FAST TRACK Folate receptor overexpression is associated with poor outcome in breast cancer Lynn C. Hartmann 1 * , Gary L. Keeney 2 , Wilma L. Lingle 2 , Teresa J.H. Christianson 3 , Bindu Varghese 4 , David Hillman 3 , Ann L. Oberg 3 and Philip S. Low 4 1 Division of Medical Oncology, Mayo Clinic Cancer Center and Mayo Clinic College of Medicine, Rochester, MN 2 Department of Pathology, Mayo Clinic Cancer Center and Mayo Clinic College of Medicine, Rochester, MN 3 Division of Biostatistics, Mayo Clinic Cancer Center and Mayo Clinic College of Medicine, Rochester, MN 4 Purdue University and Purdue Cancer Center, Department of Chemistry, West Lafayette, IN The high affinity folate receptor is a membrane-associated glyco- protein that is preferentially expressed in cancers of epithelial ori- gin and rarely expressed in normal cells. We examined its expres- sion pattern in breast cancer, utilizing a tissue microarray con- taining samples from 63 invasive breast cancers from women with divergent clinical outcomes. Thirty-three women comprised the poor outcome group with a median time to recurrence of 1.9 years. Thirty women, the good outcome group, were free of recurrence for a minimum of 7 years after diagnosis. The intensity of folate receptor staining was strongly correlated with outcome. There were two summary categories of staining intensity: weak (n 5 42) or strong (n 5 21). In the strong staining group, 17 of 21 women (81%) have recurred and their median survival is 2.4 years. In the weak staining group, 16 of 42 women (38%) have recurred. Their median survival is not estimable. After adjustment for tumor size, nodal status, ER status, adjuvant therapy, histology and tumor grade, strong staining for the folate receptor remained significantly associated with poor outcome, p < 0.001. Our work requires validation in a larger cohort, but supports the possibility of using folate receptor-targeted approaches in the management of breast cancer. ' 2007 Wiley-Liss, Inc. Key words: folate receptor; breast cancer; prognosis Folic acid and its reduced congeners are required for one carbon transfer reactions that are used in the biosynthesis of nucleotide bases, amino acids and other methylated compounds, and conse- quently, they are needed in larger quantities by proliferating cells. Physiological folates are transported into cells by either a low af- finity reduced folate carrier (K m 5 10 25 M) or a high affinity fo- late receptor (K d 5 10 210 M). 1–5 The reduced folate carrier is ubiquitously expressed and constitutes the sole folate uptake path- way for most normal cells. The high affinity receptor is a glycosylphosphatidylinositol- anchored membrane protein originally identified as a mAb-defined antigen in placenta and trophoblastic cells. It is rarely expressed or inaccessible in most normal cells. However, it is upregulated in select cancers of epithelial origin. 6 Distribution and binding stud- ies of the high affinity folate receptor have shown high expression in the majority of ovarian cancers, and lesser degrees of positivity in kidney, breast, uterine and lung cancers. 6–10 Studies in ovarian cancer have shown an association between folate receptor overex- pression and tumor aggressiveness. 7,11–13 Studies of folate receptor expression in breast cancers have found varying results, depending on the techniques used and tis- sues analyzed. Ross et al. measured mRNA for the folate receptor and found levels to be elevated in five cancers when compared with normal breast specimens; however, the degree of elevation was 10-fold less than that seen in ovarian cancer. 10 Garin-Chesa et al. used a mouse monoclonal antibody LK26 to assess folate re- ceptor expression in a variety of fresh-frozen cancers. Of 53 breast cancers studied, two showed homogeneous LK26 expression and another nine, patchy staining. 14 Parker et al. used a highly sensi- tive radioactive binding method to quantitate folate receptor expression in frozen tissue homogenates. 6 In seven breast cancers studied, three demonstrated high positivity and three low positiv- ity (one negative). The tumor specificity of the folate receptor makes it a promising target for molecular-based imaging and treatment strategies. Because of its potential as a tumor-specific target, we sought to examine the expression of the folate receptor in breast cancers in greater detail. To address this question, and to assess the possibility of a growth advantage in folate receptor-positive cells, we studied its expression in a set of good and poor-outcome breast cancers. Materials and methods Tissue samples We constructed a tissue microarray of invasive breast cancers selected from women with divergent clinical outcomes. Specifi- cally, of the 63 samples included, 30 were obtained from women who were free of recurrence for a minimum of 7 years from diag- nosis. The other 33 specimens came from women whose disease recurred less than 3.5 years after diagnosis. All cancers were diag- nosed in 1984–1985, assuring sufficient follow-up for the good outcome group. All procedures were reviewed and approved by the Mayo Clinic Institutional Review Board. Folate receptor antibody Immunohistochemistry was performed with mAb343, a mono- clonal IgG1 antibody derived by immunization with human folate receptor purified from the KB nasopharyngeal carcinoma cell line. 15 Clones were selected by radioimmunoassay for the ability to bind 125 I-labeled folic acid complexed to purified folate recep- tor and were subsequently characterized using Western blot, immunohistochemistry and flow cytometry. Tissue microarray construction, immunohistochemistry and digital imaging Arrays were constructed using a custom fabricated device to produce 0.6-mm tissue cores arrayed in a 216 core-capacity recipi- ent block. Multiple cores from each patient tumor block were incorporated into the recipient block in addition to cores of liver as fiducial markers and controls for immunohistochemistry reac- tions. Immunohistochemistry was performed on tissue microarray sections mounted on charged slides. Slides were first treated to unmask epitopes in a BioCare Decloaking chamber (BioCare Medical, Walnut Creek, CA). The remainder of the staining pro- cess was performed on a DAKO Autostainer (Carpinteria, CA) using a monoclonal antibody to the folate receptor (mAb 343) at a Grant sponsor: NIH ; Grant number: CA89581; Grant sponsor: Andersen Foundation. *Correspondence to: Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA. Fax: 1507-284-1803. E-mail: hartmann.lynn@mayo.edu Received 3 January 2007; Accepted after revision 20 March 2007 DOI 10.1002/ijc.22811 Published online 8 May 2007 in Wiley InterScience (www.interscience. wiley.com). Int. J. Cancer: 121, 938–942 (2007) ' 2007 Wiley-Liss, Inc. Publication of the International Union Against Cancer