Node-negative Primary Breast Research Articles

Abstract A124: Characterization of claudin-low breast cancers

Abstract Molecular profiling of human breast cancers has defined 5 molecular subtypes: luminal A, luminal B, HER2 over-expressing, basal-like and claudin-low. The claudin-low subtype was identified in 2007 and is characterized by low expression of claudin proteins, E-cadherin and markers of luminal differentiation, and is reported to be associated with expression of mesenchymal and cancer stem cell (CSC) markers. The prevalence of this subtype is reported to be between 7-14% and there are an excess of tumors with metaplastic and medullary-like features, an association with poor prognosis and evidence that they may be resistant to conventional chemotherapies. Using information from publicly available gene expression microarray data, we sought to identify immunohistochemical markers of the claudin-low subtype and to further describe the morphological features of claudin-low breast tumors and the overall survival characteristics of patients with these tumors, along with any associations between these tumors and CSC markers. Using the gene expression microarray datasets, we performed hierarchical clustering to assign a molecular subtype to the tumors. Differential gene expression analysis was used to identify genes that were significantly upregulated and downregulated between claudin-low tumors and other tumor subtypes as candidates for immunohistochemical markers for our formalin fixed paraffin embedded (FFPE) tumor cohort. We utilized 943 stage I or stage II, lymph node negative primary invasive breast cancers treated with breast conserving surgery and adjuvant radiation, which had FFPE tumor blocks available for tissue microarray construction. On the basis of IHC expression of ER, PR, HER2, Ki67, EGFR, CK5/6, Claudin proteins and E-cadherin, tumors were classified as luminal A, luminal B, HER2 over-expressing, basal-like or claudin-low. Kaplan-Meier methods were used to estimate overall survival, and Fisher's exact tests were used to compare tumor characteristics and expression of CSC markers (ALDH1, CD44hi/CD24low) between claudin-low and luminal A tumors. Claudin-low tumors comprised 8% of our cohort with an overall survival of 73.6% at a median follow up of 12 years, similar to that of basal-like and HER2 over-expressing subtypes. Compared to luminal A type tumors, the claudin-low tumors were statistically more likely to have circumscribed tumor margins. However, there was no statistically significant association between claudin-low subtype and the expression of CSC markers. The claudin-low subtype represents a minority of invasive breast cancers, however this group is characterized by poor prognosis, and as such the identification of these tumors is useful to determine treatment options in the clinical setting. Citation Format: Kay Dias, Anna Dvorkin-Gheva, Greg Pond, Mark Levine, Timothy Whelan, Anita Bane. Characterization of claudin-low breast cancers. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Breast Cancer Research: Genetics, Biology, and Clinical Applications; Oct 3-6, 2013; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2013;11(10 Suppl):Abstract nr A124.

P1-06-06: The Importance of CXCL10 and CXCR3-A in Breast Cancer.

Abstract Background: CXCL10 is a chemokine with chemo attractant properties which signals through G protein-coupled receptor CXCR3. At least two reported isoforms exists; CXCR3-A and CXCR3-B. CXCL10 is involved in the recruitment of leucocytes, immune modulation and angiogenesis. In animal models CXCL10 inhibits tumour growth, metastase formation and inhibits angiogenesis. The receptors have opposite features; CXCR3-A mediates proliferative or anti-apoptotic response and CXCR3-B mediates growth inhibition and apoptosis in response to ligand binding. The aim of this study was to investigate the role of CXCL10 and CXCR3-A as prognostic or tamoxifen treatment predictive factors in breast cancer. Material and Methods: A randomized tamoxifen trial comprising 1,780 breast cancer patients was conducted in Stockholm, Sweden, 1976–1990. All patients had lymph node negative primary breast cancer and were postmenopausal at the time of diagnosis. The patients were randomized to tamoxifen or no endocrine treatment. We analyzed the protein expression of CXCL10 and the receptor CXCR3A with immunohistochemistry using tissue microarrays, which were constructed from paraffin blocks originating from 912 patients. CXCL10 and CXCR3A could be scored in 793 (87%) and 735 (81%), respectively. Results: High expression of CXCL10 was associated with significant benefit of tamoxifen treatment concerning local recurrence and breast cancer survival (P=0.02 and P=0.008). For patients with high CXCL10 expression, tamoxifen decreased the risk concerning local recurrence (rate ratio (RR) = 0.56, 95% C.I. 0.33−0.97, P=0.0029) and breast cancer survival (RR 0.6 95% C.I. 0.39−0.97, P=0.023). High expression of CXCR3-A was associated with significant benefit of tamoxifen treatment concerning total recurrence and breast cancer survival (P=0.00001 and P=0.004, respectively). For patients with high CXCR3-A expression, tamoxifen decreased the risk of total recurrence (RR 0.54 95% C.I. 0.35−0.83, P=0.005) and breast cancer survival (RR 0.49 95% C.I. 0.35−0.69, P=0.00003). On the opposite, in the cohort of patients with no endocrine treatment and high CXCR3-A expression, there was an increased risk of total recurrence (P=0.003). (RR 2.07 95% C.I. 1.18−3.61, P=0.011). Conclusions: This study indicates that high expression of CXCL10 can be used as predictive markers for tamoxifen treatment. Further, high expression of CXCR3-A is related to bad prognosis and predicts tamoxifen treatment in breast cancer patients. This needs to be further evaluated. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-06-06.

Abstract 3142: Expression of C-X-C motif chemokine 10 (CXCL10) in breast cancer

Abstract Chemokines are small cytokine-like secreted proteins with chemoattractant properties which function in the recruitment of leucocytes into inflamed tissue. Besides well studied roles in the immune system, chemokines and their receptors have also been shown to play critical roles in tumorigenesis. The present study focuses on C-X-C motif chemokine 10 (CXCL10), a potent chemoattractant for stimulated T-cells and NK-cells. Both CXCL10 and its receptor CXCR3 are expressed by breast tumor cells, and accumulating data suggests roles for this chemokine and its receptor in breast cancer development and proliferation. We have previously shown CXCL10 to be significantly upregulated in ER positive as well as negative breast cancer cell lines over-expressing 17β hydroxysteroid dehydrogenase 14 (17βHSD14). Using a validated anti-CXCL10 primary antibody, we examined the relationship between CXCL10 expression and clinicopathological features in tissue microarrays comprised of tumor material from 912 breast cancer patients participating in a randomized tamoxifen trial conducted in Stockholm, Sweden, 1976-1990. All patients had lymph node negative primary breast cancer and were postmenopausal at the time of diagnosis. Results were obtained from tumors of 793 patients, of which 6% were deemed negative, 23% weak, 28% moderate, and 42% strong. Staining intensity of CXCL10 was strongly correlated with expression of 17βHSD14 (p<0.0001), however no significant correlation was observed between CXCL10 and prognostic markers such as ER, PR or tumor size. No clear associations between CXCL10-expression levels and the clinical outcome of tamoxifen treatment were observed, and moreover no prognostic value of CXCL10 was seen in systemically untreated patients. In the present study we have investigated expression of CXCL10 in a breast tumour cohort. We have previously found a functional relationship between 17βHSD14 and CXCL10 suggesting a possible regulatory loop involving this 17βHSD enzyme. A strong association between CXCL10 and 17βHSD14 in this tumor cohort further strengthens this theory. We found CXCL10 to be ubiquitously expressed in the tumors however no significant relationship between CXCL10 expression and tamoxifen treatment prediction or prognosis among patients was noted. Previous data on the action of CXCL10 in cancer is conflicting as the chemokine has been suggested to act as a growth promoter as well as an inhibitor growth. The presence of two splice variants of the CXCL10 receptor CXCR3, of which CXCR3-A has been shown to promote proliferation and CXCR3-B- to induce apoptosis, has generally not been accounted for, and varying expression levels of these receptors may to some degree explain these conflicting findings. In order to elucidate the role of CXCL10 in breast cancer, tumoral distribution and expression levels of these two splice variants of the CXCL10 receptor need to be evaluated. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3142. doi:10.1158/1538-7445.AM2011-3142

Cancer Classification from Microarray Data using Gene Feature Ranking

A significant challenge in DNA (Deoxyribo Nucleic Acid) microarray analysis can be attributed to the problem of having a large number of features (genes) but with a small number of samples in the dataset. When applying statistical methods to analyse the microarray data, particular care is required to deal with problem such as the low classification accuracy of models brought about by the small number of features that have predictive capability. To overcome these problems, proper approaches for data normalisation, feature reduction, and identifying the optimal set of genes are critical. In this paper, we apply the Gene Feature Ranking [5] method to select genes with high trust values from high dimensional cancer microarray datasets. Our contribution lies in the use of a different metric for calculating the trust values that are more domain specific for cancer datasets. By choosing a pre-defined threshold based on user's knowledge, only genes that show sufficient trustworthiness to be considered for constructing the classification model are retained. Through experimentation on three microarray datasets, namely Acute Lymphoblastic Leukemia (ALL), lymph node negative primary breast cancer, and High Grade Glioma, we are able to confirm that the classification accuracy obtained by the genes selected by the modified GFR method is consistently higher than when the method was not used.

Significance of 17bHSD Type 14 as a Predictive Factor for Adjuvant Tamoxifen Treatment Response in Breast Cancer.

Abstract Background: Peripheral sex steroid synthesis in target organs from adrenally derived steroid precursors is believed to be an ongoing event throughout life, but becomes exceedingly important as the gonadal sources of sex steroids cease, as is the case for estrogen synthesis in women after menopause. 17β hydroxysteroid dehydrogenases (17bHSDs) comprise a family of 14 enzymes catalyzing the stereospecific oxidation/reduction at position C17 of estrogens and androgens. Due to their importance in fine-tuning estrogen levels, the 17bHSDs have been studied in breast cancer. In a previous study examining mRNA of 17bHSD type 14, an enzyme driving the oxidation of estradiol into less potent estrone, from a cohort of stage 2 tumors, high levels were associated with longer disease-free survival. The present study was conducted in order to investigate whether this result could be repeated in another group of patients examining protein expression and to investigate whether 17bHSD type 14 has any prognostic or predictive importance in breast cancer patients systemically treated or untreated with tamoxifen. Material and Methods: A peptide sequence corresponding to amino acids 255 to 269 of human 17bHSD type 14 was synthesized and used for immunization of rabbits with the subsequent purification of a polyclonal antibody towards the antigen. A randomized tamoxifen trial comprising 1780 breast cancer patients was conducted in Stockholm, Sweden, 1976-1990. All patients had lymph node negative primary breast cancer and were postmenopausal at the time of diagnosis. We analyzed the protein expression of 17bHSD type 14 with immunohistochemistry using tissue microarrays, which were constructed from paraffin blocks originating from 912 patients. Results: Results were obtained from 847 patients. Among patients with estrogen receptor positive tumors, high expression of 17bHSD type 14 was associated with a significantly better benefit from tamoxifen as shown by a higher local recurrence-free probability (RR 0.38; 95% C.I. 0.19-0.77, p=0.006) when compared to patients bearing tumors with intermediate, low or negative expression levels (RR 1.20; 95% C.I. 0.54-2.60; p=0.67). The interaction test was significant (p=0.03). No significant difference in tamoxifen response was found when looking at 17bHSD type 14 in relation to breast cancer specific mortality. Furthermore, for untreated patients no prognostic significance for 17bHSD type 14 was found. Discussion: In the current study we show that intratumoral levels of 17bHSD type 14, an enzyme potentially involved in lowering the levels of potent estradiol, is a predictive factor in determining the local response to adjuvant tamoxifen treatment. The fact that patients with ER positive tumors have a greater possibility of responding well to tamoxifen when the expression levels of 17bHSD type 14 are high suggests that some patients would gain better from other treatments than tamoxifen, e.g. aromatase inhibitors and/or inhibitors of reducing 17bHSD enzymes. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 2015.

CITED2 and NCOR2 in anti-oestrogen resistance and progression of breast cancer

Background:Endocrine therapies of breast cancer are effective but ultimately fail because of the development of treatment resistance. We have previously revealed several genes leading to tamoxifen resistance in vitro by retroviral insertion mutagenesis. To understand the manner in which these genes yield tamoxifen resistance, their effects on global gene expression were studied and those genes resulting in a distinct gene expression profile were further investigated for their clinical relevance.Methods:Gene expression profiles of 69 human breast cancer cell lines that were made tamoxifen resistant through retroviral insertion mutagenesis were obtained using oligonucleotide arrays and analysed with bioinformatic tools. mRNA levels of NCOR2 and CITED2 in oestrogen receptor-positive breast tumours were determined by quantitative RT–PCR. mRNA levels were evaluated for association with metastasis-free survival (MFS) in 620 patients with lymph node-negative primary breast cancer who did not receive systemic adjuvant therapy, and with clinical benefit in 296 patients receiving tamoxifen therapy for recurrent breast cancer.Results:mRNA expression profiles of most tamoxifen-resistant cell lines were strikingly similar, except for the subgroups of cell lines in which NCOR2 or CITED2 were targeted by the retrovirus. Both NCOR2 and CITED2 mRNA levels were associated with MFS, that is, tumour aggressiveness, independently of traditional prognostic factors. In addition, high CITED2 mRNA levels were predictive for a clinical benefit from first-line tamoxifen treatment in patients with advanced disease.Conclusions:Most retrovirally targeted genes yielding tamoxifen resistance in our cell lines do not impose a distinctive expression profile, suggesting that their causative role in cell growth may be accomplished by post-transcriptional processes. The associations of NCOR2 and CITED2 with outcome in oestrogen receptor-positive breast cancer patients underscore the clinical relevance of functional genetic screens to better understand disease progression, which may ultimately lead to the development of improved treatment options.

Copy Number Alterations that Predict Metastatic Capability of Human Breast Cancer

We have analyzed the DNA copy numbers for over 100,000 single-nucleotide polymorphism loci across the human genome in genomic DNA from 313 lymph node-negative primary breast tumors for which genome-wide gene expression data were also available. Combining these two data sets allowed us to identify the genomic loci and their mapped genes, having high correlation with distant metastasis. An estimation of the likely response based on published predictive signatures was performed in the identified prognostic subgroups defined by gene expression and DNA copy number data. In the training set of 200 patients, we constructed an 81-gene prognostic copy number signature (CNS) that identified a subgroup of patients with increased probability of distant metastasis in the independent validation set of 113 patients [hazard ratio (HR), 2.8; 95% confidence interval (95% CI), 1.4-5.6] and in an external data set of 116 patients (HR, 3.7; 95% CI, 1.3-10.6). These high-risk patients constituted a subset of the high-risk patients predicted by our previously established 76-gene gene expression signature (GES). This very poor prognostic group identified by CNS and GES was putatively more resistant to preoperative paclitaxel and 5-fluorouracil-doxorubicin-cyclophosphamide combination chemotherapy (P = 0.0048), particularly against the doxorubicin compound, while potentially benefiting from etoposide. Our study shows the feasibility of using copy number alterations to predict patient prognostic outcome. When combined with gene expression-based signatures for prognosis, the CNS refines risk classification and can help identify those breast cancer patients who have a significantly worse outlook in prognosis and a potential differential response to chemotherapeutic drugs.

Internal mammary lymph drainage and sentinel node biopsy in breast cancer – a study on 1008 patients

Abstract Introduction Nowadays, axillary sentinel node (SN) biopsy is a standard procedure in the staging of breast cancer. Although the internal mammary (IM) lymph node status is a major independent prognostic factor in breast cancer patients, sampling of IM sentinel nodes (IMSNs) is not performed routinely. The aim of this study was to determine the likelihood of finding IM lymph node metastases in case of IM hotspots on lymphoscintigraphy and evaluate the relevance of IMSN biopsy as a method to improve staging. Patients and methods Between April 1997 and May 2006, a total of 1008 consecutive patients with clinically node-negative operable primary breast cancer were enrolled in a prospective study on SN biopsy. Both axillary and IMSNs were sampled, based on lymphoscintigraphy, intraoperative gamma probe detection and blue dye mapping, using 10 mCi (370 MBq) 99mTc-nanocolloid injected peritumorally, and 0.5–1.0 ml Patent Blue V injected intradermally. Results Lymphoscintigraphy showed axillary sentinel nodes in 98% (989/1008) and IMSNs in 20% of the patients (196/1008). Sampling the IM basin, as based on the results of lymphoscintigraphy, was successful in 71% of the patients (139/196) and revealed metastases in 22% (31/139). In 29% of the patients with positive IMSNs (9/31) no axillary metastases were found. Conclusion Evaluation of IMSNs improves nodal staging in breast cancer. Patients with IM hotspots on lymphoscintigraphy have a substantial risk (22%) of metastatic involvement of the IM chain. In addition, true IM node-negative patients can be spared the morbidity associated with adjuvant radiotherapy.

Bisphosphonate treatment in primary breast cancer: Results from a randomised comparison of oral pamidronate versus no pamidronate in patients with primary breast cancer

Purpose and patients. During the period from January 1990 to January 1996 a total of 953 patients with lymph node negative primary breast cancer were randomised to oral pamidronate (n=460) 150 mg twice daily for 4 years or no adjuvant pamidronate (n=493) in order to investigate whether oral pamidronate can prevent the occurrence of bone metastases and fractures. The patients received adjuvant chemotherapy, loco-regional radiation therapy, but no endocrine treatment. Results. During the follow-up period the number of patients with pure bone metastases was 35 in the control group and 31 in the pamidronate group. The number of patients with a combination of bone and other distant metastases were 22 in the control group and 20 in the pamidronate group. The hazard rate ratio for recurrence in bone in the pamidronate group compared to the control group was 1.03 (95% confidence interval 0.75–1.40) and p=0.86. No effect was observed on overall survival. In a small subgroup of 27 patients from the study, 12 of whom were treated with pamidronate a significant bone preserving effect was observed on bone mineral density in the lumbar spine, but not in the proximal femur. Conclusion. The results from the trial do not support a beneficial effect of oral pamidronate on the occurrence of bone metastases or fractures in patients with primary breast cancer receiving adjuvant chemotherapy.

Which Cyclin E Prevails as Prognostic Marker for Breast Cancer? Results from a Retrospective Study Involving 635 Lymph Node–Negative Breast Cancer Patients

To evaluate the prognostic value of cyclin E with a quantitative method for lymph node-negative primary breast cancer patients. mRNA transcripts of full-length and splice variants of cyclin E1 (CCNE1) and cyclin E2 (CCNE2) were measured by real-time PCR in frozen tumor samples from 635 lymph node-negative breast cancer patients who had not received neoadjuvant or adjuvant systemic therapy. None of the PCR assays designed for the specific splice variants of the cyclins gave additional prognosis-related information compared with the common assays able to detect all variants. In Cox multivariate analysis, corrected for the traditional prognostic factors, high levels of cyclin E were independently associated with a short distant metastasis-free survival [hazard ratio (HR), 3.40; P < 0.001 for CCNE1 and HR, 1.76; P < 0.001 for CCNE2, respectively]. After dichotomizing the tumors at the median level of 70% tumor cells, the multivariate analysis showed particularly strong results for CCNE1 in the group of 433 patients with stroma-enriched primary tumors (HR, 5.12; P < 0.001). In these tumors, the worst prognosis was found for patients with estrogen receptor-negative tumors expressing high CCNE1 (HR, 9.89; P < 0.001) and for patients with small (T1) tumors expressing high CCNE1 (HR, 8.47; P < 0.001). Our study shows that both CCNE1 and CCNE2 qualify as independent prognostic markers for lymph node-negative breast cancer patients, and that CCNE1 may provide additional information for specific subgroups of patients.

Multicenter Validation of a Gene Expression–Based Prognostic Signature in Lymph Node–Negative Primary Breast Cancer

We previously identified in a single-center study a 76-gene prognostic signature for lymph node-negative (LNN) breast cancer patients. The aim of this study was to validate this gene signature in an independent more diverse population of LNN patients from multiple institutions. Using custom-designed DNA chips we analyzed the expression of the 76 genes in RNA of frozen tumor samples from 180 LNN patients who did not receive adjuvant systemic treatment. In this independent validation, the 76-gene signature was highly informative in identifying patients with distant metastasis within 5 years (hazard ratio, [HR], 7.41; 95% CI, 2.63 to 20.9), even when corrected for traditional prognostic factors in multivariate analysis (HR, 11.36; 95% CI, 2.67 to 48.4). The actuarial 5- and 10-year distant metastasis-free survival were 96% (95% CI, 89% to 99%) and 94% (95% CI, 83% to 98%), respectively, for the good profile group and 74% (95% CI, 64% to 81%) and 65% (53% to 74%), respectively for the poor profile group. The sensitivity for 5-yr distant metastasis-free survival was 90%, and the specificity was 50%. The positive and negative predictive values were 38% (95% CI, 29% to 47%) and 94% (95% CI, 86% to 97%), respectively. The 76-gene signature was confirmed as a strong prognostic factor in subgroups of estrogen receptor-positive patients, pre- and postmenopausal patients, and patients with tumor sizes 20 mm or smaller. The subgroup of patients with estrogen receptor-negative tumors was considered too small to perform a separate analysis. Our data provide a strong methodologic and clinical multicenter validation of the predefined prognostic 76-gene signature in LNN breast cancer patients.

Prognostic Gene Expression Signatures Can Be Measured in Tissues Collected in RNAlater Preservative

Gene expression signatures have the ability to serve in both prognostic and predictive capacities in patient management. The use of RNA as the starting material and the lability of this analyte, however, dictate that tissues must be snap-frozen or stored in a solution that can maintain the integrity of the RNA. We compared pairs of snap-frozen and RNAlater preservative-suspended tissue from 30 such paired lymph node-negative breast tumors and 21 such paired Dukes' B colon tumors. We assessed the correlation of gene expression profiles and prediction of recurrence based on two prognostic algorithms. Tissues stored in RNAlater preservative generated expression profiles with excellent correlation (average Pearson correlation coefficients of 0.97 and 0.94 for the breast and colon tumor pairs, respectively) compared to those produced by tissues that were snap-frozen. The correlation in the prediction of recurrence was 97% and 95% for the breast and colon tumor pairs, respectively, between these two types of tissue handling protocols. This novel finding demonstrates that prognostic signatures can be obtained from RNAlater preservative-suspended tissues, an important step in bringing gene expression signatures to the clinic.

The prognostic value of vascular endothelial growth factor, urokinase plasminogen activator and plasminogen activator inhibitor-1 in node-negative breast cancer

The vascular endothelial growth factor (VEGF) and the plasminogen activator system play an essential role in solid tumor angiogenesis and in tumor invasion and metastasis. In the present study we investigated the relationship between patient outcome and levels of VEGF, urokinase plasminogen activator (uPA) and plasminogen activator inhibitor-1 (PAI-1) in tumor cytosols of 196 node-negative primary invasive breast cancer patients who did not receive any adjuvant therapy. The median follow-up was 65 months. VEGF, uPA and PAI-1 were measured by commercially available enzyme-linked immunosorbent assays. Cox's univariate analysis showed that pT (p = 0.0007), uPA (p = 0.0156) and PAI-1 (p = 0.0015) had a significant impact on relapse-free survival, whereas VEGF did not have any prognostic value (p = 0.18). Bivariate analysis showed significant interactions between uPA and PAI-1 (p = 0.0035) and between VEGF and PAI-1 (p = 0.006). Our study confirms that uPA and PAI-1 cytosol levels can be considered as prognostic factors for relapse-free survival in node-negative breast cancer. Moreover, the interaction between VEGF and PAI-1 warrants further investigation into the relationship between the biomarkers of angiogenesis and those of the protease cascade.

Immunoassays of urokinase (uPA) and its type-1 inhibitor (PAI-1) in detergent extracts of breast cancer tissue

Two immunoassays for quantitation of the biological markers uPA and PAI-1 were evaluated for their use with detergent extracts of breast cancer tissue. Both assays were based on murine monoclonal capture antibodies and rabbit polyclonal detector antibodies. Horseradish peroxidase-conjugated goat anti-rabbit antibodies enabled measurement of the bound antigen. The detection limit of the uPA assay was 13 pg/ml, with a linear dose-response relationship up to 350 pg/ml. The assay detected free uPA as well as uPA in complex with PAI-1 and/or with its receptor. The detection limit of the PAI-1 assay was 50 pg/ml, with a linear dose–response relationship up to 1500 pg/ml. The assay detected both free PAI-1 and uPA:PAI-1 complex. Both assays were validated for detergent extracts using immunoabsorption and recovery tests. Highly significant associations between tumour tissue uPA and PAI-1 levels and prognosis were verified in a cohort of 164 lymph node-negative primary breast cancer patients. It is concluded that the two immunoassays are well-suited for the quantitation of uPA and PAI-1 in detergent extracts of breast cancer tissues.

Prognostic significance of insulin-like growth factor-binding protein expression in axillary lymph node-negative breast cancer.

Cellular proliferation, as measured by S-phase fraction, is an important predictor of breast cancer prognosis. The insulin-like growth factors (IGFs) have been shown to regulate proliferation in both normal and neoplastic cells by interacting with specific cell surface receptors. In addition to these receptors, high-affinity extracellular binding proteins also modulate IGF action. These insulin-like growth factor-binding proteins (IGFBPs) could influence breast cancer growth and, like other biological parameters of proliferation, could be related to prognosis. To test whether IGFBP expression was related to other biological parameters and disease-free survival, we measured IGFBP expression in 238 lymph node-negative primary breast cancer specimens. Proteins were extracted from breast cancer specimens and analyzed by semiquantitative IGF-I ligand blotting for IGFBP expression. IGFBP expression levels were compared to tumor size, age, S-phase fraction, DNA ploidy, and estrogen and progesterone receptor expression by Spearman correlation. Binding protein (BP)-2, BP-3, BP-4, and BP-5 were identified in breast cancer extracts. Estrogen receptor expression was positively correlated with BP-2 (Spearman correlation coefficient, rs = .262; P = .0001), BP-4 (rs = .313; P = .0001), and BP-5 (rs = .242; P = .0002). Similar correlations between progesterone receptor and BP-2, BP-4, and BP-5 were also found. BP-3 was inversely correlated with age (rs = -.251, P = .0001). BP-4 was weakly inversely correlated with tumor size (rs = -.141; P = .0295) and S-phase fraction (rs = -.216; P = .0025). Since tumor size and S-phase fraction are powerful predictors of prognosis in node-negative breast cancer, we examined the value of BP-4 as a predictor of disease-free survival. When stratified by tumor size, patients with large (> 2 cm) tumors that expressed low levels of BP-4 had improved survival when compared with patients with large tumors and high BP-4 levels (P = .001). IGFBPs can be detected in breast cancer specimens, and their level of expression correlates with other known biological parameters of breast cancer. Large tumors with low levels of BP-4 have relatively favorable prognoses. These data suggest that the IGFBPs may play a role in breast cancer biology and that BP-4 levels, analyzed in conjunction with tumor size, may have prognostic significance.

Prognostic value of proliferating cell nuclear antigen in lymph node—negative breast cancer patients

The prognostic value of proliferating cell nuclear antigen (PCNA) has been demonstrated in recent studies of human tumors including breast cancer. PCNA immunoreactivity was assessed retrospectively in a consecutive series of 173 lymph node-negative primary breast cancer cases. The PCNA grade was determined according to estimated quartiles of nuclear immunostaining, and its association to disease-free and overall survival was studied. PCNA grade was associated significantly with nuclear grade. On univariate analysis, PCNA grade was associated significantly with both 5-year relapse-free survival rate (Grades 1-2 = 78%; Grades 3-4 = 52%; P = 0.0117) and overall survival. On multivariate analysis, T category and PCNA were associated independently and significantly with both relapse-free and overall survival. Nevertheless, the magnitude of the association of PCNA grade to prognosis was low (relative risk of recurrence in patients with PCNA Grades 3-4 versus Grades 1-2 = 2.13), and only 24% of all relapses or 28.8% of all cancer deaths observed at 5 years occurred in patients with PCNA Grades 3-4. The predictive value of the relationship of PCNA grade to prognosis was too low to be of value as an independent prognostic indicator.