Incidence of AIDS-defining and Other Cancers in HIV-positive Children in South Africa

Abstract

Background Little is known on the risk of cancer in HIV-positive children in sub-Saharan Africa. We examined incidence and risk factors of AIDS-defining and other cancers in pediatric antiretroviral therapy (ART) programs in South Africa. Methods We linked the records of five ART programs in Johannesburg and Cape Town to those of pediatric oncology units, based on name and surname, date of birth, folder and civil identification numbers. We calculated incidence rates and obtained hazard ratios (HR) with 95% confidence intervals (CI) from Cox regression models including ART, sex, age, and degree of immunodeficiency. Missing CD4 counts and CD4% were multiply imputed. Immunodeficiency was defined according to World Health Organization 2005 criteria. Results Data of 11,707 HIV-positive children were included in the analysis. During 29,348 person-years of follow-up 24 cancers were diagnosed, for an incidence rate of 82 per 100,000 person-years (95% CI 55-122). The most frequent cancers were Kaposi Sarcoma (34 per 100,000 person-years) and Non Hodgkin Lymphoma (31 per 100,000 person-years). The incidence of non AIDS-defining malignancies was 17 per 100,000. The risk of developing cancer was lower on ART (HR 0.29, 95%CI 0.09–0.86), and increased with age at enrolment (>10 versus <3 years: HR 7.3, 95% CI 2.2-24.6) and immunodeficiency at enrolment (advanced/severe versus no/mild: HR 3.5, 95%CI 1.1-12.0). The HR for the effect of ART from complete case analysis was similar but ceased to be statistically significant (p=0.078). Conclusions Early HIV diagnosis and linkage to care, with start of ART before advanced immunodeficiency develops, may substantially reduce the burden of cancer in HIV-positive children in South Africa and elsewhere.

HIV-associated malignancies in children

“…Record linkages between data of HIV-treatment programs or HIV-registries and cancer registries have been identified as a potential solution to this problem [7,8]. Three recent studies from the USA, Taiwan and South Africa (SA) have used this method to estimate cancer incidence rates in HIV-infected children [9–11]. The SA study linked data at five pediatric cART programs using probabilistic record linkage methods with four referral pediatric oncology units [9].…”

“…Three recent studies from the USA, Taiwan and South Africa (SA) have used this method to estimate cancer incidence rates in HIV-infected children [9–11]. The SA study linked data at five pediatric cART programs using probabilistic record linkage methods with four referral pediatric oncology units [9]. Amongst a total of 11,707 children, 47 prevalent and 24 incident cancer cases were identified.…”

Mapping the Epidemiology of Kaposi Sarcoma and Non-Hodgkin Lymphoma Among Children in Sub-Saharan Africa: A Review

“…In this review, the median incidence of KS alone was much higher at 67.35/100,000 patient‐years among HIV‐infected populations, which is similar to the results of a recent abstract that included more than 15,000 children with HIV on antiretroviral therapy (ART) in 15 countries in Africa and found a KS incidence of 67/100,000 patient‐years . The incidence found in this review is nearly double the incidence of a study from the South African cohort of the same study group in which the incidence of KS was 34/100,000 patient‐years . In a pre‐ART study covering parts of Uganda, Tanzania, and the Democratic Republic of Congo, the estimated lifetime incidence of KS was as high as 1,600/100,000 patient‐years …”

“…However, Bohlius et al. found an incidence of NHL of 31/100,000 patient‐years among children with HIV in South Africa, indicating the need for future studies to better characterize BL and overall NHL among HIV‐positive children throughout SSA.…”

Brief Report: Pediatric Cancer Burden and Treatment Resources Within the Pediatric IeDEA Consortium

“…Although there are limitations to the interpretation of these results, the large geographic coverage of this survey highlights the magnitude of the investment that is needed across several regions for cancer diagnosis, surgery, chemo- and radiotherapy. There are also additional epidemiologic patterns worth further investigation, such as recent evidence that a reduction in some HIV-related cancers is observed when ART is initiated prior to severe immunodeficiency 13 . Thus, collaboration with HIV clinics in establishing databases and accurate registries while continuing surveillance of treatment modalities is essential to further current understanding of the coexistence of these diseases.…”