The Amplicor HIV-1 Monitor test was compared to the nucleic acid sequence-based amplification (Nasba) assay system for the quantitation of human immunodeficiency virus (HIV) RNA in three different types of clinical samples: plasma, serum, and plasma subjected to freeze-and-thaw cycles. Each assay detected HIV RNA in the same 73 (90%) of 81 samples tested, and the quantitative results obtained with the two assays were significantly correlated. Both assays detected higher RNA levels in patients with CD4+ cell counts lower than 200 cells/mm3 than in patients with CD4+ cell counts higher than 200 cells/mm3. In addition, RNA levels in plasma higher than 5 logs predicted higher numbers of clinical events than did RNA levels in plasma lower than 5 logs. Quantitation of HIV RNA in paired plasma and serum samples showed lower HIV RNA content in serum than in the paired plasma sample, with mean differences between HIV RNA contents of plasma and serum of 0.54 and 0.28 log RNA copy/ml by the Nasba assay and the Amplicor HIV-1 Monitor assay, respectively. No significant loss of HIV RNA was detected with either assay in plasma samples subjected to multiple freeze-and-thaw cycles. These studies demonstrate that the Nasba and Amplicor assays perform similarly with plasma and serum samples. Further, the results indicate that freeze-and-thaw cycles do not result in significant loss of detectable HIV RNA.