Immunohistochemical demonstration of oestrogen and progesterone receptors: correlation of standards achieved on in house tumours with that achieved on external quality assessment material in over 150 laboratories from 26 countries

Abstract

Aims: To investigate the sensitivity of immunohistochemical (IHC) assays for oestrogen receptors (ER) and progesterone receptors (PR) achieved by laboratories on breast tumours fixed and processed in their own department, and to compare this with the degree of sensitivity they achieve on tumours circulated as part of an external quality assessment (EQA) programme.

Methods: On 10 occasions between April 1994 and June 1998, histological sections from breast cancers showing various degrees of expression of ER and PR were circulated for IHC staining to laboratories participating in the UK national external quality assessment scheme for immunocytochemistry (UK NEQAS-ICC). The staining of these tumours, in addition to that of tumours fixed and processed in the participants own laboratories (in house tumours), was assessed by a panel of four assessors, using the established UK NEQAS-ICC scoring system. For a selected assessment run, the degree of expression of participants in house tumours was evaluated by means of the semiquantitative quick score method.

Results: Although the scores awarded for the staining of in house tumours were generally higher than those awarded for the staining of UK NEQAS tumours, there was also a significant positive correlation between the two sets of scores. Using the quick score method of evaluation for one of the assessment runs, 47% of in house tumours were classified as having a high degree of ER expression. Of the remaining cases, a significant proportion initially classified as having only low or medium expression of ER were found to have higher expression when stained by the organising laboratory. The UK NEQAS-ICC centre's routine assay for hormonal receptors was found to be 90-100% efficient in achieving optimal demonstration of breast tumours from over 150 different laboratories.

Conclusions: The significant positive correlation between the results obtained on the UK NEQAS tumours and the in house tumours provides evidence for the view that results achieved on EQA material are accurate indicators of in house laboratory performance. Although most laboratories adequately detected tumours with high receptor expression, a large proportion of in house tumours classified initially by participants' staining as being of low or medium ER expression had a higher degree of expression when stained by the UK NEQAS-ICC centre. The efficiency of the organising centre's routine IHC method for ER and PR in optimally demonstrating participants in house breast tumours shows that variations in fixation and tissue preparation are not limiting factors preventing a different laboratory achieving optimal demonstration.

Figure 1 Box plot to show the relation between the UK NEQAS scores achieved by participants on the slides circulated by UK NEQAS (unshaded boxes labelled with run number and the letter "E") and the scores achieved by the same participants on their own in house control slides (shaded boxes labelled with the run number and the letter "F") between April 1994 and June 1998. The maximum score attainable was 20 and the minimum score attainable was 4. The bold line across each box indicates the median score. N, number of laboratories participating in each assessment run.

Figure 2 Results of immunohistochemistry for oestrogen receptors (ER) performed by the UK NEQAS organising laboratory on the low expressing (cytosol assay ER, 10 fmol/mg protein), ER positive, infiltrating ductal carcinoma circulated by UK NEQAS-ICC for assessment run 41.

Figure 3 High power detail of the same section shown in fig 2.

Figure 4 Results of immunohistochemistry for oestrogen receptors (ER) performed by laboratory "X" on the low expressing infiltrating ductal carcinoma shown in figs 2 and 3. The UK NEQAS score awarded to laboratory "X" for this staining was 8 out of 20. Laboratory "X" considered this tumour to be ER negative.

Figure 5 High power detail of the same section shown in fig 4.

Figure 6 Results of immunohistochemistry for oestrogen receptors (ER) performed by laboratory "X" on the high expressing in house tumour submitted by laboratory "X" for run 41 (UK NEQAS score, 16 out of 20, quick score, 4).

Figure 7 Results of immunohistochemistry for oestrogen receptors (ER) performed by the UK NEQAS organising laboratory on the tumour of laboratory "X" shown in fig 6 (UK NEQAS score, 20 out of 20, quick score, 7).

Figure 8 Results of immunohistochemistry for oestrogen receptors (ER) performed by laboratory "Y" on the high expressing in house tumour submitted by laboratory "Y" (UK NEQAS score, 13 out of 20, quick score, 3). Laboratory "Y" scored 6 out of 20 on the low expressing infiltrating ductal carcinoma shown in figs 2–5 and considered this tumour to be ER negative.

Figure 9 Results of immunohistochemistry (ICH) for oestrogen receptors (ER) performed by the UK NEQAS organising laboratory on the tumour of laboratory "Y" shown in fig 8 (UK NEQAS score, 20 out of 20, quick score, 7).

Figure 10 Box plot to compare the scores achieved by participants on their own in house breast tumours and the scores awarded for the UK NEQAS organising laboratory's immunostaining of duplicate sections of the same tumours. The plot shows seven assessment runs between February 1995 and April 1998 for which the UK NEQAS-ICC organising laboratory stained participants' in house slides. Run numbers are labelled F29 –F42. The boxes labelled "P" refer to the participants' scores, whereas the boxes labelled "N" refer to the scores awarded to the UK NEQAS organising laboratory.