Epithelial atypia in biopsies performed for microcalcifications. practical considerations about 2,833 serially sectioned surgical biopsies with a long follow-up.

This study analyzes the occurrence of epithelial atypia in 2,833 serially sectioned surgical breast biopsies (SB) performed for microcalcifications (median number of blocks per SB:26) and the occurrence of subsequent cancer after an initial diagnosis of epithelial atypia (median follow-up 160 months). Epithelial atypia (flat epithelial atypia, atypical ductal hyperplasia, and lobular neoplasia) were found in 971 SB, with and without a concomitant cancer in 301 (31%) and 670 (69%) SB, respectively. Thus, isolated epithelial atypia were found in 670 out of the 2,833 SB (23%). Concomitant cancers corresponded to ductal carcinomas in situ and micro-invasive (77%), invasive ductal carcinomas not otherwise specified (15%), invasive lobular carcinomas (4%), and tubular carcinomas (4%). Fifteen out of the 443 patients with isolated epithelial atypia developed a subsequent ipsilateral (n = 14) and contralateral (n = 1) invasive cancer. The high slide rating might explain the high percentages of epithelial atypia and concomitant cancers and the low percentage of subsequent cancer after a diagnosis of epithelial atypia as a single lesion. Epithelial atypia could be more a risk marker of concomitant than subsequent cancer.

Mammography of ductal carcinoma in situ of the breast: review of 909 cases with radiographic-pathologic correlations.

We retrospectively analysed mammographies of 909 ductal carcinoma in situ (DCIS) (1980-1999) and compared our results to those of literature. Microcalcifications were present in 75% of the cases, and soft-tissue abnormalities in 27% cases with association with calcifications in 14% of cases. Palpable masses were found in 12% of the cases and nipple discharge was present in 12% of the cases. The radiographic-pathologic correlation allowed to suspect the DCIS "aggressiveness" on radiologic signs. Granular, linear, branching and/or galactophoric topography of the microcalcifications were correlated with necrosis, grade 3, comedocarcinoma type. A number of microcalcifications higher than 20 was correlated with necrosis and grade 3. Mammographic size was correlated to histologic size. Masses were correlated with grade 1. A diagnosis strategy can be proposed with a multidisciplinar approach.

Value of microsatellite instability typing in detecting hereditary non-polyposis colorectal cancer. A prospective multicentric study by the Association Aquitaine Gastro.

AIM OF THE STUDY: To detect hereditary non-polyposis colorectal cancer (HNPCC) patients with a strategy combining clinical selection (patient age at onset of cancer less than 50 years or family history of HNPCC tumors) and microsatellite instability typing plus immunohistochemistry, leading to mismatch repair (MMR) germline mutation analysis. METHODS: Tumors were screened for microsatellite instability (MSI) and for hmlh1 and hmsh2 immunohistochemical expression. Germline mutation analysis was performed to search for MLH1 and MSH2 mutations in patients with MSI-High and MSI-Low tumors. RESULTS: 197 adenocarcinomas were studied: 164 patients were< or =50 years old, 33 were older than 50 years but had a family history of HNPCC tumors. Fifty tumors (25.4%) were MSI-High, 10 were MSI-Low (5.1%), and 130 were MS-Stable (66%). MSI typing was inconclusive in 7 (3.5%). Immunohistochemistry screening was performed on 165 tumors: sensitivity was 63.6%, specificity was 99%. Germline mutation analysis was performed in 33/60 MSI-High or Low tumors: 23 mutations were noted (70% of the tested patients). CONCLUSION: This proposed strategy of determining microsatellite instability in young colorectal cancer patients or in patients with a family history of HNPCC tumors led to an increased frequency in the detection of MMR germline mutations.

Quality of life in patients with aggressive non-Hodgkin's lymphoma. Validation of the medical outcomes study short form 20 and the Rotterdam symptom checklist in older patients.

In the elderly population, cancer treatment aims to cure and/or maintain Quality of Life (QoL). However, there is little QoL data to provide evidence for QoL benefits for some of the cancer treatments. This pilot study developed valid QoL questionnaires in French, for patients over 65 years with a diagnosis of large cell lymphoma, part of the Lymâge phase II study. They were asked to complete two questionnaires, the Medical Outcomes Study Short Form 20 (MOS SF20; generic) and the Rotterdam symptom checklist (RSCL; cancer-specific). Between June 1995 and April 1997, questionnaires were returned by 63 of 89 patients. This article reports the process undertaken to adapt the English version to a French setting, and provides the results of factor analysis, convergent and discriminant validity and reliability. Our data suggest that QoL questionnaires can be used in elderly patients. These two questionnaires are validated in French and would help us to analyse the QoL of elderly patients with the development of new treatments as done in the Lymâge study.

Pleomorphic liposarcoma: clinicopathologic, immunohistochemical, and follow-up analysis of 63 cases: a study from the French Federation of Cancer Centers Sarcoma Group.

The clinicopathologic and immunohistochemical features of 63 pleomorphic liposarcomas are presented. There were 35 men and 28 women (median age 63 years; range 18-93 years). Tumor size ranged from 2 to 23 cm (median 10 cm). Tumor locations included lower extremity (36.5%), especially the thigh (28.5%), limb girdles (17.5%), upper extremity (16%), thoracoabdominal wall (9.5%), and internal trunk (20.5%). A total of 75% were deep seated and/or extracompartmental. Histologically, lesions show a varying combination of lipogenic and nonlipogenic areas characterized by malignant fibrous histiocytoma-like, round cell liposarcoma-like, and/or epithelioid/carcinoma-like features. A pericytic pattern was focally present in 15 (24%) tumors. Eighteen (29%) lesions were grade 2, and 45 (71%) were grade 3 sarcomas. Tumor necrosis was observed in 51 (81%) cases, vascular invasion in three, and mitotic counts ranged from 3 to 124 per 10 high power fields (median 25). Lipogenic areas were S-100 protein immunoreactive, at least focally, in 20 of 42 (48%) cases. Nonlipogenic areas showed focal reactivity for smooth muscle actin (24 of 49; 49%), desmin (9 of 48; 19%), CD34 (18 of 45; 40%), S-100 protein (5 of 49, 10%), CD68 (6 of 46, 13%), and epithelial membrane antigen (13 of 49, 26.5%). Epithelioid areas showed epithelial membrane antigen (4 of 11; 36%) but not cytokeratin (0 of 11) reactivity. Treatment procedures in 51 patients consisted of simple tumorectomy (16) and wide excision (33). Five and 31 patients received neoadjuvant and adjuvant chemotherapy and/or radiation therapy, respectively. Follow-up (48 patients, range 7-276 months; median 38 months) showed a 45% local recurrence rate and a 42.5% metastasis rate, metastases occurring mostly in lungs and pleura. Seventeen patients (35%) died of disease, of whom none was metastatic at diagnosis. Five-year overall, metastasis-free, and local recurrence-free survivals were 57%, 50%, and 48%, respectively. Patient age > or =60 years, truncal tumor location, deep situation, tumor size >5 cm, vascular invasion, and incomplete tumor excision were significant adverse prognostic factors. Tumor grade and histology did not affect patient outcome. In conclusion, pleomorphic liposarcoma is a rare, often deep-seated and limb-based aggressive and metastasizing neoplasm of late adulthood. It shows a wide range of morphologic appearances, but tumor grade and histology have no effect on patient outcome.

Immunohistochemically detected lymph node metastases from breast carcinoma: practical considerations about the new American Joint Committee on Cancer classification.

BACKGROUND: The authors applied the sixth edition of the American Joint Committee on Cancer (AJCC) classification system to their previously published group of patients with breast carcinoma who had immunohistochemically detected lymph node metastases. METHODS: The original lymph node-negative slides from 218 patients, including 129 patients with infiltrating duct carcinoma (IDC) and 89 patients with infiltrating lobular carcinoma (ILC), were reviewed and then destained to perform immunohistochemistry. Each tumor cell deposit was measured. Single tumor cells could not counted or measured (because the distance separating the most distant cells could be > 2.0 mm), but they were separated into Class 1 (sparse) and Class 2 (numerous). According to the AJCC classification, isolated tumor cells (ITCs) should be classified as pN0(i+) when they measure < or = 0.2 mm and pN1mi when they measure < or = 2.0 mm but > 0.2 mm. RESULTS: ITCs were found in 13 IDCs (10%) and in 37 ILCs (41%) and corresponded in IDCs to a single deposit of a few tumor cells that measured < or = 0.2 mm (associated with a single tumor cell in 3 tumors) and were classified as pN0(i+). In ILCs, ITCs corresponded to single tumor cells with an irregular distribution throughout the entire lymph node section (24 ITCs and 13 ITCs in Class 1 and Class 2, respectively) and were difficult to classify. CONCLUSIONS: The results suggest that there are 2 categories of pN0(i+): measurable tumor cell deposits < or = 0.2 mm, which were found exclusively in IDCs; and nonmeasurable ITCs, which were found in ILCs and rarely in IDCs. The new classification is difficult to apply to ITCs that are detected by immunohistochemistry in ILCs.

Breast ductal carcinoma in situ with microinvasion: a definition supported by a long-term study of 1248 serially sectioned ductal carcinomas.

BACKGROUND: The significance of microinvasion is still debated and clinical management is controversial. The authors defined ductal carcinoma in situ with microinvasion (DCIS-MI) as DCIS with infiltration of the periductal stroma by a few tumor cells, singly (type 1) or in clusters (type 2). With this definition, the authors attempted to evaluate the clinical significance of microinvasion. METHODS: The authors compared the clinical, pathologic features, and survival (median follow-up, 7.3 years) of 1248 patients with, respectively, DCIS (722 patients), DCIS-MI with microinvasion type 1 and type 2 (243 patients), and invasive ductal carcinoma in situ with a predominant DCIS component greater than or equal to 80% of the tumor (IDC-DCIS, 283 patients). RESULTS: Microinvasion was associated with DCIS histologic type, grade, and extent (respectively, P < 10(-8), P < 10(-3), P < 10(-4)). Axillary lymph node metastases were observed in a few patients with DCIS and DCIS-MI type 1 (respectively, 1.4% and none), in 10.1% with DCIS-MI type 2 and in 27.6% with IDC-DCIS. Metastasis free and overall survival probabilities were significantly different between three groups in the following order from best to worst prognosis: 1) the group comprising DCIS and DCIS-MI type 1, 2) the DCIS-MI type 2 group, and 3) the IDC-DCIS group. CONCLUSIONS: The authors' results suggest there are two types of DCIS-MI: 1) type 1 that behaves like DCIS and should be managed as such; 2) type 2 that is less pejorative than IDC-DCIS but is more so than type 1.