Role of CD44 in nonpalpable T1a and T1b breast cancer

Role of CD44 in Nonpalpable T1 a and T1 b Breast Cancer J.S. LYZAK, MD, M.L. YAREMKO, MD, W. RECANT, MD, D.A. BAUNOCH, PHD, AND L. JOSEPH, MD Primary infiltrating ductal carcinomas (IDCs) of the breast which measure less than 0.5 cm (Tla lesions) and between 0.5 and 1.0 c m (Tlb lesions) are associated with a small risk of nodal metastasis. The role of axillary dissection in Tla and Tlb breast cancer is controversial. In the absence of axillary dissection, comparable prognostic information might be obtained by examination of the primary cancer. The adhesion molecule CD44 represents a family of transmembrane proteins that mediate cell-cell and cell-matrix interactions. Previous investigators have correlated expression of CD44 and its isoforms with prognosis in breast cancer. We investigated the value of CD44 isoform expression as a predictor of nodal metastases in nonpalpable Tla and Tlb IDC. Monoclonal antibody against the standard f o r m of CD44 (CD44s) and polyclonal antibody directed against the variant isoform (CD44v6) was tested on 34 cases of nonpalpable node-nega-

tive infiltrating ductal carcinoma (IDC) less than 1.0 cm and 9 cases of nonpalpable node-positive IDC less than 1.0 cm. The expression of CD44s was significantly decreased in node-positive Tla and Tlb IDC versus node-negative T l a and T l b IDC (11% vs 65%). In contrast, 97% of the node-negative IDC and 100% of the node-positive IDC expressed the CD44v6 isoform. We conclude that CD44s expression is significantly altered in Tla and Tlb IDC with nodal metastases but that the CD44v6 isoform does n o t correlate with nodal metastases in nonpalpable stage T l a and T l b IDC. HUM PATHOL 28:772-778. Copyright © 1997 by W.B. Saunders Company Key words: CD44, adhesion molecule, breast neoplasm, Tla, Tlb, metastases. Abbreviations: IDC, infiltrating ductal carcinoma; ER, estrogen; DAB, diaminobenzidlne.

Breast cancer is the most c o m m o n malignancy and the second most c o m m o n cause of cancer death a m o n g w o m e n in the United States. M a m m o g r a p h i c screening has increased the detection of small or n o n p a l p a b l e breast malignancies. 1 Currently, T l a and T l b tumors represent approximately 20% of all breast malignancies. 2 T l a tumors are defined as those lesions measuring less than 0.5 cm. T l b tumors are defined as those lesions measuring m o r e than 0.5 cm but not m o r e than 1.0 cm. 3 Most w o m e n with T l a and T l b tumors have a good prognosis and a relatively small risk of nodal metastases, ranging from 3% to 22%. Axillary involvem e n t is present in 4% of nonpalpable T l a and 7% of nonpalpable T l b tumors. T h e incidence rises to 6% for palpable T l a and 23% for palpable T l b tumors. 4 This raises the question of whether patients with nonpalpable T l a and T l b tumors should routinely u n d e r g o axillary dissection. Axillary dissection provides staging information for breast cancer t r e a t m e n t and prognosis, but also carries a significant risk of postoperative complications including a r m e d e m a and neuropathies. 4'6 T h e value of the information gained by p e r f o r m i n g an axillary dissection in the face of such complications is controversial and m i g h t not outweigh the small risk of nodal metastases in patients with nonpalpable T l a and T l b tumors. The operative risk becomes an additional complication in a medically c o m p r o m i s e d patient. T h e staging information gained is beneficial in choosing

p r e m e n o p a u s a l patients who may benefit f r o m adjuvant cytotoxic chemotherapy, leading some investigators to advocate obtaining nodal staging data in virtually all patients with IDC. 5 Alternative m e t h o d s to acquire these data would be desirable for optimizing prognostic information and therapeutic options. Many features of primary breast cancer have b e e n evaluated for their ability to predict the presence of axillary or distant metastases. T h e well-established parameters are t u m o r grade, size, and h o r m o n e receptor status. DNA ploidy and S-phase analysis by flow cytometry, changes in expression of the HER-2/neu oncogene, cathepsin-D enzyme, epidermal growth factor receptor, cell cycle regulatory proteins, and the nm23 protein have also b e e n described as factors in the clinical progression of breast cancer, but their prognostic value is controversial. 7.8 Expression of the cell adhesion molecule CD44 has b e e n the subject of recent interest in the t u m o r progression and survival of breast cancer patients. 9'1° CD44 is an integral t r a n s m e m b r a n e protein originally described as a lymphocyte h o m i n g receptor, which enables lymphocytes to adhere to high endothelial vela nules. Further work has shown that CD44 is expressed in epithelial and mesenchymal tissues and mediates cell-cell and cell-matrix interactions, a2-16 T h e gene for CD44 is on c h r o m o s o m e 11, at l l p 1 3 , and consists of 20 exons. Five C-terminal and five N-terminal exons are "constant." T e n intervening exons are "variable" and can be spliced in various combinations. Splicing of the five constant N-terminal exons to the five constant Cterminal exons, excluding the 10 variable exons, encodes the standard f o r m CD44 (CD44s). Alternative splicing of 1, 2, or up to 10 variable exons yields the variant isoforms (CD44v). Inclusion of the sixth variant exon generates CD44v6.17 Hyaluronic acid is a ligand for CD44s, but a specific ligand for the variant isoforms has not yet b e e n identified, aa

From the Division of the Biological Sciences, Department of Pathology at the University of Chicago Hospitals, Chicago, IL Accepted for publication October 14, 1996. Supported by the Cancer Research Foundation and the Ben G. Kissel Fund. Address correspondence and reprint requests to M.L Yaremko, MD, University of Chicago Hospitals, Department of Pathology, 5841 S. Maryland Ave, AMB S-329, MC 3083, Chicago, IL 60637. Copyright © 1997 by W.B. Saunders Company

0046-8177/97/2807-000455.00/0

772

CD44 IN NONPALPABLET1 BREASTCANCER (Lyzak et al) It has b e e n shown t h a t CD44 is u p r e g u l a t e d a n d e x p a n d s its e x p r e s s e d r e p e r t o i r e o f v a r i a n t i s o f o r m s in m a n y h u m a n tumors.19-22 Specifically, CD44 v a r i a n t isof o r m s c o n t a i n i n g e x o n v6 have b e e n s h o w n to b e sign i f i c a n t in p r e d i c t i n g p r o g n o s i s in a diverse g r o u p o f m a l i g n a n c i e s . CD44v6 e x p r e s s i o n is a s s o c i a t e d with p o o r e r survival a n d d e c r e a s e d r e c u r r e n c e - f r e e survival in low h i s t o l o g i c a l g r a d e n o n - H o d g k i n ' s l y m p h o m a . 2s P a t i e n t s with CD44v6 positive stage I to l i B cervical c a r c i n o m a also s h o w e d p o o r e r overall survival. 24 Patients with CD44v-positive o v a r i a n c a r c i n o m a s h a d a significantly s h o r t e r disease-free survival t h a n p a t i e n t s with CD44v-negative t u m o r s . 25 I n c o l o r e c t a l c a r c i n o m a , CD44v6 e x p r e s s i o n is a s s o c i a t e d with t u m o r - r e l a t e d d e a t h , i n d e p e n d e n t o f D u k e ' s stage. 26'27 T h e d a t a reg a r d i n g t h e i m p o r t a n c e o f CD44 i s o f o r m s in b r e a s t carc i n o m a is m i x e d . J o e n s u et al 1° s h o w e d t h a t CD44s exp r e s s i o n c o r r e l a t e d with i n f e r i o r o u t c o m e . Similarly, K a u f f m a n n et al 9 s h o w e d CD44v6 e x p r e s s i o n as a n i n d e p e n d e n t p o o r p r o g n o s t i c f a c t o r in b r e a s t c a n c e r . I n c o n t r a s t , F r i e d r i c h s et al 7 s h o w e d t h a t CD44v6 e x p r e s sion d i d n o t c o r r e l a t e with disease f r e e o r overall survival. T h e s e c o n t r a d i c t o r y r e p o r t s l e d us to evaluate t h e ability o f CD44 i s o f o r m e x p r e s s i o n to p r e d i c t n o d a l m e tastases. W e c o m p a r e d CD44s a n d CD44v6 a d h e s i o n m o l e c u l e e x p r e s s i o n in n o d e - p o s i t i v e a n d n o d e n e g a tive n o n p a l p a b l e T l a a n d T l b IDC. W e f o u n d t h a t CD44s e x p r e s s i o n is significantly d e c r e a s e d in n o d e positive T l a a n d T l b b r e a s t c a n c e r , w h e r e a s CD44v6 was e x p r e s s e d to a n e q u a l d e g r e e in b o t h n o d e - p o s i t i v e and node-negative nonpalpable Tla and Tlb breast cancers.

MATERIALS AND METHODS Patients a n d Breast C a n c e r s The surgical pathology reports from all 807 women who had stereotactic needle localization biopsy of the breast at the University of Chicago from April 1985 to October 1994 were reviewed. A total of 187 patients were identified with a diagnosis of IDC. Fifty-two of these (28%) had tumors measuring less than 1.0 cm, as determined by gross inspection. Fortythree had sufficient material remaining in the paraffin blocks for additional immunohistochemical analysis; 9 were excluded because material remaining was insufficient. Thirtyfour of the 43 patients (79%) were node negative, of which 7 patients had T l a and 27 patients had T l b tumors. Nine patients (21%) were node positive, 4 with T l a and 5 with T l b tumors. The mean number of lymph nodes containing metastatic tumor in the node-positive group was 12 (mean of lymph nodes examined in the node positive group = 30). All patients underwent a needle localization with excisional biopsy followed by axillary dissection (28 patients), modified radical mastectomy (14 patients), or an extended radical mastectomy (1 patient). The median age at diagnosis was 59 years (range, 28 to 80 years). Survival data were available for a median of 53 months (range, 12 to 108) in 38 patients. Only one patient with positive lymph nodes died from breast cancer. No patient without nodal metastases died of breast cancer. One patient died of esophageal cancer, and five patients were lost to follow-up.

773

Thirty-four node-negative and 9 node positive tumors were reviewed and analyzed for histological type and graded according to the criteria of Elston and Ellis.2s All tumors were infiltrating dnctal type, with 2 cases of tubular carcinoma (node negative), 2 cases of mucinous carcinoma (node negative), and 1 case of medullary carcinoma (node positive). Estrogen (ER) and progesterone receptor (PR) data were available in 12 of the node-negative cases and in 7 of the node-positive cases. Sixty-seven percent of the node-negative cases were ER positive versus 71% of the node-positive cases. Fifty-five percent of the node-negative cases were PR positive versus 50% of the node-positive cases. S-phase, ploidy, and proliferation indices had not been performed on most of the study cases because of their small size.

CD44 Immunohistochemistry Five # m - t h i c k sections of formalin-fixed, paraffin-embedded material were mounted onto Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA), baked at 60 degrees centigrade for 1 hour, cleared in xylene, and hydrated through a descending alcohol series. The hydrated tissue sections were stored in alkaline phosphatase buffer (pH 7.4) before performing the assay (Ventana Medical Systems, Tucson, AZ). For detection of CD44s, the hydrated tissue sections were microwaved (>800-W microwave) for 9 minutes (3 minutes, 3 cycles) in 0.01 m o l / L citrate buffer, p H 6.0, as a prereaction antigen-retrieval step. For CD44s and CD44v6 detection, the tissue sections were preincubated in 3% hydrogen peroxide in absolute methanol for 10 minutes at room temperature to quench endogenous peroxidase activity. To limit cross-reaction with endogenous biotin, an avidin-biotin blocking kit was used (Ventana Medical Systems). For CD44s detection, monoclonal anti-CD44s (anti-homing cell adhesion molecule [HCAM]) was used at a titer of 1:80 (Becton Dickinson, San Jose, CA). For CD44v6 immunohistochemistry, a polyclonal antibody to CD44 variant isoform exon v6 was used at a titer of 1:80 (Accurate Chemical and Scientific, Westbury, NY). Immunohistochemical assay for CD44s was performed manually using a standard indirect biotin-streptavidin-enhanced immunoperoxidase technique with diaminobenzidine (DAB) as the chromogen. The immunohistochemical assay for CD44v6 was performed on a Ventana ES automated immunostainer. The Ventana ES uses an indirect streptavidin-biotin system conjugated with horseradish peroxidase for detecting the immunocomplex and DAB substrate for localization. The sections were counterstained with hematoxylin, dehydrated through an ascending alcohol series, cleared, and coverslipped. Immunohistochemistry on a panel of normal human tissues was performed to confirm appropriate reactivity of the anti-CD44s and anti-CD44s antibodies. All were evaluated for localization and intensity of reactivity. From these tissues, we selected and prepared a single block with portions of prostate, lymph node, liver, thyroid, and brain to use as positive and negative controls for the test cases. A second positive control, of an IDC previously known to express both isoforms, was also included in the test immunoreactions. Immunohistochemistry of the IDCs was performed using the preceding as positive controls and normal mouse serum as a negative control. Interpretation of all IDCs was done without knowledge of the status of the axillary lymph nodes and graded in a semiquantitative manner using a 5-point scale: negative, weak, +1, +2, and +3. The percentage of tumor cells reacting was also scored: less than 10%, greater than, 10% but less than 90% and greater than 90%. For statistical analysis, the proportion of cases in each scoring category

HUMAN PATHOLOGY Volume28, No. 7 (July 1997) TABLE 1.

CD44 in Normal Human Tissues

Tissue Breast Myoepithelium Ductal epithelium Prostate Basal cells Ductal epithelium Tonsil Epithelium Germinal centers Thyroid Adrenal gland Brain Skeletal muscle Liver Ovary Uterus Endometrium Myometrium Cervix Ectocervix Endocervix Stomach Lung Pneumocytes Bronchial epithelium Pancreas Islets Acini Ducts Kidney Glomeruli Tubules Vasa recta Skin Epidermis Eccrine glands Dermis Placenta Trophoblast Villous stroma Other Nerve/ganglia Endothelium

CD44s

rate with either antibody. Tonsillar tissue showed strong m e m b r a n o u s labeling of the tonsillar epithelium with both anti-CD44s and anti-v6. In addition, lymphocytes within and adjacent to reactive germinal centers decorated with both antibodies. Endocrine organs were variably reactive. Adrenal gland was negative for both antibodies, whereas cytoplasmic membranes of thyroid follicular epithelium labeled well with both. Ovary, skeletal muscle, brain, and liver reacted with neither antibody. Uterine tissue showed weak focal myometrial labeling with anti-CD44s but no labeling with anti-v6. Cervix showed strong m e m b r a n o u s labeling with both antibodies and increased staining with anti-v6 at the epithelial surface. Stomach fundic mucosa decorated with anti-CD44s and showed decreased intensity with anti-v6. The anti-v6 localized to the cytoplasm o f the gastric pit epithelial cells. Cytoplasmic membranes o f lung pneumocytes and cells consistent with bronchial reserve cells labeled with anti-CD44s but not with antiv6. Pancreatic acini, and duct and rare islet cells labeled with anti-CD44s but not with anti-v6. Kidney vasa recta strongly labeled with anti-CD44s and showed no labeling with anti-v6; tubules and glomeruli were negative with both antibodies. Skin showed focal epidermal decoration for the standard and v6 isoforms. Eccrine glands focally labeled with CD44s antibody. Placental tissue labeled equivocally with anti-CD44v6 but not with antiCD44s. Lymphocytes, plasma cells, endothelium and nerve/ganglia provided internal controls in all normal tissues with staining noted for both antibodies. Similar panels of h u m a n tissues have been performed by other investigatorsJ 215

CD44v6

+1 +1

+1 Negative

+2 Negative

+2 Negative

+3 +2 +1 Negative Negative Negative Negative Negative

+3 +3 +3 Negative Negative Negative Negative Negative

Negative Weak

Negative Negative

+3 +3 +3

+2 +3 +1

+2 Negative

Negative Negative

Negative + 3* + 3*

Negative Negative Negative

Negative Negative +3

Negative Negative Negative

+ 1" +3* Negative

+ 1" Negative Negative

Negative Negative

Weak Weak

+2 +2

+1 +2

Immunohistochemical Reactions in Nonpalpable Tla and Tlb Breast Cancers

* Focal staining.

was tested. Using trend in proportions as a check, rumors with a score of negative or weak were analyzed against those tumors with a score of +1 or greater using the Fisher's exact test. RESULTS

Immunohistochemical Reactions of CD44 in Normal Human Tissues Antibodies against the standard form of CD44 and the v6 isoform both showed appropriate patterns o f reactivity and correlate well with those of Fox et a115 (Table 1). In normal breast, anti-CD44s labeled the cytoplasmic membranes of myoepithelial cells and occasional ductal cells. Anti-CD44v6 labeled the membranes of myoepithelial cells but not the ductal epithelium. Prostate showed labeling in basal cells with anti-CD44s and anti-CD44v6; however, ductal cells failed to deco774

In the node-negative group of 34 breast cancers, 19 cases expressed the CD44s antigen (65%). Decoration by antibody was accentuated along the cytoplasmic m e m b r a n e (Fig 1A). Intertumoral and intratumoral heterogeneity was p r o m i n e n t in the node-negative lesions with intensity of expression ranging from +1 (1 case) to +2 (5 cases) to +3 (13 cases) (Table 2). Within each tumor, the percentage of reactive cells varied from less than 10% (11 cases), greater than 10%, but less than 90% (5 cases) to greater than 90% of the malignant cells (3 cases) (Table 3). All 11 cases except for 1 case with 10% of the cells reacting with anti-CD44s scored an intensity of +2 or greater. In 5 of 34 cases, the expression of CD44s was not assessed as the internal control lymphocytes did not react appropriately. These cases were excluded from the statistical analysis. Among the 9 node-positive tumors, only I case (11%) expressed the CD44s antigen. This case showed +2 intensity in less than 10% of the neoplastic cells. The remaining eight cases (88%) were nonexpressors (Fig 1B). This difference in expression of the CD44s epitope in nonpalpable node-positive T l a and T l b breast cancers was significant (P = .005). Immunohistochemical studies using antibody directed against the CD44v6 isoform in node-negative

CD44 IN NONPALPABLE T1 BREASTCANCER (Lyzak et al)

FIGURE 1. (A) Anti-CD44s positive immunoreactivity in a node-negative case of IDC. (Original magnification x400.) (B) Absence of immunoreactivity with anti-CD44s in a node-positive case of IDC. Note strong decoration of an adjacent lymphocyte. (Original magnification ×400.)

breast cancers showed decoration in 30 cases (88%). T u m o r heterogeneity of CD44v6 expression was n o t e d between cells within each block of tumor tested. Intensity ranged from +1 (8 cases), to +2 (10 cases), to +3 (12 cases) (Table 4). One case expressed the antigen on less than 10% of cells, 9 cases in greater than 10%, but less than 90%, and 20 cases in greater than 90% of the cells (Table 5). Labeling of cells was predominantly cytoplasmic (Fig 2). Nine of nine node-positive cases also expressed the CD44v6 isoform. Intensity ranged from +1 (5 cases), +2 (2 cases) to +3 (2 cases); and percentage reactivity from <10% (0 cases), between 10% to 90% (2 cases) and greater than 90% (7 cases) (Fig 2B). T h e r e was no statistically significant difference between CD44v6 expression and the presence or absence of lymph node metastases (P = .36). The tumor grade was separately analyzed against CD44s and CD44v6 expression. By the criteria of Elston and Ellis, 28 21 of the node-negative cancers were well differentiated, 8 were moderately differentiated, and 5 were poorly differentiated. In contrast, n o n e of the node-positive cancers were well differentiated, 7 were moderately differentiated, and 2 were poorly differentiated. No statistically significant correlation between tumor grade and CD44v6 or CD44s expression was identified (statistical test for trend in proportions not significant). DISCUSSION

The authors show a significant difference in the expression of the standard CD44 isoform, CD44s, in TABLE 2.

Node negative Node positive

CD44s Intensity in Tla a n d Tlb IDC* Weak/0

+1

+2

+3

10 8

1 0

5 1

13 0

nonpalpable T l a and T l b breast carcinomas, which are node positive as opposed to those that are node negative. Node-negative T l a and T l b IDCs retain CD44s expression, although this expression may be focal, whereas node-positive T l a and T l b IDCs lose its expression. In contrast, the authors observed no significant difference in the expression of the v6 variant CD44 isoform, and CD44v6 between node-positive and -negative T l a and T l b IDCs. The results suggest that of the two isoforms, only determination of the standard CD44s expression contributes better staging information for the nonpalpable T l a and T l b carcinomas. These findings are similar to those that have been observed in neuroblastoma, in which higher stage tumors showed decreased CD44s expression and diseasefree survival was shorter in CD44s negative tumors. 29'3° The mechanism for this decrease in expression remains to be determined, but possible explanations include modulation of CD44 p r o m o t e r activity or alternate splicing by other oncogene products. 3° Decreased expression of CD44 isoforms has also been noted in the progression of squamous cell 31 and endometrial carcinomas. 16 DeMarzo et al ~1 also showed decreased expression of CD44 in high grade prostate carcinoma and associated nodal metastases when compared with benign nodal tissue. It has been suggested that host microenvironments may impact on CD44 isoform expression. 32 The usefulness of expression of CD44s and its isoforms for predicting nodal metastases in IDC has been investigated previously, with conflicting results. Joensu et al, 1° using a proprietary monoclonal antibody raised TABLE 3. CD44s Percentage of Cells Reacting in Tla and Tlb IDC

Node negative Node positive

* P = .005.

775

<10%

10% < X <90%

>90%

11 1

5 0

3 0

HUMAN PATHOLOGY TABLE 4.

Node negative Node positive

Volume 28, No. 7 (July 1997)

CD44v6 Intensity in T l a a n d T l b IDC*

Weak/0

+1

+2

+3

4 0

8 5

10 2

12 9

TABLE 5. CD44v6 P e r c e n t a g e of Cells Reacting in T l a a n d T l b IDC

Node negative Node positive

* P = .36.

in their laboratory against CD44s in 198 cases of breast cancer, f o u n d that tumors with greater than 50% immunoreactive cells had an inferior outcome. However, multivariate analysis showed CD44s expression did not have i n d e p e n d e n t prognostic value. Cancers with less than 50% of cells staining showed a mortality rate that a p p r o a c h e d that of the stronger expressors after longer follow-up, suggesting that those cancers with decreased n u m b e r s of positive cells progress eventually but at a slower rate. In their study, CD44s expression was associated with p o o r histology and high mitotic counts. The authors suggest that this may explain the correlation of CD44s with p o o r outcome. 1° During the course of our investigations, evidence that the CD44v6 isoform is predictive of axillary node status was r e p o r t e d by Kauffmann et al. 9 These investigators used a m o n o c l o n a l antibody against CD44v6 in 100 primary breast tumors, 12 locally r e c u r r e n t breast cancers, and 18 nodal metastases. Although they used a different m o n o c l o n a l anti-CD44v6 antibody than ours, they observed a labeling pattern similar to ours (negative in normal duct epithelium but reactive in most tumors). They f o u n d reactive cells in most of their tumors: 84% of the primary IDC, 100% of the recurrences, and 100% of the metastases. In node-positive patients, the improved survival with CD44v6-negative tumors was statistically significant. However, in nodenegative patients, CD44v6 expression did not correlate with survival. T h e authors concluded that CD44v6 expression in the primary breast t u m o r predicted low probability of survival. F u r t h e r m o r e , CD44v6 expression e m e r g e d as an unfavorable prognostic factor inde-

<10%

10% < × <90%

>90%

1 0

9 2

90 7

p e n d e n t of t u m o r size, n o d e status, p r o g e s t e r o n e receptor status, and grade. 9 These conclusions are not in keeping with o u r own data or with that of Friedrichs et al. 7 These investigators p e r f o r m e d i m m u n o h i s t o c h e m i c a l studies using antibody directed against the v6 isoform in 119 node-positive cases and 108 node-negative cases. T h e i r antibody against CD44v6 labeled b o t h ductal epithelium and myoepithelium of normal breast. CD44v6 was expressed in 47.9% of the node-positive and in 38.9% of the nodenegative cases (no significant difference). T h e y concluded that CD44v6 expression did not correlate with disease-free or overall survival. Friedrichs et al 7 also studied the relationship between t u m o r differentiation and v6 expression. Similar to our results, no statistically significant difference was n o t e d between well, moderately, or poorly differentiated tumors and CD44v6 expression, j This study supports the findings of Friedrichs et al 7 and adds that CD44v6 isoform expression does not differentiate between node-positive and negative n o n p a l p a b l e T l a and T l b breast cancers. O n e of the major differences between this study and those described previously is that we confined our analysis to a rigidly defined subset of IDC, that of nonpalpable, very small tumors. These o t h e r studies all included tumors of varying sizes, f r o m T1 to T4 lesions, and the issue of palpability was not addressed. Some differences may be ascribable to intrinsic variation in capacity for nodal metastasis in T1 versus T2 to T4 carcinomas. O f interest in our study is the expression of CD44s and CD44v6 within individual breast tumors. We ob-

FIGURE 2. (A) Antibody directed against CD44v6 shows diffuse positMty in a node-negative case of IDC. (Original magnification ×400.) (B) Anti-CD44v6 positive immunoreactivity in a node positive case of IDC (Original magnification ×400.)

776

CD44 iN NONPALPABLE T1 BREAST CANCER (Lyzak et al)

served 18 cases in which the standard form of CD44 was not detected while the CD44v6 isoform was identified. These include a higher proportion of node-positive cases. This finding has also been reported in low-grade non-Hodgkin's l y m p h o m a Y This suggests an additional mechanism in neoplastic progression wherein epitopes on CD44s are not expressed or are r e n d e r e d unrecognizable by our monoclonal antisera. Post-translational modification of CD44s, such as N- or O-linked glycosylations with the addition of chondroitin sulphate and heparan sulphate side chains, occurs, 33 but the effect this has on antibody recognition has not been studied. The standard form of CD44 may undergo regulated shedding, releasing the CD44s epitope in the N-terminal extracellular domain while leaving the v6 epitope available for immunologic recognition. Ponta et a134 describe extracellular domain shedding of CD44s with subsequent saturation of hyaluronic acid. This potentially regulated process results in an opposite effect of m e m b r a n e - b o u n d ligand. Cellular contact to extracellular matrix is effectively prohibited, thus facilitating metastasis, s4 We tested the reactivity of anti-CD44s and antiCD44v6 antibodies against a wide range of normal human tissues. Similar studies by other investigators 12-15 gave conflicting results. Our data are similar to those of Fox et a115 who n o t e d expression in brain, adrenal, lung, kidney, skeletal muscle, and e n d o m e t r i u m with anti-CD44s and endometrial staining with anti-v6. As in this study, antigen-retrieval methods were used on formalin-fixed, paraffin-embedded tissues. 15 Differences might be ascribable to variations in antigen-retrieval methods or other modifications of technique. The expression of CD44s and CD44v6 in normal breast vary among investigators. Friedrichs et al 7 and others ]3 state that both myoepithelial cells and ductal cells express the CD44v6 isoform, whereas these data as well as that of Kaufmann e t al9 and others 12'a5'25 state that normal ductal epithelium is v6 negative and only the myoepithelium expresses this antigen. In conclusion, we found that CD44v6 expression is unable to discriminate between node-positive and node-negative T l a and T l b nonpalpable breast cancers. The expression of CD44s is significantly decreased in node-positive T l a and T l b lesions. The finding of decreased CD44s expression is interesting and potentially significant, but because the n u m b e r of node-positive cases available for this study was small, m u c h larger studies n e e d to be p e r f o r m e d to solidify this observation.

REFERENCES 1. Schwartz GF, Feig SA, Patchefsky AS: Significance and staging of nonpalpable carcinomas of the breast. Surg Gynecol Obstet 166:610, 1988 2. Sinn HP, Oelmann A, Anton HW, et al: Metastatic potential of small and minimally invasive breast carcinomas. Virchows Arch 425:237-241, 1994 3. Beahrs OH, Henson DE, Hutter RVP, et al: American Joint Committee on Cancer, Manual for Staging of Cancer (ed 4). Philadelphia, Lippincott, 1992, pp 149-154

777

4. Recht A, Houlihan MJ: Axillary lymph nodes and breast cancer: A review. Cancer 76:1491-1512, 1995 5. Wise L, Johnson H: Breast Cancer: Controversies in Management. New York, NY, Futura, 1994, pp 169-201 6. Kissen MW, Querci della Rovere G, Easton D, et al: Risk of lymphoedema following the treatment of breast cancer. Br J Surg 73:580-584, 1986 7. Friedrichs K, Franke F, Lisboa B, et al: CD44 isoforms correlate with cellular differentiation but not with prognosis in human breast cancer. Cancer Res 55:5424-5433, 1995 8. Mansour EG, Ravdin PM, Dressler L: Prognostic factors in early breast carcinoma. Cancer 74:381400, 1994 (suppl) 9. Kaufmann M, Heider K, Sinn H, et al: CD44 variant exon epitopes in primary breast cancer and length of survival. Lancet 345:615-619, 1995 10. Joensu H, Klemi PK, Toikkanen S, et al: Glycoprotein CD44 expression and its association with survival in breast cancer. Am J Pathol 143:867-874, I993 11. Jalkanen S, Bargatze RF, de-los-Toyos J, et al: Lymphocyte recognition of high endothelium: Antibodies to distinct epitopes of an 85-95 kD glycoprotien antigen differentially inhibit lymphocyte binding to lymph node, mucosal or synovial endothelial cells. J Cell Biol 105:983-990, 1987 12. Mackay CR, Terpe H, Stauder R, et al: Expression and modulation of CD44 variant isoforms in humans. J Cell Biol 124:71-82, 1994 13. Terpe HJ, Stark H, Prehm P, et al: CD44 variant isoforms are preferentially expressed in basal epithelia of non-malignant human fetal and adult tissues. Histochemistry 101:79-89, 1994 14. Heider K, Hofmann M, Hors E, et al: A human homologue of the rat metastasis-associated variant of CD44 is expresssed in colorectal carcinomas and adenomatous polyps. J Cell Biol 120:227-233, 1993 15. Fox SB, FawcettJ, Jackson DG, et al: Normal human tissues, in addition to some tumors, express multiple different CD44 isoforms. Cancer Res 54:4539-4546, 1994 16. Fujita B, Yaegashi N, Ide Y, et al: Expression of CD44 in normal human versus tumor endometrial tissues: Possible implication of reduced expression of CD44 in lymph-vascular space involvement of cancer cells. Cancer Res 54:3922-3928, 1994 17. Koopman G, Griffioen AW, Ponta H, et al: CD44 splice variants: Expression on lymphocytes and in neoplasia. Res Immunol 144:750-754, 1993 18. Gunthert U, Stauder R, Mayer B, et al: Are CD44 variant isoforms involved in human tumour progression? Cancer Surv 24:1942, 1995 19. Dall P, Heider K, Sinn H, et al: Comparison ofimmunohistochemistry and RT-PCR for detection of CD44v-expression: A new prognostic factor in human breast cancer. I n t J Cancer 60:471-477, 1995 20. Iida N, Bourguignon L: New CD44 splice variants associated with human breast cancers. J Cell Physiol 162:127-133, 1995 21. Matsumara Y, Tarin D: Significance of CD44 gene products for cancer diagnosis and disease evaluation. Lancet 340:1053-1058, 1992 22. Tanabe KK, Nishi T, Saya H: Novel variants of CD44 arising from alternative splicing: Changes in the CD44 alternative splicing pattern of MCF-7 breast carcinoma cells treated with hyaluronidase. Molec Carcinog 7:212-220, 1993 23. Ristamaki R, Joensuu H, Soderstrom K, et al: CD44v6 expression in non-Hodgkin's lymphoma: An association with low histological grade and poor prognosis. J Pathol 176:259-267, 1995 24. Kainz C, Kohlberger P, Sliutz G, et al: Splice variants of CD44 in human cervical cancer stage IB to IIB. Gynecol Oncol 57:383-387, 1995 25. Uhl-Steidl M, Muller-Holzner E, Zeimet AG, etal: Prognostic value of CD44 splice variant expression in ovarian cancer. Ontology 52:400-406, 1995 26. Mulder JR, Kruyt PM, Sewnath M, et al: Colorectal cancer prognosis and expression of exon-v6-containing CD44 proteins. Lancet 344:1470-1472, 1994 27. Wielenga VJM, Heider K, Offerhaus GJA, et al: Expression of CD44 variant proteins in human colorectal cancer is related to tumor progression. Cancer Res 53:4754-4756, 1993

HUMAN PATHOLOGY

Volume 28, No. 7 (July 1997)

28. Elston CW, Ellis I0: Pathological prognostic factors in breast cancer. Histopathology 19:403-410, 1991 29. Combaret V, Lasset C, Frappaz D, et al: Evaluation of CD44 prognostic value in neuroblastoma: Comparison with the other prognostic factors. E u r J Cancer 31A:545-549, 1995 30. Terpe HJ, Christiansen H, Gonzalez M, et al: Differentiation and prognosis of neuroblastoma in correlation to the expression of CD44s. E u r J Cancer 31A:549-552, 1995 31. Demarzo A, Bradshaw C, Epstein J, et al: Total CD44 and CD44v6 expression in prostate carcinoma. Mod Pathol 9:72A, 1996

778

32. Salmi M, Gron-Virta K, Sointu P, et al: Regulated expression of exon v6 containing isoforms of CD44 in man: Downregulation during malignant transformation of tumors of squamocellular origin. J Cell Biol 122:431-442, 1993 33. Jalkanen S, Jalkanen M, Bargatze L, et al: Biochemical properties of glycoproteins involved in lymphocyte recognition of high endothelial venules in man. J Immunol 141:1615-1623, 1988 34. Ponta H, SleemanJ, Dall P, et al: CD44 isoforms in metastatic cancer. Invasion Metastasis 14:82-86, 1994