Performance of the Abbott RealTime HIV-1 assay versus the Roche Amplicor HIV-1 MonitorTM Test, v1.5, UltraSensitive assay for samples with low plasma HIV-1 RNA copy numbers

Sir, The Roche Amplicor HIV-1 Monitor Test, v1.5, UltraSensitive assay has been used to measure HIV-1 RNA levels in HIV clinical trials, but this assay has recently been discontinued. The new replacement assay, Roche TaqMan HIV-1 Test, v2.0, tends to show low levels of quantifiable HIV-1 RNA in samples undetectable by the Roche Amplicor UltraSensitive assay. An alternative assay is the Abbott RealTime HIV-1 assay, which has a lower quantification limit of 40 copies/mL. Both the Roche Amplicor UltraSensitive assay and the Abbott RealTime assay can detect traces of HIV-1 RNA, below the standard quantification limits. Recent reports have suggested that low levels of detectable HIV-1 RNA may be a predictor of virological failure, particularly for patients receiving protease inhibitor monotherapy. The aim of this study was to compare the performance of the Roche Amplicor UltraSensitive assay (used in the MONET trial) and the Abbott RealTime assay (used in the PROTEA trial) in plasma samples with HIV-1 RNA ,200 copies/mL. The MONET and PROTEA trials recruited patients with screening plasma HIV-1 RNA ,50 copies/mL; patients were tested at screening then re-tested at the baseline visit, 4–6 weeks later. During both trials, patients remained on stable triple antiretroviral combination treatment between screening and baseline. The Roche Amplicor UltraSensitive assay was used for routine measurement of HIV-1 RNA in the MONET trial. Using this assay, HIV-1 RNA results were classified as: (i) no HIV-1 RNA detected; (ii) HIV-1 RNA ,50 copies/mL, but virus detected under the quantification limit; or (iii) HIV-1 RNA ≥50 copies/mL. The Abbott RealTime assay was used for routine measurement of HIV-1 RNA in the ongoing PROTEA trial. Using the Abbott assay, HIV-1 RNA results were classified as: (i) no HIV-1 RNA detected; (ii) HIV-1 RNA ,40 copies/mL, but virus detected under the quantification limit; or (iii) HIV-1 RNA ≥40 copies/mL. During the randomized phase of both trials, samples with lowlevel elevations in HIV-1 RNA (50–100 copies/mL in MONET and 40 –200 copies/mL in PROTEA) were re-tested using the same assay. In addition, 122 samples with HIV-1 RNA ,200 copies/ mL by the Roche Amplicor UltraSensitive assay were re-tested with the Abbott RealTime assay. In the MONET trial, there were 253 patients with both screening and baseline HIV-1 RNA data available using the Roche Amplicor UltraSensitive assay. Of the 247 patients who had HIV-1 RNA ,50 copies/mL at screening, 237 (96%) had HIV-1 RNA ,50 copies/mL at the baseline visit. There were six patients with HIV-1 RNA initially ≥50 copies/mL at screening (initially screening failures), of whom 3 (50%) had HIV-1 RNA ,50 copies/mL at baseline. Between the screening and baseline visits, there was substantial fluctuation between the categories of ‘HIV-1 RNA not detected’ and ‘HIV-1 RNA ,50 copies/mL, detected’ (Table 1). Of the 212 patients with HIV-1 RNA not detected at screening, 27 (13%) had HIV-1 RNA detected under the quantification limit at baseline, while 7 (3%) had HIV-1 RNA ≥50 copies/mL at baseline. Of the 253 patients tested overall, 188 (74%) had HIV-1 RNA in the same category at both screening and baseline. Given the fluctuations in HIV-1 RNA between the categories of ‘HIV-1 RNA not detected’ and ‘HIV-1 RNA , 50 copies/mL, detected’, individual patient results in these two categories need to be interpreted with caution. During the MONET trial, 36 samples with HIV-1 RNA initially between 50 and 100 copies/mL were re-tested with the same Amplicor UltraSensitive assay; the repeat sample showed HIV-1 RNA ,50 copies/mL in 25 patients (69%). Of the 11 quantifiable re-test results, all showed HIV-1 RNA levels ,200 copies/mL. In the PROTEA trial, there were 299 patients with consecutive HIV-1 RNA measurements taken at the screening and baseline visits. Of the 292 patients who had HIV-1 RNA ,40 copies/mL at screening using Abbott RealTime, 289 (99%) had HIV-1 RNA ,40 copies/mL at baseline. There were seven patients with HIV-1 RNA ≥40; all seven patients had HIV-1 RNA ,40 copies/mL at the baseline visit. Between the screening and baseline visits, there was also substantial fluctuation between the HIV-1 RNA categories (Table 2). Of the 246 patients who had HIV-1 RNA not detected at screening, there were 30 patients (12%) with HIV-1 RNA detected, but ,40 copies/mL, at baseline, and 1 patient (0.4%) with HIV-1 RNA ≥40 copies/mL at baseline. Of the 299 patients tested overall, 226 (76%) had HIV-1 RNA in the