Predicting risk of cancer during HIV infection: the role of inflammatory and coagulation biomarkers

Objective: To investigate the relationship between inflammatory [interleukin-6 (IL-6) and C-reactive protein (CRP)] and coagulation (D-dimer) biomarkers and cancer risk during HIV infection. Design: A prospective cohort. Methods: HIV-infected patients on continuous antiretroviral therapy (ART) in the control arms of three randomized trials (N = 5023) were included in an analysis of predictors of cancer (any type, infection-related or infection-unrelated). Hazard ratios for IL-6, CRP and D-dimer levels (log(2)-transformed) were calculated using Cox models stratified by trial and adjusted for demographics and CD4(+) cell counts and adjusted also for all biomarkers simultaneously. To assess the possibility that biomarker levels were elevated at entry due to undiagnosed cancer, analyses were repeated excluding early cancer events (i.e. diagnosed during first 2 years of follow-up). Results: During approximately 24 000 person-years of follow-up (PYFU), 172 patients developed cancer (70 infection-related; 102 infection-unrelated). The risk of developing cancer was associated with higher levels (per doubling) of IL-6 (hazard ratio 1.38, P < 0.001), CRP (hazard ratio 1.16, P 0.001) and D-dimer (hazard ratio 1.17, P = 0.03). However, only IL-6 (hazard ratio 1.29, P = 0.003) remained associated with cancer risk when all biomarkers were considered simultaneously. Results for infection-related and infection-unrelated cancers were similar to results for any cancer. Hazard ratios excluding 69 early cancer events were 1.31 (P = 0.007), 1.14 (P = 0.02) and 1.07 (P = 0.49) for IL-6, CRP and D-dimer, respectively. Conclusion: Activated inflammation and coagulation pathways are associated with increased cancer risk during HIV infection. This association was stronger for IL-6 and persisted after excluding early cancer. Trials of interventions may be warranted to assess whether cancer risk can be reduced by lowering IL-6 levels in HIV-positive individuals. (C) 2013 Wolters Kluwer Health vertical bar Lippincott Williams & Wilkins