A New Strategy Based On qPCR to Optimize Detection and Quantification of Eight Herpesviruses in Patients Undergoing Allogeneic Hematopoietic Stem Cell Transplantation.

Abstract 3044

Human herpesviruses may cause severe complications after Hematopoietic Stem Cell Transplantation (HSCT) as interstitial pneumonia, encephalitis and post-transplantation lymphoproliferative disease (PTLD). Monitoring these viruses and providing precise, rapid and early diagnosis of related clinical diseases constitute an essential measure to improve outcomes. A prospective survey on the incidence and clinical features of herpesvirus infections after HSCT has not yet been performed in Brazilian patients. Additionally, the impact of most of these infections on the HSCT outcome is still unclear. We sought to develop a test based on real-time polymerase chain reaction (qPCR) to screen and quantify all known human herpesviruses (CMV, EBV, HSV1, HSV2, VZV, HHV6, HHV7 and HHV8) in plasma samples from patients submitted to HSCT. DNA purification from plasma samples has been performed with the QIAamp DNA Blood Mini Kit (manually) or with the QIAamp DNA Blood Mini QIAcube Kit and the QIAcube robot (Qiagen). At least two sets of primers previously described have been tested for each virus for the approach using SYBR Green in order to select for the sets with best efficiency and sensitivity. The sets of primers and TaqMan® probes for the hydrolysis approach have also been previously described. Lambda phage and a commercial internal positive control (IPC, Life Technologies) have been tested as internal controls. The viruses probes were labeled with FAM, while the IPC probe was labeled with VIC. All qPCR reactions have been performed in a 7900HT (Life Technologies). Infected cell cultures and plasma specimens with a known viral load/amplicon copy number have been used as controls. By august 2012, 824 whole blood and plasma were collected from 91 patients. Initially, we tested a screening approach based on three sets of triplex qPCR reactions (including the internal control) using SYBR Green and melting analysis. Although the test showed good linearity across 6 to 7 orders of magnitude in the log scale for most of the targets, the discrimination was poor for low-copy samples (≤ 10³ copies of the target/ reaction) or complex samples (positive for more than one virus). We then chose to optimize a strategy based on the use of hydrolysis probes, the gold standard in molecular pathology. Except for EBV, which has been amplified and detected in a duplex reaction along with the IPC, the other amplicons have been screened and quantified in singleplex reactions. All targets presented efficiencies between 90–100% and linearity ranging from at least 25 to 10⁸ copies per reaction. For most of the viruses the lower limit of detection (LOD) is around 5 copies of target per reaction, representing 250 copies/mL of plasma; HHV6 and VZV detection, with sensitivity around 25 copies per reaction, is under further optimization. No cross-reaction or false positive results were detected and within-run and between-run precision estimates are equal or higher than 95%. A semi-automated workflow, using the QIAcube robot for DNA extraction and the QIAgility for reaction setup is under validation. Accuracy will be assigned by testing commercial controls (Acrometrix® plasma panels and controls). Based on the precision of the test, we predict that it will be possible to use this new test to screen batches of 10 samples in 96-well plates or 46 samples in 384-well plates in singlicates for the eight known herpesviruses with high sensitivity and specificity. Only selected samples will then be submitted to fine quantification in a second round of qPCR reactions including the appropriate standard curve(s). This strategy allows for a fast and comprehensive detection and of the known hepersviruses in post-HSCT patients, while integrating the main advantages of the hydrolysis probe, which are high sensitivity and specificity.