Adenoma Versus Carcinoid Tumor of the Middle Ear: a Study of 48 Cases and Review of the Literature Kevin R
Total Page:16
File Type:pdf, Size:1020Kb
Adenoma versus Carcinoid Tumor of the Middle Ear: a Study of 48 Cases and Review of the Literature Kevin R. Torske, D.D.S., M.S., Lester D.R. Thompson, M.D. Department of Endocrine and Otorhinolaryngic–Head and Neck Pathology, Armed Forces Institute of Pathology, Washington, D.C. showed adenomas and carcinoid tumors of the mid- Carcinoid tumors and adenomas of the middle ear dle ear to be essentially indistinguishable benign are rare neoplasms of indeterminate relationship to tumors. Middle ear adenoma most correctly de- one another. Indeed, the literature is devoid of a scribes their morphologic features and clinical be- large comprehensive series that evaluates the clini- havior, although neuroendocrine adenoma of the cal, histologic, and immunophenotypic features of middle ear may be a more accurate designation. these tumors and their potential relationship. Forty-eight cases of middle ear adenoma between KEY WORDS: Adenoma, Carcinoid, Immunohisto- 1970 and 1995 were identified in the files of the chemistry, Middle ear, Prognosis. Armed Forces Institute of Pathology. All cases were Mod Pathol 2002;15(5):543–555 evaluated for cytomorphology and architectural pattern, in addition to their reactivity with various Adenomas and carcinoid tumors of the middle ear immunohistochemical reagents. Clinical follow-up are rare neoplasms with similar, if not indistin- was also obtained. A comprehensive review of the guishable, morphologies. First described in 1976 (1, literature was performed with an eye toward corre- 2) and further elaborated upon in 1980 (3), these lating any distinct differences or similarities be- tumors have been the subject of much debate with tween carcinoid tumors and adenomas of the mid- regard to their similarities, differences, etiologies, dle ear. The patients included 21 women and 27 and prognoses. Limited mostly to case reports and men, aged 20 to 80 years (mean, 45.0 y). Patients small series, the English-language literature is de- experienced hearing loss, mass, and/or pain for a void of a large comprehensive evaluation of these mean duration of 1.7 years. The mean tumor size tumors with respect to their histomorphology, im- was 0.8 cm, with six tumors extending beyond the munohistochemical reactivity, treatment out- middle ear. Histologically, the tumors were moder- comes, and clinical behavior. We elected to under- ately cellular and unencapsulated, arranged in glan- take such a study with an eye toward identifying dular, trabecular, and solid patterns composed of any distinct separation of these two tumors. small cells with “salt and pepper” nuclear chroma- tin distribution. The tumor cells were immunoreac- tive with keratin, keratin 7, chromogranin, and hu- MATERIALS AND METHODS man pancreatic polypeptide. All patients had The records of 68 patients with tumors diagnosed surgery. No patients died with their disease (mean as “middle ear adenoma,” “ceruminal gland adeno- follow-up, 15.7 y). Eight patients developed recur- ma,” and “atypical middle ear adenoma” were rences that were treated surgically and were with- identified in the files of the Otorhinolaryngic-Head out evidence of disease at last follow-up (mean, & Neck Registry at the Armed Forces Institute of 15.1 y). Our study and the review of the literature pathology from 1970 to 1995. No records of “carci- noid tumors” of the middle ear were located in the registry. This lack was considered not to be due to Copyright © 2002 by The United States and Canadian Academy of Pathology, Inc. the true absence of such a tumor but more to the VOL. 15, NO. 5, P. 543, 2002 Printed in the U.S.A. preferred designation of this tumor as “middle ear Date of acceptance: January 30, 2002. The opinions or assertions contained herein are the private views of the adenoma” by Dr. Hyams and his colleagues within authors and are not to be construed as official or as reflecting the views of the Department of Otorhinolaryngic Pathology. the Department of the Navy or Department of Defense. Address reprint requests to: Lester D.R. Thompson, M.D., Department of These 68 patients were identified in a review of Endocrine and Otorhinolaryngic–Head and Neck Pathology, Building 54, 1,703 patients with primary benign or malignant Room G066–11, Armed Forces Institute of Pathology, 6825 16th Street NW, Washington, DC 20306–6000; e-mail: [email protected]; fax: neoplasms of the ear seen in consultation during 202-782-3130. this same time period. However, 20 patients were 543 excluded from further consideration because of at The macroscopic pathologic observations noted least one of the following reasons: (1) paraffin within this study were garnered from the individual blocks were unavailable for additional sections or gross descriptions of the neoplasms by the contrib- immunophenotypic analysis (2); the original sub- uting pathologists. Hematoxylin and eosin–stained mitted case did not have sufficient demographic slides from all patients were reviewed for morpho- information supplied from which to obtain ade- logic assessment of middle-ear adenoma/carci- quate follow-up information; and (3) the cases were noid, as originally defined by both Hyams et al. and diagnosed indefinitely, using terms such as consis- Murphy et al. (2, 3) A number of macroscopic and tent with, suggestive of, or suspicious for, which histologic observations were recorded for each tu- usually indicated a lack of sufficient diagnostic ma- mor as follows: exact tumor location; tumor size terial from which to proffer a definitive diagnosis. (greatest dimension in centimeters); tumor encap- Therefore, the remaining 48 patients with middle- sulation (presence or absence); tumor extension ear adenomas (2.8% of all ear neoplasms within the (mastoid bone, eustachian tube, and/or external files of the AFIP) compose the subject of this study auditory canal); architectural pattern of growth based on adequate hematoxylin and eosin-stained (solid, trabecular, organoid, infiltrating, and/or slides to make a definitive diagnosis and adequate glandular); surface origin or pagetoid spread; pres- demographic information for initial attempt at ence or absence of necrosis; tumor cellularity (low, follow-up. Forty-three cases were obtained from moderate, or high); mitotic figures (number of mi- civilian sources, including university medical cen- totic figures per 10 high power fields [magnification ters and foreign contributors, four cases from mil- at 40ϫ with a 10ϫ objective lens using an Olympus itary hospitals, and one case from a Veterans Ad- BX40 microscope]); atypical mitotic figures (present ministration medical center. or absent, and defined by abnormal chromosome Materials within the Institute’s files were supple- spread, tripolar or quadripolar forms, circular mented by a review of the patient demographics forms, or indescribably bizarre); nuclear pleomor- (gender, age) and symptoms at presentation (mass, phism (mild, moderate or severe); presence or ab- hearing loss, bleeding, infection, pain, nerve paral- sence of nucleoli; and the presence of other micro- ysis, discharge, equilibrium changes, tinnitus, scopic pathologic findings. and/or popping), including duration. In addition, Immunophenotypic analysis was performed in we reviewed radiographic, surgical pathology, and all cases with suitable material by the standardized operative reports, and obtained follow-up informa- avidin–biotin method of Hsu et al. (4) employing tion from oncology data services by written ques- 4-m-thick, formalin-fixed, paraffin-embedded tionnaires or direct communication with the treat- sections. Table 1 documents the pertinent, com- ing physician(s) or the patient. Follow-up data, mercially available immunohistochemical antibody available for 45 patients, included information re- panel used. The analysis was performed on a single garding tumor location, presence of recurrent or representative block in each primary tumor. When metastatic disease, treatment modalities used, and required, proteolytic antigen retrieval was per- the current patient and disease status. This clinical formed by predigestion for 3 minutes with 0.05% investigation was conducted in accordance and Protease VIII (Sigma Chemical Co., St. Louis, MO) compliance with all statutes, directives, and guide- in a 0.1 M phosphate buffer, pH of 7.8, at 37 ° C. lines of the Code of Federal Regulations, Title 45, Cellular conditioning to achieve antigen enhance- Part 46, and the Department of Defense Directive ment (recovery) was performed by using formalin- 3216.2 relating to human subjects in research. fixed, paraffin-embedded tissue treated with a buff- Table 1. Immunohistochemical Panel Primary Antigen or Antibody Company Dilution Antigen Recovery Antibody Cytokeratin Boehringer Mannheim AE1/AE3 mm Biochemicals, Indianapolis, IN, 1:50 Enzyme digestion CK1 mm and DAKO, Carpinteria, CA 1:200 Keratin 7 mm DAKO 1:200 Enzyme digestion Keratin 20 mm DAKO 1:50 Enzyme digestion CAM5.2 mm Becton Dickson, San Jose, CA 1:100 Enzyme digestion Chromogranin mm Boehringer Mannheim 1:3,200 None Synaptophysin rp DAKO Neat Enzyme digestion Serotonin mm DAKO 1:25 None Neuron-specific enolase rp Ventana, Tucson, AZ Neat None Human pancreatic polypeptide mm DAKO 1:1000 None S-100 protein rp DAKO 1:800 None Vimentin mm BioGenex Labs, San Ramon, CA 1:400 Microwave recovery mm, mouse monoclonal; rp, rabbit polyclonal. 544 Modern Pathology ered citric acid solution and heated for 20 minutes changes (n ϭ 7 patients), tinnitus or popping (n ϭ in a calibrated microwave oven. Afterwards, the 6