Application of quality control planning methods for the improvement of a quantitative molecular assay

Abstract

Hepatitis B virus (HBV) DNA measurement has an important role in the diagnosis and management of patients with chronic HBV infection. In cases of chronic hepatitis B, clinical decision is based on either the absolute amount of HBV DNA level, or else the relative change in HBV DNA level. To produce high quality and comparable results, assay performance characteristics must be verified and statistical quality control methods must be planned. In this study, systematic and random error values in an assay of plasma HBV DNA were determined. Performance of the method was examined by employing a normalized operational process specifications (OPSpecs) chart. The systematic error at low and high control levels were 0.33 and 0.22 log(IU/mL) respectively. At both levels, the standard deviations (SD) of the assay were 0.17 log(IU/mL). In addition, a single rule of 12.5SD with 2 control measurements was selected as a candidate quality control method. The assay performed well and was acceptable for clinical use. Further improvement may be attained by switching to automated purification methods. In this study, the well-established discipline of statistical quality control was applied to a real-time quantitative PCR. It was concluded that by employing statistical quality control (QC) methods, which utilize long-term controls, critical changes in the measurement system could be detected.