SERUM HIV-1 RNA LEVELS AND TIME TO DEVELOPMENT OF AIDS IN THE MULTICENTER HEMOPHILIA COHORT STUDY

Objective.-To determine if the long-term incidence of the acquired imm unodeficiency syndrome (AIDS) is related to human immunodeficiency vir us type 1 (HIV-1) RNA levels measured early in HIV-1 infection. Design .-Epidemiologic cohort study. Setting.-Five hemophilia treatment cente rs in the United States. Subjects.-A total of 165 subjects with hemoph ilia and HIV-1 infection (age at HIV-1 seroconversion, 1-66 years) fol lowed from 1979 to 1995. Methods.-The HIV-1 RNA level was measured by polymerase chain reaction over a range of 200 to 1 million or more HIV -1 RNA copies/mL. in archived serum specimens collected 12 to 36 month s (median, 27 months) after the estimated date of HIV-1 seroconversion . Kaplan-Meier methods were used to examine the risk of AIDS and propo rtional hazards models were used to estimate relative hazards. Results .-The HIV-1 RNA values were similar in subjects younger than 17 years at seroconversion (median, 5214 copies/mL) and those 18 to 34 years ol d (median, 4693 copies/mL), but higher in those 35 years or older (med ian, 12 069 copies/mL) (P=.02 compared with each younger group). At 10 years after seroconversion, the proportions of subjects with AIDS wer e 72% among subjects with 100 000 or more HIV-1 RNA copies/mL measured 12 to 36 months after HIV-1 seroconversion (n=9), 52% among subjects with 10 000 to 99 999 copies/mL (n=55), 22% among subjects with 1000 t o 9999 copies/mL (n=82), and 0% among subjects with fewer than 1000 co pies/mL (n=19) (P<.001). The age-adjusted relative hazard for AIDS for subjects with 10 000 or more copies/mL was 14.3 (95% confidence inter val, 1.9-105.6) compared with subjects with fewer than 1000 copies/mL. Conclusions.-The HIV-1 RNA level during early chronic HIV-1 infection is a strong, age-independent predictor of clinical outcome; low revel s define persons with a high probability of long-term AIDS-free surviv al.