The prevalence of Kaposi sarcoma herpes virus (KSHV) in sub-Saharan Africa is among the highest in the world,1–3 and the region also bears the greatest burden of disease due to HIV.4 Infection with KSHV has been shown to lead to the development of Kaposi sarcoma (KS)5–7 and to multicentric Castleman disease8 and primary effusion lymphomas.9 KS, which is associated with significant morbidity and mortality, has now become one of the most common cancers in parts of sub-Saharan Africa and is the most common tumor in HIV-infected individuals.10 Coinfection with other viruses including cytomegalovirus, hepatitis B, and hepatitis C has been previously associated with HIV disease progression and mortality11,12 and poor CD4 cell count responses after initiation of antiretroviral therapy (ART).12,13
KSHV typically establishes a persistent latent infection in its host during which time only latent genes [of which open reading frame (Orf) 73 is 1 example] are expressed. In the presence of HIV-1 coinfection, however, immune suppression and cytokine release promote the reactivation of KSHV lytic genes, which include K8.114 and active replication and increase in KSHV viral progeny occurs. Previous in vitro studies have suggested interactions between these 2 viruses including an increase in HIV-1 viral load in the presence of KSHV15 and induced reactivation of HIV-1 replication in chronically infected cells.16 Despite this, there are few analyses describing the effect of coinfection with KSHV on HIV treatment outcomes after the initiation of ART.
We examined the effects of KSHV seropositivity on immunologic and virologic outcomes in the first year of ART among a cohort of HIV-infected adults attending a large, urban HIV care and treatment program in Johannesburg, South Africa.
Study Design and Site
This prospective cohort study was conducted at the Themba Lethu Clinic (TLC) in Johannesburg, South Africa. Currently, TLC is one of the largest treatment facilities in South Africa, with >30,000 HIV-infected adults ever enrolled in its comprehensive HIV care, management, and treatment program.17 Since the National rollout of ART in 2004, >23,000 individuals have been initiated on ART at the clinic according to the guidelines from the South African National Department of Health.18,19 Patient data at TLC are captured and stored on an electronic patient record. Patient laboratory blood tests are taken at ART initiation and monitoring laboratory tests (viral load, CD4 count, full blood count, and liver and kidney function tests) are conducted at 6 months, then yearly thereafter. Up to 3 attempts are made by clinic counselors to contact patients who do not return for scheduled clinic appointments. Information on deaths is recorded through passive surveillance and through linkage with the National Vital Registration System,20 which was last conducted in September 2011. The program participates in the International epidemiological Databases to Evaluate AIDS in Southern Africa.21
HIV-positive treatment naive patients, >18 years of age, who were assessed as ready and eligible for the initiation of ART at Themba Lethu clinic were invited to participate in the study. Participant enrollment was conducted between November 2008 and March 2009. The population was recruited from the TLC ART initiation groups. All eligible subjects attending a group counseling session before the initiation of ART were approached and invited to participate in the study on consecutive Wednesdays at TLC until the study sample size was enrolled. Patients with a history of ART use or who were unwilling to consent were excluded from the study. All the enrolled subjects provided informed consent before commencing the study procedures.
Laboratory testing for KSHV serology was conducted by the Contract Laboratory Services and National Health Laboratory Service, whereas KSHV viral load testing was performed by the Hematology and Molecular Medicine Department of the University of the Witwatersrand. Enzyme-linked immunosorbent assays that detect antibodies to lytic K8.1 and latent Orf73 KSHV recombinant protein antigens were developed in the Viral Oncology Section, ACVP, FNLCR, USA, and transferred to the Contract Laboratory Services. The enzyme-linked immunosorbent assays have good sensitivity and specificity and have been used in >30 studies internationally.22 All samples that tested serologically positive to KSHV were then tested for KSHV viral load using quantitative TaqMan polymerase chain reaction23 performed on the ABI Prism 7900 sequence detection system (Applied Biosystems, Forster City, CA). Subject and control samples were run in triplicate. The KSHV viral load assay has a linear dynamic range of 8 logs and is calibrated to detect a single copy of viral DNA in 150 ng of genomic DNA.
KSHV status at ART initiation was the exposure variable in this analysis. KSHV status was determined by venous blood samples drawn from all the study participants before the initiation of ART. Additional demographic and clinical data were extracted from the electronic patient record. Seropositivity to KSHV at the time of ART initiation was defined as a positive reaction to either lytic KSHV K8.1 or latent Orf73 antigen. The KSHV+ group was then further stratified by the presence or absence of detectable KSHV DNA. We further stratified a positive KSHV result into 3 categories: (1) positive to lytic k8.1 alone, (2) positive to latent Orf73 alone, or (3) positive to both.
We compared immunologic and virologic outcomes after ART initiation by KSHV status at treatment initiation. Outcome variables were (1) linear increase in the mean CD4 cell count after ART initiation and (2) virologic response (suppression of HIV viral load to ≤400 copies/mL) after 6 and 12 months on ART.
Demographic and clinical features of the study participants at ART initiation were stratified by KSHV status and summarized as simple proportions or medians with interquartile ranges (IQRs). The association of KSHV with a linear increase in CD4 counts from baseline to 6, 12, and 18 months was estimated using mixed linear models. We also estimated estimate CD4 trajectories by modeling the CD4 count over time using a linear regression model with CD4 cell counts after ART initiation as the outcome variable and time estimated in the models as a quadratic function using random slopes and a random intercept with an unstructured correlation matrix for repeated measures. CD4 count trajectory models for those KSHV+ and KSHV− were fit separately to allow for different curves by exposure group. Log-binomial regression was used to estimate the relative risk (RR) of KSHV on viral load suppression (<400 vs. ≥400) by 6 and 12 months on treatment, respectively. Age, gender, and baseline CD4 count were considered a priori confounders and were adjusted for in all models. Other covariates including CD3 count, CD8 count, hemoglobin level, tuberculosis treatment status, body mass index (BMI), and initiating treatment regimen were investigated for potential confounding using change-in-estimate criterion. A covariate was considered to be a confounder if the RR varied by ≥10% when the covariate was added to (or removed from) the model. The combined effect of potential confounders was assessed in the same way and was identified by the change in estimate of the RRs after adjustment for all the potential confounding factors.
Approval to conduct this study and to use data from the TLC site was granted by the Human Research Ethics Committee of the University of the Witwatersrand.
The original cohort enrolled 404 consenting adults presenting for the initiation of ART at TLC between November 2008 and March 2009. The baseline characteristics of the original cohort have been described elsewhere.24 This analysis is restricted to the 385 (95%) participants who initiated ART. The presenting features of this group at ART initiation are summarized in Table 1. The median age of the group was 38 years (IQR 32–45 years), the majority (n = 250, 65%) were women, and none had evidence of KS. The median CD4 count at ART initiation was 87 (40–149 cells per cubic millimeter) and over a third (37%) presented with a World Health Organization (WHO) stage III/IV defining condition. The majority of participants were started on standard first-line ART regimens; 86% on stavudine (d4T), lamivudine (3TC), and efavirenz and 7% on d4T, 3TC, and nevirapine. The remaining 7% were initiated on tenofovir-, zidovudine-, or lopinavir-based regimens. Alternative first-line regimens are used in situations where a preexisting condition such as peripheral neuropathy, hepatic pathology, or a planned pregnancy at ART initiation precludes the use of 1 of the standard regimens.19
The participants contributed a total of 5599.5 person months of follow-up and the mean follow-up time between the groups was similar: 13.9 months [95% confidence interval (CI): 12.9 to 14.8] for the KSHV− group compared with 15.2 months (95% CI: 14.4 to 16.1) for the KSHV+ group. Outcomes at the end of 18 months of follow-up among the KSHV+ group were similar in terms of death (7% vs. 9%), loss to follow-up (7% vs. 10%) and transfer to care at another facility (6% vs. 8%) when compared with that of the KSHV− group (Table 1).
Prevalence of KSHV
Among the study participants, 184/385 tested positive to KSHV with an overall prevalence of KSHV estimated at 48% (95% CI: 43% to 53%). Of these, 73 (39%; 95% CI: 33% to 46%) were reactive to lytic K8.1 alone, 34 (18%; 95% CI: 13% to 24%) to latent Orf73, and 77 (42%; 95% CI: 36% to 50%) to both. The groups were similar in terms of age, gender distribution, and HIV disease stage. The KSHV+ group presented with a somewhat higher median CD4 cell count (90 vs.78 cells per cubic millimeter) than their KSHV− counterparts. Among those participants with KSHV+ results, a convenience sample of 167 samples was also tested for KSHV viral load. Nineteen (11%) of these had a detectable viral load at the initiation of ART. Those reactive to both K8.1 and Orf73 were more likely to have a detectable KSHV viral load (n = 15, 20%) when compared with those reactive to lytic K8.1 (n = 2, 3%) or latent Orf73 (n = 2, 8%) alone.
CD4 Count Response
Both groups demonstrated good immune responses to treatment. The mean increase in CD4 count among the KSHV+ group by 6 months was 117 cells per cubic millimeter (95% CI: 102 to 132) compared with 107 cells per cubic millimeter (95% CI: 92 to 121) among the KSHV−. The KSHV+ group gained a similar number of cells at 6, 12 and 18 months compared with that in the KSHV− group (Table 2). The predicted CD4 trajectories from the start of ART were also similar for the groups (Fig. 1). The greatest increases occurred shortly after the treatment initiation for both groups, though the KSHV+ group retained consistently higher CD4 cell counts at all time points observed despite overall retention in care being slightly higher among those with KSHV at the end of follow-up (80% vs.73%). We also estimated differences in linear increase in CD4 count with log-transformed data to account for its nonnormal distribution. These results also showed little difference in increase in CD4 count comparing KSHV+ and KSHV− groups at 6, 12, or 18 months after ART initiation.
We then restricted the analysis to the KSHV+ group and considered the effect of a detectable KSHV viral load at ART initiation on subsequent CD4 response. Linear models adjusted for year of ART initiation, baseline WHO stage, hemoglobin, BMI, and tuberculosis status suggested little difference in the number of CD4 cells gained for those with an undetectable KSHV viral load at 6 (24.0 cells per cubic millimeter; 95% CI: −25.2 to 73.5), 12 (2.1 cells per cubic millimeter; 95% CI: −53.2 to 49.1), and 18 months on treatment (14.9 cells per cubic millimeter; 95% CI: −104.6 to 134.5) compared with those whose KSHV viral load was detectable.
HIV Viral Load Suppression
Achieving HIV virologic suppression on ART was common among both groups. By 6 months on treatment, only 4% of those with KSHV (5/143) had failed to suppress HIV viral load to <400 copies/mL while 10% (14/139) of those without KSHV had failed to achieve suppression. Among those who survived to a year on treatment, similar proportions achieving virologic suppression were noted 6% (6/109) vs. 8% (8/104). Estimates demonstrated similar virologic responses between the KSHV+ and KSHV– groups at both 6- (RR = 1.03; 95%CI:0.90-1.17) and 12-months (RR = 1.01; 95%CI:0.74-1.37) on treatment after adjustment for sex, age, CD4 count, co-infection with tuberculosis, hemoglobin and BMI (Table 3).
When we restricted the analysis to the KSHV+ group, those with antibodies to Orf73 only demonstrated the best virologic response of all the groups—100% (25/25) of this group achieved suppression of HIV viral load to <400 copies per milliliter by 6 and 12 months on treatment. All other groups achieved suppression rates >90% though, and there were no differences in the likelihood of achieving virologic suppression when we compared the KSHV− group with those reactive to lytic K8.1 (RR = 1.02; 95% CI: 0.83 to 1.25) or reactive to both antigens (RR = 1.02; 95% CI: 0.87 to 1.21) by 6 months on ART. The results were similar at 12 months of treatment.
The clinical effect of KSHV infection on immunologic and virologic outcomes in HIV-positive patients after ART initiation is unclear. In this prospective cohort study, we followed a group of KSHV+ adults without clinical evidence of KS and a group of KSHV− HIV-infected adults for 18 months after ART initiation. We found that, overall, immunologic and virologic responses to the first year of ART were similar between those coinfected with KSHV compared with their KSHV− counterparts. In general, KSHV infection, did not seem to have a negative impact on CD4 cell count recovery during the first 18 months on ART, though the impact of detectable KSHV viral load on CD4 cell count reconstitution requires further investigation.
The clinical course of infection with KSHV in the presence of an intact immune system is typically indolent and asymptomatic. Like all herpes viruses, KSHV establishes a persistent latent infection in its host and the number of KSHV-infected cells is controlled by the intact immune system.25 In the presence of HIV-1 coinfection, however, this course is altered. There is evidence that the immune suppression caused by HIV-1 and cytokine release promotes the reactivation of KSHV lytic genes14 including K8.1. This increase in viral progeny eventually results in the destruction of the host cell and progression to the development of KS, and the often-aggressive course seen in HIV-1 positive individuals.26,27
Although the pathogenesis from KSHV infection to clinical disease is complex and includes expansion of latently infected cells, the transition between latency and lytic replication may play a role in the development of progression to clinical disease and possibly even transmission of the virus.28,29 Although the detection of antibodies to lytic antigens is only a rough approximation of active lytic replication of KSHV (we also did not find a strong correlation between K8.1 antibody detection and detectable KSHV viral load in peripheral blood in this study), we did observe a poorer immunological response in terms of the number of CD4 cells gained within the first year on ART among those reactive to both lytic K8.1 and Orf73. Poor immunologic response to treatment and low CD4 cell counts have previously been associated with an increased risk of KS morbidity and mortality.30–32
The debate about the effect of KSHV on HIV-1 among those coinfected is ongoing. In vitro and in vivo studies suggested an interaction between these 2 viruses and an increase in HIV-1 viral load in the presence of KSHV.15 Conversely, there is evidence that KSHV infection is associated with the inhibition of HIV infection of CD4 cells expressing CCR3 receptors, largely mediated through beta chemokines.33,34 In this population, we observed little difference in the HIV viral load before ART initiation between the groups and a similar risk of failure to suppress the viral load at 6 and 12 months after ART initiation when comparing KSHV+ with KSHV−. Other in vitro work also suggested that KSHV increased HIV-1 replication in acutely infected cells and also induced the reactivation of HIV-1 replication in chronically infected cells.16 Our results suggest that those with KSHV coinfection achieved comparable virologic and immunologic responses with highly active antiretroviral therapy when compared with those without KSHV. Although the numbers were small and the results were somewhat imprecise, we noted that when stratified by reactivity to K8.1 alone, Orf73 alone, or both antigens, all 3 groups gained a similar number of CD4 cells over the first 18 months of ART and were at a similar risk of failing to achieve a 50-cell increase and a 100-cell increase at 6 and 12 months, respectively. The linear models for those KSHV+ and KSHV− were fit separately, allowing for different curves by exposure group. Despite this, the curves remained parallel suggesting that the only difference between the groups were the differences in baseline CD4 count at treatment initiation, and these remained almost perfectly consistent over time. When we restricted the analysis to the KSHV-infected group only, there was a trend to an association between detectable KSHV viral load and poor reconstitution of CD4 cells after ART initiation when compared with those where the virus was undetectable at initiation of ART. We emphasize here that the numbers in this subanalysis were small, limiting our power to detect statistically significant differences. Broadly speaking, our results concur with previous work among a cohort of long term nonprogressor MSMs, which also concluded that there was no impact of KSHV on the progression of HIV-1 infection in terms of CD4 cell count decline, HIV-1 viral load increase or CD4 cell viramia.35 The authors postulate that KSHV acts as an opportunistic agent rather than an HIV-cofactor among coinfected individuals. It is also possible that the effects are not apparent in the short term but may manifest later.
Our study has several strengths and limitations that should be considered when interpreting these findings. One important strength is that, unlike prior studies that have been conducted in the absence of any ART, our study investigates the clinical impact of the interaction between HIV and KSHV in the presence of ART. Early stage KS has been successfully treated with ART,36 yet the exact mechanism through which ART reduces the tumor is unknown. ART has been shown to reduce KSHV viral loads to undetectable levels.37 CD8+ lymphocytes are involved in the cellular immune response to several viruses including HIV-1 and herpes virus; these virus-specific cytotoxic T lymphocytes respond to both lytic and latent antigens of KSHV. Evidence of ART-induced immune reconstitution to KSHV (indicated by an undetectable KSHV viral load) was associated with an increase in CD8+ lymphocytes, suggesting restoration of these cells may be important in the control of KSHV infection25 and ultimately control or even prevention of KS. All models in this analysis were adjusted for CD8 cell count at the initiation of ART to attempt to account for this. However, as the timing of this immune restoration to KSHV in relation to immune restoration to HIV-1 has not yet been established (estimates vary from as little as 12 weeks37 to within 125 to 2years38 of initiating ART) it is possible that residual confounding exists in our estimates.
We note that with observational studies, particularly cohort studies, the potential for bias due to the differences in retention and follow-up exists. Although we cannot exclude this possibly completely, we do note that the mean follow-up time was very similar for both KSHV+ and KSHV− groups. Additionally, the proportions lost to follow-up were similar between the groups and <10% in both groups.
Our findings are also strengthened by the relatively large sample size. Although previous work has been conducted predominantly among small cohorts of mostly European males, we investigate this relationship among a relatively large sample of predominantly black heterosexual adults in an urban South African setting with a high prevalence of HIV and KSHV coinfection.
However, although the sample is large compared with that in previous clinical work in this field, we note that the limited precision of some of our estimates might mean that small differences were not detected. In addition, our sample is limited to individuals attending care at 1 urban facility in Johannesburg, South Africa. This may mean our results have limited generalizability to other nonurban settings or populations of men who have sex with men. Results may be different in the developed world especially, considering the differences in ART regimens used in these settings and the geographical and population differences1–3,6,39–41 in KSHV prevalence previously demonstrated.
We demonstrate good HIV treatment outcomes after ART initiation among a group of treatment naive, HIV-infected adults initiating ART at a large urban clinic in South Africa. In this population with a high KSHV prevalence, coinfection with KSHV does not seem to negatively impact the immunologic recovery or virologic response to ART in the first 18 months of treatment. The impact of detectable KSHV viral load on CD4 cell count recovery after the initiation of ART requires further investigation.
The authors thank all the study participants, doctors, and nurses at the TLC.
1. Klaskala W, Brayfield BP, Kankasa C, et al.. Epidemiological characteristics of human herpesvirus-8 infection in a large population of antenatal women in Zambia. J Med Virol. 2005;75:93–100.
2. Adjei A, Armah HB, Gbagbo F, et al.. Seroprevalence of HHV-8, CMV, and EBV among the general population in Ghana, West Africa. BMC Infect Dis. 2008;8:111.
3. Malope-Kgokong BI, Macphail P, Mbisa G, et al.. Kaposi's sarcoma associated-herpesvirus (KSHV) seroprevalence in pregnant women in South Africa. Infect Agent Cancer. 2010;5:14.
4. UNAIDS. Global Report. UNAIDS Report on the Global AIDS Epidemic. Joint United Nations Programme on HIV/AIDS (UNAIDS). Geneva, Switzerland: Joint United Nations Programme on HIV/AIDS. 2010.
5. Sarid R, Olsen SJ, Moore PS. Kaposi sarcoma-associated herpesvirus: epidemiology, virology, and molecular biology. Adv Virus Res. 1999;5:2139–2232.
6. Sitas F, Carrara H, Beral V, et al.. Antibodies against human herpesvirus 8 in black South African patients with cancer. N Engl J Med. 1999;340:1863–1871.
7. Whitby D, Howard MR, Tenant-Flowers M, et al.. Detection of Kaposi sarcoma associated herpesvirus in peripheral blood of HIV-infected individuals and progression to Kaposi sarcoma. Lancet. 1995;346:799–802.
8. Soulier J, Grollet L, Oksenhendler E, et al.. Kaposi sarcoma-associated herpesvirus-like DNA sequences in multicentric Castleman's disease. Blood. 1995;86:1276–1280.
9. Cesarman E, Chang Y, Moore PS, et al.. Kaposi sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas. N Engl J Med. 1995;332:1186–1191.
10. Boshoff C, Weiss R. AIDS-related malignancies. Nat Rev Cancer. 2002;2:373–382.
11. Covacs A, Schluchter M, Easley K, et al.. Cytomegalovirus infection and HIV-1 disease progression in infants born to HIV-1–infected women. New Engl J Med. 1999;341:77–84.
12. Greub B, Ledergerber B, Battegay M, et al.. Clinical progression, survival, and immune recovery during antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet. 2000;356:1800–1805.
13. Law WP, Duncombe CJ, Mahanontharit A, et al.. Impact of viral hepatitis co-infection on response to antiretroviral therapy and HIV disease progression in the HIV-NAT cohort. AIDS. 2004;18:1169–1177.
14. Mercader M, Taddeo B, Panella JR, et al.. Induction of HHV-8 lytic cycle replication by inflammatory cytokines produced by HIV-1-infected T cells. Am J Pathol. 2000;156:1961–1971.
15. Mercader M, Nickoloff BJ, Foreman KE. Induction of human immunodeficiency virus 1 replication by human herpesvirus 8. Arch Pathol Lab Med. 2001;125:785–789.
16. Caselli E, Galvan M, Cassai E, et al.. Human herpesvirus 8 enhances human immunodeficiency virus replication in acutely infected cells and induces reactivation in latently infected cells. Blood. 2005;106:2790–2797.
17. Fox MP, Maskew M, Macphail AP, et al.. Cohort profile: the Themba Lethu clinical cohort, Johannesburg, South Africa. Int J Epidemiol. 2012 doi: 10.1093/ije/dys029.
18. South African National Ministry of Health. National Antiretroviral Treatment Guidelines. 1st ed. Pretoria, South Africa: National Department of Health Republic of South Africa; 2004.
19. South African National Ministry of Health. The South African Antiretroviral Treatment Guidelines. Pretoria, South Africa: National Department of Health Republic of South Africa; 2010.
20. Fox MP, Brennan A, Maskew M, et al.. Using vital registration data to update mortality among patients lost to follow-up from ART programmes: evidence from the Themba Lethu Clinic, South Africa. Trop Med Int Health. 2010;15:405–413.
21. Egger M, Ekouevi DK, Williams C, et al.. Cohort Profile: the international epidemiological databases to evaluate AIDS (IeDEA) in sub-Saharan Africa. Int J Epidemiol. 2012;41:1256–1264.
22. Mbisa GL, Miley W, Gamache CJ, et al.. Detection of antibodies to Kaposi sarcoma associated herpesvirus: a new approach using K8.1 ELISA and a newly developed recombinant LANA ELISA. J Immunol Methods. 2010;356:39–46.
23. de Sanjosé S, Marshall V, Solà J, et al.. Prevalence of Kaposi sarcoma-associated herpesvirus infection in sex workers and women from the general population in Spain. Int J Cancer. 2002;98:155–158.
24. Maskew M, MacPhail AP, Whitby D, et al.. Prevalence and predictors of Kaposi sarcoma herpesvirus seropositivity: a cross-sectional analysis of HIV-infected adults initiating ART in Johannesburg, South Africa. Infect Agents Cancer. 2011;6:22.
25. Wilkinson J, Cope A, Gill J, et al.. Identification of Kaposi sarcoma-associated herpesvirus (KSHV)-specific cytotoxic T-lymphocyte epitopes and evaluation of reconstitution of KSHV-specific responses in human immunodeficiency virus type 1-Infected patients receiving highly active antiretroviral therapy. J Virol. 2002;76:2634–2640.
26. Aoki Y, Tosato G. HIV-1 Tat enhances Kaposi sarcoma-associated herpesvirus (KSHV) infectivity. Blood. 2004;104:810–814.
27. Lukac DM, Kirshner JR, Ganem D. Transcriptional activation by the product of open reading frame 50 of Kaposi sarcoma-associated herpesvirus is required for lytic viral reactivation in B cells. J Virol. 1999;73:9348–9361.
28. Adang LA, Parsons CH, Kedes DH. Asynchronous progression through the lytic cascade and variations in intracellular viral loads revealed by high-throughput single-cell analysis of Kaposi sarcoma-associated herpesvirus infection. J Virol. 2006;80:10073–10082.
29. Taylor MM, Chohan B, Lavreys L, et al.. Shedding of human herpesvirus 8 in oral and genital secretions from HIV-1-seropositive and -seronegative Kenyan women. J Infect Dis. 2004;190:484–488.
30. Chêne G, Sterne JA, May M, et al.. Prognostic importance of initial response in HIV-1 infected patients starting potent antiretroviral therapy, analysis of prospective studies. Lancet. 2003;362:679–686.
31. Carrieri MP, Raffi F, Lewden C, et al.. Impact of early versus late adherence to highly active antiretroviral therapy on immuno-virologic response, a 3-year follow-up study. Antivir Ther. 2003;8:585–594.
32. Lawn S, Harries A, Anglaret X, et al.. Early mortality among adults accessing antiretroviral treatment programmes in sub-Saharan Africa. AIDS. 2008;22:1897–1908.
33. Boshoff C, Endo Y, Collins PD, et al.. Angiogenic and HIV-inhibitory functions of KSHV-encoded chemokines. Science. 1997;278:290–294.
34. Lusso P. Chemokines and viruses: the dearest enemies. Virology. 2000;273:228–240.
35. Ait-Arkoub Z, Robert-Visse C, Calvez V, et al.. No influence of human herpesvirus 8 infection on the progression of HIV-1 infection in initially asymptomatic patients. AIDS. 2003;17:1394–1396.
36. Aversa SML, Cattelan AM, Salvagno L, et al.. Treatments of AIDS-related Kaposi sarcoma. Crit Rev Oncol/Hematol. 2005;53:253–265.
37. Gill J, Bourboulia D, Wilkinson J, et al.. Prospective study of the effects of antiretroviral therapy on Kaposi sarcoma—associated herpesvirus infection in patients with and without Kaposi sarcoma. J Acquir Immune Defic Syndr. 2002;31:384–390.
38. Bourboulia D, Aldam D, Lagos D, et al.. Short- and long-term effects of highly active antiretroviral therapy on Kaposi sarcoma-associated herpesvirus immune responses and viraemia. AIDS. 2004;18:485–493.
39. Malope BI, MacPhail P, Mbisa G, et al.. No evidence of sexual transmission of Kaposi sarcoma herpesvirus in a heterosexual South African population. AIDS. 2008;22:519–526.
40. Smith N, Sabin C, Gopal R, et al.. Serologic evidence of human herpesvirus 8 transmission by homosexual but not heterosexual sex. J Infect Dis. 1999;180:600–606.
41. Mbulaiteye SM, Biggar RJ, Pfeiffer RM, et al.. Water, socioeconomic factors, and human herpesvirus 8 infection in Ugandan children and their mothers. J Acquir Immune Defic Syndr. 2005;38:474–479.