Comparison of commonly used real-time PCR methods to detect and quantify HIV1 proviral DNA in infected patients

Abstract number: P542

Rozera G., Abbate I., Bruselles A., Bartolini B., D'Offizi G., Nicastri E., Tommasi C., Capobianchi M.

Objectives: Different PCR methods have been widely used to quantify cell associated HIV-1 DNA. The clinical use of HIV-1 DNA may be limited by the lack of an international standard to calibrate the different methods. In addition, genetic variability of HIV-1 subtypes may profoundly influence the accuracy of the assays. The aim of the study was to evaluate three non commercial commonly used Real-time PCR to quantify proviral HIV DNA in lymphomonocytes from infected patients.

Methods: Parallel evaluation with 3 assays to quantify HIV-1 DNA targeting pol, gag and LTR regions was performed in 93 lymphomonocytes from 43 patients. All the patients harboured HIV-1 B subtype and were in the chronic phase. Twenty-three patients were successfully treated with HAART and underwent CD4-guided treatment interruption; the remaining 20 patients included subjects with virological failure because of multiple-drugs resistant HIV strains. An additional real-time PCR targeting a cellular gene (hTERT) was done to refer the HIV-1 DNA copies to 1 million lymphomonocytes. Full-length sequencing of the virus present in a representative patient was performed by massive parallel pyrosequencing (GS-FLX platform, Roche).

Results: A statistically significant difference between the mean values of HIV-1 DNA in clinical samples obtained by pol, gag and LTR real-time PCR was observed. Only the LTR-targeting PCR was able to detect HIV-1 DNA in all the samples from all the patients, with a number of copies always at least 1 Log higher than that obtained by the two other methods. A stronger correlation between viraemia (HIV-1 RNA) and proviral load was found when HIV-1 DNA was measured with LTR real-time PCR as compared to pol real-time PCR, while the correlation was not significant with gag real-time PCR. Full-length sequencing of the virus, present in a representative patient, clearly showed the simultaneous presence of multiple mismatches in the primers and probes of pol real-time PCR. Conversely, only 1 mismatch was observed in one of the primers of gag real-time PCR.

Conclusions: The data strongly suggest that HIV-1 DNA quantification may be very different, depending on the genome target region. The possible use of HIV-1 DNA as a marker to predict disease progression and treatment outcome in infected patients depends on standardisation of the laboratory methods to minimise variability in genome recognition by the different molecular tools.