Despite recent declines in the incidence of Kaposi's sarcoma (KS) , it remains the most common malignancy among individuals with AIDS . Although infection with human herpesvirus 8 (HHV-8) is required for the development of KS [2,3], current strategies for the treatment of AIDS-KS generally do not include antiherpesvirus drugs, but instead emphasize the use of HAART , treatment of KS lesions locally , and administration of chemotherapeutic agents [5,6]. Antiherpesvirus drugs, such as ganciclovir or valganciclovir, mitigate HHV-8 replication and may be beneficial in the management of KS [7,8], particularly among patients for whom the standard treatment options are contraindicated or ineffective. Although established KS lesions have not shown good response to treatment with antiherpesvirus drugs, there is evidence that these agents can help prevent the development of new KS or stabilize existing KS .
Because of limited and conflicting reports in the literature, we carried out the largest longitudinal study to date that examines the association between KS disease progression and the presence and viral load of HHV-8 in 96 men with or at high risk for developing KS. Previous reports on this study described risk factors for KS at enrollment , and monitored the frequency of HHV-8 detection and clinical outcome in short time intervals that spanned 3 months . The current study will describe how HHV-8 viral load is associated with KS progression and regression in 6-month intervals that span 2 years.
Study participants and evaluation of Kaposi's sarcoma
We enrolled 96 HIV-seropositive men at two HIV clinics in Atlanta, Georgia, between April 1999 and April 2001. Men were eligible for inclusion if they had a diagnosis of KS or were without KS but seropositive for HHV-8. At each visit the men were evaluated for their KS clinical features, HIV viral load, CD4 cell count, antiretroviral therapy use, and blood and oral fluids were collected as described . The visits included a baseline visit and four additional visits at 6-month intervals over 2 years.
At each visit, the patient's KS was evaluated by a physician and changes since the previous visit were noted and corroborated by the patients' self-reported changes in their KS. At each follow-up visit we classified patients as KS progressors if they had at least one of the following developments since their last study visit: increase in number or size of lesions, new visceral involvement, new oropharynx involvement, increase in edema, or increase in ulceration. We classified patients as regressors if they showed improvement in at least one of the clinical markers above, with no progression. We classified patients as having stable KS if they had no worsening or improvement of clinical symptoms.
Human herpesvirus 8 serologic testing
Antibodies to HHV-8 were measured in serum using two enzyme-linked immunosorbent assays (ELISA) based on peptides to the K8.1 and orf 65 viral open reading frames, and an immunofluorescence assay based on the HHV-8-infected BCBL cell line as described .
Human herpesvirus 8 and HIV viral load testing
Peripheral blood mononuclear cells (PBMC) were separated from whole blood with Lymphocyte Separation Medium (ICN Biomedicals, Aurora, Ohio, USA) following the manufacturer's instructions. DNA was extracted from PBMC and oral fluid using Qiagen mini kits (Valencia, California, USA). The HHV-8 PCR used Taqman reagents and each 50 μl reaction had extracted DNA from approximately 150 000 PBMC determined by spectrophotometric reading of DNA or 200 μl of oral fluid, 600 nmol/l each primer, and 200 nmol/l probe for the viral orf 25, performed in duplicate with a sensitivity of 5–10 genome copies per PCR reaction or approximately 30–60 genome copies per 1 million cells as described . Positive reactions were confirmed with a second PCR amplification using the HHV-8 orf 37 primers and probes . HIV viral loads were measured with reverse transcription-PCR Taqman as described .
We used statistical models to assess the associations between different variables and to control for potential confounding. Except where otherwise stated, model parameters were estimated using generalized estimating equations (GEE). To assess the statistical validity of the models, we performed goodness-of-fit and collinearity assessments. In all analyses, HHV-8 viral load was log10 transformed.
Of the 96 enrollees, 92 were men who have sex with men and four were injection drug users; 47 had a clinical diagnosis of KS. The 1-year, 1.5-year and 2-year study completion rates were 67, 52 and 44%, respectively, for patients with KS and 64, 59 and 53%, respectively, for patients without KS. Mortality during follow-up was 13% for patients with KS and 6% for patients without KS. HAART adherence was 77.6% for patients with KS and 73.3% for patients without KS.
Human herpesvirus 8 viral shedding in peripheral blood mononuclear cells and oral fluid
HHV-8 DNA was detected in PBMC or oral fluid of 63% (60/96) of study participants at one or more clinic visits (Table 1). In multivariate models that adjusted for presence of KS, CD4 cell count, and HIV viral load, the presence of HHV-8 DNA (‘shedding’) in PBMC was associated with the presence of HHV-8 DNA in oral fluid [adjusted odds ratio (aOR), 5.1; 95% confidence interval (CI), 2.7–9.7; P < 0.001], and HHV-8 viral load in PBMC was positively associated with HHV-8 viral load in oral fluids (slope coefficient = 0.18; standard error (se) = 0.04; P < 0.001).
The presence of HHV-8 DNA in PBMC was more common in men with KS than in men without KS (aOR = 3.4; 95% CI, 1.8–6.4; P < 0.001), adjusted for CD4 cell count and HIV viral load. Similarly, HHV-8 viral loads in PBMC were significantly higher in patients with KS (mean = 294 100 genome copies per million cells) than in patients without KS (mean = 726 genome copies per million cells) (P < 0.001). In contrast, neither the presence of HHV-8 DNA (P = 0.20) nor viral load (P = 0.51) in oral fluids differed significantly between patients with and without KS.
Individuals who shed HHV-8 DNA in PBMC or oral fluid at one time point were more likely to shed virus at other time points. In fact, of all the clinical and immunological parameters we tested, HHV-8 shedding at a previous visit was the best predictor for shedding at future visits in PBMC (aOR = 15.1; 95% CI = 6.9–32.9) and oral fluid (aOR = 11.0; 95% CI = 4.8–25.5). In addition, the magnitude of HHV-8 viral loads at earlier visits was predictive of that in PBMC at subsequent visits (slope coefficient = 0.62; P < 0.001) and oral fluid (slope coefficient = 0.63; P < 0.001), adjusting for KS status, CD4 cell count and HIV viral load.
HHV-8 DNA presence or viral load was not significantly associated with HIV viral load, CD4 T-cell count, HAART, or race. However, as in our previous studies [10,11], we found an inverse association between HHV-8 ORF65 antibody titer and the presence of HHV-8 DNA. Serum ORF65 titers > 25 600 were inversely associated with the presence of HHV-8 DNA in PBMC (OR = 0.81; 95% CI = 0.45–1.5), and to an even greater degree, oral fluid (OR = 0.34; 95% CI = 0.13–0.91). A number of other variables were examined for associations with HHV-8 antibody titer levels, but no significant associations were found.
Clinical change in Kaposi's sarcoma status: factors associated with clinical progression and regression
Among the 47 patients with active KS or a history of KS without apparent lesions at enrollment, 16 patients had visits at which their KS had progressed (24 patient-visits), 40 patients had visits at which their KS remained stable (78 patient-visits), and 25 patients had visits at which their KS had regressed (53 patient-visits). In comparison with other KS patients, patients with worsening KS (clinical progressors) were more likely to have HHV-8 DNA in PBMC (OR = 2.7; 95% CI = 1.1–6.8) or oral fluid (OR = 2.2; 95% CI = 1.0–5.0) (Fig. 1a). In addition, clinical progressors were more likely to have HHV-8 viral loads greater than 10 000 genome copies per million cells in comparison with nonprogressors (OR = 7.7; 95% CI = 2.3–25.4; P = 0.001). Mean HHV-8 viral loads in PBMC and oral fluids were highest among progressors, intermediate among patients with stable KS, and lowest among regressors (Fig. 1b). Specifically, for every 10-fold increase in HHV-8 DNA copy number in PBMC, the risk of KS progression increased by approximately 30% (OR = 1.3; P = 0.02); similarly for every 10-fold increase in HHV-8 DNA copy number in oral fluid the risk of KS progression increased by 20% (OR = 1.2; P = 0.11).
We describe a significant association between changes in KS disease severity and HHV-8 DNA presence and viral load. Patients with progressing KS had a 2.7-fold higher likelihood of having HHV-8 DNA in blood and higher viral loads than patients with stable or regressing KS. We had insufficient data to confirm a threshold viral load for disease progression; however, patients with progressing KS had a 7.7-fold higher likelihood of having HHV-8 viral loads greater than 10 000 genome copies per million cells compared to patients with regressing KS. These findings are consistent with a rise in active HHV-8 replication leading to new KS lesion formation.
Several previous studies have examined HHV-8 viral load as a correlate of KS severity at a given visit [10,14–19]. A unique aspect of this study is that we examined HHV-8 load as a correlate of how KS severity changed from one visit to the next. Although we could not demonstrate causality, our data show that KS is more likely to worsen when HHV-8 is present at high levels in PBMC.
Our findings add to evidence suggesting that clinicians could use HHV-8 viral load testing in PBMC in patients with KS to assess the risk of further disease progression. For KS patients who are nonresponsive to HAART and who have high or increasing HHV-8 viral loads, it may be appropriate to consider antiherpes treatments such as ganciclovir or valganciclovir. As HHV-8 load appears to be associated with KS disease progression, reductions in HHV-8 viral load through the use of antiherpes agents may help stabilize the KS and protect against further progression.
Study limitations should be noted. We could not confirm that HHV-8 viral load predicted the first onset of KS because none of the 49 patients without KS, all of whom were on HAART, developed KS during the 2 years of follow-up. Nevertheless, others have shown that among individuals who are seropositive for both HIV and HHV-8, the presence of HHV-8 DNA in PBMC predicts KS . Another limitation of our study was that, although patients were followed for 2 years, assessments of KS clinical severity and HHV-8 viral load were conducted at 6-month intervals. Important KS clinical events or peaks in HHV-8 viral load could have occurred between study visits and been missed. A third limitation is the possibility of bias from loss to follow-up.
In conclusion, we describe a significant positive association between changes in KS severity and HHV-8 viral loads. Although the precise temporality of HHV-8 viral load and changes in KS disease are likely to be variable within individuals, monitoring of HHV-8 viral load among patients at greatest risk of KS might be useful for informing treatment decisions.
The authors would like to thank the study participants for their time, and Karen McCaustland and Nathan Kow for their contributions to this project.
Consent: Informed consent was obtained from all study participants in accordance with the guidelines for human experimentation of the US Department of Health and Human Services.
Conflicts of interest: None of the coauthors have financial interests in any product that could potentially be related to the work presented in this report.
Financial support: This study was supported by the CDC Opportunistic Infections Program.
Disclaimer: The results and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
1. Bernstein WB, Little RF, Wilson WH, Yarchoan R. Acquired immunodeficiency syndrome-related malignancies in the era of highly active antiretroviral therapy. Int J Hematol 2006; 84:3–11.
2. Cannon MJ, Laney AS, Pellett PE. Human herpesvirus 8
: current issues. Clin Infect Dis 2003; 37:82–87.
3. Dukers NH, Rezza G. Human herpesvirus 8 epidemiology
: what we do and do not know. AIDS 2003; 17:1717–1730.
4. Aversa SM, Cattelan AM, Salvagno L, Crivellari G, Banna G, Trevenzoli M, et al
. Treatments of AIDS-related Kaposi's sarcoma
. Crit Rev Oncol Hematol 2005; 53:253–265.
5. Sturzl M, Zietz C, Eisenburg B, Goebel FD, Gillitzer R, Hofschneider PH, Bogner JR. Liposomal doxorubicin in the treatment of AIDS-associated Kaposi's sarcoma
: clinical, histological and cell biological evaluation. Res Virol 1994; 145:261–269.
6. Welles L, Saville MW, Lietzau J, Pluda JM, Wyvill KM, Feuerstein I, et al
. Phase II trial with dose titration of paclitaxel for the therapy of human immunodeficiency virus-associated Kaposi's sarcoma
. J Clin Oncol 1998; 16:1112–1121.
7. Martin DF, Kuppermann BD, Wolitz RA, Palestine AG, Li H, Robinson CA. Oral ganciclovir for patients with cytomegalovirus retinitis treated with a ganciclovir implant. Roche Ganciclovir Study Group. N Engl J Med 1999; 340:1063–1070.
8. Robles R, Lugo D, Gee L, Jacobson MA. Effect of antiviral drugs used to treat cytomegalovirus end-organ disease on subsequent course of previously diagnosed Kaposi's sarcoma
in patients with AIDS. J Acquir Immune Defic Syndr Hum Retrovirol 1999; 20:34–38.
9. Klass CM, Offermann MK. Targeting human herpesvirus
-8 for treatment of Kaposi's sarcoma
and primary effusion lymphoma. Curr Opin Oncol 2005; 17:447–455.
10. Cannon MJ, Dollard SC, Black JB, Edlin BR, Hannah C, Hogan SE, et al
. Risk factors for Kaposi's sarcoma
in men seropositive for both human herpesvirus 8
and human immunodeficiency virus. AIDS 2003; 17:215–222.
11. Laney AS, Dollard SC, Jaffe HW, Offermann MK, Spira TJ, Gunthel CJ, et al
. Repeated measures study of human herpesvirus 8
(HHV-8) DNA and antibodies in men seropositive for both HHV-8 and HIV. AIDS 2004; 18:1819–1826.
12. Stamey FR, Patel MM, Holloway BP, Pellett PE. Quantitative, fluorogenic probe PCR assay for detection of human herpesvirus 8
DNA in clinical specimens. J Clin Microbiol 2001; 39:3537–3540.
13. Martro E, Cannon MJ, Dollard SC, Spira TJ, Laney AS, Ou CY, Pellett PE. Evidence for both lytic replication and tightly regulated human herpesvirus 8
latency in circulating mononuclear cells, with virus loads frequently below common thresholds of detection. J Virol 2004; 78:11707–11714.
14. Campbell TB, Borok M, Gwanzura L, MaWhinney S, White IE, Ndemera B, et al
. Relationship of human herpesvirus 8
peripheral blood virus load and Kaposi's sarcoma
clinical stage. AIDS 2000; 14:2109–2116.
15. Mendez JC, Procop GW, Espy MJ, Paya CV, Smith TF. Detection and semiquantitative analysis of human herpesvirus 8
DNA in specimens from patients with Kaposi's sarcoma
. J Clin Microbiol 1998; 36:2220–2222.
16. van der Kuyl AC, Polstra AM, van den Burg R, Jan Weverling G, Goudsmit J, Cornelissen M. Cytomegalovirus and human herpesvirus 8
DNA detection in peripheral blood monocytic cells of AIDS patients: correlations with the presence of Kaposi's sarcoma
and CMV disease. J Med Virol 2005; 76:541–546.
17. Boneschi V, Brambilla L, Berti E, Ferrucci S, Corbellino M, Parravicini C, Fossati S. Human herpesvirus 8
DNA in the skin and blood of patients with Mediterranean Kaposi's sarcoma
: clinical correlations. Dermatology 2001; 203:19–23.
18. Marcelin AG, Gorin I, Morand P, Ait-Arkoub Z, Deleuze J, Morini JP, et al
. Quantification of Kaposi's sarcoma
in blood, oral mucosa, and saliva in patients with Kaposi's sarcoma
. AIDS Res Hum Retroviruses 2004; 20:704–708.
19. Lebbe C, Agbalika F, de Cremoux P, Deplanche M, Rybojad M, Masgrau E, et al
. Detection of human herpesvirus 8
and human T-cell lymphotropic virus type 1 sequences in Kaposi sarcoma. Arch Dermatol 1997; 133:25–30.
20. Engels EA, Biggar RJ, Marshall VA, Walters MA, Gamache CJ, Whitby D, Goedert JJ. Detection and quantification of Kaposi's sarcoma
to predict AIDS-associated Kaposi's sarcoma
. AIDS 2003; 17:1847–1851.