Assessment of HIV-1 RNA concentration is widely used for monitoring antiretroviral therapies. Tests are, however, expensive and require technically advanced equipment and highly trained personnel. Increasing availability of antiretroviral treatment in resource-poor settings calls for simple and inexpensive virus tests. HIV-1 p24 antigen tests were frequently used before the availability of nucleic acid tests (NAT). Two simple modifications, heat-mediated destruction of test-interfering antibodies and increased sensitivity achieved by signal amplification, have shaped the p24 antigen test into a tool that rivals NAT. This improved p24 antigen test, for which all reagents are available from Perkin Elmer Life Sciences, was evaluated in clinical studies in comparison with the most sensitive PCR methods available at a given time. In a prospective study over 4 years, HIV-1 infection among 859 samples from 307 infants born to HIV-positive mothers in Switzerland was detected as sensitively by p24 antigen assay as by PCR for viral DNA or RNA: 100% sensitivity of all methods after 10 days of age; 99.2% diagnostic specificity of p24 after neutralization (RNA, 98.6%). A study conducted in Dar es Salaam (Tanzania) found 123 of 125 samples from 76 PCR-positive infants positive for p24 antigen (sensitivity = 98.7%). In 169 infected Swiss adults with a median CD4+ T-cell count of 140 cells/microliter followed for a median of 2.7 years, p24 at baseline correlated as well as or better than HIV-1 RNA with the ensuing CD4+ T-lymphocyte decline and was independently predictive of progression to clinical AIDS (P = 0.043) and survival (P = 0.032). RNA predicted AIDS (P < 0.005), but not survival (P = 0.19). Another study of first-visit samples from 496 mostly black IVDU in the U.S. with a median CD4+ count of 518 cells/microliter showed equally strong prediction of progression to clinical AIDS for p24 antigen, HIV-1 RNA, and CD4+ T-lymphocyte concentrations at baseline. Treatment-associated changes in p24 and RNA levels in adults and children correlated well in three Swiss studies. The half-life of p24 antigen in the first phase of effective treatment was 1.6 +/- 0.4 days (RNA, 1.7 +/- 0.8). A second, slower decay phase had a half-life of 42 +/- 16 days. One study suggested that a strategy involving a somewhat more frequent testing for p24 antigen permitted detection of viral failures significantly earlier than tests for HIV-1 RNA conducted at 3-month intervals, while at the same time significantly saving on costs. Experience from three studies indicates that the p24 antigen test recognizes viruses of subtypes A-G and O, as well as some recombinant isolates, but leaves open the possibility that some non-B p24 antigens may be suboptimally detected. This improved p24 antigen test provides diagnosis of pediatric HIV infection, prediction of prognosis and treatment monitoring in quality comparable to tests for HIV-1 RNA, but at much lower costs. There is no problem with sample instability and no need for cumbersome nucleic acid extraction. The test is validated for subtype B, but requires further studies for non-B subtypes.