BIRC5 (survivin): a pejorative prognostic marker in stage II/III breast cancer with no response to neoadjuvant chemotherapy.

PURPOSE: Neoadjuvant systemic therapy (NAC) is currently used in the treatment of stage II/III breast cancer. Pathological complete response as a surrogate endpoint for clinical outcomes is not completely validated for all subgroups of breast cancers. Therefore, there is a need for reliable predictive tests of the most effective treatment. METHODS: We used a combination of predictive clinical, pathological, and gene expression-based markers of response to NAC in a prospective phase II multicentre randomized clinical trial in breast cancer patients, with a long follow-up (8 years). This study concerned the subpopulation of 188 patients with similar levels of pathological response rates to sequential epirubicin/cyclophosphamide and docetaxel to determine predictive marker of pCR and DFS. We used a set of 45 genes selected from high throughput analysis and a standardized RT-qPCR. We analyzed the predictive markers of pathological complete response (pCR) and DFS in the overall population and DFS the subpopulation of 159 patients with no pCR. RESULTS: In the overall population, combining both clinical and genomic variables, large tumor size, low TFF1, and MYBL2 overexpression were significantly associated with pCR. T4 Stage, lymphovascular invasion, negative PR status, histological type, and high values of CCNB1 were associated with DFS. In the no pCR population, only lymphovascular invasion and high values of BIRC5 were associated with DFS. CONCLUSIONS: We confirm the importance of ER-related and proliferation genes in the prediction of pCR in NAC-treated breast cancer patients. Furthermore, we identified BIRC5 (survivin) as a main pejorative prognostic factor in patients with breast cancers with no pCR. These results also open perspective for predictive markers of new targeted therapies.

Identifying trajectory clusters in breast cancer survivors' supportive care needs, psychosocial difficulties, and resources from the completion of primary treatment to 8 months later.

PURPOSE: This study aimed to chart patterns of simultaneous trajectories over 8 months in breast cancer survivors' (BCS) supportive care needs, psychological distress, social support, and posttraumatic growth. Clusters of BCS among these trajectories were identified and characterized. METHODS: Of 426 BCS study participants, 277 (65%) provided full assessments in the last week of primary cancer treatment and 4 and 8 months later. Latent trajectories were obtained using growth mixture modeling for patients who responded to all scores for at least one time point (n = 348). Then, classification of BCS was performed by hierarchical agglomerative clustering on axes derived from a multiple factor analysis of trajectory assignments. Self-esteem, attachment security, and satisfaction with care were assessed at baseline. RESULTS: Four trajectory clusters were identified, including two BCS subgroups (63%) with low needs and low psychological distress. Two others (37%) exhibited high or increasing needs and concerning levels of psychological distress. These latter clusters were characterized by higher insecure attachment, lower satisfaction with care, and either lower education or younger age, and having undergone chemotherapy. CONCLUSION: More than a third of BCS present unfavorable patterns in supportive care needs over 8 months after primary cancer treatment. Identified psychosocial and cancer care characteristics point to targets for enhanced BCS supportive care.

Histological grade concordance between diagnostic core biopsy and corresponding surgical specimen in HR-positive/HER2-negative breast carcinoma.

BACKGROUND: The identification and validation of suitable predictive and prognostic factors are a challenge to improve the treatment scheme selection. Discordances in histological grade can be established between core biopsy and surgical specimens. This is important in HR-positive/HER2-negative subgroup where histological grade identifies patients at high risk and is a strong determinant for treatment scheme. METHODS: A total of 350 consecutive invasive breast carcinoma biopsies were assessed and compared with surgical specimens in Institut Curie, Paris, France. Clinical, radiological and pathological data were recorded. RESULTS: Histological grade concordance rate in the HR+/HER2- group was 75%. A grade underestimation was mainly due to mitotic index misgrading (23%). Large tumours (P<0.05), premenopausal patients (P=0.005) and non-ultrasound-guided biopsies (P=0.04) were risk factors for misgrading. The highest discordance was found in tumours that required chemotherapy (39%, P<0.05), and it was related to an underestimation of histological grade on core biopsies (94%). CONCLUSIONS: Histological grade in HR+/HER2- group is important to identify patients with poor prognosis and start a systemic therapy. Histological grade discordance was correlated with an underestimation of mitotic index and factors probably associated with intratumor heterogeneity (premenopausal status, tumour size and the type of core biopsy performed). But such discordance did not appear to modify the therapeutic decision, because systemic treatment decision-making also integrates other variables. Determining histological grade in core biopsy can be especially important in HR-positive/HER2-negative subgroup where it identifies patients at high risk and is a strong determinant of the treatment scheme.

Prediction of axillary lymph node status in male breast carcinoma.

BACKGROUND: To evaluate whether predictive factors of axillary lymph node metastasis in female breast cancer (BC) are similar in male BC. PATIENTS AND METHODS: From January 1994 to May 2011, we recorded 80 non-metastatic male BC treated at Institut Curie (IC). We analysed the calibration and discrimination performance of two nomograms [IC, Memorian Sloan-Kettering Cancer Center (MSKCC)] originally designed to predict axillary lymph node metastases in female BC. RESULTS: About 55% and 24% of the tumours were pT1 and pT4, respectively. Nearly 46% demonstrated axillary lymph node metastasis. About 99% were oestrogen receptor positive and 94% HER2 negative. Lymph node status was the only significant prognostic factor of overall survival (P = 0.012). The area under curve (AUC) of IC and MSKCC nomograms were 0.66 (95% CI 0.54-0.79) and 0.64 (95% CI 0.52-0.76), respectively. The calibration of these two models was inadequate. CONCLUSIONS: Multi-variate models designed to predict axillary lymph node metastases for female BC were not effective in our male BC series. Our results may be explained by (i) small sample size (ii) different biological determinants influencing axillary metastasis in male BC compared with female BC.

Cost comparison of axillary sentinel lymph node detection and axillary lymphadenectomy in early breast cancer. A national study based on a prospective multi-institutional series of 985 patients 'on behalf of the Group of Surgeons from the French Unicancer Federation'.

BACKGROUND: Our objective was to assess the global cost of the sentinel lymph node detection [axillary sentinel lymph node detection (ASLND)] compared with standard axillary lymphadenectomy [axillary lymph node dissection (ALND)] for early breast cancer patients. PATIENTS AND METHODS: We conducted a prospective, multi-institutional, observational, cost comparative analysis. Cost calculations were realized with the micro-costing method from the diagnosis until 1 month after the last surgery. RESULTS: Eight hundred and thirty nine patients were included in the ASLND group and 146 in the ALND group. The cost generated for a patient with an ASLND, with one preoperative scintigraphy, a combined method for sentinel node detection, an intraoperative pathological analysis without lymphadenectomy, was lower than the cost generated for a patient with lymphadenectomy [€ 2947 (σ = 580) versus € 3331 (σ = 902); P = 0.0001]. CONCLUSION: ASLND, involving expensive techniques, was finally less expensive than ALND. The length of hospital stay was the cost driver of these procedures. The current observational study points the heterogeneous practices for this validated and largely diffused technique. Several technical choices have an impact on the cost of ASLND, as intraoperative analysis allowing to reduce rehospitalization rate for secondary lymphadenectomy or preoperative scintigraphy, suggesting possible savings on hospital resources.

Establishment and characterisation of a new breast cancer xenograft obtained from a woman carrying a germline BRCA2 mutation.

BACKGROUND: The BRCA2 gene is responsible for a high number of hereditary breast and ovarian cancers, and studies of the BRCA2 biological functions are limited by the lack of models that resemble the patient's tumour features. The aim of this study was to establish and characterise a new human breast carcinoma xenograft obtained from a woman carrying a germline BRCA2 mutation. METHODS: A transplantable xenograft was obtained by grafting a breast cancer sample into nude mice. The biological and genetic profiles of the xenograft were compared with that of the patient's tumour using histology, immunohistochemistry (IHC), BRCA2 sequencing, comparative genomic hybridisation (CGH), and qRT-PCR. Tumour response to standard chemotherapies was evaluated. RESULTS: Histological profile identified the tumour as a basal-like triple-negative breast cancer. Targeted BRCA2 DNA sequencing of the xenograft showed the presence of the mutation previously identified in the carrier. Comparative genomic hybridisation array profiles of the primary tumour and the xenograft revealed a high number of similar genetic alterations. The therapeutic assessment of the xenograft showed sensitivity to anthracyclin-based chemotherapy and resistance to docetaxel. The xenograft was also highly sensitive to radiotherapy and cisplatin-based treatments. CONCLUSIONS: This study describes a new human breast cancer xenograft obtained from a BRCA2-mutated patient. This xenograft provides a new model for the pre-clinical drug development and for the exploration of the drug response biological basis.

Disseminated tumor cells and the risk of locoregional recurrence in nonmetastatic breast cancer.

BACKGROUND: In early breast cancer patients, bone marrow (BM)-disseminated tumor cells (DTCs) were associated with distant metastasis and locoregional recurrence. Our aim was to determine whether BM DTC detection could be related to specific locoregional dissemination of cancer cells, according to radiotherapy volumes. PATIENTS AND METHODS: The relationship between locoregional recurrence-free survival (LRFS) and DTC detection was evaluated according to the various locoregional volumes irradiated after surgery. RESULTS: BM DTCs were detected in 94 of 621 stage I-III breast cancer patients (15%) and were not associated with axillary node status. Eighteen patients (2.9%) experienced locoregional recurrence (median follow-up 56 months), of whom eight (44%) were initially BM DTC positive. BM DTC detection was the only prognostic factor for LRFS [P = 0.0005, odds ratio = 5.2 (2.0-13.1), multivariate analysis]. In BM DTC-positive patients, a longer LRFS was observed in those who were given adjuvant hormone therapy (P = 0.03) and radiotherapy to supraclavicular nodes (SCNs)/internal mammary nodes (IMNs) (P = 0.055) (multivariate analysis; interaction test: P = 0.028). CONCLUSIONS: The presence of DTC in BM may be associated with a different pattern of locoregional cancer cell dissemination and influences LRFS. The possible reseeding of the primary cancer area by DTC could be prevented by systemic hormone therapy but also by SCN/IMN irradiation.

Prognosis of women with stage IV breast cancer depends on detection of circulating tumor cells rather than disseminated tumor cells.

BACKGROUND: At metastatic relapse, detection of circulating tumor cells (CTC) in peripheral blood is predictive of poor survival of breast cancer patients. Detection of disseminated tumor cells (DTC) in bone marrow (BM) is an independent prognostic factor in early breast cancer. We evaluated the prognostic value of DTC detection in the BM of metastatic breast cancer patients. MATERIALS AND METHODS: BM aspirates from 138 patients were screened for DTC with the pancytokeratin mAb A45-B/B3, according to the ISHAGE classification. One hundred and ten patients (80%) were enrolled before first-line treatment. Thirty-seven patients were simultaneously screened for CTC in the blood. RESULTS: DTC detection rate in the BM was 59%. DTC were associated with bone metastasis (P = 0.0001), but not with a poorer overall survival. Adverse significant prognostic factors were hormone receptor negativity (P = 0.0004) and more than one line of chemotherapy (P = 0.002). CTC detection in the subgroup of 37 metastatic patients was associated with shorter survival (P = 0.01). CONCLUSIONS: Detection of CTC but not BM DTC had a prognostic significance in stage IV breast cancer patients. CTC in blood are a more reliable and a less invasive tool to evaluate prognostic and monitor tumor response in this metastatic setting.

Estrogen-Receptor, Progesterone-Receptor and HER2 Status Determination in Invasive Breast Cancer. Concordance between Immuno-Histochemistry and MapQuant™ Microarray Based Assay.

BACKGROUND: Hormone receptor status and HER2 status are of critical interest in determining the prognosis of breast cancer patients. Their status is routinely assessed by immunohistochemistry (IHC). However, it is subject to intra-laboratory and inter-laboratory variability. The aim of our study was to compare the estrogen receptor, progesterone receptor and HER2 status as determined by the MapQuant™ test to the routine immuno-histochemical tests in early stage invasive breast cancer in a large comprehensive cancer center. PATIENTS AND METHODS: We retrospectively studied 163 invasive early-stage breast carcinoma with standard IHC status. The genomic status was determined using the MapQuant™ test providing the genomic grade index. RESULTS: We found only 4 tumours out of 161 (2.5%) with discrepant IHC and genomic results concerning ER status. The concordance rate between the two methods was 97.5% and the Cohen's Kappa coefficient was 0.89. Comparison between the MapQuant™ PR status and the PR IHC status gave more discrepancies. The concordance rate between the two methods was 91.4% and the Cohen's Kappa coefficient was 0.74. The HER2 MapQuant™ test was classified as « undetermined ¼ in 2 out of 163 cases (1.2%). One HER2 IHC-negative tumour was found positive with a high HER2 MapQuant™ genomic score. The concordance rate between the two methods was 99.3% and the Cohen's Kappa coefficient was 0.86. CONCLUSION: Our results show that the MapQuant™ assay, based on mRNA expression assay, provides an objective and quantitative assessment of Estrogen receptor, Progesterone receptor and HER2 status in invasive breast cancer.

GST pi gene is frequently coamplified with INT2 and HSTF1 proto-oncogenes in human breast cancers.

The glutathione S-transferase gene (GST pi) is located on the same chromosome band (11q13) as proto-oncogenes INT2 and HSTF1 which are frequently amplified in breast cancer. Using the Southern blot technique, we looked for the amplification of the GST pi gene in 17 fresh tumors from human mammary carcinoma. The tumors were preselected because either they had an amplification of the INT2 proto-oncogene detected by dot blot, or their karyotypes exhibited or did not exhibit homogeneously staining regions, a cytogenetic character indicating amplification. Coamplification of GST pi, HSTF1 and INT2 was observed in five tumors, and coamplification of GST pi and HSTF1 without amplification of INT2 in another tumor. We also observed coamplification of GST pi, INT2, HSTF1 in the mammary carcinoma cell line MDA/MB134, whereas GST pi alone was amplified in the mammary epithelial cell line HBL100. These results indicate that INT2, HSTF1 and GST pi belong to the same large amplicon. Since GST pi is involved in intracellular detoxication and since chemotherapeutic drugs are among its substrates, it will be of interest to study GST pi gene expression as well as the response to chemotherapy in patients presenting this amplicon.

Breast inflammatory gigantomastia in a context of immune-mediated diseases.

CONTEXT: Localized breast lesions have been described in lupic or diabetic patients. However, the description of breast gigantomastia in women presenting with autoimmune diseases has not been reported. SETTING: The study took place within the Department of Endocrinology and Reproductive Medicine, Necker Hospital, Paris, France. PATIENTS: We describe eight patients with inflammatory gigantomastia, occurring in a context of immune-mediated diseases: myasthenia, chronic arthritis, or thyroiditis. MAIN OUTCOME MEASURES: Together with hormonal, immunological, and breast magnetic resonance imaging (MRI) evaluation, breast histology enabled us to perform immunocytochemical and indirect immunofluorescence studies. Control sera were obtained from patients with (n = 10) and without (n = 7) antinuclear antibodies. RESULTS: Six of the eight patients developed gigantomastia either at puberty or during pregnancy. Neither a hormonal oversecretion nor a specific immunological pattern was observed. All patients except one presented antinuclear antibodies. Histological study revealed a diffuse, stromal hyperplasia and a severe atrophy of the lobules. A rarefaction of adipocytes was also noted, as previously suggested on MRI. There was a perilobular lymphocytic infiltrate made of CD3+ lymphocytes. Study of sera from five of six cases of gigantomastia showed a nuclear immunofluorescence pattern in normal mammary ductal and lobular glandular epithelium, as well as in kidney and intestine epithelial cells. In control sera, a nuclear signal was observed only when antinuclear antibodies were present. CONCLUSIONS: We suggest that breast tissue may be a target tissue in autoimmune diseases, this process being favored by the hormonal milieu. However, the precise mechanism of such association is not individualized. The fact that stromal hyperplasia is the main histological feature justifies the search for the involvement of growth factors in such a process.

Human papillomavirus genotype as a major determinant of the course of cervical cancer.

PURPOSE: To determine whether the prognosis of invasive cancers of the uterine cervix is related to the type of human papillomavirus (HPV) associated with the tumor. PATIENTS AND METHODS: Two hundred ninety-seven patients with invasive cervical cancer were prospectively registered from 1986 to 1994. HPV typing was performed on DNA extracted from frozen tumor specimens by means of Southern blot hybridization (SBH) and polymerase chain reaction (PCR) techniques. The median follow-up was 38 months. RESULTS: HPV sequences were detected in 246 patients (83%): 150 patients had HPV16, 31 patients had HPV18, and 14 patients had one of the intermediate-oncogenic-risk HPV types (HPV31, 33, 35, 52, 58). In 51 patients, HPV type remained undetermined, and in 51 patients, no viral sequences were found. No significant associations were observed between virologic data and tumor stage or node status. The 5-year disease-free survival (DFS) rate was 100% for patients with intermediate-risk HPV-associated tumors, 58% for patients with HPV16-positive tumors, and 38% for patients with HPV18-positive tumors (P = .02). In multivariate analysis, patients with HPV18-associated tumors had a relative risk (RR) of death 2.4 times greater (95% confidence interval [CI], 1.29-4.59) than that for patients with HPV16, and 4.4 times greater (95% CI, 3.48-5.32) than that for patients with a tumor associated with a viral type different from HPV16/18. CONCLUSION: The prognosis for invasive cancers of the uterine cervix is dependent on the oncogenic potential of the associated HPV type. HPV typing may provide a prognostic indicator for individual patients and is of potential use in defining specific therapies against HPV-harboring tumor cells.

Inter-laboratory quality control for hormone-dependent gene expression in human breast tumors using real-time reverse transcription-polymerase chain reaction.

Quantitative reverse transcription-polymerase chain reaction (RT-PCR) used to detect minor changes in specific mRNA concentrations may be associated with poor reproducibility. Stringent quality control is therefore essential at each step of the protocol, including the PCR procedure. We performed inter-laboratory quality control of quantitative PCR between two independent laboratories, using in-house RT-PCR assays on a series of hormone-related target genes in a retrospective consecutive series of 79 breast tumors. Total RNA was reverse transcribed in a single center. Calibration curves were performed for five target genes (estrogen receptor (ER)alpha, ERbeta, progesterone receptor (PR), CYP19 (aromatase) and Ki 67) and for two reference genes (human acidic ribosomal phosphoprotein PO (RPLPO) and TATA box-binding protein (TBP)). Amplification efficiencies of the calibrator were determined for each run and used to calculate mRNA expression. Correlation coefficients were evaluated for each target and each reference gene. A good correlation was observed for all target and reference genes in both centers using their own protocols and kits (P < 0.0001). The correlation coefficients ranged from 0.90 to 0.98 for the various target genes in the two centers. A good correlation was observed between the level of expression of the ERalpha and the PR transcripts (P < 0.001). A weak inverse correlation was observed in both centers between ERalpha and ERbeta levels, but only when TBP was the reference gene. No other correlation was observed with other parameters. Real-time PCR assays allow convenient quantification of target mRNA transcripts and quantification of target-derived nucleic acids in clinical specimens. This study addresses the importance of inter-laboratory quality controls for the use of a panel of real-time PCR assays devoted to clinical samples and protocols and to ensure their appropriate accuracy. This can also facilitate exchanges and multicenter comparison of data.

Increased expression of prolactin receptor gene assessed by quantitative polymerase chain reaction in human breast tumors versus normal breast tissues.

The role of PRL in human breast tumorigenesis is not well understood. One of the limitations is the difficulty of accurately measuring PRL receptors (PRLR) in human tissues. We established a quantitative PCR method (Q-PCR) in T-47D human breast cancer cells and applied it to 29 patients, 25 of whom presented with either cancer or fibroadenoma. Four patients underwent a mammoplasty, and normal epithelial cells were cultured before Q-PCR. In T-47D cells, 31 x 10(6) messenger RNA molecules were detected per microgram of total RNA. In all patients, expression of the PRLR gene was detected, varying from 1500 to 1 x 10(6) molecules/microgram of RNA in normal tissues and from 4500 to 34.7 x 10(6) molecules/microgram of RNA in tumors. PRLR expression was always greater in tumor than in normal contiguous tissue and similar in cultured mammary epithelial cells and normal breast tissues. Estradiol and progesterone receptor-negative tumors expressed low levels of PRLR transcripts, similar to normal breast tissue from menopausal women. Immunocytochemical analysis of PRLR confirmed stronger staining in almost all tumor samples compared with normal tissues. A messenger RNA encoding locally produced human PRL was also identified by RT-PCR in every sample tested. Our results confirm PRLR gene expression in all tissues studied, and moreover, indicate that this expression is increased in human breast tumors vs. normal contiguous tissues.

Prolonged G2 phase of breast cancer cells and chromosome damage.

5-bromodeoxyuridine incorporation was studied in metaphase chromosomes from 24 breast tumour specimens, including 23 adenocarcinomas. In these 23 cases, a slow cell cycle was observed, with a long (8 h) G2 phase. This slowing of the cell cycle, which was poorly related to the degree of polyploidy, was significantly related to the number of chromosome anomalies: the cell cycle was particularly slow when many rearranged chromosome were observed. These in vitro findings during the first cell cycle cannot easily be transposed to the in vivo situation. By analogy with Fanconi anaemia, in which both chromosome lesions and a long G2 phase are detected, a DNA repair defect and/or high DNA mutagenesis might exist in breast cancer cells.

ERBB2 overexpression in breast carcinomas: no positive correlation with complete pathological response to preoperative high-dose anthracycline-based chemotherapy.

The predictive value of ERBB2 amplification/expression to doxorubicin use is controversial. Preoperative chemotherapy, followed by the pathological assessment of tumour response to treatment provide optimal conditions for the evaluation of the predictive value of biological parameters. We report here data on the predictive value of ERBB2 in a series of 54 cases of breast cancer treated by preoperative high-dose anthracycline-based chemotherapy. Our series consisted of 26 women presenting an inflammatory breast cancer (IBC) and of 28 women with poor prognosis primary cancer (PPPC). Patients received a total of four cycles with doxorubicin (75 mg/m(2) for IBC or 70 mg/m(2) for PPPC) and cyclophosphamide (6 g/m(2) for IBC or 1400 mg/m(2) for PPPC), every 21 days. ERBB2 expression was determined by immunohistochemistry (clone CB11) performed on a tumour biopsy taken before chemotherapy. All patients underwent surgery as a second step of treatment, and the tumour response was assessed on pathological specimens. A complete pathological response was observed in 24 of the 54 cases (44%) (95% confidence interval (CI), 31-57). Pathological complete response was positively correlated with high histological grade (P=0. 02) and with the absence of oestrogen (P=0.003) or progesterone (P=0. 02) receptor expression. ERBB2 overexpression was found in 18 of the 54 cases (33%). A complete pathological response was observed in 33% of these cases (6/18). This figure was not significantly different from the 50% rate of complete response observed for tumours with no detectable ERBB2 expression (18/36). In this small series, ERBB2 overexpression was not a significant predictive marker of the pathological response to high-dose doxorubicin-based chemotherapy.

Wide metastatic spreading in infiltrating lobular carcinoma of the breast.

The aim of this study was to determine whether the metastatic potential of breast cancer could be related to phenotypic characteristics of the tumour. Therefore, we compared the metastatic patterns of invasive lobular (ILC) and ductal (IDC) carcinomas. In ILC, we also analysed this pattern according to the histological subtype of the primary and the E-cadherin (EC) expression level. Metastatic ILC cases (n=96) were retrospectively analysed and classified into classical, alveolar, solid, tubulo-lobular, signet ring cells or pleomorphic subtypes. Anatomical distribution of metastases was detailed for every patient and compared with that registered for IDC (n=2749). Immunostaining of EC (HECD1 antibody) was performed in 82 cases. Histologically, 78 of the 96 cases (81%) corresponded to classical ILC. The pleomorphic subtype was observed in 14 cases (15%), a rate that was higher than that expected. Others corresponded to alveolar (2 cases), signet ring cell (1 case) and solid (1 case) subtypes. EC was undetectable in 72/82 cases (88%). The rate of multiple metastases was higher in ILC (25.0%) than in IDC (15.8%) (P=0.016). Metastases were found more frequently in ILC than in IDC in the bone (P=0.02) and/or in various other sites (peritoneum, ovary, digestive tract, skin em leader ) (P<0.001). In ILC, no significant link was found between the localisation(s) of metastases, the histological subtype and the EC status in the primary. In conclusion, in breast carcinomas, the frequency of multiple metastasis was found to be higher in ILC than IDC. This fact may be related to the phenotypic trait of discohesive small cells which characterises ILC. EC loss, observed in most cases of ILC, may result in alterations in cell-cell adhesion and a preferential growth at metastatic sites. A high rate of pleomorphic tumours was observed in the group of metastatic ILC, but the pattern of metastatic site(s) was not related to the histological subtype of the primary.

Causes of inconsistency in diagnosing and classifying intraductal proliferations of the breast. European Commission Working Group on Breast Screening Pathology.

It is now widely recognised that classifying ductal carcinoma in situ (DCIS) of the breast and diagnosing atypical ductal hyperplasia are associated with significant interobserver variation. Two possible reasons for this inconsistency are differences in the interpretation of specified histological features and field selection where morphology is heterogeneous. In order to investigate the relative contribution of these two factors to inconsistent interpretation of intraductal proliferations, histological sections of 32 lesions were sent to 23 European pathologists followed 3 years later by images of small parts of these sections. Kappa statistics for diagnosing hyperplasia of usual type, atypical ductal hyperplasia and ductal carcinoma in situ were 0.54, 0.35 and 0.78 for sections and 0.47, 0.29 and 0.78 for images, respectively, showing that most of the inconsistency is due to differences in morphological interpretation. Improvements can thus be expected only if diagnostic criteria or methodology are changed. In contrast, kappa for classifying DCIS by growth pattern was very low at 0.23 for sections and better at 0.47 for images, reflecting the widely recognised variation in the growth pattern of DCIS. Higher kappa statistics were obtained when any mention of an individual growth pattern was included in that category, thus allowing multiple categories per case; but kappa was still higher for images than sections. Classifying DCIS by nuclear grade gave kappa values of 0.36 for sections and 0.49 for images, indicating that intralesional heterogeneity has hitherto been underestimated as a cause of inconsistency in classifying DCIS by this method. More rigorous assessment of the proportions of the different nuclear grades present could lead to an improvement in consistency.

Pathological work-up of sentinel lymph nodes in breast cancer. Review of current data to be considered for the formulation of guidelines.

Controversies and inconsistencies regarding the pathological work-up of sentinel lymph nodes (SNs) led the European Working Group for Breast Screening Pathology (EWGBSP) to review published data and current evidence that can promote the formulation of European guidelines for the pathological work-up of SNs. After an evaluation of the accuracy of SN biopsy as a staging procedure, the yields of different sectioning methods and the immunohistochemical detection of metastatic cells are reviewed. Currently published data do not allow the significance of micrometastases or isolated tumour cells to be established, but it is suggested that approximately 18% of the cases may be associated with further nodal (non-SN) metastases, i.e. approximately 2% of all patients initially staged by SN biopsy. The methods for the intraoperative and molecular assessment of SNs are also surveyed.

Local control and survival of breast cancer treated by limited surgery followed by irradiation.

Between 1960 and 1978, 324 patients with early breast cancer were treated by lumpectomy with or without axillary dissection followed by radiation therapy. All were followed for a minimum of 5 years. All patients were, retrospectively, classified T1, T2, N0, N1a, in the TNM (U.I.C.C.) Classification. The retrospective analysis of the local-regional patterns of failure revealed that young age (less than or equal to 32 years) and premenopausal status were associated with an increased rate of local failure, whereas tumor size and location showed no influence. No pathological features were associated with an increased risk of local recurrence, whether pathological subtypes, Scarff Bloom and Richardson grading, intraductal associated component, or vascular involvement. The absolute 5 year disease-free survival rate was 87% in patients who recurred and 93% in those who did not. The absolute 10 year disease-free survival rates were 75 and 82%, respectively. Therefore, these results confirm that loco-regional failure does not significantly influence the disease-free survival.