Adenoid cystic carcinoma of the trachea and main-stem bronchus

Adenoid cystic carcinoma of the trachea and main-stem bronchus

J THORAC CARDIOV ASC SURG 1988;96:271-7 Adenoid cystic carcinoma of the trachea and main-stem bronchus A clinical, histopathologic, and immunohisto...

5MB Sizes 0 Downloads 79 Views

J

THORAC CARDIOV ASC SURG

1988;96:271-7

Adenoid cystic carcinoma of the trachea and main-stem bronchus A clinical, histopathologic, and immunohistochemical study Twelve cases of adenoid cystic carcinoma of the trachea and main-stem bronchus were histologically analyzed, and the results were examined with reference to the growth pattern of the tumor and the prognosis. The tumors were histologically classified into tubular, cribriform, and solid subtypes. Three histologic grades were established: grade I, tumors with tubular and cribrifonn subtypes but without solid subtype; grade II, tumors with tubular and cribrifonn subtypes in which the solid subtype comprised less than 20 % of the area; grade III, tumors in which the solidsubtype comprisedmore than 20 % of the area. Three gross infiltrating types were established: type I, entirely intraluminal; type II, predominantly intraluminal; type III, predominantly extraluminal. In most cases histologic grade correlated with gross tumor type; that is, grades, I, II, and ill were grossly types I, II, and III, respectively, The tumors infiltrating along the tracheobronchial wall were of the tubular or cribrifonn subtype, but not of the solid subtype. In two patients who died of distant metastasis, the histologic studies revealed the solid subtype. Immunohistochemical analysis demonstrated that the tubular subtype was the most differentiated fonn and the solidsubtype, the most undifferentiated fonn. The histologic subtype of adenoid cystic carcinoma of the tracheobronchial tree was an important factor in the growth pattern of the tumor and the prognosis.

Hiroaki Nomori, MD,a Shizuka Kaseda, MD,a Kouichi Kobayashi, MD,a Tsuneo Ishihara, MD,a Noboru Yanai, MD,b and Chikao Torikata, MD,c Tokyo and Ibaragi, Japan

Because of the advances made in tracheobronchial surgery,':' resection of adenoid cystc carcinoma of the trachea and main-stem bronchus has recently been attempted. However, because of the rarity of adenoid cystic carcinoma, its clinicopathologic features have never been reported. Adenoid cystic carcinoma is a specific variant of adenocarcinoma that occurs most commonly in the major and minor salivary glands. In adenoid cystic carcinoma of the salivary glands, several authors have demonstrated a correlation between the histologic pat-

From the Departments of Surgery' and Pathology,' School of Medicine of Keio University, Tokyo, Japan, and the Department of Surgery, Seiransou National Sanatorium Hospital: Ibaragi, Japan. Received for publication July 31, 1987. Accepted for publication Dec. 2, 1987. Address for reprints: Hiroaki Nomori, MD, Department of Surgery, School of Medicine, Keio University, 35 Shinanomachi, Shinjukuku Tokyo 160, Japan.

tern of the tumor and its clinical behavior.r" The most characteristic histologic pattern is the cribriform subtype, in which the cells are arranged in nests or sheets that are fenestrated by round or oval spaces. Occasionally, the tumors have a predominantly solid cellular growth and an anaplastic appearance. Eby, Johnson, and Barker? found that the solid histologic pattern of adenoid cystic carcinoma of the salivary gland was associated with distant metastasis and a worse prognosis than the cribriform type. In contrast, Perzin, Gullane, and Clairmont" identified the tubular subtype as the most differentiated form of adenoid cystic carcinoma, because its prognosis is better than that of the cribriform type. The purpose of this study was to examine whether the histologic pattern of adenoid cystic carcinoma of the tracheobronchial tree was correlated with the growth pattern of the tumor and the prognosis. Immunohistochemical analysis was also performed to determine the characteristics and direction of differentiation of each histologic subtype. 271

272

The Journal of Thoracic and Cardiovascular Surgery

Nomori et al.

Fig. 1. Tubular subtype of adenoid cystic carcinoma. (Hematoxylin and eosin, X200.)

Fig. 2. Cribriform subtype of adenoid cystic carcinoma. (Hematoxylin and eosin, x200.)

Table I. Characteristics of patients with adenoid cystic carcinoma of the trachea and main-stem bronchus Case No.

Age. sex

Tumor location

I 2 3 4 5 6 7 8 9 10

39, M 28, M 29, M 42, F 44,M 52, M 68, M 42, F 47, F 26, F

11

47, F

12

37, M

Lower trachea Middle trachea Middle trachea Lower trachea Lower trachea Upper trachea Upper trachea Middle trachea Lower trachea Left main bronchus Right main bronchus Right main bronchus

Operative procedure

Surgical margin

Postop. radiation (Gy)

+

45

+

30

+

63 78 87 86 49 49 21 9 4 41

Wedge resection Sleeve resection Sleeve resection Carinal resection CarinaI resection Sleeve resection Sleeve resection Sleeve resection CarinaI resection Pneumonectomy

+ + +

50 36 46 40 60

Sleeve pneumonectomy

+

45

Carinal resection

Patients and methods Twelve patients with adenoid cystic carcinoma of the trachea and main-stem bronchus who had undergone resection at Keio University Hospital and Seiransou National Sanatorium Hospital between 1964 and 1987 were investigated. The characteristics of the patients are presented in Table I. There were seven men and five women with an age range between 26

Infiltrating type

Histologic type

II II II I III II I II II 1II

II 1II II I 1II II I I I 1II

78 rno, alive

II

II

36 mo, alive

II

II

Outcome mo, dead mo, dead mo, alive mo alive rno, alive mo, alive mo, alive mo, alive rno, alive rno, dead

and 68 years. Nine tumors were located in the trachea and three in the main bronchus. Two tumors in the trachea were located in the upper third, three in the middle third, and four in the lower third of the trachea. Except for the first patient in 1964, who underwent wedge resection of the trachea, the eight patients with adenoid cystic carcinoma of the trachea underwent sleeve resection of the trachea or carinal resection. Three

Volume 96 Number 2 August 1988

Adenoid cystic carcinoma

273

Table II. Immunohistochemical staining of normal tracheobronchial gland and adenoid cystic carcinoma Histologic type Normal gland Acinar cells Myoepithelial cells Duct epithelial cells Adenoid cystic carcinoma Tubular subtype Cribriform subtype Solid subtype

Antibodies Keratin

++ ++ ++ +

SC

S/OO

Lactoferrin

++

++

++

++ +

++ +

++ +

SC, Secretory component; ++, intensely staining; +, weakly staining; -, negative staining.

patients with adenoid cystic carcinoma of the main-stem bronchus underwent pneumonectomy, sleeve pneumonectomy, and carinal resection, respectively. Eight patients received postoperative adjuvant radiotherapy. Sections stained with hematoxylin and eosin were examined, and three histologic subtypes were identified: tubular, cribriform, and solid, The tubular subtype was characterized by single-lumen tubular units with smaller nests than the cribriform subtype (Fig. 1). The cribriform subtype was characterized by tumor cells arranged in nests or sheets fenestrated by round and oval spaces (Fig. 2), The solid subtype was characterized by large cellular nests that had no glandular spaces (Fig, 3), Three histologic grades were classified according to the area percentage of each histologic subtype: grade I, a tumor composed of tubular and cribriform subtypes without solid subtype; grade II, a tumor with tubular and cribriform subtypes in which the solid subtype comprised less than 20% of the area; grade III, a tumor in which the solid subtype comprised more than 20% of the area. The histologic grade in each case was determined in the section of the main tumor from the primary site. From the histologic section of the main tumor, three gross infiltrating types were classified as follows: type I (entirely intraluminal type), in which the tumor appears to be an entirely intraluminal and semipedunculated mass and scarcely invades beyond the cartilage plate (Fig. 4); type II (predominantly intraluminal type), in which the tumor is an intraluminal mass with extension beyond the cartilage plate (Fig, 5); and type III (predominantly extraluminal type), in which the tumor predominantly extends beyond the cartilage plate to form an extraluminal projection (Fig, 6), Immunohistochemical study was performed on formalinfixed paraffin-embedded tissue by the avidin-biotin-peroxidase complex (ABC) method, as previously described.r" Antibodies used in this study were rabbit antisera to human secretory component (SC), S I00 protein, lactoferrin, and keratin. Antibodies to SC, Sloo protein, and lactoferrin were used as markers to acinar cell differentiation, and antibody to keratin was used as the marker to epithelial differentiation, These antisera were purchased from Dakopatts, Ltd., Copenhagen. After deparaffinization, slides were treated with hydrogen peroxide (0.5%) in methanol to abolish endogenous

Fig. 3. Solid subtype of adenoid cystic carcinoma. (Hematoxylin and eosin, x2oo,)

peroxidase activity and then incubated sequentially with (I) normal serum; (2) primary antibody of keratin (1/200), SC (1/100), sioo protein (1/100), and lactoferrin (1/100); (3) biotinyl antirabbit immunoglobulin G antibody; and (4) complexes of avidin OH and biotinyl-peroxidase (Vectastain ABC kit; Vector Laboratories. Burlingame, Calif.). Antibody localization was observed by incubating the slides for 5 minutes with a freshly prepared solution of 3,3' -diarninobenzidine tetrahydrochloride (Wako Dotite Chemicals, Tokyo) and 0,005% hydrogen peroxide in Tris buffer 0.5 mol/L, pH 7.6. Slides were counterstained with methyl green. Specimens of normal trachea and bronchus were used as controls.

Results Table I shows the relation between the histologic grade and gross infiltrating type for each case, Histologically, four cases were classified as grade I, five as grade II, and three as grade III. Among the gross infiltrating types, two cases were classified as type I, eight as type II, and two as type III. In most cases the histologic grade correlated with the gross infiltrating type; that is, histologic grades I, II, and III were grossly types I, II, and III, respectively, Most adenoid cystic carcinomas that were histologically composed of the tubular and cribriform subtypes without the solid subtype presented

The Journal of Thoracic and Cardiovascular

2 7 4 Nomori et al.

Surgery

Fig. 4. Gross infiltrating type I (entirely intraluminal type).

r?~~~-'.,

'II

Fig. S. Gross infiltrating type II (predominantly intraluminal type).

Fig. 6. Gross infiltrating type III (predominantly extraluminal type).

Volume 96 Number 2 August 1988

Adenoid cystic carcinoma

275

Fig. 7. Immunohistochemical studies for adenoid cystic carcinoma. A, Secretory component in tubular and cribriform subtypes. B, S I00 protein in tubular and cribriform subtypes. C, Lactoferrin in tubular and cribriform subtypes. D, Keratin in tubular and cribriform subtypes. The tubular subtype was more intensely stained with these antibodies than the cribriform subtype. (Original magnification X200.)

as an intraluminal mass and almost never extended beyond the cartilage plate. All of the adenoid cystic carcinomas that were histologically composed of the solid subtype extended beyond the cartilage plate, and tumors that were composed of a more solid subtype infiltrated more extensively into the extraluminal region. All of the specimens showed infiltration of the tumors in the submucosa or perineural lymphatic spaces along the tracheobronchial wall. The tumors that showed this

local infiltration along the tracheobronchial wall were composed of the tubular or cribriform subtype, and not of the solid subtype. Table I shows the postoperative prognosis of each case. Patients 1 and 2 died of distant metastasis 63 and 79 months, respectively, after the operation. Their respective histologic grades were II and III. In patient 10, a part of the tumor could not be resected grossly, and after the operation the remaining tumor extended to the trachea. Despite postoperative radiotherapy and laser

The Journal of Thoracic and Cardiovascular Surgery

27 6 Nomori et of.

therapy, the patient died of respiratory failure caused by the tracheal tumor 41 months after the operation. This case was grade III histologically. Table II shows the immunohistochemical findings of the normal tracheobronchial gland and each histologic subtype of adenoid cystic carcinoma. In the normal tracheobronchial gland, serous cells of acini were stained for SC, SIOO protein, and lactoferrin, whereas the mucus cells were not. The myoepithelial cells around acini and the epithelial cells of ducts were stained intensely for keratin. In adenoid cystic carcinoma, the tubular and cribriform subtypes were both stained for SC, SIOOprotein, lactoferrin, and keratin. Staining with these antibodies was more intense in the tubular than in the cribriform subtype (Fig. 7). The solid subtype was not stained with these antibodies. However, the tumor cells of solid area mixed with the cribriform pattern, which was classified as the cribriform subtype in this study, were stained weakly for keratin. Discussion Adenoid cystic carcinoma occurs most commonly in the major and minor salivary glands, and its histologic subtypes are classified into tubular, cribriform, and solid subtypes." In this study three histologic grades were defined according to the percentage of each histologic subtype in the tumor, and they were found to correlate with the gross infiltrating type. Tumors composed of more solid histologic subtypes showed more extensive extraluminal growth. In adenoid cystic carcinoma of the salivary gland, Szanto and associates' reported that tumors of a predominantly solid pattern were larger than those of other subtypes. It was suggested that the solid subtype of adenoid cystic carcinoma might grow more extensively and invasively than the cribriform and tubular subtypes. In adenoid cystic carcinoma of the salivary gland, the cribriform and tubular subtypes were associated with more frequent local recurrence than the solid subtype.' In this study tumors that infiltrated the submucosa or perineural lymphatic spaces along the tracheobronchial wall were histologically identified as the cribriform or tubular subtype, but not the solid subtype. It was suggested that the cribriform and tubular subtypes might be associated with local recurrence after resection of adenoid cystic carcinoma of the trachea and bronchus. Histopathologic studies of adenoid cystic carcinoma of the salivary gland have shown that the overall prognosis in terms of distant metastasis and survival is worst for the solid subtype.r" In this study the two patients who died of distant metastasis both had tumors

of the solid subtype histologically. It thus appears that, in adenoid cystic carcinoma of the trachea and bronchus, tumors of the solid subtype are particularly associated with distant metastasis. Adenoid cystic carcinoma of the salivary gland is immunohistochemically stained with antibodies to keratin, SC, SIOO protein, and lactoferrin.":" In this study the tubular and cribriform subtypes were stained with these antibodies, whereas the solid subtype was not. In the normal tracheobronchial gland, serous cells of acini were stained for SC, SIOO protein, and lactoferrin, and myoepithelial cells and duct epithelium were stained for keratin. Our immunohistochemical study demonstrated that adenoid cystic carcinoma of the trachea and bronchus showed differentiation to the acinar cells, myoepithelial cells, and/or duct epithelial cells of the tracheobronchial gland. According to the staining activity of each histologic subtype, the tubular subtype appeared to be the most differentiated and the solid subtype the most undifferentiated form of adenoid cystic carcinoma. On the basis of our data, we believe that the histologic subtype of adenoid cystic carcinoma in the tracheobronchial tree is an important factor in the growth pattern of the tumor and the prognosis. We thank Dr. H. Iri and Dr. M. Tashiro of the Division of Surgical Pathology, Central Laboratories, Keio University, for use of the formalin-fixed paraffin-embedded tissue for the immunohistochemical study. REFERENCES I. Ishihara T, Nemoto E, Kikuchi K, Kato R, Kobayashi K. Does pleural bronchial wrapping improve wound healing in right sleeve lobectomy? J THORAC CARDIOVASC SURG 1985;89:665-72. 2. Pearson FG, Todd TRJ, Cooper JD. Experience with primary neoplasms of the trachea and carina. J THORAC CARDIOVASC SURG 1984;88:511-8. 3. Grillo HC. Tracheal tumors: surgical management. Ann Thorac Surg 1978;26: 112-25. 4. Szanto PA, Luna MA, Tortoledo ME, White RA. Histologic grading of adenoid cystic carcinoma of the salivary glands. Cancer 1984;54: I062-9. 5. Nascimento AG, Amaral ALP, Prado LAF, Kligerman J, Silveira TRP. Adenoid cystic carcinoma of salivary glands. Cancer 1986;57:312-9. 6. Matsuba HM, Spector GJ, Thawley SE, Simpson JR, Mauney M, Pikul FJ. Adenoid cystic salivary gland carcinoma. Cancer 1986;57:519-24. 7. Eby LS, Johnson DS, Barker HW. Adenoid cystic carcinoma of the head and neck. Cancer 1972;29: 1160-8. 8. Perzin KH, Gullane P, Clairmont AC. Adenoid cystic

Volume 96 Number 2 August 1988

carcinoma arising in salivary glands. Cancer 1978;42:26582. 9. Nomori H, Shimosato Y, Kodama T, Morinaga S, Nakajima T, Watanabe S. Subtypes of small cell carcinoma of the lung. Hum Pathol 1986;17:604-13. 10. Nomori H, Watanabe S, Nakajima T, Shimosato Y, Kameya T. Histiocytes in nasopharyngeal carcinoma in relation to the prognosis. Cancer 1986;57:100-5. II. Wick MR, Swanson PE. Primary adenoid cystic carcinoma of the skin. Am J Dermatopathol 1986;8:2-13.

Adenoid cystic carcinoma

277

12. Kahn HJ, Baumal R, Marks A, Dardick I, Nostrand AWP. Myoepithelial cells in salivary gland tumors. Arch Pathol Lab Med 1985;109:190-5. 13. Warner TFCS, Sea IS, Azen EA, Hafez GR, Zarling T A. Immunohistochemistry of acinic cell carcinomas and mixed tumors of salivary glands. Cancer 1985;56: 2221-7.