[1995] Feasibility of Levey-Jennings Quality Control Charts for Monitoring Estrogen Receptor (ER) and Progesterone Receptor (PR) Immunohistochemistry Testing

Heather Neufeld, Richard Berendt, Dragana Pilavdzic, Barry Ziola, John Krahn, Carol Cheung, Blake Gilks, John Garratt, Svein Carlsen, Emina Torlakovic. University of Saskatchewan, Saskatoon, SK, Canada; Cross Cancer Institute/University of Alberta, Edmonton, AB, Canada; Jewish General Hospital, McGill University, Montreal, QC, Canada; University Health Network, University of Toronto, Toronto, ON, Canada; Vancouver General Hospital, University of British Columbia, Vanocuver, BC, Canada; Lions Gate Hospital, Vancouver, BC, Canada

Background: Clinical immunohistochemistry (IHC) techniques are not yet fully standardized. Quality Control (QC) chart analysis has traditionally been applied to liquid-based laboratory testing but not to clinical immunohistochemistry (IHC) laboratories. Image analysis (IA) introduced a true quantitative aspect to interpretation of IHC results, which now makes possible, the theoretical application of control charts to clinical immunohistochemistry quality control. In this project, we tested the real world feasibility of applying QC charts in monitoring the daily performance of positive controls in clinical IHC.
Design: IA was used to evaluate both usual human tissue-based and cell line-based positive controls for estrogen receptor (ER) and progesterone receptor (PR). Several series of on-slide tissue and cell-line positive controls were cut onto the same slide, stained for ER and/or PR, digitally scanned (Aperio ScanScope), and then assessed using IA algorithms. The readings of sequential control samples were plotted as Levey-Jennings QC charts.
Results: Levey-Jennings charts consisting of IA-generated H-scores were created for both tissue-based and cell line-based positive controls. Cell line-based positive controls were a better reflection of true protocol variations since they do not incorporate error due to tissue heterogeneity that is an intrinsic component of human tissue-based positive controls.
Conclusions: Levey-Jennings QC charts are feasible for clinical IHC. They are an excellent way to observe trueness, reproducibility, and repeatability of the IHC protocols.
Category: Quality Assurance

Tuesday, March 5, 2013 9:30 AM

Poster Session III # 276, Tuesday Morning


Close Window