Joško Bezić

Her-2/neu assessment for gastric carcinoma: validation of scoring system.

BACKGROUND/AIMS Gastric cancer is the second leading cause of cancer mortality in the world. Amplification of HER-2/neu oncogene has become an important biomarker for identifying patients who respond to HER-2 targeting therapy. A number of studies have analyzed HER-2/neu overexpression in gastric carcinoma, and the rate of HER2 positivity is variable, ranging from 6% to 35%. METHODOLOGY In our study HER-2/neu expression was assessed on 73 samples of primary gastric cancer, using immunohistochemistry. For 19 patients preoperative biopsy samples and resected specimens were available. Additionally, internal ring study was performed to estimate intraobserver variability of IHC scoring among pathologists at our department. RESULTS HER-2/neu overexpression was found in 10 (13.6%) of the tested samples, and it was more common in intestinal (22.5%) than the diffuse type (3.7%). Not one of the 6 analyzed mixed type tumors showed HER-2/neu expression. For the paired samples (preoperative biopsy samples and resected specimens) the concordance rate for HER-2/neu expression was 94.7%. CONCLUSIONS According to high concordance rate in paired samples we consider it appropriate to evaluate HER2 expression on biopsy specimens, especially in unresectable cases, and to re-evaluate it on resected specimens if available, due to high heterogeneity of a gastric cancer.

Minimal Breast Cancer in Split Region of Croatia on the Eve of the National Mammographic Screening Program

To the Editor: As a result of the worldwide use of screening mammography, many breast biopsies are now performed for small, usually nonpalpable, mammographically detected abnormalities. The breast biopsies of screened abnormalities contain a large number of socalled minimal breast cancers. This term includes all non-invasive cancers (Tis) and invasive cancers up to 1 cm in diameter (minimal invasive cancers, T1a,b) (1,2). In our previous work on the minimal breast cancers in Split region (period 1997–2001), the proportion of Tis and T1a,b cancers was 2.78% and 15.16%, respectively (3). These low proportions were concordant to the proportions of detected minimal breast cancers in developed countries in the prescreening period. Therefore, we advocated the introduction of regional and national breast cancer early detection programs (3). The national breast cancer early detection program started in Croatia in the second half of 2006 under the auspices of Ministry of Health and Social Affairs. The screening method consists of mammography in 50–69-year-old women, with 2-year screening interval. The main goals of this program are reduction in breast cancer mortality for 25% over a 5-year period starting with the introduction of the program, detection of higher percentage of cancers at an early stage, and improvement of the life quality of the patients with breast cancer (4). We wanted to explore basic pathohistologic characteristics of the breast cancer detected in Split region before the introduction of screening program, particularly the presence of the minimal breast cancers that are expected to be increasingly detected during screening program. Therefore, the results presented here may be used in future evaluation of screening success. The pathohistologic data of 2,141 consecutively operated breast carcinomas in Clinical Hospital Center Split in the period from 1997 to 2006 were retrieved from the data base of Institute of Pathology and Cytology, Clinical Hospital Split, Croatia. The data (tumor size, histologic types of invasive and noninvasive tumors, differentiation grade, axillary lymph node status, and hormonal receptor expression) were collected over two 5-year periods (1997–2001 ⁄ 2002– 2006), and statistically correlated to explore possible changing trends. According to the greatest diameter, the invasive tumors were divided in the following groups: the tumors with the diameter of £2 cm, 2–5 cm, and >5 cm. The cancers with the diameter of £2 cm were additionally divided in T1a,b tumors (diameter of £1 cm; microinvasive carcinomas (T1mic) are also included in this group), and T1c tumors (diameter of 1.1–2 cm) (5). Invasive and non-invasive tumors were histologically classified according to the WHO Classification of breast tumors (5). The grade of invasive tumors was assessed according to Elston and Ellis, and the grade of non-invasive ones according to the classification proposed by a group of European pathologists (6,7). The patients with one or more tumor positive lymph nodes were considered node positive (N+). Estrogen (ER) and progesterone receptor (PgR) status were determined mainly biochemically in the period 1997–2001, using the dextran-coated charcoal method (DCC) with cut-off level of 5 fmol ⁄ mg of protein for ER, and 10 fmol ⁄ mg of protein for PgR. In the period 2002–2006, the hormonal status was determined immunohistochemicaly using the standard avidin-biotin complex method. ER and PgR were considered positive, if there was nuclear staining in more than 10% of neoplastic cells. Address correspondence and reprint requests to: Joško Bezić, MD, Institute of Pathology, Forensic Medicine and Cytology, Clincal Hospital Center Split, Spinčićeva 1, 21 000 Split, Croatia, or e-mail: jb@mefst.hr.

Mammographic screening has failed to improve pathohistological characteristics of breast cancers in Split region of Croatia

The national breast cancer screening program has been introduced in Croatia in the second half of 2006 with mammography as the screening method. We investigated the impact of screening mammography on the basic pathohistological characteristics of breast cancers retrieved from the database of large community hospital. The data were collected in the period following the initiation of national mamographic screening program (2007-2011), and compared with the data collected in the period before the program introduction (2002-2006) to explore the possible changing trends. In the screening period 1,320 breast cancers were diagnosed, while in the prescreening period 1,204 breast cancers were diagnosed (p=0.02). We found the reduction of mean tumor size (p=0.039), decrease of the diagnosed non-invasive cancers (p=0.001), and the higher percentage of the diagnosed ductal (p=0.0003) and grade 3 (p=0.038) invasive cancers in the screening period. We also noticed higher percentage of diagnosed advanced breast cancers with unknown tumor size in the screening period (p=0.027). We concluded that the implementation of national screening program did not improve the pathohistological features of breast cancers in our region probably due to low response rate to screening invitation.

Can we identify the group of small invasive (T1a,b) breast cancers with minimal risk of axillary lymph node involvement? A pathohistological and DNA flow cytometric study

The goal of this study was to identify a group of small (≤1cm) breast cancers (T1a,b) with a particularly low probability of axillary lymph node metastases, where routine axillary staging may be unnecessary. We retrospectively analyzed 152 T1a,b breast carcinomas with axillary dissection surgically removed at Clinical Hospital Center Split (Croatia) in the period from 1997 to 2006. The analysis included 40 T1a,b cancers with, and 112 T1a,b cancers without axillary lymph node metastases. The basic morphological features of cancers were investigated histologically, while hormone receptors and HER2/neu were investigated immunohistochemically with an additional CISH analysis of HER2/neu 2+ cases. The ploidy and S-phase fraction were determined by DNA flow cytometry. The association of the investigated features with the likelihood of axillary lymph node metastases was analyzed by univariate and multivariate analysis. The univariate analysis showed that lymph node metastases were associated with tumor size (T1a/T1b; p=0.026), histological type (ductal/non-ductal; p=0.014), lymphovascular invasion (p<0.001), HER2/neu expression (p=0.04), ploidy (p=0.027), and combined values of ploidy and S-phase fraction (p=0.025). The lymphovascular invasion was the only independent factor associated with axillary nodal metastases (p=0.01). In the group of T1a,b cancers without lymphovascular invasion, HER2/neu expression (p=0.021) and combined values of ploidy and S-phase fraction (p=0.016) were independent factors associated with axillary lymph node metastases. This study showed that diploid T1a,b cancers with low S-phase fraction, which are also without lymphovascular invasion and HER2/neu amplification, represented the group of cancers with a low probability of axillary lymph node metastases.

Combined Fibroadenoma and Tubular Adenoma of the Breast: Rare Presentation that Confirms Common Histogenesis

“low grade” fibrosarcoma-like aspect (Fig. 3). Mitotic activity was not exceeding three mitoses/10 HPF. Immunohistochemically, tumor cells were extensively positive for CD34 (Fig. 4). No additional adjuvant treatment was done. After 24 months, the patient underwent to a breast reconstruction. Actually, our patient is well 30 months after our surgery. The localization of DFSP in the skin of the breast (DFSP-Br) is rare. In our review of the last 10 years, only 21 cases of DFSP-Br were reported and it commonly rushes between the second to fifth decades of life. Sometimes, differentiating DFSP from other dermatofibromas using CD34 and the other antigens immunochemistry remains difficult; therefore, in these cases, a search of some genetic alterations may be useful. About the therapy, the recommended treatment of DFSP is a wide surgical excision with pathologic negative margins. Despite optimal surgical management, local recurrences are detected in 25% of patients. A problem not to be underestimated in case of DFSP-Br is the esthetic one, which can be felt particularly in the case of reoperation for recurrence, especially in small breasts. Since wide excision usually causes noteworthy distortion and leaves patient with significant cosmetic problems, reconstructive procedure is required in almost every istances. Due to the significant recurrences, even if with free pathologic margins, a long-term postoperative follow-up is mandatory.

Near-Diploid Hyperploidy in Early Breast Cancer (T1a,b) is Associated with Higher Risk of Lymph Node Involvement

Due to the worldwide implementation of the mammographic screening program early breast cancer (T1a,b) has become more prevalent form of breast cancer. Although T1a,b breast cancers are generally associated with excellent prognosis, some of them, particularly those with lymph node involvement, has unfavourable outcome. Searching for additional prognostic factors, we investigated DNA content of 163 T1a,b cancers measured by DNA flow cytometry, and correlated it with regional lymph node status. T1a,b cancers were divided into four ploidy classes based on their DNA index (DI): hypodiploid (DI < 0.95), diploid (DI 0.95–1.05), low-hyperploid (DI 1.06–1.3), and high-hyperploid (DI > 1.3). Diploid T1a,b cancers were associated with negative lymph node status (p = 0.003). Among aneuploid cancers only low-hyperploid tumors were associated with positive lymph node status (p = 0.03). The histopathological features of low-hyperploid group of T1a,b cancers did not differ from the other three ploidy groups of cancers, except for lower S-phase fraction of tumor cells in low-hyperploid group compared to high-hyperploid group (p = 0.01). Our data showed that near-diploid hyperploid T1a,b cancers are associated with higher risk of lymph node involvement despite similar clinicopathological features shared with other ploidy classes of T1a,b tumors.

Pathobiology of HER 2 positive small invasive (T1a, b) breast cancer: "A wolf in sheep's clothing".

Small invasive breast cancers (cancers with maximum diameter <1cm, T1a,b) become more prevalent form of breast cancer as a result of the introduction of breast cancer mammographic screening programs. Although associated with an excellent prognosis, T1a,b breast cancers are heterogeneous group of tumors with prognostically unfavorable subset of cases, primarily those with axillary lymph node metastases. To determine if the HER2 overexpression is associated with the prognostically unfavorable traditional clinicopathological features in this group of breast cancers, clinicopathological features (age, tumor size, histological type, histological grade, nodal status, hormone receptor status, proliferation index, lymphovascular invasion, ploidy) of 38 HER2 positive T1a,b cancers were compared with those of the control group consisting of 315 HER2 negative T1a,b cancers. The comparison of clinicopathological features was made using χ2 and t-test. HER2 positive T1a,b breast cancers were significantly associated with higher tumor grades (p<0.001), negative hormone receptors (p=0.008), presence of lymphovascular invasion (p=0.025), high proliferation index (p<0.001), and abnormal DNA content (p=0.04). We also noticed the higher frequency of lymph node positive cases in the HER2 positive group of cancers (p=0.05). There were no differences in age, tumor size and histological type between investigated groups. Our group of HER2 positive T1a,b breast cancers was associated with many unfavorable traditional prognostic factors, demonstrating that this subtype of early breast cancer has an aggressive biological phenotype which may have potential benefit from adjuvant chemo and immunotherapy.

Breast fibroadenoma with pseudoangiomatous (PASH-like) stroma

Pseudoangiomatous stromal hyperplasia (PASH) is a breast stromal change, histologically characterized by anastomosing, slit-like spaces lined by slender myofibroblasts and surrounded by dense collagenous stroma. Mass forming cases clinically and radiologically simulate fibroadenoma. A middle aged women presented with unpalpable breast nodule discovered on ultrasound examination. The ultrasound characteristics were typical for fibroadenoma, while fine-needle aspiration cytology was inconclusive. The histological examination of the lumpectomy specimen showed fibroadenoma with peculiar stromal alteration consistent with pseudoangiomatous stromal hyperplasia. The presented case of fibroadenoma with pseudoangiomatous hyperplasia within its stroma demonstrates the relationship between these two entities not only clinically and radiologically, but also histologically.

Osteonevus of Nanta Revisited: Clinicopathological Features of 33 Cases.

To the Editor: Conventional nevus with secondary ossification (osteonevus of Nanta) is seldom encountered in routine dermatopathological practice. This type of nevus is usually located in the upper part of the body, with the predilection in females. Histopathologically, the ossification is usually in the form of the small islands of compact lamellar or structureless bone at the base of the lesion. In some cases, trabecular bone with bone marrow and adipocytes may also be present. The ossification is believed to be the result of inflammatory process in surrounding dermis, but the production of osteogenic growth factors by nevus cells may also have an impact in pathogenesis. We investigated the clinicopathological features of osteonevi, prospectively identified in the two-year period. This is the first prospective study of the osteonevi, and the first study with the statistical comparison of the clinicopathological features of the nevi with and without ossification. Histopathological sections of all types of nevi diagnosed in populationbased practice were prospectively screened for the presence of ossification by two pathologists (J.B. and I.K.) in 2year period (March 1, 2013–February 28, 2015). The ethnic background of all patients with nevi was European. The presence of minimal amount of mature bone was required for the inclusion of the nevus in the investigated group. During this period, pathologists prospectively evaluated 2.122 nevi, with 33 osteonevi found. The control group consisted of all other types of nevi consecutively diagnosed in the first 3 months of the investigated period. The tissue of surgically excised nevi was fixed in neutralbuffered formalin, grossly sectioned, and embedded in paraffin for routine histopathological examination. The sections were not decalcified because the foci of ossification were small and not suspected grossly. The slides were stained with hematoxylin and eosin. After the histopathological detection of ossification on initial slides, additional multiple levels were examined in each case, with no special stains performed. The following features of all nevi were recorded: age and sex of the patients, site of the occurrence, maximum diameter, histopathological type, and the presence of histopathological features of associated chronic inflammation. In osteonevi, the type and the position of bone formation were recorded. According to the type, the ossification was divided in the compact type and type with central cavitation and visible fatty bone marrow within. According to the site, the ossification was divided in intralesional (ossification surrounded by nevus cells) and extralesional type. The comparison of the clinicopathological features of osteonevi and control group of nevi was made using x2 and t test. The significance of the differences was calculated as P value, with probabilities lower than 5% (P , 0.05) considered significant. Statistical analysis was performed using the Statistical Package for Social Sciences, version 15.0 for Windows (SPSS Inc, Chicago, IL). Among 2.122 histopathologically evaluated nevi, 33 (1.56%) osteonevi were found. The clinicopathological features of the diagnosed osteonevi and the control group of nevi are shown in Table 1. There were 17 male and 16 female cases of osteonevi. Most of the osteonevi were solitary but two patients had two lesions each (both males). All cases of osteonevi were from nonglabrous skin with most cases located in the head and neck region, with additional cases from the back, and the extremities. Among the head and neck situated cases, 18 cases were located on the face, 4 on the forehead, 2 on

Prognostic Significance of Ploidy and DNA Index in Patients with Epithelial Ovarian Carcinoma

Paraffin-embedded tissue from sixty-one ovarian epithelial cancers were analyzed by histology and flow cytometry.The results of DNA analysis were correlated with clinicopathologic data.For univariate survival analysis we used the Kaplan-Meier model and log-rank test to asses the differences between the groups.For multivariate analysis, Cox proportional hazard regression model was used to examine several parameters simultaneously. Aneuploidy was significantly associated with advanced clinical stage, menopausal status, high grade, serous histological type and vascular invasion. Univariate analysis showed that aneuploidy and DNA index were associated with shorter overall survival. Multivariate analysis showed only vascular invasion as an independent prognostic factor. Although abnormal DNA content was associated with traditional clinicopathological characteristics for aggresive tumors and shorter overall survival, aneuploidy and high DNA index had no independent prognostic value in our study.