Autoantibody signatures: progress and perspectives for early cancer detection.

Becoming invasive is a crucial step in cancer development, and the early spread of tumour cells is usually undetected by current imaging technologies. In patients with cancer and no signs of overt metastases, sensitive methods have been developed to identify circulating autoantibodies and their antigen counterparts in several cancers. These technologies are often based on proteomic approaches, and recent advances in protein and antibody microarrays have greatly facilitated the discovery of new antibody biomarkers in sera from cancer patients. Interestingly, in a clinical application setting, combinations of multiple autoantibody reactivities into panel assays have recently been proposed as relevant screening tests and validated in several independent trials. In addition, autoantibody signatures seem to be particularly relevant for early detection of cancer in high-risk cancer patients. In this review, we highlight the concept that immunogenic epitopes associated with the humoural response and key pathogenic pathways elicit serum autoantibodies that can be considered as relevant cancer biomarkers. We outline the proteomic strategies employed to identify and validate their use in clinical practice for cancer screening and diagnosis. We particularly emphasize the clinical utility of autoantibody signatures in several cancers. Finally, we discuss the challenges remaining for clinical validation.

Serum protein signature may improve detection of ductal carcinoma in situ of the breast.

Ductal carcinoma in situ (DCIS) of the breast is part of a spectrum of preinvasive lesions that originate within normal breast tissue and progress to invasive breast cancer. The detection of DCIS is important for the reduction of mortality from breast cancer, but the diagnosis of preinvasive breast tumors is hampered by the lack of an adequate detection method. To identify the changes in protein expression during the initial stage of tumorigenesis and to identify the presence of new DCIS markers, we analysed serum from 60 patients with breast cancer and 60 normal controls using mass spectrometry. A 23-protein index was generated that correctly distinguishes the DCIS and control groups with sensitivities and specificities in excess of 80% in two independent cohorts. Two candidate peptides were purified and identified as platelet factor 4 (PF-4) and complement C3a(desArg) anaphylatoxin (C3a(desArg)) using liquid chromatography-tandem mass spectrometry (LC-MS/MS). In an independent serum set of 165 patients, PF-4 and C3a(desArg) were significantly upregulated in DCIS compared with non-cancerous controls, as validated using western blot and enzyme-linked immunosorbent assay. We conclude that our serum protein-based test, used in conjunction with image-based screening practices, could improve the sensitivity and specificity of breast cancer detection.

Pemphigus vulgaris antigen mRNA quantification for the staging of sentinel lymph nodes in head and neck cancer.

BACKGROUND: Molecular diagnosis has been proposed to enhance the intra-operative diagnosis of sentinel lymph node (SLN) invasion in head and neck squamous cell carcinoma (HNSCC). Although cytokeratin (CK) mRNA quantification with real-time reverse transcriptase-PCR (QRT-PCR) has produced encouraging results, the more discriminating markers remain to be identified. METHODS: Pemphigus vulgaris antigen (PVA), squamous cell carcinoma antigen (SCCA), and CK17 mRNA were quantified using QRT-PCR, and the results were compared with an extensive histopathological examination of the entire SLNs on 78 SLNs harvested from 22 patients with HNSCC. RESULTS: SCCA and CK17 quantification showed significantly higher mRNA values for macrometastases (MAs) than for either negative or isolated tumour cell (ITC) SLNs (P<0.01). Pemphigus vulgaris antigen allowed the discrimination of all MAs and micrometastases from both negative and ITC SLNs (P<0.001). For the neck staging of patients, considering metastatic vs non-metastatic status, receiver-operating characteristic curve analysis found areas under the curve of 93.8, 97.9, and 100% for CK17, SCCA, and PVA, respectively. With PVA, a cutoff value of 562 copies per 100 ng of cDNA permitted the correct distinction between patients with positive as opposed to negative neck nodes in all cases. CONCLUSION: PVA seems to be a highly promising marker for accurate intra-operative SLN staging in HNSCC by QRT-PCR.

Lack of complete 1p19q deletion in a consecutive series of 12 WHO grade II gliomas involving the insula: a marker of worse prognosis?

1p19q codeletion in WHO grade II glioma (GIIG) is claimed to be a marker of better outcome and chemosensitivity. Through the molecular study of 12 insular GIIG, our goal was to investigate a possible anatomo-molecular relationship in insula, specific brain sub-region, known to be involved with high frequency in GIIG. Loss of heterozygosity on 1p and 19q chromosomes was assessed in all tumors. None complete deletion of the 1p and 19q arms chromosomes and partial deletions in 3 patients were retrieved in the 11 WHO grade II oligodendrogliomas and one oligo-astrocytoma analyzed. Discrepancy between our results and the high reported incidence of 1p19q codeletion in oligodendrogliomas could mean that insular GIIG has a specific molecular pattern. We hypothesize that prognosis of insular GIIG is worse than in other locations, with reduced chemosensitivity. Thanks to minimization of surgical risks, surgery might be a first intention therapeutic option proposed in the GIIG.

Proteomic analysis of RCL2 paraffin-embedded tissues.

Histopathological diagnosis in most of the world's hospitals is based upon formalin-fixed and paraffin-embedded (FFPE) tissues. Although this standard fixation and embedding procedure keeps the tissue in excellent form for morphological and immunohistological analysis, FFPE is inappropriate for nucleic acids and protein studies. We investigated the potential value of RCL2, a new non-toxic fixative, for sparing proteins preserved in paraffin-embedded tissues. Normal colonic mucosa tissue was fixed in RCL2 prior to paraffin embedding (RCL2P), and then processed for quality and quantity of protein conservation, as compared to frozen and FFPE tissues using complementary proteomic analysis approaches. Using 4 different protein extraction protocols, RCL2P tissue consistently showed the highest protein yield. Similar protein patterns were observed with RCL2P and frozen tissues using mono and bi-dimensional electrophoresis. Moreover, membrane, cytoplasmic and nuclear proteins, as well as phosphorylated proteins, were successfully detected using western-blot. Furthermore, protein patterns observed by mass spectrometry analysis after laser-captured microdissection were found to be identical for frozen and RCL2-fixed tissues. At last, immunohistochemistry using various antibodies showed comparable results between both tissue storage methods. We concluded that RCL2 has great potential for performing both morphological and molecular analyses on the same archival paraffin-embedded tissue sample, and can be a new method for investigating protein biomarkers.

Proteomics-based identification of HSP60 as a tumor-associated antigen in early stage breast cancer and ductal carcinoma in situ.

The detection of autoantibodies in cancer patients has been shown to constitute an excellent tool for early diagnosis. Because breast cancer still lacks early diagnostic markers, we investigated novel tumor-associated antigens and related autoantibodies in sera from patients with early stage breast cancer compared to autoimmune disease, other cancers, and healthy volunteers, using a proteomics-based approach. Among the 26 protein antigens specifically recognized by early stage breast cancer sera, we focused on Heat Shock Protein 60 (HSP60). Using ELISA, we investigated the frequency of autoantibodies directed against this protein in the sera of 240 individuals, comprising patients with either ductal carcinoma in situ (DCIS) ( n = 49) or early stage breast cancer ( n = 58), other cancers ( n = 20), autoimmune disease ( n = 20), and healthy subjects ( n = 93). Autoantibodies directed against HSP60 were present in 16/49 (31%) early stage breast cancer and 18/58 (32.6%) DCIS patients, compared to 4/93 (4.3%) healthy subjects. In particular, autoantibodies were present in 11/23 patients (47.8%) with high-grade DCIS, compared to 5/26 (19.2%) with low-grade DCIS. HSP60 mRNA levels were significantly higher in primary breast cancer compared to healthy breast tissues. Using immunohistochemistry, we found that HSP60 expression gradually increases from normal through DCIS to invasive tissues. Our results indicate that HSP60 autoantibodies may be of interest in terms of clinical utility for the early diagnosis of breast cancer and more particularly in DCIS. Moreover, HSP60 overexpression during the first steps of breast carcinogenesis may be functionally correlated to tumor growth and/or progression.

Anti-growth factor activities of benzothiophenes in human breast cancer cells.

We have tested the effects of two Eli-Lilly compounds, LY 117, 018 and raloxifene, on E2-regulated and IGF-I-induced proliferation or AP-1 activity in human breast cancer cells. We now demonstrate that both molecules have strong antiestrogenic and anti-growth factor inhibitory effects in MCF7 cells. They were as potent as ICI 182, 780 and more efficient than OH-Tam to prevent estradiol action whereas their inhibition on IGF-I stimulation was less than with ICI 182, 780 and equivalent to that of OH-Tam. Moreover, raloxifene was the most efficient molecule to prevent IGF-I-induced AP-1 activity, with a significant effect observed with a concentration as low as 5 x 10(-11)M in the presence of IGF-I alone. Similar dose-response curves were obtained with a combined treatment of IGF-I and E2 with a 2log shift. Their action on IGF-I-induced proliferation was completely abrogated in MCF7 transfectants in which the expression of an antiestrogen-regulated protein tyrosine phosphatase, PTPL1, was abolished by antisense RNA transfection. Accordingly, they were both able to dose-dependently regulate the expression of PTPL1 and to interfere with the PI3-K/Akt pathway by drastically decreasing Akt phosphorylation exclusively in wild-type PTPL1 expressing cells. Our data altogether demonstrate that raloxifene has a potent inhibitory effect on IGF-I action, with a drastic effect on AP-1 triggered responses as well as on Akt phosphorylation, suggesting that it might be a useful therapeutic agent in tumors in which these signalling pathways become constitutively active.

Aromatase expression in ovarian epithelial cancers.

Our study focused on aromatase cytochrome P450 (CYP19) expression in ovarian epithelial normal and cancer cells and tissues. Aromatase mRNA expression was analyzed by real-time PCR in ovarian epithelial cancer cell lines, in human ovarian surface epithelial (HOSE) cell primary cultures, and in ovarian tissue specimens (n=94), including normal ovaries, ovarian cysts and cancers. Aromatase mRNA was found to be expressed in HOSE cells, in BG1, PEO4 and PEO14, but not in SKOV3 and NIH:OVCAR-3 ovarian cancer cell lines. Correlation analysis of aromatase expression was performed according to clinical, histological and biological parameters. Aromatase expression in ovarian tissue specimens was higher in normal ovaries and cysts than in cancers (P<0.0001). Using laser capture microdissection in normal postmenopausal ovaries, aromatase was found to be predominantly expressed in epithelial cells as compared to stromal component. Using immunohistochemistry (IHC), aromatase was also detected in the epithelium component. There was an inverse correlation between aromatase and ERalpha expression in ovarian tissues (P<0.001, r=-0.34). In the cancer group, no significant differences in aromatase expression were observed according to tumor histotype, grade, stage and survival. Aromatase activity was evaluated in ovarian epithelial cancer (OEC) cell lines by the tritiated water assay and the effects of third-generation aromatase inhibitors (AIs) on aromatase activity and growth were studied. Letrozole and exemestane were able to completely inhibit aromatase activity in BG1 and PEO14 cell lines. Interestingly, both AI showed an antiproliferative effect on the estrogen responsive BG1 cell line co-expressing aromatase and ERalpha. Aromatase expression was found in ovarian epithelial normal tissues and in some ovarian epithelial cancer cells and tissues. This finding raises the possibility that some tumors may respond to estrogen and provides a basis for ascertaining an antimitogenic effect of AI in a subgroup of ovarian epithelial cancers.

Transcriptional and posttranscriptional regulation of fibulin-1 by estrogens leads to differential induction of messenger ribonucleic acid variants in ovarian and breast cancer cells.

Fibulin-1 is an extracellular matrix protein overexpressed in epithelial ovarian and breast cancers. In estrogen receptor (ER)-positive ovarian and breast cancer cell lines, fibulin-1 mRNA levels are markedly increased by estrogens. Transfection experiments using fibulin-1 promoter constructs indicate that 17beta-estradiol (E2) increases fibulin-1 gene transcription and that ERalpha is more potent than ERbeta to mediate E2 regulation of the transfected fibulin-1 promoter. Using SL2 cells devoid of specificity protein 1 (Sp1) and site-directed mutagenesis of GC boxes, we evidenced that the E2 regulation occurs through a proximal specificity protein 1 binding site. In addition, we show that fibulin-1C and -1D mRNAs, the two major fibulin-1 splicing variants, are differentially induced by E2. The induction of both mRNAs variants is direct and independent of a newly synthesized protein intermediate. Interestingly, actinomycin D chase experiments demonstrate that E2 treatment selectively shortens the fibulin-1D mRNA half-life. This indicates that estrogens affect differentially the stability of fibulin-1 variants and may explain the lower accumulation of fibulin-1D mRNA on E2 treatment. In conclusion, our data show that estrogens, via ERalpha, are key regulators of fibulin-1 expression at both the transcriptional and posttranscriptional levels. The preferential induction of the fibulin-1C variant, which is overexpressed in ovarian and breast cancer, might play an important role in estrogen-promoted carcinogenesis.

17beta-estradiol downregulates beta3-integrin expression in differentiating and mature human osteoclasts.

The increased bone resorption observed after estrogen withdrawal is responsible for bone loss and may lead to osteoporosis. The mechanism by which estradiol inhibits bone resorption is known to involve decreased osteoclastogenesis, however, the effect on osteoclast adhesion remains unclear. We examined the in vitro effect of estradiol and raloxifene on human osteoclast differentiation and function. Human peripheral blood mononuclear cells were cultured with M-CSF/RANK-L for 18 days, and we evaluated bone resorption, the expression of the protein and mRNA of the integrins, c-jun and c-fos in the presence or absence of estradiol. In this human model, beta3-integrin expression increased at the mRNA and protein levels during osteoclast differentiation, whereas that of beta5-integrin did not. We found that estradiol and raloxifene directly inhibited bone resorption on bone slices by 50%, and decreased the expression of beta3-integrin mRNA (60%) and protein (20%) in a time-dependent manner. Moreover, the mRNAs of c-fos and c-jun were both diminished by estradiol and raloxifene, particularly in early osteoclasts, but also to a lesser extent in mature cells. These findings suggest that the direct inhibitory action of estradiol on bone resorption may affect human osteoclast differentiation through downregulation of c-fos and c-jun and adhesion through modulation of beta3-integrin.

Novel germline RET mutation segregating with papillary thyroid carcinomas.

The RET proto-oncogene is responsible for inherited medullary thyroid cancer syndromes. RET is also found mutated in sporadic medullary thyroid cancer (MTC) and rearranged in sporadic papillary thyroid carcinomas. Here, we describe a previously unreported germline RET mutation at codon 603 in exon 10 associated with both MTC and nonmedullary thyroid cancer (NMTC) in a kindred. RET may thus not be excluded as a potential candidate for predisposition to some forms of NMTC.

Analysis of the potential contribution of estrogen receptor (ER) beta in ER cytosolic assay of breast cancer.

Estrogen receptor (ER) content is the most useful parameter for predicting hormone response therapy in breast cancer. Assays available for detecting ER in breast tumor cytosol are ligand-binding assay (LBA), which detects both ERalpha and ERbeta, and the enzymatic immunoassay (EIA), in which monoclonal antibodies are directed against ERalpha. As shown in several studies, the 2 assays correlate and both are used routinely. However, some discrepancies between the 2 assays were found and explanations remain controversial. We evaluated ERalpha and ERbeta mRNA coexpression in breast tumors in order to study whether the presence of ERbeta could account for differences between LBA and EIA in the determination of ER protein level. Using HeLa cell lines transfected with either ERalpha or ERbeta, we confirmed that EIA, using H222 and D547 monoclonal antibodies, recognizes only ERalpha expression, whereas LBA detects both isoforms. In 119 breast tumor cytosols, the correlation between ER-EIA and ER-LBA was high (r = 0.72), although some discrepancies were found. When analyzing ER mRNA expression of samples with higher LBA values, no overexpression of ERbeta mRNA relatively to ERalpha mRNA were observed. There was a difference in ERbeta/ERalpha ratio between ER-negative and ER-positive samples, with a 10-fold increased median ratio in ER-negative samples (p = 0.01). We thus confirmed that the major form of ER in breast cancer is the ERalpha at both the protein and mRNA levels. Moreover, our data do not support the hypothesis that ERbeta expression could explain differences between LBA and EIA in the determination of ER protein level.

A prospective prognostic study of the hormonal milieu at the time of surgery in premenopausal breast carcinoma.

BACKGROUND: Despite numerous studies, the influence of timing at surgery in relation to the menstrual cycle on the prognosis of breast carcinoma is still controversial. Most studies are retrospective, and the reliability of the menstrual history data is limited by the lack of hormonal assessment at the time of surgery. The authors prospectively studied the influence of the menstrual cycle phase as determined by circulating hormones at the time of surgery on the outcome of breast carcinoma. METHODS: A population of 360 premenopausal women with nonmetastatic breast carcinoma operated on from 1992 to 1995 was analyzed. Serum estradiol, progesterone, follicle-stimulating hormone (FSH), and luteinizing hormone (LH) levels were assayed the day of surgery to define the menstrual cycle phase (follicular, n = 186; ovulatory, n = 24; luteal, n = 150). The mean follow-up was 48 months. RESULTS: There were no relations between the menstrual phase at surgery and tumor size, cathepsin D level, Scarff-Bloom-Richardson grade, Pg receptor (PgR), and the number of positive lymph nodes. The mean estrogen receptor level was higher during the follicular phase than in the ovulatory and luteal phases (P < 0.02). Univariate analysis of recurrence free survival (RFS) and overall survival (OS) showed no relations with the menstrual phase or the level of estradiol and progesterone at the time of surgery. High LH or FSH levels (above the medians) were associated with shorter RFS (P = 0.02 and P = 0.04, respectively) or OS (P < or = 0.01 and P = 0.01, respectively). In multivariate analysis, lymph node status, PgR status and LH level were the most significant parameters for predicting OS. There appeared to be no survival differences between menstrual cycle groups after stratification by lymph node status. CONCLUSIONS: This prospective study showed a lack of prognostic value of timing at surgery in relation to the menstrual period or to estrogen and progesterone levels in premenopausal breast carcinoma. Conversely, high gonadotropin levels could predict OS independently of other prognostic factors.

Neoadjuvant tamoxifen for operable breast cancer: a need for phase III studies?

We conducted a case-control study to analyze the effect of neoadjuvant tamoxifen on steroid receptors and histologic grade and to evaluate the feasibility of phase III studies in operable breast cancer. Between 1987 and 1990, 107 patients without clinical metastases who had had no chemotherapy preoperatively, were treated preoperatively with 20 mg/day of tamoxifen for 3 weeks. Of them, 92 were matched with controls for age at diagnosis, year of diagnosis, presence or absence of lymph node involvement, and preoperative radiotherapy. The percentage of ER1 tumors (P = .03) and the mean and median ER levels (P<.001 for both) were lower in the tamoxifen group than in the control group. In six patients analyzed longitudinally, the mean ER decreased from 52 to 19 fmol/mg protein. The difference in relapse-free survival between the two groups was not significant (mean follow-up 87 months). This study suggests a decrease in ER content in patients treated with neoadjuvant tamoxifen. This change may thus be taken into account when ER determination is performed after tamoxifen therapy is started. Further randomized trials should determine whether patients with operable breast cancer benefit from neoadjuvant tamoxifen treatment.

Semiquantitative reverse transcription-polymerase chain reaction to evaluate the expression patterns of genes involved in the oestrogen pathway.

The increasing number of factors to be taken into account in the oestrogen transcriptional process has created a need to develop a rapid screening method to evaluate their role in physiology and pathology. Molecular biology techniques enable gene expression studies at the mRNA level with small amounts of tissues. We therefore developed a semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR) technique using fluorescent oligonucleotides to analyse simultaneously a large panel of interrelated genes involved in the oestrogen transcriptional pathway using a moderately expressed housekeeping gene, the hypoxanthine phosphoribosyltransferase gene (HPRT), as the reference gene. Expression levels of oestrogen receptors (ERalpha, ERbeta), cofactors AIB1, RIP140, SMRT and the Fas-associated protein-tyrosine phophatase-1 (FAP-1) genes were evaluated in breast, endometrial and ovarian cancer cell lines and in three ERalpha-positive and three ERalpha-negative breast cancer tumours. This technique provides a rapid and reliable way to quantitate simultaneously numerous mRNAs of genes involved in the oestrogen pathway from small amounts of tissues.

Dissociated overexpression of cathepsin D and estrogen receptor alpha in preinvasive mammary tumors.

The role of estrogen as a promoter agent of sporadic breast cancer has been considered by assaying, in benign breast disease (BBD) and in situ carcinomas (CIS), 2 markers, the estrogen receptor alpha (ERalpha) and cathepsin D (cath-D) involved in estrogen action on mammary tissue. ERalpha and cath-D were assayed by quantitative immunohistochemistry using an image analyzer in 170 lesions of varying histological risk (94 BBD and 76 CIS), and in "normal" glands close to these lesions. The ERalpha level increased significantly in proliferative BBD with atypia (P < .001), in non-high-grade CIS (P < .001), and in adjacent "normal" glands. ERalpha level was decreased in high-grade ductal CIS (DCIS) and also in adjacent "normal" glands. Cath-D level increased in ductal proliferative BBD (P < or = .01) and in high-grade DCIS (P < or = .003), but not in the other lesions. After menopause, ERalpha level was increased (P = .012) but not cath-D level. According to Mac Neman test, the high-grade DCIS were predominantly ERalpha negative and cath-D positive (P = .0017), and the other CIS were predominantly ERalpha positive and cath-D negative (P = .0002). The 2 markers are overexpressed early in premalignant lesions, but independently. This dissociation suggests a branched model of mammary carcinogenesis involving 1 estrogen-independent pathway with high cath-D and low ERalpha levels (including high-grade DCIS) and 1 estrogen-dependent pathway, with high ERalpha level (including proliferative BBD with atypia and low-grade DCIS). We propose that ERalpha-negative breast cancers may develop directly from high-grade DCIS and that ERalpha assay in preinvasive lesions should be considered in prevention trials with antiestrogens.

[Anti-estrogens, selective estrogen receptor modulators (SERM), tibolone: modes of action]. / Antioestrogènes, SERM, tibolone: mécanisme d'action.

The regulation of estrogen-induced cellular effects is a multistep molecular process. The diversity of estrogen and anti-estrogen effects on cellular functions is also modulate by tissue and gene specificity. This diversity may be explained by different levels of molecular regulation, including the presence of two distinct estrogen receptor isoforms (ER alpha and ER beta), their binding to activator or corepressor transcriptional proteins, and their affinity to different DNA binding domains of target genes (estrogen responsive element or API). These mechanisms may account for the specific responses to estrogens or anti-estrogens according to tissue, cell or gene level. The anti-estrogen tamoxifen, in vitro, inhibits the estrogen-induced proliferation of breast cancer cells and, in vivo, enhances long-term prognosis of patients having ER positive breast cancer when it is used as an adjuvant treatment. The partial agonist effect of anti-estrogens such as raloxifene, is observed only on bones and vessels but not in endometrium. Tibolone is another class of ER modulators which acts as a prodrug. It is metabolized in compounds activating nuclear receptors (androgen and progesterone receptors) which modify the ER level and estrogen metabolism. The improvement of current knowledge on the cellular mechanisms of estrogen and anti-estrogens action should allow the elaboration new therapeutic approaches on specific functions involved in estrogen-dependent pathologies.

Time at surgery during menstrual cycle and menopause affects pS2 but not cathepsin D levels in breast cancer.

Many studies have addressed the clinical value of pS2 as a marker of hormone responsiveness and of cathepsin D (Cath D) as a prognostic factor in breast cancer. Because pS2 and Cath D are both oestrogen induced in human breast cancer cell lines, we studied the influence of the menstrual cycle phase and menopausal status at the time of surgery on the levels of these proteins in breast cancer. A population of 1750 patients with breast cancer, including 339 women in menstrual cycle, was analysed. Tumoral Cath D and pS2 were measured by radioimmunoassay. Serum oestradiol (E2), progesterone (Pg), follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels at the day of surgery were used to define the hormonal phase in premenopausal women. There was a trend towards a higher mean pS2 level in the follicular phase compared with the luteal phase (17 ng mg(-1) and 11 ng mg(-1) respectively, P = 0.09). Mean pS2 was lower in menopausal patients than in women with cycle (8 ng mg(-1) and 14 ng mg(-1) respectively, P = 0.0001). No differences in mean Cath D level were observed between the different phases of the menstrual cycle, or between pre- and post-menopausal women. In the overall population, pS2 was slightly positively associated with E2 and Pg levels and negatively associated with FSH and LH, probably reflecting the link between pS2 and menopausal status. In premenopausal women, no association was found between pS2 and E2, Pg, FSH or LH levels. There were no correlations between Cath D level and circulating hormone levels in the overall population. However, in the subgroup of premenopausal women with ER-positive (ER+) tumours, E2 was slightly associated with both pS2 and Cath D, consistent with oestrogen induction of these proteins in ER+ breast cancer cell lines. There are changes in pS2 level in breast cancer throughout the menstrual cycle and menopause. This suggests that the choice of the pS2 cut-off level should take the hormonal status at the time of surgery into account. In contrast, the level of Cath D is unrelated to the menstrual cycle and menopausal status.