Masaoka Staging is of Prognostic Relevance in Type B3 / C Thymomas

Transcription

1 Masaoka Staging is of Prognostic Relevance in Type B3 / C Thymomas T. LUEBKE 1, K.T.E. BECKURTS 1, C. WICKENHAUSER 2, P. SCHNEIDER 1, A. H. HOELSCHER 1 and S.P. MOENIG 1 1 Department of Visceral and Vascular Surgery and 2 Institute of Pathology, University of Cologne, Joseph-Stelzmann-Straße 9, Cologne, Germany Abstract. Background: The purpose of this study was to correlate the Ki67 labelling index (LI) with the Masaoka classification and the new WHO-classification in type B3 / C thymomas. Patients and Methods: Fourteen patients with type B3 / C thymomas were evaluated, and archived specimens were histologically reclassified according to Masaoka staging, the new WHO classification and the Ki-67 LI in a retrospective analysis. Results: Four patients presented with Masaoka stage II disease (all WHO-type B3), 1 patient had stage III (WHOtype C), 6 stage IVa (3 WHO-type B3 and 3 WHO-type C), and another 3 patients stage IVb (all WHO-type C). The statistical analysis revealed a significant correlation between Masaoka staging and Ki-67 LI (II, III vs. IV; p=0.007). As well, WHO-classification correlated significantly with Ki-67 LI (B3 vs. C; p=0.015). Masaoka staging (II, III vs. IV) correlated significantly with survival status (p=0.0237) in patients with type B3 / C thymoma whereas WHOclassification did not (p=0.3266). Between survivors and nonsurvivors there was no statistically significant correlation concerning Ki-67 LI (p=0.075). Conclusion: Our study indicated that the Masaoka staging system is of prognostic relevance in type B3 / C thymomas. Thymomas represent 5 10% of all mediastinal tumours. Thymomas are derived from the epithelial cells of the thymus and constitute a group of tumours with a very heterogeneous histology and varied biological behaviour. Type B3 and C thymomas (s) are characterized by malignant cytological findings (1) and a Correspondence to: Stefan P. Moenig, MD, Department of Visceral and Vascular Surgery, University of Cologne, Joseph-Stelzmann- Straße 9, Cologne, Germany. Tel: , Fax: , Stefan.Moenig@medizin.uni-koeln.de Key Words: Thymoma,, thymoma classification, Ki-67 LI. clinical course determined by early and frequent metastases and poor survival (2, 3). They account for less than 0.06% of thymic neoplasms (4). Because of the rare incidence of type B3 and C thymomas, so far there have been no greater prospective randomized studies defining prognostic variables or testing the efficacy of different treatment modalities for patients with these neoplasms (5 11). In 1999, a WHO committee (headed by J. Rosai) achieved consensus to use the morphological criteria, though not the nomenclature, of the "histogenetic classification" by Müller- Hermelink as the morphological basis of the new WHOapproved Histological Classification of Tumours of the Thymus. Thymomas are now labelled as type A, AB, B1, B2, B3 and C thymomas. Type B3 thymomas (welldifferentiated s) are equivalent to category I malignant thymomas (Figure 3). Type C thymomas were previously called "s" (12) (Figure 4). They should be subclassified following the rules of general tumour pathology. Type B3 / C thymomas are currently staged by the Masaoka classification, which takes into consideration macroscopic and histological findings and the grade of invasion in relation to the tumour capsule (3) and they are classified histologically according to the new WHO classification. Cases of type B3 / C thymomas should be reported using the same histological classification and staging to develop optimized therapies for individual patients considering both tumour stage and histology. In spite of numerous previous studies about expression of oncogene proteins in human cancers, very little is known about the biological characteristics of (13-21). Ki67 is a widely used proliferation marker which has been proved for diagnostic and prognostic relevance in various solid tumours, but which has rarely been investigated in type B3 / C thymomas (22-25). Therefore, the purpose of this study was to correlate the Ki67 labelling index (LI) in type B3 / C thymomas with the Masaoka classification and the new WHO-classification for thymomas /2004 $

2 Figure 1. Survival curves according to the World Health Organization histological classificaton system among patients with type B3 / C thymomas. WHO: World Health Organization Figure 2. Survival curves according to the Masaoka staging system among patients with type B3 / C thymomas. Masaoka: Masaoka stage Patients and Methods The criteria for diagnosis of type C thymoma was an epithelial cell thymic tumour with clear-cut cytological features of malignancy. Type C thymomas included tumours defined as type II malignant thymomas, as described by Rosai and colleagues (26); type B3 thymomas included well-differentiated s, as described by Müller-Hermelink et al. (25). Clinical and histopathological staging was performed according to the criteria published by Masaoka and colleagues (1). The type B3 / C thymomas were classified and staged by two independent pathologists, who had no knowledge of the outcome of the patients. Ki67 immunostaining. Three-Ìm-thick sections of formalin-fixed paraffin-embedded blocks were cut. The slides were pre-treated with microwave heating in 10 mm citric acid buffer (ph 6.0) for 4 x 5 min. Immunostaining with monoclonal Ki67 antibody (final dilution: 1 : 50, Dako, Copenhagen, Denmark) was applied. The slides were incubated with primary antibodies at 4ÆC overnight. The bound Ki67 antibody was detected with streptavidin-biotin complex (Dako) using 3,3 -diaminobenzidine tetrahydrochloride as chromogen, which yielded a brown product (Figure 5). The slides were examined without knowledge of the patients data. At least 1000 epithelial cells were counted. Any nuclear staining of Ki67, regardless of its intensity, was considered to be positive. The LI was expressed as a percentage of positive nuclei. The Ki67 LI was correlated with the Masaoka classification and the new WHO-classification. Follow-up data were available for all patients. The minimal postoperative observation time (follow-up) of all patients was 2.3 years and extended up to 9.3 years (median follow-up: 5.8 years). Statistical analysis. The statistical difference of the average value was examined by Student s t-test. The survival rate was calculated with the method of Kaplan and Meier. Kaplan-Meier survival curves were generated. The statistical difference in survival was determined with the log-rank test. The correlation between Masaoka stage and WHO-type, respectively, and Ki67 LI was calculated according to Spearman. Statistical analyses were performed with the commercially available personal computer program SPSS (SPSS, Inc., Chicago, IL, USA). Significance was defined as a p value less than Results Demographics. The analysis included 14 patients 11 men and 3 women with a mean age of 47.6 years (range: 32 82) undergoing treatment, including surgery at the Department of Surgery, University of Cologne, Germany, from January 1, 1990 to December 31, None of the patients showed associated myasthenia gravis (MG) and 4 of the 7 patients with a type B3 thymoma had a paraneoplastic syndrome. Surgical procedure. All tumours were resected through median sternotomy. Combined resection of the mediastinal pleura and lung was conducted in two patients. The pericardium and the left upper pulmonary lobe or the right upper pulmonary lobe, respectively, were resected in another two patients. The pleura, left upper pulmonary lobe, left sternoclavicular joint, left clavicula and the left brachiocephalic vein were resected in one patient, and the mediastinal pleura, pericardium, and the lung in yet another. Complete resection was done in 9 patients. Incomplete resection was done in 5 patients (Table I). Staging and classification. The number of patients according to the Masaoka staging system (Table II) with each 4114

3 Luebke et al: Type C / B3 Thymoma and Masaoka Staging Figure 3. Type B3 thymoma (survey). Figure 5. Type C thymoma, Ki67 immunostaining, original magnification x25. Radiotherapy and/or chemotherapy. Postoperative external radiation therapy was offered to all patients with Masaoka stage II disease and the patient with stage III disease, to 5 patients with stage IVa disease and 2 patients with stage IVb disease. One patient with stage IVa disease obtained no radiation- or chemotherapy and another patient with stage IVb disease had only immuno-chemotherapy. The postoperative radiotherapy varied in doses between 40 and 60 Gy. The survival rate of patients who received postoperative radiotherapy was not statistically significantly different from that of non-radiated patients. Figure 4. Type C thymoma (survey). histological tumour type is denoted in Table III. The histological tumour type in the 7 patients with type C thymoma was in 3 cases an epidermoid non-keratinizing carcinoma, in 1 case an epidermoid and in 3 cases a lymphoepithelioma-like carcinoma. Morbidity and mortality. The 30-day mortality rate was 0%. With regards to postoperative complications, we had one case of postoperative bleeding, one pneumothorax and two cases of pleural effusion. Major cardiovascular, pulmonary, gastrointestinal and technical morbidity did not occur. Follow-up. The median follow-up was months ( months). Of 9 patients with stage IV disease, 4 died during follow-up (44%). All patients in stage II and III disease survived during follow-up. (overall mortality rate: 26.6%). Only 1 out of 7 patients with WHO-type C thymoma survived during follow-up, whereas 4 of the 7 patients with WHO-type B3 thymoma survived in the same period of time (Table IV). Ki67 LI. In stage II disease the mean LI was 3.75% (2 5%). The stage III showed an LI of 10%. Stage IVa disease was characterized by a mean LI of 14.4% (5 30%) and stage IVb had a mean LI of 18.3% (8 35%). The statistical analysis revealed a significant correlation between Masaoka staging and Ki-67 LI (II, III vs. IV; p=0.007). As well, the WHO-classification correlated significantly with Ki-67 LI (B3 vs. C; p=0.015). Masaoka staging (II, III vs. IV) correlated significantly with survival status (p=0.0237) in patients with type B3 / C thymoma, whereas the WHO-classification did not (p=0.3266). Between survivors and non-survivors there was no 4115

4 Table I. Patient characteristics. Age/ R / combined RTX/ Masa Histology WHO-typ Ki-67 LI (%) Survival (months) Rec Gender resection CTX M 52 R0 RTX III epidermoid non- C , alive N M 53 R1 / pleura, 46 Gy IVb epidermoid non- C d Y pericardium, lung M 49 R2 RTX IVa epidermoid C d Y F 37 R1 / pleura, lung, RTX IVb lymphoepithelioma-like C d Y sternoclavicular joint, carcinoma clavicula, BCV M 40 R0 RTX IVa epidermoid non- C 5 45 d Y M 82 R1 RTX IVa lymphoepithelioma- C 9 33, d Y like carcinoma M 56 R1 / pericardium, CTX IVb lymphoepithelioma- C d Y lung like carcinoma M 32 R0 46 Gy IVa well-differentiated B d N M 52 R0 / pericardium, 61,2 Gy IVa well-differentiated B d U lung F 53 R IVa well-differentiated B d U M 53 R0 RTX II well-differentiated B , alive N M 37 R0 RTX II well-differentiated B , alive N F 34 R0 RTX II well-differentiated B , alive U M 37 R0 / pleura, lung RTX II well-differentiated B , alive N M, male; F, female; RTX, radiotherapy; CTX, chemotherapy; Masa, Masaoka stage; WHO, World Health Organization; LI, label index; Rec, recurrence; Y, yes; N, no; U, unknown; Gy, Gray; D, day; BCV,.brachiocephalic vein Table II. Masaoka classification system. Stage Macroscopic and histological findings I Macroscopic completely encapsuled, histologic no infiltration of the capsule II Macroscopic invasion of the parathymic fat tissue or of the pleura mediastinalis or microscopic invasion of the capsule III Macroscopic invasion of the adjacent organs IVa Diffuse infiltration of pleura or pericardium IVb Lymphogenic or hematogenic metastases Table III. Masaoka stage with reference to the World Health Organization histological classification system. WHO tumour type Stage B3 C (%) II III IVa IVb WHO: World Health Organization statistically significant correlation concerning Ki-67 LI (p=0.075). Survival curves according to the WHO histological classification system are shown in Figure 1 and the survival curves according to the staging system of Masaoka are shown in Figure 2. Discussion Thymic carcinomas are a rare type of malignant tumours, comprising only 0.06% of thymic neoplasms (4). They have a poorer prognosis and are more invasive than oridinary 4116

5 Luebke et al: Type C / B3 Thymoma and Masaoka Staging Table IV. Tumour recurrence with reference to the World Health Organization classification system. WHO tumour type Status B3 C R0 7 2 R1 0 4 R2 0 1 Died of tumour 3 6 Recurrence of tumour 0* 6 WHO: World Health Organization thymoma (3, 26 29). Thus, prognostic statements can only be made on the basis of greater patient collectives. Blumberg et al. (3) reported 43 patients with s over a 45- year period, which is one of the largest single institutional reports to date. He found that prognosis for patients with was dependent solely on tumour invasion of the innominate vessels. The five-year survival rate is quoted at 31% to 50% (25, 26, 29). Schneider et al. (30) evaluated 21 cases of type C thymomas where the mean 10-year survival rate of R0-resected cases was 51.4%±20.4%. Previous studies showed that a complete resection suggests a favourable result (3, 31, 32). On the other hand, Hsu et al. (28) showed a lack of efficacy of complete resection, even though they could demonstrate longer median survival. The staging system of Masaoka et al. (1), originally described for thymomas, has been adopted for s but lacks universal support (3) (Table II). The prognostic usefulness of the Masaoka classification for patients with has not yet been fully investigated (30). A recent clinical study of 1320 patients with thymic epithelial tumours by Kondo et al. (33) showed that the Masaoka clinical stage was an excellent predictor of the prognosis not only of thymoma but also of, but not thymic carcinoid. In general, thymic carcinomas can occur throughout all the Masaoka stages. However, the heterogeneous group of type C thymoma are frequently detected in advanced stages III and IV (30). In discordance with Blumberg et al., other authors (2, 30, 34-40) found that the Masaoka and Müller-Hermelink classifications were significant and independent prognostic indicators of overall and disease-free survival in patients with epithelial thymic tumours. In our study, Masaoka staging (II, III vs. IV) correlated significantly with survival status (p=0.0237) in patients with type B3 / C thymomas, whereas WHOclassification did not (p=0.3266). In a clinicopathological study by Chen et al. (39), WHO histological subtype was an independent prognosis factor but only in Masaoka stage I and II thymomas. To date, only two studies exist that examine the clinical features as well as postoperative survival of patients with type B3 / C thymomas with reference to the WHO histological classification (39, 41). Okumura et al. (35, 41) adopted the WHO classification in his evaluation only for patients with type A, AB, B1, B2 and B3 tumours but not with type C thymomas. He found that the WHO histological classification system was a significant independent prognostic factor in type A, AB, B1, B2 and B3 thymomas. Chen et al. (39) could demonstrate that type B2, B3 and C thymomas had a worse prognosis than type A, AB and B1 thymomas. There are many studies correlating Ki67 expression with clinical outcome. Even after taking into consideration the inevitable publication bias in favour of studies showing significance, it is clear that Ki67 is of prognostic value for many types of malignant tumours. The effectiveness of using Ki67 antibody as a tool to measure proliferative activity has been widely investigated in a variety of tumours, but rarely in thymoma, and in (22). The Ki67 murine IgG1 monoclonal antibody recognizes a nonhistone protein of the nuclear matrix of greater than 300 kd that is preferentially expressed during the late G 1 -, S-, G 2 - and M- phases of the cell cycle. Cells in G 0 - and early G 1 -phases consistently lack antigen expression (42). Salakou et al. (43) evaluated Ki67 expression in two s. A higher cell proliferation rate was found in cases of s compared to thymomas and hyperplasia. However, they did not evaluate Masaoka staging or WHO-classification. They showed, as well, that Ki67 expression was directly correlated to the size of the thymus and also the weight of the organ. Pan et al. (23) showed a statistically significant positive correlation between Ki67 LI and stage of thymomas based on both Lattes- Bernatz and Müller-Hermelink-Kirchner classifications. In detail, there were statistically significant differences of LIs between stage I and stage III and between stage I and stage IV thymomas. The stage III/IV tumours, predominantly epithelial thymoma and WDTC, had higher Ki67 LIs (23). The study of Pan et al. (23) seemed to demonstrate a trend that patients with thymomas of higher LIs had slightly worse survival, but this difference lost significance in both univariate and multivariate survival analysis, possibly due to the small number of cases. In a previous study with eight patients, Tomita et al. (39) could show that the expression rate of Ki67 was higher in s (7.01±6.37 5) than in thymomas (2.57±4.52%), indicating the aggressiveness of thymic carcinoma. In our analysis, there was a highly significant correlation between Masaoka staging and Ki67 LI (II, III vs. IV; p=0.007). Between survivors and non-survivors there was no statistical significant correlation concerning Ki67 LI (p=0.075), although this value indicated a trend towards an association between a lower Ki67 LI and longer patient survival. Since Ki67 is related to tumour proliferation, one can suggest that higher expression rates of this protein might indicate the aggressiveness of. The negative correlation with Ki67 may be caused by the fact that the expression of Ki67 varies greatly during the cell cycle. 4117

6 Furthermore, the negative correlation may be caused by the small sample size of a tumour entity which is known to have great variations in histology and growth pattern even within a single thymoma. However, tissue growth (especially during tumour development) is not only a result of cell proliferation but also of enhanced cell survival (through inhibition of apoptosis) or it results from a combination of both mechanisms (43, 44). Tumour cells with high proliferative activity might be those having an imbalance in proapoptotic and antiapoptotic signals making them more vulnerable to apoptotic death, whereas tumour cells with low proliferation might be those being more in balance and which therefore will survive (45). Additionally, growth rates measured by Ki67 do not reflect true proliferation because the time required for cell cycle completion cannot be measured. Real tumour growth is the result of a balance of proliferation and cell loss. These features are not taken into account by Ki67 quantitation. Perhaps, the difference in Ki67 LI of type B3 and type C thymomas reflects a true difference in these two tumour types. However, we should interpret the results from the series with caution because, in cases of small sample size, several parameters influence the prognostic significance, for example histological tumour type, the macroscopic and microscopic pattern of tumour infiltration, the proliferative activity of the tumour, the degree of surgical resection (R0, R1, R2) and application of radiotherapy and / or chemotherapy. Taking into consideration these restrictions, our results seem to suggest that, in thymic carcinoma, the proliferative activity measured by Ki67 LI has clinical relevance. Additionally, the immunohistochemical study performed by Tomita et al. (31) revealed that the expression rates of p53, bcl-2, CEA and EMA in thymic carcinomas were higher than those in thymomas which might be implicated in the malignant potential of. On the other hand, the two cases of s presented by Salakou et al. (43) had fewer cells positive for bcl- 2 protein or bcl-2 mrna. In this study, there was no statistically significant correlation between bcl-2 expression and thymus weight and size and proliferation activity. Furthermore, in the study by Tomita et al. (31) 75% of the s were positive for nm23-h1 expression. The expression of nm23-h1 is thought to play a specific biological role in suppression of tumour metastasis (46). They (47) also showed that microvessel density and expression of VEGF and FGF-2 in s were higher than those in thymomas. The authors concluded that angiogenesis might also be related to the aggressiveness of. In conclusion, the outcome of treatment must be investigated using the same histological typing and staging. We found that the Masaoka staging system may be of prognostic relevance in (thymoma type B3 / C) in certain stages, since all patients who have died so far during follow-up suffered from stage IV disease according to Masaoka. Herewith, this study is restricted only to type B3 and type C thymomas. The immunohistochemical results concerning Ki67 may indicate the malignant potential of. Acknowledgements The authors gratefully appreciate the assistance of A. Marx, MD, Department of Pathology, University of Würzburg, Germany, in reviewing selected histological slides. References 1 Masaoka A, Monden Y, Nakahara K and Tanioka T: Followup study of thymomas with special reference to their clinical stages. Cancer 48: , Quintanilla-Martinez L, Wilkins EW Jr, Choi N, Efird J, Hug E and Harris NL: Thymoma. Histologic subclassification is an independent prognostic factor. Cancer 74: , Blumberg D, Burt ME, Bains MS, Downey RJ, Martini N, Rusch V and Ginsberg RJ: Thymic carcinoma: current staging does not predict prognosis. J Thorac Cardiovasc Surg 115: 303-8; discussion 308-9, Wick MR, Scheithauer BW, Weiland LH and Bernatz PE: Primary s. Am J Surg Pathol 6: , Akoum R, Brihi E, Chammas S and Abigerges D: Results of radiation therapy for thymoma based on a review of 27 patients. Mol Immunol 39: , Gawrychowski J, Rokicki M, Gabriel A, Lackowska B and Czyzewski D: Thymoma - the usefulness of some prognostic factors for diagnosis and surgical treatment. Eur J Surg Oncol 26: 203-8, Giaccone G: Treatment of thymoma and. Ann Oncol 11: 245-6, Sonobe M, Nakagawa M, Ichinose M, Ikegami N, Nagasawa M and Shindo T: Thymoma. Analysis of prognostic factors. Jpn J Thorac Cardiovasc Surg 49: 35-41, Venuta F, Rendina EA, Longo F, De Giacomo T, Anile M, Mercadante E, Ventura L, Osti MF, Francioni F and Coloni GF: Long-term outcome after multimodality treatment for stage III thymic tumors. Ann Thorac Surg 76: ; discussion 1872, Hernandez-Ilizaliturri FJ, Tan D, Cipolla D, Connolly G, Debb G and Ramnath N: Multimodality therapy for (TCA): results of a 30-year single-institution experience. Am J Clin Oncol 27: 68-72, Daniele O and Fornasiero A: Ifosfamide in thymic neoplasms. Oncology 65: 44-5, Muller-Hermelink HK and Marx A: Thymoma. Curr Opin Oncol 12: , Tateyama H, Eimoto T, Tada T, Hattori H, Murase T and Takino H: Immunoreactivity of a new CD5 antibody with normal epithelium and malignant tumors including thymic carcinoma. Am J Clin Pathol 111: , Tateyama H, Eimoto T, Tada T, Inagaki H, Hattori H and Takino H: Apoptosis, bcl-2 protein, and Fas antigen in thymic epithelial tumors. Mod Pathol 10: , Tateyama H, Eimoto T, Tada T, Mizuno T, Inagaki H, Hata A, Sasaki M and Masaoka A: p53 protein expression and p53 gene mutation in thymic epithelial tumors. An immunohistochemical and DNA sequencing study. Am J Clin Pathol 104: ,

7 Luebke et al: Type C / B3 Thymoma and Masaoka Staging 16 Hayashi Y, Ishii N, Obayashi C, Jinnai K, Hanioka K, Imai Y and Itoh H: Thymoma: tumour type related to expression of epidermal growth factor (EGF), EGF-receptor, p53, v-erb B and ras p21. Virchows Arch 426: 43-50, Chen FF, Yan JJ, Jin YT and Su IJ: Detection of bcl-2 and p53 in thymoma: expression of bcl-2 as a reliable marker of tumor aggressiveness. Hum Pathol 27: , Chen FF, Yan JJ, Chang KC, Lai WW, Chen RM and Jin YT: Immunohistochemical localization of Mcl-1 and bcl-2 proteins in thymic epithelial tumours. Histopathology 29: 541-7, Hino N, Kondo K, Miyoshi T, Uyama T and Monden Y: High frequency of p53 protein expression in but not in thymoma. Br J Cancer 76: , Fukai I, Masaoka A, Hashimoto T, Yamakawa Y, Mizuno T, Tanamura O, Hirokawa K and Ueda R: An immunohistologic study of the epithelial components of 81 cases of thymoma. Cancer 69: , Fukai I, Masaoka A, Hashimoto T, Yamakawa Y, Niwa H, Kiriyama M and Eimoto T: Differential diagnosis of thymic carcinoma and lung carcinoma with the use of antibodies to cytokeratins. J Thorac Cardiovasc Surg 110: , Baldus SE, Zirbes TK, Weingarten M, Fromm S, Glossmann J, Hanisch FG, Monig SP, Schroder W, Flucke U, Thiele J, Holscher AH and Dienes HP: Increased galectin-3 expression in gastric cancer: correlations with histopathological subtypes, galactosylated antigens and tumor cell proliferation. Tumour Biol 21: , Pan CC, Ho DM, Chen WY, Huang CW and Chiang H: Ki67 labelling index correlates with stage and histology but not significantly with prognosis in thymoma. Histopathology 33: 453-8, Gilhus NE, Jones M, Turley H, Gatter KC, Nagvekar N, Newsom-Davis J and Willcox N: Oncogene proteins and proliferation antigens in thymomas: increased expression of epidermal growth factor receptor and Ki67 antigen. J Clin Pathol 48: , Kirchner T, Schalke B, Marx A and Muller-Hermelink HK: Evaluation of prognostic features in thymic epithelial tumors. Thymus 14: , Suster S and Rosai J: Thymic carcinoma. A clinicopathologic study of 60 cases. Cancer 67: , Chung DA: Thymic carcinoma-analysis of nineteen clinicopathological studies. Thorac Cardiovasc Surg 48: 114-9, Hsu CP, Chen CY, Chen CL, Lin CT, Hsu NY, Wang JH and Wang PY: Thymic carcinoma. Ten years' experience in twenty patients. J Thorac Cardiovasc Surg 107: , Truong LD, Mody DR, Cagle PT, Jackson-York GL, Schwartz MR and Wheeler TM: Thymic carcinoma. A clinicopathologic study of 13 cases. Am J Surg Pathol 14: , Schneider PM, Fellbaum C, Fink U, Bollschweiler E and Prauer HW: Prognostic importance of histomorphologic subclassification for epithelial thymic tumors. Ann Surg Oncol 4: 46-56, Tomita M, Matsuzaki Y and Onitsuka T: Relationship between expression of cancer-related proteins and tumor invasiveness in thymoma. Eur J Cardiothorac Surg 21: 596, Liu W, Tong T, Ji Z and Zhang Z: Long-term prognostic analysis of thymectomized patients with myasthenia gravis. Chin Med J (Engl) 115: 235-7, Kondo K and Monden Y: Therapy for thymic epithelial tumors: a clinical study of 1,320 patients from Japan. Ann Thorac Surg 76: ; discussion 884-5, Blumberg D, Port JL, Weksler B, Delgado R, Rosai J, Bains MS, Ginsberg RJ, Martini N, McCormack PM, Rusch V et al: Thymoma: a multivariate analysis of factors predicting survival. Ann Thorac Surg 60: ; discussion 914, Okumura M, Ohta M, Tateyama H, Nakagawa K, Matsumura A, Maeda H, Tada H, Eimoto T, Matsuda H and Masaoka A: The World Health Organization histologic classification system reflects the oncologic behavior of thymoma: a clinical study of 273 patients. Cancer 94: , Lardinois D, Rechsteiner R, Lang RH, Gugger M, Betticher D, von Briel C, Krueger T and Ris HB: Prognostic relevance of Masaoka and Muller-Hermelink classification in patients with thymic tumors. Ann Thorac Surg 69: , Pescarmona E, Rendina EA, Venuta F, D'Arcangelo E, Pagani M, Ricci C, Ruco LP and Baroni CD: Analysis of prognostic factors and clinicopathological staging of thymoma. Ann Thorac Surg 50: 534-8, Kuo TT and Lo SK: Thymoma: a study of the pathologic classification of 71 cases with evaluation of the Muller- Hermelink system. Hum Pathol 24: , Chen G, Marx A, Wen-Hu C, Yong J, Puppe B, Stroebel P and Mueller-Hermelink HK: New WHO histologic classification predicts prognosis of thymic epithelial tumors: a clinicopathologic study of 200 thymoma cases from China. Cancer 95: 420-9, Matsushima S, Yamamoto H, Egami K, Suzuki S and Tanaka S: Evaluation of the prognostic factors after thymoma resection. Int Surg 86: 103-6, Okumura M, Ohta M, Miyoshi S, Mori T, Yasumitsu T, Nakahara K, Iuchi K, Tada H, Maeda H and Matsuda H: Oncological significance of WHO histological thymoma classification. A clinical study based on 286 patients. Jpn J Thorac Cardiovasc Surg 50: , Gerdes J: Ki-67 and other proliferation markers useful for immunohistological diagnostic and prognostic evaluations in human malignancies. Semin Cancer Biol 1: , Salakou S, Tsamandas AC, Bonikos DS, Papapetropoulos T and Dougenis D: The potential role of bcl-2, bax, and Ki67 expression in thymus of patients with myasthenia gravis, and their correlation with clinicopathologic parameters. Eur J Cardiothorac Surg 20: , Hannun YA: Apoptosis and the dilemma of cancer chemotherapy. Blood 89: , Evan G: Cancer - a matter of life and cell death. Int J Cancer 71: , Hartsough MT, Clare SE, Mair M, Elkahloun AG, Sgroi D, Osborne CK, Clark G and Steeg PS: Elevation of breast carcinoma Nm23-H1 metastasis suppressor gene expression and reduced motility by DNA methylation inhibition. Cancer Res 61: , Tomita M, Matsuzaki Y, Edagawa M, Maeda M, Shimizu T, Hara M and Onitsuka T: Correlation between tumor angiogenesis and invasiveness in thymic epithelial tumors. J Thorac Cardiovasc Surg 124: 493-8, Received May 26, 2004 Revised August 9, 2004 Accepted October 14,