RBC alloantibody prevalence and specificity in a Western Norwegian tertiary hospital.

BACKGROUND AND OBJECTIVE: Although several studies focus on red blood cell (RBC) alloantibody distribution in selected patient populations, few address the specificity and frequency in all relevant groups. This study reports alloantibody frequency, distribution and the relationship to age and gender in blood donors, pregnant women and potential recipients of blood products. METHODS: This historical cohort study included 55 462 consecutive antibody screening tests from a tertiary Western Norwegian Hospital. Descriptive statistics were performed, and the results were compared with the literature. RESULTS: The detection and immunisation frequency for the whole cohort were 0·39 and 0·51%, respectively, whereas the RBC alloantibody prevalence was 0·73%. The most frequent RBC alloantibodies were anti-E (20·1%), anti-M (18·7%), anti-K (9·8%), anti-D (8·9%) and anti-Fy(a) (7·0%). In pregnant women, the most frequent RBC alloantibodies were anti-M, anti-D and anti-Le(a) (20·8, 18·9 and 18·9%, respectively), whereas there was no anti-K detected. Anti-E and anti-M were the dominating RBC alloantibodies in the pre-transfusion testing of in-hospital patients (24·1 and 17·1%, respectively). Eighteen (9·2%) persons in the total cohort had two RBC alloantibodies, six persons had three alloantibodies, and two persons had four alloantibodies. Rh and K typing to prevent future immunisations was only performed in 21·0% of the individuals who presented with a new alloantibody; despite that, 50·5% of the detected alloantibodies had such specificities. CONCLUSIONS: The immunisation frequency and the level of anti-K are low compared to national and international studies. Rh and K phenotype-matched blood transfusions might be a feasible future strategy to further decrease RBC alloantibodies.

Flotillins as regulators of ErbB2 levels in breast cancer.

Amplification and overexpression of the receptor tyrosine kinase ErbB2 occur in up to 30% of human breast cancers, and high ErbB2 levels are correlated with poor prognosis for breast cancer patients. In contrast to the epithelial growth factor receptor (ErbB1), ErbB2 is not downregulated by ligand-induced mechanisms. Here we show that flotillins are involved in the stabilization of ErbB2 at the plasma membrane. In SKBR3 breast cancer cells and breast cancer tissue, a positive correlation between flotillin and ErbB2 expression levels could be demonstrated. Moreover, the tissue microarray analyses of biopsies from 194 patients diagnosed with carcinomas of the breast showed that flotillin-2 emerged as a potential predictor of prognosis in breast cancer. Depletion of flotillin-1 and flotillin-2 leads to internalization and degradation of ErbB2. Furthermore, flotillin-1 and -2 were found to be in a molecular complex with ErbB2 and Hsp90. The depletion of one of these proteins results in disruption of this complex, followed by destabilization of ErbB2 at the membrane, and its internalization and degradation. As a consequence, ErbB2-triggered downstream signalling is inhibited. Our data demonstrate a novel mechanism for interfering with ErbB2 signalling, which potentially can have clinical impact.

Long-term follow-up of the SBG 9401 study comparing tailored FEC-based therapy versus marrow-supported high-dose therapy.

BACKGROUND: The purpose was to investigate adjuvant marrow-supportive high-dose chemotherapy compared with an equitoxicity-tailored comparator arm. PATIENTS AND METHODS: Five hundred and twenty-five women below the age of 60 years with operated high-risk primary breast cancer were randomised to nine cycles of granulocyte colony-stimulating factor supported and individually tailored FEC (5-fluorouracil, epirubicin, cyclophosphamide), (n = 251) or standard FEC followed by marrow-supported high-dose therapy with CTCb (cyclophosphamide, thiotepa, carboplatin) therapy (n = 274), followed by locoregional radiotherapy and tamoxifen for 5 years. RESULTS: There were 104 breast cancer relapses in the tailored FEC group versus 139 in the CTCb group (double triangular method by Whitehead, P = 0.046), with a median follow-up of all included patients of 60.8 months. The event-free survival demonstrated 121 and 150 events in the tailored FEC- and CTCb group, respectively [P = 0.074, hazard ratio (HR) 0.804, 95% confidence interval (CI) 0.633-1.022]. Ten patients in the tailored FEC regimen developed acute myeloid leukaemia (AML)/myelodysplasia (MDS). One hundred deaths occurred in the tailored FEC group and 121 in the CTCb group (P = 0.287, HR 0.866, 95% CI 0.665-1.129). CONCLUSION: The update of this study shows an improved outcome linked to the tailored FEC treatment in relation to breast cancer relapse, but also an increased incidence of AML/MDS.

Adjuvant endocrine treatment (goserelin vs tamoxifen) in pre-menopausal patients with operable node positive stage II breast cancer. A prospective randomized national multicenter study.

AIMS: To evaluate possible differences in effect on time to recurrence and overall survival in node positive pre-menopausal breast cancer patients (age < or = 50 years) receiving LHRH analogue or tamoxifen as adjuvant endocrine treatment. METHODS: Between January 1989 and July 1994, 320 patients with node positive (pN(+)) and hormone receptor positive or receptor status unknown tumors were included and randomized in a national multicenter study to receive either tamoxifen or goserelin as adjuvant treatment for two years. Primary surgical treatment was employed according to current standards. Final follow-up was completed as of December 2000. Time to events were estimated by the Kaplan-Meier method, and compared by the log rank test. Relative risks were estimated by the Cox's proportional hazards model. RESULTS: No differences in time to first recurrence or overall survival were observed between treatment groups. Proportions of patients in each group having a second breast cancer were also similar. CONCLUSIONS: Standard adjuvant treatment with tamoxifen as compared to adjuvant LHRH analogue therapy employed in this group of breast cancer patients gave similar outcomes, but the number of patients was too small to formally exclude a potentially clinically relevant difference in survival.

Fulvestrant, formerly ICI 182,780, is as effective as anastrozole in postmenopausal women with advanced breast cancer progressing after prior endocrine treatment.

PURPOSE: To compare the efficacy and tolerability of fulvestrant (formerly ICI 182,780) and anastrozole in postmenopausal women with advanced breast cancer progressing after prior endocrine treatment. PATIENTS AND METHODS: Patients (n = 451) with advanced breast cancer were randomized to receive fulvestrant 250 mg as a once-monthly (one x 5 mL) intramuscular injection or an oral dose of anastrozole 1 mg in this open, parallel-group, multicenter trial. The primary end point was time to progression (TTP). Secondary end points included objective response (OR) rates, defined as complete response (CR) or partial response (PR), duration of response (DOR), and tolerability. RESULTS: Patients were followed for a median period of 14.4 months. In terms of TTP, fulvestrant was as effective as anastrozole (hazard ratio, 0.98; confidence interval [CI], 0.80 to 1.21; P =.84). Median TTP was 5.5 months for fulvestrant and 5.1 months for anastrozole. OR rates showed a numerical advantage for fulvestrant (20.7%) over anastrozole (15.7%) (odds ratio, 1.38; CI, 0.84 to 2.29; P =.20). Clinical benefit rates (CR + PR + stable disease > or = 24 weeks) were 44.6% for fulvestrant and 45.0% for anastrozole. Median DOR was 14.3 months for fulvestrant and 14.0 months for anastrozole. Both treatments were well tolerated, with 3.2% and 1.3% of fulvestrant- and anastrozole-treated patients, respectively, withdrawn from treatment because of an adverse event. CONCLUSION: Fulvestrant was as effective as anastrozole. These data confirm that fulvestrant is an additional, effective, and well-tolerated treatment for advanced breast cancer in postmenopausal women whose disease progressed on prior endocrine therapy.

Effects of dose-intensive chemotherapy and radiotherapy on serum n-terminal proatrial natriuretic peptide in high-risk breast cancer patients.

By using N-terminal proatrial natriuretic peptide (proANP) in serum as a marker of cardiac function, we compared the cardiac side effects of two intensive adjuvant treatment regimens for breast cancer. Patients received either 9 cycles of FEC (5-fluorouracil, epirubicin and cyclophosphamide) where the doses of epirubicin and cyclophosphamide were escalated according to the leucocyte nadir (n = 49, FEC-group) or three cycles of FEC followed by high-dose chemotherapy with alkylating agents (n = 56, CTCb-group) given with the support of peripheral blood stem cells support. Both groups received adjuvant radiotherapy. Serial measurements of proANP were performed up to three years after treatment. Mean proANP values in the FEC-group was on average 19% higher than in the CTCb-group (p = 0.002). The proANP levels showed a significant association with the cumulative dose of epirubicin (p < 0.001) but not with cyclophosphamide (p = 0.151) and 5-FU (p = 0.160). The pharmacokinetics of epirubicin was studied at the first and third chemotherapy course. The proANP levels after treatment were significantly related to the AUC (p = 0.034) and Cmax(p = 0.037) of epirubicin. Left-sided chest irradiation was associated with on average 12% higher proANP values than right-sided (p = 0.031). We conclude that dose-escalated FEC causes a stronger increase in proANP than 3 FEC followed by high-dose CTCb-treatment. Increase of proANP levels might represent an early sign of cardiotoxicity secondary to chemotherapy and radiation treatment. Long-time follow-up is necessary to determine the clinical significance of these findings.

Genetic susceptibility and environmental estrogen-like compounds.

Environmental chemicals with estrogenic activities have been suggested to be able to interact with the endocrine system. Endogenous estrogen is synthesized in the ovarian theca cells of premenopausal women or in the stromal adipose cells of the breast of postmenopausal women and minor quantities in peripheral tissue. These cells, as well as breast tissue, express all the necessary enzymes for this synthesis, CYP17, CYP11a, CYP19, 17-beta-hydroxysteroid hydrogenase, steroid sulfatase as well as enzymes further hydroxylating estradiol, such as CYP1A1, CYP3A4, CYP1B1, catechol-o-methyltransferase (COMT). Polymorphisms in these enzymes may have a possible role in the link between environmental estrogens and hormone-like substances and the interindividual risk of breast cancer.

Quantitative analysis of chromosomal CGH in human breast tumors associates copy number abnormalities with p53 status and patient survival.

We present a general method for rigorously identifying correlations between variations in large-scale molecular profiles and outcomes and apply it to chromosomal comparative genomic hybridization data from a set of 52 breast tumors. We identify two loci where copy number abnormalities are correlated with poor survival outcome (gain at 8q24 and loss at 9q13). We also identify a relationship between abnormalities at two loci and the mutational status of p53. Gain at 8q24 and loss at 5q15-5q21 are linked with mutant p53. The 9q and 5q losses suggest the possibility of gene products involved in breast cancer progression. The analytical techniques are general and also are applicable to the analysis of array-based expression data.

Elevated CA125 in breast cancer--A sign of advanced disease.

The serum tumor markers CA 125, MUC1 and CEA were measured in 221 breast cancer patients over a period of 2 years. Patients examined on at least three occasions were included in the study. Thirty-three patients had increasing or continuously high concentrations of CA 125. Thirty (91%) of these had involvement of the pleura, either as pleural metastasis or metastasis in surrounding tissue i.e. bone structures in the thorax cavity or lung parenchyma. MUC1 and CEA were elevated in 27 (82%) and 24 (73%) of the 33 patients, respectively. Increased concentrations of these two markers did not relate to the site of metastasis. However, the three tumor markers complemented each other in detecting early metastases. Increased CA 125 was associated with metastasis in or near the pleura, and in stage IV breast cancer it was related to poor prognosis.

The potential for aromatase inhibition in breast cancer prevention.

Although SERMs are currently being evaluated and are approved for breast cancer prevention in several countries, aromatase inhibitors and inactivators may represent interesting options in this setting. The encouraging results revealing these drugs to be superior to conventional therapy in metastatic breast cancer confirm their therapy efficacy and suggest that they may also have a role in adjuvant therapy and even for breast cancer prevention. Secondly, whereas the bulk of "high-risk" breast cancer patients with confirmed founder mutations in the BRCA1 or BRCA2 genes develop their cancers earlier in life (during the premenopausal period), 75-80% of all breast cancers, in general, develop in postmenopausal women. Thus, in considering prevention of breast cancer in moderate-risk groups, strategies for prevention in postmenopausal women may play an important role. Also, among high-risk patients who have not developed breast cancer by the time of the menopause, aromatase inhibition could be a feasible option. Considering the potential hazards of long-term use of SERMs, switching to an aromatase inhibitor or inactivator in this setting may be beneficial. Finally, the observation that postmenopausal estrogen levels are related to subsequent risk of breast cancer in the general population underlines the potential for estrogen suppression as a preventive strategy. Results from ongoing studies examining the toxicity of aromatase inhibitors and inactivators in postmenopausal women will set the stage for future trials that explore them as preventive treatment options.

Screening breast cancer patients for Norwegian ATM mutations.

483 Norwegian breast cancer patients were screened for six different ataxia telangiectasia mutated (ATM) mutations previously found to account for 83% of the disease alleles in Norwegian ataxia telangiectasia (AT) patients. Only one carrier was found. These results provide no evidence in favour of an excess risk of breast cancer associated with heterozygosity for classical AT mutations, but remain consistent with a maximum 2.4-fold increased risk.

Tailored fluorouracil, epirubicin, and cyclophosphamide compared with marrow-supported high-dose chemotherapy as adjuvant treatment for high-risk breast cancer: a randomised trial. Scandinavian Breast Group 9401 study.

BACKGROUND: Chemotherapy drug distribution varies greatly among individual patients. Therefore, we developed an individualised fluorouracil, epirubicin, cyclophosphamide (FEC) regimen to improve outcomes in patients with high-risk early breast cancer. We then did a randomised trial to compare this individually tailored FEC regimen with conventional adjuvant chemotherapy followed by consolidation with high-dose chemotherapy with stem-cell support. METHODS: 525 women younger than 60 years of age with high-risk primary breast cancer were randomised after surgery to receive nine cycles of tailored FEC to haematological equitoxicity with granulocyte colony-stimulating factor (G-CSF) support (n=251), or three cycles of FEC at standard doses followed by high-dose chemotherapy with cyclophosphamide, thiotepa, and carboplatin (CTCb), and peripheral-blood stem-cell or bone-marrow support (n=274). Both groups received locoregional radiation therapy and tamoxifen for 5 years. The primary outcome measure was relapse-free survival, and analysis was by intention to treat. FINDINGS: At a median follow-up of 34.3 months, there were 81 breast-cancer relapses in the tailored FEC group versus 113 in the CTCb group (double triangular method p=0.04). 60 deaths occurred in the tailored FEC group and 82 in the CTCb group (log-rank p=0.12). Patients in the CTCb group experienced more grade 3 or 4 acute toxicity compared with the tailored FEC group (p<0.0001). Two treatment-related deaths (0.7%) occurred in the CTCb group. Six patients in the tailored FEC group developed acute myeloid leukaemia and three developed myelodysplastic syndrome. INTERPRETATION: Tailored FEC with G-CSF support resulted in a significantly improved relapse-free survival and fewer grade 3 and 4 toxicities compared with marrow-supported high-dose chemotherapy with CTCb as adjuvant therapy of women with high-risk primary breast cancer.

Genetic variants of CYP19 (aromatase) and breast cancer risk.

The effect of a SNP in exon 10 of CYP19 on tumor mRNA levels and splice variants were studied and correlated with clinical parameters and risk of breast cancer. In the vast majority of breast cancers, the estrogen levels modulate the tumor growth and depend on the activity of CYP19. Patients (n=481) and controls (n=236) were genotyped by T-tracks in a single sequencing reaction (SSR). The frequency of TT genotypes was significantly higher in patients versus controls (P=0.007) particularly among those with stage III and IV disease (P=0.004) and with tumors larger than 5 cm (P=0.001). A significant association between presence of the T allele and the level of aromatase mRNA in the tumors was observed (P=0.018), as well as with a switch from adipose promoter to ovary promoter (P=0. 004). Previously, we reported a rare polymorphic allele of CYP19 (repeat (TTTA)12) to be significantly more frequent in breast cancer patients than in controls. Here we describe another polymorphism, a C - T substitution in exon 10 of the CYP19 gene which is in strong linkage disequilibrium with the (TTTA)n polymorphism but with higher frequency of the variant allele. Our data suggest that the T-allele of the CYP19 gene is associated with a 'high activity' phenotype. Oncogene (2000) 19, 1329 - 1333.

Lovastatin inhibits G1/S transition of normal human B-lymphocytes independent of apoptosis.

Lovastatin is a potent inhibitor of protein prenylation, and it has been reported to have pleiotropic cellular effects. In the present study we have elucidated the effects of lovastatin on cell cycle progression and apoptosis of normal human B-lymphocytes. When added to B-lymphocytes stimulated with anti-immunoglobulin (anti-mu) and SAC, lovastatin (20 microM) inhibited the cells in the late G1 phase of the cell cycle. Thus, no early activation parameters such as Ca(2+) flux or MYC induction were affected by lovastatin, whereas progression of cells into the second cell cycle as well as DNA synthesis was markedly reduced. We therefore examined the effects of lovastatin on components of the cell cycle machinery responsible for regulating the G1/S transition. We demonstrated that pRB phosphorylation, cdk2 activity needed for this phosphorylation, and the levels of cyclin A, D, and E were inhibited after 24 h of lovastatin treatment, while the levels of p27(Kip1) were elevated. There was no effect on p21(Cip1), cyclin D2, cdk4, and cdk6. These data are consistent with the cells being inhibited by lovastatin between 24 and 32 h into G1. Lovastatin added to stimulated B-cells in late G1 still inhibited the DNA synthesis by 60%, but at this point only minor effects were noted on the cell cycle machinery. We therefore looked for induced apoptosis as an explanation for reduced S-phase entry of the cells. However, despite the ability to enhance the apoptosis of unstimulated B-cells from 48 to 61% as judged by the TUNEL method, lovastatin only marginally affected apoptosis when administered to stimulated B-cells. Thus, it appears that accelerated apoptosis cannot account for the effect of lovastatin on cell cycle progression.

CYP17 and breast cancer risk: the polymorphism in the 5' flanking area of the gene does not influence binding to Sp-1.

The ability of a motif of the CYP17 5' untranslated region, created by a polymorphic T to C substitution, to bind to the human transcription factor Sp-1 was investigated. No binding of any of the polymorphic alleles was observed in electromobility shift assay. No other sequence within +1 to +100 of each of the CYP17 alleles formed complex with the Sp-1 or enhanced binding to the polymorphic CACC box. Genotyping of 510 breast cancer patients and 201 controls revealed no difference in genotype frequencies. Age at onset, tumor grade, lymph node status and distant metastases, stage, and estrogen and progesterone receptor status were not associated with the CYP17 genotype.

Comparison of an orally disintegrating ondansetron tablet with the conventional ondansetron tablet for cyclophosphamide-induced emesis in cancer patients: a multicenter, double-masked study. Ondansetron Orally Disintegrating Tablet Emesis Study Group.

A total of 427 cancer patients receiving cyclophosphamide chemotherapy participated in this multicenter, double-masked, double-dummy, parallel-group, randomized study comparing the antiemetic efficacy and safety of an 8-mg conventional ondansetron tablet (OT, n = 212) taken twice daily with an 8-mg orally disintegrating ondansetron tablet (ODT, n = 215) taken twice daily for 3 days. In the primary efficacy analysis, complete or major control of emesis (0 to 2 emetic episodes) between days 1 and 3 was seen in 80% of OT and 78% of ODT patients. The 90% confidence interval for the differences between treatments was -8.6% to 4.4% (defined interval of equivalence, +/-15%), showing that the formulations were equivalent. In the secondary efficacy analysis, no significant differences were observed in the rates of complete control of emesis (no episodes of emesis) over 3 days (63% and 64% of the respective groups) and on day 1 (84% and 81%, respectively) and in the complete control of nausea over 3 days (37% and 43%, respectively) and on day 1 (59% and 61% of patients, respectively). The taste of ODT was acceptable to the majority of patients (89%) who received it. OT and ODT were both well tolerated. Thus 8 mg ODT twice daily represents a palatable, well-tolerated, and effective antiemetic treatment for the control of cyclophosphamide-induced emesis and nausea and provides equivalent treatment to OT 8 mg twice daily.

Single tube multiplex polymerase chain reaction genotype analysis of GSTM1, GSTT1 and GSTP1: relation of genotypes to TP53 tumor status and clinicopathological variables in breast cancer patients.

Glutathione S-transferases are involved in the conjugation of a number of human carcinogens. The frequencies of the deletion alleles coding for GSTM1, and GSTT1, related to deficient conjugation of xenobiotics, as well as a recently reported variant in the exon 5 of GSTP1 were investigated in this study. A multiplex polymerase chain reaction based method for a rapid and high throughput genotype analysis of all three GSTM1, GSTT1 and GSTP1 genes in a single tube was developed. Leukocyte DNA from two hundred and thirty-nine (n = 239) breast cancer patients were genotyped. Tumors from a subset of these breast cancer patients (n = 131) have previously been investigated for mutations in the TP53 gene, levels of p53 protein accumulation and loss of heterozygosity at several loci on chromosome 17. When genetic alterations in the tumors were analyzed with respect to glutathione S-transferase genotypes, a significantly higher proportion of the patients with a G allele (GG + AG) of the GSTP1 had loss of heterozygosity at the TP53 gene locus mapping to 17p, compared with non-G allele carriers (74% versus 29%) (P = 0.018). The patients carrying the G allele of GSTP1 also had more frequently mutations in the TP53 gene in their tumor (38%), compared with patients with the AA genotype (21%) (P = 0.055). G allele carriers had predominantly deletion or transversion mutations in the TP53 gene (5 of 7 and 5 of 6 respectively). A higher frequency of the G allele carriers was observed among patients with negative lymph node status (P = 0.0004). A higher proportion of the patients with positive lymph node status at the time of diagnosis had a combined GSTM1 null/GSTT1 null genotype (P = 0.05). Patients who were homozygous for the deleted GSTM1 allele were found to have a significantly shorter overall survival (P = 0.036).

Dosage of adjuvant G-CSF (filgrastim)-supported FEC polychemotherapy based on equivalent haematological toxicity in high-risk breast cancer patients. Scandinavian Breast Group, Study SBG 9401.

BACKGROUND: Conventional dosages of cytostatics in mg/m2 will cause marked variations in systemic exposure, resulting in over- and under-treatment, at least with respect to side effects. PATIENTS AND METHODS: We are conducting a randomized adjuvant study for breast cancer patients younger than 60 years of age with > or = 70% risk of recurrence within five years. The first 89 consecutive patients who have received nine courses q three weeks of individually dose-escalated and G-CSF (filgrastim)-supported FEC (5-fluorouracil (5-FU), epirubicin, and cyclophosphamide) therapy given with ciprofloxacin prophylaxis were included in this analysis. Six different FEC dose levels were used for treatment at equivalent haematological toxicity. Dose modifications were based on white blood cell and platelet counts on days 8, 11/12, 15, and 22. RESULTS: Eighty-three of 89 patients completed all nine courses. The median epirubicin and cyclophosphamide doses were 782 mg/m2 (range 0-994 mg/m2) and 10.330 mg/m2 (range 0-14.460 mg/m2), respectively. Patients treated at the two highest dose levels experienced NCl grade 0 or 1 toxicities in 73% to 92% of the courses. Three patients have developed acute myeloid leukaemia, and two of them have demonstrated abnormalities compatible with topoisomerase II-poison-related karyotypic changes. CONCLUSIONS: Tailored adjuvant G-CSF-supported FEC polychemotherapy will make it possible for all patients to be treated at equivalent levels of haematological toxicity with significantly higher doses without a marked increase in other organ toxicities.

A rare CYP19 (aromatase) variant may increase the risk of breast cancer.

The aromatase P450 (coded by the CYP19 gene) is responsible for the rate limiting step in the metabolism of C19 steroids to estrogens and is expressed in most breast carcinomas. A polymorphic tetranucleotide repeat (TTTA)n in intron 5, about 80 nucleotides downstream of exon 4 has previously been described. The allele frequencies of the polymorphic repeat were studied in series of 182 sporadic and 185 familial breast cancer patients as well as in 252 healthy control individuals. Five different alleles containing 7, 8, 9, 11 and 12-TTTA-repeats were detected. A relatively rare allele (A1) containing the longest repeat (TTTA)12 was found significantly more frequently in breast cancer patients than in control individuals. This indicates that individuals carrying the A1 allele of CYP19 may have an increased risk of developing breast cancer, OR 2.42 (95% confidence interval [CI] 1.03-5.80). The higher frequency was observed in both sporadic and familial patients, although when each of the groups was compared to the control group only a borderline significance was seen. A higher frequency of A1 allele carriers was also found in the group of patients with positive estrogen receptor and progesterone receptor positive tumors. These data suggest that the CYP19 gene may be involved as a low penetrance gene in breast cancer susceptibility.

Expression of stromelysin-3 in medullary carcinoma of the breast.

Matrix metalloproteinases are belived to play a central role in the invasion and metastasis of human cancers by mediating the degration of extracellular matrix components. Expression of stromelysin-3 (ST3), a new member of matrix metalloproteinase family, was investigated by in situ hybridization in 32 tissue samples of medullary carcinoma of the breast. Eighty-four per cent of the cases showed ST3 gene expression in stromal cells adjacent to tumor cells. Therefore, expression of ST3 in stromal cells may be expected to play a significant role in the destruction of extracellular matrix and invasion of medullary carcinoma of the breast.