European proficiency testing program for molecular detection and quantitation of hepatitis B virus DNA

External quality control of hepatitis B virus (HBV) DNA detection remains a n important issue. This study reports and compares the results obtained fro m two different proficiency panels for both the qualitative and quantitativ e assessment of HBV DNA. The panels were designed by the European Union Qua lity Control Concerted Action, prepared by Boston Biomedica, Inc., and dist ributed in May 1999 (panel 1) and February 2000 (panel 2). Each contained t wo negative samples and six positive samples with 10(3) to 10(7) copies/ml (panel 1) or 10(3) to 2 X 10(6) copies of HBV DNA per ml (panel 2). For pan el 1, 42 laboratories submitted 20 qualitative (all in-house PCRs) and 37 q uantitative (87% commercial assays) data sets. For panel 2, 51 laboratories submitted 25 qualitative (all in-house PCRs) and 47 quantitative (94% comm ercial assays) data sets. Five data sets (8.8%) in panel I and two data set s (2.8%) in panel 2 contained totals of six and two false-positives, respec tively, corresponding to false-positive result rates of 5.3% for panel 1 an d 1.4% for panel 2. The false-negative result rates of 10.5% for panel 1 an d 17.4% for panel 2 were dependent on the detection levels of the assays em ployed as well as panel composition. In the qualitative analysis of all dat a sets, 47.4% (panel 1) and 51.4% (panel 2) had all samples correct. An ade quate or better score (all correct or only the weak-positive sample missed) was obtained with 77.2% of the panel 1 samples and 68.1% of the panel 2 sa mples. In the quantitative analysis, 57.1% (panel 1) and 42.6% (panel 2) of the data sets achieved an adequate or better score (positive results withi n the acceptable range of the geometric mean +/- 0.5 log(10) of all positiv e results). These results demonstrate that while the qualitative performanc e of HBV detection has considerably improved compared to that of a previous ly published HBV proficiency study, the detection levels of many commercial quantitative assays are still too high to allow adequate quantitation of a ll relevant clinical samples.