EXPERIENCE AND REASON |
HIV and AIDS Malignancy Branch
National Cancer Institute
National Institutes of Health
Bethesda, MD 20892
Medical Illness Counseling Center
Chevy Chase, MD 20815
HIV and AIDS Malignancy Branch
National Cancer Institute
National Institutes of Health
Bethesda, MD 20892
Childrens National Medical Center
Washington, DC 20010
| ABSTRACT |
|---|
|
|
|---|
A retrospective review of 107 patient records of children with HIV infection on HAART treatment protocols was conducted. Four patients were identified who responded to protease inhibitor-containing HAART therapy with sustained viral load suppression, and stable immunologic and medical parameters, yet demonstrated significant cognitive decline. Such discordance between biological and clinical markers previously has been reported in adults with HIV disease but not in children. This observed decline in neurocognitive functioning despite stable medical parameters suggests that HAART regimens that are effective for systemic disease may not be as effective for the central nervous system (CNS), perhaps because the antiretrovirals do not penetrate adequately into the CNS.
Of note, 3 of these 4 patients did not have zidovudine (ZDV) included in their HAART regimen. The only patient who was treated with ZDV containing regimen received 90 mg/m2 every 6 hours, which is at the lower end of the recommended ZDV pediatric full-dose range (90 mg/m2 to 120 mg/m2). Two of the 4 patients began ZDV at 120 mg/m2 every 6 hours following the decline in their cognitive test scores and subsequently showed improved or stable functioning as evidenced by the results of follow-up psychometric testing.
Long-term prospective studies using both systemic and CNS measures are necessary to further investigate the effects of HAART in children with HIV disease. Longitudinal cognitive assessments of children receiving HAART appear indicated to identify cognitive decline and to provide appropriate therapeutic intervention when manifestations of HIV-related CNS disease progression occur.
Key Words: HIV HAART
Abbreviations: HIV, human immunodeficiency virus HAART, highly active antiretroviral therapy CNS, central nervous system ZDV, zidovudine AIDS, acquired immunodeficiency syndrome d4T, stavudine NVP, nevirapine CSF, cerebrospinal fluid ddI, didanosine NFV, nelfinavir NCI, National Cancer Institute IQ, intelligence quotient CT, computed tomography SD, standard deviation GCI, General Cognitive Index WISC-III, Wechsler Intelligence Scale for Children, Third Edition FISQ, full-scale intelligence quotient VIQ, Verbal Intelligence Quotient PIQ, Performance Intelligence Quotient CDC, Centers for Disease Control and Prevention NRTI, nucleoside reverse transcriptase inhibitor
Children infected with HIV type-1 (HIV-1) are at high risk for developmental delay and deterioration in their cognitive functioning, primarily associated with the effects of HIV-1 infection of the brain.1 The most severe form of HIV-related CNS disease, HIV encephalopathy, has been a common acquired immunodeficiency syndrome (AIDS)-defining condition in the pediatric population.2,3 Psychometric testing has documented global delays in cognitive ability as well as specific deficits in selective domains, such as psychomotor speed and expressive language.47 Low scores on measures of cognitive ability are correlated with markers of advanced HIV disease, such as low CD4 lymphocyte percentages and high plasma viral RNA concentrations.3,8 In a large cohort of infants, children, and adolescents, low neuropsychological test scores and presence of motor dysfunction were more predictive of HIV disease progression than these medical surrogate markers.9
The use of combination antiretroviral therapy, including a protease inhibitor or non-nucleoside reverse transcriptase inhibitor (HAART), can reduce viral loads to low levels, often below the limits of detection, and can lead to significant increases or normalization of the CD4+ cell count.10 In addition, treatment with antiretroviral therapy, particularly including ZDV, can improve neurobehavioral deficits in children.1113 The widespread use of HAART has likely led to a decline in the number of HIV-infected children with severe CNS manifestations.4,14 Most school-aged children being treated with combination therapy generally function within normal limits, do well in school, and may only exhibit subtle abnormalities in selective areas.15,16
The CNS appears to function as an independent compartment with regard to viral replication and antiretroviral drug pharmacokinetics.13 Although some drugs, such as ZDV, stavudine (d4T), abacavir, efavirenz, nevirapine (NVP), and indinavir, penetrate into the CNS,17 their concentrations in cerebrospinal fluid (CSF) may still be well below the concentrations typically achieved in plasma.18 Other drugs, such as didanosine (ddI)19 and nelfinavir (NFV),20 penetrate poorly into the CSF. This lack of penetration into the CNS by some antiretrovirals may mean that certain drug combinations that effectively inhibit viral replication in the periphery do not inhibit viral replication as effectively in the CNS, making it a sanctuary for HIV-infected cells. Thus, some drugs that are beneficial for treating systemic HIV disease are not as effective against HIV-related CNS dysfunction12,13 and may leave the developing brain of children relatively unprotected against the effects of HIV.
Studies of adults have suggested that there can be a discordance between the apparent control of HIV disease in the periphery versus the CNS. Adults in the Multicenter AIDS Cohort Study group showed a decreased incidence of AIDS dementia complex from 21.1/1000 person-years in 19901992 to 14.7/1000 person-years in 19961997.21 The Australian adult AIDS surveillance data from 19921997 showed that the proportion of AIDS dementia complex relative to other AIDS defining illness increased from 4.4% to 6.5%, even as the median CD4+ cell count improved.22 These findings suggests that HAART provided systemic benefits but might provide less protection against AIDS dementia complex than other AIDS-defining illnesses. In the pediatric literature, no comparable research has described a potential discordance between CNS and systemic disease among patients treated with HAART regimens. However, such discordance is possible because children generally have higher viral loads and a greater incidence of HIV CNS disease.23,24 Thus, the efficacy of current HAART regimens to treat both systemic and CNS disease in children needs to be investigated.
In our pediatric HIV clinical program at the HIV and AIDS Malignancy Branch at the National Cancer Institute (NCI), we initially observed 2 patients treated with HAART who demonstrated significant cognitive decline, as assessed by serial neuropsychologic testing, while maintaining undetectable or stable, low viral loads. These observations prompted us to conduct a retrospective review of all the patients in our program treated with a HAART regimen to determine if other of our patients showed a discordance between well-controlled peripheral HIV disease and worsening CNS disease. Our review revealed an additional 2 patients. This current report presents these 4 cases with discordance between indicators of their CNS and peripheral disease.
All 4 patients had vertically acquired HIV infection and presented with significant changes in their neuropsychological functioning (ie, a significant decline in their intelligence quotient [IQ] scores), but had no corresponding changes in their immunologic status or plasma HIV RNA concentrations. Some of these patients showed neuropsychological improvement following modification of their antiretroviral regimen. These cases suggest that practitioners caring for children with HIV should carefully monitor the neuropsychological status of their patients, even those with apparently well-controlled HIV disease.
| METHODS |
|---|
|
|
|---|
0.5 log10 from baseline with absolute plasma viral load <2000 viral RNA copies/ml at time of cognitive change.
20 points on either the Verbal (VIQ) or Performance IQ (PIQ). The drop in IQ must have occurred within 1 year and could not be attributed to other factors (eg, change in tests, behavioral/emotional factors, lack of appropriate schooling, acute illness, medication effects, other CNS diseases, etc). This decrease in cognitive function corresponded to the criteria established for a classification of CNS dysfunction established by the Neuropsychology Group at the HIV and AIDS Maligancy Branch, NCI.27
CD4+ Lymphocyte Counts and Viral Loads
Enumeration of CD4+ lymphocytes was determined by standard 2-color or 3-color flow cytometry techniques using an Epics XL flow cytometer (Beckman Coulter, Hialeah, FL). Lymphocyte staining was performed using a modification of the Centers for Disease Control and Preventions (CDC) whole blood method.28 The procedure used to determine viral loads was the Roche Diagnostics Amplicor HIV-1 Monitor Test. The ultrasensitive and standard methods were used along with internal controls provided with each kit.
Neuroimaging
A pediatric neurologist reviewed each patients CT scan of the brain for the presence and severity of cortical atrophy, white matter abnormalities, intracerebral calcifications, and other lesions.
Cognitive Testing
Baseline neurocognitive scores were obtained before protocol enrollment. A psychologist administered an age-appropriate test of general cognitive functioning, which included the McCarthy Scales of Childrens Ability25 for children between the ages 2.4 through 8.6 years, or the WISC-III26 for the children from 6 through 16.9 years. Two patients were tested with the McCarthy and 2 were evaluated using the WISC-III. Each measure provided standard scores as follows:
Behavioral Assessment
At each visit, the psychologist conducted a clinical interview with the parent/caregiver regarding the childs daily activities to account for any possible behavioral changes seen in the home or at school since the childs last neuropsychological evaluation. Parents/caregivers also were asked to rate their childs behavior in the past month using a standardized behavior rating scale.29
| CASE REPORTS |
|---|
|
|
|---|
|
Patient B
Patient B was a 6.2-year-old black male, CDC class A1. Past treatment consisted of several NRTI regimens for 65 months. He was treated with ritonavir (400 mg/m2 every 12 hours), d4T (1 mg/kg every 12 hours), and ddI (135 mg/m2 every 2 hours) 2 months before enrollment in a phase I therapeutic vaccine study of glycoprotein 120-depleted killed whole virus. Patient B exhibited a significant 23-point increase in his FSIQ, including 19 points on his VIQ and 24 points on his PIQ at his 6-month follow-up evaluation. The degree of improvement in his IQ scores was beyond what one could expect as a result of test familiarity or practice effects.26,30 At that time, it was difficult to distinguish whether the improved IQ test scores were attributable to the antiretroviral therapy or to environmental factors. Approximately 2 months before starting the protocol, Patient B was removed from an allegedly neglectful foster home and placed in a new foster home that was reportedly more stable and nurturing. The positive influence of his placement in a more nurturing home environment and school setting possibly may have contributed to his improved test scores. However, at his 18-month follow-up, Patient Bs FSIQ score declined significantly by 16 points, with a 22-point drop in his VIQ. Although the initial improvement in test scores may have been associated with an affirmative change in his environment, some transient CNS benefit from his antiretrovirals may have also played a role in the increase and may partially explain the subsequent decline in his test performance. Close monitoring on the same antiretroviral regimen and 2 subsequent neuropsychological assessments have revealed no further decline in cognitive functioning.
Patient C
Patient C was a 2.5-year-old white male, CDC Class A2, with no prior HIV treatment. He enrolled in a phase I protocol in which he initially was treated with ritonavir (400 mg/m2 every 12 hours), and 2 months later ZDV (90 mg/m2 every 6 hours) and ddI (90 mg/m2 every 12 hours) were added to his regimen. He exhibited a drop in hematocrit and hemoglobin levels after 4 months of study, presumed to be attributable to an acute viral illness, although ZDV-induced bone marrow suppression could not be ruled out. He was treated with ferrous sulfate and his anemia resolved within a month. Because of concerns regarding ZDV-related bone marrow suppression, all his antiretroviral drugs were withheld for 2 weeks. When his regimen was restarted, ZDV was reintroduced at a lower dose (60 mg/m2 every 6 hours) for 1 month. After his anemia resolved, his ZDV dose was increased back to the initially prescribed level. At his 12-month follow-up visit, Patient Cs GCI was 24 points lower than his baseline score, which was considered to be a significant decline. All other scaled scores on the McCarthy (Verbal, Perceptual-Performance, Quantitative, Memory, and Motor) also were substantially lower. The patient reportedly continued to be adherent to the prescribed antiretroviral therapy and interviews did not reveal any significant changes in the home or school environment. After the decrease in his cognitive test scores, Patient Cs ZDV was further increased to the upper end of the full-dose range (120 mg/m2 every 6 hours). Two follow-up cognitive evaluations administered 6 months apart showed no further decline in functioning. The patients IQ scores remained in the Average range although he had scored in the High Average range before the decline. The family decided to seek medical care in their locality and there has been no further follow-up.
Patient D
Patient D was a 4-year-old black male with vertically acquired HIV infection, CDC Class B2. Prior antiretroviral treatment included combination NRTIs for 8 months. He started a regimen of NFV (26 mg/kg every 8 hours) and d4T (4 mg/kg every 12 hours) in conjunction with enrollment on the phase I therapeutic vaccine protocol mentioned earlier. At his 6-month follow-up testing, Patient Ds GCI fell significantly by 17 points. He also was observed to have a further decline in his speech and expressive language functions, despite speech therapy provided in the interim. The decline in language function was a particular concern because expressive language has been shown to be more susceptible to the effects of the HIV than receptive language.6,31 The parent did not report any factors that could explain these changes in functioning. A neurologic examination 3 months after the change in cognitive scores was significant for possible apraxia (left upper extremity) and the brain CT scan was remarkable for mild cortical atrophy. Apparently, cognitive decline preceded other parameters of HIV disease progression. Although it was recommended that the patient change his antiretroviral therapy to include ZDV, lamivudine, and NFV, the patient was lost to follow-up and additional outcome data are not available.
| DISCUSSION |
|---|
|
|
|---|
This observed decline in neurocognitive functioning despite stable systemic disease suggests that viral replication in the CNS may not be well-controlled in some patients, even when treated with an effective HAART regimen, perhaps because the antiretrovirals do not penetrate adequately into the CNS.18,33 ZDV enters CSF better than other antiretrovirals,18 and may improve neurocognitive dysfunction, particularly when used in combination therapy.3437 In this case series, 3 of the 4 patients were not treated with a ZDV-containing regimen during the period preceding their cognitive changes. Although patient C was treated with ZDV at 90 mg/m2 every 6 hours, this dose is at the lower end of the full-dose range for children (90 mg/m2 to 120 mg/m2 every 6 hours) and it might have been less than adequate for him. Even as follow-up information was not available for 1 of the 4 patients, the 2 children (A and C) who were changed to a regimen that included ZDV (120 mg/m2 every 6 hours) demonstrated improved or stable cognitive test scores during subsequent evaluations.
Almost all (103 of 107; 97%) of the children reviewed for this study demonstrated at least stable cognitive function during treatment with HAART regimens, some of which included ZDV (78%), and some that did not. An analysis of the total group of patients (N = 107) found a significant increase (F = 143.85; P < .001) in FSIQ (91.4 ± 1.6 to 94.4 ± 1.7) from baseline to after 6 months of HAART. Thus, despite limited CNS penetration of some antiretroviral drugs, HAART may have a protective effect on the brain by reducing the viral load burden in the periphery, thus minimizing the risk of infected cells going into the CNS. Factors unrelated to antiretroviral drugs, such as host genetic factors,38 also may influence the extent of HIV CNS disease. In addition, other medical, psychosocial, and environmental factors possibly could have contributed to the decline in their cognitive scores, although such factors were examined.
This case study shows that some patients may exhibit significant declines in cognitive test scores even when their HIV disease is apparently well-controlled as assessed by CD4+ T cell subsets, plasma HIV-1 load, and CT brain scans. Altering a patients antiretroviral regimen to include a CNS penetrating drug, such as ZDV, may partially reverse CNS HIV disease. Thus, practitioners caring for children with HIV disease should carefully monitor their cognitive status through repeated psychological assessments.
| ACKNOWLEDGMENTS |
|---|
We thank Dr Lauren Wood, Principal Investigator of the Immunogen Protocol; Dr Rohan Hazra for his valuable comments during the initial phase of this project; and Dr Robert Yarchoan, Chief of the HIV and Malignancy Branch, NCI. We also wish to thank Drs Staci Martin and Patricia Klaas for their support; Judy Zuckerman, RN, for her help in the selection of the 4 patients discussed; and the patients and families who participated in this study.
| FOOTNOTES |
|---|
Reprint requests to (M.A.T.-T.) National Institutes of Health, Building 10, Rm 10S255, 10 Center Dr, Bethesda, MD 20892-1868. E-mail: tamulam{at}mail.nih.gov
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. C. Lindsey, K. M. Malee, P. Brouwers, M. D. Hughes, and for the PACTG 219C Study Team Neurodevelopmental Functioning in HIV-Infected Infants and Young Children Before and After the Introduction of Protease Inhibitor-Based Highly Active Antiretroviral Therapy Pediatrics, March 1, 2007; 119(3): e681 - e693. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Crowe, D. De Matteo, M. Irwin, G. Kent, and V. McClain Coercive Treatment of HIV-Positive Children Is Not Justified Pediatrics, December 1, 2005; 116(6): 1605 - 1606. [Full Text] [PDF] |
||||
![]() |
M. C. Shanbhag, R. M. Rutstein, T. Zaoutis, H. Zhao, D. Chao, and J. Radcliffe Neurocognitive Functioning in Pediatric Human Immunodeficiency Virus Infection: Effects of Combined Therapy Arch Pediatr Adolesc Med, July 1, 2005; 159(7): 651 - 656. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||