ABSTRACT
Together with CD4(+)-cell counts as an indicator of immune function, the use of human immunodeficiency virus type 1 (HIV-1) RNA levels as a direct marker of viral load has gained widespread attention for evaluation of patient clinical status. Results obtained with other HIV-1 markers for this purpose are often inconsistent. This study examined the relationship between various HIV-1 markers by using clinical specimens (plasma) from HIV-1-infected individuals at different stages of disease progression and supernatant fluid from four human T-lymphocyte cell lines chronically infected with HIV-1. Cell culture specimens were collected periodically over 7 days and were tested for HIV-1 RNA levels with a nucleic acid amplification assay, for p24 with an enzyme-linked immunosorbent assay, and for reverse transcriptase activity by isotope uptake. An increase in the level of each marker was observed over the 7-day period with each of the four HIV-1 strains tested (LAV1, HTLV-IIIB, MN, and ARV2); with these specimens, the frequency of detection for each marker was 100%. In the clinical specimens, HIV-1 RNA was detected more often (143 of 183 specimens [78%]) than was p24 (87 of 183 [48%]); little correlation between the levels of the two markers was seen. In these clinical specimens evaluated, CD4(+)-cell counts were better correlated with the frequency and levels of HIV-1 RNA than with p24. In specimens (n = 38) collected serially from six HIV-1-infected subjects, HIV-1 RNA was detected more often (33 of 38 [85%]) than p24 (23 of 38 [59%]). When reported by the assays used, the levels of both HIV-1 markers fluctuated over time for each of the subjects. Although the markers correlated in the in vitro systems studied, the observed differences in the correlation of levels and frequencies of HIV-1 markers in vivo indicate that p24 has less clinical utility than does viral load testing when used in conjunction with CD4(+)-cell counts as a measure of immune system functioning.