Search

Breast cancer is a heterogeneous disease whose evolution is difficult to predict by using classic histoclinical prognostic factors. Prognostic classification can benefit from molecular analyses such as large-scale expression profiling. Using immunohistochemistry on tissue microarrays, we have monitored the expression of 26 selected proteins in more than 1,600 cancer samples from 552 consecutive patients with early breast cancer. Both an unsupervised approach and a new supervised method were used to analyze these profiles. Hierarchical clustering identified relevant clusters of coexpressed proteins and clusters of tumors. We delineated protein clusters associated with the estrogen receptor and with proliferation. Tumor clusters correlated with several histoclinical features of samples, including 5-year metastasis-free survival (MFS), and with the recently proposed pathophysiologic taxonomy of disease. The supervised method identified a set of 21 proteins whose combined expression significantly correlated to MFS in a learning set of 368 patients (P < 0.0001) and in a validation set of 184 patients (P < 0.0001). Among the 552 patients, the 5-year MFS was 90% for patients classified in the “good-prognosis class” and 61% for those classified in the “poor-prognosis class” (P < 0.0001). This difference remained significant when the molecular grouping was applied according to lymph node or estrogen receptor status, as well as the type of adjuvant systemic therapy. In multivariate analysis, the 21-protein set was the strongest independent predictor of clinical outcome. These results show that protein expression profiling may be a clinically useful approach to assess breast cancer heterogeneity and prognosis in stage I, II, or III disease.

Supplementary Figure from Protein Expression Profiling Identifies Subclasses of Breast Cancer and Predicts Prognosis

Purpose: To test the efficacy of selective therapy against cyclooxygenase-2 in combination with a low-dose regimen of a cytotoxic agent in the treatment of juvenile hematopoietic malignancies in the experimental model, Friend disease. Experimental Design: Juvenile erythroleukemic mice (n = 8) received no treatment, celecoxib (1600 mg/kg/d), vinblastine (0.5 μg/g twice weekly), vehicle controls, or celecoxib + vinblastine combination (n = 9) over a 6-month period from time of tumor induction. Overt toxicity was assessed daily and recorded weekly. Results: Among randomly selected mice from celecoxib treatment groups, plasma concentrations ranged from 2 to 6 μmol/L. As a single agent, celecoxib was not associated with any apparent toxicity. Monotherapy with vinblastine, however, caused early mortality marked by severe diarrhea, lethargy, and weight loss. At the tested doses, neither vinblastine nor celecoxib enhanced survival as monotherapies. Coadministration of these two drugs alleviated the overt toxicity associated with vinblastine and resulted in a significant increase in survival (P < 0.05). Survivors sampled throughout the study showed a trend to decreased weight loss and hematocrit levels among all groups, but significance was evidenced earlier in the vinblastine monotherapy group overall (P < 0.05). Despite similar degree of splenomegaly, histologic analysis revealed preserved splenic mantle architecture from mice given combination therapy compared with those sampled from mice on all other monotherapies, exhibiting a more diffuse burden of blasts and destruction of germinal centers. Conclusion: We propose that addition of a selective cyclooxygenase-2 inhibitor to a modified low-dose conventional chemotherapeutic regimen protects juvenile mice with Friend disease from succumbing to low-dose cytotoxicity, in part, by neutralizing acute inflammatory responses.

Cyclooxygenase 2 (COX-2) is an inducible enzyme involved in the production of prostaglandins and thromboxanes during inflammation. There are now several lines of evidence indicating that increased expression of COX-2 plays a functional role in the development and progression of malignant epithelial cancers. However, there is only limited data regarding the role of COX-2 in melanoma pathogenesis. In the present work, we retrospectively examined lesions through out the development of melanoma and metastatic disease (dysplastic nevi n = 10, melanoma in situ n = 4, stage II melanoma n = 10, stage III n = 4, stage IV n = 3, stage V n = 2, melanoma metastasis lymph nodes n = 13 metastasis to other sites n = 3). COX-2 was consistently observed in keratinocytes, dermal fibroblasts, and inflammatory cells in regions adjacent to benign evi and primary cutaneous melanomas. However, no COX-2 staining was detected in the nevi nor in the primary skin melanoma cells. In addition, COX-2 was undetected in all vertical and radial growth phase cases Interestingly, 13 out of 13 of the lymph node metastasis expressed extremely high levels of COX-2 in overlying epithelium and inflammatory cells, and COX-2 was strongly detected in the metastatic cancer cells per se. For additional information on the expression of COX-2 in malignant melanoma, we determined the expression of COX-2 protein in several different melanoma cell lines. We found that 3We found that 5 out of 7 of the melanoma cells over expressed COX-2 compared to normal melanocytes. Collectively, these data suggest that COX-2 may play a functional role in metastases of melanoma, and treatment with COX-2 inhibitors may be efficacious for malignant melanoma.

To understand the potential role of cyclooxygenase (COX) in normal and inflammatory human diseases, we characterized the expression of COX-1 and COX-2 in biopsies of osteoarthritis, atherosclerosis, and cancer. Tissues were prepared for immunohistochemistry by standard methods, and representative cases assayed via Western blot and quantitative RT-PCR. COX-2 was not detected in normal human tissues with few exceptions. Moderate to marked COX-2 was observed in the macula densa (MD) and thick ascending limb (TAL) in human fetal kidneys, but was not detected in neonatal and adult MD and TALs. Low level, constitutive COX-2 was detected in colonic epithelium, peribronchial glands, and pancreatic ductal epithelium. Low to moderate COX-2 was detected basally in the cortex, hippocampus, hypothalamus, and spinal cord, and in reproductive tissues during ovulation, implantation and labor. No COX-2 was detected in the existing vasculature in normal tissues, and was also not expressed throughout the ductus arteriosus. COX-2 was markedly induced in human tissues of osteoarthritis, atherosclerosis and cancer. COX-2 was prominently expressed in the synovium, fibrocartilage of osteophytes, and in the blood vessels in the osteoarthritic (OA) knee joint. COX-2 was also prominently detected in the macrophages/foam cells in atherosclerotic plaques, and in the endothelium overlying and immediately adjacent to the fibrofatty lesion. Moderate- to intense COX-2 expression was consistently observed in the inflammatory cells, neoplastic lesions, and blood vessels in all epithelial-derived human cancers studied. In contrast, COX-1 was relatively ubiquitously observed in both normal and pathophysiological conditions. These data collectively imply COX-2 plays an important role in mediating a variety of inflammatory diseases, and imply COX-2 inhibitors may be effective in the prevention and/or treatment of OA, heart disease, and epithelial cancers.