Cancer risk in people infected with human immunodeficiency virus in the United States.

Data are limited regarding cancer risk in human immunodeficiency virus (HIV)-infected persons with modest immunosuppression, before the onset of acquired immunodeficiency syndrome (AIDS). For some cancers, risk may be affected by highly active antiretroviral therapy (HAART) widely available since 1996. We linked HIV/AIDS and cancer registries in Colorado, Florida and New Jersey. Standardized incidence ratios (SIRs) compared cancer risk in HIV-infected persons (initially AIDS-free) during the 5-year period after registration with the general population. Poisson regression was used to compare incidence across subgroups, adjusting for demographic factors. Among 57,350 HIV-infected persons registered during 1991-2002 (median CD4 count 491 cells/mm(3)), 871 cancers occurred during follow-up. Risk was elevated for Kaposi sarcoma (KS, SIR 1,300 [n = 173 cases]), non-Hodgkin lymphoma (NHL, 7.3 [n = 203]), cervical cancer (2.9 [n = 28]) and several non-AIDS-defining malignancies, including Hodgkin lymphoma (5.6 [n = 36]) and cancers of the lung (2.6 [n = 109]) and liver (2.7 [n = 14]). KS and NHL incidence declined over time but nonetheless remained elevated in 1996-2002. Incidence increased in 1996-2002 compared to 1991-1995 for Hodgkin lymphoma (relative risk 2.7, 95%CI 1.0-7.1) and liver cancer (relative risk infinite, one-sided 95%CI 1.1-infinity). Non-AIDS-defining cancers comprised 31.4% of cancers in 1991-1995, versus 58.0% in 1996-2002. For KS and NHL, risk was inversely related to CD4 count, but these associations attenuated after 1996. We conclude that KS and NHL incidence declined markedly in recent years, likely reflecting HAART-related improvements in immunity, while incidence of some non-AIDS-defining cancers increased. These trends have led to a shift in the spectrum of cancer among HIV-infected persons.

Enhanced Collection of Laboratory Data in HIV Surveillance Among 5 States with Confidential Name-based HIV Infection Reporting, 2005-2006.

Laboratory data reported through HIV surveillance can provide information about disease severity and linkage to care; however these measures are only as accurate as the quality and completeness of data reported. Using data from five states that implemented enhanced collection of laboratory data in HIV surveillance from 2005-2006, we determined completeness of reporting, stage of disease at diagnosis, the most common opportunistic illnesses (OI) at diagnosis, and linkage to medical care. Methods to enhance laboratory reporting included increasing active surveillance efforts, identifying laboratories not reporting to HIV surveillance, increasing electronic reporting, and using laboratory results from auxiliary databases. Of 3,065 persons ≥13 years of age diagnosed with HIV, 35.5% were diagnosed with stage 3 (AIDS) and 37.7% progressed to stage 3 within 12-months after diagnosis. Overall, 78.5% were linked to care within 3 months; however, a higher proportion of persons with ≥1 CD4 or viral load test was found among whites compared with blacks/African Americans (82.1% vs 73.6%, p<0.001). Few (12.3%) had an OI within 3 months of diagnosis. The completeness of laboratory data collected through surveillance was improved with enhanced reporting and provided a more accurate picture of stage of disease and gaps in linkage to care. Additional interventions are needed to meet the goals of the National HIV/AIDS Strategy on linkage to care and the reduction of HIV-related disparities.