Development of Polyclonal and Monoclonal Antibodies to Iodinated Contrast Media

Two independent systems were developed to produce antibodies to diatrizoate. In the rabbit model polyclonal antibodies were produced that showed cross-reactivity to certain other contrast agents, analogues and serum proteins. In the mouse model monoclonal antibodies were produced that reacted only with diatrizoate, metrizamide, and ovalbumin. These studies demonstrated the potential antigenicity of contrast medium.

Biological and clinical implications of heat shock protein 27,000 (Hsp27): a review

Heat shock and other environmental and pathophysiologic stresses stimulate synthesis of heat shock proteins (Hsps). These proteins enable the cell to survive and recover from stressful conditions by as yet uncompletely understood mechanisms. Hsp27 is an important small Hsp (molecular weight, 27,000) found in human cells--both cancer cells and normal cells. This protein, besides its putative role in thermotolerance, is of special clinical interest because of recent data suggesting it may also play a role in drug resistance. In adults, Hsp27 is found particularly in several cell types such as breast, uterus, cervix, placenta, skin, and platelets. Although low-molecular-weight (small) Hsps have been found to be involved in embryogenesis of Xenopus and Drosophila, they have not been detected in human fetal organs. Regulation of expression of the Hsp gene (also known as HSPB1) has been considered a paradigm of gene regulation and is actively being studied in both prokaryotes and eukaryotes. In prokaryotes, the major Hsp genes are transcriptionally regulated by positively and negatively acting transcription factors. In eukaryotes, the genes encoding Hsps contain a regulatory DNA motif (inverted repeats of the pentameric sequence nGAAn) known as the heat shock element. Hsp27 may function as a molecular chaperone and in signal transduction pathways of different cell regulators, and Hsp27 and other Hsps may be active in development of resistance to stressful conditions and agents including cytotoxic drugs. Study findings indicate that some but not all estrogen-positive breast cancers express Hsp27, and overexpression of Hsp27 has been associated with both good and poor prognosis. In endometrial carcinomas, the presence of Hsp27 is correlated with the degree of tumor differentiation as well as with the presence of estrogen and progesterone receptors. Studies suggest, however, that detection of Hsp27 should not be considered to be a method for identifying hormone-responsive tumors or detecting estrogen receptors. Hsp27 seems to be a biochemical marker of estrogenic endometrial response. In patients with cervical cancer, Hsp27 is predominantly expressed in well-differentiated and moderately differentiated squamous cell carcinomas. In addition, expression of Hsp27 seems to be a negative prognostic factor for gastric cancer. Different isoforms of Hsp27 have been found in lymphoid tissue of patients with acute lymphoblastic leukemia, and the protein has also been associated with viral infections. These aspects are summarized and discussed in the present review.

DNA ploidy, S-phase, and steroid receptors in more than 127,000 breast cancer patients

Several potential prognostic factors are available today for patients with breast cancer, and many more are being identified and studied. To evaluate the clinical utility of these factors, it will be necessary to measure them on a large number of patients, and then follow these patients so that multivariate survival analyses can be performed. The Oncology Research Network was established in 1986 by the University of Texas Health Science Center at San Antonio and Nichols Institute Reference Laboratories in order to evaluate the clinical utility of new prognostic factors for patients with primary breast cancer. The first generation of prognostic factors included steroid receptors, along with DNA ploidy and S-phase fraction determined by flow cytometry. Currently, laboratory results have been obtained from more than 127,000 patients, and follow-up information is available on a subset of more than 25,000 of these patients. S-phase fraction was related to the ploidy status of the tumor. An increased incidence of aneuploidy and higher S-phase fractions were found in estrogen and progesterone receptor negative tumors, tumors from patients with positive axillary lymph nodes, tumors greater than 2 cm in diameter, and patients younger than 35 years of age. Preliminary survival analyses suggest that S-phase fraction and DNA ploidy, in combination with other prognostic factors, are powerful predictors of early disease relapse. The Oncology Research Network provides an important resource for examining the clinical significance of new laboratory assays and for expediting improvements in existing laboratory techniques.

Glutathione transferase GST pi in breast tumors evaluated by three techniques

The glutathione transferases are involved in intracellular detoxification reactions. One of these, GST pi, is elevated in some breast cancer cells, particularly cells selected for resistance to anticancer agents. We evaluated GST pi expression in 60 human breast tumors by three techniques, immunohistochemistry. Northern hybridization, and Western blot analysis. There was a significant positive correlation between the three methods, with complete concordance seen in 64% of the tumors. There was strong, inverse relationship between GST pi expression and steroid receptor status with all of the techniques utilized. In addition, there was a trend toward higher GST pi expression in poorly differentiated tumors, but no correlation was found between tumor GST pi content and DNA ploidy or %S-phase. GST pi expression was also detected in adjacent benign breast tissue as well as infiltrating lymphocytes; this expression may contribute to GST pi measurements using either Northern hybridization or Western blot analysis. These results suggest that immunohistochemistry is the method of choice for measuring GST pi in breast tumors.

Insulin-like growth factor-I receptors are overexpressed and predict a low risk in human breast cancer

IGF-I receptor (IGFR) content and its prognostic significance were evaluated in human breast cancer specimens using a sensitive and specific radioimmunoassay (V. Pezzino et al., Metabolism, 40: 861, 1991). The prognostic significance of IGFR expression was investigated by two different approaches: (a) detectable IGFR content was measured in 82% of specimens in a consecutive series of 184 human breast cancers and in 32% of 19 normal breast tissues. The average IGFR content in breast cancer was nearly 10-fold higher than the value observed in normal breast tissue (7.6 +/- 0.8 versus 0.8 +/- 0.1 ng/0.1 mg protein, mean +/- SEM; P < 0.001). IGFR content was positively correlated with estrogen (ER) and insulin receptor content (r = 0.269 and 0.515, respectively, Pearson correlation) but not with progesterone receptors (PR). No significant correlation was observed between IGFR content and a variety of tumor parameters (tumor size, lymph node involvement, grade) and host characteristics (age, body mass index, menopausal status); (b) IGFR content was measured in a noncontinuous series of 265 primary breast cancer specimens subdivided into 136 high-risk and 129 low-risk specimens on the basis of being either negative (ER-/PR-/aneuploid/high S-phase) or positive (ER+/PR+/diploid/low S-phase) for four well-established prognostic factors. IGFR levels were significantly higher in the low-risk group (6.4 +/- 0.4 ng/0.1 mg protein, mean +/- SEM) than in the high-risk group (3.6 +/- 0.5; P < 0.0001, Wilcoxon sum rank test). In summary, our data indicate that there is an elevated IGFR content in most human breast cancers compared with normal breast tissue and that an elevated IGFR content is a favorable prognostic indicator.

The 28-kDa protein whose phosphorylation is induced by protein kinase C activators in MCF-7 cells belongs to the family of low molecular mass heat shock proteins and is the estrogen-regulated 24-kDa protein

We have previously reported the presence of a 28-kDa protein in human mammary adenocarcinoma MCF-7 cells, whose phosphorylation by phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) and permeant diacylglycerol 1,2-dioctanoyl-sn-glycerol was correlated to growth arrest induced by the protein kinase C (PKC) activators. We now investigate the possible identity of this protein with the estrogen-regulated "24-kDa" protein shown as related to the mammalian heat shock protein 27 (Fuqua, S. A. W., Blum-Salingaros, M., and McGuire, W. L. (1989) Cancer Res 49, 4126-4129). 32P-Labeled 28-kDa protein from TPA-treated MCF-7 cells was immunoprecipitated with a 24-kDa-specific monoclonal antibody. Immunoblots from cell extracts fractionated by two-dimensional isoelectric focusing/SDS-polyacrylamide gel electrophoresis demonstrated that TPA induced the conversion of a 28-kDa isoform "a" (pI 6.7) to a more acidic isoform "b" (pI 6.2). Two-dimensional gel analysis of [3H]leucine-labeled MCF-7 cell extracts demonstrated that conversely to TPA, which induced only phosphorylation of 28-kDa protein, heat shock induced both synthesis (increase of isoform a) and phosphorylation (conversion of isoforms a to b) of the protein. 32P labeling of MCF-7 cells allowed demonstration of the presence of an extra phosphoisoform "c" (pI 5.9) upon TPA as well as heat shock treatment. When cells were pretreated with the bisindolylmaleimide GF109203X, a selective inhibitor of PKC, the heat shock-induced phosphorylation was unchanged, while the TPA effect was almost abolished, suggesting that the heat shock-activated protein kinase was very likely different from PKC. However, peptide mapping of the 28-kDa phosphoprotein suggested identical sites of phosphorylation upon TPA and heat shock stimulation. Partial amino acid sequencing of the 28-kDa protein revealed identity with both the 24-kDa protein and the mammalian HSP27. The fact that estrogens and PKC, respectively, regulate expression and phosphorylation of this 24/28-kDa protein strongly argues for its key role in MCF-7 cell proliferation and differentiation.

Expression of insulin-like growth factor binding proteins in human breast cancer correlates with estrogen receptor status

The insulin-like growth factors (IGFs) have been implicated in the growth regulation of human breast cancer. Since the IGFs are associated with specific binding proteins (IGFBPs) which may modulate receptor/ligand interactions, production of IGFBPs by breast cancer cells could alter their IGF-dependent growth. This study examined the expression of IGFBPs 4, 5, and 6 in eight breast cancer cell lines (BCCLs) using ribonuclease (RNase) protection assays. IGFBP-4 mRNA was detected in all BCCLs studied. IGFBP-5 expression was higher in estrogen receptor (ER) positive cells, while IGFBP-6 mRNA was detected in only two ER negative BCCLs. We also found that E2 treatment enhanced the expression of IGFBPs 2, 4, and 5 in T47-D cells. We next studied IGFBP mRNA expression in 40 primary breast tumors. All tumors expressed mRNA for IGFBPs 2-6 but none expressed IGFBP-1 message. IGFBP-3 expression was higher in ER negative tumors, while that of IGFBP-4 and -5 was higher in ER positive specimens. These differences were statistically significant (P < .05). Ligand blot analysis of tumor extracts confirmed the presence of IGFBPs in breast cancer tissues. Thus, differential IGFBP expression in ER positive and negative tumors suggests an important role for this protein in breast cancer biology.

Estrogen receptor, progesterone receptor, and HER-2/neu protein in breast cancers from pregnant patients

BACKGROUND

Because the occurrence of breast cancer during pregnancy is uncommon and because the high levels of estrogens and progestins associated with pregnancy could cause false-negative results from ligand binding assays (LBA), the actual incidence of steroid hormone receptor positivity in tumors from this subset of women is unclear.

METHODS

Estrogen receptor (ER) and progesterone receptor (PgR) were determined using LBA methods in 15 tumors from 15 pregnant patients with breast cancer. In addition, immunohistochemistry was done for ER, PgR, pS2, heat shock protein 27 (hsp27), and HER-2/neu on 12 of the 15 tumors.

RESULTS

Five of 15 (33%) tumors were positive for ER by LBA, compared with 52% of tumors from age-matched nonpregnant patients. Six of 12 (50%) were ER-positive by immunohistochemistry. For PgR, 7 of 15 (47%) tumors were positive by LBA, compared with 42% of tumors from nonpregnant patients. Ten of 12 (83%) stained positive for PgR. By LBA, 67% of tumors studied were positive for ER or PgR or both, as opposed to 57% of tumors from the nonpregnant comparison group. Two other estrogen receptor-mediated proteins, pS2 and hsp27, were present by staining in 8 of 12 (67%) and 10 of 12 (83%) of tumors, respectively. Seven of 12 tumors (58%) had positive staining for HER-2/neu, whereas only 16% of age-matched nonpregnant patients had positive-staining tumors.

CONCLUSION

By LBA, the incidence of ER and PgR in breast tumors from pregnant women was not significantly different from that of tumors from nonpregnant age-matched patients. Some ER-negative tumors were PgR, pS2, or hsp27 positive, indicating that an intact estrogen response system was operative although ER was not detectable by standard LBA.

Heat shock protein hsp70 in patients with axillary lymph node-negative breast cancer: prognostic implications

BACKGROUND

Cell synthesis of heat shock (stress-response) proteins is increased by a variety of environmental and pathophysiological stressful conditions. The 70-kd heat shock protein (hsp70) is thought to be involved in protein-protein interactions including those of the protein products of the human c-myc oncogene and the p53 (also known as TP53) tumor suppressor gene.

PURPOSE

The purpose of this study was to investigate whether elevated hsp70 expression may be an indicator of biological stress experienced by a breast cancer and may, therefore, predict disease outcome.

METHODS

Levels of hsp70 were determined by Western blot analysis in primary breast tumors from patients with negative axillary lymph nodes. We performed exploratory data analyses on a set of 162 primary breast cancers and constructed prognostic indexes of hsp70 expression levels. The optimal cutpoint for hsp70 expression was considered to be the value yielding the greatest separation for disease-free survival for the resulting two groups of patients. That cutpoint was then validated in a set of 345 tumors by univariate and multivariate analyses. Data were analyzed for overall survival, disease-free survival, tumor size, and patient age, as well as estrogen receptor and progesterone receptor status, ploidy (DNA content), and percentage of cells in S phase as determined by flow cytometry.

RESULTS

Expression of hsp70 emerged as a useful prognostic factor, both in univariate and in multivariate analyses. Patients whose tumors had high expression of hsp70 had significantly shorter disease-free survival (P = .006). The other statistically significant factors were S-phase fraction (P = .008) and tumor size (P = .01). For patients who received adjuvant therapy, hsp70 was the only independent predictor of disease recurrence (P = .05). For those with tumors 1-3 cm in diameter, hsp70 (P = .008) and S-phase fraction (P = .02) were statistically significant predictors of recurrence.

CONCLUSIONS

Measurement of hsp70 expression in primary tumors from patients with node-negative breast cancer may be useful in identifying patients at high risk for disease recurrence and thus may affect decisions regarding treatment after surgery.

IMPLICATIONS

Future studies should be performed to determine if detection of hsp70 by immunohistochemistry can be used to predict clinical outcome and to better understand the relationships between hsp70 and the effects of various treatment modalities.

How to integrate steroid hormone receptor, flow cytometric, and other prognostic information in regard to primary breast cancer

A large group of patients with node-positive breast cancer was divided into a training set (n = 851) and a validation set (n = 432) to demonstrate techniques for integrating steroid hormone receptor status, DNA flow cytometric findings, and other prognostic factors to predict patient survival. Multivariate analyses showed that estrogen receptor status, the number of involved axillary lymph nodes, patient age, S-phase fraction, progesterone receptor status, and tumor size were significant predictors of survival in patients with node-positive breast cancer. Techniques for optimizing and validating a cut point for a new prognostic factor and for examining alternative representations of prognostic factors were demonstrated. Prognostic indexes were created that could be used to identify patients with very good or very poor prognoses.

Prognostic indicators in early breast cancer

Approximately 115,000 new cases of axillary node negative breast cancer were diagnosed in this country last year. Since about 20-30% of these patients will ultimately relapse and die of their disease, adjuvant systemic therapy has been advocated for this group to decrease the relapse rate and prolong survival. However, although most clinical trials have demonstrated a modest impact on disease recurrence, the available data have failed to show consistent improvements in overall survival and does not justify the generalized use of systemic treatment in this patient subgroup. For this reason, a plethora of prognostic factors have been described to identify those patients with a higher risk of recurrence to concentrate therapeutic options in this specific group. Of all the disease prognosticators studied, tumor size, nuclear grade, and proliferative indexes appear to correlate well with tumor recurrence. In addition, biologic characteristics of primary tumors such as the presence of hormone and growth factor receptors, secretion of specific polypeptides and proteases, expression of proto-oncogenes, and abnormalities in tumor suppressor genes have been shown to be potentially useful as prognostic indicators in patients with early breast cancer. Despite these provocative data, larger clinical trials are necessary before incorporating these parameters in the routine evaluation of patients with axillary node negative breast cancer.

Association of p53 protein expression with tumor cell proliferation rate and clinical outcome in node-negative breast cancer

BACKGROUND

The p53 (also known as TP53) tumor suppressor gene encodes for a nuclear phosphoprotein thought to regulate proliferation of normal cells. Most p53 mutations result in a nonfunctional protein that accumulates in tumor cell nuclei. These common mutations appear to be involved in the development and/or progression of several neoplastic diseases including human breast cancer.

PURPOSE

Our purpose was to investigate the relationships between levels of mutant p53 protein expression, tumor cell proliferation rate, and clinical outcome in patients with node-negative breast cancer.

METHODS

Expression of mutant p53 protein was evaluated by frozen-section immunohistochemistry (IHC) and light microscopy in 700 breast cancers from axillary lymph node-negative patients with long-term follow-up (median, 54 months). The immunostaining signal was expressed as the sum of scores representing the proportion and staining intensity of negative and positive tumor cell nuclei (ranges, 0 and 2-8, respectively). Statistical comparisons were made between levels of p53 protein expression and disease-free survival, overall survival, and tumor proliferation rate expressed as the percentage of cells in the S phase (%S phase) as determined by flow cytometry.

RESULTS

Of the 700 tumors, 362 (52%) showed positive nuclear immunostaining (IHC score > 0). Proliferation rates were significantly higher (P = .0001) in positive tumors (median %S phase, 7.1%) than in negative tumors (4.1%). In a univariate cutpoint analysis, negative tumors (n = 388) versus low-positive tumors (IHC score = 2-6; n = 263) versus high-positive tumors (IHC score > 6; n = 99) showed progressively reduced disease-free survival (80% versus 72% versus 58% at 5 years, respectively; P < or = .05 for all pairwise comparisons). Analogous results for overall survival were 88% versus 84% versus 74%; only the result for negative versus high positive tumors was significant (P = .003). In a multivariate analysis, expression of p53 protein and high %S phase were independently associated with reduced disease-free survival (P = .008 and .01, respectively).

CONCLUSIONS

Expression of mutant p53 protein was associated with high tumor proliferation rate, early disease recurrence, and early death in node-negative breast cancer. Despite the strong direct correlation between accumulation of p53 protein and tumor proliferation rate, both factors were independently associated with poor prognosis, suggesting that p53 may have other biological functions in addition to cell-cycle regulation.

IMPLICATIONS

This test, when combined with other prognostic factors, may enhance our ability to identify node-negative breast cancer patients at high risk for early disease recurrence and/or death, for whom the use of adjuvant chemotherapy is unequivocally justified.

Estrogen receptor mutations in breast cancer

It is fairly well accepted that the presence of estrogen receptor (ER) identifies those breast cancer patients with a lower risk of relapse and better overall survival [Clark and McGuire, 1988], and the measurement of ER has become a standard assay in the clinical management of breast cancer. Receptor status also provides a guideline for those tumors which may be responsive to hormonal intervention [McGuire 1978; Osborne et al., 1980; Rose et al., 1985]. But only about half of ER-positive patients will respond to the various hormonal therapies available, and of those who do initially respond, most will eventually develop hormonally unresponsive disease following a period of treatment even though ER is often still present. Loss of ER from initially ER-positive tumors biopsied again at a later date has been estimated at only 19% [Gross et al., 1984]. Obviously the simple measurement of ER presence by ligand-binding assays does not provide us with an adequate estimate of the functional state of the receptor. In 1985 Sluyser and Mester hypothesized that the loss of hormone dependence of certain breast tumors may be due to the presence of mutated or truncated steroid receptors that activate transcription even in the absence of hormone [Sluyser and Mester, 1985]. Based on the recent identification of several ER sequence variants in human breast cancer cell lines and tumor specimens, we would now like to propose that some of these identified mutations play a role in receptor dysfunction in vivo, and will review those ER mutations which may prove to be important in breast cancer progression.

Prognostic factors and therapeutic decisions in axillary node-negative breast cancer

The incidence of node-negative breast cancer is increasing, and it is becoming a more commonly encountered clinical problem. Prognostic factors such as tumor size, histopathological characteristics, receptor status, proliferative markers, biochemical alterations, and gene abnormalities can be useful in assessing the risk of cancer recurrence after primary local therapy. By accurately assessing the risk of recurrence, informed decisions can be made about whether or not to treat with adjuvant systemic therapy.

Neural Network Analysis of DNA flow cytometry histograms

A pattern recognition system based on Neural Network Analysis, a form of artificial intelligence, was used to search DNA flow cytometry histograms for features that correlated with breast cancer patients' risk of relapse. DNA flow cytometry histograms and clinical follow-up information from 796 breast cancer patients were used to train a Neural Network to predict the clinical outcome of patients in a separate independent set of 794 patients. Median follow-up in this patient data base was short, 23 months. Neural Network Analysis resulted in a model that evaluated DNA flow cytometry histograms differently than conventional analysis, which categorizes the histograms by ploidy and S-phase fraction. Neural Network Analysis appeared to identify low risk and high risk subsets of patients as accurately as conventional analysis. Neural Network Analysis placed heavy emphasis on the region to the right of the diploid G2/M peak, where a subpopulation of nuclei with high DNA content is seen even in many histograms scored as diploid by conventional techniques. The number of nuclei in this region was found to be a powerful predictor of patient outcome, and multivariate analysis showed that the number of nuclei in this region and the S-phase fraction both were independently predictive of relapse. This pilot study suggests that conventional analysis (based on a mechanistic interpretation of regions in flow cytometry histograms) might be used in conjunction with and improved by pattern recognition systems or insights derived from them.

Measurement of steroid hormone receptors in breast cancer patients on tamoxifen

Estrogen (ER) and progesterone receptor (PgR) positive breast tumors often respond to tamoxifen, but ultimately progress as they become tamoxifen resistant. An accurate assessment of receptor status in specimens from tamoxifen-resistant patients could help to understand potential mechanisms of resistance and to predict response to second line hormonal therapies. However, since tamoxifen itself can affect ER and PgR determinations, assay results can be misleading. We measured ER and PgR by both ligand binding (LBA) and immunohistochemical (IHC) assays in 34 tumors from patients on tamoxifen, 30 of whom were displaying resistance to the drug. These tumors were classified into several receptor phenotypes. Eleven patients, 8 of whom were clearly progressing, expressed both receptors while on tamoxifen. ER was significantly less often negative when measured by IHC, suggesting that ER status by LBA was falsely negative in this group due to receptor occupancy by tamoxifen. Six patients had no detectable ER by LBA or IHC but still expressed PgR. The presence of PgR suggests that ER could still be functional, though undetectable, in these tumors, or that PgR is constitutively expressed by them. Finally, 12 patients were ER and PgR-negative by both assays, suggesting hormonal independence as the mechanism for resistance in this group. In a subset of patients with receptor assays both prior to tamoxifen and at the time of progression while taking the drug, we found that most ER-positive tumors converted to an apparent ER-negative status when assayed by LBA, while PgR status frequently remained unchanged. The continued expression of ER and/or PgR in many patients with tumor progression on tamoxifen indicates that mechanisms for resistance other than receptor loss are common in breast cancer.

The ER-positive/PgR-negative breast cancer phenotype is not associated with mutations within the DNA binding domain

We have used in vitro DNA binding assays as a measure of estrogen receptor (ER) function in human breast tumors. We found that the majority of ER+ (25 ER+/progesterone receptor [PgR]+, and 25 ER+/PgR-) tumors we examined were capable of binding consensus estrogen response element (ERE) oligonucleotides in this assay system. We found significant proteolytic activity in many of the tumors such that protease inhibitors were found to be essential during the preparation of tumor extracts. We next applied direct sequence analysis of the ER DNA binding domain of several of these tumors, and determined that the ER+/PgR- breast tumors did not contain mutations within the DNA binding domain which might explain their apparent discordant receptor phenotype. We did identify an alternatively spliced ER variant missing exon 3 of the DNA binding domain. This variant was unable to function as a transcriptional inducer of an estrogen-responsive reporter in a yeast assay system. Furthermore, the exon 3 ER deletion variant was expressed at equivalent levels in all of the ER+ breast tumors, so that it does not appear to be involved in the evolution of the ER+/PgR- breast cancer phenotype.

Prognostic significance of p53 gene alterations in node-negative breast cancer

Mutations in the p53 gene can play a role in the transformation of normal to malignant cells. Because these mutations are more frequently reported later in the course of transformation, their presence could reflect a greater malignant potential of the tumor and, thus, an increased probability of metastasis and recurrence after local therapy. In a pilot study using single-stranded conformation polymorphism analysis (SSCP), 200 node-negative breast tumors were examined for mutations in the region encompassing exons 5 through 9 of the p53 gene. Exons 5 through 9 were tested because they contain 80-90% of known p53 gene mutations. The tumors ranged in size from 1 to 3 cm. 28 tumors were found to have an abnormal band pattern on both initial and repeat analysis. 4 of these tumors were sequenced; 3 contained a p53 mutation and the 4th had a rare neutral polymorphism. Disease-free survival (DFS) at 5 years for women with tumors having an abnormal SSCP analysis was 57% (+/- 10%), compared to a 79% (+/- 3%) DFS for the group with a normal pattern. By the log rank test, this difference was highly significant, p < or = 0.01. The relative risk of recurrence for the group with an abnormal SSCP pattern was 2.2. In a multivariate analysis including ER, PgR, ploidy, S-phase, age, and tumor size, an abnormal p53 by SSCP analysis and patient age were the only factors that independently predicted DFS at 5 years.

CONCLUSION

Women with node-negative breast cancer who have tumors with alterations in the p53 gene, as indicated by SSCP analysis, have a significantly poorer prognosis and a higher rate of relapse at 5 years. The prognostic significance is maintained in a multivariate analysis including many established prognostic factors.

Expression of HER-2/neu oncoprotein in human breast cancer: a comparison of immunohistochemical and western blot techniques

Three hundred and one primary breast cancers from patients with tumor infiltrated lymph nodes were analyzed for the presence of HER-2/neu oncoprotein by two procedures: Western blot (WB) and immunohistochemistry (IHC). Overexpression of this protein was found by WB in 16.6% of the tumors, and by IHC in 16.3%. Concordance between the two methods was found in 95% of tumors (286/301). In 7 cases we found HER-2/neu by IHC but not by WB, while the opposite was found in the remaining 8 patients. This discrepancy was found mainly in samples with HER-2/neu values just above the cut points and were therefore close to the sensitivity limits of the procedures used here. This study helps to define the parameters that should be considered to evaluate the immunostaining for HER-2/neu as positive (i.e., membrane staining, IHC score of 2 or more). The results obtained by both techniques were correlated with several currently used prognostic factors. Higher HER-2/neu protein expression was found in tumors lacking estrogen or progesterone receptors, in tumors with high S-phase fraction and in patients with more than 3 positive lymph nodes. In contrast, no relationship was found between overexpression of this protein and tumor size, ploidy, or age of the patient. Patients with elevated HER-2/neu expression showed a significantly worse overall survival by both methods, IHC (p = 0.05) and WB (p = 0.001). In conclusion, there is very high agreement between IHC and WB when measuring expression of HER-2/neu and both techniques showed prognostic significance.

Prognosis and treatment decisions in patients with breast cancer without axillary node involvement

BACKGROUND

Every month, treatment decisions must be made for more than 6000 patients with breast cancer without axillary node involvement in the United States. Approximately 70% of these patients will survive more than 10 years after surgery and/or radiation treatment without additional systemic adjuvant therapy. If we had good methods to identify patients who are destined to have a recurrence of their disease, only those patients should receive adjuvant therapy.

METHODS

The authors reviewed the literature supporting the use of currently available prognostic factors for patients with node-negative breast cancer, and formulated a framework on which prognostic factor information can be based to help make these treatment decisions.

RESULTS

The steps involved in making treatment decisions are: use prognostic factors to determine the recurrence probability; calculate the expected treatment benefit; and weigh the expected benefits against the potential risks.

CONCLUSIONS

Prognostic factors can be used to help make treatment decisions for patients with breast cancer without axillary node involvement. However, the final treatment decision must take into account all aspects of the patient and her disease, and the physician must help the patient evaluate her prognostic factors, arrive at an understanding of her particular risk of recurrence, and weigh the potential benefits and risks of adjuvant therapy.

Overexpression of HER-2/neu and its relationship with other prognostic factors change during the progression of in situ to invasive breast cancer

Using permanent-section immunohistochemistry, we investigated the role of HER-2/neu in the development and progression of human breast cancer by measuring its overexpression in a series of hyperplastic (n = 30), dysplastic (n = 15), and malignant neoplastic (n = 708) lesions of ductal epithelium and by evaluating the relationships between overexpression and clinicopathologic features known to have prognostic significance in these lesions. The neoplasms included pure ductal carcinoma in situ (DCIS; n = 59) and infiltrating ductal carcinoma (IDC; n = 649). The latter were all node negative and stratified into IDC combined (n = 237) or not combined (n = 412) with a "significant amount" of DCIS (defined as DCIS greater than or equal to 10% of total tumor cellularity). Overexpression of HER-2/neu was not observed in any of the hyperplastic or dysplastic lesions. In contrast, it was present in 56% of pure DCIS and in 77% of the comedo subtype of this group. Only 15% of IDC overexpressed HER-2/neu. However, the rate of overexpression was significantly higher in the subset of IDC combined with DCIS compared with the subset of IDC not combined with DCIS (22% v 11%, respectively; P less than .0001). These results are consistent with the hypothesis that HER-2/neu plays a more important role in initiation than in progression of ductal carcinomas. They also suggest that overexpression decreases within individual tumors as they evolve from in situ to increasingly invasive lesions or, alternatively, that many invasive carcinomas arise de novo (ie, without progressing through a significant in situ stage) by mechanisms not involving HER-2/neu. In addition, overexpression of HER-2/neu was associated with several poor prognostic features (younger patient age, premenopause, negative estrogen receptor status, negative progesterone receptor status, and high nuclear grade) in the subset of IDC combined with DCIS. With one exception (negative estrogen receptor status) these associations were lost in IDC not combined with DCIS, also suggesting that the role of HER-2/neu changes during the progression of human breast cancer.

Why do so many prognostic factors fail to pan out?

Although there can be many reasons that one study fails to confirm the results of another, the consequences of data exploration and the potential for spuriously significant results are often overlooked. A series of simulation experiments were designed to mimic the characteristics of relapse-free survival data that might be encountered in a prognostic factor study of node-negative breast cancer patients. Each simulated dataset of 500 or 250 cases was divided into a training set, used to select the "best" prognostic factor cutpoint, and a validation set, used to confirm the cutpoint. Testing multiple cutpoints markedly increased the risk of making a Type I error. The power to detect even small true differences was substantial, and increased as the number of cutpoints increased. Regardless of the number of cutpoints tested on the training sets, the Type I error rate on an independent validation data set was quite stable and the power of the validation set to detect true differences was not related to the number of cutpoints. Validation power closely approximated that predicted for a simple two group comparison. It is therefore recommended that exploratory analyses of prognostic factors formally employ some method of adjusting for increased Type I errors, such as independent validation sets, ad hoc adjustment factors, or other statistical methods of estimating the true risk.

Prognosis and treatment decisions in patients with breast cancer without axillary node involvement

BACKGROUND

Every month, treatment decisions must be made for more than 6000 patients with breast cancer without axillary node involvement in the United States. Approximately 70% of these patients will survive more than 10 years after surgery and/or radiation treatment without additional systemic adjuvant therapy. If we had good methods to identify patients who are destined to have a recurrence of their disease, only those patients should receive adjuvant therapy.

METHODS

The authors reviewed the literature supporting the use of currently available prognostic factors for patients with node-negative breast cancer, and formulated a framework on which prognostic factor information can be based to help make these treatment decisions.

RESULTS

The steps involved in making treatment decisions are: use prognostic factors to determine the recurrence probability; calculate the expected treatment benefit; and weigh the expected benefits against the potential risks.

CONCLUSIONS

Prognostic factors can be used to help make treatment decisions for patients with breast cancer without axillary node involvement. However, the final treatment decision must take into account all aspects of the patient and her disease, and the physician must help the patient evaluate her prognostic factors, arrive at an understanding of her particular risk of recurrence, and weigh the potential benefits and risks of adjuvant therapy.

Regulation of insulin-like growth factor-binding protein (IGFBP) expression by breast cancer cells: use of IGFBP-1 as an inhibitor of insulin-like growth factor action

BACKGROUND

The insulin-like growth factors (IGFs) play an important role in normal growth and development. Evidence suggests they may also regulate the growth of several cancer cell types. This regulation is mediated by interactions between the receptors and ligands. There is now ample evidence to suggest that these interactions are also influenced by extracellular IGF-binding proteins (IGFBPs). Six different IGFBPs have been cloned. Some species may act to inhibit the mitogenic effects of the IGFs. Since breast cancer cells are responsive to the IGFs, it is possible that regulated expression of the IGFBPs affects tumor growth. Furthermore, inhibitory binding proteins could be used as neutralizers of IGF action.

PURPOSE

We conducted this study to fully characterize the expression and hormonal regulation of IGF-binding protein expression in human MCF-7 breast cancer cells and to test the ability of purified IGFBP-1 to inhibit IGF-I action.

METHODS

We used ribonuclease protection assays and Western ligand blotting to examine IGFBP expression in MCF-7 cells. The effect of IGF-I, IGFBP-1, and 17 beta-estradiol on serum-free cell growth was also studied.

RESULTS

MCF-7 cells expressed IGFBP-2, IGFBP-4, and IGFBP-5 RNA and protein. These cells are dependent on estrogen for growth. In short-term culture, IGF-I can substitute for estrogen. Concomitant addition of IGF-I and estrogen enhanced stimulation above the level achieved by either factor alone. Estrogen also increased IGFBP production, making it unlikely that the IGFBPs induced by estrogen in MCF-7 cells could function as major inhibitors of IGF action. In contrast, exogenous addition of IGFBP-1 could block IGF-I-induced mitogenesis; this effect was reversible by excess IGF-I.

CONCLUSIONS

The studies suggest that cancer cell growth may be regulated by endogenous IGFBP expression. Furthermore, the exogenous addition of the IGFBP-1 blocked IGF-I action and potentially could be used as a pharmacologic inhibitor of IGF action.

Abnormal estrogen receptor in clinical breast cancer

We have discovered a number of estrogen receptor variants in clinical breast cancer tissues. We have base-pair insertions, transitions, and deletions of exons 3, 5 and 7. Using a transactivation assay we have discovered receptors with outlaw function consisting of both dominant-positive receptors which are transcriptionally active in the absence of estrogen, and dominant-negative receptors which are transcriptionally inactive themselves but prevent normal estrogen receptor function.

Prognostic significance of progesterone receptor levels in estrogen receptor-positive patients with metastatic breast cancer treated with tamoxifen: results of a prospective Southwest Oncology Group study

PURPOSE

Southwest Oncology Group (SWOG) protocol 8228 is a prospective trial designed to investigate the prognostic significance of progesterone receptor (PgR) levels in estrogen receptor (ER)-positive breast cancer patients who were treated with tamoxifen. This study was undertaken because the value of PgR measurements in advanced breast cancer had been assessed previously only in studies that were small, retrospective, or included heterogeneously treated patients.

METHODS

Receptor assays were performed only in the laboratories that met strict quality control guidelines. Of the 398 patients entered, 342 patients were eligible and assessable for the study end points of objective clinical response, time to treatment failure, and overall survival.

RESULTS

Multivariate analysis shows that elevated PgR levels significantly and independently correlated with increased probability of response to tamoxifen, longer time to treatment failure, and longer overall survival. Overall response rate (defined as complete response [CR], partial response [PR], or stable disease [SD] for greater than 6 months) in this trial was 54%. Response rates to tamoxifen were 43%, 53%, and 61% in subsets of patients with less than 10, 10 to 99, and more than 100 fmol/mg PgR, respectively. Exploratory subset analysis using PgR and other prognostic variables identified ER-positive patient subsets with response rates to tamoxifen ranging from 24% (premenopausal patients) to 86% (postmenopausal patients with ER greater than 38 and PgR greater than 329 fmol/mg). No groups of ER-positive patients were identified who had such a low response rate as to absolutely preclude considering the use of tamoxifen. Multivariate analysis showed the independent, statistically significant predictors were: for response to tamoxifen, menopausal status, PgR, and ER; for time to treatment failure, menopausal status, disease-free interval (DFI), PgR, and ER; and for overall survival DFI, PgR, ER, site of disease, and history of adjuvant therapy.

CONCLUSION

We conclude that knowledge of PgR levels together with other clinical information can improve the pretreatment assessment of ER-positive breast cancer patients with metastatic disease.

Uptake of the noncytotoxic transport probe procainamide in the Chinese hamster ovary model of multidrug resistance

Many of the cytotoxic substrates of the multidrug transporter are organic cations. Cimetidine, procainamide, and tetraethylammonium bromide were used in a Chinese hamster ovary model of multidrug resistance, to study handling of noncytotoxic cationic transport probes. Cimetidine and procainamide, but not tetraethylammonium, accumulated to a greater extent (5-fold) in the sensitive CHOAUXB1 (AB) cell line than in the resistant CHRC5 (C5) cell line. Accumulation of both cimetidine and procainamide was significantly increased by verapamil in C5 but not AB. Procainamide accumulation in both AB and C5 was temperature dependent and occurred by passive diffusion. Diltiazem, nifedipine, rifampin, tamoxifen, rhodamine, and ethidium also increased procainamide accumulation in C5 but not AB. Azide in glucose-free medium increased procainamide accumulation in C5, and this was reversed when glucose, but not 3-O-methylglucose, was added. Procainamide efflux rates were similar in AB and C5 and not affected by verapamil or azide. The initial rate of procainamide uptake was higher in AB than in C5, and both verapamil and azide increased the initial rate of procainamide uptake in C5. Thus, differences in accumulation of the noncytotoxic transport probe procainamide in the colchicine-sensitive and colchicine-resistant components of the Chinese hamster ovary cell line mimic the accumulation of known cytotoxic substrates for the multidrug transporter, such as colchicine, vinblastine, and doxorubicin. The differential accumulation of procainamide is due to differences in rates of drug influx, rather than efflux. Since procainamide influx is passive and decreased accumulation in the resistant line appears to parallel M(r) 170,000 glycoprotein presence and activity, we would speculate that decreased procainamide accumulation may be due to an indirect effect of the M(r) 170,000 glycoprotein, such as its effect on intracellular pH.

Response of human breast cancer cells to heat shock and chemotherapeutic drugs

Previous studies have shown that certain chemotherapeutic drugs are less effective on tumor cells when cells have been previously exposed to hyperthermia. In the present study, we have evaluated whether specific modifications in heat shock protein (hsp) expression are associated with resistance to anticancer drugs. RNA levels for hsp90, hsp70, and hsp27 were studied by Northern and slot blots, while proteins were studied by two-dimensional gel electrophoresis, in MCF-7/BK and MDA-MB-231 breast cancer cells. The sensitivities of these cells to doxorubicin, colchicine, 5-fluorouracil, cisplatin, actinomycin D, and methotrexate were tested by clonogenic assays. These techniques were applied to both cell lines before (control) and after heat shock. The study revealed that elevated hsp70 and hsp27 levels were associated with doxorubicin resistance. In addition, the presence of phosphorylated hsp27 isoforms was also associated with doxorubicin resistance. The study showed that elevated hsps were not associated with multidrug resistance. Heat shock did not induce P170 glycoprotein mRNA overexpression or resistance to the other drugs tested. We also found that the level of doxorubicin protection conferred by the overexpression of hsp was lower than that obtained in cells expressing a multidrug resistance phenotype (MDA-A1R cells). In these cells, heat shock did not confer additional doxorubicin resistance and hsp27 phosphorylation was deficient. Our studies suggest that specific hsps are associated with doxorubicin resistance in certain human breast cancer cells and that this mechanism seems to be independent of the multidrug resistance system.

Identification of estrogenic tamoxifen metabolite(s) in tamoxifen-resistant human breast tumors

PURPOSE

We have shown previously that acquired tamoxifen resistance in an in vivo experimental model is associated with reduced tamoxifen accumulation, isomerization of trans-4-hydroxytamoxifen, and tamoxifen-stimulated tumor growth. The purpose of this study is to isolate and verify the presence of estrogenic tamoxifen metabolites in human breast tumors using high-performance liquid chromatography (HPLC) and mass-spectrometry (MS) techniques.

PATIENTS AND METHODS

In the present study, we used HPLC and MS to identify the presence of estrogenic metabolites in tumor samples excised from athymic nude mice and in human breast tumors isolated from patients receiving adjuvant tamoxifen therapy.

RESULTS

We identified the presence of metabolite E, a known estrogenic metabolite of tamoxifen, in tamoxifen-resistant MCF-7 human breast tumors implanted in athymic nude mice, as well as in tumors from patients with clinical resistance. Additionally, we separated another estrogenic metabolite, bisphenol, by HPLC, and this was also tentatively confirmed by MS analysis.

CONCLUSION

These data suggest that cellular tamoxifen metabolism to estrogenic metabolites may in part contribute to stimulating the growth of hormone-responsive breast tumors following prolonged exposure to tamoxifen. Further evaluation of the relationship between cellular metabolism and acquired tamoxifen resistance is warranted.

Progesterone antagonists: tumor-inhibiting potential and mechanism of action

A new approach for the treatment of breast cancer could be the use of progesterone antagonists. These compounds were originally developed for the inhibition of progesterone-dependent processes and have been shown to be effective in inhibition of nidation and interruption of pregnancy. Although the roles of progesterone and the progesterone receptor in control of cell growth remain unclear, it was found in progesterone receptor positive mammary carcinoma cell lines that the antiprogestin, Mifepristone, had an inhibitory effect on cell growth and a growth-inhibiting action on the DMBA-induced mammary carcinoma of the rat. We have shown that the progesterone antagonists, Onapristone and ZK 112993, which possess a reduced antiglucocorticoid activity compared to Mifepristone, exert a strong tumor-inhibiting effect in a panel of hormone-dependent mammary tumor models. The effects of these compounds were in some systems superior to those of tamoxifen or high dose progestins and comparable to ovariectomy. Although prerequisites for their antiproliferative potency are an affinity to the progesterone receptor as well as a sufficient number of available receptors in the tumors, the strong tumor inhibiting potential of the antiprogestins cannot be explained by a classical anti-hormonal mechanism. Surprisingly, the antitumor activity is evident in spite of elevated serum levels of ovarian and pituitary hormones. It was established by morphometric procedures that treatment with Onapristone triggers differentiation of the mitotically active polygonal tumor epithelial cell towards secretory active glandular structures and acini. All our quantitative light and electron microscopic data indicate that the antitumor action of antiprogestins is accompanied by the initiation of terminal differentiation leading to (apoptotic) cell death. Finally, our flow cytometry studies revealed an accumulation of the tumor cells in the G0G1 phase of the cell cycle, which may result from induction of differentiation since a differentiation-specific G1 arrest has already been proposed for other stem cell systems. It can be concluded from these data that the progesterone receptor antagonists differ in their mode of action from compounds used in established endocrine treatment strategies for mammary carcinoma. The ability of progesterone antagonists like Onapristone to reduce the number of cells in S-phase may offer a significant clinical advantage, since it is established that the S-phase fraction is a highly significant predictor of disease-free survival among axillary node-negative patients with diploid mammary tumors.

Tamoxifen and the isomers of 4-hydroxytamoxifen in tamoxifen-resistant tumors from breast cancer patients

PURPOSE

The antiestrogen tamoxifen is effective in therapy for breast cancer. However, its use is limited by the eventual development of acquired tamoxifen resistance in many patients. The mechanisms responsible for tamoxifen resistance remain unknown; loss of estrogen receptor (ER), selection of hormone-independent breast cancer clones, or alterations in serum tamoxifen levels after long-term use do not explain acquired resistance in most patients. Using an experimental model in which human breast cancer cells develop resistance in athymic mice treated with tamoxifen, we have recently shown that acquired resistance is associated with markedly reduced cellular concentrations of tamoxifen and by isomerization of the trans-4-hydroxy metabolite to the less potent cis isomer.

MATERIALS AND METHODS

Using a sensitive high-performance liquid chromatography (HPLC) assay, we have now measured levels of tamoxifen and its major metabolites in a series of 14 tumors from patients treated with tamoxifen. The duration of therapy ranged from 1 month to 6 years.

RESULTS

Tumor tamoxifen levels varied over a wide range. Low concentrations were observed in tumors from eight patients, all demonstrating progressive disease at the time of biopsy after a minimum duration of treatment of 6 months. Six tumors had moderate to high tamoxifen levels, two from patients responding to tamoxifen, one from a patient with stable disease, and three from patients with disease progression. Both the cis and trans isomers of the potent antiestrogenic metabolite 4-hydroxy-tamoxifen were detected in 11 tumors. Six tumors had high ratios of the cis to trans isomer (1.10:2.06), all from patients not responding to tamoxifen. The five tumors with low cis:trans ratios included the two tumors from responding patients and three from patients with progression. All but one of the 11 nonresponding patients had either a low tumor tamoxifen level, a high cis:trans ratio, or both.

CONCLUSION

This study clearly demonstrates a wide range of tumor tamoxifen levels and accumulation of the less antiestrogenic cis isomer of 4-hydroxytamoxifen in some patients on tamoxifen therapy. Additional study is necessary to determine if these metabolic profiles are related to the development of tamoxifen resistance.

A demonstration that breast cancer recurrence can be predicted by neural network analysis

Neural Network Analysis, a form of artificial intelligence, was successfully used to predict the clinical outcome of node-positive breast cancer patients. A Neural Network was trained to predict clinical outcome using prognostic information from 1008 patients. During training, the network received as input information tumor hormone receptor status, DNA index and S-phase determination by flow cytometry, tumor size, number of axillary lymph nodes involved with tumor, and age of the patient, as well as length of clinical followup, relapse status, and time of relapse. The ability of the trained Network to determine relapse probability was then validated in a separate set of 960 patients. The Neural Network was as powerful as Cox Regression Modeling in identifying breast cancer patients at high and low risk for relapse.

Inhibition of estrogen receptor action by a naturally occurring variant in human breast tumors

It is fairly well accepted that the presence of estrogen receptor (ER) and progesterone receptor (PgR) identifies breast cancer patients with a lower risk of relapse and better overall survival. But patients with discordant receptors, the ER+/PgR- phenotype, are often intermediate in clinical response. We focused upon this group of patients and have identified a truncated ER which is abundant in some ER+/PgR- breast tumors and which inhibits the binding of wild-type ER to its cognate response element. This variant interferes in a dominant negative manner with wild-type ER function and may represent a mechanism for modulation of estrogen responsiveness.

Estrogen receptor variants in clinical breast cancer

We have used the screening techniques of chemical mismatch cleavage, single stranded conformational polymorphism, and gel retardation to discover a number of estrogen receptor RNA variants in clinical breast cancer tissues. We have found basepair insertions, transitions, and deletions as well as alternative splicing, yielding deletions of exon 3, 5, or 7. Using a yeast transactivation assay we have discovered receptors with outlaw function, including both a dominant-positive receptor, which is transcriptionally active in the absence of estrogen, and a dominant-negative receptor, which is itself transcriptionally inactive, but prevents the action of normal estrogen receptor. These variants could have clinical significance, helping to explain breast tumor behavior and patient outcome.

Progesterone receptor gene restriction fragment length polymorphisms in human breast tumors

We examined the progesterone receptor (PgR) gene in tissue from both primary human breast tumors and normal placentas, detecting restriction fragment length polymorphisms (RFLPs) with the restriction endonucleases Pst I/Sst I and HindIII. There was a general agreement of the Pst I and Sst I polymorphisms in any individual tumor, suggesting that they define two alleles in the human PgR locus, one being characterized by a deletion of about 300 base pairs with respect to the other. Both primary human breast tumor specimens (n = 36) and human term placentas (n = 48) displayed similar allele frequencies and typical mendelian distribution of these Pst I/Sst I alleles. The previously reported HindIII PgR RFLP was also investigated in 132 breast tumors. The HindIII PgR gene RFLP did not display typical mendelian distribution in the breast tumors; the factors affecting the HindIII allele frequencies are presently unknown. Neither the HindIII RFLP nor the deletion defined by Pst I and Sst I correlated with PgR expression as determined by a ligand-binding assay, suggesting that neither is related to the heterogeneity of PgR expression seen in breast tumors.

Hormones and breast cancer

The study of steroid hormone receptors in human breast cancer provides insight into tumor biology and forms the rationale for many therapeutic modalities. Clinical assays for estrogen and progesterone receptors improve the selection of patients who may benefit from hormonal therapy. In addition, receptor content in breast cancer tissue is useful for determining prognosis. Together with other new prognostic factors, hormone receptors help to distinguish among patients with primary breast cancer who are at risk for early recurrence.

Follow-up study of HER-2/neu amplification in primary breast cancer

Amplification of the HER-2/neu oncogene was determined in 362 tumors from patients with primary breast cancer (185 node-positive patients and 177 node-negative patients). The overall amplification rate was 33% (30% for node-negative patients; 31% for patients with 1-3 positive nodes; 40% for patients with greater than 3 positive nodes). Gene copy number was not associated with axillary lymph node status, steroid receptor status, or patient age but was weakly correlated with the size of the primary tumor. Amplification of the HER-2/neu gene did not correlate with either disease-free or overall survival in univariate or multivariate analyses. The results were unambiguously negative for patients with node-negative disease. Although the univariate results for node-positive patients were marginally significant (P = 0.07), the significance was not retained in multivariate analyses. Thus, while HER-2/neu amplification may be biologically important in primary breast cancer, it will only be of marginal utility as a prognostic factor for predicting clinical outcome.

Variant human breast tumor estrogen receptor with constitutive transcriptional activity

Since progesterone receptor (PgR) is normally induced by estrogen, breast cancer lacking estrogen receptor (ER) would also be expected to lack PgR. However, a small percentage of breast cancers are ER- yet PgR+. These tumors might possess an ER which is defective in estrogen binding but is still functional in stimulating estrogen-responsive genes such as PgR. We have now detected such a variant, lacking exon 5 of the hormone-binding domain, using complementary DNA amplified by the polymerase chain reaction. This variant was the predominate ER RNA expressed in three ER-/PgR+ tumors. Furthermore, the variant ER constitutively activates transcription of a normally estrogen-dependent gene construct in yeast cells. The variant ER could explain the expression of PgR in certain tumors and have therapeutic implications.