The NHS breast screening programme (pathology) EQA: experience in recent years relating to issues involved in individual performance appraisal.

BACKGROUND: The original role of the National Health Service breast screening programme (pathology) external quality assessment (EQA) scheme was educational; it aimed to raise standards, reinforce use of common terminology, and assess the consistency of pathology reporting of breast disease in the UK. AIMS/METHODS: To examine the performance (scores) of pathologists participating in the scheme in recent years. The scheme has evolved to help identify poor performers, reliant upon setting an acceptable cutpoint. Therefore, the effects of different cutpoint strategies were evaluated and implications discussed. RESULTS/CONCLUSIONS: Pathologists who joined the scheme improved over time, particularly those who did less well initially. There was no obvious association between performance and the number of breast cancer cases reported each year. This is not unexpected because the EQA does not measure expertise, but was established to demonstrate a common level of performance (conformity to consensus) for routine cases, rather than the ability to diagnose unusual/difficult cases. A new method of establishing cutpoints using interquartile ranges is proposed. The findings also suggest that EQA can alter a pathologist's practice: those who leave the scheme (for whatever reason) have, on average, marginally lower scores. Consequently, with the cutpoint methodology currently used (which is common to several EQA schemes) there is the potential for the cutpoint to drift upwards. In future, individuals previously deemed competent could subsequently be erroneously labelled as poor performers. Due consideration should be given to this issue with future development of schemes.

Impact of a national external quality assessment scheme for breast pathology in the UK.

BACKGROUND: This article presents the results and observed effects of the UK National Health Service Breast Screening Programme (NHSBSP) external quality assurance scheme in breast histopathology. AIMS/METHODS: The major objectives were to monitor and improve the consistency of diagnoses made by pathologists and the quality of prognostic information in pathology reports. The scheme is based on a twice yearly circulation of 12 cases to over 600 registered participants. The level of agreement was generally measured using kappa statistics. RESULTS: Four main situations were encountered with respect to diagnostic consistency, namely: (1) where consistency is naturally very high-this included diagnosing in situ and invasive carcinomas (and certain distinctive subtypes) and uncomplicated benign lesions; (2) where the level of consistency was low but could be improved by making guidelines more detailed and explicit-this included histological grading; (3) where consistency could be improved but only by changing the system of classification-this included classification of ductal carcinoma in situ; and (4) where no improvement in consistency could be achieved-this included diagnosing atypical hyperplasia and reporting vascular invasion. Size measurements were more consistent for invasive than in situ carcinomas. Even in cases where there is a high level of agreement on tumour size, a few widely outlying measurements were encountered, for which no explanation is readily forthcoming. CONCLUSIONS: These results broadly confirm the robustness of the systems of breast disease diagnosis and classification adopted by the NHSBSP, and also identify areas where improvement or new approaches are required.

Expression of epidermal growth factor receptors on human cervical, ovarian, and vulval carcinomas.

We describe the properties of two monoclonal antibodies produced to a synthetic peptide consisting of residues 985 to 996 from the cytoplasmic domain of the epidermal growth factor (EGF) receptor. We have examined a group of ten human tumors including cervical, ovarian, and vulval carcinomas for expression of EGF receptors by immunohistological staining using one of these antibodies and another monoclonal antibody to the extracellular domain of the molecule. The tumors were examined using a sensitive amplified enzyme system and a less sensitive indirect staining method. There was generally a good correlation in staining intensity with the two monoclonal antibody reagents. Both antibodies showed strong staining of squamous cell carcinomas and usually weak or heterogeneous patterns with the adenocarcinomas. Samples of each tumor were solubilized in detergent and analyzed for the presence of functional EGF receptors by immunoprecipitation and autophosphorylation. Three of the squamous cell tumors gave labeled bands, Mr 170,000, on sodium dodecyl sulfate:polyacrylamide gels. DNA was extracted from seven of the tumors and digested with two restriction endonucleases, and the fragments were analyzed on Southern blots using probes representing the extracellular and cytoplasmic domains of the molecule. The tumor DNA showed no apparent rearrangements or amplifications when compared to the EGF receptor gene in human placental DNA. These results suggest that there is a high level of EGF receptors on some squamous cell tumors.

c-erbB-3 protein expression in ductal carcinoma in situ of the breast.

c-erbB-3, A recently identified member of the type I tyrosine kinase receptor family, has been shown to be overexpressed in invasive ductal carcinoma of breast. In this study, expression of the c-erbB-3 protein was examined in 57 cases of pure ductal carcinoma in situ of the breast (DCIS) by immuno-cytochemical methods. Staining was either absent (17 cases), present at levels equivalent to that found in adjacent normal tissue (20) or greater than in normal tissue (20). In most cases the pattern of staining was cytoplasmic, but in 4 cases with the most intense reaction there was also focal membrane staining. In the same series of cases, c-erbB-2 protein had previously been shown to be overexpressed in 28 of 57 cases, c-erbB-2 overexpression was correlated with normal level of c-erbB-3, and lack of c-erbB-2 expression was correlated with c-erbB-3 overexpression.

Sequence variants of the estrogen receptor (ER) gene found in breast cancer patients with ER negative and progesterone receptor positive tumors.

Thirteen pairs of tumor and blood DNAs from breast cancer patients with estrogen receptor (ER) negative and progesterone receptor (PgR) positive tumors were screened for mutation analysis using SSCP method. Although neither germline nor somatic mutation of the ER gene in this series was detected, we found two types of sequence variants in exon 1 and exon 4, indicating two silent mutations in codon 10 (TCT to TCC) and codon 325 (CCC to CCG), respectively. These variants were recognized as polymorphic sites. Although the frequency of these polymorphic sites was not correlated with hormone receptor status, the variant in codon 325 tended to be seen more frequently in breast cancer patients than in non-cancer control cases (P = 0.057).

Amphiregulin and cripto overexpression in breast-cancer - relationship with prognosis and clinical and molecular-variables.

The expression of amphiregulin (AR) and cripto (CR-1), two recently identified growth factor peptides, has been evaluated in 196 mammary carcinomas, 13 biopsies of normal breast and 17 benign breast lesions using immunohistochemical methods. Strong immunostaining with AR antibody indicating elevated protein expression was seen in more than 90% of cases of mammary carcinoma whereas with the CR-1 antibody strong staining was seen only in about 77% of cases. No significant relationship could be demonstrated between AR and CR-1 overexpression and traditional prognostic factors such as tumour grade, nodal status and steroid hormone receptor status.

Expression of the hypervariable PUM locus in normal and malignant lung: the tumor-associated epitopes are present but masked in normal tissue.

A single highly polymorphic gene locus PUM codes for a family of mucin-type glycoproteins present in human urine. These glycoproteins can be detected after electrophoresis using a group of monoclonal antibodies which show marked tumour specificity on immunohistology and include the HMFG and Ca antibodies (Swallow et al., 1986, 1987). Here we show by electrophoretic analysis of lung specimens and urine samples from nine individuals, that the PUM locus is expressed both in malignant and in normal lung. In contrast immunohistology of frozen sections of normal lung showed very little staining using the same antibodies, occasional reactive type 2 pneumocytes alone staining, whilst the carcinoma material showed strong staining in each case. However, after formalin fixation much more staining was observed in normal lung, all type 1 and 2 pneumocytes being stained. These observations suggest a difference in accessibility of the epitopes in normal and malignant lung, rather than a difference in expression of the PUM gene.

Breast lymphomas: a clinicopathologic review.

Primary lymphoma is an uncommon tumor in the breast. Review of the literature shows two distinct clinicopathologic groups. One, which affects young women, is frequently bilateral, is often associated with pregnancy, and is a Burkitt-type lymphoma. The second group affects older women, is usually B-cell non-Hodgkins-type lymphoma, and presents clinical features identical to carcinoma of the breast. Three recent studies have suggested that up to half of the cases in the latter group belong to the category of lymphomas arising from the mucosa-associated lymphoid tissues. We have identified nine cases of primary lymphoma from the files of Guys Hospital Clinical Oncology Breast Unit in the 16-year period from 1974 to 1990. The clinical features have been reviewed and the tumors have been evaluated both on a morphologic and an immunohistochemical basis, and seven of nine of the cases have been screened for t[14;18] translocation using the polymerase chain reaction. All the tumors occurred in women older than 50 years and who presented with features of mammary carcinoma. One tumor was true histiocytic lymphoma; the remaining eight cases were B-cell lymphomas. Seven of the latter cases were high-grade B-cell lymphomas and one was a true follicular lymphoma. None of our cases showed the features of lymphoma arising in mucosa-associated lymphoid tissue.

Study of interlaboratory reliability and reproducibility of estrogen and progesterone receptor assays in Europe. Documentation of poor reliability and identification of insufficient microwave antigen retrieval time as a major contributory element of unreliable assays.

Immunohistochemical assays for estrogen receptors (ERs) and progesterone receptors (PRs) have not been surveyed for technical validity. In the present study, the reliability of the immunohistochemical assay for ER and PR was evaluated using data from 105 laboratories participating in external quality assessment (EQA) during a 2-year period. Technical variables associated with reliable immunostaining were analyzed. The efficiency of the antigen retrieval step was identified as the single most important contributory factor influencing the overall reproducibility of the assays. Reliable assays were found in 24 (36%) of 66 laboratories participating in continual EQA, including the majority of centers known to have clinically validated results. Inadequate assay sensitivity, with subsequent weak staining, was the main cause of poor and variable results by laboratories using microwave antigen retrieval; too short a heating time was identified as the principal contributory factor. Extension of the heating time resulted in significant improvement regardless of all other variables in the immunohistochemical protocol. Continual participation in EQA is an effective means for identifying and ameliorating variables that influence the reliability of immunohistochemical assays for predictive markers, thereby assisting in technical validation and standardization.

AUA1: new immunocytochemical marker for detecting epithelial cells in body cavity fluids.

Further to detailed genetic and biochemical characterisation of AUA1 as a surface glycoprotein present on epithelial cells, the antibody against AUA1 was used as an immunocytochemical marker of epithelial cells in body cavity fluids in an attempt to improve the diagnosis made on routine staining. AUA1 was initially tested in 144 morphologically clear cut effusions. It was positive in 46 of 52 (88%) carcinomas and negative in 82 of 84 (98%) benign effusions, including technically inadequate or poorly cellular preparations. There were no false positive results. AUA1 was subsequently used more selectively--that is, in 42 of 175 (24%) of morphologically difficult fluids. AUA1 provided essential diagnostic information in 15 of 42 (36%) and confirmed diagnosis in 17 of 42 (40%), thus enabling accurate diagnosis in a further 32 of 42 (76%) of the difficult cases. The total diagnostic accuracy was therefore 94.3%. AUA1 is a reliable immunocytochemical marker for detecting epithelial cells in body fluids. Its use improves diagnostic accuracy of morphological assessment in difficult cases.

Monoclonal antibody to cytokeratin for use in routine histopathology.

CAM 5.2 is a murine monoclonal antibody, raised against the colon carcinoma cell line HT29, which recognises lower molecular weight intracellular cytokeratin proteins within secretory epithelia. Extensive indirect immunohistochemical studies have confirmed that this antibody stains formalin fixed (and freshly frozen) normal and malignant human tissue in a consistent manner. Reliable staining of conventionally processed pathological tissues provides more accurate identification and staging of human malignant epithelial diseases.

Angiogenesis and inflammation in invasive carcinoma of the breast.

AIM: To investigate the relation between angiogenesis and inflammation in invasive carcinoma of the breast. METHODS: Sections from 75 invasive carcinomas of the breast were stained using immunohistochemistry for von Willebrand factor, CD3, CD8, CD45RO, CD45RA, CD20, CD68, and c-erbB-2. Tumour vascularity was assessed by counting vessels in the three most vascular areas, and calculating the average (x 400 magnification, field 0.168 mm2). Each pattern of inflammation was scored semiquantitatively. RESULTS: The main pattern of inflammation was a diffuse infiltrate of macrophages, and to a lesser extent T cells. Perivascular and perilobular clusters of B and T cells were noted at the edge of the carcinomas, but were less prominent than the diffuse inflammation. Diffuse inflammation, particularly macrophages, was associated with high tumour grade, tumour necrosis, large tumour size, and c-erbB-2 expression. Perivascular and perilobular inflammation also increased with tumour grade. Tumour vascularity increased slightly with intensity of diffuse inflammation (Spearman's rank correlation coefficient rs = 0.17, p = 0.08), and was inversely related to perilobular inflammation (rs = -0.23, p = 0.03). CONCLUSIONS: The correlations between inflammation and vascularity were weak in this study (r2 about 0.04) and thus there was no evidence of an important relation. Discrepancies between this and other studies may be resolved by studying expression of angiogenic cytokines and proteolytic enzymes by tumour infiltrating inflammatory cells, and their relation to tumour vascularity.

Reliability of immunohistochemical demonstration of oestrogen receptors in routine practice: interlaboratory variance in the sensitivity of detection and evaluation of scoring systems.

AIMS: To investigate interlaboratory variance in the immunohistochemical (IHC) detection of oestrogen receptors so as to determine the rate of false negatives, which could adversely influence the decision to give adjuvant tamoxifen treatment. METHODS: To ensure that similar results are obtained by different institutions, 200 laboratories from 26 countries have joined the UK national external quality assessment scheme for immunocytochemistry (NEQAS-ICC). Histological sections from breast cancers having low, medium, and high levels of oestrogen receptor expression were sent to each of the laboratories for immunohistochemical staining. The results obtained were evaluated for the sensitivity of detection, first by estimating threshold values of 1% and 10% of stained tumour cells, and second by the Quick score method, by a panel of four assessors judging individual sections independently on a single blind basis. The results were also evaluated using participants' own threshold values. RESULTS: Over 80% of laboratories were able to demonstrate oestrogen receptor positivity on the medium and high expressing tumours, but only 37% of laboratories scored adequately on the low expressing tumour. Approximately one third of laboratories failed to register any positive staining in this tumour, while one third showed only minimal positivity. CONCLUSIONS: There is considerable interlaboratory variability, especially in relation to the detection of breast cancers with low oestrogen receptor positivity, with a false negative rate of between 30% and 60%. This variability appears to be caused by minor differences in methodology that may be rectified by fine adjustment of overall technique.

A monoclonal antibody that detects HLA-D region antigen in routinely fixed, wax embedded sections of normal and neoplastic lymphoid tissues.

We describe the use of a monoclonal antibody (TAL-IB5) to HLA-D region alpha-chains that reacts well with HLA-D positive cells in normal and neoplastic lymphoid tissues fixed in routine fixatives and embedded in paraffin wax in the conventional fashion. This antibody should prove to be useful in routine histological investigations of lymphoid and possibly other neoplasms as well as other non-neoplastic conditions where the immune system plays an important part.

Sentinel lymph node biopsy in impalpable breast cancer.

Sentinel lymph node biopsy has been investigated using combined radioactive colloid and supra vital blue dye in 27 patients with impalpable breast cancers. Sentinel nodes were identified in 25 cases (93%). Seven patients had involved nodes of whom all had a positive sentinel node. Sentinel node biopsy is ideally suited for use in impalpable breast cancers.

The significance of intramammary nodes in primary breast cancer.

The use of sentinel node biopsy in primary breast cancer raises many new controversies with regard to extra-axillary nodes. Three cases with intramammary nodes are discussed in relation to sentinel node biopsy.

Radioguided occult lesion localisation (ROLL) and sentinel node biopsy for impalpable invasive breast cancer.

AIM: The aim of this study was to simplify the technique of ROLL and sentinel node biopsy without compromising tumour excision and sentinel node biopsy. METHODS: Twenty patients with impalpable primary invasive breast carcinoma underwent an injection of 99mTc-nanocolloid mixed with radiographic contrast medium Iohexol into the centre of the lesion under ultrasound or stereotactic guidance pre-operatively. No guidewire localisation was performed. Under general anaesthesia, a periareolar intradermal/subcutaneous injection of patent blue-V dye was performed. The sentinel node was identified by blue-stained lymphatics and node and a hot spot on the gamma probe. Surgical excision of the primary tumour was then carried out using the gamma probe. RESULTS: In eight of 20 cases an immediate re-excision was carried out and on histological assessment, all 20 patients were clear of invasive disease at the margins. In two patients, in situ disease was present at the margins and a further re-excision was therefore performed. The sentinel node was identified in all cases. In all, five of 20 patients were node positive on routine HE staining. In a further two patients, tumour cells were identified by immunohistochemistry with CAM5.2 antibody. Completion axillary clearance in six patients confirmed that the sentinel node was the only positive node. CONCLUSIONS: This modification of the previously described ROLL technique is feasible and safe and does not compromise tumour excision or sentinel node detection.

The classification of ductal carcinoma in situ and its association with biological markers.

One hundred five cases of pure ductal carcinoma in situ (DCIS) seen in the Guys Hospital breast unit between 1975 and 1991 were reviewed and reclassified using a modified histologic classification based on cytological features as well as histological architecture. The expression of p53 protein, cerbB2 protein, progesterone receptor, and a proliferation antigen KiS1, all factors reported to be of prognostic significance in invasive ductal carcinoma, was also evaluated using immunohistochemical methods. The mode of presentation of these cases was noted, and its relationship to biological markers and histologic type was also assessed. Good interobserver agreement was achieved by two independent observers using the modified histologic classification. Strong correlation was seen between histologic pattern and biological markers as well as between the individual markers. Poorly differentiated DCIS was associated with a high proliferation rate, the presence of cerbB2 and p53 protein and the absence of progesterone receptors. Well-differentiated DCIS showed the reverse, and the intermediate group showed an intermediate pattern. Paget's disease of the nipple was only seen in association with poorly differentiated DCIS, but no other significant association was noted between mode of presentation and DCIS type.

Peroxidase labelling immunocytochemistry: a comparison of eleven commercially-available avidin-biotin systems.

Eleven commercially produced avidin-biotin peroxidase labelling systems employed in immunocytochemistry were compared by titrating a monoclonal and a polyclonal antibody onto routinely prepared formalin-fixed paraffin wax sections of tonsil and appendix. The cost per test for each labelling system was also calculated.