Application of anti-leukocyte common antigen and anti-cytokeratin antibodies to the biopsy diagnosis of gastric large cell lymphoma

Application of Anti-Leukocyte Common Antigen and Anti-Cytokeratin Antibodies to the Biopsy Diagnosis of Gastric Large Cell Lymphoma PATRICK J. DEAN, MD,*t SHAMIM M. MOINUDDIN, MD,* AND LORI D. EMERSON, MD* Definitive diagnosis of gastric large cell lymphoma and its distinction from anaplastic carcinoma in endoscopic biopsy material may be problematic. To assess the utility of immunohistochemical studies in routinely processed, paraffin-embedded tissue in this situation, we applied immunostaining for leukocyte common antigen (LCA) and cytokeratin in 17 cases diagnosed on biopsy as undifferentiated malignant tumor but proved on resection to be primary gastric large cell lymphoma. Clinical and endoscopic features failed to distinguish lymphoma from carcinoma in these cases. Immunoreactivity for LCA occurred in 15 cases (88 per cent) and was correctly and readily interpreted on blinded evaluation. Open review increased the yield to 16 cases (94 per cent). Tumor cells were uniformly negative for cytokeratin; however, staining of adjacent epithelium for cytokeratin provided additional confirmation of the lymphoid nature of the tumor. The one case in which excessive background staining precluded interpretation consisted of a single biopsy specimen of necrotic tumor. We conclude that antibodies to LCA and cytokeratin are sensitive, specific, and reliable diagnostic adjuncts that are useful in the definitive biopsy diagnosis of gastric large cell lymphoma. HUM PATHOL18:918--923, 1987.

that biopsy diagnosis o f large cell l y m p h o m a o f the stomach would represent a logical setting in which to test the applicability o f these immunohistochemical markers. T o this end, we applied i m m u n o s t a i n i n g for LCA a n d cytokeratin to 17 cases diagnosed on biopsy as u n d i f f e r e n t i a t e d m a l i g n a n t t u m o r b u t proved on resection to be primary gastric large cell lymphoma.

MATERIALSAND METHODS

Identification of undifferentiated malignant t u m o r in endoscopic biopsy samples o f gastric mucosa poses little difficulty. However, the distinction o f anaplastic c a r c i n o m a f r o m large cell l y m p h o m a in small tissue fragments may be a formidable task. 1-H Because gastric l y m p h o m a s typically simulate carcin o m a in their clinical presentation and endoscopic appearance, definitive diagnosis depends on accurate pathologic interpretation o f the biopsy specimen. Despite the use o f c o m p l e m e n t a r y techniques such as h i s t o c h e m i c a l stains f o r m u c i n , electron microscopy, 6,8 a n d i m m u n o p h e n o t y p i n g , 8 simple biopsy d i a g n o s i s o f gastric large cell l y m p h o m a r e m a i n s problematic. Recent advances in the application o f i m m u n o h i s t o c h e m i c a l t e c h n i q u e s to p a r a f f i n - e m b e d d e d tissue have b r o u g h t increased precision to the diagnosis o f u n d i f f e r e n t i a t e d m a l i g n a n t neoplasms. In particular, use o f antibodies directed against leukocyte c o m m o n antigen (LCA) 12-14 and low-molecularweight cytokeratin 15-18 has facilitated the separation o f epithelial f r o m l y m p h o i d tumors. We postulated From the Departments of Pathology, *Baptist Memorial Hospital and tUniversity of Tennessee, Memphis, Tennessee. Accepted for publication 25 October 1986. Address correspondence and reprint requests to Dr. Dean: Department of Pathology, Baptist Memorial Hospital, 899 Madison Avenue, Memphis, TN 38146. 0046-8177/87 $0.00 + .25

T h e surgical pathology files o f the Baptist Memorial Hospital for the years 1979 t h r o u g h 1985 were reviewed to identify biopsy cases classified as u n d i f f e r e n t i a t e d malignant neoplasm o f the stomach. Seventeen cases met the following criteria for selection: 1) a definitive diagnosis o f l y m p h o m a had not been r e n d e r e d ; 2) histochemical stains for mucin including mucicarmine, periodic a c i d - S c h i f f , a n d alcian blue were negative; 3) on gastric resection the t u m o r was large cell l y m p h o m a ; 4) the t u m o r was j u d g e d clinically a n d pathologically to be a primary gastric neoplasm. Six gastric biopsy cases were chosen as controls. T h r e e were anaplastic carcinomas shown on resection to be gland-forming, m u c i n - p r o d u c i n g adenocarcinomas, a n d three were secondary large cell lymphomas in patients with previous or c o n c u r r e n t lymphomas at o t h e r anatomic sites (one in Waldeyer's ring, one in mediastir/al lymph nodes a n d lung, and o n e in t h y r o i d g l a n d ) . Six r e s e c t e d g a s t r i c lymp h o m a s - - t h r e e fixed in formalin a n d three in B5 sol u t i o n - a l s o served as controls. T h e original light microscopic slides were available for review in all cases. In three study cases, the tissue had been fixed in formalin; the remaining 14 study cases as well as the six control cases were fixed in Hollande's modification o f Bouin's solution. All cases were e m b e d d e d in paraffin, cut at 4 I~m, and stained with hematoxylin-eosin. Tissue sections for immunohistochemistry were cut at 4 I~m a n d placed in a 60~ oven overnight to maximize a d h e r e n c e o f tissue to glass slides. Sections were then deparaffinized, hydrated, a n d treated with p r e h e a t e d 0.1 p e r c e n t P r o n a s e ( C a l b i o c h e m Behring, San Diego, California) in phosphate-buffered saline for 15 minutes in a 37~ oven prior to application o f t h e first blocking solution. Mouse

918

LCA AND CYTOKERATININ G,~TRIC LARGECELLLYMPHOMA(Dean et al.) TABLE t.

Selected Clinical Data from 17 Cases of Primary Gastric Large Cell Lymphoma

Age (yr)/Sex

Symptoms and Signs

Endoscopic Appearance

Location*

Endoscopic Impression

1

4 I/M

4-cm nodular mass

LC

Probable lymphoma

2

7 I/F

10-cm ulcer

LC

Carcinoma or lymphoma

3

17/F

5-cm ulcer

LC

Malignant ulcer

4

34/M

Malignant tumor

61/M 70/F 58lM

LC C C

Carcinoma Carcinoma Carcinoma

8 9

75/M 65lF

4-cm nodular mass, rigid folds 5-cm ulcer 7-cm nodular mass 10-cm circumferential ulcerated mass 2-cm polyp Large nodular mass

C

5 6 7

A GC

Polyp Carcinoma or iymphoma

10 11 12

73lM 73lM 7 I/F

Nodular mass 6-cm ulcer 5-cm ulcer

C LC LC

Probable lymphoma Carcinoma Carcinoma

13

76lF

2-cm ulcer

LC

14 15

64/M 62lF 63lM

7-cm ulcerated mass 2-cm mass, thick folds 8-cm ulcerated mass

GC C

16 17

57/F

Epigastric pain, weight loss Epigastric pain, weight loss, anorexia Epigastric pain, nausea and vomiting Epigastric pain, nausea and vomiting Epigastric pain Dysphagia, weight loss Epigastric pain, weight loss, early satiety Anemia, weight loss Epigastric pain, weight loss, anorexia, anemia Dysphagia, early satiety Epigastric pain, anemia Epigastric pain, nausea and vomiting, anorexia Epigastric pain, anorexia Epigastric pain Epigastric pain, dysphagia Epigastric pain, weight loss Epigastric pain, nausea, weight loss

3-cm ulcer

LC

Probable malignant ulcer Carcinoma Carcinoma or lymphoma Carcinoma or lymphoma Carcinoma

Case

GG

* c, cardia; LC, lesser curvature; GC, greater curvature; A, antrum.

RESULTS

m o n o c l o n a l anti-LCA (Dako Corp., Santa Barbara, California) at a I:10 dilution and commercially prediluted m o u s e m o n o c l o n a l anti-cytokeratin (Becton Dickinson, M o u n t a i n View, California) antibodies were employed. Slides were processed by the avidinbiotin-peroxidase c o m p l e x m e t h o d (Vectastain ABC Mouse IgG Kit, V e c t o r L a b o r a t o r i e s , B u r l i n g a m e , California). T h e p e r o x i d a s e reaction p r o d u c t was localized using as c h r o m o g e n 3 , 3 ' - d i a m i n o b e n z i d i n e tetrahydrochloride (Aldrich Chemical Company, Milwaukee, Wisconsin). All sections were c o u n t e r stained with hematoxylin. B l i n d e d e v a l u a t i o n o f the i m m u n o h i s t o c h e m i cally stained tissue sections was p e r f o r m e d by one o f us (SMM) who had no knowledge o f clinical information, a p p e a r a n c e o f t u m o r o n h e m a t o x y l i n - e o s i n staining, o r n a t u r e o f the antisera used. T h e reviewer was asked w h e t h e r t u m o r could be identified in the tissue sections, w h e t h e r t u m o r cells exhibited a positive o r n e g a t i v e s t a i n i n g r e a c t i o n , a n d to specify r e a s o n s f o r e q u i v o c a l i n t e r p r e t a t i o n s if such occurred. C o m p l e t e clinical records including t u m o r registry files were available for review in all cases. T h e y were analyzed to d e t e r m i n e clinical and endoscopic findings and to c o n f i r m that all study cases were prim a r y gastric l y m p h o m a s .

Clinical Findings

S e l e c t e d clinical d a t a f o r the 17 p a t i e n t s are shown in T a b l e 1. Patients r a n g e d in age f r o m 17 to 76 years, with a m e a n o f 61 years. T h e r e were nine m e n a n d eight women. F o u r t e e n patients p r e s e n t e d with u p p e r a b d o m i n a l pain, a n d eight had s u f f e r e d weight loss r a n g i n g f r o m 3 to 13.5 kg. Disturbances o f eating w e r e c o m m o n and included a n o r e x i a in six patients, n a u s e a o r vomiting in f o u r , d y s p h a g i a in three, a n d early satiety in two. T h r e e patients had blood loss a n e m i a (hemoglobin, 8.0, 8.5, and 9.1 g/dl) with H e m o c c u l t - p o s i t i v e stools (Smith-Kline Diagnostics, Sunnyvale, California). T e n d i f f e r e n t gastroenterologists p e r f o r m e d the gastroscopies in these 17 patients. Five t u m o r s were located in the gastric cardia, and eight were c e n t e r e d along the lesser curvature, three o n the g r e a t e r curvature, a n d o n e in the a n t r u m . Seven lesions had the a p p e a r a n c e o f ulcers with n o d u l a r margins; they r a n g e d in size f r o m 2 to 10 cm. T e n t u m o r s were mass lesions, o f which o n e was a polyp. Sixteen cases were r e c o g n i z e d as malignant neoplasms. In only two cases was a diagnosis o f l y m p h o m a f a v o r e d e n d o scopically. In f o u r additional cases, the possibility o f l y m p h o m a was raised but the t u m o r could not be dis-

919

HUMANPATHOLOGY Volume18, No. 9 [September 1987] Selected Pathologic Features and Results of Immunohistochemical Staining in 17Cases of Primary Gastric Large Cell Lymphoma

TABLE 2.

Case

Total Biopsies

Biopsies Containing Tumor

Biopsy Diagnosis

Tumor Immunohistochemistry* LCA

1

7

5

2 3 4 5

6 3 13 7

4 1 11 7

6

9

8

7 8 9

15 5 14

12 4 6

10 ll 12

1

1

3 9

3 2

13 14

7 6

2 5

15 16

5 5

4 2

17

6

3

Poorly differentiated malignant neoplasm, favor lymphoma Undifferentiated malignant tumor Undifferentiated neoplasm Undifferentiated malignant tumor Undifferentiated malignant tumor, probably lymphoma Poorly differentiated malignant tumor, favor lymphoma Undifferentiated malignant neoplasm Malignant tumor, favor lymphoma Poorly differentiated malignant neoplasm, favor lymphoma Undifferentiated malignant tumor Malignant tumor, favor lymphoma Undifferentiated malignant tumor, favor lymphoma Undifferentiated malignant tumor Poorly differentiated malignant tumor, probably lymphoma Undifferentiated malignant tumor Undifferentiated malignant tumor, favor lymphoma Undifferentiated malignant tumor

Ck

+ + + + + "4-

+ + U + +

U

+ + + +

* Blinded review. LCA, leukocyte common antigen; Ck, cytokeratin; • equivocal; U, uninterpretable.

t i n g u i s h e d f r o m c a r c i n o m a . Seven cases were thought to be carcinoma.

Pathologic Findings Selected pathologic features are presented in table 2. T h e number of biopsies obtained per patient ranged from one to 15 (mean, seven); t u m o r was present in multiple biopsy samples in 15 cases. In 14 cases, the t u m o r o c c u r r e d within viable mucosa, whereas it was present solely enmeshed in necrotic debris and granulation tissue in three cases. T u m o r cells were f o u n d in n o n c o h e s i v e clusters in the lamina propria. In five cases, linear strands of cells bounded by reticulin produced a pseudotrabecular pattern. T u m o r cells o f six cases were located adjacent to glands that showed atypical reactive foveolar epithelium, thus contributing to the confusion with carcinoma. T u m o r present in ulcerated debris occurred in solid but noncohesive sheets. T h e 17 cases were initially encountered by seven different pathologists; four cases were seen in consultation by two pathologists. L y m p h o m a was suspected but not definitively diagnosed in nine cases. T h e remaining eight cases were classified as undifferentiated malignant tumor without further comment.

Immunohlstochemlcal Staining Controls

T u m o r cells of the biopsied and resected lymphomas stained intensely with antisera to LCA. The 920

reaction p r o d u c t was localized to the cytoplasmic m e m b r a n e in a linear or finely granular pattern. Where t u m o r cells were densely packed together, the intensity of the staining reaction gave rise to a latticelike network. Nonneoplastic lymphocytes also stained with LCA and served as internal controls. Epithelial cells were uniformly negative for LCA. Hollande-, formalin-, and B5-fixed tissues yielded staining of comparable intensity. T u m o r cells of the cases of lymphoma were uniformly negative for cytokeratin. By contrast, adjacent epithelia stained intensely f o r cytokeratin. T h e reaction product was a granular precipitate confined to the cytoplasm. T u m o r cells of the three biopsied and resected carcinomas stained for cytokeratin and were negative for LCA. Staining of neoplastic and nonneoplastic epithelial cells was comparable, as were results with Hollande- and formalin-fixed tissues. Study Cases, Blinded Review

T u m o r cells of 15 cases (88 per cent) were correctly interpreted as LCA positive and cytokeratin negative (fig. 1). Staining of tumor cells was comparable in p a t t e r n and intensity to that in the control cases. Interpretation was described as straightforward and unequivocal. Thirteen of these 15 cases contained intact epithelium that stained for cytokeratin and offered further support for the lymphoid origin of the tumor cells. Recognition of LCA-positive tumor cells was accomplished most easily within intact mucosa (figs. la, b, and c); however, the presence of exudate or gran-

LCAAND CYTOKERAT1N IN GASTRIC LARGE CELL LYMPHOMA (Dean et al.)

'~ 9

-',~ ~ 1 ~ ' , ~ ' ~

-+ ~I""~&;~

-

9

w,'~

~ ~

.*

't"

~

~

~ * ~ ~'~'~'~- ~,~ .,

~IP

"

.

.

.

:~

i

.

-r*.~

.

.

' "

.

.....

. . . .

.

~

4,.-~,,~,~.'-~;,'~

~ . ~ ; ~ B ~ , ~ - ,

";~..+"

.

e"

1~-

j

.

.

.

.

......

~m~K~'.

.,.

~-

~

" ~'ql~'3

~.-d~ ~

,t.--

9 9 ;..llr~'.~ ,--

",~'"~.

,..~'~-~.'

~

.

.

. . . .

!lrr,~,;2l# -' - ~

..

."

-

",, '-:

"-"

*

-.

'-'

9"

,~

~

-~

.

~--~'~ ': ~ "

. .~-s

~

-~:"-" -~ " " '? ~"....

~3":~;~ * ~ - ~ * ":~ '~%~- " ~ * * : '~,,,._ ~, ",-;#x~':~."::/,_,L+;"-.._ ~.It~

.

-~t,~

K~

: ~

~.~."~L'~; ~, .,* <: ";~ ~.~l

",,"~'i ~

9

, j~

: ,>,~.r<,.

- ~ J- ' ~ ' "~ . ~ . ~" ' ; ' ~ ,~r '~"

~,.

~'~ ~' l~J ~" 9

. ~. ,. l. . . ",*

~"~

"L;~'

~"

~,J~

~~

"~ " ~"

~* 9 "~s

-"

"

"4:'

. ' ~9 ~ ""~ " "~ ' "" .~ " " ~-,l~ i --:.'~'~ "

FIGURE 1. Undifferentiated malignant neoplasm in endoscopic biopsy of the gastric mucosa, a, The noncohesive, luster of tumor cells occupies the lamina propria. Reactive foveolar epithelium lines the gastric gland to the right of the field. [Hematoxylin-eosin ;fain. x 480.] b, Tumor cells exhibit circumferential m e m b r a n e immunoreactivily for leukocyte c o m m o n antigen 0-CAl. Epithelial cells to the right a n d left of the field are nonreactive. [Immunoperoxidase technique, hematoxylin counterstain, x 480.} c, Intense cyJ/oplasmic immunoreactiviJy for cytokeratin occurs in foveolar epithelium. Tumor cells are uniformly negative. [Immunoperoxidase technique, hematoxylin counterstain. x 480.] d, Tumor o b t a i n e d from the base of a maUgnant gastric ulcer9 Viable mucosa was not sampled 9 Necrotic debris from tumor diaJhesis is present to the left of the field. (HematoxyUn-eosin stain, x 235.} e, Interpretable immunoreactiviJy for LCA is evident despite tumor necrosis. [Immunoperoxidase technique, hematoxyIin counterstain, x 235.] f, Immunostaining for cytokeratin demonstrates a b s e n c e of reactivity. [Immunoperoxidase technique, hematoxylin counterstain, x 235.] i

staining. T h e s e two cases were reviewed in conjunction with the h e m a t o x y l i n - e o s i n - s t a i n e d sections. O n e case was resolved as l y m p h o m a , b r i n g i n g the n u m b e r o f correctly interpreted cases to 16 (94 per c e n t ) . . T h e i m m u n o s t a i n i n g p a t t e r n o f case 10, in which only o n e biopsy f r o m an area o f t u m o r necrosis was obtained, remained uninterpretable.

ulation tissue did not preclude accurate interpretation. In two cases that did not contain intact mucosa, LCA-positive t u m o r cells were confidently identified a d m i x e d with necrotic debris (figs. ld, e, a n d f). Several artifacts were e n c o u n t e r e d . T w o cases contained crushed t u m o r which p r o d u c e d a nonspecific staining pattern. Edge e n h a n c e m e n t o f immunostaining occurred in five cases. Six cases exhibited strong b a c k g r o u n d uptake o f stain in areas o f tissue necrosis. However, well-preserved t u m o r unaffected by artifact was present for evaluation in these cases. T w o cases y i e l d e d equivocal results, b o t h o f which w e r e a t t r i b u t e d to excessive b a c k g r o u n d

DISCUSSION

Most biopsied gastric neoplasms can b e diagnosed on hematoxylin-eosin-stained sections in con921

HUMAN PATHOLOGY

Volume 18, No. 9 (September 1987)

junction with histochemical stains for mucin. However, when tumor cells are anaplastic and poorly cohesive and mucin stains are negative, the differentiation of carcinoma from lymphoma may be problematic. Multiple investigators have recognized the difficulty of classification of the undifferentiated malignant tumor, especially when it is encountered in endoscopic biopsy material, l-x/Saraga et al.4 identified tumor cells as malignant in 11 of 16 biopsies of gastric lymphomas, but in only two cases could carcinoma be excluded. Likewise, Radaszkiewicz and Dragosics3 suspected lymphoma in only 12 of 23 endoscopic biopsies of gastric lymphomas. Several authors have advocated the use of ancillary techniques, including electron microscopy6,s and immunophenotyping,8 to secure the diagnosis when lymphoma is suspected on biopsy. Although these procedures are effective, they require repeat endoscopy for the procurement of additional tumor tissue. T h e development of monoclonal antibodies and immunohistochemical techniques applicable to paraffin-embedded tissue has afforded the pathologist a n o t h e r approach to the differential diagnosis of poorly differentiated neoplasms. In particular, antibodies directed against leukocyte common antigen (LCA) 12-14 and cytokeratinlS-]Shave been shown to be sensitive, specific, and reliable for the distinction o f epithelial from lymphoid neoplasms. Kurtin and Pinkus ~3 demonstrated immunoreactivity for LCA in 93 per cent of lymphomas; false-positive staining of nonhematopoietic neoplasms was not encountered. Battifora and Trowbridge 12 reported results for L C A similar to those o f Kurtin and Pinkus. Makin et al. 17 described almost 100 per cent positivity of formalinfixed epithelial t u m o r s for cytokeratin with uniformly negative staining o f lymphomas. In the present study, we applied immunohistochemistry for LCA and cytokeratin to the problem of definitive biopsy diagnosis of primary gastric large cell l y m p h o m a . Unequivocal i m m u n o s t a i n i n g of tumor cells for LCA occurred in 15 of 17 cases (88 per cent). That correct interpretation was rendered by a reviewer blinded to all clinical information, to the hematoxylin-eosin-stained appearance o f the tumor, and to the nature o f the applied antibody serves to underscore the quality of staining produced with antisera to LCA. Staining of adjacent epithelium for cytokeratin provided additional support for the lymphomatous nature of the tumor. Several factors contributed to the ease of immunohistochemical interpretation. Sections were dried overnight to minimize washing o f tissue from the slides. Enzymatic digestion with Pronase enhanced antigen-antibody binding. Optimal dilution o f LCA antisera was determined prior to staining. In addition, the amount and quality of tissue procured by biopsy were m a j o r d e t e r m i n a n t s o f successful stammg. T h e mean number of biopsy samples obtained per patient was seven, and viable mucosa and multiple biopsy specimens containing t u m o r were present in most cases. 922

A n u m b e r o f immunoperoxidase artifacts, including enhancement of staining about the edges o f biopsy fragments, staining of crushed tissue, and uptake of stain by necrotic and inflamed tissue, were encountered. Since tumor was invariably present in multiple biopsy samples, artifact compromised interpretation in only one case. We attribute this failure to inadequate tissue procurement, for the one biopsy obtained from this patient was taken from an area of tumor necrosis. Although these 17 cases occurred at one institution over a six-year period, we believe that they reflect the general experience with this entity. Ten diff e r e n t gastroenterologists p e r f o r m e d the endoscopies, seven d i f f e r e n t pathologists were initially confronted with the biopsied tissue, and the diagnostic dilemma--carcinoma or l y m p h o m a m w a s not resolved prior to gastric resection. Advantages o f the immunohistochemical approach are evident. Staining is sensitive and specific. T h e t u r n a r o u n d time is usually equivalent to that for mucin histochemical stains. Staining of paraffin-embedded tissues averts the necessity of repeat endoscopy for procurement of additional material. In summary, immunohistochemical staining for LCA and cytokeratin is a reliable and useful adjunct in the evaluation of the undifferentiated malignant tumor in endoscopic biopsy material and offers the opportunity for definitive diagnosis o f p r i m a r y gastric large cell lymphoma.

REFERENCES

1. Kahn LB, Seizer G, Kaschula ROC: Primary gastrointestinal lymphoma. A clinicopathologicstudy of fifty-sevencases. Dig Dis 17:219, 1972 2. MorsonBC, Dawson IMP: Non-eplthelialtumours. In Gastrointestinal Pathology,2nd ed. Oxford, BlackwellScientific, 1979 3. RadaszkiewiczT, DragosicsB: Primarylymphomasof the gastrointestinal tract. A clinicopathologicstudy of 60 cases. Pathol Res Pract 169:353, 1980 4. Saraga P, Hurlimann J, Ozzello L: Lymphomasand psuedolymphomas of the alimentarytract: an immunohistochemical study with clinicopathologiccorrelations. HUMPATnOL 12:713, 1981 5. Rotterdam H, Sommers SC, WayeJD: Stomach. In Biopsy Diagnosis of the DigestiveTract. New York, Raven Press, 1981, p 45 6. Kay S: The Stomach. In Silverberg SG (ed): Principles and Practice of Surgical Pathology. New York, John Wiley & Sons, 1983, p 819 7. Haggitt RC: Malignant lymphoma,large cell type. In Endoscopic Biopsies of tile Esophagus and Stomach. American Society of Clinical PathologistsCommissionon Continuing Education, 1983, p 68 8. AntonioliDA: Current concepts in carcinomaof the stomach. In A'ppelman HD (ed): Pathology of the Esophagus, Stomach, and Duodenum. New York, Churchill Livingstone, 1984, p 121 9. Platz CE: Lymphoid proliferations of the stomach. In Appelman HD (ed). Pathologyof the Esophagus,Stomach,and Duodenum. New York, Churchill Livingstone,1984, p 243 10. Kahn LB, Mir R: Lymphoidproliferationsof tile gastrointestinal tract. In Levin B, Riddell RH (eds): Frontiers in Gastrointestinal Cancer. New York, Elsevier, 1984, p 19

LCA AND CYi'OKERATININ GASTRICLARGECELL LYMPHOMA (Dean et al.)

11. Fine C.,, Ma CK: Alimentary tract. In Kissane JM (ed): Anderson's Pathology, 8th ed. St. Louis, C. V. Mosby Co., 1985, p 1055 12. Battifora H, Trowbridge IS: A monocolonal antibody useful for differential diagnosis between malignant lymphoma and nonhematopoietic neoplasms. Cancer 51:816, 1983 13. Kurtin PJ, Pinkus GS: Leukocyte common antigen--a diagnostic discriminant between hematopoietic and nonhematopoietic neoplasms in paraffin sections using monoclonal antibodies: correlation with immunologic studies and ultrastructural localization. HUM PATHOL 16:353, 1985 14. Warnke RA, Rouse RV: Limitations encountered in the application of tissue section immunodiagnosis to the study of

923

15. 16. 17. 18.

lymphomas and related disorders. HUM PATHOL 16:326, 1985 Said JW: Immunohistochemical localization of keratin proteins in tumor diagnosis. HUM PATHOL 14:1017, 1983 Gatter KC, Alcock C, Heryet A, et ah The differential diagnosis of routinely processed anaplastic tumors using monoclonal antibodies. Am J Clin Pathol 82:33, 1984 Makin CA, Bobrow LG, Bodmer WF: Monoclonal antibody to cytokeratin for use in routine histopathology. J Clin Pathol 37:975, 1984 Gown AM, Vogel AM: Monoclonal antibodies to human intermediate filament proteins. Am J Clin Pathol 84:413, 1985