Interest in the decreased risk that a diagnosis of Kaposi's sarcoma (KS) confers on AIDS dementia complex (ADC) was rekindled following the demonstration that a human herpes virus 8 (HHV8)-encoded chemokine, viral macrophage inflammatory protein (vMIP)-II, blocks the entry of HIV-1 into CD4 cell lines, such as microglia, that express CCR3 [1,2]. Boshoff et al. hypothesized that patients with KS, or high HHV8 load, may be less likely to develop ADC . The majority of epidemiological studies have subsequently demonstrated a protective association between KS and ADC [3–8]. No association was seen in one case–control study; however, the authors concluded that in order to make reliable statements concerning the protective effects that complete neurological, neuropathological and clinical data are required .
We have demonstrated previously that HHV8 antibodies are a strong risk factor for KS in the Amsterdam Cohort Studies on HIV-1 infection [10,11] and note that the influence of HHV8 antibodies on ADC has been examined in two studies; no association was seen in one study whereas a higher geometric mean titer of HHV8 antibodies was seen in HIV patients without ADC than in those with ADC [7,8].In our prospective study we focus on a single HIV transmission group with a uniform risk for KS to investigate whether a diagnosis of KS or the presence of HHV8 antibodies is associated with a low risk for ADC and find no evidence for a protective effect of HHV8 infection or KS against ADC among homosexual men.
Study population and study period
The study population consisted of 599 HIV-1 infected homosexual men who participated in the Amsterdam Cohort Studies on HIV infection and AIDS between December 1984 and December 1996.
Serologic assays and study design
The enzyme immunoassay system and Western blot assays for the detection of HHV8 and HIV antibodies respectively and the study design, have been described in a previous publication .
The diagnoses of definite ADC were made according to the 1987 Centers for Disease Control and Prevention surveillance case definition . Those patients who met the clinical criteria for definite ADC but lacked either cerebrospinal fluid, computed tomography or autopsy findings to exclude other illnesses, were termed provisional ADC. Central nervous system diseases including toxoplasmosis, progressive multifocal leukoencephalopathy, and cerebral lymphoma were excluded by computed tomography or magnetic resonance imaging; metabolic and toxic encephalopathies were excluded by extensive blood or urine tests.
All participants were homosexual men who were either HIV seropositive at enrolment in the cohort or who became HIV seropositive during follow-up and the date of HIV seroconversion was taken as the start of observation time. Prevalences of KS, HHV8 antibodies and ADC were calculated for the 599 participants at the end of the study period. For individuals who tested positive for HHV8, the date of seroconversion was determined by testing earlier samples. The risk of ADC as predicted by a diagnosis of KS or the presence of HHV8 antibodies was estimated using a time-dependent Cox proportional hazards model. In this model, observation time started at either the moment of HIV-1 seroconversion or the date of study entry for HIV-1 seropositive participants; the date of HHV8 seropositivity or KS diagnosis were time-dependent factors. In addition, to adjust for deterioration of the immune system, the moment that the CD4 cell count dropped below 100 × 106 or 50 × 106 cells/l was included in the time-dependent model. Finally, we entered a time-dependent variable for the use of antiretroviral medication to adjust for the protective effect of antiretroviral medication on the occurrence of ADC. Data were censored at 31 December 1996.
Cross sectional analysis
Of the 599 participants 290 (48.4%) had HHV8 antibodies and 99 (16.5%) had KS. Thirty (5.0%) of the 599 participants met the criteria for definite or provisional ADC, of whom 19 (3.2%) were diagnosed with definite ADC. ADC was diagnosed after KS in five (5.2%) of 97 participants, in two participants before KS, and in 25 (5.0%) of 502 participants who were never diagnosed with KS [odds ratio (OR), 1.5; 95% confidence interval (CI), 0.5–3.7]. ADC was diagnosed in 14 (4.8%) of 290 HHV8 positive individuals and 16 (5.2%) of 309 HHV8 negative individuals (OR, 0.93, 95% CI; 0.41–2.07). All ADC diagnoses were after HHV8 seroconversion.
KS was diagnosed in five (16.7%) of 30 participants with ADC and in 92 (16.2%) of 569 participants without ADC (OR, 1.0; 95% CI, 0.4–2.9). Antibodies against HHV8 were detected in 14 (46.7%) of 30 participants with ADC and 276 (48.5%) of 569 participants without ADC (OR, 0.9; 95% CI, 0.4–2.0). For comparison, the OR between KS and HHV8 seropositivity was 8.9 (95% CI, 4.3–18.3) .
The hazards ratio for participants with KS to develop ADC was 2.7 (95% CI, 0.92–7.96;p = 0.07) and only definite ADC was 3.5 (95% CI, 1.00–12.26;P = 0.05). These risks do not change significantly when adjusting for low CD4 cell counts. When adjusting the model for the use of antiretroviral medication, the risks for KS and ADC were 2.0 (95% CI, 0.66–5.77;P = 0.23) and 2.6 (95% CI, 0.73–9.12;P = 0.14) for definite ADC; the risk for ADC decreased when antiretroviral therapy was used: 0.19 (95% CI, 0.07–0.57;P = 0.003) and 0.26 (95% CI, 0.07–0.90;P = 0.03) for definite ADC. Inclusion of the variables on KS, CD4 cell count and use of antiretroviral medication in the same model does not significantly alter the results reported above (data not shown).
The risk for participants with HHV8 to develop ADC was 0.84 (95% CI, 0.41–1.75;P = 0.65) and 0.69 (95% CI, 0.28–1.73;P = 0.43) for definite ADC. Again, adjustment for low CD4 cell counts shows no changes. When adjusting this model for the use of antiretroviral medication, the risk for HHV8 positive patients to develop ADC was 0.85 (95% CI, 0.41–1.77;P = 0.66) and 0.69 (95% CI, 0.27–1.73;P = 0.42) for definite ADC. The hazards ratio of antiretroviral medication use for developing ADC was 0.18 (95% CI, 0.06–0.53;P = 0.002) and 0.23 (95% CI, 0.07–0.81;P = 0.02) for definite ADC. Inclusion of all variables on HHV8 serostatus, low CD4 cell count and use of antiretroviral medication does not significantly alter the results reported above (data not shown).
The hazards ratios for KS and HHV8 are presented separately above as not all KS patients are HHV8 seropositive in this study and not all HHV8 positive individuals developed KS. The risk of having either KS or HHV8 and developing ADC or definite ADC, having adjusted for low CD4 cell count, use of antiretroviral or anti-herpesviral medication, and age was similar to the results reported above (data not shown).
Several groups have shown that there is a negative or protective association between KS and ADC in HIV-1 infected individuals. In contrast we did not find a significant difference in occurrence of ADC in HIV-1 infected patients with or without KS. In contrast, our analyses show that KS, although not statistically significant, is associated with an increased risk for ADC in a cohort study of HIV-1 infected homosexual men and we interpret KS to be a marker of disease progression (Table 1). Once antiretroviral medication is initiated, the risk of ADC dropped dramatically to 20% of participants without antiretroviral medication; this finding has been reported previously by Portegies et al.. When all variables on KS, CD4 cell count and medication were analysed, KS still provided no protection against ADC and, although not statistically significant, patients with KS appear to have twice the risk for ADC as compared to those without KS.
A similar analysis based on HHV8 seropositivity showed no significant difference in HHV8 seroprevalence between those with and without ADC and also yielded no significant protective effect for HHV8 antibodies against ADC. An analysis using KS and HHV8 antibodies also demonstrated a lack of protective effect. Despite disagreement on the negative association between KS and ADC, we agree with Rezza et al. that there is no association between HHV8 antibodies and ADC and consider that the higher geometric mean titer in the study by Dupin et al. reflects a decline in CD4 cell count and is not necessarily involved in ADC pathogenesis .The previously reported OR of 8.9 (95% CI, 4.3–18.3) between HHV8 antibodies and KS in this same group helps to validate our analysis .
We consider that the strengths of this study lie in investigating the association between KS and ADC by adjusting for immune deterioration by means of low CD4 cell count and the use of antiretroviral medication and also in focusing on one HIV transmission group with a uniform risk for KS. There are several other differences between published studies in this field: geographic location; study populations; prevalences of KS, HHV8 and ADC in the study group; study design; comparison of ADC in HIV transmission groups which have differing risks for KS; use of different clinical or pathological criteria for defining ADC; and approach to adjusting for the confounding effects of changes in CD4 cell count, HIV-1 RNA load and administration of medication. The association between KS and ADC in AIDS patients may be explained by combining people from different HIV-1 transmission groups with dissimilar disease risks; HIV-1 infected homosexual men are more likely to have KS than any other HIV-1 transmission group, such as drug users, whereas injecting drug use is contentiously associated with ADC [1,7,14]. Under these circumstances KS would appear to protect against ADC in a group of AIDS patients composed of homosexual men and drug users.
In individuals with untreated HIV-1 infection, brain tissue is commonly infected with HIV-1 whereas KS is a rare neurological event [15,16]. Studies to detect HHV8 by PCR in cerebrospinal fluid or brain tissue have yielded conflicting results; HHV8 DNA was detected in none of the cerebrospinal fluid samples from nine patients with ADC whereas another study reported that 19 (63.3%) of 30 healthy brains contained detectable HHV8 [17,18]. This would suggest that only small amounts of HHV8, if any at all, are present in brain tissue and that vMIP-II is unlikely to be produced in adequate amounts to block HIV-1 entry into microglia at the CCR3 receptor. It is possible to counter-argue that vMIP-II is produced at or before the blood–brain barrier and we are not able to comment on differences in HHV8 viral load, KS tumor burden or vMIP-II in patients with and without ADC, as suggested by Brew . We, however, have found no evidence that KS or HHV8 protects against ADC in a prospective cohort study of HIV-1 infected homosexual men and therefore cannot confirm the epidemiological association upon which the search for novel KS-associated or HHV8-derived therapies for ADC is based.
1. Chiesi A, Vella S, Dally LG. et al
. Epidemiology of AIDS dementia complex in Europe.AIDS in Europe Study Group.
J Acquir Immune Defic Syndr Hum Retrovirol 1996, 11: 39–44.
2. Boshoff C, Endo Y, Collins PD. et al
. Angiogenic and HIV-inhibitory functions of KSHV
-encoded chemokines. Science 1997, 278: 290–294.
3. Liestoel K, Goplen AK, Dunlop O, Bruun JN, Maehlen J. Kaposi's sarcoma
and protection from HIV dementia. Science 1998, 280: 361–362.
4. Baldeweg T, Catalan J, Gazzard BG, Weiss RA, Boshoff C. Kaposi's sarcoma
and protection from HIV dementia. Science 1998, 280: 362.362.
5. Dore GJ, Grulich A, Law MG, Brew BJ, Kaldor JM. Kaposi's sarcoma
and protection from AIDS dementia complex
. AIDS 1998, 12: 2354–2355.
6. Mary-Krause M, Billaud B, and the Clinical Epidemiology
Group from Centres d'Information et de Soins de l'Immunodeficience Humanine. Impact of Kaposi's sarcoma
in occurrence of opportunistic central nervous system disease. AIDS 1999, 13: 1985–1987.
7. Rezza G, Dorrucci M, Andreoni M. et al
. Does HHV-8 have a protective role on the development of HIV encephalopathy? Neurology 1999, 53: 2032–2036.
8. Dupin N, Marcelin AG, Caumes E. et al
. Relationship between Kaposi's sarcoma
, Kaposi's sarcoma
-associated herpesvirus and AIDS dementia complex
. AIDS 2000, 14: 333–335.
9. Neuenburg JK, Brodt HR, Herndier BG, Schlote W. Is there really a correlation between AIDS dementia and Kaposi's sarcoma
. AIDS 2000, 14: 94–95.
10. Renwick N, Halaby T, Weverling GJ. et al
. Seroconversion for human herpesvirus 8 during HIV infection is highly predictive of Kaposi's sarcoma
. AIDS 1998, 12: 2481–2488.
11. Goudsmit J, Renwick N, Dukers NH. et al
. Human herpesvirus 8 infections in the Amsterdam Cohort Studies (1984-1997): analysis of seroconversions to ORF65 and ORF73. Proc Natl Acad Sci USA 2000, 97: 4838–4843.
12. Anonymous. Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome.Council of State and Territorial Epidemiologists; AIDS Program, Center for Infectious Diseases.
MMWR 1987, 36 (suppl 1): 1S–15S.
13. Portegies P, de Gans J, Lange JM. et al
. Declining incidence of AIDS dementia complex
after introduction of zidovudine treatment. BMJ 1989, 299: 819–821.
14. Beral V, Peterman TA, Berkelman RL, Jaffe HW. Kaposi's sarcoma
among persons with AIDS: a sexually transmitted infection? The Lancet 1990, 335: 123–128.
15. Levy RM, Bredesen DE, Rosenblum ML. Neurological manifestations of the acquired immunodeficiency syndrome (AIDS): experience at UCSF and review of the literature. J Neurosurg 1985, 62: 476–495.
16. Nuovo G, Gallery F, MacConnell P, Brau A. In situ
detection of polymerase chain reaction-amplified HIV-1 nucleic acids and tumor necrosis factor-alpha RNA in the central nervous system. Am J Pathol 1994, 144: 659–666.
17. Brink NS, Sharvell Y, Howard MR, Fox JD, Harrison MJ, Miller RF. Detection of Epstein–Barr virus and Kaposi's sarcoma
-associated herpesvirus DNA in CSF from persons infected with HIV who had neurological disease. J Neurol Neurosurg Psychiatry 1998, 65: 191–195.
18. Chan PK, Ng HK, Cheung JL, Cheng AF. Survey for the presence and distribution of human herpesvirus 8 in healthy brain. J Clin Microbiol 2000, 38: 2772–2773.
19. Brew BJ. Does HHV-8 have a neuroprotective role on the development of HIV encephalopathy? Neurology 2000, 55: 459–460.