which were preserved at -70°C, were tested in pairs on the same ELISA plate. The intra-assay variability for mBDNF and proBDNF was notably low, recorded at 2.7% and 2.2% respectively, confirming the precision of the measurements. Additionally, the laboratory technicians analyzing the samples were kept unaware of the participants' clinical and cognitive status to eliminate any potential for bias in the results. Neuropsychological tests Data collected as part of the ACTG study was accessed and used in this analysis. During the ACTG study, participants underwent a battery of four neuropsychological tests – grooved pegboard (both hands)[ 29 ], finger tapping (both hands)[ 30 ], timed gait[ 31 ] and semantic verbal fluency tests[ 32 ]. These tests were administered primarily by physicians, with training and quality assurance conducted annually, complemented by biannual centralized training sessions. Sites underwent initial quality evaluations before full enrollment, ensuring readiness. Data collection was standardized through translated and back-translated case forms to resolve inconsistencies. Ongoing monitoring included automated range checks and follow-ups on outliers, safeguarding data accuracy and reliability[ 26 , 27 ]. The neuropsychological assessments were conducted at baseline and then followed up every 24 weeks. We utilized neuropsychological data at baseline, 24, 48 and 96 weeks. Test scores were normalized into z-scores, utilizing demographic adjustments (site, age, sex, and education-adjusted) based on data from the International Neurocognitive Normative Study ACTG A5271, encompassing a wide array of at-risk populations[ 33 ]. These z-scores were then averaged to formulate a composite cognitive z-score, which serves as a measure reflective of cognitive performance, with higher scores denoting higher cognitive performance. Predictors To investigate predictors of trajectory group membership, we analyzed key demographic and clinical variables. These included age (in years), sex (male/female), years of education, study site (Harare/Johannesburg), baseline CD4 count (cells/mm³), baseline CD8 count (cells/mm³), CD4/CD8 ratio, and log-transformed plasma HIV RNA levels. These variables were chosen based on their relevance to HIV-related neurocognitive outcomes and their potential to influence neurotrophic factors. Following the identification of trajectory groups, regression models were employed to determine the associations between these predictors and trajectory group membership. Statistical analysis We included participants who had completed at least two neuropsychological assessments across the study period. To explore distinct patterns over time, we applied group-based trajectory modelling (GBTM) to each variable (log-transformed serum mBDNF, log-transformed proBDNF, and the composite cognitive z-score) using an iterative approach to determine the optimal number and shape (linear, quadratic, or cubic) of trajectories[ 18 , 34 ]. Log transformation was applied to mBDNF and proBDNF values to normalize their distributions prior to modelling. GBTM assumes that the population comprises discrete subgroups that follow distinct developmental trajectories. The selection of the best-fitting model for mBDNF, proBDNF, and the composite cognitive z-sore was grounded in a comprehensive evaluation of model fit diagnostics and a set of criteria aimed at minimizing the risk of identifying spurious groups [ 35 ]. These criteria included the Bayesian Information Criterion (BIC) for model fit, with higher BIC values signifying a better fit; an Average Posterior Probability (APP) of group assignments above 0.70; a minimum group membership probability of about 5% of the sample; odds of correct classification exceeding 5; tight 99% confidence intervals around estimated group memberships; and evidence of statistically significant groups (P < 0.05), along with the requisites of graphical clarity, clinical relevance, and theoretical interpretability. Subsequently, participants were categorized into trajectory groups based on their maximum posterior probability derived from the final models. We conducted descriptive analyses to outline baseline characteristics of the cohort and further analyzed differences in demographic and HIV-related variables across mBDNF and proBDNF trajectory groups using Wilcoxon rank-sum tests for continuous variables and chi-square tests for categorical variables. Following these univariate comparisons, multivariable logistic or multinomial regression was utilized to ascertain significant predictors of group membership, setting a significance threshold at P < .05. Additionally, dual trajectory models elucidated the relationship between BDNF levels and cognitive performance, presenting conditional and joint probabilities to highlight the mutual influence of mBDNF and the composite cognitive z-score, as well as proBDNF and the composite cognitive z-score trajectories[ 16 ]. This comprehensive analysis leverages the “lcmm” package[ 36 ] within R software version 4.3.1, facilitating an in-depth exploration of the longitudinal relationships between neurotrophic factors and cognitive trajectories in our cohort. The annotated R code for the dual trajectory analysis is available in Supplementary Material S1. RESULTS Of the 158 participants with baseline neuropsychological data and serum samples, 4 were excluded due to lack of follow-up, yielding a baseline analytical sample of 154 participants (Supplementary Material S2). In our statistical analysis, we incorporated data from participants who completed at least two neuropsychological assessments. Missing data were assessed for key variables including serum mBDNF, proBDNF, and the composite cognitive z-score. Specifically, missing values for mBDNF ranged from 1 at baseline to 27 at week 96, while for proBDNF, missing values ranged from 11 at week 24 to 47 at week 96. Notably, no missing values were observed for the composite cognitive z-score. The models incorporated all available data using maximum likelihood estimation, a robust approach for longitudinal analyses, ensuring reliable and unbiased parameter estimation despite the presence of missing data[37]. The median baseline age was 35 years with an interquartile range (IQR) of 10 years; the majority, 66.2%, were female, and the median educational attainment was 11 years (IQR: 3). The baseline median cognitive z-score was 0.19 with an IQR of 1.01 [Table1, Panel A]. Trajectories of Serum mBDNF Based on fit indices and our a priori criteria, we identified two trajectories of mBDNF over 96 weeks post-ART. The best-fitting model was a two-group cubic model, with trajectory shapes that captured a “Stable ascent” (N = 129, 83.9%) and a “Peak with gradual decline” (N = 25, 16.1%) (Figure 1). Supplementary material S3 provides detailed information on model fit indices and diagnostics of assignment accuracy. Table 1 (Panel B) presents baseline descriptive statistics for the mBDNF trajectory groups. Participants in the stable ascent group tended to be younger, with lower CD8 cell counts at baseline and higher mBDNF levels than those in the "peak with gradual decline" group. However, there were no significant differences between the groups regarding years of education, sex, plasma viral load, CD4 count, and baseline cognitive performance. Furthermore, the results of multivariable logistic regression analysis, as shown in Supplementary material S4, indicate that baseline age, CD8 count, and baseline serum mBDNF were significant predictors of mBDNF trajectory group membership. Trajectories of Serum ProBDNF We identified two trajectories of serum proBDNF over 96 weeks post-ART that best fit the data. The optimal model was a two-group cubic model, resulting in trajectories labelled as “Gradual Increase” (N = 132, 85.7%) and “Gradual Decline” (N = 22, 14.3%) (Figure 2). Supplementary material S5 provides detailed information on model fit indices and diagnostics of assignment accuracy. Table 1 (Panel C) presents baseline descriptive statistics for the proBDNF trajectory groups. Participants were only significantly different (p < 0.05) based on study site. Furthermore, study site remained significant (Log odds -0.07, Std error = 0.04, p = 0.01) in the multivariable logistic regression model (Supplementary material S6). Trajectories of Cognitive Performance (Composite Cognitive Z-scores) Our analysis delineated three primary composite cognitive z-score trajectories that resonate with the patterns observed in our data. The best-fitting model was a three-group cubic model, reflecting the following patterns: “Low Baseline–Slow Improvement” (N = 36, 24.4%), “Gradual Improvement” (N = 106, 67.7%), and “Late Surge” (N = 12, 7.7%) (Figure 3). Supplementary Material S7 provides comprehensive details on the model's goodness-of-fit and trajectory classification accuracy. While all three trajectories demonstrated improvement over 96 weeks post-ART, the “Low Baseline–Slow Improvement” group began with the lowest cognitive z-scores at baseline but showed steady gains, converging closely with the “Late Surge” group by approximately week 48. Table 2 delineates the baseline characteristics distinguishing the cognitive trajectory groups. Notably, the "Low Baseline-Slow Improvement" group was significantly younger and had higher serum proBDNF levels compared to the "Gradual Improvement" group (p = 0.02). This group also displayed considerably lower composite cognitive z-scores relative to both the "Gradual Improvement" and "Late Surge" groups (p < 0.001). Multivariable logistic regression further identified baseline age (Log odds = -0.21, Std error = 0.07, p < 0.01) and baseline composite cognitive z-scores (Log odds = -1.89, Std error = 0.61, p < 0.001) as significant variables differentiating the "Low Baseline-Slow Improvement" group from the "Gradual Improvement” and "Late Surge" groups, underscoring the importance of these factors in predicting cognitive trajectory post-ART initiation (Supplementary material S8). Interrelationships Across the Trajectory Groups of Serum mBDNF and Cognitive performance In examining the association between mBDNF trajectories and cognitive trajectories, distinct conditional probabilities emerged. Table 3a delineates the conditional probabilities of cognitive performance based on serum mBDNF trajectories. It shows individuals with a peak and decline mBDNF trajectory have a 68% probability of gradual cognitive improvement, with none exhibiting a late surge. Those with a stable ascent mBDNF trajectory show a 68% probability of gradual cognitive improvement, a lower probability of 22.52% for low baseline-slow improvement, and a 9.5% probability of a late surge. Table 3b reflects the conditional probabilities of serum mBDNF trajectories given the cognitive trajectory groups. In the group with low baseline and slow improvement in composite cognitive z-scores, a 22.4% chance existed of following a peak and decline pattern in mBDNF levels, whereas a stable rise in mBDNF levels was highly probable at 77.6%. Within the gradual improvement cognitive trajectory, 16.9% are likely to have a peak and decline mBDNF trajectory, but a higher probability of 83.1% is observed for a stable ascent mBDNF trajectory. Notably, every individual within the late surge cognitive trajectory group is associated with a stable ascent mBDNF pattern, as indicated by the 100% probability. Table 3c displays the joint probabilities for serum mBDNF and cognitive trajectories. The table highlights that individuals with a stable ascent mBDNF trajectory have the highest joint probability of 56.5% for being in the gradual improvement cognitive trajectory, a lower probability of 18.7% for the low baseline-slow improvement group, and 7.9% for the late surge group. For the peak and decline mBDNF trajectory, the joint probability is 5.4% for the low baseline-slow improvement group and 11.5% for the gradual improvement group, with a 0% probability for the late surge cognitive trajectory, indicating no occurrence of this mBDNF pattern in the late surge cognitive category. Interrelationships Across the Trajectory Groups of Serum proBDNF and Cognitive Performance Table 4a provides the conditional probabilities of cognitive trajectories based on serum proBDNF levels. Individuals with a Gradual Decline in proBDNF have a 72.4% probability of falling into the Gradual improvement cognitive trajectory, while the probability of them experiencing a Late surge is 0%. For those with a Gradual Increase in proBDNF, there is a 67.7% probability for Gradual improvement and a notably lower 22.9% for Low Baseline-Slow improvement, with a small 9.4% for Late surge. Table 4b shows the conditional probabilities of serum proBDNF trajectories given the cognitive trajectory group. In the Low Baseline-Slow improvement cognitive trajectory, 82.7% of individuals are characterized by a Gradual Increase in proBDNF, while only 17.3% show a Gradual Decline. The Gradual improvement cognitive group has a high probability of 84.4% for a Gradual Increase in proBDNF and a lower probability of 15.6% for a Gradual Decline. The Late surge trajectory is exclusively linked to a Gradual Increase in proBDNF, with a probability of 100%. Table 4c presents the joint probabilities of serum proBDNF levels and cognitive trajectories. The highest joint probability is for a Gradual Increase in proBDNF with the Gradual improvement cognitive group at 57.7%, while the lowest is the absence of any individuals (0%) with a Gradual Decline in proBDNF in the Late surge cognitive trajectory. Additionally, there is a moderate joint probability of 19.5% for a Gradual Increase in proBDNF with the Low Baseline-Slow improvement cognitive group and a slight 8% with the Late surge group. The Gradual Decline proBDNF trajectory is least common, with a joint probability of 4.1% for the Low Baseline-Slow improvement group and 10.7% for the Gradual improvement group. Interrelationships Between Serum mBDNF and proBDNF Trajectories Supplementary Material S9 provides detailed insights into the inter-relationship between serum mBDNF and proBDNF trajectories through conditional and joint probabilities. Table S7a presents the conditional probabilities of serum proBDNF trajectories based on mBDNF trajectories. Among individuals in the Stable Ascent mBDNF trajectory, 82.7% were associated with the Gradual Increase proBDNF trajectory, while 17.3% followed the Gradual Decline proBDNF trajectory. Conversely, for those in the Peak with Gradual Decline mBDNF trajectory, 86.2% belonged to the Gradual Decline proBDNF trajectory, and only 13.8% were aligned with the Gradual Increase proBDNF trajectory. Table S7b reflects the reverse conditional probabilities of mBDNF trajectories given proBDNF trajectories. Individuals in the Gradual Decline proBDNF trajectory had a 27.6% probability of being in the Stable Ascent mBDNF trajectory and a 72.4% probability of following the Peak with Gradual Decline mBDNF trajectory. In contrast, among those in the Gradual Increase proBDNF trajectory, 77.1% were likely to belong to the Stable Ascent mBDNF trajectory, while 22.9% followed the Peak with Gradual Decline mBDNF trajectory. Table S7c displays the joint probabilities of trajectory membership for serum mBDNF and proBDNF. The most frequent trajectory pairing was Stable Ascent mBDNF with a Gradual Increase in proBDNF, occurring in 70.3% of participants. This was followed by a Stable Ascent of mBDNF with a Gradual Decline in proBDNF (12.7%). The least common pairings were Peak with Gradual Decline mBDNF with either Gradual Increase proBDNF (2%) or Gradual Decline proBDNF (14.9%). DISCUSSION This study is the first to examine the dynamic interplay between serum mBDNF and proBDNF and cognitive trajectories in people living with HIV after antiretroviral therapy initiation. Over 96 weeks, we identified two distinct serum mBDNF trajectories—stable ascent and peak with gradual decline—and two for proBDNF—gradual increase and gradual decline—each associated with three cognitive trajectories: low baseline with slow improvement, gradual improvement, and late surge. Notably, stable mBDNF ascent and gradual proBDNF increases were more commonly observed among participants with gradual cognitive improvement, while elevated proBDNF levels appeared in those with late cognitive surges. These findings suggest that mBDNF and proBDNF levels may be linked to neurocognitive improvement following ART initiation. Our analysis reveals an association between serum mBDNF trajectories and cognitive trajectory, particularly noting that a stable ascent in mBDNF levels correlates with better cognitive trajectories. A stable ascent in mBDNF levels was predominantly observed in participants who demonstrated gradual cognitive improvement and in those experiencing late cognitive surges. This pattern may reflect the neurotrophic support provided by increasing mBDNF levels. Conversely, the trajectory of peaking followed by a gradual decline in mBDNF levels shows a lower likelihood of cognitive gains and is absent among those with late surges, suggesting its potential as an indicator of limited cognitive improvement. These observations are consistent with the proposed role of mBDNF in supporting neuroplasticity and counteracting neurodegeneration triggered by viral proteins[ 38 – 40 ]. The relationship between peripheral BDNF levels and central nervous system (CNS) function is complex. BDNF is primarily synthesized in the CNS but is also present peripherally and can cross the blood-brain barrier (BBB) in both directions[ 41 ]. Studies suggest that peripheral BDNF levels may reflect central BDNF activity, though this correlation can be influenced by various factors[ 42 ]. In the context of HIV, the virus can compromise BBB integrity, potentially altering BDNF transport dynamics[ 43 ]. Additionally, ART regimens have been associated with changes in BDNF signalling pathways[ 44 ]. While ART effectively reduces viral load and neuroinflammation, it may not fully restore BDNF signalling deficits, which may contribute to persistent neurocognitive challenges in people living with HIV[ 45 ]. These factors underscore the intricate interplay between peripheral BDNF dynamics, ART, and CNS neuroplasticity in PLWH. This observation emphasizes the need for carefully designed and monitored cognitive enhancement strategies for people with HIV, particularly in settings with limited resources[ 46 ]. It points out the importance of methods that ensure a sustained increase in mBDNF levels for ongoing cognitive improvement, through various means such as exercise, nutritional supplements, or cognitive training[ 9 , 47 ]. Evidence, including studies focused on HIV, shows that regular exercise boosts BDNF levels and response[ 48 – 50 ]. However, personalizing these interventions to match individual preferences, abilities, the nature and intensity of the activity, and considering time and resource availability, is essential for preserving these cognitive gains over time[ 51 ]. Unexpectedly, we found that "gradual increase" trajectories in serum proBDNF were linked to positive cognitive trajectories, diverging from our original hypothesis and the traditional understanding of proBDNF's role in cognitive decline through mechanisms like long-term depression and neuronal atrophy[ 52 , 53 ]. This finding contrasts with proBDNF's known effects on spine retraction, dendritic pruning, and neuronal death[ 53 – 55 ]. However, aligning with our previous reports and supported by both preclinical and clinical studies, improved cognitive outcomes have been observed when increases in mBDNF are accompanied by rises in proBDNF levels[ 56 – 63 ]. This indicates that moderate increases in proBDNF levels could indicate ongoing neural remodelling and serve as a source for mBDNF production, potentially explaining its link to cognitive improvement[ 56 ]. Additionally, the beneficial impact of rising serum proBDNF levels might be most evident in the early phases of cognitive decline, suggesting a phase-specific role[ 58 ]. This emerging evidence suggests that the interaction between proBDNF and mBDNF in cognitive function may be more complex than the "yin-yang" hypothesis of opposing roles, leaning towards a nuanced understanding within the “continuum-sorting” hypothesis that emphasizes homeostatic balance[ 62 , 63 ]. Across 96 weeks, specific covariates appeared to influence mBDNF, proBDNF, and cognitive trajectories. The stable ascent in mBDNF levels was more common among younger individuals, those with lower CD8 counts, and those with higher baseline mBDNF, suggesting a possible association between immunological status, neurotrophic support, and age. Younger individuals may possess greater neuroplastic capabilities, reflected in their mBDNF profiles[ 64 , 65 ]. In addition, lower CD8 counts, indicative of reduced immune activation or inflammation, could provide a more conducive environment for mBDNF production and function[ 66 ]. Starting with higher mBDNF levels suggests an already established robust neurotrophic framework, essential for ongoing neuroplastic processes, potentially serving as a biomarker for cognitive reserve amidst HIV-associated cognitive challenges[ 67 ]. This relationship underscores the potential significance of considering biological and immunological factors in understanding mBDNF trajectories. Our analysis revealed two main serum proBDNF trajectories: "Gradual Increase" and "Gradual Decline." Study site emerged as a critical factor influencing these trajectories, suggesting that location-specific factors, possibly environmental enrichment or variations in clinical practices, may affect proBDNF dynamics. Logistic regression analysis further confirmed the study site's pivotal role in determining proBDNF trajectories. These findings highlight the need to account for site-level confounders in future research on HIV-related cognitive outcomes[[ 68 ]. All cognitive trajectory groups in our study demonstrated improvement over time, consistent with prior evidence on the neurocognitive benefits of sustained ART in individuals with acute HIV infection [ 23 ]. Our analysis revealed that the "Low Baseline-Slow Improvement" group, characterized by a younger demographic and lower initial cognitive scores, showed a slower improvement trajectory. This was unexpected, as younger age is typically associated with greater neuroplasticity. One possible explanation is that younger individuals may face greater challenges with ART adherence, as reported in prior studies, which could blunt cognitive gains despite their biological advantage[ 69 ]. Given our cohort's relatively young average age (39 years) and age-adjusted NPZ-6 scores, the data suggests that slightly older individuals within this young cohort may possess greater cognitive reserve, enabling them to achieve quicker cognitive gains post-ART. This reserve and an increased ability to develop compensatory mechanisms could explain the more rapid cognitive improvements observed post-ART in these slightly older individuals. This suggests a complex interplay between age, baseline cognitive status, and the trajectory of cognitive improvement in HIV[ 46 , 70 ]. Our study's methodology, employing a dual group-based modelling approach, represents a significant strength by probabilistically linking serum BDNF isoforms and cognitive performance trajectories, offering a nuanced view of longitudinal data. While the identified trajectories represent a spectrum of an underlying continuous phenomenon, caution is necessary for drawing causal inferences. The repeated measurement of serum mBDNF and proBDNF adds specificity to understanding their neurocognitive effects over time in people with HIV. However, the absence of tissue-level or neuroimaging data leaves the direct functional impact of BDNF on cognition uncertain. Future integration of neuroimaging could help clarify these mechanisms and link peripheral BDNF dynamics to brain-level activity. These findings contribute to the understanding of BDNF dynamics in HIV-associated neurocognitive impairment. The stable ascent of mBDNF suggests its potential as a biomarker for identifying individuals at risk of cognitive decline or tracking improvement. Exploring how mBDNF and proBDNF interact with HIV-specific factors, such as chronic inflammation and immune activation, could guide personalized interventions to sustain cognitive improvement. Approaches like tailored exercise programs or nutritional strategies may enhance neuroplasticity and improve long-term outcomes. Limitations The specific profile of our cohort— (young individuals living with HIV in a resource-limited setting, significantly immunosuppressed, part of a larger clinical trial and without significant comorbidities)—may restrict the generalizability of our findings. Moreover, since participants were recruited from two urban centers with relatively well-established clinical infrastructure, the applicability of these results to rural or less-resourced settings across sub-Saharan Africa may be limited. Additionally, a larger sample size may have unveiled more detailed group distinctions and supported more comprehensive models that could include a broader range of confounders. While our models accounted for several key demographic and clinical variables, there are additional covariates—such as inflammatory markers, ART adherence, psychosocial stressors, substance use history, ART regimen, depression, and nutritional status—that may influence BDNF expression and cognitive outcomes but were not included in this analysis. As such, residual confounding from unmeasured or excluded variables remains a possibility and may affect the interpretation of some observed relationships. The neuropsychological battery employed in this study focused predominantly on motor and psychomotor speed, due to practical constraints related to testing time, staffing, and available infrastructure in the resource-limited setting. While these measures were feasible and clinically relevant, they do not capture higher-order cognitive domains such as memory, attention, and executive function, which are more frequently affected in HIV-associated neurocognitive disorders. This limited scope reduces the ability to fully interpret the cognitive significance of observed BDNF trajectories. Future studies should incorporate more comprehensive neurocognitive assessments and clinical criteria to better characterize the breadth of cognitive outcomes in this population. Finally, relying on serum BDNF measurements rather than CNS-specific assessments limits the ability to directly infer CNS activity, as serum levels may not fully reflect central neurotrophic dynamics. Implications for Clinical Practice and Future Research This study underscores the potential of mBDNF and proBDNF as biomarkers for monitoring cognitive trajectories in people with HIV. Clinically, the findings suggest prioritizing interventions that sustain mBDNF levels, such as tailored exercise programs, nutritional strategies, or cognitive training, which have shown promise in enhancing neuroplasticity. Personalized approaches are particularly crucial in resource-limited settings to maximize cognitive gains post-ART initiation. Future research should validate these findings in larger, diverse cohorts and explore mechanisms linking peripheral BDNF dynamics with brain-level activity through neuroimaging and CNS-specific assessments. Additionally, the role of site-specific and biological factors, such as immune activation and inflammation, in shaping BDNF trajectories warrants further investigation to refine intervention strategies. Such efforts could lead to targeted, individualized care aimed at improving neurocognitive resilience and long-term outcomes in HIV populations. Declarations Ethics Approval This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of University of Kwazulu-Natl, Durban South Africa (August 18, 2022/Number BREC/00003101/2021), University of the Witwatersrand, Johannesburg, South Africa (July 28, 2021/04081), and Medical Research Council of Zimbabwe, Harare, Zimbabwe (September 10, 2021/MRCZ/E/299). Consent to Participate Informed consent was obtained from all individual participants included in the study. Consent to Publish The authors confirm that participants provided consent for publication. The data analyzed was anonymized and presented collectively. Competing Interest The authors have no relevant financial or non-financial interests to disclose. Funding Research reported in this publication was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under Award Number UM1 AI068634, UM1 AI068636 and UM1 AI106701. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Henry Michael received research support from the College of Health Sciences, University of KwaZulu-Natal, Durban, South Africa. Author Contribution All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by HM, AR and NX. TS, SR, SD and FO supervised all aspects of the project. The first draft of the manuscript was written by HM and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Acknowledgement The authors thank the study investigators and participants of the ACTG A5175, A5199, and A5271 studies for providing the samples and data for this work. The authors also thank the support staff at the African Health Research Institute for sample management and testing. Data Availability The authors confirm that all data are fully available upon request from [email protected] with the written agreement of Advancing Clinical Therapeutics Globally for HIV/AIDS and Other Infections (ACTG). Study specimens were obtained from the ACTG Repository (www.specimenrepository.org). References UNAIDS & Global, H. I. V. & AIDS statistics — Fact sheet, UNAIDS 2023 Epidemiol. Estim. (2023). https://www.unaids.org/en/resources/fact-sheet (accessed March 5, 2024). Lerner, A. M., Eisinger, R. W. & Fauci, A. S. Comorbidities in Persons With HIV: The Lingering Challenge. JAMA 323 , 19–20. https://doi.org/10.1001/jama.2019.19775 (2020). Wang, Y. et al. Global prevalence and burden of HIV-associated neurocognitive disorder. Neurology 95 , e2610–e2621. https://doi.org/10.1212/WNL.0000000000010752 (2020). Zenebe, Y., Akele, B., Selassie, M. W. & Necho, M. Neurocognitive Impairment and Associated Factors Among People Living with HIV: A Systematic Review and Meta-Analysis of African Studies. Neuropsychiatr Dis. Treat. 19 , 673–687. https://doi.org/10.2147/NDT.S377636 (2023). McArthur, J. C. & Johnson, T. P. Chronic inflammation mediates brain injury in HIV infection: relevance for cure strategies. Curr. Opin. Neurol. 33 , 397–404. https://doi.org/10.1097/WCO.0000000000000807 (2020). Barokar, J. et al. Neurocognitive impairment is worse in HIV/HCV-coinfected individuals with liver dysfunction. J. Neurovirol . 25 , 792–799. https://doi.org/10.1007/s13365-019-00767-6 (2019). Thompson, E. C. et al. Internalized HIV-Related Stigma and Neurocognitive Functioning Among Women Living with HIV. AIDS Patient Care STDS . 36 , 336–342. https://doi.org/10.1089/apc.2022.0041 (2022). Mekuriaw, B., Belayneh, Z., Teshome, W. & Akalu, Y. Prevalence and variability of HIV/AIDS-associated neurocognitive impairments in Africa: a systematic review and meta-analysis. BMC Public. Health . 23 , 997. https://doi.org/10.1186/s12889-023-15935-x (2023). Nicastri, C. M. et al. BDNF mediates improvement in cognitive performance after computerized cognitive training in healthy older adults., Alzheimer’s Dement. (New York, N. Y.) 8 e12337. (2022). https://doi.org/10.1002/trc2.12337 Mizoguchi, Y., Yao, H., Imamura, Y., Hashimoto, M. & Monji, A. Lower brain-derived neurotrophic factor levels are associated with age-related memory impairment in community-dwelling older adults: the Sefuri study. Sci. Rep. 10 , 16442. https://doi.org/10.1038/s41598-020-73576-1 (2020). Cechova, K. et al. Ratio of serum proBDNF to BDNF and its association with cognitive performance and brain morphometry in mild cognitive impairment. Alzheimer’s Dement. 16 , e046340. https://doi.org/https://doi.org/10.1002/alz.046340 (2020). Falasca, K. et al. Cytokines, Hepatic Fibrosis, and Antiretroviral Therapy Role in Neurocognitive Disorders HIV Related. AIDS Res. Hum. Retroviruses . 33 , 246–253. https://doi.org/10.1089/AID.2016.0138 (2017). Fazeli, P. L. et al. Differential Associations Between BDNF and Memory Across Older Black and White Adults With HIV Disease, JAIDS J. Acquir. Immune Defic. Syndr. 89 (2022). https://journals.lww.com/jaids/fulltext/2022/02010/differential_associations_between_bdnf_and_memory.2.aspx Míguez-Burbano, M. J. et al. Brain derived neurotrophic factor and cognitive status: the delicate balance among people living with HIV, with and without alcohol abuse. Curr. HIV Res. 12 , 254–264. https://doi.org/10.2174/1570162x12666140721121238 (2014). Wang, C. S., Kavalali, E. T. & Monteggia, L. M. BDNF signaling in context: From synaptic regulation to psychiatric disorders. Cell 185 , 62–76. https://doi.org/10.1016/j.cell.2021.12.003 (2022). Xie, H., McHugo, G. J., He, X. & Drake, R. E. Using the Group-Based Dual Trajectory Model to Analyze Two Related Longitudinal Outcomes. J. Drug Issues . 40 , 45–61. https://doi.org/10.1177/002204261004000104 (2010). Nagin, D. S. & Tremblay, R. E. Analyzing developmental trajectories of distinct but related behaviors: a group-based method. Psychol. Methods . 6 , 18–34. https://doi.org/10.1037/1082-989x.6.1.18 (2001). Nagin, D. S., Jones, B. L., Passos, V. L. & Tremblay, R. E. Group-based multi-trajectory modeling, Stat. Methods Med. Res. 27 , 2015–2023. https://doi.org/10.1177/0962280216673085 (2018). Treble-Barna, A. et al. Acute Brain-Derived Neurotrophic Factor DNA Methylation Trajectories in Cerebrospinal Fluid and Associations With Outcomes Following Severe Traumatic Brain Injury in Adults., Neurorehabil. Neural Repair. 35 , 790–800. https://doi.org/10.1177/15459683211028245 (2021). Miguez, M. J., Chan, W., Espinoza, L., Tarter, R. & Perez, C. Marijuana use among adolescents is associated with deleterious alterations in mature BDNF. AIMS Public. Heal . 6 , 4–14. https://doi.org/10.3934/publichealth.2019.1.4 (2019). Patel, P. B. et al. Trajectory Analysis of Cognitive Outcomes in Children With Perinatal HIV., Pediatr. Infect. Dis. J. 38 , 1038–1044. https://doi.org/10.1097/INF.0000000000002427 (2019). Brouillette, M. J. et al. Identifying Neurocognitive Decline at 36 Months among HIV-Positive Participants in the CHARTER Cohort Using Group-Based Trajectory Analysis. PLoS One . 11 , e0155766. https://doi.org/10.1371/journal.pone.0155766 (2016). Chan, P. et al. Cognitive trajectories after treatment in acute HIV infection. Aids 35 , 883–888. https://doi.org/10.1097/QAD.0000000000002831 (2021). Fitzgerald, K. C. et al. Factors Predicting Detrimental Change in Declarative Memory Among Women With HIV: A Study of Heterogeneity in Cognition. Front. Psychol. 11 10.3389/fpsyg.2020.548521 (2020). https://www.frontiersin.org/journals/psychology/articles/ Robertson, K. R. et al. Human Immunodeficiency Virus-associated Neurocognitive Impairment in Diverse Resource-limited Settings., Clin. Infect. Dis. Off Publ Infect. Dis. Soc. Am. 68 1733–1738. https://doi.org/10.1093/cid/ciy767 . (2019). Robertson, K. et al. Improved neuropsychological and neurological functioning across three antiretroviral regimens in diverse resource-limited settings: AIDS Clinical Trials Group study a5199, the International Neurological Study. Clin. Infect. Dis. Off Publ Infect. Dis. Soc. Am. 55 , 868–876. https://doi.org/10.1093/cid/cis507 (2012). Robertson, K. et al. A multinational study of neurological performance in antiretroviral therapy-naïve HIV-1-infected persons in diverse resource-constrained settings. J. Neurovirol . 17 , 438–447. https://doi.org/10.1007/s13365-011-0044-3 (2011). Campbell, T. B. et al. Efficacy and safety of three antiretroviral regimens for initial treatment of HIV-1: a randomized clinical trial in diverse multinational settings. PLoS Med. 9 , e1001290. https://doi.org/10.1371/journal.pmed.1001290 (2012). Lafayette Instrument, G. & Pegboard (n.d.). https://lafayetteevaluation.com/products/grooved-pegboard Reitan, R. M. & Tucson, S. The Halstead-Reitan neuropsychological test battery: theory and clinical interpretation, Second edition. Arizona : Neuropsychology Press, [1993] ©1993, n.d. https://search.library.wisc.edu/catalog/9910209689302121 Robertson, K. R. et al. Timed Gait Test: Normative Data for the Assessment of the AIDS Dementia Complex, J. Clin. Exp. Neuropsychol. 28 1053–1064. https://doi.org/10.1080/13803390500205684 . (2006). Strauss, O., Sherman, E., Spreen & E. M. S., & A compendium of neuropsychological tests: administration, norms, and commentary 3rd edn (Oxford University Press, 2006). Robertson, K. et al. International neurocognitive normative study: neurocognitive comparison data in diverse resource-limited settings: AIDS Clinical Trials Group A5271. J. Neurovirol . 22 , 472–478. https://doi.org/10.1007/s13365-015-0415-2 (2016). Nagin, D. S. & Odgers, C. L. Group-Based Trajectory Modeling in Clinical Research. Annu. Rev. Clin. Psychol. 6 , 109–138. https://doi.org/10.1146/annurev.clinpsy.121208.131413 (2010). Mésidor, M., Rousseau, M. C., O’Loughlin, J. & Sylvestre, M. P. Does group-based trajectory modeling estimate spurious trajectories? BMC Med. Res. Methodol. 22 , 194. https://doi.org/10.1186/s12874-022-01622-9 (2022). Proust-Lima, C., Philipps, V. & Liquet, B. Estimation of Extended Mixed Models Using Latent Classes and Latent Processes: The R Package lcmm. J. Stat. Softw. 78 , 1–56. https://doi.org/10.18637/jss.v078.i02 (2017). Schafer, J. L. & Graham, J. W. Missing data: Our view of the state of the art. Psychol. Methods . 7 , 147–177. https://doi.org/10.1037/1082-989X.7.2.147 (2002). Abassi, M. et al. Cerebrospinal fluid biomarkers and HIV-associated neurocognitive disorders in HIV-infected individuals in Rakai. Uganda J. Neurovirol . 23 , 369–375. https://doi.org/10.1007/s13365-016-0505-9 (2017). Mocchetti, I., Bachis, A., Campbell, L. A. & Avdoshina, V. Implementing Neuronal Plasticity in NeuroAIDS: the Experience of Brain-derived Neurotrophic Factor and other Neurotrophic Factors. J. Neuroimmune Pharmacol. 9 , 80–91. https://doi.org/10.1007/s11481-013-9488-y (2014). Bachis, A., Wenzel, E., Boelk, A., Becker, J. & Mocchetti, I. The neurotrophin receptor p75 mediates gp120-induced loss of synaptic spines in aging mice. Neurobiol. Aging . 46 , 160–168. https://doi.org/10.1016/j.neurobiolaging.2016.07.001 (2016). Tikhonova, M. A. et al. Neurospecific Molecules Measured in Periphery in Humans: How Do They Correlate with the Brain Levels? A Systematic Review. Int. J. Mol. Sci. 23 https://doi.org/10.3390/ijms23169193 (2022). Cain, S. W. et al. Circadian Rhythms in Plasma Brain-derived Neurotrophic Factor Differ in Men and Women. J. Biol. Rhythms . 32 , 75–82. https://doi.org/10.1177/0748730417693124 (2017). Atluri, V. S. R. et al. Effect of human immunodeficiency virus on blood-brain barrier integrity and function: an update. Front. Cell. Neurosci. 9 , 212. https://doi.org/10.3389/fncel.2015.00212 (2015). Tan, C. S. & Reeves, R. K. HIV on the brain: is neurosignalling damage irreversible even on antiretroviral therapy? AIDS 35 , 1503–1504. https://doi.org/10.1097/QAD.0000000000002957 (2021). Maxi, J. K. et al. Antiretroviral therapy administration reduces neuroinflammation without restoring brain-derived neurotrophic factor signaling in alcohol-administered simian immunodeficiency virus-infected macaques. AIDS 35 (2021). https://journals.lww.com/aidsonline/fulltext/2021/07150/antiretroviral_therapy_administration_reduces.2.aspx Narsi, K., Tomita, A. & Ramlall, S. Cognitive Reserve and Its Determinants in Newly HIV Diagnosed Antiretroviral-Naive Adults From Periurban and Informal Settlements: Evidence From an HIV Hyperendemic South African Setting. J. Acquir. Immune Defic. Syndr. 85 , 387–393. https://doi.org/10.1097/QAI.0000000000002457 (2020). Gravesteijn, E., Mensink, R. P. & Plat, J. Effects of nutritional interventions on BDNF concentrations in humans: a systematic review. Nutr. Neurosci. 25 , 1425–1436. https://doi.org/10.1080/1028415X.2020.1865758 (2022). Szuhany, K. L. & Otto, M. W. Assessing BDNF as a mediator of the effects of exercise on depression. J. Psychiatr Res. 123 , 114–118. https://doi.org/10.1016/j.jpsychires.2020.02.003 (2020). Khoramipour, K. et al. Combined training in patient with aids: improved quality of life and preserved BDNF. Sport Sci. Health . 18 , 39–46. https://doi.org/10.1007/s11332-021-00771-6 (2022). De la Rosa, A. et al. Long-term exercise training improves memory in middle-aged men and modulates peripheral levels of BDNF and Cathepsin B. Sci. Rep. 9 , 3337. https://doi.org/10.1038/s41598-019-40040-8 (2019). Grace, J. M., Semple, S. J. & Combrink, S. Exercise therapy for human immunodeficiency virus/AIDS patients: Guidelines for clinical exercise therapists. J. Exerc. Sci. Fit. 13 , 49–56. https://doi.org/10.1016/j.jesf.2014.10.003 (2015). Zhang, X. Y. et al. The interplay between BDNF and oxidative stress in chronic schizophrenia. Psychoneuroendocrinology 51 , 201–208. https://doi.org/http://dx.doi.org/10.1016/j.psyneuen.2014.09.029 (2015). Buhusi, M., Etheredge, C., Granholm, A. C. & Buhusi, C. V. Increased Hippocampal ProBDNF Contributes to Memory Impairments in Aged Mice, Front. Aging Neurosci. 9 10.3389/fnagi.2017.00284 (2017). https://www.frontiersin.org/articles/ Je, H. S. et al. ProBDNF and Mature BDNF as Punishment and Reward Signals for Synapse Elimination at Mouse Neuromuscular Junctions, J. Neurosci. 33 9957 LP – 9962. (2013). https://doi.org/10.1523/JNEUROSCI.0163-13.2013 Yang, J. et al. proBDNF negatively regulates neuronal remodeling, synaptic transmission, and synaptic plasticity in hippocampus. Cell. Rep. 7 , 796–806. https://doi.org/10.1016/j.celrep.2014.03.040 (2014). Nilsson, J., Ekblom, M., Moberg, M. & Lövdén, M. The role of acute changes in mBDNF, cortisol and pro-BDNF in predicting cognitive performance in old age. Sci. Rep. 13 , 9418. https://doi.org/10.1038/s41598-023-35847-5 (2023). Håkansson, K. et al. Serum proBDNF availability predicts memory gains in the FINGER multimodal lifestyle intervention. Alzheimer’s Dement. 17 , e052946. https://doi.org/https://doi.org/10.1002/alz.052946 (2021). Peng, S., Wuu, J., Mufson, E. J. & Fahnestock, M. Precursor form of brain-derived neurotrophic factor and mature brain-derived neurotrophic factor are decreased in the pre-clinical stages of Alzheimer’s disease. J. Neurochem . 93 , 1412–1421. https://doi.org/10.1111/j.1471-4159.2005.03135.x (2005). Michalski, B. & Fahnestock, M. Pro-brain-derived neurotrophic factor is decreased in parietal cortex in Alzheimer’s disease. Brain Res. Mol. Brain Res. 111 , 148–154. https://doi.org/10.1016/s0169-328x(03)00003-2 (2003). Sun, W. et al. Dorsolateral Striatal proBDNF Improves Reversal Learning by Enhancing Coordination of Neural Activity in Rats. Mol. Neurobiol. 57 , 4642–4656. https://doi.org/10.1007/s12035-020-02051-9 (2020). An, L., Li, X., Tang, C., Xu, N. & Sun, W. Hippocampal proBDNF facilitates place learning strategy associated with neural activity in rats. Brain Struct. Funct. 223 , 4099–4113. https://doi.org/10.1007/s00429-018-1742-x (2018). Diniz, C. R. A. F., Casarotto, P. C., Resstel, L. & Joca, S. R. L. Beyond good and evil: A putative continuum-sorting hypothesis for the functional role of proBDNF/BDNF-propeptide/mBDNF in antidepressant treatment. Neurosci. Biobehav Rev. 90 , 70–83. https://doi.org/https://doi.org/10.1016/j.neubiorev.2018.04.001 (2018). Pettorruso, M. et al. Enhanced peripheral levels of BDNF and proBDNF: elucidating neurotrophin dynamics in cocaine use disorder. Mol. Psychiatry . https://doi.org/10.1038/s41380-023-02367-7 (2024). Erickson, K. I. et al. Brain-derived neurotrophic factor is associated with age-related decline in hippocampal volume. J. Neurosci. Off J. Soc. Neurosci. 30 , 5368–5375. https://doi.org/10.1523/JNEUROSCI.6251-09.2010 (2010). Dogliotti, G., Galliera, E., Licastro, F. & Corsi, M. M. Age-related changes in plasma levels of BDNF in Down syndrome patients. Immun. Ageing . 7 , 2. https://doi.org/10.1186/1742-4933-7-2 (2010). Míguez-Burbano, M. J. et al. Beyond the Brain: The Role of Brain-Derived Neurotrophic Factor in Viroimmune Responses to Antiretroviral Therapy among People Living with HIV with and without Alcohol Use. J. Int. Assoc. Provid AIDS Care . 13 , 454–460. https://doi.org/10.1177/2325957414535253 (2014). Beeri, M. S. & Sonnen, J. Brain BDNF expression as a biomarker for cognitive reserve against Alzheimer disease progression. Neurology 86 , 702–703. https://doi.org/10.1212/WNL.0000000000002389 (2016). Giacobbo, B. L. et al. Long-term environmental modifications affect BDNF concentrations in rat hippocampus, but not in serum., Behav. Brain Res. 372 , 111965. https://doi.org/10.1016/j.bbr.2019.111965 (2019). Tchakoute, C. T., Rhee, S. Y., Hare, C. B., Shafer, R. W. & Sainani, K. Adherence to contemporary antiretroviral treatment regimens and impact on immunological and virologic outcomes in a US healthcare system. PLoS One . 17 , e0263742. https://doi.org/10.1371/journal.pone.0263742 (2022). Judd, A. et al. Cognitive Function in Young Persons With and Without Perinatal HIV in the AALPHI Cohort in England: Role of Non-HIV-Related Factors. Clin. Infect. Dis. Off Publ Infect. Dis. Soc. Am. 63 , 1380–1387. https://doi.org/10.1093/cid/ciw568 (2016). Tables Tables 1 to 4 are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Tables.docx SUPLEMENTARYMATERIALv3.pdf Cite Share Download PDF Status: Published Journal Publication published 03 May, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Accepted 21 Apr, 2025 Reviews received at journal 20 Apr, 2025 Reviews received at journal 10 Apr, 2025 Reviewers agreed at journal 10 Apr, 2025 Reviewers agreed at journal 09 Apr, 2025 Reviewers invited by journal 08 Apr, 2025 Submission checks completed at journal 07 Apr, 2025 First submitted to journal 05 Apr, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4307577","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":440733447,"identity":"81357aa5-6f53-47d4-87c8-32d08b3e931f","order_by":0,"name":"HENRY MICHAEL","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDElEQVRIiWNgGAWjYDCCA1CaDYgZP1TYyBmAuQYW+LQwNsC0MEucSTM2YGAGaZEgrAUMeNsOJ24Aa2HArYXvePPzBx/33Mnnk25++ECCjTl9O3v/0Q0/CiQY+Nu7E7BpkTxzzLBxxrNnlm0yx4wNCnjYcnf2HGa72QN0mMSZsxuwaTG4kcPYzHPgsAGbRA6bhIQET+6GG8lsN3iAWgwkcglqYf8BVJluANRy8w+RWtgYeBIMEkBabuOzBeSXmTMOPANqSTOWljiQYLjhzGGz2zIGEjy4/AIMsQcfPhy4YyA/I/nhx4///ssbHG98dvPNHxs5/vZerFqg4ACmEA8e5Ti0jIJRMApGwSiAAQDjaGZ6iuA6EgAAAABJRU5ErkJggg==","orcid":"","institution":"University of KwaZulu-Natal","correspondingAuthor":true,"prefix":"","firstName":"HENRY","middleName":"","lastName":"MICHAEL","suffix":""},{"id":440733448,"identity":"43ba3827-1723-4a97-9511-993d460c4fff","order_by":1,"name":"Antony Rapulana","email":"","orcid":"","institution":"Africa Health Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Antony","middleName":"","lastName":"Rapulana","suffix":""},{"id":440733449,"identity":"dea5052d-6cc1-487f-9613-fd0d65f403e7","order_by":2,"name":"Theresa Smit","email":"","orcid":"","institution":"Africa Health Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Theresa","middleName":"","lastName":"Smit","suffix":""},{"id":440733450,"identity":"c43738e5-7b7b-46a7-b614-c1606962bff0","order_by":3,"name":"Njabulo Xulu","email":"","orcid":"","institution":"Africa Health Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Njabulo","middleName":"","lastName":"Xulu","suffix":""},{"id":440733451,"identity":"148c6c3a-af51-4e69-954b-cb8b74c3b669","order_by":4,"name":"Sivapragashini Danaviah","email":"","orcid":"","institution":"Eduvos","correspondingAuthor":false,"prefix":"","firstName":"Sivapragashini","middleName":"","lastName":"Danaviah","suffix":""},{"id":440733452,"identity":"5d62aa14-84c7-44f0-9ea7-105aba28725c","order_by":5,"name":"Suvira Ramlall","email":"","orcid":"","institution":"University of KwaZulu-Natal","correspondingAuthor":false,"prefix":"","firstName":"Suvira","middleName":"","lastName":"Ramlall","suffix":""},{"id":440733454,"identity":"4ee94730-751e-4206-a389-0980847d803b","order_by":6,"name":"Frasia Oosthuizen","email":"","orcid":"","institution":"University of KwaZulu-Natal","correspondingAuthor":false,"prefix":"","firstName":"Frasia","middleName":"","lastName":"Oosthuizen","suffix":""}],"badges":[],"createdAt":"2024-04-22 17:49:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4307577/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4307577/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-99569-6","type":"published","date":"2025-05-03T15:57:31+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80582626,"identity":"5ce06b71-ef86-4f2e-84da-53aac5a1543b","added_by":"auto","created_at":"2025-04-14 23:35:25","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":124784,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted trajectory groups for serum mBDNF. “Peak with gradual decline” group (16.1 %) is represented in red; “Stable Ascent” group (83.9%) is represented in blue. The shaded regions around the trajectories represent the 95% confidence intervals, illustrating the uncertainty or variability in the model's predictions over time. Narrower shading indicates greater confidence in the predicted values. Abbreviation: mBDNF, mature brain-derived neurotrophic factor\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4307577/v1/d1b27a6367cbaa45ff64f88d.png"},{"id":80582308,"identity":"1723552c-772c-47dd-aa8c-9b04fcb4f914","added_by":"auto","created_at":"2025-04-14 23:27:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":102188,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted trajectory groups for serum proBDNF. “Gradual Decline” group (14.2 %) is represented in red; “Gradual Increase” group (85.7%) is represented in blue. The shaded regions around the trajectories represent the 95% confidence intervals, illustrating the uncertainty or variability in the model's predictions over time. Narrower shading indicates greater confidence in the predicted values. Abbreviation: proBDNF, precursor brain derived neurotrophic factor\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4307577/v1/02944d9ab463c2cf7398c7ca.png"},{"id":80582311,"identity":"ced964cc-3ac5-4251-b333-900280f6514d","added_by":"auto","created_at":"2025-04-14 23:27:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":121516,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted trajectory groups for cognitive z-score. “Low Baseline-Slow Improvement” group (24.52 %) is represented in red; “Gradual improvement” group (67.74%) is represented in blue; “Late Surge” group (7.74%) is represented in green. The shaded regions around the trajectories represent the 95% confidence intervals, illustrating the uncertainty or variability in the model's predictions over time. Narrower shading indicates greater confidence in the predicted values. \u0026nbsp;Clinically, the identification of a 'Late Surge' trajectory is of interest as it suggests that cognitive gains may manifest later in treatment for a subset of individuals, highlighting the potential for improvement even among those with initially limited progress.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4307577/v1/a06f5e39d6c80d5f4fe1d4be.png"},{"id":81987764,"identity":"1d4c7cb7-e639-4e5c-a87a-fedce76d32f9","added_by":"auto","created_at":"2025-05-05 16:05:47","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1190044,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4307577/v1/6f53620b-8edd-4105-b467-f3bc5c9c44da.pdf"},{"id":80582873,"identity":"b1091c39-7c3d-44ad-ba07-b26173a92bd1","added_by":"auto","created_at":"2025-04-14 23:43:26","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":30433,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-4307577/v1/ca2212b89ca903da328bbd20.docx"},{"id":80582629,"identity":"cec011c5-29cf-46eb-8d25-3bc474d4e2c2","added_by":"auto","created_at":"2025-04-14 23:35:26","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":213058,"visible":true,"origin":"","legend":"","description":"","filename":"SUPLEMENTARYMATERIALv3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4307577/v1/6431390e4024634236a48a02.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Dual Trajectories of Serum Brain-Derived Neurotrophic Factor and Cognitive Function in People Living with HIV","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eDespite a 69% drop in HIV-related deaths since 2004 due to wider antiretroviral therapy (ART) access[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e], individuals with HIV still face significant risks of comorbidities, notably HIV-associated neurocognitive impairment[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Affecting 43% of the global HIV community[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] and 45% in sub-Saharan Africa[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], this condition stems from various factors such as chronic immune activation, inflammation, ART side effects, opportunistic infections, social determinants, and stigma[\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. HIV-associated neurocognitive impairment compromises treatment adherence and quality of life and increases healthcare demands and costs[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. This underscores the critical need for the discovery and targeting of new biological pathways, and the development of biomarkers for early detection, prevention, tailored interventions, and monitoring treatment effectiveness.\u003c/p\u003e \u003cp\u003eBrain-derived neurotrophic factor (BDNF) is increasingly recognized as a potential biomarker for cognitive changes in various neuropathologic conditions due to its critical roles in synaptic plasticity, neurogenesis, and neuronal differentiation[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Extensive evidence from the general population highlights the benefits of mature BDNF (mBDNF) on cognitive and motor learning via mechanisms like long-term potentiation[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Conversely, mBDNF\u0026rsquo;s precursor, proBDNF, is associated with long-term depression and apoptotic pathways, suggesting a nuanced role in brain function[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Throughout this article, the term \"BDNF\" without a specific prefix is used to refer either to the overall category of BDNF-related proteins or in reference to instances where distinctions among the protein's forms are not made clear in cited studies.\u003c/p\u003e \u003cp\u003eAlthough longitudinal research on BDNF and cognitive function in HIV is limited, various cross-sectional studies have examined the relationship between BDNF levels and cognition among people living with HIV, with inconsistent results. For example, an Italian study reported that higher BDNF levels in serum were linked to slower performance on the Grooved Pegboard Test[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Conversely, a U.S. study found that increased serum BDNF levels improved learning and delayed recall in older African-American individuals living with HIV[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Another study highlighted a nuanced relationship between plasma BDNF levels and cognitive performance in individuals with HIV, suggesting that moderate, not high, BDNF levels might be protective[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe variability in the impact of BDNF on cognitive function across individuals living with HIV may stem from previous studies' lack of distinction between the effects of mBDNF and its precursor, proBDNF. This oversight complicates comparisons with the broader body of literature. Additionally, inadequate consideration of temporal dynamics, potential confounders, cognitive domains assessed, and variability in sample sites and types has likely contributed to inconsistent findings, both within HIV research and more generally[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWhile mixed-effect models are adept at analyzing longitudinal data, they presuppose a uniform trajectory across individuals, which may simplify the intricate dynamics between BDNF and cognitive performance [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Group-based trajectory modeling (GBTM) addresses this limitation by allowing for the identification of distinct subgroups within the population that share similar patterns over time. This method is particularly advantageous for capturing heterogeneity and uncovering co-occurring or divergent trajectories of biomarkers and cognitive outcomes, offering deeper insights into underlying processes. By leveraging GBTM, this study aims to unravel the intertwined trajectories of mBDNF, proBDNF, and cognition, providing a more nuanced understanding of their relationships.\u003c/p\u003e \u003cp\u003eTowards this end, our study adopts Group-based Dual Trajectory Modeling (GBDTM) to explore the intertwined trajectories of mBDNF, proBDNF, and cognitive performance in adults living with HIV in sub-Saharan Africa. GBDTM enhances traditional univariate group-based trajectory modelling by simultaneously examining two distinct outcomes and their evolution over time, thereby capturing their potential comorbidity or heterotypic continuity[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. This method, rooted in finite growth mixture modelling, identifies groups of individuals sharing similar patterns over time, offering a nuanced understanding of the interplay between biological markers and cognitive performance[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Although previous research has shed light on BDNF's impact on patient outcomes[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] and complex cognitive phenotypes in HIV[\u003cspan additionalcitationids=\"CR22 CR23\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], it has not linked these elements. Our study is pioneering in applying GBDTM to unravel the concurrent development trajectories of mBDNF, proBDNF, and cognition among individuals with HIV, marking a significant step forward in understanding their dynamic relationship.\u003c/p\u003e \u003cp\u003eIn this study, we posed four research questions: (1) What are the distinct trajectories of mBDNF, proBDNF, and cognitive performance in a cohort of people living with HIV over 96 weeks post-ART initiation? (2) What proportion of the cohort follows each identified trajectory? (3) What sociodemographic and clinical factors influence the trajectories of mBDNF and proBDNF? (4) How are the trajectories of mBDNF and proBDNF linked to changes in cognitive performance when analyzed together? We hypothesize that an upward trajectory in cognitive performance will correlate with rising levels of mBDNF and a decrease in proBDNF levels.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003eThis study was a secondary analysis of serum specimens and data from the Johannesburg and Harare sites of the ACTG A5199, the International Neurological Study (INS) from 2006\u0026ndash;2009. The ACTG 5199 study evaluated the neurological and neuropsychological effects of three randomly assigned antiretroviral regimens in resource-limited settings and the study methods have been detailed in previous reports[\u003cspan additionalcitationids=\"CR26\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCohort description\u003c/h2\u003e \u003cp\u003eThis analysis focused on ART-na\u0026iuml;ve individuals with CD4 counts under 300, who were part of the ACTG A5175 multinational trial assessing the efficacy and safety of antiretroviral treatments (ClinicalTrials.gov, NCT00084136) [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Participants were required to have a Karnofsky performance score of at least 70 and no more than a week\u0026rsquo;s experience with antiretroviral therapy prior to the study's commencement. Those with recent substance or alcohol abuse, severe psychiatric illnesses, hospitalization within the fortnight preceding study entry, or any condition likely to affect study participation or results interpretation were excluded. Follow-up included neurocognitive and neurological evaluations every 24 weeks over a period of up to 192 weeks, with this study reporting on data and specimens collected within the first 96 weeks.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMeasurements\u003c/h3\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eQuantitative ELISA Analysis of Serum mBDNF and proBDNF Levels\u003c/h2\u003e \u003cp\u003eQuantitative analysis of serum mBDNF and proBDNF levels was conducted using a specific enzyme-linked immunoassay (ELISA) kit, adhering to the standardized protocols provided by Biosensis Pty Ltd., Thebarton, SA, Australia. To maintain consistency and reduce variability between assays, samples, which were preserved at -70\u0026deg;C, were tested in pairs on the same ELISA plate. The intra-assay variability for mBDNF and proBDNF was notably low, recorded at 2.7% and 2.2% respectively, confirming the precision of the measurements. Additionally, the laboratory technicians analyzing the samples were kept unaware of the participants' clinical and cognitive status to eliminate any potential for bias in the results.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eNeuropsychological tests\u003c/h3\u003e\n\u003cp\u003eData collected as part of the ACTG study was accessed and used in this analysis. During the ACTG study, participants underwent a battery of four neuropsychological tests \u0026ndash; grooved pegboard (both hands)[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e], finger tapping (both hands)[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], timed gait[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] and semantic verbal fluency tests[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. These tests were administered primarily by physicians, with training and quality assurance conducted annually, complemented by biannual centralized training sessions. Sites underwent initial quality evaluations before full enrollment, ensuring readiness. Data collection was standardized through translated and back-translated case forms to resolve inconsistencies. Ongoing monitoring included automated range checks and follow-ups on outliers, safeguarding data accuracy and reliability[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. The neuropsychological assessments were conducted at baseline and then followed up every 24 weeks. We utilized neuropsychological data at baseline, 24, 48 and 96 weeks. Test scores were normalized into z-scores, utilizing demographic adjustments (site, age, sex, and education-adjusted) based on data from the International Neurocognitive Normative Study ACTG A5271, encompassing a wide array of at-risk populations[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. These z-scores were then averaged to formulate a composite cognitive z-score, which serves as a measure reflective of cognitive performance, with higher scores denoting higher cognitive performance.\u003c/p\u003e\n\u003ch3\u003ePredictors\u003c/h3\u003e\n\u003cp\u003eTo investigate predictors of trajectory group membership, we analyzed key demographic and clinical variables. These included age (in years), sex (male/female), years of education, study site (Harare/Johannesburg), baseline CD4 count (cells/mm\u0026sup3;), baseline CD8 count (cells/mm\u0026sup3;), CD4/CD8 ratio, and log-transformed plasma HIV RNA levels. These variables were chosen based on their relevance to HIV-related neurocognitive outcomes and their potential to influence neurotrophic factors. Following the identification of trajectory groups, regression models were employed to determine the associations between these predictors and trajectory group membership.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eWe included participants who had completed at least two neuropsychological assessments across the study period. To explore distinct patterns over time, we applied group-based trajectory modelling (GBTM) to each variable (log-transformed serum mBDNF, log-transformed proBDNF, and the composite cognitive z-score) using an iterative approach to determine the optimal number and shape (linear, quadratic, or cubic) of trajectories[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Log transformation was applied to mBDNF and proBDNF values to normalize their distributions prior to modelling. GBTM assumes that the population comprises discrete subgroups that follow distinct developmental trajectories. The selection of the best-fitting model for mBDNF, proBDNF, and the composite cognitive z-sore was grounded in a comprehensive evaluation of model fit diagnostics and a set of criteria aimed at minimizing the risk of identifying spurious groups [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. These criteria included the Bayesian Information Criterion (BIC) for model fit, with higher BIC values signifying a better fit; an Average Posterior Probability (APP) of group assignments above 0.70; a minimum group membership probability of about 5% of the sample; odds of correct classification exceeding 5; tight 99% confidence intervals around estimated group memberships; and evidence of statistically significant groups (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), along with the requisites of graphical clarity, clinical relevance, and theoretical interpretability.\u003c/p\u003e \u003cp\u003eSubsequently, participants were categorized into trajectory groups based on their maximum posterior probability derived from the final models. We conducted descriptive analyses to outline baseline characteristics of the cohort and further analyzed differences in demographic and HIV-related variables across mBDNF and proBDNF trajectory groups using Wilcoxon rank-sum tests for continuous variables and chi-square tests for categorical variables. Following these univariate comparisons, multivariable logistic or multinomial regression was utilized to ascertain significant predictors of group membership, setting a significance threshold at P\u0026thinsp;\u0026lt;\u0026thinsp;.05.\u003c/p\u003e \u003cp\u003eAdditionally, dual trajectory models elucidated the relationship between BDNF levels and cognitive performance, presenting conditional and joint probabilities to highlight the mutual influence of mBDNF and the composite cognitive z-score, as well as proBDNF and the composite cognitive z-score trajectories[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. This comprehensive analysis leverages the \u0026ldquo;lcmm\u0026rdquo; package[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e] within R software version 4.3.1, facilitating an in-depth exploration of the longitudinal relationships between neurotrophic factors and cognitive trajectories in our cohort. The annotated R code for the dual trajectory analysis is available in Supplementary Material S1.\u003c/p\u003e \u003c/div\u003e"},{"header":"RESULTS","content":"\u003cp\u003eOf the 158 participants with baseline neuropsychological data and serum samples, 4 were excluded due to lack of follow-up, yielding a baseline analytical sample of 154 participants (Supplementary Material S2). In our statistical analysis, we incorporated data from participants who completed at least two neuropsychological assessments. Missing data were assessed for key variables including serum mBDNF, proBDNF, and the composite cognitive z-score. Specifically, missing values for mBDNF ranged from 1 at baseline to 27 at week 96, while for proBDNF, missing values ranged from 11 at week 24 to 47 at week 96. Notably, no missing values were observed for the composite cognitive z-score. The models incorporated all available data using maximum likelihood estimation, a robust approach for longitudinal analyses, ensuring reliable and unbiased parameter estimation despite the presence of missing data[37]. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe median baseline age was 35 years with an interquartile range (IQR) of 10 years; the majority, 66.2%, were female, and the median educational attainment was 11 years (IQR: 3). The baseline median cognitive z-score was 0.19 with an IQR of 1.01 [Table1, Panel A].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrajectories of Serum mBDNF\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on fit indices and our a priori criteria, we identified two trajectories of mBDNF over 96 weeks post-ART. The best-fitting model was a two-group cubic model, with trajectory shapes that captured a “Stable ascent” (N = 129, 83.9%) and a “Peak with gradual decline” (N = 25, 16.1%) (Figure 1). Supplementary material S3 provides detailed information on model fit indices and diagnostics of assignment accuracy. Table 1 (Panel B) presents baseline descriptive statistics for the mBDNF trajectory groups. Participants in the stable ascent group tended to be younger, with lower CD8 cell counts at baseline and higher mBDNF levels than those in the \"peak with gradual decline\" group. However, there were no significant differences between the groups regarding years of education, sex, plasma viral load, CD4 count, and baseline cognitive performance.\u003c/p\u003e\n\u003cp\u003eFurthermore, the results of multivariable logistic regression analysis, as shown in Supplementary material S4, indicate that baseline age, CD8 count, and baseline serum mBDNF were significant predictors of mBDNF trajectory group membership.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrajectories of Serum ProBDNF\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe identified two trajectories of serum proBDNF over 96 weeks post-ART that best fit the data. The optimal model was a two-group cubic model, resulting in trajectories labelled as “Gradual Increase” (N = 132, 85.7%) and “Gradual Decline” (N = 22, 14.3%) (Figure 2). Supplementary material S5 provides detailed information on model fit indices and diagnostics of assignment accuracy. Table 1 (Panel C) presents baseline descriptive statistics for the proBDNF trajectory groups. Participants were only significantly different (p \u0026lt; 0.05) based on study site. Furthermore, study site remained significant (Log odds -0.07, Std error = 0.04, p = 0.01) in the multivariable logistic regression model (Supplementary material S6).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTrajectories of Cognitive Performance (Composite Cognitive Z-scores)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur analysis delineated three primary composite cognitive z-score trajectories that resonate with the patterns observed in our data. The best-fitting model was a three-group cubic model, reflecting the following patterns: “Low Baseline–Slow Improvement” (N = 36, 24.4%), “Gradual Improvement” (N = 106, 67.7%), and “Late Surge” (N = 12, 7.7%) (Figure 3). Supplementary Material S7 provides comprehensive details on the model's goodness-of-fit and trajectory classification accuracy. While all three trajectories demonstrated improvement over 96 weeks post-ART, the “Low Baseline–Slow Improvement” group began with the lowest cognitive z-scores at baseline but showed steady gains, converging closely with the “Late Surge” group by approximately week 48.\u003c/p\u003e\n\u003cp\u003eTable 2 delineates the baseline characteristics distinguishing the cognitive trajectory groups. Notably, the \"Low Baseline-Slow Improvement\" group was significantly younger and had higher serum proBDNF levels compared to the \"Gradual Improvement\" group (p = 0.02). This group also displayed considerably lower composite cognitive z-scores relative to both the \"Gradual Improvement\" and \"Late Surge\" groups (p \u0026lt; 0.001). Multivariable logistic regression further identified baseline age (Log odds = -0.21, Std error = 0.07, p \u0026lt; 0.01) and baseline composite cognitive z-scores (Log odds = -1.89, Std error = 0.61, p \u0026lt; 0.001) as significant variables differentiating the \"Low Baseline-Slow Improvement\" group from the \"Gradual Improvement” and \"Late Surge\" groups, underscoring the importance of these factors in predicting cognitive trajectory post-ART initiation (Supplementary material S8).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInterrelationships Across the Trajectory Groups of Serum mBDNF and Cognitive performance\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn examining the association between mBDNF trajectories and cognitive trajectories, distinct conditional probabilities emerged. Table 3a delineates the conditional probabilities of cognitive performance based on serum mBDNF trajectories. It shows individuals with a peak and decline mBDNF trajectory have a 68% probability of gradual cognitive improvement, with none exhibiting a late surge. Those with a stable ascent mBDNF trajectory show a 68% probability of gradual cognitive improvement, a lower probability of 22.52% for low baseline-slow improvement, and a 9.5% probability of a late surge.\u003c/p\u003e\n\u003cp\u003eTable 3b reflects the conditional probabilities of serum mBDNF trajectories given the cognitive trajectory groups. In the group with low baseline and slow improvement in composite cognitive z-scores, a 22.4% chance existed of following a peak and decline pattern in mBDNF levels, whereas a stable rise in mBDNF levels was highly probable at 77.6%. Within the gradual improvement cognitive trajectory, 16.9% are likely to have a peak and decline mBDNF trajectory, but a higher probability of 83.1% is observed for a stable ascent mBDNF trajectory. Notably, every individual within the late surge cognitive trajectory group is associated with a stable ascent mBDNF pattern, as indicated by the 100% probability.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 3c displays the joint probabilities for serum mBDNF and cognitive trajectories. The table highlights that individuals with a stable ascent mBDNF trajectory have the highest joint probability of 56.5% for being in the gradual improvement cognitive trajectory, a lower probability of 18.7% for the low baseline-slow improvement group, and 7.9% for the late surge group. For the peak and decline mBDNF trajectory, the joint probability is 5.4% for the low baseline-slow improvement group and 11.5% for the gradual improvement group, with a 0% probability for the late surge cognitive trajectory, indicating no occurrence of this mBDNF pattern in the late surge cognitive category.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInterrelationships Across the Trajectory Groups of Serum proBDNF and Cognitive Performance\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable 4a provides the conditional probabilities of cognitive trajectories based on serum proBDNF levels. Individuals with a Gradual Decline in proBDNF have a 72.4% probability of falling into the Gradual improvement cognitive trajectory, while the probability of them experiencing a Late surge is 0%. For those with a Gradual Increase in proBDNF, there is a 67.7% probability for Gradual improvement and a notably lower 22.9% for Low Baseline-Slow improvement, with a small 9.4% for Late surge.\u003c/p\u003e\n\u003cp\u003eTable 4b shows the conditional probabilities of serum proBDNF trajectories given the cognitive trajectory group. In the Low Baseline-Slow improvement cognitive trajectory, 82.7% of individuals are characterized by a Gradual Increase in proBDNF, while only 17.3% show a Gradual Decline. The Gradual improvement cognitive group has a high probability of 84.4% for a Gradual Increase in proBDNF and a lower probability of 15.6% for a Gradual Decline. The Late surge trajectory is exclusively linked to a Gradual Increase in proBDNF, with a probability of 100%.\u003c/p\u003e\n\u003cp\u003eTable 4c presents the joint probabilities of serum proBDNF levels and cognitive trajectories. The highest joint probability is for a Gradual Increase in proBDNF with the Gradual improvement cognitive group at 57.7%, while the lowest is the absence of any individuals (0%) with a Gradual Decline in proBDNF in the Late surge cognitive trajectory. Additionally, there is a moderate joint probability of 19.5% for a Gradual Increase in proBDNF with the Low Baseline-Slow improvement cognitive group and a slight 8% with the Late surge group. The Gradual Decline proBDNF trajectory is least common, with a joint probability of 4.1% for the Low Baseline-Slow improvement group and 10.7% for the Gradual improvement group.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInterrelationships Between Serum mBDNF and proBDNF Trajectories\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary Material S9 provides detailed insights into the inter-relationship between serum mBDNF and proBDNF trajectories through conditional and joint probabilities. Table S7a presents the conditional probabilities of serum proBDNF trajectories based on mBDNF trajectories. Among individuals in the Stable Ascent mBDNF trajectory, 82.7% were associated with the Gradual Increase proBDNF trajectory, while 17.3% followed the Gradual Decline proBDNF trajectory. Conversely, for those in the Peak with Gradual Decline mBDNF trajectory, 86.2% belonged to the Gradual Decline proBDNF trajectory, and only 13.8% were aligned with the Gradual Increase proBDNF trajectory.\u003c/p\u003e\n\u003cp\u003eTable S7b reflects the reverse conditional probabilities of mBDNF trajectories given proBDNF trajectories. Individuals in the Gradual Decline proBDNF trajectory had a 27.6% probability of being in the Stable Ascent mBDNF trajectory and a 72.4% probability of following the Peak with Gradual Decline mBDNF trajectory. In contrast, among those in the Gradual Increase proBDNF trajectory, 77.1% were likely to belong to the Stable Ascent mBDNF trajectory, while 22.9% followed the Peak with Gradual Decline mBDNF trajectory.\u003c/p\u003e\n\u003cp\u003eTable S7c displays the joint probabilities of trajectory membership for serum mBDNF and proBDNF. The most frequent trajectory pairing was Stable Ascent mBDNF with a Gradual Increase in proBDNF, occurring in 70.3% of participants. This was followed by a Stable Ascent of mBDNF with a Gradual Decline in proBDNF (12.7%). The least common pairings were Peak with Gradual Decline mBDNF with either Gradual Increase proBDNF (2%) or Gradual Decline proBDNF (14.9%).\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThis study is the first to examine the dynamic interplay between serum mBDNF and proBDNF and cognitive trajectories in people living with HIV after antiretroviral therapy initiation. Over 96 weeks, we identified two distinct serum mBDNF trajectories\u0026mdash;stable ascent and peak with gradual decline\u0026mdash;and two for proBDNF\u0026mdash;gradual increase and gradual decline\u0026mdash;each associated with three cognitive trajectories: low baseline with slow improvement, gradual improvement, and late surge. Notably, stable mBDNF ascent and gradual proBDNF increases were more commonly observed among participants with gradual cognitive improvement, while elevated proBDNF levels appeared in those with late cognitive surges. These findings suggest that mBDNF and proBDNF levels may be linked to neurocognitive improvement following ART initiation.\u003c/p\u003e \u003cp\u003eOur analysis reveals an association between serum mBDNF trajectories and cognitive trajectory, particularly noting that a stable ascent in mBDNF levels correlates with better cognitive trajectories. A stable ascent in mBDNF levels was predominantly observed in participants who demonstrated gradual cognitive improvement and in those experiencing late cognitive surges. This pattern may reflect the neurotrophic support provided by increasing mBDNF levels. Conversely, the trajectory of peaking followed by a gradual decline in mBDNF levels shows a lower likelihood of cognitive gains and is absent among those with late surges, suggesting its potential as an indicator of limited cognitive improvement. These observations are consistent with the proposed role of mBDNF in supporting neuroplasticity and counteracting neurodegeneration triggered by viral proteins[\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe relationship between peripheral BDNF levels and central nervous system (CNS) function is complex. BDNF is primarily synthesized in the CNS but is also present peripherally and can cross the blood-brain barrier (BBB) in both directions[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Studies suggest that peripheral BDNF levels may reflect central BDNF activity, though this correlation can be influenced by various factors[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. In the context of HIV, the virus can compromise BBB integrity, potentially altering BDNF transport dynamics[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Additionally, ART regimens have been associated with changes in BDNF signalling pathways[\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. While ART effectively reduces viral load and neuroinflammation, it may not fully restore BDNF signalling deficits, which may contribute to persistent neurocognitive challenges in people living with HIV[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. These factors underscore the intricate interplay between peripheral BDNF dynamics, ART, and CNS neuroplasticity in PLWH.\u003c/p\u003e \u003cp\u003eThis observation emphasizes the need for carefully designed and monitored cognitive enhancement strategies for people with HIV, particularly in settings with limited resources[\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. It points out the importance of methods that ensure a sustained increase in mBDNF levels for ongoing cognitive improvement, through various means such as exercise, nutritional supplements, or cognitive training[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. Evidence, including studies focused on HIV, shows that regular exercise boosts BDNF levels and response[\u003cspan additionalcitationids=\"CR49\" citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. However, personalizing these interventions to match individual preferences, abilities, the nature and intensity of the activity, and considering time and resource availability, is essential for preserving these cognitive gains over time[\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eUnexpectedly, we found that \"gradual increase\" trajectories in serum proBDNF were linked to positive cognitive trajectories, diverging from our original hypothesis and the traditional understanding of proBDNF's role in cognitive decline through mechanisms like long-term depression and neuronal atrophy[\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. This finding contrasts with proBDNF's known effects on spine retraction, dendritic pruning, and neuronal death[\u003cspan additionalcitationids=\"CR54\" citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]. However, aligning with our previous reports and supported by both preclinical and clinical studies, improved cognitive outcomes have been observed when increases in mBDNF are accompanied by rises in proBDNF levels[\u003cspan additionalcitationids=\"CR57 CR58 CR59 CR60 CR61 CR62\" citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e]. This indicates that moderate increases in proBDNF levels could indicate ongoing neural remodelling and serve as a source for mBDNF production, potentially explaining its link to cognitive improvement[\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e]. Additionally, the beneficial impact of rising serum proBDNF levels might be most evident in the early phases of cognitive decline, suggesting a phase-specific role[\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. This emerging evidence suggests that the interaction between proBDNF and mBDNF in cognitive function may be more complex than the \"yin-yang\" hypothesis of opposing roles, leaning towards a nuanced understanding within the \u0026ldquo;continuum-sorting\u0026rdquo; hypothesis that emphasizes homeostatic balance[\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e, \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAcross 96 weeks, specific covariates appeared to influence mBDNF, proBDNF, and cognitive trajectories. The stable ascent in mBDNF levels was more common among younger individuals, those with lower CD8 counts, and those with higher baseline mBDNF, suggesting a possible association between immunological status, neurotrophic support, and age. Younger individuals may possess greater neuroplastic capabilities, reflected in their mBDNF profiles[\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e]. In addition, lower CD8 counts, indicative of reduced immune activation or inflammation, could provide a more conducive environment for mBDNF production and function[\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e]. Starting with higher mBDNF levels suggests an already established robust neurotrophic framework, essential for ongoing neuroplastic processes, potentially serving as a biomarker for cognitive reserve amidst HIV-associated cognitive challenges[\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e]. This relationship underscores the potential significance of considering biological and immunological factors in understanding mBDNF trajectories.\u003c/p\u003e \u003cp\u003eOur analysis revealed two main serum proBDNF trajectories: \"Gradual Increase\" and \"Gradual Decline.\" Study site emerged as a critical factor influencing these trajectories, suggesting that location-specific factors, possibly environmental enrichment or variations in clinical practices, may affect proBDNF dynamics. Logistic regression analysis further confirmed the study site's pivotal role in determining proBDNF trajectories. These findings highlight the need to account for site-level confounders in future research on HIV-related cognitive outcomes[[\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAll cognitive trajectory groups in our study demonstrated improvement over time, consistent with prior evidence on the neurocognitive benefits of sustained ART in individuals with acute HIV infection [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Our analysis revealed that the \"Low Baseline-Slow Improvement\" group, characterized by a younger demographic and lower initial cognitive scores, showed a slower improvement trajectory. This was unexpected, as younger age is typically associated with greater neuroplasticity. One possible explanation is that younger individuals may face greater challenges with ART adherence, as reported in prior studies, which could blunt cognitive gains despite their biological advantage[\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e]. Given our cohort's relatively young average age (39 years) and age-adjusted NPZ-6 scores, the data suggests that slightly older individuals within this young cohort may possess greater cognitive reserve, enabling them to achieve quicker cognitive gains post-ART. This reserve and an increased ability to develop compensatory mechanisms could explain the more rapid cognitive improvements observed post-ART in these slightly older individuals. This suggests a complex interplay between age, baseline cognitive status, and the trajectory of cognitive improvement in HIV[\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOur study's methodology, employing a dual group-based modelling approach, represents a significant strength by probabilistically linking serum BDNF isoforms and cognitive performance trajectories, offering a nuanced view of longitudinal data. While the identified trajectories represent a spectrum of an underlying continuous phenomenon, caution is necessary for drawing causal inferences. The repeated measurement of serum mBDNF and proBDNF adds specificity to understanding their neurocognitive effects over time in people with HIV. However, the absence of tissue-level or neuroimaging data leaves the direct functional impact of BDNF on cognition uncertain. Future integration of neuroimaging could help clarify these mechanisms and link peripheral BDNF dynamics to brain-level activity.\u003c/p\u003e \u003cp\u003eThese findings contribute to the understanding of BDNF dynamics in HIV-associated neurocognitive impairment. The stable ascent of mBDNF suggests its potential as a biomarker for identifying individuals at risk of cognitive decline or tracking improvement. Exploring how mBDNF and proBDNF interact with HIV-specific factors, such as chronic inflammation and immune activation, could guide personalized interventions to sustain cognitive improvement. Approaches like tailored exercise programs or nutritional strategies may enhance neuroplasticity and improve long-term outcomes.\u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eThe specific profile of our cohort\u0026mdash; (young individuals living with HIV in a resource-limited setting, significantly immunosuppressed, part of a larger clinical trial and without significant comorbidities)\u0026mdash;may restrict the generalizability of our findings. Moreover, since participants were recruited from two urban centers with relatively well-established clinical infrastructure, the applicability of these results to rural or less-resourced settings across sub-Saharan Africa may be limited. Additionally, a larger sample size may have unveiled more detailed group distinctions and supported more comprehensive models that could include a broader range of confounders. While our models accounted for several key demographic and clinical variables, there are additional covariates\u0026mdash;such as inflammatory markers, ART adherence, psychosocial stressors, substance use history, ART regimen, depression, and nutritional status\u0026mdash;that may influence BDNF expression and cognitive outcomes but were not included in this analysis. As such, residual confounding from unmeasured or excluded variables remains a possibility and may affect the interpretation of some observed relationships.\u003c/p\u003e \u003cp\u003eThe neuropsychological battery employed in this study focused predominantly on motor and psychomotor speed, due to practical constraints related to testing time, staffing, and available infrastructure in the resource-limited setting. While these measures were feasible and clinically relevant, they do not capture higher-order cognitive domains such as memory, attention, and executive function, which are more frequently affected in HIV-associated neurocognitive disorders. This limited scope reduces the ability to fully interpret the cognitive significance of observed BDNF trajectories. Future studies should incorporate more comprehensive neurocognitive assessments and clinical criteria to better characterize the breadth of cognitive outcomes in this population.\u003c/p\u003e \u003cp\u003eFinally, relying on serum BDNF measurements rather than CNS-specific assessments limits the ability to directly infer CNS activity, as serum levels may not fully reflect central neurotrophic dynamics.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eImplications for Clinical Practice and Future Research\u003c/h2\u003e \u003cp\u003eThis study underscores the potential of mBDNF and proBDNF as biomarkers for monitoring cognitive trajectories in people with HIV. Clinically, the findings suggest prioritizing interventions that sustain mBDNF levels, such as tailored exercise programs, nutritional strategies, or cognitive training, which have shown promise in enhancing neuroplasticity. Personalized approaches are particularly crucial in resource-limited settings to maximize cognitive gains post-ART initiation.\u003c/p\u003e \u003cp\u003eFuture research should validate these findings in larger, diverse cohorts and explore mechanisms linking peripheral BDNF dynamics with brain-level activity through neuroimaging and CNS-specific assessments. Additionally, the role of site-specific and biological factors, such as immune activation and inflammation, in shaping BDNF trajectories warrants further investigation to refine intervention strategies. Such efforts could lead to targeted, individualized care aimed at improving neurocognitive resilience and long-term outcomes in HIV populations.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics Approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of University of Kwazulu-Natl, Durban South Africa (August 18, 2022/Number BREC/00003101/2021), University of the Witwatersrand, Johannesburg, South Africa (July 28, 2021/04081), and Medical Research Council of Zimbabwe, Harare, Zimbabwe (September 10, 2021/MRCZ/E/299).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInformed consent was obtained from all individual participants included in the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Publish\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors confirm that participants provided consent for publication. The data analyzed was anonymized and presented collectively.\u003c/p\u003e\n\u003ch2\u003eCompeting Interest\u003c/h2\u003e\n\u003cp\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eResearch reported in this publication was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under Award Number UM1 AI068634, UM1 AI068636 and UM1 AI106701. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Henry Michael received research support from the College of Health Sciences, University of KwaZulu-Natal, Durban, South Africa.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eAll authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by HM, AR and NX. TS, SR, SD and FO supervised all aspects of the project. The first draft of the manuscript was written by HM and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eThe authors thank the study investigators and participants of the ACTG A5175, A5199, and A5271 studies for providing the samples and data for this work. The authors also thank the support staff at the African Health Research Institute for sample management and testing.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe authors confirm that all data are fully available upon request from [email protected] with the written agreement of Advancing Clinical Therapeutics Globally for HIV/AIDS and Other Infections (ACTG). Study specimens were obtained from the ACTG Repository (www.specimenrepository.org).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eUNAIDS \u0026amp; Global, H. I. V. \u0026amp; AIDS statistics \u0026mdash; Fact sheet, UNAIDS 2023 Epidemiol. Estim. (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.unaids.org/en/resources/fact-sheet\u003c/span\u003e\u003cspan address=\"https://www.unaids.org/en/resources/fact-sheet\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (accessed March 5, 2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLerner, A. M., Eisinger, R. W. \u0026amp; Fauci, A. S. Comorbidities in Persons With HIV: The Lingering Challenge. \u003cem\u003eJAMA\u003c/em\u003e \u003cb\u003e323\u003c/b\u003e, 19\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1001/jama.2019.19775\u003c/span\u003e\u003cspan address=\"10.1001/jama.2019.19775\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, Y. et al. Global prevalence and burden of HIV-associated neurocognitive disorder. \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e95\u003c/b\u003e, e2610\u0026ndash;e2621. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1212/WNL.0000000000010752\u003c/span\u003e\u003cspan address=\"10.1212/WNL.0000000000010752\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZenebe, Y., Akele, B., Selassie, M. W. \u0026amp; Necho, M. Neurocognitive Impairment and Associated Factors Among People Living with HIV: A Systematic Review and Meta-Analysis of African Studies. \u003cem\u003eNeuropsychiatr Dis. Treat.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 673\u0026ndash;687. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2147/NDT.S377636\u003c/span\u003e\u003cspan address=\"10.2147/NDT.S377636\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcArthur, J. C. \u0026amp; Johnson, T. P. Chronic inflammation mediates brain injury in HIV infection: relevance for cure strategies. \u003cem\u003eCurr. Opin. Neurol.\u003c/em\u003e \u003cb\u003e33\u003c/b\u003e, 397\u0026ndash;404. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/WCO.0000000000000807\u003c/span\u003e\u003cspan address=\"10.1097/WCO.0000000000000807\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarokar, J. et al. Neurocognitive impairment is worse in HIV/HCV-coinfected individuals with liver dysfunction. \u003cem\u003eJ. Neurovirol\u003c/em\u003e. \u003cb\u003e25\u003c/b\u003e, 792\u0026ndash;799. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13365-019-00767-6\u003c/span\u003e\u003cspan address=\"10.1007/s13365-019-00767-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThompson, E. C. et al. Internalized HIV-Related Stigma and Neurocognitive Functioning Among Women Living with HIV. \u003cem\u003eAIDS Patient Care STDS\u003c/em\u003e. \u003cb\u003e36\u003c/b\u003e, 336\u0026ndash;342. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/apc.2022.0041\u003c/span\u003e\u003cspan address=\"10.1089/apc.2022.0041\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMekuriaw, B., Belayneh, Z., Teshome, W. \u0026amp; Akalu, Y. Prevalence and variability of HIV/AIDS-associated neurocognitive impairments in Africa: a systematic review and meta-analysis. \u003cem\u003eBMC Public. Health\u003c/em\u003e. \u003cb\u003e23\u003c/b\u003e, 997. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12889-023-15935-x\u003c/span\u003e\u003cspan address=\"10.1186/s12889-023-15935-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNicastri, C. M. et al. BDNF mediates improvement in cognitive performance after computerized cognitive training in healthy older adults., Alzheimer\u0026rsquo;s Dement. (New York, N. Y.) 8 e12337. (2022). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/trc2.12337\u003c/span\u003e\u003cspan address=\"10.1002/trc2.12337\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMizoguchi, Y., Yao, H., Imamura, Y., Hashimoto, M. \u0026amp; Monji, A. Lower brain-derived neurotrophic factor levels are associated with age-related memory impairment in community-dwelling older adults: the Sefuri study. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 16442. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-020-73576-1\u003c/span\u003e\u003cspan address=\"10.1038/s41598-020-73576-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCechova, K. et al. Ratio of serum proBDNF to BDNF and its association with cognitive performance and brain morphometry in mild cognitive impairment. \u003cem\u003eAlzheimer\u0026rsquo;s Dement.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e, e046340. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1002/alz.046340\u003c/span\u003e\u003cspan address=\"10.1002/alz.046340\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFalasca, K. et al. Cytokines, Hepatic Fibrosis, and Antiretroviral Therapy Role in Neurocognitive Disorders HIV Related. \u003cem\u003eAIDS Res. Hum. Retroviruses\u003c/em\u003e. \u003cb\u003e33\u003c/b\u003e, 246\u0026ndash;253. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1089/AID.2016.0138\u003c/span\u003e\u003cspan address=\"10.1089/AID.2016.0138\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFazeli, P. L. et al. Differential Associations Between BDNF and Memory Across Older Black and White Adults With HIV Disease, JAIDS J. Acquir. \u003cem\u003eImmune Defic. Syndr.\u003c/em\u003e \u003cb\u003e89\u003c/b\u003e (2022). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://journals.lww.com/jaids/fulltext/2022/02010/differential_associations_between_bdnf_and_memory.2.aspx\u003c/span\u003e\u003cspan address=\"https://journals.lww.com/jaids/fulltext/2022/02010/differential_associations_between_bdnf_and_memory.2.aspx\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026iacute;guez-Burbano, M. J. et al. Brain derived neurotrophic factor and cognitive status: the delicate balance among people living with HIV, with and without alcohol abuse. \u003cem\u003eCurr. HIV Res.\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e, 254\u0026ndash;264. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2174/1570162x12666140721121238\u003c/span\u003e\u003cspan address=\"10.2174/1570162x12666140721121238\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, C. S., Kavalali, E. T. \u0026amp; Monteggia, L. M. BDNF signaling in context: From synaptic regulation to psychiatric disorders. \u003cem\u003eCell\u003c/em\u003e \u003cb\u003e185\u003c/b\u003e, 62\u0026ndash;76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cell.2021.12.003\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2021.12.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXie, H., McHugo, G. J., He, X. \u0026amp; Drake, R. E. Using the Group-Based Dual Trajectory Model to Analyze Two Related Longitudinal Outcomes. \u003cem\u003eJ. Drug Issues\u003c/em\u003e. \u003cb\u003e40\u003c/b\u003e, 45\u0026ndash;61. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/002204261004000104\u003c/span\u003e\u003cspan address=\"10.1177/002204261004000104\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagin, D. S. \u0026amp; Tremblay, R. E. Analyzing developmental trajectories of distinct but related behaviors: a group-based method. \u003cem\u003ePsychol. Methods\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, 18\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/1082-989x.6.1.18\u003c/span\u003e\u003cspan address=\"10.1037/1082-989x.6.1.18\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagin, D. S., Jones, B. L., Passos, V. L. \u0026amp; Tremblay, R. E. Group-based multi-trajectory modeling, Stat. \u003cem\u003eMethods Med. Res.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 2015\u0026ndash;2023. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0962280216673085\u003c/span\u003e\u003cspan address=\"10.1177/0962280216673085\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTreble-Barna, A. et al. Acute Brain-Derived Neurotrophic Factor DNA Methylation Trajectories in Cerebrospinal Fluid and Associations With Outcomes Following Severe Traumatic Brain Injury in Adults., Neurorehabil. \u003cem\u003eNeural Repair.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 790\u0026ndash;800. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/15459683211028245\u003c/span\u003e\u003cspan address=\"10.1177/15459683211028245\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiguez, M. J., Chan, W., Espinoza, L., Tarter, R. \u0026amp; Perez, C. Marijuana use among adolescents is associated with deleterious alterations in mature BDNF. \u003cem\u003eAIMS Public. Heal\u003c/em\u003e. \u003cb\u003e6\u003c/b\u003e, 4\u0026ndash;14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3934/publichealth.2019.1.4\u003c/span\u003e\u003cspan address=\"10.3934/publichealth.2019.1.4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatel, P. B. et al. Trajectory Analysis of Cognitive Outcomes in Children With Perinatal HIV., Pediatr. \u003cem\u003eInfect. Dis. J.\u003c/em\u003e \u003cb\u003e38\u003c/b\u003e, 1038\u0026ndash;1044. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/INF.0000000000002427\u003c/span\u003e\u003cspan address=\"10.1097/INF.0000000000002427\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrouillette, M. J. et al. Identifying Neurocognitive Decline at 36 Months among HIV-Positive Participants in the CHARTER Cohort Using Group-Based Trajectory Analysis. \u003cem\u003ePLoS One\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, e0155766. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0155766\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0155766\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan, P. et al. Cognitive trajectories after treatment in acute HIV infection. \u003cem\u003eAids\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 883\u0026ndash;888. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/QAD.0000000000002831\u003c/span\u003e\u003cspan address=\"10.1097/QAD.0000000000002831\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFitzgerald, K. C. et al. Factors Predicting Detrimental Change in Declarative Memory Among Women With HIV: A Study of Heterogeneity in Cognition. \u003cem\u003eFront. Psychol.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fpsyg.2020.548521\u003c/span\u003e\u003cspan address=\"10.3389/fpsyg.2020.548521\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.frontiersin.org/journals/psychology/articles/\u003c/span\u003e\u003cspan address=\"https://www.frontiersin.org/journals/psychology/articles/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobertson, K. R. et al. Human Immunodeficiency Virus-associated Neurocognitive Impairment in Diverse Resource-limited Settings., \u003cem\u003eClin. Infect. Dis. Off Publ Infect. Dis. Soc. Am.\u003c/em\u003e \u003cb\u003e68\u003c/b\u003e 1733\u0026ndash;1738. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/cid/ciy767\u003c/span\u003e\u003cspan address=\"10.1093/cid/ciy767\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobertson, K. et al. Improved neuropsychological and neurological functioning across three antiretroviral regimens in diverse resource-limited settings: AIDS Clinical Trials Group study a5199, the International Neurological Study. \u003cem\u003eClin. Infect. Dis. Off Publ Infect. Dis. Soc. Am.\u003c/em\u003e \u003cb\u003e55\u003c/b\u003e, 868\u0026ndash;876. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/cid/cis507\u003c/span\u003e\u003cspan address=\"10.1093/cid/cis507\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobertson, K. et al. A multinational study of neurological performance in antiretroviral therapy-na\u0026iuml;ve HIV-1-infected persons in diverse resource-constrained settings. \u003cem\u003eJ. Neurovirol\u003c/em\u003e. \u003cb\u003e17\u003c/b\u003e, 438\u0026ndash;447. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13365-011-0044-3\u003c/span\u003e\u003cspan address=\"10.1007/s13365-011-0044-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCampbell, T. B. et al. Efficacy and safety of three antiretroviral regimens for initial treatment of HIV-1: a randomized clinical trial in diverse multinational settings. \u003cem\u003ePLoS Med.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, e1001290. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pmed.1001290\u003c/span\u003e\u003cspan address=\"10.1371/journal.pmed.1001290\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLafayette Instrument, G. \u0026amp; Pegboard (n.d.). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://lafayetteevaluation.com/products/grooved-pegboard\u003c/span\u003e\u003cspan address=\"https://lafayetteevaluation.com/products/grooved-pegboard\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReitan, R. M. \u0026amp; Tucson, S. The Halstead-Reitan neuropsychological test battery: theory and clinical interpretation, Second edition. Arizona : Neuropsychology Press, [1993] \u0026copy;1993, n.d. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://search.library.wisc.edu/catalog/9910209689302121\u003c/span\u003e\u003cspan address=\"https://search.library.wisc.edu/catalog/9910209689302121\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobertson, K. R. et al. Timed Gait Test: Normative Data for the Assessment of the AIDS Dementia Complex, \u003cem\u003eJ. Clin. Exp. Neuropsychol.\u003c/em\u003e \u003cb\u003e28\u003c/b\u003e 1053\u0026ndash;1064. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13803390500205684\u003c/span\u003e\u003cspan address=\"10.1080/13803390500205684\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStrauss, O., Sherman, E., Spreen \u0026amp; E. M. S., \u0026amp; \u003cem\u003eA compendium of neuropsychological tests: administration, norms, and commentary\u003c/em\u003e 3rd edn (Oxford University Press, 2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobertson, K. et al. International neurocognitive normative study: neurocognitive comparison data in diverse resource-limited settings: AIDS Clinical Trials Group A5271. \u003cem\u003eJ. Neurovirol\u003c/em\u003e. \u003cb\u003e22\u003c/b\u003e, 472\u0026ndash;478. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13365-015-0415-2\u003c/span\u003e\u003cspan address=\"10.1007/s13365-015-0415-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagin, D. S. \u0026amp; Odgers, C. L. Group-Based Trajectory Modeling in Clinical Research. \u003cem\u003eAnnu. Rev. Clin. Psychol.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 109\u0026ndash;138. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1146/annurev.clinpsy.121208.131413\u003c/span\u003e\u003cspan address=\"10.1146/annurev.clinpsy.121208.131413\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026eacute;sidor, M., Rousseau, M. C., O\u0026rsquo;Loughlin, J. \u0026amp; Sylvestre, M. P. Does group-based trajectory modeling estimate spurious trajectories? \u003cem\u003eBMC Med. Res. Methodol.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e, 194. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12874-022-01622-9\u003c/span\u003e\u003cspan address=\"10.1186/s12874-022-01622-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProust-Lima, C., Philipps, V. \u0026amp; Liquet, B. Estimation of Extended Mixed Models Using Latent Classes and Latent Processes: The R Package lcmm. \u003cem\u003eJ. Stat. Softw.\u003c/em\u003e \u003cb\u003e78\u003c/b\u003e, 1\u0026ndash;56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18637/jss.v078.i02\u003c/span\u003e\u003cspan address=\"10.18637/jss.v078.i02\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchafer, J. L. \u0026amp; Graham, J. W. Missing data: Our view of the state of the art. \u003cem\u003ePsychol. Methods\u003c/em\u003e. \u003cb\u003e7\u003c/b\u003e, 147\u0026ndash;177. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1037/1082-989X.7.2.147\u003c/span\u003e\u003cspan address=\"10.1037/1082-989X.7.2.147\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbassi, M. et al. Cerebrospinal fluid biomarkers and HIV-associated neurocognitive disorders in HIV-infected individuals in Rakai. \u003cem\u003eUganda J. Neurovirol\u003c/em\u003e. \u003cb\u003e23\u003c/b\u003e, 369\u0026ndash;375. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13365-016-0505-9\u003c/span\u003e\u003cspan address=\"10.1007/s13365-016-0505-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMocchetti, I., Bachis, A., Campbell, L. A. \u0026amp; Avdoshina, V. Implementing Neuronal Plasticity in NeuroAIDS: the Experience of Brain-derived Neurotrophic Factor and other Neurotrophic Factors. \u003cem\u003eJ. Neuroimmune Pharmacol.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 80\u0026ndash;91. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11481-013-9488-y\u003c/span\u003e\u003cspan address=\"10.1007/s11481-013-9488-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBachis, A., Wenzel, E., Boelk, A., Becker, J. \u0026amp; Mocchetti, I. The neurotrophin receptor p75 mediates gp120-induced loss of synaptic spines in aging mice. \u003cem\u003eNeurobiol. Aging\u003c/em\u003e. \u003cb\u003e46\u003c/b\u003e, 160\u0026ndash;168. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neurobiolaging.2016.07.001\u003c/span\u003e\u003cspan address=\"10.1016/j.neurobiolaging.2016.07.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTikhonova, M. A. et al. Neurospecific Molecules Measured in Periphery in Humans: How Do They Correlate with the Brain Levels? A Systematic Review. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/ijms23169193\u003c/span\u003e\u003cspan address=\"10.3390/ijms23169193\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCain, S. W. et al. Circadian Rhythms in Plasma Brain-derived Neurotrophic Factor Differ in Men and Women. \u003cem\u003eJ. Biol. Rhythms\u003c/em\u003e. \u003cb\u003e32\u003c/b\u003e, 75\u0026ndash;82. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0748730417693124\u003c/span\u003e\u003cspan address=\"10.1177/0748730417693124\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAtluri, V. S. R. et al. Effect of human immunodeficiency virus on blood-brain barrier integrity and function: an update. \u003cem\u003eFront. Cell. Neurosci.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 212. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fncel.2015.00212\u003c/span\u003e\u003cspan address=\"10.3389/fncel.2015.00212\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTan, C. S. \u0026amp; Reeves, R. K. HIV on the brain: is neurosignalling damage irreversible even on antiretroviral therapy? \u003cem\u003eAIDS\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 1503\u0026ndash;1504. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/QAD.0000000000002957\u003c/span\u003e\u003cspan address=\"10.1097/QAD.0000000000002957\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaxi, J. K. et al. Antiretroviral therapy administration reduces neuroinflammation without restoring brain-derived neurotrophic factor signaling in alcohol-administered simian immunodeficiency virus-infected macaques. \u003cem\u003eAIDS\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e (2021). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://journals.lww.com/aidsonline/fulltext/2021/07150/antiretroviral_therapy_administration_reduces.2.aspx\u003c/span\u003e\u003cspan address=\"https://journals.lww.com/aidsonline/fulltext/2021/07150/antiretroviral_therapy_administration_reduces.2.aspx\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNarsi, K., Tomita, A. \u0026amp; Ramlall, S. Cognitive Reserve and Its Determinants in Newly HIV Diagnosed Antiretroviral-Naive Adults From Periurban and Informal Settlements: Evidence From an HIV Hyperendemic South African Setting. \u003cem\u003eJ. Acquir. Immune Defic. Syndr.\u003c/em\u003e \u003cb\u003e85\u003c/b\u003e, 387\u0026ndash;393. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/QAI.0000000000002457\u003c/span\u003e\u003cspan address=\"10.1097/QAI.0000000000002457\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGravesteijn, E., Mensink, R. P. \u0026amp; Plat, J. Effects of nutritional interventions on BDNF concentrations in humans: a systematic review. \u003cem\u003eNutr. Neurosci.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 1425\u0026ndash;1436. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/1028415X.2020.1865758\u003c/span\u003e\u003cspan address=\"10.1080/1028415X.2020.1865758\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSzuhany, K. L. \u0026amp; Otto, M. W. Assessing BDNF as a mediator of the effects of exercise on depression. \u003cem\u003eJ. Psychiatr Res.\u003c/em\u003e \u003cb\u003e123\u003c/b\u003e, 114\u0026ndash;118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jpsychires.2020.02.003\u003c/span\u003e\u003cspan address=\"10.1016/j.jpsychires.2020.02.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhoramipour, K. et al. Combined training in patient with aids: improved quality of life and preserved BDNF. \u003cem\u003eSport Sci. Health\u003c/em\u003e. \u003cb\u003e18\u003c/b\u003e, 39\u0026ndash;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11332-021-00771-6\u003c/span\u003e\u003cspan address=\"10.1007/s11332-021-00771-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe la Rosa, A. et al. Long-term exercise training improves memory in middle-aged men and modulates peripheral levels of BDNF and Cathepsin B. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 3337. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-019-40040-8\u003c/span\u003e\u003cspan address=\"10.1038/s41598-019-40040-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrace, J. M., Semple, S. J. \u0026amp; Combrink, S. Exercise therapy for human immunodeficiency virus/AIDS patients: Guidelines for clinical exercise therapists. \u003cem\u003eJ. Exerc. Sci. Fit.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 49\u0026ndash;56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jesf.2014.10.003\u003c/span\u003e\u003cspan address=\"10.1016/j.jesf.2014.10.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, X. Y. et al. The interplay between BDNF and oxidative stress in chronic schizophrenia. \u003cem\u003ePsychoneuroendocrinology\u003c/em\u003e \u003cb\u003e51\u003c/b\u003e, 201\u0026ndash;208. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/http://dx.doi.org/10.1016/j.psyneuen.2014.09.029\u003c/span\u003e\u003cspan address=\"10.1016/j.psyneuen.2014.09.029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuhusi, M., Etheredge, C., Granholm, A. C. \u0026amp; Buhusi, C. V. Increased Hippocampal ProBDNF Contributes to Memory Impairments in Aged Mice, Front. \u003cem\u003eAging Neurosci.\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fnagi.2017.00284\u003c/span\u003e\u003cspan address=\"10.3389/fnagi.2017.00284\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.frontiersin.org/articles/\u003c/span\u003e\u003cspan address=\"https://www.frontiersin.org/articles/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJe, H. S. et al. ProBDNF and Mature BDNF as Punishment and Reward Signals for Synapse Elimination at Mouse Neuromuscular Junctions, J. Neurosci. 33 9957 LP \u0026ndash; 9962. (2013). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1523/JNEUROSCI.0163-13.2013\u003c/span\u003e\u003cspan address=\"10.1523/JNEUROSCI.0163-13.2013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, J. et al. proBDNF negatively regulates neuronal remodeling, synaptic transmission, and synaptic plasticity in hippocampus. \u003cem\u003eCell. Rep.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 796\u0026ndash;806. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.celrep.2014.03.040\u003c/span\u003e\u003cspan address=\"10.1016/j.celrep.2014.03.040\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNilsson, J., Ekblom, M., Moberg, M. \u0026amp; L\u0026ouml;vd\u0026eacute;n, M. The role of acute changes in mBDNF, cortisol and pro-BDNF in predicting cognitive performance in old age. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 9418. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-023-35847-5\u003c/span\u003e\u003cspan address=\"10.1038/s41598-023-35847-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026aring;kansson, K. et al. Serum proBDNF availability predicts memory gains in the FINGER multimodal lifestyle intervention. \u003cem\u003eAlzheimer\u0026rsquo;s Dement.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, e052946. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1002/alz.052946\u003c/span\u003e\u003cspan address=\"10.1002/alz.052946\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeng, S., Wuu, J., Mufson, E. J. \u0026amp; Fahnestock, M. Precursor form of brain-derived neurotrophic factor and mature brain-derived neurotrophic factor are decreased in the pre-clinical stages of Alzheimer\u0026rsquo;s disease. \u003cem\u003eJ. Neurochem\u003c/em\u003e. \u003cb\u003e93\u003c/b\u003e, 1412\u0026ndash;1421. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1471-4159.2005.03135.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1471-4159.2005.03135.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMichalski, B. \u0026amp; Fahnestock, M. Pro-brain-derived neurotrophic factor is decreased in parietal cortex in Alzheimer\u0026rsquo;s disease. \u003cem\u003eBrain Res. Mol. Brain Res.\u003c/em\u003e \u003cb\u003e111\u003c/b\u003e, 148\u0026ndash;154. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/s0169-328x(03)00003-2\u003c/span\u003e\u003cspan address=\"10.1016/s0169-328x(03)00003-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun, W. et al. Dorsolateral Striatal proBDNF Improves Reversal Learning by Enhancing Coordination of Neural Activity in Rats. \u003cem\u003eMol. Neurobiol.\u003c/em\u003e \u003cb\u003e57\u003c/b\u003e, 4642\u0026ndash;4656. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12035-020-02051-9\u003c/span\u003e\u003cspan address=\"10.1007/s12035-020-02051-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAn, L., Li, X., Tang, C., Xu, N. \u0026amp; Sun, W. Hippocampal proBDNF facilitates place learning strategy associated with neural activity in rats. \u003cem\u003eBrain Struct. Funct.\u003c/em\u003e \u003cb\u003e223\u003c/b\u003e, 4099\u0026ndash;4113. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00429-018-1742-x\u003c/span\u003e\u003cspan address=\"10.1007/s00429-018-1742-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiniz, C. R. A. F., Casarotto, P. C., Resstel, L. \u0026amp; Joca, S. R. L. Beyond good and evil: A putative continuum-sorting hypothesis for the functional role of proBDNF/BDNF-propeptide/mBDNF in antidepressant treatment. \u003cem\u003eNeurosci. Biobehav Rev.\u003c/em\u003e \u003cb\u003e90\u003c/b\u003e, 70\u0026ndash;83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.1016/j.neubiorev.2018.04.001\u003c/span\u003e\u003cspan address=\"10.1016/j.neubiorev.2018.04.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePettorruso, M. et al. Enhanced peripheral levels of BDNF and proBDNF: elucidating neurotrophin dynamics in cocaine use disorder. \u003cem\u003eMol. Psychiatry\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41380-023-02367-7\u003c/span\u003e\u003cspan address=\"10.1038/s41380-023-02367-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErickson, K. I. et al. Brain-derived neurotrophic factor is associated with age-related decline in hippocampal volume. \u003cem\u003eJ. Neurosci. Off J. Soc. Neurosci.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 5368\u0026ndash;5375. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1523/JNEUROSCI.6251-09.2010\u003c/span\u003e\u003cspan address=\"10.1523/JNEUROSCI.6251-09.2010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDogliotti, G., Galliera, E., Licastro, F. \u0026amp; Corsi, M. M. Age-related changes in plasma levels of BDNF in Down syndrome patients. \u003cem\u003eImmun. Ageing\u003c/em\u003e. \u003cb\u003e7\u003c/b\u003e, 2. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/1742-4933-7-2\u003c/span\u003e\u003cspan address=\"10.1186/1742-4933-7-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026iacute;guez-Burbano, M. J. et al. Beyond the Brain: The Role of Brain-Derived Neurotrophic Factor in Viroimmune Responses to Antiretroviral Therapy among People Living with HIV with and without Alcohol Use. \u003cem\u003eJ. Int. Assoc. Provid AIDS Care\u003c/em\u003e. \u003cb\u003e13\u003c/b\u003e, 454\u0026ndash;460. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/2325957414535253\u003c/span\u003e\u003cspan address=\"10.1177/2325957414535253\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeeri, M. S. \u0026amp; Sonnen, J. Brain BDNF expression as a biomarker for cognitive reserve against Alzheimer disease progression. \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e86\u003c/b\u003e, 702\u0026ndash;703. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1212/WNL.0000000000002389\u003c/span\u003e\u003cspan address=\"10.1212/WNL.0000000000002389\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiacobbo, B. L. et al. Long-term environmental modifications affect BDNF concentrations in rat hippocampus, but not in serum., Behav. \u003cem\u003eBrain Res.\u003c/em\u003e \u003cb\u003e372\u003c/b\u003e, 111965. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.bbr.2019.111965\u003c/span\u003e\u003cspan address=\"10.1016/j.bbr.2019.111965\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTchakoute, C. T., Rhee, S. Y., Hare, C. B., Shafer, R. W. \u0026amp; Sainani, K. Adherence to contemporary antiretroviral treatment regimens and impact on immunological and virologic outcomes in a US healthcare system. \u003cem\u003ePLoS One\u003c/em\u003e. \u003cb\u003e17\u003c/b\u003e, e0263742. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0263742\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0263742\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJudd, A. et al. Cognitive Function in Young Persons With and Without Perinatal HIV in the AALPHI Cohort in England: Role of Non-HIV-Related Factors. \u003cem\u003eClin. Infect. Dis. Off Publ Infect. Dis. Soc. Am.\u003c/em\u003e \u003cb\u003e63\u003c/b\u003e, 1380\u0026ndash;1387. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/cid/ciw568\u003c/span\u003e\u003cspan address=\"10.1093/cid/ciw568\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 to 4 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"HIV/AIDS, Cognition, Brain-derived neurotrophic factor, Group-based trajectory modelling, sub-Saharan Africa","lastPublishedDoi":"10.21203/rs.3.rs-4307577/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4307577/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study aimed to identify the interrelationships between mature BDNF (mBDNF), precursor BDNF (proBDNF) trajectories, and cognitive performance in individuals with HIV from sub-Saharan Africa over 96 weeks following antiretroviral therapy (ART) initiation. Using data from 154 participants in the ACTG 5199 study (ClinicalTrials.gov NCT00096824, 2005-06-23) in Johannesburg and Harare (2006\u0026ndash;2009), we measured serum mBDNF and proBDNF levels via ELISA and assessed cognitive performance with neuropsychological tests. Group-based trajectory modelling indicated two mBDNF trajectories\u0026mdash;\"Stable Ascent\" (83.9%) and \"Peak with Gradual Decline\" (16.1%)\u0026mdash;and two proBDNF trajectories\u0026mdash;\"Gradual Increase\" (85.7%) and \"Gradual Decline\" (14.3%). These were linked to three cognitive trajectories: \"Low Baseline-Slow Improvement,\" \"Gradual Improvement,\" and \"Late Surge.\" The \"Stable Ascent\" mBDNF group showed a significant probability of \"Gradual Improvement\" (68%) in cognitive performance and a \"Late Surge\" (9.5%). In contrast, the \"Peak with Gradual Decline\" mBDNF trajectory saw no \"Late Surge.\" A \"Gradual Increase\" in proBDNF corresponded to a 67.7% chance of \"Gradual Improvement\" in cognition. Findings suggest BDNF isoforms as potential biomarkers for cognitive interventions in HIV, emphasizing that stable or increasing BDNF levels post-ART are linked to favourable cognitive outcomes. Further research is needed to develop BDNF-based cognitive health strategies to improve outcomes for people with HIV.\u003c/p\u003e","manuscriptTitle":"Dual Trajectories of Serum Brain-Derived Neurotrophic Factor and Cognitive Function in People Living with HIV","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-14 23:19:21","doi":"10.21203/rs.3.rs-4307577/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Accepted","date":"2025-04-21T11:08:50+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-20T17:41:12+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-10T19:18:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"293817803630588941922553911295368346644","date":"2025-04-10T15:55:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"16828391493550843984648347198393144265","date":"2025-04-09T15:47:27+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-08T10:18:27+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-08T03:36:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-04-06T00:08:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4dd55532-370e-46c4-9f04-b3375300735f","owner":[],"postedDate":"April 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":46925498,"name":"Biological sciences/Neuroscience"},{"id":46925499,"name":"Health sciences/Biomarkers"},{"id":46925500,"name":"Health sciences/Neurology"}],"tags":[],"updatedAt":"2025-05-05T16:01:04+00:00","versionOfRecord":{"articleIdentity":"rs-4307577","link":"https://doi.org/10.1038/s41598-025-99569-6","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-05-03 15:57:31","publishedOnDateReadable":"May 3rd, 2025"},"versionCreatedAt":"2025-04-14 23:19:21","video":"","vorDoi":"10.1038/s41598-025-99569-6","vorDoiUrl":"https://doi.org/10.1038/s41598-025-99569-6","workflowStages":[]},"version":"v1","identity":"rs-4307577","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4307577","identity":"rs-4307577","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}