Sequence diversity in the glycoprotein B gene complicates real-time PCR assays for detection and quantification of cytomegalovirus.

Real-time quantitative PCR systems (Q-PCR) for the rapid detection and quantification of microorganisms in clinical specimens employ oligodeoxyribonucleotide primers and probes for specificity, which makes them vulnerable to false negatives caused by sequence diversity in the template. Schaade et al. (J. Clin. Microbiol. 39:3809, 2001) reported a sequence variant (C630T) in the cytomegalovirus (CMV) glycoprotein B (gB) gene that, although detectable in their Q-PCR assay, could not be accurately quantified. In an effort to evaluate the impact of CMV sequence variants in our patient population by use of a similar Q-PCR assay, we surveyed 54 isolates of CMV, each from a different patient. We detected evidence for the C630T variant in 4 of 54 (7.4%) patients. Furthermore, isolates from two additional patients were completely negative in the test. Sequencing of these false-negative isolates revealed multiple mutations within the probe hybridization sites. A Q-PCR that targeted the CMV polymerase gene instead of gB detected all 54 isolates. We suggest that Q-PCR assays for viral load be rigorously tested on large panels of viral isolates to assess the impact of sequence diversity on detection as well as quantification.