HER2 status in breast cancer: experience of a Spanish National Reference Centre

Transcription

1 Clin Transl Oncol (2011) 13: DOI RESEARCH ARTICLES HER2 status in breast cancer: experience of a Spanish National Reference Centre Marta Cuadros Carlos Cano Francisco Javier López Paloma Talavera Inmaculada García-Peréz Armando Blanco Ángel Concha Received: 18 March 2010 / Accepted: 16 October 2010 Abstract Background Accurate HER2 testing is of great clinical value for the identification of breast cancer patients who are eligible for trastuzumab therapy. The aim of this study is to review breast carcinomas diagnosed from 2001 to 2007 at a Spanish National Reference Centre for HER2 testing, evaluating the agreement between HER2 immunohistochemical (IHC) tests and fluorescence in situ hybridisation (FISH) tests. Methods Demographic and clinical information was obtained from 2751 breast carcinoma patients. HER2 IHC and FISH tests were performed both in a local laboratory and in the reference centre. The HER2 IHC0/1+, IHC2+, IHC3+ and FISH-positive patients comprised 64%, 20%, 16% and 24% of the available population, respectively (results from the reference centre). Using statistical approaches, we evaluated the agreement between: (1) HER2 IHC and FISH tests, and (2) results provided by the local and the reference laboratories. Results The data confirmed a statistically significant relation between HER2 overexpression and amplification. We also found that instances of polysomy 17 and heterogeneous patterns of HER2 expression (heterogeneous staining distribution in different areas of the same tumour) are more frequently observed in HER2-positive tumours. Finally, since the diagnoses were made from 2001 to 2007, we could also observe a rising agreement rate between laboratories/pathologists with time. Conclusions HER2 testing is most accurate when performed by experienced pathologists and at a high-volume reference laboratory. Polysomy 17 and HER2 heterogeneous staining patterns should also be considered for a better understanding of the variation in the anti-her2 therapeutic response. Keywords Breast neoplasms HER2 Immunohistochemistry In situ hybridisation Fluorescence Descriptive analysis M. Cuadros ( ) C. Cano F.J. López A. Blanco Department of Computer Science and Artificial Intelligence University of Granada C/ Daniel Saucedo Aranda, s/n ES Granada, Spain marta@decsai.ugr.es M. Cuadros I. García-Peréz Á. Concha Department of Pathology Hospital Universitario Virgen de las Nieves Granada, Spain P. Talavera Á. Concha Tissue and Tumor Bank Department of Pathology Hospital Universitario Virgen de las Nieves Granada, Spain Background Breast cancer is the most common cancer in women in developed countries, accounting for 27% of cancers in European women. Spain has one of the lowest rates in Europe, in terms both of incidence (estimated age-standardised rate of 51 per 100,000 people) and mortality (16 per 100,000) [1]. Breast cancer death rates have been dropping steadily in Spain because of earlier detection and the existence of better treatments. However, breast cancer rates have risen about 30% in the past 25 years in Western countries, due in part to increased screening, which detects the cancer in earlier stages [2]. Breast cancer survival depends on prognostic factors, which are distributed differently in the population. The

2 2 Clin Transl Oncol (2011) 13: most widely used prognostic factors are primary tumour size, lymph node status, tumour histological grade, and tumour receptor status, such as hormone receptors and human epidermal growth factor receptor 2 (HER2) [3]. HER2, oestrogen receptors (ER) and progesterone receptors (PgR) play an important role in the development and progression of breast cancer. Furthermore, the response to endocrine and targeted anti-her2 therapies depends on the presence of ER and PgR [4], and HER2 overexpression [5], respectively. The aim of this study is to review the breast carcinomas diagnosed from September 2001 to December 2007 in a Spanish National Reference Centre for HER2 testing, and to perform a comparative analysis between immunohistochemical (IHC) results and fluorescence in situ hybridisation (FISH) data. The population under study comprises 2751 patients diagnosed with breast carcinoma, for which HER2 test results, as well as other tumour characteristics and demographic data, are available. Methods Patient identification Data from 2751 patients with breast carcinoma diagnosed from September 2001 to December 2007 were supplied by one of the Spanish National Reference Centres for HER2 testing (Hospital Universitario Virgen de las Nieves, HUVN). During the first years of the study (before 2002), most of the cases reviewed by this centre correspond to aggressive or retrospective tumours (considering intervention and diagnosis date). However, equivocal cases are added after 2002, producing a dataset more representative of the population. The data were contributed by five local hospitals from eastern Andalucia and submitted to the Pathology Department of the Hospital Virgen de las Nieves for diagnosis confirmation. Local laboratories were not subject to rigorous quality control procedures and did not rely on very experienced pathologists. The interpretation of results may thus be a key factor in these low-volume testing laboratories, making them suitable for diagnosis confirmation at the HUVN Reference Centre. Submitted datasets were subjected to standard and comprehensive edit checks, and verification and validation procedures before being inputted into the database. In addition, HER2 status (evaluated both by IHC and FISH tests) was studied by the reference pathology laboratory (Department of Pathology at HUVN). Demographic data, stage, histology, and immunohistochemical markers, such as ER, PgR, p53, ki67 (data not shown) and HER2 status, were recorded. The impact of socio-economic and racial factors was not considered because 100% of the population was covered by the Spanish Health System. HER2 testing methodologies at reference laboratory HER2 status of breast tumours was evaluated by both IHC and FISH analysis of paraffin-embedded tissue on conventional slides, which were submitted by each local laboratory. The reference pathology laboratory used FDA approval kits and the HER2 test results were interpreted according to manufacturers' test kit protocols. IHC analysis for HER2 protein overexpression was performed using the HercepTest (Dako Diagnostics) in an automatic stainer (TechMate Horizon, Dako Diagnostic) and the degree of staining was scored by two blinded and independent pathologists (A.C. and P.T.). All cases without a consensus diagnosis were reviewed jointly on a multi-headed microscope and discussed by the two expert pathologists until a final diagnosis was agreed. IHC was scored on a qualitative scale from 0 to 3, based on the interpretation of the staining intensity, with 0 and 1 classified as negative, 2 as equivocal, and 3 as positive. Cytoplasmic staining was considered to be nonspecific. Heterogeneous pattern of staining for HER2 was also defined as the heterogeneous HER2 staining distribution in different areas of the same tumour. The criteria used for IHC scoring followed the recommendations for HER2 testing by the American Society of Clinical Oncology and the College of American Pathologists (ASCO/CAP). Particularly, heterogeneous pattern of staining cases were defined as those satisfying one of these conditions [6]: (a) the sample shows a difference in IHC score over 1 in more than 30% of infiltrating tumour cells; (b) the sample shows a difference in IHC score over 2 in 10 30% of infiltrating tumour cells; (c) the sample shows a difference in IHC score over 3 in <30% of infiltrating tumour cells. For the detection of Her2/neu amplification we used the HER2 FISH pharmdx kits (Dako Diagnostics) and followed the recommendations of the ASCO/CAP. The FISH testing methodology is a semiquantitative method based on the computation of the average ratio of Her2/neu signals to CEP17 signals in non-overlapping interphase nuclei of the lesion. Tumours with a Her2/neu:CEP17 ratio 2:1 were considered positive for gene amplification. The Her2/ neu:cep17 signals were visualised by two blinded and independent pathologists (A.C. and P.T.). This test also allowed a simultaneous determination of chromosome 17 copies. Polysomy 17 was defined as the ocurrence of three or more copy numbers of chromosome 17 centromere per cell. Statistical data analysis A descriptive analysis was performed for demographic, histological and immunohistochemical data. The statistical analysis was performed using SPSS 13.0 software (SPSS Inc., Chicago, IL). We examined the association between IHC2+, IHC3+, FISH-positive cases and other variables using 2 tests. Cut-off values for the different biomarkers were established before the statistical analysis.

3 Clin Transl Oncol (2011) 13: Table 1 Concordance between IHC and FISH tests by the reference laboratory according to the registration date of the breast tumours HercepTest FISH >3 Before p< p< p< Results We presented a retrospective descriptive study of 2751 breast tumours diagnosed between September 2001 and December HER2 testing was performed in 1918 breast tumour samples using IHC and FISH at the reference pathology laboratory at HUVN. Most of the samples (1025/1918, 53%) were collected at the HUVN, the rest of them being submitted by five other local hospitals from eastern Andalucia. Our study had some limitations. Most notably, most of the samples included in the earlier stages of the study (before 2002) were submitted by the local laboratories and mainly correspond to retrospective, aggressive or equivocal HER2 IHC cases. Patients demographics Age was reliably confirmed for 829 out of the 2751 patients. The age range was from 20 to 91 years with a mean age of 56.88±13.31 years. Men with breast cancer were excluded from this analysis. The most common type of breast carcinoma was infiltrating ductal carcinoma not otherwise specified (IDCnos). The majority of patients presented a histological grade higher than I. Concordance between IHC and FISH results In order to estimate the impact of retrospective cases, we divided the dataset into three subgroups according to the registration date: (1) before 2002 (n=303) (this set was characterised by a higher number of retrospective breast tumours); (2) (n=821); and (3) (n=794). For all the samples, IHC and FISH tests were performed at the HER2 reference laboratory. We only considered the cases in which the FISH test was successfully performed. The percentage of IHC0/1+ cases increased from 118 (before 2002), to 423 ( ) and to 579 ( ). In contrast, we observed a significant decrease in the number of IHC2+ tumours from to (Table 1). The concordance rates between IHC and FISH were 84%, 96% and 97% in IHC0/1+ tumours, and 63%, 87% and 95% in IHC3+ tumours for the three different time periods, respectively. Most of IHC2+ tumours were classified as FISH-negatives. Concordance between the reference laboratory and the local laboratories for HER2 findings The level of agreement between the local and reference laboratories for HER2 IHC testing was examined in 1226 breast tumour patients (Table 2). Most of the IHC local results were obtained using HercepTest (35.6% of available data). These results showed that for the 744 patients for which the reference laboratory reported a IHC 0/1+, 463 were assigned the same score (IHC0/1+) by the local laboratory. Similarly, 122 out of 189 patients classified as IHC3+ by the local laboratory were confirmed as IHC3+ by the reference centre; and 293 out of 641 cases initially reported as IHC2+ were also confirmed as IHC2+ by the reference centre (Table 2a). Therefore, the agreement rates between the local and reference laboratories were 62% for IHC0/1+, 65% for IHC3+ and 46% for IHC2+, showing that the local laboratories reported a high number of IHC2+ tumours which were later classified as IHC0/1+ by the reference laboratory. Finally, we examined the agreement rate between the FISH results from the reference laboratory and the IHC results from the local laboratories (Table 2b). The agreement between the local and reference laboratories was higher for IHC0/1+ samples (91%) than for IHC3+ (78%) samples. As expected, we also noted that the concordance between HER2 IHC and FISH results was higher when both tests were performed at the reference laboratory: 95% for IHC0/1+ and 88% for IHC3+ (Table 2c).

4 4 Clin Transl Oncol (2011) 13: Table 2 Testing methods used by reference and local laboratories to determine HER2 status IHC local laboratories Total reference laboratory (a) IHC reference laboratory Total local laboratories (b) FISH reference laboratory > Total local laboratories (c) FISH reference laboratory > Detailed analysis of IHC0/1+, IHC2+, IHC3+ and FISH-positive breast tumours Table 3 shows a detailed comparative analysis of IHC0/1+, equivocal HER2 (IHC2+) and HER2 overexpressed/amplified cases (IHC3+ and/or FISH-positive confirmed by the reference laboratory). We also incorporated into the analysis other features of the tumours such as the grade, number of copies of chromosome 17 and the presence of heterogeneous staining patterns. Some observations derived from this analysis follow: A total of 1708 cases were classified as HER2 IHC0/1+. FISH tests could be carried out for 1120 of these cases, 56 of them (56/1120, 5%) presenting HER2 gene amplification. We found 196 (196/1129, 17%) of these tumours presenting polysomy 17. Furthermore, 34 (34/1708, 2%) IHC-negative cases showed an HER2 heterogeneous staining pattern. A total of 530 tumours were classified as HER2 IHC2+. FISH tests were successfully carried out for 505 of these cases, obtaining 368 cases (368/505, 73%) classified as FISH-negative and 137 (137/505, 27%) as FISHpositive. An analysis of the prevalence of polysomy 17 in the IHC2+ tumours showed that 154 tumours (154/505, 33%) presented polysomy 17. Most of the IHC2+ tumours (91%, 482/530) did not show HER2 heterogeneous staining pattern. A total of 442 breast tumours were classified as HER2 IHC3+. We got FISH scores for 293 of these samples, only 12% (35/293) of them being classified as FISH-negative. The rate of discordance was much higher than the recommended threshold (maximum of 5%) due to the first cases which were included in the study before considering the ASCO recommendations. We also found 36% (106/293) of cases presenting polysomy 17. There was a HER2 heterogeneous staining pattern of 13% (56/442). From the total number of breast cancer tumours analysed by FISH, 24% (462/1960) amplified HER2, around 38% (174/462) showed polysomy 17 and 19% (85/462) exhibited an HER2 heterogeneous staining pattern. HER2 overexpression/amplification correlates with polysomy 17 and heterogeneous staining We performed a statistical analysis for IHC, FISH, chromosome 17 and heterogeneous pattern data. The findings confirmed a high level of concordance between the results provided by HER2 IHC and FISH techniques (p<0.0001). In addition, the copy number of chromosome 17 in overexpressed/amplified HER2 breast tumours was significantly higher (p<0.0001), indicating a general association of polysomy 17 with HER2 overexpression/amplification. Interestingly, heterogeneity of HER2 staining pattern was associated with high levels of HER2 protein (p<0.0001), DNA HER2 amplification (p<0.0001) and polysomy 17 (p<0.0001). Discussion Tumour markers are becoming increasingly important in breast cancer research due to their impact on prognosis, treatment and survival, and because of their relation to breast cancer subtypes [7]. Routine clinical management of breast cancer patients relies on traditional prognostic factors, which include ER, PgR and HER2. Hormone-receptor-negative breast tumour patients have a worse prognosis because hormonal therapy cannot be used in treatment and they receive traditional systemic chemotherapy [8]. Similarly, HER2 positivity is of great clinical value in breast tumours for the identification of those patients who are eligible for trastuzumab therapy. This study included a large sample of breast tumour patients from the Spanish population, thus allowing us to

5 Clin Transl Oncol (2011) 13: Table 3 Characteristics of breast cancer patients included in this study All samples (n=2751) IHC0/1+ (n=1120) IHC2+ (n=530) IHC3+ (n=442) FISH+ (n=462) Grade I 143 (17%) 110 (22%) 22 (12%) 9 (5%) 8 (6%) II 384 (45%) 220 (44%) 108 (57%) 52 (32%) 49 (35%) III/IV 329 (38%) 160 (34%) 59 (31%) 103 (63%) 84 (59%) IHC (27%) 696 (41%) 0 (0%) 0 (0%) 9 (3%) (37%) 1012 (59%) 0 (0%) 0 (0%) 47 (10%) (20%) 0 (0%) 530 (100%) 0 (0%) 137 (30%) (16%) 0 (0%) 0 (0%) 442 (100%) 258 (57%) FISH (26%) 428 (38%) 66 (13%) 2 (1%) 0 (0%) (50%) 636 (57%) 302 (60%) 33 (11%) 0 (0%) (8%) 42 (4%) 80 (16%) 21 (7%) 147 (32%) >3 315 (16%) 14 (1%) 57 (11%) 237 (81%) 315 (68%) Chromosome 17 Monosomy 117 (6%) 54 (5%) 33 (7%) 27 (9%) 44 (9%) Diploidy 1374 (70%) 879 (78%) 308 (61%) 164 (56%) 244 (53%) Polysomy 469 (24%) 196 (17%) 154 (33%) 102 (35%) 174 (38%) Heterogeneous Yes 138 (5%) 34 (2%) 48 (9%) 56 (13%) 85 (18%) No 2613 (95%) 1674 (98%) 482 (91%) 386 (87%) 377 (82%) know the tumour features of women with IHC2+ or HER2 overexpression/amplification breast cancer. However, this study was not without limitations: (a) The data was derived from several different sources (laboratories) and lacking some information. (b) Diagnosis of tumours and immunohistochemical tests were carried out by several laboratories without confirmation from the reference laboratory (except for HER2 testing, which was repeated at the reference centre). (c) In some cases the breast cancer samples were not representative of the population due to the fact that some retrospective cases with inadequate tissue fixation, especially aggressive and equivocal HER2 IHC scoring were submitted to the reference laboratory in the first stages of this study. (d) Old cases were analysed before the ASCO recommendations were observed [6, 9, 10]. These could increase the percentage of HER2-positive tumours, affecting the correlation between HER2 testing results [11]. The most common histological type of the sample was primary ductal infiltrating carcinoma and the majority of the tumours presented a histological grade higher than I (Table 3). IHC0/1+, IHC2+, HER2 protein overexpression and HER2 DNA amplification were found in 64%, 20%, 16% and 24% of the samples, respectively. There was a relatively good correlation between IHC and FISH (p<0.0001) in all categories except for the IHC2+ and 27% (137/530) of IHC2+ showed HER2 amplification [12 14]. In this study, breast tumour samples were divided into three subgroups according to the registration date. The impact of retrospective cases on the accuracy and the agreement rates was also determined. The comparison between the IHC and FISH findings showed higher agreement rates in the last years of the study than at the beginning of it, indicating the important role of experience of the pathologists and laboratories. The HER2 status assigned by the reference laboratory was more accurate than those submitted by the local laboratories, also confirming the importance of relying on experienced pathologists and a rigorous high-volume laboratory for HER2 testing. These results comply with the American Society of Clinical Oncology and the College of American Pathologists recommendations for HER2 testing. These recommendations also indicate that HER2 testing must be done in an accredited laboratory or in a laboratory that meets the accreditation and proficiency testing requirements specified by the ASCO [6, 9, 10]. However, quantitative PCR techniques based on the quantitative evaluation of HER2 messenger RNA (mrna) have been described that may provide alternatives to these methods. Furthermore, real-time PCR techniques have demonstrated the increased abundance of HER2 transcripts as well as gene amplification in breast cancer samples [15]. Instances of chromosome 17 polysomy, as well as HER2 amplification, were associated with adverse pathological and clinical outcomes in breast cancer [16]. In our study, we reported 24% of cases presenting polysomy 17, a lower percentage than in similar studies [17 19]. According to the status of the HER2 gene, the polysomy 17 was more frequent in IHC3+ and FISH-positive cases, supporting a general association between an increased chromosome 17 copy number and HER2 gene overexpression and amplification [20, 21]. Polysomy 17 would have an additive effect in these tumours and contribute to the overall increase of HER2 in the tumour cells [22]. On the other hand, we showed that polysomy 17 was also significantly associated with heterogeneous HER2 staining patterns, suggesting that breast tumours with chromosome 17 polysomy are more likely to show intratumoral heterogeneity.

6 6 Clin Transl Oncol (2011) 13: In addition, IHC3+ and FISH-positive breast tumours exhibited a more heterogeneous HER2 staining pattern (13% and 18%, respectively) than HER2 IHC2+ breast tumours. The heterogeneity of HER2 expression should be considered a true biological phenomena, arising from different populations of cells in the same tumour, and not a technical problem due to a poor fixation or the type of fixative. The polysomy 17 and HER2 heterogeneity may contribute to a better understanding of HER2-positive tumours and its variation in therapeutic responses. Furthermore, this could also explain the conflicting data in studies about the prognostic and predictive role of HER2 status in breast cancer patients [23]. In conclusion, this study indicates that HER2 testing should only be performed in reliable, high-volume laboratories. The roles of polysomy 17 and HER2 heterogeneity in breast tumour patients should not be underestimated, although more studies are required to study the true clinical implications of these biological events. Acknowledgements This work has been carried out as part of projects P08-TIC-4299 and CTS2200 of JA (Junta de Andalucía), Sevilla and TIN of DGICT (Dirección General de Investigación Científica y Tecnológica), Madrid. Conflict of interest The authors declare that they have no conflict of interest relating to the publication of this manuscript. References 1. Ferlay J, Bray F, Pisani P, Parkin DM (2004) LOBOCAN 2002: cancer incidence, mortality and prevalence worldwide [IARC CancerBase No 5, version 20]. IARC Press, Lyon 2. Pollan M, Ramis R, Aragones N et al (2007) Municipal distribution of breast cancer mortality among women in Spain. BMC Cancer 7:78 3. Esteva FJ, Sahin AA, Cristofanilli M et al (2002) Molecular prognostic factors for breast cancer metastasis and survival. Semin Radiat Oncol 12: Early Breast Cancer Trialists' Collaborative Group (EBCTCG) (2005) Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 365: Vogel CL, Cobleigh MA, Tripathy D et al (2002) Efficacy and safety of trastuzumab as a single agent in first-line treatment of HER2-overexpressing metastatic breast cancer. J Clin Oncol 20: Sauter G, Lee J, Bartlett JM et al (2009) Guidelines for human epidermal growth factor receptor 2 testing: biologic and methodologic considerations. J Clin Oncol 27: Bauer KR, Brown M, Cress RD et al (2007) Descriptive analysis of estrogen receptor (ER)- negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: a populationbased study from the California Cancer Registry. Cancer 109: Guarneri V, Broglio K, Kau SW et al (2006) Prognostic value of pathologic complete response after primary chemotherapy in relation to hormone receptor status and other factors. J Clin Oncol 24: Wolff AC, Hammond ME, Schwartz JN et al (2007) American Society of Clinical Oncology/ College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. Arch Pathol Lab Med 131: Wolff AC, Hammond ME, Schwartz JN et al (2007) American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol 25: Al-Dujaily EA, Al-Janabi AA, Pierscionek T, Yasseen AA (2008) High prevalence of HER-2/neu overexpression in female breast cancer among an Iraqi population exposed to depleted uranium. J Carcinog 7:8 12. Kitano Y, Umemura S, Ohbayashi H et al (2007) Assessment of a new anti-her2 monoclonal antibody, SV2-61gamma: a best concordance with HER2 FISH. Appl Immunohistochem Mol Morphol 15: Dybdal N, Leiberman G, Anderson S et al (2005) Determination of HER2 gene amplification by fluorescence in situ hybridization and concordance with the clinical trials immunohistochemical assay in women with metastatic breast cancer evaluated for treatment with trastuzumab. Breast Cancer Res Treat 93: Tsuda H, Akiyama F, Terasaki H et al (2001) Detection of HER-2/neu (c-erb B-2) DNA amplification in primary breast carcinoma. Interobserver reproducibility and correlation with immunohistochemical HER-2 overexpression. Cancer 92: Cuadros M, Talavera P, López FJ et al (2010) Realtime RT-PCR analysis for evaluating the Her2/neu status in breast cancer. Pathobiology 77: Merola R, Mottolese M, Orlandi G et al (2006) Analysis of aneusomy level and HER-2 gene copy number and their effect on amplification rate in breast cancer specimens read as 2+ in immunohistochemical analysis. Eur J Cancer 42: Downs-Kelly E, Yoder BJ, Stoler M et al (2005) The influence of polysomy 17 on HER2 gene and protein expression in adenocarcinoma of the breast: a fluorescent in situ hybridization, immunohistochemical, and isotopic mrna in situ hybridization study. Am J Surg Pathol 29: Hyun CL, Lee HE, Kim KS et al (2008) The effect of chromosome 17 polysomy on HER-2/neu status in breast cancer. J Clin Pathol 61: Zhao J, Wu R, Au A et al (2002) Determination of HER2 gene amplification by chromogenic in situ hybridization (CISH) in archival breast carcinoma. Mod Pathol 15: Chibon F, de Mascarel I, Sierankowski G et al (2009) Prediction of HER2 gene status in Her2 2+ invasive breast cancer: a study of 108 cases comparing ASCO/CAP and FDA recommendations. Mod Pathol 22: Vanden Bempt I, Van Loo P, Drijkoningen M et al (2008) Polysomy 17 in breast cancer: clinicopathologic significance and impact on HER-2 testing. Clin Oncol 26: Lal P, Salazar PA, Ladanyi M, Chen B (2003) Impact of polysomy 17 on HER-2/neu immunohistochemistry in breast carcinomas without HER-2/ neu gene amplification. Mol Diagn 5: Moriki T, Takahashi T, Ueta S et al (2004) Hormone receptor status and HER2/neu overexpression determined by automated immunostainer on routinely fixed cytologic specimens from breast carcinoma: correlation with histologic sections determinations and diagnostic pitfalls. Diagn Cytopathol 30: