ANTICANCER RESEARCH 27: 215-218 (2007)

Fas/Fas- Expressions in Primary Breast are Significant Predictors of its Skeletal Spread

MAREK BE˛BENEK1, DANUTA DUS’ 2 and JOANNA KOZ’ LAK2

1First Deptartment of Surgical Oncology, Regional Comprehensive Cancer Center, 53-413 Wroclaw; 2Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, 53-114 Wroclaw, Poland

Abstract. Background: Bones belong to the most frequent ligand status may have a significant impact on patient survival localizations of breast cancer metastases. Several studies on (10-12). Consequently, the aim of our study was to determine female breast malignancies have indicated that Fas/Fas-ligand if Fas/Fas-ligand can serve as predictors of skeletal status may have a significant impact on survival. Hence, the aim dissemination in radically-treated breast cancer patients. of our study was to determine if these molecules might serve as the predictors of skeletal dissemination in radically-treated breast Patients and Methods cancer patients. Patients and Methods: Tumor samples from 147 radically-treated breast cancer patients were studied Tumor samples from 147 breast cancer patients of the Regional Comprehensive Cancer Center, Wroclaw, Poland, treated radically immunohistochemically for Fas/Fas-ligand expression. Results: at the local Surgical Department between 1998 and 2000 were Both Fas and Fas-ligand expression in the primary tumor were studied. The median age of the patients was 59 (range, 35-81 considerably less frequent among breast cancer patients with years). Detailed characteristics of the group studied are given in bone metastases compared to women without skeletal spread. Table I. All the patients had undergone surgery (radical Moreover, negative staining for Fas or the lack of Fas-ligand mastectomy or breast conserving surgery with lymph node expression proved to be significant predictors for the survival free dissection) with adjuvant treatment conducted in accordance with from bone metastases under univariate analysis. Conclusion: obligatory standards. The follow-up period amounted to 5 years. Breast tumor specimens, freshly obtained at the time of surgery, Our results suggest that the probability of bone metastases may were immediately placed in an RPMI1640 medium (Sigma, be assessed on the basis of Fas/Fas-ligand expression in primary Poland), snap-frozen and stored at –80ÆC. For purposes of breast cancer. Consequently, their determination seems crucial standard histological studies (pTNM and perilymphatic fat for further prognosis and determination of adjuvant treatment. infiltration), postoperative specimens were also fixed in 10% formalin and subsequently stained with hematoxilin and eosin. Various studies have revealed that both Fas and its ligand - The 4-6 Ìm cryostat sections were immunostained for Fas and FasL play an important role in cancer biology. The FasL using the Biotin-Streptavidin-Peroxidase method as follows. Air-dried (overnight) and fixed in cold acetone (10 min), the significance of the expression of these molecules in primary sections were incubated with a peroxidase-blocking solution (Dako tumors for the clinical outcome of malignancies with various A/S, Glostrup, Denmark) for 30 min. They were then incubated localizations is, however, not fully understood (1-8). with anti-Fas or anti-Fas-ligand monoclonal (1:100, Bones belong to the most frequent localizations of breast Santa Cruz Biotechnology, Inc. CA, USA) overnight at 4ÆC. The cancer metastases (9). Skeletal dissemination of mammary sections were then incubated (37ÆC) with a biotin-labeled tumors is in turn related to a significantly worse prognosis. secondary mouse anti-goat IgG (1:50, Dako A/S) and Accordingly, there is a need for new markers to assess the streptavidin-biotin-peroxidase (1:50, Dako A/S) for 30 min each. Tissue was stained for 5 min with 0.05% 3,3’-diaminobenzidine probability of bone dissemination in breast cancer patients. tetrahydrochloride (DAB) and then counterstained with Although their conclusions are still controversial, several haematoxylin, dehydrated and mounted. The expression of the studies on female breast have indicated that Fas/Fas- molecules studied was analyzed under microscopy. The localization and the intensity of the color reaction were compared with negative controls (normal breast tissue, Figure 1). Samples were considered positive when at least 10% of breast cells were Correspondence to: Marek Be˛benek, MD, Ph.D., First Department stained (Figures 2 and 3). of Surgical Oncology, Regional Comprehensive Cancer Center, pl. Moreover, ER and PR status were determined using the Dako Hirszfelda 12, 53-413 Wroclaw, Poland. Tel: +48 71 368 93 00, A/S Cytomation (mouse anti-human estrogen 1D5- Fax: +48 71 368 93 09, e-mail: [email protected] clone and mouse anti-human progesterone-receptor clone PGR 636) and immunohisochemical staining for p53 status was Key words: Fas, Fas-ligand, breast cancer, bone metastases. performed using monoclonal anti-p53 (DO-7, Dako A/S).

0250-7005/2007 $2.00+.40 215 ANTICANCER RESEARCH 27: 215-218 (2007)

Table I. Patient and tumor characteristics in breast cancer in women with evidence of bone metastases. In patients whose breast (+) or without (–) bone metastases (Mann-Whitney U, p≤0.05). cancer spread to bones: i) Fas expression in the primary tumor was significantly less frequent, with 6/18 (33%) cases Characteristics Total Bone Bone p-value (n=147) metastases metastases compared to 85/129 (66%) in women without skeletal (+) (–) metastates (p=0.017); ii) Fas-ligand expression in the (n=18) (n=129) primary tumor was significantly rarer (4/18 or 22% of the cases) compared to 66/129 (51%) in women without skeletal spread (p=0.028, Table I). Age at diagnosis ≤50 years 48 (33%) 7 (39%) 41 (32%) 0.548 Similar relationships were, however, not observed when >50 years 99 (67%) 11 (61%) 88 (68%) metastatic and non-metastatic women were compared ≤60 years 79 (54%) 10 (56%) 69 (53%) 0.870 without regard to the location of metastases (data not >60 years 68 (46%) 8 (44%) 60 (47%) shown). Menopausal status Women with and without bone metastases were, however, premenopausal 48 (33%) 6 (33%) 42 (33%) 0.948 postmenopausal 99 (67%) 12 (67%) 87 (67%) similar in terms of the age of diagnosis, menopausal status, Histological type size of primary tumor, regional lymph node involvement, ductal ca 108 (74%) 16 (89%) 92 (71%) 0.115 the presence of perilymphatic fat infiltration, and the lobular ca 39 (26%) 2 (11%) 37 (29%) expressions of estrogen and progesterone receptors and p53 pT (Table I). 0+1 35 (24%) 3 (17%) 32 (25%) 0.104 2+3 112 (76%) 15 (83%) 97 (75%) Univariate analysis showed that both negative staining for Histological grade Fas and the lack of Fas-ligand expression are significant G1+G2 103 (70%) 15 (83%) 88 (68%) 0.191 predictors for the survival free from bone metastases. Median G3 44 (30%) 3 (17%) 41 (32%) survival free from bone metastases amounted to 14 and 27 Fatty tissue infiltration months in Fas-negative and Fas-positive patients, respectively + 52 (35%) 9 (50%) 43 (33%) 0.167 – 95 (65%) 9 (50%) 86 (67%) (p=0.005). Similarly, the aforementioned parameter of Lymph node status survival was shorter in Fas-ligand (–) patients (median, 15.5 + 49 (33%) 4 (22%) 45 (35%) 0.287 months) compared to the Fas-ligand (+) ones (median, 28 – 98 (67%) 14 (78%) 84 (65%) months, p=0.048). Among other clinicopathological Estrogen receptor parameters tested, only the relationship between p53 + 79 (54%) 12 (67%) 67 (52%) 0.256 – 68 (46%) 6 (33%) 62 (48%) expression and the survival free from skeletal spread, though Progesterone receptor insignificant, was worth of attention (p=0.06). + 48 (33%) 7 (40%) 41 (32%) 0.636 – 99 (67%) 11 (60%) 88 (68%) Discussion Fas + 91 (62%) 6 (33%) 85 (66%) 0.017 – 56 (38%) 12 (67%) 44 (34%) Our study revealed that the lack of Fas or Fas-ligand Fas-ligand expression in primary breast cancer was associated with + 70 (48%) 4 (22%) 66 (51%) 0.028 increased risk of its skeletal spread. The lack of significant – 77 (52%) 14 (78%) 63 (49%) association between Fas/Fas-ligand expression and the p53 evidence of breast cancer metastases, if the latter were + 82 (56%) 8 (44%) 74 (57%) 0.330 – 65 (44%) 10 (56%) 55 (43%) analyzed irrespective of their location, suggests that the relationship revealed in our study is specific only for bones as a target for tumor spread. This hypothesis, however, requires verification in a larger population of patients with The association between Fas or Fas-ligand expression and breast cancer spread to various locations. Metastases clinicopathological variables was tested using the Mann-Whitney U- located in other parts of the body than in the bones test. Survivals free from bone metastases were obtained by the Kaplan- occurred so rarely in our material that it was impossible Meier method and compared with the log- test. Calculations were for them to be examined as an independent group in performed using Statistica 5, Version 97 (StatSoft®, Poland) software and statistical significance was defined as p≤0.05. statistical analysis. Our results give a valuable contribution to understanding Results the role of the Fas/Fas-ligand system in breast cancer biology. The results of previous studies indicate that the Bone metastases were recorded in 18/147 (12%) breast absence of Fas in primary tumor is characteristic for breast cancer patients. There was a statistically significant malignancies of the more aggressive phenotype. We have relationship between Fas or Fas-ligand expressions and already revealed that regional lymph node involvement is

216 Be˛benek et al: Fas/Fas-ligand in Breast Cancer as Predictors of its Skeletal Spread

Figure 1. Normal breast tissue (negative control).

Figure 2. Positive immunohistochemical staining of Fas in breast cancer tissue.

Figure 3. Positive immunohistochemical staining of Fas-ligand in breast cancer tissue.

217 ANTICANCER RESEARCH 27: 215-218 (2007) more frequent in patients in whom Fas expression is absent 5 Ragnarsson GB, Mikaelsdottir EK, Vidarsson H, Jonasson JG, in primary breast cancer tissues (13). Also in the present Olafsdottir K, Kristjansdottir K, Kjartansson J, Ogmundsdottir study, the axillary lymph nodes of patients whose tumors HM and Rafnar T: Intracellular Fas ligand in normal and malignant breast epithelium does not induce in Fas- were deficient in Fas, were also more frequently involved sensitive cells. Br J Cancer 83: 1715-1721, 2000. (data not shown) though the aforementioned relationship 6 Shiraki K, Naoki T, Isselbacher KJ and Takahashi H: was insignificant. Expression of Fas ligand in liver metastases of human colonic Some authors claim a relationship exists between Fas adenocarcinomas. Proc Natl Acad Sci 94: 6420-6425, 1997. expression and the duration of disease-free survival. 7 Sjostrom J, Blomqvist C, von Boguslawski K, Bengtsson NO, Mottolese et al. (10) revealed that disease-free survival was Mjaaland I, Malmstrom P, Ostenstadt B, Wist E, Valvere V, significantly longer in patients with Fas-positive tumors Takayama S, Reed JC and Saksela E: The predictive value of bcl-2, bax, bcl-xL, bag-1, fas, and fasL for compared to those with Fas-negative breast cancer tissues. response in advanced breast cancer. Clin Cancer Res 8: 811- These results were further confirmed by Reimer et al. (11) 816, 2002. and Botti et al. (12), who found that a FasL: Fas ratio >1 8 Yakirevich E, Maroun L, Cohen O, Izhak OB, Rennert G and was related to significantly shorter disease-free survival. Resnick MB: Apoptosis, proliferation, and Fas (APO-1, Our univariate analysis revealed that Fas, and to a lesser CD95)/Fas ligand expression in medullary carcinoma of the extent also Fas-ligand, modulate the duration of survival free breast. J Pathol 192: 166-173, 2000. from skeletal metastases. A similar association was, however, 9 Rubens RD and Mundy GR: Cancer and the skeleton. London: Marin Dunitz, 2000. not demonstrated when disease-free survival was analyzed 10 Mottolese M, Buglioni S, Bracalenti C, Cardarelli MA, irrespective of recurrence location. These observations seem Ciabocco L, Giannarelli D, Botti C, Natali PG, Concetti A and to be another demonstration of the significance of Fas/Fas- Venanzi FM: Prognostic relevance of altered Fas (CD95)- ligand expression as specific predictor for the skeletal spread system in human breast cancer. Int J Cancer 89: 127-132, 2000. of breast cancer. 11 Reimer T, Herrnring C, Koczan D, Richter D, Gerber B, Previous studies on the role of Fas/Fas-ligand in skeletal Kabelitz D, Friese K and Thiesen HJ: FasL: Fas ratio – a metastasis of breast cancer were limited to interactions prognostic factor in breast carcinomas. Cancer Res 60: 822-828, 2000. between metastatic cells and structural components of 12 Botti C, Buglioni S, Benevolo M, Giannarelli D, Papaldo P, bones. Cell-culture studies revealed that Fas-ligand present Cognetti F, Vici P, Di Filippo F, Del Nonno F, Venanzi FM, in metastatic cells of breast cancer stimulates apoptosis in Natali PG and Mottolese M: Altered expression of FAS system is Fas-positive osteoblasts (14-16). related to adverse clinical outcome in stage I-II breast cancer Our results in turn suggest that the probability of bone patients treated with adjuvant anthracycline-based chemotherapy. metastases may be assessed on the basis of Fas/Fas-ligand Clin Cancer Res 10: 1360-1365, 2004. ˛ expression in primary breast cancer. Consequently, the 13 Bebenek M, Dus’ D and Koz’lak J: Fas and Fas ligand as prognostic factors in human breast carcinoma. Med Sci Monit determination of these molecules in surgically-removed 12: 457-461, 2006. tissues seems crucial in the context of a patient’s prognosis 14 Mastro AM, Gay CV and Welch DR: The skeleton as a unique and the selection of further adjuvant treatment. environment for breast cancer cells. Clin Exp Metastasis 20: 275-284, 2003. References 15 Mastro AM, Gay CV, Welch DR, Donahue HJ, Jewell J, Mercer R, DiGirolamo D, Chislock EM and Guttridge K: 1 Nagata S and Goldstein P: The Fas death factor. Science 267: Breast cancer cells induce osteoblast apoptosis: a possible 1449-1456, 1995. contributor to bone degradation. J. Cell Biochem 91: 265-276, 2 Bernstorff WV, Glickman JN, Odze RD, Farraye FA, Joo HG, 2004. Goedegebuure PS and Eberlein TJ: Fas (CD95/APO-1) and Fas 16 Mercer RR, Miyasaka C and Mastro AM: Metastatic breast ligand expression in normal pancreas and pancreatic tumors. cancer cells suppress osteoblast adhesion and differentiation. Implications for immune privilege and immune escape. Cancer Clin Exp Metastasis 21: 427-435, 2004. 94: 2552-2560, 2002. 3 Moller P, Koretz K, Leithauser F, Bruderlein S, Henne C, Quentmeier A and Krammer PH: Expression of APO-1 (CD95), a member of the NGF/TNF receptor superfamily, in normal and neoplastic colon epithelium. Int J Cancer 57: 371- 377, 1994. 4 Nagarkatti N and Davis BA: Tamoxifen induces apoptosis in Received June 19, 2006 Fas+ tumor cells by upregulating the expression of Fas ligand. Revised November 1, 2006 Cancer Chemother Pharmacol 51: 284-290, 2003. Accepted November 6, 2006

218