A117

Objective

To compare the incidence rate of all non-AIDS cancers combined among HIV-infected individuals with the general population rate and to estimate modification of the association by gender and AIDS diagnosis.

Design

A meta-analysis was carried out using data from 11 studies of non-AIDS cancer incidence in HIV-infected individuals.

Methods

Standardized incidence rate ratios (SIR) for non-AIDS cancers in HIV-infected individuals compared to the general population and 95% confidence limits (CL) were abstracted from each study. The results were assessed for between-study heterogeneity and funnel plot asymmetry. Meta-regression was used to estimate modification of the general-population SIR by gender and AIDS.

Results

There was pronounced heterogeneity between studies, thus, an overall pooled estimate was not calculated. The SIR of non-AIDS cancers in HIV-infected individuals, adjusted for AIDS and study design was greater among men (2.3; 95% CL: 1.7, 3.1) than women (1.5; 95% CL: 0.9; 2.3; ratio of SIRs = 1.6; 95% CL: 0.7, 3.2). The SIR also appeared slightly greater among those with an AIDS diagnosis; however, when the analysis was adjusted for gender and study design, the SIRs for those with (1.88; 95% CL: 1.34, 2.66) and without AIDS (1.98; 95% CL: 1.35, 2.88) were similar (ratio of SIRs: 0.95; 95% CL: 0.46, 1.96).

Conclusions

The incidence rate of non-AIDS cancers was elevated among HIV-infected individuals. Additionally, HIV-infected men appeared to have a greater SIR for non-AIDS cancers than HIV-infected women. However, no difference was seen between the SIRs for non-AIDS cancers among those with and without AIDS.

Citation Information: Cancer Prev Res 2008;1(7 Suppl):A117.

Seventh AACR International Conference on Frontiers in Cancer Prevention Research-- Nov 16-19, 2008; Washington, DC