Modeling Canadian quality control test program for steroid hormone receptors in breast cancer: diagnostic accuracy study

Teresa Pérez, Nikita Makrestsov, John Garatt, Emina Torlakovic, C Blake Gilks, Susan Mallett

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

The Canadian Immunohistochemistry Quality Control program monitors clinical laboratory performance for estrogen receptor and progesterone receptor tests used in breast cancer treatment management in Canada. Current methods assess sensitivity and specificity at each time point, compared with a reference standard. We investigate alternative performance analysis methods to enhance the quality assessment. We used 3 methods of analysis: meta-analysis of sensitivity and specificity of each laboratory across all time points; sensitivity and specificity at each time point for each laboratory; and fitting models for repeated measurements to examine differences between laboratories adjusted by test and time point. Results show 88 laboratories participated in quality control at up to 13 time points using typically 37 to 54 histology samples. In meta-analysis across all time points no laboratories have sensitivity or specificity below 80%. Current methods, presenting sensitivity and specificity separately for each run, result in wide 95% confidence intervals, typically spanning 15% to 30%. Models of a single diagnostic outcome demonstrated that 82% to 100% of laboratories had no difference to reference standard for estrogen receptor and 75% to 100% for progesterone receptor, with the exception of 1 progesterone receptor run. Laboratories with significant differences to reference standard identified with Generalized Estimating Equation modeling also have reduced performance by meta-analysis across all time points. The Canadian Immunohistochemistry Quality Control program has a good design, and with this modeling approach has sufficient precision to measure performance at each time point and allow laboratories with a significantly lower performance to be targeted for advice.This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially. http://creativecommons.org/licenses/by-nc-nd/4.0/.

Original languageEnglish
Pages (from-to)679-687
JournalApplied Immunohistochemistry & Molecular Morphology
Volume24
Issue number10
Early online date18 Jul 2015
DOIs
Publication statusPublished - Nov 2016

Keywords

  • breast cancer
  • estrogen receptor
  • progesterone receptor
  • meta-analysis
  • sensitivity and specificity

Fingerprint

Dive into the research topics of 'Modeling Canadian quality control test program for steroid hormone receptors in breast cancer: diagnostic accuracy study'. Together they form a unique fingerprint.

Cite this