_, AMERICAN SOCIETY OF CLINICAL PATHOLOGISTS Commissi on on Continuing Education Council on Anatomic Pathology 1994 SPRING ANATOMIC PATHOLOGY SLIDE SEMINAR TUMORS AND TUMOR-LIKE CONDITIONS OF THE MEDIASTINUM The Sheraton Seattle Hotel and Towers Thursday, April 14, 8:30AM-12 :00NOON CASE HISTORIES PRESENTERS: JUAN ROSAI, M.D. Chairman, Department of Pathology Memorial Sloan-Kettering Cancer Center Professor of Pathology Cornell University Medical School New York, New York DAVID S. KLIMSTRA, M.D. Assistant Attending Pathologist Memori al Sloan-Kettering Cancer Center Assistant Professor of Pathology Cornell University Medical School New York, New York MODERATOR: Y AO SHI FU, M.D. Professor of Paholo. gy and Chief of Surgical Pathology University of California at Los Angeles Medical Center Los Angeles. California PLEASE BRING THIS PROTOCOL TO THE SEMINAR
Transcript
_,
AMERICAN SOCIETY OF CLINICAL PATHOLOGISTS Commission on Continuing Education
Council on Anatomic Pathology
1994 SPRING ANATOMIC PATHOLOGY SLIDE SEMINAR
TUMORS AND TUMOR-LIKE CONDITIONS OF THE MEDIASTINUM
The Sheraton Seattle Hotel and Towers Thursday, April 14, 8:30AM-12:00NOON
CASE HISTORIES
PRESENTERS:
JUAN ROSAI, M.D. Chairman, Department of Pathology
Memorial Sloan-Kettering Cancer Center Professor of Pathology
Cornell University Medical School New York, New York
DAVID S. KLIMSTRA, M.D. Assistant Attending Pathologist
Memorial Sloan-Kettering Cancer Center Assistant Professor of Pathology
Cornell University Medical School New York, New York
MODERATOR:
Y AO SHI FU, M.D. Professor of Paholo.gy and Chief of Surgical Pathology
University of California at Los Angeles Medical Center Los Angeles. California
PLEASE BRING THIS PROTOCOL TO THE SEMINAR
Case Histories
Case 1: A 69 year old woman presented with shortness of breath. She was initially treated for pneumonia; however, a chest radiograph revealed a tumor in the right upper lung zone. Mediastinoscopy showed the tumor to arise from the anterior mediastinum. A frozen section diagnosis of "neuroendocrine tumor" was made, and the mass was excised. Grossly the 3cm tumor appeared encapsM~ated. On cut section the tissue was pink to tan and fleshy, with bands of white fibrous tissue dividing it into lobules.
Case 2: A 30 year old female first presented with cough and chest pain radiating to the back in 1986. A large anterior mediastinal mass was resected following a brief course of radiation therapy. The patient was well until 7/92 when she developed alopecia. A chest CT revealed local recurrence of tumor in the ante.rior mediastinum, which was re-excised. Grossly the tumor was firm and tan-white with a multinodular appearance. A portion of lung tissue was attached.
Case 3: A. 59 year old male with a nine year history of progressive spinal/ cerebellar degeneration of unclear etiology presented with shortness of breath. A chest radiograph revealed a 9cm left anterior mediastinal mass. A fine needle aspirate was non-diagnostic and resulted in a pneumothorax requiring chest tube placement. One month later the patient underwent median sternotomy with resection of the tumor. Grossly, the mass weighed 55 grams and measured 9x6.Sx3.5cm. Cut sections revealed a tan to yellow, gritty, infiltrative tumor with cystic cavities measuring up to 2cm in diameter.
Case 4.: A 42 year old male with a past history of excision of a parathyroid adenoma presented in 1990 with a sensation of "tightness" in the chest. There were no other symptoms. An anterior mediastinal mass was noted on chest radiograph. An open biopsy was performed and a diagnosis was established. The patient declined treatment for two years but then consented to resection when the mass was found to be increasing in size. Grossly, the tumor was lOcm in greatest diameter and covered by a thi."l capsule. The tissue was tan and soft with multiple small foci of necrosis and calcification.
Of note, the patient's brother had excisions of pa.rathyroid and pituitary adenomas as well as a pancreatic endocrine tumor, his mother died of a widely metastatic endocrine neoplasm.
Case 5: A 75 year old male was noted to have a slowly growing mass in the left posterior mediastinum five years prior to surgery. However, a marked increase in size was noted and the tumor was resected. The patient was asymptomatic and laboratory studies were within normal limits. Grossly the tumor measured 20cm and weighed 1340 gm. It was partially encapsulated and consisted of firm, nodular white tissue with myxoid areas.
Case 6: The patient is a 72 year old white female who was noted to have an incidental mediastinal mass on a chest radiograph performed prior to repair of a rectocele in 1990. The patient refused further evaluation of the mass at that time, but presented in 1992 with increasing shortness of breath and anterior chest pain. On cr, the tumor measured 16cm and involved the anterior mediastinum and right hemithorax. Two components to the mass were identified, one of which had the appearance of adipose tissue. The tumor was resected uneventfully. Grossly the tumor was circumscribed and soft, with a.reas of yellow, fatty tissue alternating with more fum, white-tan, lobulated areas.
Cue 7: A 62 year old man presented with a right supraclavicular mass which was found to be contiguous with a large anterior mediastinal mass. A biopsy was interpreted as neoplastic, possibly paraganglioma or thymoma. The tumor was resected in two pieces measuring 9cm and llcm and weighing 320gm in aggregate. The cut section revealed multinodular yellow soft tumors with central hemorrhage and necrosis.
Case 8: A 66 year old man underwent a coronary artery bypass operation several years before presentation. On follow-up he was found to have a Scm welldemarcated nodular mass exactly in the region when the pleura had been stripped to expose the internal mammary artery. Several1-2mm nodules were also noted on the pleura adjacent to the main mass. The remainder of the pleura was unremarkable. There was no history of asbestos exposure or pleural effusion. The tumor was excised. Grossly the mass was brown-tan and homogeneous. There was no hemorrhage or necrosis.
Case 9: A 56 year old man presented in 2/92 with 2 months of cough, fatigue and dysphagia. Chest radiograph revealed a 13x10cm left anterior mediastinal mass. Biopsy via mediastinoscopy showed ectatic blood vessels in a fibrotic stroma, and the diagnosis of cancerous hemangioma was made. The patient received 2160 rads of radiation with no response. In 6/92 a more substantial biopsy was taken. At that time, the tumor filled the left chest with extension to the diaphragm. Grossly it appeared pink to gray with areas of necrosis.
Case 10: An 18 year old male, previously healthy, presented in 8/91 with fevers, night sweats, chest pain, and dyspnea on exersion. A left anterior mediastinal mass was detected on CT scan. The HCG level was 479 and the AFP level was 6900. A fine needle aspirate was performed, and malignant cells consistent with germ cell tumor were found. The patient received six cycles of VIP chemotherapy which was completed on 12/13/91. The HCG and AFP levels returned to normal, but there was no radiographic change in the appearance of the tumor. On 1/ 20/92, a transverse sternotomy and bilateral thoracotomies were performed, with resection of the tumor. Grossly, the 320 gm tumor measured 10x9x6.3cm and was circumscribed, lobulated, and surrounded by a thin capsule. Cut sections revealed numerous small cystic spaces within a tan, hemorrhagic tumor with focal myxoid areas and calcifications. The cysts contained thick mucinous material.
Case 1 2 3 4 5 6 7 8 9
10
T umor and Tumor-Like Conditions of the Mediastinum
A 69 year old woman presented with shortness of breath. She was initially treated for pneumonia; however, a chest radiograph revealed a tumor in the right upper lung zone. Mediastinoscopy showed the tumor to arise from the anterior mediastinum. A frozen section diagnosis of "neuroendocrine tumor" was made, and the mass was excised.
GROSS DESCRIPTION:
The 3cm tumor appeared encapsulated. On cut section the tissue was pink to tan and fleshy, with bands of white fibrous tissue dividing it into lobules.
HJSTOLOGICAL FEATURES:
This mediastinal neoplasm represents a thymoma exhibiting a well-vascularized pattern and a profusion of rosettes. There is a great predomince of epithelial cells over the lymphocytes, the former having a shape ranging from round to oval to spindle. The tumor capsule is violated in several areas by tumor growth, and there are several nodules of tumor located immediately outside in the adjacent mediastinal fat.
COMMENT:
This thymoma is designated in the various classifications as spindle, rosette-making, or medollary.(3.S.6) Because of its capsular invasion, it should be regarded as a malignant thymoma type I and classified as stage ll (microscopic invasion into the capsule) in the Masaoka's system.(7) This translates as a T2NOMO in the recently proposed staging system for this neoplasm.<t2) In our opinion, the tumor represented by this case is a neoplasm of nonfunctional, post-mature thymic epithelial cells which do not match phenotypically either medullary or cortical cells, but rather those of the effete epithelial cells seen in the involuted thymus of adult life. This type of thymoma is characterized by the presence of rosettes, gland-like spaces, foci of storiform arrangement, and scantiness or absence of lymphocytes.< tO) These lymphocytes are usually of mature T -cell type, i.e., different phenotypically to those present in other thymomas, which are characterized by a phenotype corresponding to that of cortical thymocytes.(8) Nearly all cases of the thymoma type represented by the Seminar case are encapsulated (stage I) or only minimally invasive (stage m. <9> As a result the prognosis following excision is excellent.
Overall, about 70 to 80 percent of thymomas are encapsulated.(lO) Direct local invasion (first into the capsule and mediastinal fat, later into adjacent structures, such as lung) or implants into the pleural or pericardia! surfaces are considerably more common events in thymoma than distant metastases. When local invasion is extensive, it is already apparent to the surgeon at the time of the t.horncotomy. It is therefore important for the pathologist to know the surgical findings, while remembering that fibrous adhesions resulting from secondary inflammatory, necrosis, and multilocular cystic changes can give the surgeon the false impression that the tumor is invasive. The term "minimal invasion" refers to the findings in this case, i.e., complete capsular breaks or tumor islands in the mediastinal fat. Presence of large nerves surrounded by tumor also represent indirect evidence of invasion.
In regard to the rosenes which are so prominently displayed by this tumor, it is important to distinguish them from the rosette-like structures seen in thymic carcinoid. The Iauer, in contrast to those of thymoma, have a well-defined lumen and therefore represent evidence of glandular differentiation. The type of thymoma represented by the Seminar case is very r.u-ely associated
with myasthenia gravis or any other of the paraneoplastic syndromes that have been reponed with thymoma as a group. In regard to the classification scheme, we would like to point out that although this thymoma is designated in some classification schemes as "medullary" (implying a histogenetic relationship with the normal medulla of the thymus),(9) no consistent relation has yet been found berween the proposed types of thymoma (including this one) and the known phenotypeS of normal thymic epithelial cells.0.2.4,11) This criticism notwithstanding, credit should be given to the authors of recent publications of the subject for increasing our awareness of the statistical relationship between the cytoarchitectural features of thrrooma and behavior, particularly in regard to the fact that tumors composed exclusively or predommantly of bland-appearing spindle/oval cells are nearly always encapsulated and associated with an excellent prognosis, and that in tumors composed of round/polygonal epithelial cells, aggressiveness is directly related both to the numerical predominance of epithelial cells over the lymphocytes and to the presence and degree of atypia in those epithelial cells.(9)
DIAGNOSIS:
MAI.:.IGNANT THYMOMA TYPE I, SPINDLE ("MEDULLARY") TYPE
REFERENCES:
I. Eimoto T, Teshima K, Shirakusa T, Takeshita M, OkamUTa H, Naito H, Mitsui T, Kikuchi l: Heterogeneity of epithelial cells and reactive components in thymomas. An ultrastructural and immunochemical study. Ultrasrruct Patholl0:157-173, 1986.
2. Giraud F, Fabien N, Auger C. Girod C. Loire R, Monier JC: Human epithelial thymic tumours: Heterogeneity in immunostaining of epithelial cell markers and thymic honnones. Thymus 15:15-29, 1990.
3. Hammond EH, Flinner, RL: The diagnosis of thymoma: A review. Ultrastruct Pathol 15:419-438, 1991.
4. Komstein MJ, Hoxie JA, Levinson AI, Brooks JJ: Immunohistology of human thymomas. Arch Pathol Lab Med 109:460-463, 1985.
5. Levine GD. Rosai J: Thymic hyperplasia and neoplasia. A review of current concepts. Hum Pathol9:495-5 15, 1978.
6. Lewis IE, Wick MR, Schcithauer BW, Bematz PE, Taylor WF: Thymoma. A clinicopathologic review. Cancer 60:2727-2743, 1987.
7. Masaoka A, Monden Y, Nakahara K, Tanioka T: Follow-up study of thymomas with special reference to their clinical stages: Cancer 48:2485-2492, 1981.
8. Mokhtar N, Hsu S-M, Lad RP, Haynes BF, Jaffe ES: Thymoma. Lymphoid and epithelial components mirror the phenotype of normal thymus. Hum Pathol 15:378-384, 1984.
9. Quintanllla-Martinez L, Wilkins E, Ferry J, Harris N: Thymoma-morphologic subclassification correlates with invasiveness and immunohistologic features. A study of 122 cases. Hum Pathol 24:958-969, 1993.
I 0. Rosai J: The pathology of thymic neoplasia. International Academy of Pathology Monograph No. 29, Baltimore, 1987, Williams & Wilkins.
II. van der Kwast TH, van Vliet E, Cristen E, van Ewijk W, van der Heul RO: An immunohistologic study of the epithelial and lymphoid components of six thymomas. Hum Pathol l 6:1001- l008, 1985.
12. Yamakawa Y, Masaoka A, Hashimoto T, Niwa H. Mizuno T, Fuji Y, Nakahara K: A tentative tumor-node metastasis classification of thymoma_ Cancer 68: 1984-1987, 1991.
CASE #2
aJ.NICAL HISTORY:
A 30 year old female first presented with cough and chest pain radiating to the back in 1986. A large anterior mediastinal mass was resected following a brief course of radiation therapy. The patient was well until 7/92 when she developed alopecia A chest cr revealed local recurrence of tumor in the anterior mediastinum, which was re-excised.
GROSS DESCRIPTION:
The tumor was finn and tan-white, with a multinodular appearance. A portion of lung tissue was attached.
HISTOLOGICAL FEATURES:
This thymic neoplasm is clearly malignant by virtue of the fact that it is growing outside its original confines and extending into lung. However, it retains the classical cytoarchitectural features of a lhymoma, and in all likelihood the lymphocytes present in it are COl positive, immature cortical thymocytes. The neoplastic epithelial cells have in areas the typical appearance of those described in the thymoma type variously designated as epithelial or cortical. In others areas, they show marked cytologic atypia, associated with squamous differentiation and mitotic activity.
COMME!IIT:
The interpretation and description of this thymic neoplasm might create some problems for those authors who like to draw a sharp separation between thymoma and thymic carcinoma, because it would be regarded as a thymoma on the basis of the architectural features and as a thymic carcinoma on the basis of the cytologic atypia.(3) It seems to us lhat this case is a perfect demonstration of the existence of thymic neoplasms that fall in between those two extremes. (2.6) In this regard, the recent proposal to add the category of "well-differentiated thymic carcinoma" for a tumor type that is included among the thymo!JUIS in all publications on the subject (a fact that may not be immediately apparent to all readers) may not be particularly useful. (1.5) From the point of view of staging, the present neoplasm corresponds to a stage ill in Masaoka's system (gross invasion intO a neighboring organ)C4) and as a T3NOMO in the recently proposed TNM staging system.(9) As already stated, direct local invasion or implants are considerably more common events in thymoma than distant metastases. The latter, which are exceptional, have been documented in mediastinal and cervical lymph nodes, lung, liver, bone (particularly spine), ovary (we have seen three examples), and other sites. Usually they develop months or years after the invasive thymoma has been detected and treated; sometimes they are noted at presentation, and exceptionally represent the first clinical manifestation of the disease.
The treatment of malignant thymoma associated with gross invasion or implants usually consists in a combination of excision and radiation therapy. When distant metastases are present, chemotherapy has been added; combination regimes containing cis-platinum have shown the best results. Conventional thymic carcinoma of either squamous or lymphoepithelioma-like type are also treated with surgery plus ra.diation therapy, with chemotherapy added in cases of massive local disease or distant spread.(8) The prognosis of fully encapsulated thymoma following surgical excision is excellent regardless of microscopic type. The prognosis of invasive thymoma depends greatly on the completeness of the original excision. It also correlates with the degree of invasiveness, a fact that has been taken into account for the various staging systems for these tumors. The
prognosis drops significantly for the tumors showing gross invasion or implants and even more so for the few cases associated with distant metastases. The prognosis of thymic carcinoma depends a great deal on the microscopic subtype. It is very aggressive for the nonkeratinizing carcinoma (including the lymphoepithelioma-like tumors), sarcomatoid carcinoma, clear-cell carcinoma, and undifferentiated (anaplastic) carcinoma; intermediate for squamous cell carcinoma; and relatively indolent for the rare mucoepidermoid and basaloid carcinomas. (7.8)
DIAGNOSIS:
MALIGNANT THYMOMA TYPE 1.5
REFERENCES:
1. Kirchner T, Schalke B, Buchwald J, Ritter M, Marx A, Miiller-Hermelink HK: Well-differentiated thymic carcinoma. An organotypicallow-grade carcinoma with relationship to cortical thymoma. Am J Surg Patholl6:1153-1169, 1992.
2, Levine GD, Rosai: Thymic hyperplasia and neoplasia. A review of current concepts. Hum Pathol9:495-515, 1978.
3. Lewis JE, Wick MR, Scheithauer BW, BernatzPE, Taylow WF: Thymoma. A clinicopathologic review. Cancer 60:2727-2743, 1987.
4. Masaoka A, Monden Y, Nahahara K, Tanioka T. Follow-up study of thymomas with special reference to their clinical stages. Cancer 48:2485-2492, 1981.
5. Miiller-Hermelink HK, Marino M, Palestro G: Pathology of thymic epithelial tumors. In Miiller-Hermelink HK (ed): Currem topics in pathology. The human thymus. Histophysiology and pathology. vol. 75, New York, 1986, SpringerVerlag, pp. 207-268.
6. Rosai J: The pathology of thymic neoplasia. International Academy of Pathology Monograph No. 29, Baltimore, 1987, Williams & Wilkins.
7. Snover DC, Levine GD, Rosai J: Thymic carcinoma. Five distinctive histological variants. Am J Surg Pathol6:451-470, 1982.
8. Suster S, Rosai J: Thymic carcinoma. A clinicopathologic study of 60 cases. Cancer 67:1025-1032, 1991.
9. Yamakawa Y, Masaoka A, Hashimoto T, Niwa H, Mizuno T, Fuji Y, Nakahara K: A tentative tumor-node metastasis classficiation of thymoma. Cancer 68: 1984-1987, 1991.
CASE #3
CLINICAL IDSTORY:
A 59 year old male with a nine year history of progressive spina!/cerebellar degeneration of unclear etiology presented with shortness of breath. A chest radiograph revealed a 9cm left antericr mediastinal mass. A fine needle aspirate was non-diagnostic and resulted in a pneumothorax requiring chest tube placement. One month later the patient underwent median sternotomy with n:section of the tumor.
GROSS DESCRIPTION:
The mass weighed 55 grams and measured 9x6.5x3.5cm. Cut sections revealed a tan to yellow, gritty, infiltrative tumor with cystic cavities measuring up to 2cm in diameter.
HISTOLOGICAL FEATURES:
The tumor cells are arranged in variably sized nests separated by bands of focally cellular, focally hyalinized stroma. Scattered throughout the tumor are cystic slructures which are lined by cuboidal to squamous epithelium with interspersed goblet cells. The more solid areas show similar cell types, with an admixture of clear cells and intermediate cells. There is accumulation of exiiliCellular mucin which in areas extends into the stromal tissue, resulting in a fibre-inflammatory reaction. Mucicarmine staining highlights the large goblet cell component of this neoplasm. Cytologically, the tumor is well-differentiated, with minimal nuclear pleomorphism.
COMMENT:
These features are diagnostic of mucoepidermoid carcinoma of the thymus. Only three cases of this tumor have been reponed;(4.6.7) it is one of the many types of thymic carcinoma (others being well-dlfferentiated squamous cell carcinoma, basaloid carcinoma, lymphoepithelioma-like carcinoma, neuroendocrine carcinoma, anaplastic carcinoma, sarcomatoid carcinoma, and clear cell carcinoma).0-2.6.9) Histologically, the tumor is essentially indistinguishable from mucoepidermoid carcinoma of the salivary glands; therefore, the possibility of metastasis should always be considered. The reported cases have occurred in adult patients and have pursued an indolent clinical course, belonging to the low-grade histology group of thymic carcinomas defined by Suster and Rosai,<6l which also includes well-differentiated squamous cell carcinoma and ba.~aloid carcinoma. The histogenesis of mucoepidermoid carcinoma in the thymus is not entirely clear. The possibility that these tumors arise either from aberrant salivary tissue in mediastinal teratomas or from congenital cysts of the thymus (such as bronchogenic cysts) has been suggested, and one reported case did coexist with a thymic cyst lined by respiratory epithelium.(?) An alternative (and probably more likely) explanation is that mucoepidermoid carcinomas arise from the thymic epithelium. It is well known that mucinous metaplasia of the thymic epithelium can occur, and cases of multilocular thymic cyst(S) containing goblet cells do recur.
Entities to be considered in the differential diagnosis are primarily epithelial thymomas containing gland-like spaces, mediastinal teratomas containing glandular and squamous elements, and highgrade thymic carcinomas showing squamous and glandular differentiation (adenosqua.mous carcinoma). <2•3•8>
DIAGNOS IS:
MUCOEPIDERMOID CARCINOMA OF THE THYMUS
REFERENCES:
l. Hasserjian RP, Klimstra DS, Rosai J. Carcinoma of the thymus with clear cell features. Lab Invest 70: !lOA, 1994.
2. Kuo T, Chang JP, Lin FJ, Wu WC, Chang CH. Thymic carcinomas. Histopathological varieties and immunohistochemical study. Am J Surg Path 14:24-34,1990.
3. Matsuno Y, Mukai K, Noguchi M, Sato Y, and Shirnosato Y. Histochemical and immunohistochemical evidence of glandular differentiation in thymic carcinoma. Acta Pathol Jpn 39:433-438, 1989.
4. Snover DC, Levine GD, Rosai J. Thymic carcinoma. Five distinctive histological variants. Am J Surg Path 6:451-470, 1982.
5. Suster S, Rosai J. Multilocul.ar thymic cyst. An acquired reactive process. Study of 18 cases. Am J Surg Path 15:388-398, 1991.
6. Suster S, Rosai J. Thymic carcinoma. A clinicopathologic study of 60 cases. Cancer 67:1025- 1032, 1991.
7. Tanaka M, Shirnokawa R, Matsubara 0 , Aoki N, Kamiyama R, Kasuga T, Hatakeyama S. Mucoepidermoid carcinoma of the thymic region. Acta Pathol Jpn 32:703-712, 1982.
8. Truong LD, Mody DR, Cagle PT, Jackson-York GK, Schwanz MR, Wheeler TM. Thymic carcinoma. A clinicoopathologic study of 13 cas:es. Am J Surg Path 14:151-166, 1990.
9. Wal.ker AN, Mills SE, Fechne~ RE. Thymomas and thymic carcinomas. Semin Diagn Pathol 7:250-265, 1990.
CASE #4
CLINICAL HISTORY:
A 42 year old male with a past history of excision of a parathyroid adenoma presented in 1990 with a sensation of "tightness" in the chest. There were no other symptoms. An anterior mediastinal mass was noted on chest radiograph. The lungs were free of masses. An open biopsy was performed and a diagnosis was established. The patient declined treatment for two years but then consented to resection when the mass was found to be increasing in size.
Of note, the patient's brother had excisions of parathyroid and pituitary adenomas as well as a pancreatic endocrine tumor, and his mother died of a widely metastatic endocrine neoplasm.
GROSS DESCRJPTION:
The tumor was lOcm in gneatest diameter and covered by a thin capsu1e. The tissue was tan and soft with multiple small foci of necrosis and calcification. A portion of lung was densely adherent to one aspect, and lobulated adipose tissue suggestive of residual thymus gland was present elsewhere.
illSTOLOGICAL FEATURES:
The tumor is very cellular, with solid sheets and nests of small cells separated by a scanty but well· vascularized stroma. In areas the nesting growth pattern is pronounced. The central portions of the nests exhibit comedo-type necrosis with frequent calcification. Within the nests (and in the more solid areas) the cells are arranged in rosettes with central lumina. At the periphery of the tumor there is infiltration into the surrounding adipose tissue, lung, and residual thymus; in these areas the cells are growing individually and in thin trabeculae. Cytologically, the cells are uniform with minimal amphophilic cytoplasm. Nuclei are round to oval and lack pleomorphism. The chromatin is dispersed and nucleoli are inconspicuous. Mitoses are easily found and range from 10·15 per 10 high power fields. Vascular invasion is present.
Immunohistochemical staining reveals the tumor cells to be positive for keratin (Cam 5.2), chromogranin, synaptophysin, and neuron specific enolase. Stains for SIOO protein, neurofilaments, and parathyroid hormone are negative.
COMMENr:
This tumor is clearly an endocrine neoplasm, and the clinical presentation and gross and microscopic features establish it as a thymic primary, i.e., thymic carcinoid. Although the features of thymic carcinoids overlap somewhat those of pulmonary carcinoid tumors, endocrine neoplasms including carcinoid and small cell carcinoma have been accepted to arise primarily within the thymus,(7.16,19) although these tumors were probably classified as thymomas until25 years ago. The cells of origin are presumbly the small population of endocrine cells which have been observed within the thymuses of animals(3.5) and humans<l7).
Thymic carcinoids typically arise in young adults (mean = 45), with males being more commonly affected(l6). Although most of the patients present with symptoms related only to the mass itself, approximately one third suffer from paraneoplastic endocrinopathies, most commonly Cushing's syndrome0.4.8) due to ACTH secretion by the tumor. In addition, cases of thymic carcinoids associated with MEN I have been reponed00. t4); the tumor in the seminar case clearly belongs to this group. Rare cases associated with MEN II have also been described(lll.
The histologic features of the seminar case are typical, although trabecular formations may be more pronounced. Of note, the presence of "aggressive" endocrine features (such as necrosis and increased mitoses) is common, although nuclear pleomorphism is generally not marked. Some histologic variants have been described, including spindle cell carcinoid(9) and patterns mimi.ck:ing medullary thyroid carcinoma<17). Immunohistoehemically the tumors are1JOsitive for keratin and general endocrine markers such as chromogranin, synaptophysin , and NSE. Peptides such as ACI'H, somatostatin, cholecystokinin, and serotinin may also be detecfed(7.19l.
In comparison with pulmonary carcinoids, thymic carcinoid tumor is an aggressive neoplasm. Local invasion is common, up to 75% of patients develop metastases,(l6) and death from rumor is not infrequent, usually within a few years. Those tumors occurring in the context of MEN may be more aggressive. (14)
It is usually not difficult to exclude the possibility of mediastinal metastasis from a pulmonary primary on clin.ical grounds. Most metastasizing pulmonary carcinoid tumors are detectable radiographically (although at least one.case of a metastasizing pulmonary "tumorlet"(6) has been reponed), and if bronchoscopic examination of the bronchial tree has also failed to identify a pulmonary tumor, an anterior mediastinal carcinoid rumor can be regarded as most likely thymic in origin. In addition, there are some hlstologic differences between thymic and pulmonary carcinoids. The former tumors often have larger nests of cells, and the central necrosis and calcification is characteristic. In addition, rosette formation is more common in thymic primaries. Ftnally, the general appearance of thymic carcinoids is that of a more aggressive endocrine neoplasm than the typical pulmonary carcinoid tumor. In fact, the tenn "atypical carcinoid" would probably beuer suit thymic carcinoids, given the presence of mitoses and necrosis and the relatively aggressive behavior.
Small cell carcinomas within the anterior mediastinum may be more difficult to establish as thymic primaries becuase of the well-known propensity of pulmonary small cell carcinomas to metastasize widely while still very small. Thus, careful radiographic and bronchoscopic evalulation of the lungs must be performed before the diagnosis of primary thymic small cell carcinoma can be confmned. (18)
In addition to metastases from the lung (or other site.s), tumors in the differential diagnosis include other primary endocrine tumors of the mediastinum. Tumors (and hyperplasias) of parathyroid glands are encountered in the mediastinum and may be associated with the thymus, a not unelC])CCted phenomenon given the common embryologic origin of the two structures.07) While hyperplasia and adenoma are by far the most commonly occurring conditions, cases of primary mediastinal parathyroid carcinoma have been reponed as welL03) Parathyroid lesions with chief cell predominance are most likely to be confused with thymic carcinoids. The clinical history of hypercalcemia and immunohistochemical detection of parathyroid hormone can confirm the diagnosis. Paragangliomas also occur in the mediastinum.Cl2 .. JS.t7) Prominence of a "zellballen" growth pattern, pleomorphism of the nuclei and the lack of necrosis and mitotic activity are features of paraganglioma, and immunohistochemistry may be helpful as well. While both paragangliomas and carcinoids are positive for chromogranin, paragangliomas lack keratin staining and may express neurofilaments.<12} The finding of abundant S-100 protein-positive sustentacular cells around the nestS of tumor cells also suppons the diagnosis of paraganglioma, although similar cells have occasionally been seen in carcinoid tumors.<2) Nonendocrine neoplasms can also simulate thymic carcinoid. The rosette-like patterns which may be seen in thymomas may cause confusion with thymic carcinoid, and spindle cell carcinoids may be mistaken for spindle cell thymomas. In addition to the lack of immunohistochemical staining for endocrine markers, thymomas typically Jack necrosis and relatively abundant mitoses, and scattered lymphocyteS are usually present (at
least focally). In some cases, thymic carcinoids may also mimic germ cell tumors or lymphomas (IS) immunohistochemistry is again of great utility in these distinctions.
DIAGNOSIS:
THYMIC CARCINOID TUMOR
REFERENCES:
1. Baker J, Holdaway IM, Jagusch M, Kerr AR, Donald RA., Pullan PT. Ectopic secretion of ACTH and met-enkephalin froma thymic carcinoid. J Endocrinol Invest 5.:33-37, 1982.
2. Barbareschi M, Frigo B, Mosca L , Carboni N, Arrigoni GP, Leonardi E, Wilander E, Siegal GP, Shiro BC. Bronchial carcinoids with S-100 positive sustentacular cells. Pathol Res Pract 186:212-222, 1990.
3. Csaba G. An attempt to demonstrate endocrine cells in the rat thymus. Endocrinol Exper 7:99-105, 1973.
4. Gartner LA, Voorhess ML. AdrenocorticotrOpic hormone.-producing thymic carcinoid in a teenager. Cancer71:106- ll, 1993.
5. Hi!kanson R, Larsson Ll, Suodler F. Peptide and amine producing endocrine-like cells in the chicken thymus. A chemical, histochemical and electron microscopic study. Histochemistry 39:25-34, 1974.
6. Hausman DR, Weimann RB. Pulmonary tumorlet with hilar lymph node metastases. Cancer 20:1515-1519, 1967.
7. Herbst WM, Kummer W, Hofmann W, Otto H, Heym C. Carcinoid tumors of the thymus. An immunohistochemical study. Cancer 60:2465-2470, 1987.
8. Hunttakoon M, Lin F, Heitz PU, Tomita T. Thymic carcinoid tumor with Cushing' s syndrome. Arch Pathol Lab Med 108:551-554, 1984.
2. Levine GD, Rosai J. A spindle cell variant of thymic carcinoid tumor. Arch Pathol Lab Med 100:293-300, 1976.
10. Manes JL, Taylor HB. Thymic carcinoid in familial multiple endocrine adenomatosis. Arch Pathol 95:252-255, 1973.
11. Marchevsky AM, Dikman SH. Medi~tinal carcinoid with an incomplete Sipple's syndrome. Cancer 43:2497-2501, 1979.
12. Moran, CA, Suster S, Fishback N, Koss MN. Mediastinal paragangliomas. A clinicopathologic and immunohistochemical study of 16 cases. Cancer 72:2358-2364, 1993.
B . Murphy MN, Glennon PG, Diocee MS, Wick MR, Cavers DJ. Nonsecretory parathyroid carcinoma of the mediastinum. Cancer 58:2468-2476, 1986.
'
14. Rosai J, Higa E, Davie J. Mediastinal end~ne neoplasm in patients with multiple endocrine adenomatosis. Cancer 29:1075-1083, 1972.
15. Rosai, J, Levine G, Weber, WR, Higa E. Carcinoid tumors and oat cell carciomas of the thymus. Pathol Ann 11:201-226, 1976.
16. Wick MR, Bematz PE, Carney JA, Brown LR. Primary mediastinal carcinoid tumors. Am J Surg Pathol6:195-205, 1982.
17. Wick MR, Rosai J. Neuroendocrine neoplasms of the mediastinum. Seminars in Diagn Pathol 8:35-51, 1991.
18. Wick MR, Scheithauer BW. Oat-cell carcinoma of the thymus. Cancer 49:1652-1657, 1982.
!9. Wick MR, Scheithauer BW. Thymic carcinoid, A histologic, immunohistochemical, and ultrastructural study of 12 cases. Cancer 53:475-484, 1984.
CASE #5
CLINICAL HISTORY:
A 75 year old male was noted to have a slowly growing mass in the left posterior mediastinum five years prior to surgery. However, a marked increase in size was noted and the tumor was resected. The patient was asymptomatic and laboratory studies were within n:ormallimits.
GROSS DESCRIPTION:
The tumor measured 20cm and weighed 1340 gm. It was partially ·encapsulated and consisted of finn, nodular white tissue with myxoid areas.
HISTOLOGICAL FEATURES:
This mediastinal neoplasm is a typical example of the entity currently known as solitary fibrous tumor. It is possible that some of these tumors grow into the.mediastinum from the medial pleura, but we. think that most originate from the mediastinal (including thymic) stroma.(l2) Their microscopic appearance and immunohistochemical profile are similar to !hose of their pleural counterparts, out a higher proportion of their reported cases have run an aggressive clinical course. The most striking mi<.,Toscopic feature is the alternation of byper- and hypocellular foci, both of them composed of oval to spindle cells growing in illdefinei:l fascicles and separated by keloid-type collagen. Cellularity can be very high, but !here is usually little pleomorphism. Necrosis and mitotic activity can be seen and represent two of the most important criteria for predicting aggressive behavior.
COMMENT:
Although this I)Jmor was regarded for many years as originating from mesothelial cells on !he basis of tissue culture studies and designated as solitary fibrous tumor, it has become increasingly apparent that it is instead made up of fibroblasts/myofibroblasts having phenoi)'Jlical features equivalent to those of submesothelial conoective tissue cells.(l.3.5.l The tumor cells are probably responsible for the large amounts of collagen deposited in lhese tUmors. Immunohistochentically, they are strongly reactive for vimentin, focally and inconstantly positive for actin, and negative for desmin, SlOO protein, and keratin.02l We have recently found that they also show a consistent, strong, and widespread positivity for CD34, a feature that they also share with dermatofibrosarcoma protuberans, other "fibrohistocytic" tumors, some peripheral nerve neoplasms, endothelial tumors, and ~ithelioid sarcoma.(IO,l!)
Other locations in which this neoplasm has been reported (in addition to pleura and mediastinum) include peritoneum, liver, lung (without pleurd! connection), upper respiratory tract and orbit.(4,6,7,l3, 14,15.16) There is also a remarkable similarity between this entity a~ reported in those sites and at least some cases of the tumor of the male and felllllle breast known as myofibroblastoma(9) (including the CD34 positivity that we have documented in some of these cases). Some· examples of these solitary fibrous tumors (especially when large) are associated with tumor-related hypoglycemia, which promptly disappears following removal of the neoplasm. (8) Practically all of the mesenchymal tumors that we have had the opportunity to review that were associated with this paraneoplastic syndrome have belonged to this tumor type, although many have been diagnosed or reported as fibrosarcoma, leiomyosarcoma, or hemangiopericytoma. The hypoglycemia has been found to be due to the secretion by the tumor of an insulin-like growth factor II. (2)
DIAGNOSIS:
SOLITARY FIBROUS T UMOR
REFERENCES:
1. Dalton WT, Zolliker AS, McCaughey WTE, Jacques J, Kannerstein M: Localized primary tumors of the pleura. Cancer 44:1465-1475, 1979.
2. Daughaday WH, Emanuele MA, Brooks MH, Barbato AL, Kapadia M, Rotwein P: Synthesis and secretion of insulin-like growth factor II by a leiomyosarcoma with associated hypoglycemia. N Eng! J Med 319:1434- 1440, 1988.
9. Wargotz ES, Weiss SW, Norris HJ: Myofibroblastoma of the breast. Sixteen cases of a distinctive benign mesenchymal tumor. Am J Surg Patho1 11:493-502, 1987.
10. Weiss S, Nickoloff B: CD-34 is expressed by a clistinctive cell population in peripheral nerve, nerve sheath rumors, and related lesions. Am J Surg Pathol 17:1039-1045, 1993.
11 . Westra WH, Gerald WL, Rosai J: Two novel observation on solitary fibrous tumor: Consistent CD34 reactivity and occurrence in the orbit. Am J Surg Pathol (in press).
12. Witkin GB, Rosai J: Solitary fibrous tumor of the mediastinum. Am J Surg Patho1 13:547-557, 1989.
13. Witkin GB, Rosai J. Solitary fibrous tumor of the upper respiratory tract. Am J Surg Patholl5:842-848, 1992..
14. Young RH, Clement PB, McCaughey WTE: Solitary fibrous tumors ('fibrous mesotheliomas') of the peritoneum. Arch Pathol Lab Med 114:493-495, 1990.
15. Yousem SA, Flynn SD. Intrapulmonary localized fibrous tumor. Am J Clin Pathol 89:365-369, 1988.
16. Zukerberg LR, Rosenberg AE, Randolph G, Pilch BZ, Goodman ML: Solitary fibrous tumor of the nasal cavity and paranasal sinuses. Am J Surg Patholl5: 126-130, 1991.
CASE #6
CLINICAL HISTORY:
The patient is a 72 year old white female who was noted to have an incidental mediastinal mass on a chest radiograph performed prior to repair of a rectocele in 1990. The patient refused funher evaluation of the mass at that time, but presented in 1992 with increasing shortness of breath and anterior chest pain. On cr, the rumor measured 16cm and involved the anterior mediastinum and right hemithorax. Two components to the mass were identified, one of which had the appearance of adipose tissue. The tumor was resected uneventfully.
GROSS DESCRIPTION:
The tumor was circumscribed and soft, with areas of yellow, fatty tissue alternating with more firm, white-tan, lobulated areas.
HISTOLOOICAL FEATURES:
At low power the tumor has a geographic appearance. Cords and nests of thymic tissue are separated by bands of fibroadipose tissue with focal sclerosis. The thymic tissue contains abundant lymphocytic in many areas, and scattered Hassell's corpuscles can be seen. Focally the lymphocytes are absent, and the thymic tissue is limited to thin strips of epithelial cells lacking keratinization. The fibroadipose tissue contains large, atypical cells which are most numerous in the more sclerotic regions. Focally the nuclear atypia involves the adipocytes, and scattered multi vacuolated lipoblasts are presenl In some areas the thymic tissue is absent, and the fatty component becomes more myxoid. Spindle cells predominate elsewhere. Atypical cells are more abundant in these regions, although the lesion retains a low-grade appearance throughout, lacking hypercellularity, increased mitoses, and necrosis.
COMMENT:
This tumor contains elements of benign thymic tissue (both the epithelial and lymphocytic components) and well differentiated liposarcoma (focally sclerosing). The abundance of thymic tissue within the lesion and the intimate admixture of the two components suggests that the malignant component not only arose within the thymus gland but induced a proliferation of the thymic epithelium as well. Tumors with this appearance have been referred to as "thymoliposarcomas", in analogy with their benign counterpart, the thymoliporna .<B.tO) Only two such cases have been described; both appeared to arise de novo, having no evidence of a pre• existing thymolipoma.
Although malignant lipomatous tumors with this distinctive appearance are very uncommon, liposarcomas of the thymus and anterior mediastinum are probably the most frequent mesenchymal tumors of that region.<4.tO.li,J6.t9) Over 80 cases of mediastinal liposarcoma have been reported, with the majority occurring in the antero-superior mediastinum. Some can be grossly or microscopically proven to have arisen within the thymus (25% of anterior mediastinal liposarcomas}, but in these cases the thymic tissue consists only of remnants compressed around the periphery of the tumor. It is possible that an even greater percentage of anterior mediastinal liposarcoma.~ arise within the thymus, and some of the larger tumors may completely replace the gland.
Mediastinal liposarcomas occur over a wide age range, including children0.14) and elderly adults; the mean age is 42 years.(JO,J6) Males are more commonly affected. Presenting symptoms are often non-specific but may include pain, respiratory difficulty, and superior vena cava syndrome.
The tumors may reach very large sizes (I!P to 40.0 em), with extension into the thoracic cavities. Grossly, the appearance 1~ similat to that of liposarcomas occurring elsewhere and depends upon the histologic type.
The histologic features of mediastinal liposarcomas are similar to those of liposarcomas occurring in the retroperitoneum or extremities. All histologic types occur (well differentiated lipoma-like and sclerosing, myxoid, round cell, and pleomorphic), with most cases being low grade (89% ). The well differentiated types consititute 57%, while 32% of cases are the myxoid type.OO) One interesting feature of the sclerosing variant is the frequent presence of large dense aggregates of lymphoid cells which occasionally contain follicles. Within this background, the scattered atypical tumor cells may easily be mistaken for the Reed-Sternberg cells of nodular sclerosis Hodgkin's disease. These aggregates probably represent reactive lymphocytic components rather than residual thymic lymphocytes, since an equal admixture of B-cells and T -cells is found.
The ctinical course of mediastinal liposarcomas parallels that of similar tumors within the retroperitoneum.OO,l6,17) Patients with low grade tumors experience reP.ellted recurrences and usually die of local disease, often .after many years. The rare high grade liposarcomas are capable of distant metastases as well as locally aggressive behavior. Treatment has largely been surgical, and the multilobu.lated growth pattern of these tumors make complete excision difficult. Chemotherapy and radiation therapy have been employed in some cases, but there are inadequate data to assess their efficacy.
One of the entities in the differential diagnosis (especially in case of thymoliposaroma) is thymolipoma. In these tumors, there is intimate admixture of thymic elements with mature adiposetissue, and no malignant cells are seen in the stromal component. In addition spind.le cell elements, sclerosis, and myxoid areas are not seen. Thymolipomas are con:sidered completely benign; cases associated with myasthenia gravis and aplastic anemia have been reported. The possibility of lipomas occurring in the mediastinum must also be considered, but (as in the reiroperitoneum) the minimal degree of atypia which may be present in well differentiated, lipoma-like liposarcomas suggests that the diagnosis of lipoma must be made· with great caution. Multivacuolated cells, cytologic atypia, or areas of sclerosis sho.uld raise suspicions that. cells diagnostic of liposarcoma are likely to be present. The differentia.! diagnosis aJso includes other sarcomas which can occur in the mediastinuli'l. MFH, leiomyosarcoma,(l.7.15.20) chondrosarcoma,<3.12,13) angiosarcoma,(S) and osteosarcoma<6,9) have all been described, and each has distinctive histologic (and possibly immunohistochemical) features while lacking lipoblasts. The possibility that a mediastina.! sarcoma has arisen from an underlying teratoma must be.considered; this transformation appears to be more frequent in mediastinal germ cell tumors than in tbose occurring elsewhere. Malignant peripheral nerve sheath tumors also occur, usually in the posterior mediastinum. Of course, the possibilities of sarcomatoid carcinoma (of thymic or other origin) or sarcomatoid epithelial mesothe.lioma must be con~idered for high grade spind.le cell neoplasms before the diagnosis of sarcoma can be confirmed. Solitary fibrous tumors can also involve the mediastinum and may mimic the more sclerotic areas of liposarcomas. The recently described positivity of solitary fibrous tumors for CD34 may be helpful in this differential.
DIAGNOSIS:
THYMIC LIPOSARCOMA ("THYMOLIPOSARCOMA ")
REFERENCES:
I. Bernheim J, Griffel B, Versano S, Bruderrnan I. Mediastina.! leiomyosarcoma in the wail of a bronchia.! cyst. Arch Pathol Lab Med I 04:221, 1980.
2. Castleberry RP, Kelly DR, Wilson ER, Cain WS, Salter MR. Childhood · liposarooma. Repon of a case and review of the literature. Cancer 54:579-584, 1984.
3. Chetty R. Extraske1etal mesenchymal chondrosarcoma of the mediastinum. Histopathology 17:261-263, 1990.
4. Dogan R, Ayrancioglu K, Aksu 0. Primary mediastinalliposarooma. A repon of a case and review of the literature. Eur J Cardiothorac Surg 3:367-370, 1989.
5. Gibbs AR, Johnson NF, Giddings JC, Powell DE, Jasani B. Primary angiosarcoma of the mediastinum. Hum Pathol 15:687-691, 1984.
6. Greenwood SM, Meschter SC. ExtraSkeletal osteogenic sarcoma of the mediastinum. Arch Pathol La,b 'Med 113:430-433, 1989.
7. Gupta S, Jindal SK, Vashisht R, Singh H, Malik·SK. Leiomyosarcoma of the mediastinum. Eur J Resp Dis 64:69-71, 1983.
8. Havlicek F, Rosai J. A sarcoma of thymic stroma with features of liposarcoma. Am J Clin Pathol 82:217-224, 1984. .
9. Ikeda T, Ishihara T, Yoshimatsu H, Kikuchi K, Murakami M. Primary osteogenic sarooma of the mediastinum. Thorax. 29:582-588, 1974.
10. Klimstra DS, Moran CA, Koss M, Rosai J. Liposarcoma of the anterior mediastinum: A clinicopathologic series of 27 cases. Am J Surg Pathol, 1994 (In press).
11. McLean TR, Almassi GH, Hackbanh DA, Janyan NA, Potish RA. Mediastinal involvement by myxoid liposarcoma. Ann Thorac Surg47:920-921, 1989.
12. Montresor E, Abrescia F, Bertrand C. Piazzola E, Girardi C. Iacono C, Pucheni V. Mediastinal chondrosarcoma. Case repon. Acta Chirurg Scand 156:733-6, 1990.
13. Phillips OW, Chooug M. Chondrosarcoma presenting as an anterior mediastinal mass. Clin\ Radio143:63-64, 1991.
14. Plukker JT, Joosten HJ, Rensing JB, Van Haelst UJ. Primary liposarcoma of the mediastinum in a child.J Surg Oncol37:257-263, 1988.
15. Rasaremam R, Panabokke RG. Leiomyosarcoma of the mediastinum. Br J Dis Chest 69:63-69, 1985.
16. Razzuk MA, Urschel HC, Race OJ, Kingsley WB, Paulson DL. Liposarcoma of the mediastinum. Case repon and review of the literature. 1 Thor Cardiovasc Surg 61:819-826. 1971.
17. Schweitzer DL, A guam AS. Primary liposarcoma of the mediastinum. Repon of a case and review of the literature. J Thorac Cardiovasc Surg 74:83-97, 1977.
18. Shibata K, Koga Y, Onitsuka T, WakeN, Ishii K, Selciya R, Sumiyoshi A. Primary liposarcoma of the mediastinum--a case report and review of the literature. Jpn J Surg 16:277-283, 1986.
19. Standerfer RJ, Armistead SH, Paneth M. Liposarcoma of the mediastinum: report of two cases and review of the literature. Thorax 36:693-694, 1981.
20. Sunderrajan EV, Luger AM, Rosenholtz MJ, Maltby 10. Leiomyosarcoma in the mediastinum presenting as superior vena cava syndrome. cancer 53:2553-2556, 1984.
CASE 4#7
CLINICAL HISTORY:
A 62 year old man presented with a right supraclavicular mass which was found to be contiguous with a large anterior mediastinal mass. A biopsy was interpreted as neoplastic, possibly paraganglioma or thymoma. The tumor was resected in two pieces measuring 9cm and llcm and weighing 320gm in aggregate.
GROSS DESCRIPTION:
The cut section revealed multinodular yellow soft tumors with central hemorrhage and necrosis.
HISTOLOGICAL FEATURES:
This mediastinal tumor seems to be arising from a lymph node, judging from the fact that large amounts of residual lymph node strucnires are present in some of the sections. It is in these areas ~}!at the relationship between the tumor cells and the non-neoplastic lymphoid component can be better appreciated. The tumor cells are represented by plump elements with large vesicular nuclei and prominent nuclei which are arranged individually or in small clusters around the normal lymphoid cells. In other areas, they form larger clusters that eventually result in an obliteration of the lymph node architecture. In these solid areas, the tumor cells also acquire a greater degree of atypicality• with nuclear hyperchromasia and mitotic activity. Focally, the close intermingling of neoplastic cells and lymphocytes is reminiscent of that seen in thymoma, to the point that some areas would be difficult to distinguish from some in Seminar Case 2. lrnmunohistochemically, the tumor cells of this case are negative for all epithelial and endothelial markers. They are instead positive for vimentin and CD21. The combination of cytological, architectural, and immunohistochemical features strongly suggest that this lesion represents a neoplasm of true reticular cells, speCifically those associated with foUicular B cell zones, which have been variously referred to as dendritic reticular cells and dendritic follicular cells.
COMMENT:
The fact that there is a family of cells related to the immune system that have as a main function that of presenting antigen to the lymphoid elements is now widely accepted. (3)
The original term "reticulum cell" derives from the fa<:t that these cells have elongated cytoplasmic processes joining each other and resulting in the formation of a network ("reticulum"). Later on, this was also linked to the association with (and presumed fonnatioJI of) reticulin fibers by these cells. The old concept of reticulum cell sarcoma stems from the beijef that the tumors so designated originated from this segment of the immune system. Although it was subsequently demonstrated that the large majority of tumors diagnosed in the past as reticulum cell sarcomas represent large-cell lymphomas of either B or T cell type, there is now good evidence that true reticulum cell sarcomas do indeed exist, and that some of them have the phenotype of dendritic cells whereas others have the phenotype of interdigitating reticulum cells.~2.4.6,9) The diagnosis is based on a oombination of features which include conventional morphology, enzyme histochemistry, electron microscopy, and immunohistochemistry. The majority of the dendritic reticulum Gell tumors that we have seen have involved cervical lymph nodes,, but cases of this entity have also been reponed in extranodal sites, such as oral cavity and skin.(I,S) At the eleGtron microscopic level, the most important distinguishing feature ofthese cells is the presence of long cytoplasmic prolongations and the fact that these prolongations are joined by complex desmosomes.(4) As a matter of fa<:t, the dendritic; reticulum cell is the only normal element in lymph nodes that contains complex intercellular junctions of this type. (3)
lmmunohistoohemically, the most important markers are R4/23 (wllich can be denwnstrated only in frozen sections), KiM4, CD21, and CD35.(1,5,7} S100 protein reactivity is inconstant. In the nonnal situation, the latter reactivity .is associated with interdigitating rather then dendritic reticulum cells, but in the neoplastic counterparts the resulls with this marker have been rather erratic. (9)
Dendritic reticulum cell tumor is a low-grade malignancy which is characterized by a teni!ency to local recurrence and eventually (in some cases) distant spread. There is a tendency for.progressive pleomorphism, increased mitotic activity, and necrosis in the recurrent lesions. In the later stages, it may impossible to distinguish tllis tumor from an anaplastic large cell lymphoma or a non-lymphoid malignancy. (4) At the H&B level, one of the most important distinguishing features is the close intermingling of tumor cells and lymphocytes. This feature, plus the fact that the tumor cells have nuclei of predominantly oval shape. and vesicular appearance ·are responsible for some cases of this entity having been misdiagnosed in the past as ectopic cervical thymomas.
DIAGNOSIS:
CONSISTENT WITH TRUE RETICULUM CELL TUMOR
REFERENCES:
I. Chan JKC, Tsang WYW, Tang SK, Lee A WM: Follicular dendritic cell tumors of the oral cavity. Am J Surg Pathol18:148-157, 1994.
3. Lennert K, Feller AC. Histopathology of non-Hodgkin's lymphomas (based on the updated Kiel classification). 2nd ed. Berlin: Springer-Verlag, 1992.
4. Monda L, Warnke R, Rosai J: A primitive lymph node malignancy with features suggestive of dendritic reticulum cell differentiation. A report of 4 cases. Am J Pathol 122:562-572, 1986.
5. Naeim M, Gerdes J, Abdulaziz A, Stein H, Mason DY: Production of a monoclonal antibody reactive with human dendritic cells and its use in the immunohistological analysis of lymphoid tissue. J Clio Patbol36:167-175, 1983.
6. Pallesen G, Myhre-Jensen 0: Immunophenotypic analysis of neoplastic cells in follicular dendritic cell sarcoma. Leukemia 1:549-557, 1987.
7. Parwaresch MR, Radzun HJ, Hansmann ML, Peters KP: Monoclonal antibody Ki-M4 specifically recognizes human dendritic reticulum cells (follicular dendritic cells) and their possible precursors in blood. Blood 62:585-590, 1983.
8. Strickler JG, Parkin JL, Schmidt CM, Hurd DD: Primary skin malignancy with features suggestive of dendritic reticulum cell differentiation. Ultrastruct Pathol 14:273-282, 1990.
9. Weiss LM, Berry GJ, DoJ;fman RF, Banks P, Kaiserling E, Curtis J, Rosai J, Warnke R: Spindle cell neoplasms of lymph nodes of probable reticulum cell lineage. Am J Surg Pathol 14:405-414, 1990.
CASE #8
CLINlCAL HISTORY:
A 66 year old man underwent a coronary artery bypass operation several years before presentation. On follow-up he was found to have a 5 em well-demarcated nodular mass exactly in the region when the pleura had been stripped to expose the internal mammary artery. Several l-2 mm nodules were also noted on the pleufl\ adjacent to the main mass. The remainder of the pleura was unremarkable. There was no history of asbestos exposure or pleural effusion. The tumor was excised.
GROSS DESCRIPTION:
The mass was brown-tan and homogeneous. There was no hemorrhage or necrosis.
HISTOLOOICALFEA TURES:
The tumor consists of uniform, epithelial-appearing cells with minimal stroma within the lesion. Numerous islands of lymphocytes and plasma cells are present. Architecturally the cells are arranged in solid nests and sheets. Branching cleft-like spaces are present between groups of cells, and the cells projecting into these spaces have a "hob-nailed" appearance. Scattered papillary and tubular formations are present. In many areas the cells appear somewhat discohesive. On high power, individual cells are separated from their neighbors by clear spaces. The nuclei are uniform with mild atypia, although prominent nucleoli are conspicuous. Mitoses are detectable but not frequent At the periphery the tumor is partially surrounded by a fibrous pseudocapsule, although focally nests of cells have invaded into the adjacent fat.
Immunohistochemically the rumor cells are diffusely positive for keratin (Cam 5.2 and AEl) but stain negatively for CEA, LeuMI, and B72.3. No mucin is detectable. By electron microscopy the cells have the characteristics of mesothelial cells, with numerous long microvilli.
COMMENT:
The pathologic fearures of this tumor are clearly those of a mesothelial proliferation. Histologically, the growth pattern (with sheets, papillae, and rubules) and cytologic features are typical, the immunohistochemical profile and lack of mucin are characteristic, and the ultrastructural appearance completes the picrure. The fundamental questions about this rumor are twofold: first, is it a neoplastic proliferation, and second, if neoplastic, is it benign or malignant.
Evidence to suggest that this tumor may be a reactive proliferation exists, and arguments could also be made that it is a malignant neoplasm; however, the concept of a benign neoplastic proliferation of mesothelial cells is not well accepted to occur in the pleural cavity. Indeed, grading of epithelial mesotheliomas has not even been worthwhile, since well differentiated examples usually pursue the same rapidly aggressive course as their poorly differentiated counterparts. Thus, once a mesothelial proliferation in the pleura is determined to be neoplastic, it is by definition malignant. Benign mesothelial tumors do occur in the peritoneum (especially the pelvis), but these tumors are characteristic clinicopathologic entities (adenomatoid tumors and multicystic mesotheliomas) and do not share histologic features with the seminar case. The only benign tumor of the pleura to be classified as a mesothelioma is the localized (or solitary) fibrous "mesothelioma". This tumor is now known to be of mesenchymal rather than mesothelial derivation.(2) Under the name of solitary fibrous tumor, it (along with its malignant counterpart, which is a true sarcoma) is now classified separately from true mesothelial neoplasms,CS> and should especially be distinguished
from sarcomatoid mesotheliomas, which are epithelial neoplasms with a sarcomatoid growth P.attem (analogous to sarco~atoid carcinomas of many organs).
If this tumor is a reactive proliferation, it would be regarded as an (extreme) example of nodular mesothelial hyperplasia. This condition was fJCSt described as an incidental finding in hernia sacs,(! I) but similar proliferations can be seen in other mesothelial-lined areas. ln fact, the mesothelial lining of the pericardium is well known to undergo florid reactive proliferation, to the point that the diagnosis of primary pericardia! rnesothelioma(6.9) should be made with great caution. The features of the current case which suggest a reactive proliferation include the relative Jack of atypia and mitotic activity (indeed, nodular mesothelial hyperplasia may contain more abundant mitoses than the seminar case), the. localized nature of the tumor, and the temporal association of the mass with the previous surgery in that area. Recently there has been recognition that manipulation of the tissue in the region of the heart and pericardium may result in the proliferation of mesothelial and other cells, resulting in a lesion which may resemble mesothelial neoplasms (i.e., mesotlielioma) or other twliors.(7,t2) The similarity of these proliferations to epithelioid hemangioma has been noted,<?> but immunohistochemical positivity for keratin (with negative staining for factor VIII) points toward their mesothelial nature. Other investigators have noted the admixture of monocytes with the mesothelial cells in these lesions and have suggested the te,rm "mesothelial/monocytic incidental cardiac excresences" or "cardiac MICE':; proliferation reactive to cardiac catheterization was favored.(l2) Similar tissue fragments were also found in the suction tips of cardiac bypass pump filters, suggesting that the the "tumors" are not even proliferative lesions but rather artifactual accumulations of mesothelial cells and other debris mechanically removed during surgery or other manipulation. (3) Another point is that mesothelial cells may be found as incidental lymph node inclusions in the mediastinum, OJ another nonneoplastic phenomenon. which could be. confused with mesothelioma. Thus, there is ample precedent for reactive mesothelial proliferations in this region, especially following surgery.
Despite these points, there are many features of the lesion in question which differ from the reactive proliferations mentioned. This tumor arose three years after surgery. Although continued proliferation of a cardiac "MICE" could be consid~red, this is clearly not an "incidental" lesion. No monocyte component was present, and the proliferation had a monotonous appearance. Although the histologic characteristics are not dissimilar to those of nodular mesothelial hyperplasia, the size of the lesion far exceeds any examples we ha.ve seen. Also, there was focal invasion.into surrounding tissues as well as several additional pleural nodules noted at surgery (not biopsied). Thus, the conservative interpretation would have to be that this tumor represents an early (i.e., currently localized) malignant epithelial mesothelioma which would be expected to progress to diffuse pleural involvement with time. The primary location in the mediastinum would be unusual but bas been reported.(4)
Obviously, one would have to consider thymoma in the differential diagnosis of this lesion, although the growth pattern is highly suggestive of mesothelial cells, and the architectural features of thymoma are lacking. The opposite phenomenon (i.e., the pleural growth of thymoma simulating mesothelioma) has also been described(8,10) and may even occur in the absence of a mediastinal component.
The patient in question is alive without known recurrence two years after surgery; however, we have seen several cases of mesothelioma recur after extended intervals. Probably only long-term follow-up will ultimately determine the true nature of this tumor.
I. Brooks JSJ, Livolsi VA, Pietra GG. Mesothelial cell inclusions in mediastinal lymph nodes mimicking metastatic carcinoma. Am J Clin Pathol 93:741-748, 1990.
2. Carter D, Otis CN. Three types of spindle cell tumors of the pleura: fibroma, sarcoma, and sarcomatoid mesothelioma. Am J Surg Parhol 12:747-753, 1988.
3. Courtice RW, Stinson W A, Walley VM. Tissue fragments recovered at cardiac surgery masquerading as tumoral proliferations. Evidence suggesting iatrogenic or anefactual origin and common occurrence. Am J Surg Pathol 18:167-174, 1994.
5. England DM, Hochholzer L, McCarthy MJ. Localized benign and malignant fibrous tumors of the pleura. Am J Surg Pathol 13:640-658, 1989.
6. Fukuda T, Ishikawa H, Ohnishi Y, Tachikawa S, Oguma F, Kasuya S, Sakashita I. Malignant spindle cell tumor of the pericardium. Evidence of sarcomatous mesothelioma with aberrant antigen expression. Acta Parhol Jpn 39:750-754, 1989.
7. Luthringer DJ, Virmani R, Weiss SW, Rosai J. A distinctive cardiovascular lesion resembling histiocytoid (epithelioid) hemangioma: evidence suggesting mesothelial participation. Am J Surg Pathol 14:993-1000, 1990.
8. Moran CA, Travis WD, Rosado-de-Christenson M, Koss MN, Rosai J. Thymomas presenting as pleural tumors. Am J Surg PAthol 16:138-144, 1992.
9. Nararnoto A, Itoh N, Nakano M, Sbigematsu H. An autopsy case of tuberous sclerosis associated with primary pericardial mesothelioma. Acta Pathol Jpn 39:400-406, 1989.
10. Payne Jr. CB, Morningstar WA, Chester EH. Thymoma of the pleura masquerading as diffuse mesothelioma. Am Rev RespirDis 94:441-446, 1960.
11. Rosai J, Dehner LP. Nodular mesothelial hyperplasia in hernia sacs. A benign reactive condition simulating a neoplastic process. Cancer 35:165-175, 1975.
A 56 year old man presented in 2/92 with 2 months of cough. fatigue and dysphagia. Chestradiograph revealed a 13xl0cm left anterior mediastinal mass. Biopsy via mediastinoscopy showed ectatic blood vessels in a fibrotic s1l:'Qma, and the diagnosis of cavernous hemangioma was made. The patient received 2160 rads of radiation with no response. In 6/92 a more substantial biopsy was taken. At that time, the tumor filled the left chest with extension to the diaphragm.
GROSS DESCRIPTION:
Grossly the tissue appeared pink to gray with areas of necrosis.
HISTOLOGICAL FEATURES:
This neoplasm represents a malignant germ cell tumor showing a combination of epithelial and mesenehymal elements. The epithelial component is arranged in the form of cords and tubules lined by cells with markedly hyperchromatic nuclei and high mitotic activity. Some of these tubules have a primiitive neuroepithelial l!Ppearance. They are separated by an abundant edematous stroma. The mesenchymal element of the neoplasm is sometimes seen encircling the tubules and is formed by oval to spindle cells which are mitotically active.
COMMENT:
Mediastinal germ cell neoplasms are interesting for a variety of reasons. One relates to their evolution, which is largely a function of their morphologic makeup. Mature teratomas.are nearly always curable simply by extirpation. Seminomas are c.ured in the majority of the cases with radiation therapy. All the others (malignant nonseminomatous germ cell tumors) require a combination of raaiation therapy and chemotherapy, the results being similar but not identical to the testicular tumors of equivalent moxphology.02) An event of dire prognosis in media~tinal germ cell tum_ors is the emergence of a sarcomatous component of "somatic" type. It is our impression that this event is more common in this location than in the gonadal site. (5) It may take the form of angiosarcoma, rhabdomyosarcoma, or other types of sarcomas. This component is not susceptible to treatment with germ-cell-type protocols, confulning the rule that tumors respond to chemotherapy depending on their phenotypical features rather than their histogenesis. (5) Parenthetically, another type of sar·coma that we have observed with germ cell tumors is represented by the emergence of a high-grade sarcoma (sometimes with rhabdomyoblastic differentiation) in spermatocytic seminoma, a t11mor which occurs exclusively in the tes9-s and which has never been seen associated with a nonseminomatous germ cell tumox: component. (13)
The other fascinating aspect of mediastinal germ cell tumors concerns the.ir histogenesis. Specifically, the question that has been asked ever since this entity was first described is how to explain the presence of a germ cell tumor outside sites where normal gem cells are known to reside, i.e., the male and female gonads.OO) Several possibilities have been proposed:
1. They could r«Jlresent metastases from a gonadal or germ cell tumor which has not been detected or which has undergone spontaneous regression;(6)
2. They could be of somatic cell derivation but endowed with the capacity to differentiate along germ cell type features;(3,7)
3. They could represent the neoplastic alteration of a structure which -although still embryonal and "primitive" - is composed of cells which have progressed beyond the germ cell stage; (2)
4. They could have arisen from germ cells which were located outside the gonads.
0f these four possibilities, the currently favored one is the fourth, at least for the mediastinallocarion.O.IO) There is little doubt that most if not all retroperitoneal germ cell tumors in adults repres~nt lymph node metastases from undetected or regressed gonadal primaries. This does not seem to be the case for the mediastinal tumors which, almost without exception, arise from within the thymus. The main problem with this interpretation is the fact that the presence of normal germ cells in the thymus bas yet to be conclusively demonstrated. This is a strong indictment indeed considering the fact that the history of pathology is replete with attribution of tumors to hypothetical rests that were either figments of the imagination or- when real - totally unrelated to the tumors for whose occurrence they were blamed. However, the opposite argument can be made, in the sense that sometimes existence of a tumor of a certain site has preceded a description of the corresponding normal cell and has actually prefigured the discovery of the latter.
If germ cells are normally present in the thymus, why has nobody conclusively identified them at those sites? The claims in the literature on this matter are few and unconvincing. (I) · One could argue that they may be very difficult to detect because of their inconstancy, small density, and Jack of easily identifiable features in routinely stained sections. An alternative explanation can be suggested for the apparent absence of germ cells at this site, and that is the possibility of them developing into somatic cell of one type or another. Could one think of a possible candidate for this event in the thymus? The epithelial cells and lymphocytes of this organ seem out of the question, being that their branchial pouch and bone marrow derivations, respectively, are well established,(l4) However, this organ contains another cell type of unknown origin and function for which a thymic location seems most incongruous: the myoid cell. (4) This is a cell which has the phenotypical features of a skeletal (strlated) muscle cell and which is found scattered in the medullary portion of the organ. Suggested pr:ogenitors for these cells include the thymic epithelium, (9.l5) neural crest, (8) and prechordal mesoderm,(! t) but none of them have been 'conclusively proven. Could thymic myoid ce11s be of germ cell origin and thus provide an indirect evidence for the existence of the latter? The observation that foci of skeletal muscle differentiation are much more common in thymic than in gonadal germ cell tumor provides a little bit of evidence in favor of this· interpretation,.
DIAGNOSIS:
MALIGNANT TERATOMA
REFbRENCES: .,
l. Friedman NB, Van de Velde RL: Germ cell tumors in man, pleiotropic mice and continuity of germ plasm and somatoplasm. Hum Pathol 12:772-776, 1981.
2. Gonzalez-Crussi F: Extragonadal teratomas. In: Atlas.ofTumor Pathology, Second Series, Fascicle 18, Washington, DC, 1982, Armed Forces Institute of Pathology.
3. Green DM: Mucoid carcinoma of the breast with choriocarcinoma in itS metastases. Histopathology 16:504-506, 1990.
4. Hayward AR: Myoid cells in l;he human foetal thymus. J Pathol106:45-49, 1972.
5. Manivel C, Wick MR, Abenoza P, Rosai J: The occurrence of sarcomatou.s components in primary mediastinal germ cell tumors. Am J SurgPathol10:7ll-717, 1986.
6. Meares EM, Briggs EM: Occult seminoma of the testis masquerading as primary extragonadal germinal neoplasms. Cancer 30:300-306, 1988.
7. Morton KD, Burnett RA: Choriocarcinoma arising in transitional cell carcinoma of bladder: a case report. Histopathology 12:325-328, 1988.
8. Nakamura H, Ayer-Le Lievre C: Neural crest and thymic myoid cells. CurrTop Dev Biol20: 111-115, 1986.
9. Raviola E, Raviola G: Striated muscle cells in the thymus of reptiles and birds: an electron microscope study. Am J A nat 121:623, 1967.
10. Rosai J, Parkash V, Reuter VB: On the. origin of mediastinal germ cell tumors in males. Int J Surg Pathol (in press, July 1994).
11. Seifer R, Christ B: On the differentiation and origin of myoid cells in the avian thymus. AnatEmbryoll81:287-298, 1990.
12. Toner GC, Geller N, Lin SY, Bosl OJ: Extragonadal and poor risk nonseminomatous germ cell tumors. Survival and prognostic features. Cancer 67:2049-2057, 1991.
13. True LD, Otis CN, Delprado W, Scully RE, Rosai J: Spermatocytic seminoma of testis with sarcomatous transformation. A report of five cases. Am J Surg Patholl2:75-82, 1988.
14. Weller GL: Development of the thyroid, parathyroid and thymus glands in man. Contr Embryo! Carnegie Inst 24:93-140, 1933.
15. Zoltowska A: Myoid and epithelial cell differentiation in myasthenic thymuses. Thymus 17:237-248, 1991.
CASE #10
CLlNICALHISTORY:
An 18 year old male, previously healthy, presented in 8/91 with fevers, night sweats, chest pain, and dyspnea on exersion. A left anterior mediastinal mass was detected on CT scan. The HCG level was 479 and the AFP level was 6900. A fine needle aspirate was performed, and malignant cells consistent with germ cell tumor were found. The patient received six cycles of VIP chemotherapy which was completed on 12/13/91. The HCG and AFP levels returned to normal, but there was no radiographic change in the appearance of the tumor. On 1/20/92, a transverse sternotomy and bilateral thoracotomies were performed, with resection of the tumor.
GROSS DESCRIPTION:
The :320 gm nunor measured 10x9x6.3cm and was circumscribed, lobulated, and surrounded by a ibln capsule. Cut sections revealed numerous small cystic spaces within a tan, hemorrhagic tumor with fOCJ!l myxoid areas and calcifica.tions. The cysts contained thick mucil:tous material.
HISTOLOGICAL FEATURES:
Microscopically, the tumor shows teratomatous elements with a large component of endodermal structures, including well-formed large bowel mucosa, and mesoderm with innumerable lobules of cartiliige and bone. No other germ cell tumor elements are recognized. On closer examination, the mesenchymal tissue show widespread abnormalities including hypercellularity, nuclear pleomorphism, and abundant mitotic activity. The nuclear pleomorphism is quite striking, and involves both nonspecific fibrous tissues, cartilage, smooth muscle, and focally exhibits skeletal muscle differentiation. However, there is no large nodular proliferation of the pleomorphic spindle cells. Infiltration is not seen either within the stroma of the teratoma or into the adjacent tissues. Focal atypia is also noted in the glandular component of the teratoma
COMMENT:
Th.e fundamental nature of this neoplasm is obviously that of a germ cell tumor, in this case primary within the anterior mediastinum. This is second to the gonads as the most common site for germ cell tumors to develop, and the large majority of affected patients are young adult males.<4.8) Teratomas are the most common germ cell tumors to present in the mediastinum, although all germ cells tumor types (other than spermatocytic seminoma) have been deS<-Tibed in that location. Obviously, the possibility that these lesions may represent metastases from an occult gonadal primary should be excluded clinically. The tumors are thought to originate from primordial germ cells which are misplaced during migration in embryogenesis.
This case is of interest for two reasons. As evidenced by the preoperative elevation of HCG and AFP, this neoplasm most likely arose as a mixed geJID cell tumor, with elements of embryonal carcinoma and/or yolk sac tumor being irradicated by preoperative chemotherapy. The mature feratomatous comJ?Onents, being less sensitive to chemotherapy, remained. The perhaps more interesting question is that of the significance of the wide.spread nuclear pleomorphism. It is well accepted that chemotherapy itself may induce such changes both within the stromal and epithelial elements of teratomas. (3) On the other hand, it is equally well-documented that somatic neoplasms can arise in teratomas,(l.2.7,9.10) with sarcomas including rhabdomyosarcoma and angiosarcoma being particularly common in the mediastinum. (6) These. sarcomas may arise de novo or may follow trealtilent of the primary germ cell tumor with chemotherapy. Interestingly, tbe somatic neoplasms exhibit the same genetic abnormality (isochromosome 12p) that is commonly seen in germ cell tumors. It is of considerable importance to distinguish chemotherapy-related atypia from
somatic-type sarcoma. The presence of widespread nuclear changes without alterations in the architecture (i.e., chmeotli'erapy effect) has been associated with a slightly higher recurrence rate (not statistically significant)Pl the recurrences consisting of germ cell elements. On the other band, the occurrence of an invasive sarcomatous element is associated with a markedly worse prognosis,(3.5) with dissemination of the sarcomatous elements and resistance to germ cell chemotherapy. Treatment for those patients should therefore contain chemotherapy which is effective in sarcomas, in adclition to germ cell therapy. It is also important to distinguish this phenomenon from immature teratoma, w)lere the imlll!lture elements (usually of neural derivation) resemble the· normal immature tissues encountered during embryogenesis. Residual immature eJements after chemotherapy alsO impart a wot:Se prognosis, but not as dismal as that of sarcomatous transformation. In the seminar case, the lack of an invasive growth pattern in the pleomorphic mesenchymal elements and the presence of similar degrees of atypia in stromal cells of many clifferent types (as well as some epithelial elements) suggests that the changes are chemotherapy related. We have seen similar changes in many case of mediastinal teratomas as well as retroperitoneal metastases of germ cell tumors following chemotherapy, and despite the extreme degrees of atypia, no recurrences of sarcomatous elements have been seen, and these patients do not receive sarcoma-type chemotherapy.
DIAGNOSIS:
CYSTIC TERATOMA WITH CHEMOTHERAPY EFFECT
REFERENCES:
I. Ahlgren AD, Simrell CR, Triche TJ, Ozols R, Barsky SH. Sarcoma arising in residua:! testicular teratoma after cytoreductive chemotherapy. Cancer 54:2015· 2018, 1984.
2. Ahmed T, Bosl G, Hajdu Sl. Teratoma with malignant transformation in germ cell tumors in men. Cancer 56:860-863, 1985.
3. Davey DD, Ulbright TM, Loehrer PJ, Einhorn LH, Donohue JP, Williams SO. The significance of atypia within teratomatous metastases after chemotherapy for malignant germ cell tumors. Cancer 59:533-539, 1987.
4. Dehner LP. Germ cell tumors of the mediastinum. Semin Diagn Pathol 7(4):266-284, 1990.
5. Gonzalez-Vela JL, Savage PD, Manivel JC, Torkelson RN, Kennedy BJ. Poor prognosis of mecliastinal germ cell cancers containing sarcomatous components. Cancer 66: 1114-1116, 1990.
6. Manivel C, Wick MR, Abenoza P, Rosai J. The occurrence of sarcomatous components in primary mediastinal germ cell tumors. AmJ Surg Pathol 10(10):711-717, 1986.
7,. Nagabara N, Kitamura H, Kanisawa M, Ikeda A, ShirafK, Matsushita K. A testicular teratoma with rhabdomyosarcoma and seminoma. Acta Pathol Jpn 41(9):707-711., 1991.
8. Nichols CR, Saxman S, Wllliams SO, Loehrer PJ, Miller ME, Wright C, Einhorn LH. Primary mecliastinal nonseminomatous germ cell tumot:S . . Cancer 65:1641-1646,1990.
9. Ulbright TM, Clark SA, Einhorn AH. Angio~arcoma associated with germ cell tumors. Hum Pathol 16(3):268-272, 1985.
10. Ulbright TM, Loehrer PJ, Roth LM, Einhorn LH, William SD, Clark SA. The development of non-germ cell malignancies within germ cell tumors. A clinicopathologic study of 11 cases. Cancer 54:1824-1833, 1984.
case 1 2 3 4 ~ 6 7 8 9
10
Tumor and Tumor-Like Conditions of the Mediastinum
Juan Rosai, M.D. Department of Pathology Yale University School of MediciJ1e 310 Cedar Street New Haven, Connecticut 06510-8070
Dear Juan:
On behalf of the Anatomic Council of the ASCP, I thank you for agreeing to be the Chief Pre lector for the 1994 Spring Slide Seminar on "Mediastinum." It is my understanding that you are required to have a co-prelectqr whose name should be submitted to me so t hat i t can be approved by t he Council no later than November, 1991.
You wil l be receivi ng specific information regarding the seminar from Or. George F. Stevenson who is the Senior Vice President for Educational Activiti es Planning.
Again, let me express our appreciation for your taking on this program.
REF /njk
cc: George F. Stevenson, M. 0. Sara J. Hollingsworth, MT(ASCP)
Best personal regards ,
d~ Robert E. Fechner, M.D. ·chairman, Anatomic Counci l
HEALTH SCIENCES CENTER BOX 214, CHARlOTTESVILLE, VIRGINIA 22908 804-924-5'127
Yale University
January 3, 1991
Robert E. Fechner, M. D . Chairman, Anatomic ·Council University Qf Virginia Health Sciences center surgical Pathology Box 214 €harlottesville, VA 22908
Dear Bob:
Juan Rcmai; M.D. Proftuor & Direuor of Anat9mic Pathology Dtpartmtlll of Pdthology Sthool of Mtdiliut ) J o Ctdt~r Stru t P. O . Box J JJJ Ntw Havttl. Conntctiwt o65t0-8070
Thanks for your letter of November 29 and your invitation for me to be the Chief Prelector for the 1994 Spring Slide Seminar on "Mediastinum". This is an invitation that I am very glad and honored to accept. I will be sending you the name of the co-pr e lector some time before November 1991. I will be waiting to receive more specific instructions regarding the seminar from Dr. Stevenson.
Thank you again for having chosen me for this important task.
Best personal regards,
Juan Rosai, M. D. Professor of Pathology Director of Anatomic Pathology
;rR: rnl
0E6RGE .F. S'££VENSON, M.D.
·sENIOR V1CE .PRESLDENT
EDOCATIONAL ACTJVITTES PLANN!'.N"G
January 18, 1991
Juan Rosa i , ~1D Department of Pathology 'l.ale Univetsity School of Medicine 310 Cedar Street New Haven, CT 06510-8070
Dear Dr. Rosa i:
I was delighted to learn from Dr. Robert Fechner, Chairman of the CCE Council on Anatomic Pathology, that you have agreed to serve as prelector of the ASCP 1994 Spring AP Slide Seminar on Mediastinum.
Enclosed is the first fact sheet on this seminar. It will be updated as necessary as we proceed, with deletion of completed items and addition of new material, as appropriate. Please let us know if there are any probl ems.
I am sure that the officers of the Soci e.ty would wish to join me in thanking you for accepting this assignment, now one of the most important to ASCP.
Stevenson, MD
GFS:ce
Enclosure
cc: K. Ireland, MD, CCE Commissioner R. Fechner , MD, Chairman, Council on Anatomic Pathology S. Moran, Project Manager
AMERICAN SOCIET Y OF CLIN I CAL PATHOLOGISTS 210 0 WEST HA RRIS ON S TREET • OHJ C AOO. ILLINOIS O<Hll~-3708 • l31,~) 7 38 - lS!iti
Robert E. Fechner, M.D. Director, Surgical Pathology University of Virginia Medical School Health Sciences Center, Box 214 Charlottesville, VA 22908
Dear Bob:
December 26, 1991
In checking the section of "Things to Do" in my 1991 calendar, I realized that one of them was to give you the name of the co-pre lector for the 1994 Spring slide seminar on mediastinal tumors.
I propose for that individual to be Dr. David Klimstra. David has accepted a junior staff position at Memorial Sloan-Kettering Cancer Center, and is developing an expertise in the field of pulmonary and mediastinal pathology. He is an excellent speaker and an extremely promising academic pathologist. have already approached him regarding this matter, and he has tentatively agreed to participate, pending your approval.
Best personal regards,
Juan Rosai, M.D. Chairman, Department of Pathology
•''"'''' Sci•,~~ Vnin!l'*ily l W. Saat btkaon Parle. R<>ad
~CIL CHAIRMEN
...,E.T«hnu.~m
~Pathology
lltl ».. Jot~es, ~m '""""htbolo!IY
!Din!L J!i3ner, biD C"""llloloiO'
CAd R.l\leldJil>e'l!. ~m .... IOioiO'
fro A. Shulman, MD Ttllllfu.si.oo Medicine
JaMph A. KruJ~bt, ~m ataical Ch~mi&try
B. Scbifman, 1\o,.O iiao~lology
laH. Hilbornt!, MD !pKW1'opiclf
litbul D.O. McNeely, MO WcaJ lnformuticl5
OCPill'\' CO~L\IISSJONERS
t.MtU K. Srynes, MD
Mn1 W. ~ff.K~n.na, MD
.ftmeS. NON.nchuk, MD
JlQJECT DJR£CJ'ORS
llous A. BonfialJo, MD
~B. Cbandor, MD
....,. E. )tills. liD
Jeitt lt. t\dson, MD
l"nf P. Sher, ~m
January 15, 1992
. Juan Rosai , M.D. Chai rman , Department of Pathology Memorial Sloan-Kettering Cancer Center 1275 York Avenue New York, New York 10021
Dear Juan:
Dr. David Kl imstra will be fine for your co-pre lector for the 1994 Spring Slide Seminar on mediastinal tumors. Although I do not know him, your recommendation is more than enough. Further communications regarding the details of the seminar will be coming from ASCP headquarters in Chicago. If there is anything I can do, please let me know.
Best personal regards,
6~ Robert E. Fechner, M.D.
REF/ nj k
cc: George F. Stevenson, M.D . Sondra Moran MT(ASCP)
AMERICAN SOCIE'l'Y OF CLINICA L PATHOLOGI STS 2 100 Wll:.S 'l' t-IAI'll:.! .t SON S'I'R J£Bl 1J' • C H lt:AOO. I C...L I NO J S 00 61 2· 3 7 98 • ca I ::!1 7:1~- t :-l :~ O
![Index [] · MSBI 002 16cm MSBI 003 16cm MSBI 004 16cm MSBI 001 16cm 6. ID Bracelets SKID 001 16cm SKID 002 16cm SKID 003 16cm SKID 004 16cm SKID 005 16cm SKID 006 16cm 7. Brooches](https://static.documents.pub/doc/80x56/5f67f3ed65595c74fc237528/index-msbi-002-16cm-msbi-003-16cm-msbi-004-16cm-msbi-001-16cm-6-id-bracelets.jpg)
