HIV Articles  
HIV & Alzheimers: are HIV+ At Greater Risk for Alzheimers - pdf full text attached
  Cognitively unimpaired HIV-positive subjects do not have increased 11C-PiB A case-control study
Published online before print June 9, 2010
Download the PDF here
NEUROLOGY 2010;75:111-115
B.M. Ances, MD, PhD, J.J. Christensen, BA, M. Teshome, MD, J. Taylor, BA, C. Xiong, PhD, P. Aldea, BS, A.M. Fagan, PhD, D.M. Holtzman, MD, PhD, J.C. Morris, MD, M.A. Mintun, MD and D.B. Clifford, MD
From the Departments of Neurology (B.M.A., M.T., A.M.F., D.M.H., J.C.M., D.B.C.), Radiology (J.J.C., J.T., P.A., M.A.M.), Biostatistics (C.X.), and Developmental Biology (A.M.F., D.M.H.), Alzheimer's Disease Research Center (C.X., A.M.F., D.M.H., J.C.M., M.A.M.), and Hope Center for Neurological Disorders (B.M.A., A.M.F., D.M.H., J.C.M.), Washington University School of Medicine, St. Louis, MO.
Address correspondence and reprint requests to Dr. Beau M. Ances, Department of Neurology, Washington University in St. Louis, Box 8111, 660 South Euclid Ave., St. Louis, MO 63110
Objectives: Diagnostic challenges exist for differentiating HIV dementia from Alzheimer disease (AD) in older HIV-infected (HIV+) individuals. Similar abnormalities in brain amyloid-ß42 (Aß42) metabolism may be involved in HIV-associated neuropathology and AD. We evaluated the amyloid-binding agent 11C-Pittsburgh compound B (11C-PiB), a biomarker for Aß42 deposition, in cognitively unimpaired HIV+ (n = 10) participants and matched community controls without dementia (n = 20).
Methods: In this case-control study, all participants had an 11C-PiB scan within 2 years of concomitant CSF studies and neuropsychometric testing. Statistical differences between HIV+ and community controls for demographic and clinical values were assessed by {chi}2 tests. Participants were further divided into either low (<500 pg/mL) or normal (≥500 pg/mL) CSF Aß42 groups with Student t tests performed to determine if regional differences in fibrillar amyloid plaque deposition varied with CSF Aß42.
Results: Regardless of CSF Aß42 level, none of the HIV+ participants had fibrillar amyloid plaques as assessed by increased 11C-PiB mean cortical binding potential (MCBP) or binding potential within 4 cortical regions. In contrast, some community controls with low CSF Aß42 (<500 pg/mL) had high 11C-PiB MCBP with elevated binding potentials (>0.18 arbitrary units) within cortical regions.
Conclusions: Cognitively unimpaired HIV+ participants, even with low CSF Aß42 (<500 pg/mL), do not have 11C-PiB parameters suggesting brain fibrillar amyloid deposition. The dissimilarity between unimpaired HIV+ and preclinical AD may reflect differences in Aß42 production and/or formation of diffuse plaques. Future longitudinal studies of HIV+ participants with low CSF Aß42 and normal 11C-PiB are required.
Abbreviations: Aß42 = amyloid-ß42; AD = Alzheimer disease; ART = antiretroviral therapy; CDR = Clinical Dementia Rating; CHARTER = CNS Highly Activated Retroviral Therapy Effects Research; GDS = global deficit score; HAND = HIV-associated neurocognitive disorder; LP = lumbar puncture; MCBP = mean cortical binding potential; PiB = Pittsburgh compound B; ROI = region of interest; WUSTL = Washington University in St. Louis.
HIV-associated neuroinflammation can occur despite virologic control with antiretroviral therapy (ART).1 The prevalence of HIV-infected (HIV+) participants >50 years old has risen as life expectancy increases with ART. If current trends continue, more than 50% of all HIV+ individuals will be >50 years old by 2015.2 Age is a risk factor for HIV-associated neurocognitive disorder (HAND) and Alzheimer disease (AD). As HIV+ participants age, clinicians face the challenge of differentiating individuals at risk for HAND from those with AD.
Genetic, biochemical, and animal models and autopsy studies have demonstrated a critical role for brain amyloid-ß42 (Aß42) aggregation in AD.3 Similar neuropathologic abnormalities occur with HIV. Postmortem HIV+ subjects have increased brain Aß42 and tau deposition compared to age-matched community controls.4 Decreased CSF Aß42 is observed in subjects with AD and some unimpaired community controls with fibrillar Aß42.3 Subjects with HAND have CSF Aß42 levels similar to participants with mild AD.1,5
Reduced CSF Aß42 (<500 pg/mL) correlates with increased fibrillar amyloid deposition using the PET amyloid binding agent N-methyl-[11C]2-(4-methylaminophenyl)-6-hydroxybenzothiazole (11C-PiB) in subjects with AD and unimpaired community controls with preclinical AD.3 It remains unknown if a similar relationship exists for HIV. We investigated if low CSF Aß42 levels were predictive of increased 11C-PiB binding potentials in cognitively unimpaired HIV+ participants.
We observed that cognitively unimpaired HIV+ participants, even with low CSF Aß42 (<500 pg/mL), did not have increased 11C-PiB that might indicate fibrillar brain amyloid deposition. However, community controls with a low CSF Aß42 were more likely to have elevated 11C-PiB MCBP (>0.18 arbitrary units).3 Unimpaired community controls with increased 11C-PiB MCBP may have preclinical AD.8 Within a 2-year retrospective interval during which we followed the HIV+ participants, even those with low CSF Aß42 had no significant changes in cognition (GDS = 0.18 at LP and GDS = 0.31 at subsequent 11C-PiB). Our findings suggest that 11C-PiB MCBP differs in cognitively unimpaired HIV+ individuals compared to community controls with low CSF Aß42. In the setting of HIV, low CSF Aß42 may not reliably predict fibrillar Aß brain deposits as it does in preclinical AD.9 As the HIV+ population ages, this distinction could be diagnostically important. It remains necessary to understand whether fibrillar Aß seen with increased 11C-PiB is present in patients with HAND. This would assist in differentiating HAND from AD. While APOE status was not determined for participants, future studies investigating the impact of genetic risk factors on 11C-PiB MCBP and CSF Aß42 in HIV+ participants are required.2
Both 11C-PiB and CSF Aß42 levels are biomarkers of brain amyloid deposition in patients with AD and antecedent measures of impairment in community controls with preclinical AD.8 A strong inverse correlation exists between these biomarkers. The lack of correlation between CSF Aß42 and 11C-PiB MCBP in unimpaired HIV+ participants could result from decreased Aß42 production, increased intraneuronal Aß42 deposition leading to reduced extracellular concentrations, or more extracellular Aß42 amyloid but in a diffuse, nonfibrillar Aß form.4,9,10 In each instance, relatively normal 11C-PiB would occur. Future longitudinal examination, especially a larger sample of HIV+ participants with low CSF Aß42 and normal 11C-PiB, are required to understand whether observed low CSF Aß42 represents an aggregation of diffuse oligomeric forms (11C-PiB-negative) that eventually become substantial fibrillar (11C-PiB-positive) deposits,1,5 or simply the low normal end of CSF Aß42 in HIV+ participants.9 Our findings reinforce the importance of understanding amyloid metabolism in HIV-associated neuropathology, while confirming that low CSF Aß42 is not simply a manifestation of early fibrillar Aß deposition in the brain.
Demographic and clinical variables were similar (table). Neither group had significant cognitive impairment. HIV+ participants, even those with low CSF Aß42 (<500 pg/mL), did not have increased fibrillar amyloid plaques using 11C-PiB (figure 1A). In contrast, community controls with low CSF Aß42 had more fibrillar amyloid plaques (figure 1B). Several community controls had 11C-PiB measures indistinguishable from a typical AD pattern.7 These unimpaired community controls may have preclinical AD.8
We assessed the relationship between fibrillar amyloid deposition using 11C-PiB and CSF Aß42 for HIV+ participants and community controls. A 2 x 2 matrix was created using CSF (Aß42 <500 pg/mL) and 11C-PiB MCBP (<0.18 arbitrary units) (figure 2A). All HIV+ participants were located in the left upper and lower quadrants. Community controls fell within 3 boxes: left upper and lower quadrants and right lower quadrant. Half of the community controls had low CSF Aß42 and high 11C-PiB MCBP. A mismatch existed between CSF Aß42 and 11C-PiB MCPB for HIV+ participants and community controls (left lower quadrant).
Binding potentials were assessed within ROIs to determine degree of variation in fibrillar amyloid deposition. Binding potentials were elevated for community controls with low CSF Aß42 compared to other groups within all areas. HIV+ participants including those with low CSF AB42 (n = 4) had binding potentials similar to community controls with normal CSF Aß42 (figure 2B).
Participants. HIV+ participants (n = 10) (3959 years of age) with confirmed serologic status were selected from the CNS Highly Activated Retroviral Therapy Effects Research (CHARTER) cohort at Washington University in St. Louis (WUSTL). Four participants with low CSF Aß42 levels (<500 pg/mL) and 6 with normal CSF Aß42 levels (≥500 pg/mL) were contacted. We selected community controls (n = 20) (4463 years of age) of similar sex and education from memory and aging studies at the WUSTL Alzheimer's Disease Research Center (2 controls for every HIV+ subject). We received approval from the WUSTL ethical standards committee on human experimentation for experiments using human subjects. In this case-control study, written informed consent was obtained from all subjects participating in this study. The recommendations of the Strengthening the Reporting of Observational Studies in Epidemiology criteria were followed whenever applicable.6
All participants had an 11C-PiB scan within 2 years of concomitant lumbar puncture (LP) and neuropsychometric testing. For HIV+ participants, cognition was assessed at the time of scan and approximately 2 years prior to LP. Cognition was evaluated in HIV+ subjects using the previously validated global deficit score (GDS) with impairment deemed significant if GDS ≥0.5.5 For community controls, impairment was assessed by the Clinical Dementia Rating (CDR) scale with impairment noted if CDR >0.3
CSF evaluation. CSF collection used previously described methods.3 CSF Aß42 was analyzed using a commercial enzyme-linked immunosorbent assay (Innogenetics, Ghent, Belgium). Samples were kept on ice with assays performed on aliquots after a single thaw.
Imaging. Participants underwent 11C-PiB as previously described.7 Tracer was injected into the antecubital vein with a 60-minute 3-dimensional dynamic PET scan performed. Each subject had a T1-weighted anatomic scan with 11C-PiB images corrected for head motion and registered to this scan.3 The cerebellum was used as a reference as amyloid deposition has not been observed within this area in community controls.7 Logan graphical analyses were performed and 11C-PiB distribution volume calculated for the prefrontal, lateral temporal, precuneus, and gyrus rectus.11C-PiB binding potentials for each region of interest (ROI) and the mean cortical binding potential (MCBP) were calculated.7
Statistical analysis. Statistical differences between HIV+ and community controls for demographic and clinical values were assessed by {chi}2 tests. Participants were divided into either low (<500 pg/mL) or normal (≥500 pg/mL) CSF Aß42 groups using previously defined criterion with excellent sensitivity (100%) and good specificity (84%) for predicting subjects at risk for dementia.8 An analysis of variance with Bonferroni correction for multiple comparisons assessed if regional differences in fibrillar amyloid plaque deposition varied with CSF Aß42.
  iconpaperstack view older Articles   Back to Top