Comparative frequencies of HIV low-level viremia between real-time viral load assays at clinically relevant thresholds

Abstract

Background The introduction of new real-time PCR HIV-1 assays with higher sensitivity and broader dynamic range has resulted in detection of low-level viremia (LLV) (≥50 copies/mL) in some patients who previously had undetectable HIV-1 viral load (VL) (<50 copies/mL) with end-point PCR assays. It is therefore important to compare the performance of end-point and newer real-time PCR assays at medically relevant decision points. Objectives The study compared the results obtained with the end-point COBAS ® AMPLICOR HIV-1 MONITOR Test, v1.5 to those obtained by three real-time PCR assays COBAS ® AmpliPrep/COBAS ® TaqMan ® HIV-1 Test; the COBAS ® AmpliPrep/COBAS ® TaqMan ® HIV-1 Test, v2.0; and the Abbott RealTi me™ HIV-1 test. Study design A total of 391 plasma specimens from HIV-1-infected patients from three US cities were tested with all four assays. The correlation and concordance of results between real-time and end-point PCR assays were calculated. Results There was a consistent and similar proportion (11.8–14.0%) of HIV-1 VL ≥ 50 copies/mL with the three real-time PCR assays for specimens recording <50 copies/mL on the end-point PCR assay. The real-time PCR assays correlated with the end-point PCR assay within generally accepted limits, but consistently quantified higher than the end-point PCR assay between 50 and 200 copies/mL. Discrepancies in results were associated with patient CD4+ cell count and antiviral medication class. Conclusions The clinical interpretation of VL results from real-time PCR assays should take into account their higher sensitivity at the lower quantitation range when assessing patients for disease progression and monitoring response to therapy in HIV-1-infected patients, in line with current treatment guidelines.