Specimen radiography as predictor of resection margin status in non-palpable breast lesions.

AIM: This study aimed to evaluate the role of specimen radiography in predicting margin status for non-palpable breast malignancies. METHODS: We retrospectively reviewed the clinical and pathological data together with specimen radiographs of 164 women with ductal carcinoma in situ, who were referred to our centre between January 1997 and December 2000. In all cases microcalcifications were discovered on mammography. Lesions were localized preoperatively using a guide-wire. Specimen radiography findings and clinicopathological data were correlated with pathological findings. RESULTS: Findings comprised 122 pure ductal carcinomas in situ (74%) and 42 mixed carcinomas, both infiltrating and in situ (26%). On the specimen radiographs, the lesions were close (<1mm) to one edge of the lumpectomy in 34 (21%) cases. Histologically, there were 103 positive resection margins (<1mm, 63%) and only 61 negative margins (> or =1mm, 37%). On univariate analysis, factors associated with positive resection margins were found to be distance from microcalcifications to edge of lesion on specimen radiographs, and radiological multifocality. On multivariate analysis (logistic regression), a radiological margin <5mm and multifocality were the only risk factors for close histological margins. Radiological margins were not associated with surgical findings. CONCLUSION: Our results demonstrate that there is a correlation between specimen radiographs and histological results. The clinical relevance of this should be evaluated in a prospective study.

Quality of life after sentinel lymph node biopsy in early breast cancer.

AIM: This study assessed the effects of multiple therapeutic factors on quality of life (QOL) in the treatment of breast cancer. METHODS: We surveyed 179 recurrence-free women with early breast cancer who had undergone a sentinel lymph node procedure, between January 1999 and June 2001. Age, tumour size, breast and axillary procedure, nodal status, chemotherapy, supra-clavicular fossa radiotherapy, and hormone therapy were tested as possible factors associated with poor QOL. RESULTS: Information on QOL was obtained for 148 out of 179 patients. Age less than 55 years and chemotherapy were factors associated with impairment of physical well-being. Tumour size was associated with poor socio-familial well-being. Factors associated with altered arm subscale scores were age <55, axillary procedure, nodal status, chemotherapy and supra-clavicular fossa radiotherapy. Unexpectedly, sentinel lymph node (SLN) procedure delayed the onset of chemotherapy if the metastatic status of SLN was not diagnosed intra-operatively. CONCLUSION: Efforts are needed to improve the QOL of young patients. Axillary procedure affects only QOL related to arm morbidity.

Late local recurrences in a randomised trial comparing conservative treatment with total mastectomy in early breast cancer patients.

BACKGROUND: A randomised trial was conducted comparing wide lumpectomy and breast irradiation with modified radical mastectomy. As the follow-up was long (mean duration 22 years), we analysed the variation in the effect of treatment over time. PATIENTS AND METHODS: The trial included 179 patients with a breast cancer measuring

Prognostic factors in primary breast sarcomas: a series of patients with long-term follow-up.

PURPOSE: To describe the pathologic characteristics and prognostic factors of primary breast sarcomas (PBSs). PATIENTS AND METHODS: We reviewed the clinical records and pathologic slides of 83 women with PBS treated in our institution between 1954 and 1991, with a median follow-up of 7.8 years. The majority of patients had undergone surgical treatment. RESULTS: The main histologic type was malignant fibrohistiocytoma (n = 57). For the whole population, the 10-year overall survival (OS) and disease-free survival (DFS) rates were 62% and 50%, respectively. For Fédération Nationale des Centres de Lutte Contre le Cancer grade 1, 2, and 3 tumors, the 10-year OS and DFS rates were 82% and 61%, 62% and 51%, and 36% and 25%, respectively (P =.00007 and.004, respectively). For tumors measuring less than 5 cm, 5 to 10 cm, and more than 10 cm, the 10-year OS and DFS rates were 76% and 66%, 68% and 55%, and 28% and 15%, respectively (P =.002 and.009, respectively). In the multivariate analysis, the tumor size and histologic grade were correlated with the 10-year DFS rate (P =.04 and.01, respectively), but only the histologic grade was correlated with OS (P =.01). Angiosarcoma was the only histologic type significantly associated with a poorer outcome in the multivariate analysis. CONCLUSION: PBSs have the same clinical history and prognostic factors as sarcomas arising at other sites. Therefore, it is legitimate to use a similar treatment strategy for PBS as for other sarcomas.

Sentinel lymphadenectomy without systematic axillary dissection in breast cancer patients: predictors of non-sentinel lymph node metastasis.

AIMS: To identify factors predicting metastatic involvement of non sentinel axillary lymph nodes in breast cancer patients who underwent sentinel lymph node (SLN) biopsy followed by complete axillary dissection only in case of metastatic sentinel lymph node. METHODS: A prospective database including 165 breast cancer patients who underwent SLN biopsy without further complete axillary dissection in case of non-metastatic SLN was reviewed. Primary tumor size, pathologic grade, lymphatic invasion in the primary tumor, estrogen receptor status, tumor size in the SLN and number of metastatic SLNs were tested as possible predictors of metastatic involvement of non-SLN. RESULTS: The sentinel lymph node detection rate was 97% (160/165 patients). The mean number of SLNs per patient was 1.8 (range: 1-5). Fifty patients (31.3%) had a metastatic axillary SLN: 10 of the 42 patients with T1a or T1b breast tumors and 40 of the 118 patients with T1c< or = 15mm tumors. Fifteen of the 50 patients with metastatic SLN had metastatic non-SLN. Primary tumor size, tumor size in the SLN, pathologic grade, estrogen receptor status and age were not significantly associated with metastatic involvement of non-SLN. Number of metastatic SLNs fell short of reaching statistical significance (P: NS). Lymphatic invasion in the primary tumor was the only factor significantly associated with the presence of tumor in the non SLN (P<0.01). CONCLUSION: In our series, only lymphatic invasion in the primary tumor was correlated with metastases detection in the non-SLN. We could not identify a subset of patients without metastatic non-SLN in patients with metastatic SLN.

Predictive factors for local recurrence in 2006 patients with surgically resected small breast cancer.

BACKGROUND: Analyses of predictive factors for local recurrences are important, as an increasing number of patients with early breast cancer opt for a breast-conserving procedure. This study investigates whether factors predictive of local recurrence differ between patients treated with conservative or radical surgery. PATIENTS AND METHODS: Two thousands and six patients with invasive breast carcinoma (< or =25 mm) were included. Of these patients, 717 were treated conservatively (lumpectomy and breast irradiation) and 1289 were treated with total mastectomy. All patients had axillary dissection and received lymph node irradiation if axillary nodes were positive. Most patients did not receive adjuvant chemotherapy or additive hormonal treatments. The mean duration of follow-up was 20 years. The main end point was the total local recurrence rate. The risk factors of local recurrence were estimated by multivariate analyses and interaction tests were used for intergroup comparisons. RESULTS: Statistically significant predictive factors for mastectomized patients were histological grade, extensive axillary node involvement (10 nodes or more), and inner quadrant tumors, which were of borderline significance. Young age, however, was not a prognostic indicator for local recurrence. The main statistically significant factor for patients treated with a conservative approach was young age (< or =40 years). These younger patients had a five-fold increased risk of developing a breast recurrence compared with patients older than 60 years. CONCLUSIONS: Younger patients with early breast cancer treated with breast-conserving surgery should in particular be followed up at regular intervals so that any sign of local failure can be diagnosed early.

[Sentinel node biopsy in breast cancer. Will this lead to the end of lymphadenectomy for small tumors without suspected axillary adenopathy?]. / Le ganglion sentinelle dans le cancer du sein. Est-ce la fin du curage pour les petites tumeurs NO?

Sentinel node (SN) biopsy in breast cancer is still in a crucial stage of evaluation. Many teams have obtained excellent results using this method, with a detection rate always higher than 90% and a false negative rate between 0 and 8%, in prospective series. The main question is to know if lymphadenectomy can now be avoided when the SN is negative. The answer will come from the results of the two ongoing trials comparing sentinel node biopsy to axillary lymphadenectomy. But their results will be available only in two or three years. However, many teams, as at Institute Gustave Roussy, are now applying the technique routinely, because of the excellent results obtained during their learning curve. But there are some methodological differences between teams, which can influence the detection and false negative rates. Thus, several methodological standards remain to be defined. This review enable us to clarify a certain number of questions. Today, SN biopsy can only be performed by trained teams, with prospective evaluation of their results or participation in phase III trials.

[Basal cell carcinoma of the vulva: 21 cases]. / Carcinome basocellulaire de la vulve: 21 cas.

INTRODUCTION: Basal cell carcinoma (BCC) accounts for 2 to 3 p. 100 of all vulvar malignancies. PATIENTS AND METHODS: We report a retrospective study of 21 cases treated from 1937 to 1999. RESULTS: Vulvar BCC's occurred in elderly patients (average age: 66 years), with mean delay to diagnosis of 5.5 years. Five patients were referred for recurrence. Lesions were located on the external hairy side of the labia majus, except one located on the internal side. In 4 cases a preexisting risk factor was identified: 2 patients had previously received radiation therapy and 2 other patients had multiple disseminated BCC. Mean BCC diameter was 2 cm. Pathological data were similar to skin BCC, with one case of mixed tumor (BCC and squamous cell carcinoma). The treatment was surgical excision for 19 BCCs. Local recurrence risk was high. Only one patient died of visceral dissemination of the disease. DISCUSSION: More than 250 cases of vulvar BCC have been reported in the literature. Clinical, pathological and follow up data are similar to results in the present series. Treatment of choice consists of surgical excision with tumor-free margins. Because of local recurrence risk and possible association with other primary cancers in this age group, long term follow-up is necessary.

Loss of heterozygosity, allele silencing and decreased expression of p73 gene in breast cancers: prevalence of alterations in inflammatory breast cancers.

The p73 gene is a p53 homologue located at 1p36-33, a region submitted to deletions in breast cancer (BC) and putatively imprinted. To study whether p73 was associated with breast carcinogenesis, loss of heterozygosity (LOH), allele expression and transcript levels were assessed in 59 BC, including 39 BC presenting no inflammatory symptoms (NBC) and 20 inflammatory BC (IBC). IBC is a rare but aggressive form of cancer with a very poor prognosis. Normal breast epithelium (BE) and lymphocytes from patients were used as controls. StyI polymorphism generating GC and/or AT alleles was used to select 22 heterozygous patients. p73 LOH was significantly higher in IBC than in NBC [five of eight cases (62%) versus two of 14 cases (14%); Fisher's exact test, P=0.05]. p73 was biallelically expressed in all BE. In contrast, 12 of 16 (75%) BC were monoallelically expressed, showing that allele silencing was significantly associated with breast carcinogenesis (P=0.012), AT being the preferential silent allele (10 out of 12 tumours). p73 mRNA levels in NBC and IBC were two- and threefold lower than in BE, respectively, suggesting that decreased expression could be related to tumour aggressiveness. In conclusion, LOH, allele silencing and decreased expression of the p73 gene may play a role in breast carcinogenesis.

European guidelines for quality assurance in the surgical management of mammographically detected lesions. European Breast Cancer Working Group.

The European Guidelines developed for mammography screening have contributed to the general discussion on quality assurance and the important tasks of the health professionals dealing with breast cancer screening. The cooperation of each medical discipline is of utmost importance in order to achieve optimal results and eventually a mortality reduction. The following guidelines are based on the British NHS quality assurance guidelines for surgeons in breast cancer screening and modified to meet the different needs in the European Countries. The term "surgeon" denotes a medical doctor trained and involved in the surgical treatment of breast diseases. The members of the working group who participated in order to adapt these guidelines are listed above. We hope that this document will contribute towards a more comprehensive approach of breast cancer screening-detected lesions throughout Europe.

[Negative second-look laparatomy in ovarian cancers. Survival analysis and prognostic factors from 64 cases treated at the Gustave Roussy Institute]. / Seconde laparotomie négative dans les cancers de l'ovaire. Analyse de la survie et des facteurs pronostiques à partir de 64 cas traités a l'Institut Gustave-Roussy.

This study describes 64 cases of ovarian adenocarcinoma seen in the Gustave-Roussy Institut between 1978 and 1988 and who had a negative second-look laparotomy. The median age was 51 years (30-74). FIGO stages were: I: 7 (11%); II: 3 (5%); III: 39 (61%); IV: 3 (5%); and undetermined: 12 (19%). There were 53% of serous type, 14% of endometrioid type, 13% of undifferencied type, 8% of clear cells type, 3% of mucinous type, and 9% of mixed type tumors. There were 50% of grade 3 tumors. Initial debulking surgery was as complete as possible in 59 patients, with a residual tumor after surgery superior or equal to 2 cm in 25 patients. Post second-look surgery treatment (n = 57) consisted of chemotherapy (CT) alone in 22 patients (34%), radiotherapy (RT) alone in 31 patients (49%), and CT associated with RT in 4 patients (6%). Median follow-up is 100 months. The overall survival rates at 3 and 5 years were respectively 86 and 81%, and disease free survival rates 70 and 61%. Among the 64 patients, 26 relapsed (39%). Median time to relapse was 96 months. There is a statistical difference in the survival between patients who had no or inferior to 2 cm residual tumor and others. Residual tumor was the only factor to be significant in univariate and multivariate analysis of survival.

Conservative treatment versus mastectomy in early breast cancer: patterns of failure with 15 years of follow-up data. Institut Gustave-Roussy Breast Cancer Group.

PURPOSES: A randomized trial was conducted to compare tumorectomy and breast irradiation with modified radical mastectomy. We have analyzed the patterns of failure in each arm of the trial and the prognostic factors that have an independent effect on treatment failures and overall survival. PATIENTS AND METHODS: The trial included 179 patients with breast cancer of up to 20 mm in diameter at macroscopic examination. Eighty-eight patients had conservative management and 91 a mastectomy. All patients had axillary dissection with frozen-section examination. For patients with positive axillary nodes (N+), a second randomization was performed: lymph node irradiation versus no further regional treatment. Patterns of failure were determined by a competing-risk approach and multivariate analysis. A prognostic-score was determined by multivariate analysis. RESULTS: Overall survival, distant metastasis, contralateral breast cancer, new primary malignancy, and locoregional recurrence rates were not significantly different between the two surgical groups, or between lymph node irradiation groups. Most recurrences appeared during the first 10 years. Three distinct prognostic groups were determined taking into account age, tumor size, histologic grading, and number of positive axillary nodes. CONCLUSION: Long-term results support conservative treatment with limited surgery and systematic breast irradiation as a safe procedure for the management of small breast cancers. Four easily obtainable clinical and histologic factors may be combined in a prognostic score that is highly predictive of overall and event-free survival.

Ductal carcinoma in situ: MR imaging-histopathologic correlation.

PURPOSE: To correlate histopathologic and magnetic resonance (MR) imaging findings of ductal carcinoma in situ (DCIS). MATERIALS AND METHODS: Thirty-six women with DCIS underwent preoperative contrast material-enhanced subtraction dynamic MR imaging. Concomitant early contrast enhancement in the breast parenchyma with normal vessels was considered a positive finding. The size and shape of early enhancement were correlated with the size and density packing of ducts involved by DCIS. Tumor angiogenesis in the stroma that surrounded the ducts was evaluated with immunoperoxidase staining. RESULTS: Early contrast enhancement was demonstrated in 34 patients with DCIS but not in two patients with comedo-type DCIS. Tumor angiogenesis was demonstrated in the stroma. The size and morphology of contrast-enhanced lesions significantly correlated with the size (P = .0085) and density packing of ducts involved by DCIS (P = .012). CONCLUSION: Contrast enhancement on dynamic MR images of DCIS may be due to the presence of tumor angiogenesis in the stroma.

c-erbB2 gene amplification and protein expression in ovarian epithelial tumors: evaluation of their respective prognostic significance by multivariate analysis.

c-erbB2 gene amplification or over-expression has been reported in ovarian cancer, but their prognostic value remains conflicting. To investigate the respective prognostic significance of c-erbB2 gene amplification and protein over-expression, tumor samples were obtained from 65 patients with ovarian adenocarcinoma (9 FIGO stage I, 7 stage II, 38 stage III and II stage IV) followed up for a median period of 71 months. c-erbB2 gene amplification (> or = 2.5 a.u.) was detected in 9/65 (14%) adenocarcinomas and in none of 5 benign and 8 borderline ovarian epithelial tumors also analyzed. Specimens from 52 of the 65 adenocarcinomas were available for immunohistochemical analysis. c-erbB2 protein expression was observed in 23/52 (44%) adenocarcinomas. No correlation was found between c-erbB2 gene copy number and protein expression. There was no correlation of c-erbB2 gene copy number or protein expression with any of the clinico-pathological factors analyzed (i.e., FIGO stage, histological type, histological grade and residual tumor). On univariate analysis, c-erbB2 gene amplification was associated with poorer survival (p = 0.04). However, in the multivariate analysis of clinico-pathological factors and c-erbB2 gene copy number, c-erbB2 gene amplification did not retain any independent prognostic significance (p = 0.19). No significant survival difference was found between patients with and without c-erbB2 protein over-expression in univariate or multivariate analyses. Therefore, neither c-erbB2 gene amplification nor c-erbB2 protein over-expression appears to be a significant prognostic marker in patients with ovarian carcinoma.

[Diagnosis of local breast cancer recurrences by magnetic resonance imaging]. / Diagnostic des récidives locales du cancer du sein par imagerie par résonance magnétique.

The aim of this study is to determine the diagnostic value of magnetic resonance imaging in the diagnosis of local recurrence of breast cancer. From 1991 to 1994, 61 women were studied prospectively using magnetic resonance imaging. All examinations were made on a 1.5 Testa machine with T1 weighted images and after gadolium-dota injection and dynamic images (T1 weighted sequences every 47 seconds during injection of a gadolinium-dota bolus). All the pre-injection, images in the dynamic series were subtracted from the images after injection. A surgical biopsy was obtained in 39 patients yielding a diagnosis of local recurrence (n = 28) or a benign lesion (n = 11). Among the 28 local recurrences, pathology examination reported invasive cancer in 22 and intra-ductal carcinoma in 6. In 22 patients with normal magnetic resonance imaging, follow-up examinations were performed every 6 months. There were no local recurrences within a delay of 6 to 36 months. Twenty-six of the 28 patients with a local recurrence, cystosteatonecrosis and surgery scar tissue less than 6 months old showed contrast uptake 1 min and 34 s after gadolinium injection during the dynamic sequence. This product uptake yielded nodular images within the invasive carcinomas and linear images in the intraductal cancers. In all cases, it is easier to visualize this contrast uptake in subtraction images. In conclusion, magnetic resonance imaging is a simple reliable method for the diagnosis of local recurrence of breast cancer.

[Contribution of expert systems in clinical practice: apropos of the Penelope experience, an expert system in assisted diagnosis and therapy of ovarian adenocarcinoma]. / Apport des systèmes experts dans la pratique clinique: à propos de l'expérience Pénélope, système expert d'aide au diagnostic et à la thérapeutique des adénocarcinomes de l'ovaire.

An expert system (ES) for Diagnosis and Therapy of ovarian adenocarcinoma has been developed at the Institut Gustave-Roussy. From surgical and histological results, clinical examination and additional investigative reports, the system presents a synthesis and then determines the stage of the disease. The system than proposes therapeutic indications adapted to the characteristics of the illness and of the patient, and edits a report at the end of the ES consultation. This experience allowed us to specify the field of ES applications in oncology. As tools for diagnosis and therapy, they cannot act as a substitute for the know-how of the physician, as too many medical decisions remain difficult to formalize in the ES. On the other hand the use of artificial intelligence techniques appears to be useful for establishing coherent data bases, which are necessary pre-requisites for clinical research in oncology. The integration of the system in the Hospital Information System is the guarantee of its use in current clinical practice.