1. ¹Epidemiology & Genetics Unit, Department of Health Sciences, Area 3, Seebohm Rowntree Building, Heslington, York YO10 5DD, UK
2. ²Cancer Research UK Cancer Epidemiology Unit, University of Oxford, Oxford, UK
3. ³Department of Public Health, University of Oxford, Oxford, UK
4. ⁴MRC Clinical Trials Unit, University College London, London, UK
5. ⁵Centre for Sexual Health and HIV Research, University College London, London, UK
6. ⁶Wohl Virion Centre, Division of Infection and Immunity, University College London, London, UK
7. ⁷Cancer Research UK Viral Oncology Group, Wolfson Institute of Medical Sciences, London, UK

Correspondence: Dr R Newton, E-mail: rob.newton@egu.york.ac.uk

Received 18 November 2005; Revised 14 March 2006; Accepted 15 March 2006.

Abstract

Antibody titres against Kaposi's sarcoma associated herpesvirus (KSHV or human herpesvirus 8 (HHV-8)) and Epstein–Barr virus (EBV) were examined in people who subsequently developed Kaposi's sarcoma and non-Hodgkin's lymphoma, within randomised controlled trials of antiretroviral therapy in adults infected with the human immunodeficiency virus-1 (HIV). For each case of Kaposi's sarcoma (n=189) and each case of non-Hodgkin's lymphoma (n=67), which developed after randomisation, one control was randomly selected from other trial participants, after matching for age, sex, ethnicity, mode of HIV transmission, type of treatment received and period of follow-up. Using sera taken an average of two and a half years before the diagnosis of cancer, titres of antibodies against KSHV latent (LANA) and lytic (K8.1) antigens and against EBV (VCA) antigens were investigated in relation to subsequent risks of cancer by calculating odds ratios (OR) using conditional logistic regression. Latent antibodies against KSHV were detectable among 38% (72 out of 189) of Kaposi's sarcoma cases and 12% (23 out of 189) of their controls (OR=4.4, 95% confidence intervals (CI) 2.3–8.3, P<0.001). The OR for Kaposi's sarcoma increased with increasing antilatent KSHV antibody titre (²₁ for trend=32.2, P<0.001). Lytic antibodies against KSHV were detectable among 33% (61 out of 187) of Kaposi's sarcoma cases and 19% (36 out of 187) of their controls (OR=2.0, 95% CI 1.2–3.4, P=0.003) and the OR for Kaposi's sarcoma increased with increasing antilytic KSHV antibody titre (²₁ for trend=6.2, P=0.02). Virtually, all cases and controls had anti-EBV antibodies detected and the OR for non-Hodgkin's lymphoma associated with a doubling of the anti-EBV antibody titre was estimated to increase by a multiplicative factor of 1.3 (95% CI 0.9–1.7, P=0.1). Kaposi's sarcoma was not associated with antibody levels against EBV (P=0.4) and non-Hodgkin's lymphoma was not associated with antibodies against KSHV (latent P=0.3; lytic P=0.5). Adjustment for CD4 count at the time of sample collection made no material difference to any of the results. In conclusion, among human immunodeficiency virus infected people, high levels of antibodies against KSHV latent and lytic antigens are strongly associated with subsequent risk of Kaposi's sarcoma but not non-Hodgkin's lymphoma. Antibody titre to EBV does not appear to be strongly associated with subsequent risk of Kaposi's sarcoma or non-Hodgkin's lymphoma in HIV infected people.