Incremental improvement in the efficacy of standard chemotherapy and adjuvant endocrine agents, as well as earlier detection of new breast tumors, have together altered breast cancer diagnosis and treatment during the past 25 years. Advances in therapy have led to rising 5-year survival rates and encouraging reduction in disease mortality.[1] However, traditional chemotherapy achieves its desired effects by targeting all rapidly dividing cells. Desired antitumor effects typically come at the expense of nonmalignant cells, specifically those in a high rate of turnover in the gastrointestinal tract and bone marrow. Therefore, the therapeutic index is significantly narrowed as efforts to obtain aggressive tumor eradication are weighed against achieving a tolerable side-effect profile. Greater understanding of the molecular biology of cancer has allowed novel additions to the chemotherapeutic armamentarium-"targeted" agents that exploit the characteristics unique to cancer cells for their eradication, rather than relying on the more universal "cytodestruction" of standard chemotherapy. Targeted therapies therefore maximize the therapeutic index by improving the efficacy of the anticancer treatment while also reducing toxicities to surrounding noncancerous host cells. Targeting HER Family Receptors: Monoclonal Antibodies Epidermal growth factor receptor (EGFR) is a transmembrane tyrosine kinase receptor important for normal cellular development, damage repair, and survival.[2] Each of the four receptors in the EGFR family-HER1, HER2/neu/ErbB-2, HER3, and HER4-has distinct ligand specificity, but all four possess a homologous transmembrane portion connected to an intracellular tyrosine kinase domain. Once bound to ligand, HER proteins must form either homodimers (eg, HER1/HER1) or heterodimers (eg, HER2/HER3) in order to activate intracellular phosphorylation. The only exception is HER2, which has no naturally existing ligand of its own, and is present only in low levels in normal human tissue as compared to HER1.[3] Instead, it acts as the preferred cofactor for ligand-bound HER1, HER3, and HER4 proteins, increasing the number of initiation stimuli for downstream signaling[4] (Figure 1). In tumor cells, EGFR is upregulated, resulting in increased activation of secondary messenger pathways and cell hyperproliferation.[5] Overampli fication of the EGFR gene has been shown in a variety of human cancers including kidney, bladder, colon, pancreas, lung, rectum, and breast.[6] In addition, high levels of EGFR have been correlated with poor disease prognosis and lower survival among cancer patients.[7] Because of its widespread expression and relevant role in tumor development, the HER family receptors were some of the first targets to be selected for "rational" drug development. The HER2/c-erbB2 gene is amplified in 25% to 30% of invasive breast tumors.[7] Preclinical experiments suggest that high levels of HER2 may forecast a poor prognosis in the same manner that large tumor size, high histopathologic grade, and lack of ER+ and PR+ expression are associated with negative outcomes in breast and ovarian cancers. Trastuzumab(Drug information on trastuzumab) (Herceptin) is the first example of successful targeted therapy for breast cancer directed against the extracellular domain of HER2.[8] By blocking HER2, this recombinant humanized monoclonal antibody prevents kinasemediated activation of the ras/raf/ MAPK and PI3K pathways,[9] therefore inhibiting the mechanisms that initiate tumor growth. Trastuzumab is currently used in the treatment of metastatic breast cancer, either alone[10] or in the presence of taxane chemotherapy.[11] Other phase II clinical trials have shown efficacy when trastuzumab was used with vinorelbine (Navelbine)[12] and gemcitabine(Drug information on gemcitabine) (Gemzar).[13] Another recently completed phase III trial demonstrated that the combination of trastuzumab and paclitaxel(Drug information on paclitaxel) with carboplatin(Drug information on carboplatin) (Paraplatin) resulted in improved response rate and time to progression as compared to trastuzumab and paclitaxel alone in women newly diagnosed with HER2-overexpressing breast cancers.[14] The successes of trastuzumab offer encouraging proof of principle for further targeted therapy development: by identifying unique characteristics of tumor cells, the tumor phenotype can be abrogated without major adverse effects on nonmalignant cells. The low but clinically relevant incidence of trastuzumab-associated cardiomyopathy is a notable exception.[15] Tumors with higher levels of HER2 by immunohistochemistry (IHC) (ie, 3+ score) or with HER2 gene amplification using fluorescence in situ hybridization (FISH) have been shown to respond more convincingly to trastuzumab than those with less HER2 expression (IHC 2+ score). Initial phase II studies testing single-agent trastuzumab enrolled women with any HER2 expression at all, IHC 2+ or 3+ score.[16] While overall response to trastuzumab varied from 15% to 38%, the clinical benefit rate increased to 48% when only examining tumors that scored IHC 3+ for HER2 expression. In the absence of HER2 gene amplification, patients with tumors that exhibit IHC 2+ overexpression of HER2 do not appear to derive benefit from trastuzumab.[17] However, even among IHC 3+ tumors, barely 50% respond to trastuzumab.[ 18] Of those patients who benefit initially from treatment with trastuzumab, most progress again within 9 months.[17] A newer antibody, pertuzumab, has shown promising preclinical efficacy in breast cancer cell lines, and is currently being tested in phase I clinical trials. In contrast to trastuzumab's exclusive specificity for HER2, pertuzumab blocks all HER-mediated signal transduction by interfering with transmembrane receptor dimerization.[19] Combined treatment with both antibodies is proposed to exhibit synergistic inhibition of EGFR signal transduction, while increasing the number of HER2-expressing tumors that will respond to anti-HER therapies.[ 19,20] In hopes of further improving the scope and efficacy of HER-targeted agents, a number of other anti-EGFR monoclonal antibodies are currently in development. For example, cetuximab(Drug information on cetuximab) (Erbitux) is a recombinant chimeric antibody directed against EGFR/HER1 receptor with an affinity greater than twice that of any of its natural ligands. Cetuximab has been shown to induce dimerization, internalization, and downregulation of the EGFR.[21] It has been shown to successfully inhibit tumor growth in many cancer lines including head and neck, colorectal, and pancreatic cancer.[22] While preclinical results in breast cancer cells were also promising, a small study of 13 women treated with paclitaxel and cetuximab did not demonstrate any promising antitumor activity and did encounter significant dermatologic toxicity (personal communication, A. Seidman, 2005). In contrast to the aforementioned human-engineered antibodies, geldanamycin is a naturally occurring ansamycin antibiotic also under investigation as a potential HER2- targeted agent (Figure 2). Produced by the bacteria Streptomyces hygroscopicus during fermentation, geldanamycin has been shown to inhibit tumor growth in rodent fibroblasts by inhibiting intracellular tyrosine kinase phosphorylation.[23] Geldanamycin and its derivative, 17-N-allylamino- 17-demethoxy geldanamycin (17- AAG), appear to block tyrosine kinase activity by inhibiting Hsp-90, a ubiquitous chaperone protein that stabilizes signaling proteins including EGFR.[24] Preclinical studies show that geldanamycin significantly inhibits tumor cell lines that overexpress HER-2, and in mouse xenografts, trastuzumab and 17-AAG demonstrated more superior tumor inhibition than either agent alone (unpublished data, D. Solit). Currently, 17-AAG in combination with trastuzumab is being studied in a phase I/II trial at Memorial Sloan- Kettering Cancer Center (MSKCC) in patients with advanced breast cancer. This trial will examine the extent to which the mechanisms of action of the two drugs acting in concert can slow tumor growth, and will also determine the activity of 17-AAG alone in trastuzumab-resistant tumors. Targeting EGFR: Anti-EGFR Tyrosine Kinase Inhibitors Tyrosine kinase inhibitors (TKIs) are the second main strategy for targeting HER-mediated signaling. These small-molecule inhibitors are modeled after imatinib(Drug information on imatinib) mesylate (Gleevec), a well-validated TKI directed against the bcr-abl translocation that drives the development of 95% of chronic myeloid leukemias. By inhibiting the constitutive tyrosine kinase activity of this oncogene, imatinib therapy alone is sufficient to stop the malignant transformation of cells in chronic myeloid leukemia.[25] Likewise, a TKI bound to the intracellular ATP-binding pocket of the EGFR disrupts downstream phosphorylation and signaling pathways in solid tumor cell lines. By design, anti- EGFR TKIs offer potential advantages over anti-EGFR antibodies, as they are several-fold smaller (~400 Daltons, as compared to ~150,000-Dalton monoclonal antibodies) and therefore provide improved tumor penetration. In addition, they are administered orally rather than by intravenous infusion.[26] However, TKIs are less specific for malignant cells, and can cause mild to moderate toxicities including skin rash and diarrhea, as well as edema and headaches.[27] Anti-EGFR TKIs have been moderately successful in achieving significant clinical response among patients with solid tumors. Gefitinib(Drug information on gefitinib) (Iressa) is a reversible TKI that has demonstrated safety and tolerability in doseescalation studies of a variety of solid tumors,[28-30] including preclinical breast cancer tumor models.[31] Large multicenter phase II trials confirmed antitumor activity in patients with advanced non-small-cell lung cancer (NSCLC),[32,33] motivating the US Food and Drug Administration (FDA) to approve gefitinib as third-line treatment for patients with NSCLC following failure of both docetaxel(Drug information on docetaxel) (Taxotere)- and platinum-based therapies. Further studies by Lynch et al demonstrated that patients with particular mutations within the EGFR gene experience a more rapid, dramatic response when taking gefitinib, suggesting a more specific niche for TKIs that may improve their efficacy in treating patients with NSCLC.[34] Emerging understanding of the biology underlying hormone-dependent breast cancers demonstrates crucial "cross-talk" between EGFR-mediated and estrogen receptor pathways, suggesting a possible role for gefitinib in the treatment of breast cancer as well. Shou et al found that while breast cancer cells overexpressing HER2 seemed resistant to tamoxifen(Drug information on tamoxifen)'s ER-inhibitory effects, using gefitinib in addition to tamoxifen enhanced antitumor effects, presumably by inhibiting HER2-mediated EGFR phosphorylation and cross-activation of ER signaling pathways.[35] Three recent phase II studies in women with metastatic breast cancer demonstrated very slight clinical response or disease stabilization with gefitinib monotherapy. For example, Albain et al reported a partial response in 1 of 63 women enrolled in the trial, and stable disease in 8 others. However, 24% (15 patients) continued to receive treatment after the required 2 months for participation regardless of outcome, and 5 of 12 patients reported improvement in bone pain,[36] offering some encouragement for further investigation of its use for breast cancer. The recent FDA approval of erlotinib (Tarceva) a second anti-EGFR TKI, for the treatment of refractory stage IIIB or IV non-small-cell lung cancer, provides motivation for the investigation of this TKI for use in breast cancer as well.[37] Targeting Angiogenesis Angiogenesis is another necessary aspect of cancer development. As tumors grow, they are increasingly dependent on new blood vessel formation for adequate oxygen and nutrients. A highly conserved, homodimeric member of the PDGF super- family, vascular endothelial growth factor (VEGF) is the most potent and specific glycoprotein driving tumor angiogenesis.[38] This heparinbinding glycoprotein also mediates vascular permeability and induces endothelial migration. VEGF catalyzes signal transduction pathways very similar to those of EGFR by binding one of three extracellular tyrosine kinase receptors: flt-1 (VEGFR-1), KDR (VEGFR-2), or flt-4 (VEGFR-3). Intracellular signaling results in neovascular formation (Figure 3). In tumor cells, overexpression of VEGF results in abnormal, leaky vessels, with blind sacs and variations in flow as compared to normal vasculature.[39] Bevacizumab(Drug information on bevacizumab) (Avastin) is a recombinant monoclonal antibody that binds directly to VEGF so that it is unable to bind any of its usual VEGF-Rs, thereby inhibiting angiogenesis. Preclinical studies with bevacizumab confirmed that VEGF does result in a reduction of tumor microvessel density, as well as a delay in tumor growth.[40] Other studies in murine models affirm that antiangiogenic agents will potentiate the antitumor effects of standard chemotherapy.[41] VEGF is highly expressed in the majority of cancers, including breast cancer.[42,43] In both node-positive and node-negative breast cancers, VEGF has been found to serve as a marker of larger tumors, p53 mutations, and poor tumor differentiation.[ 44] Phase I clinical trials demonstrate that bevacizumab is well tolerated, in comparison to many cytotoxic chemotherapies. Common side effects include hypertension, epistaxis, and proteinuria. More serious side effects such as thromboembolism and pulmonary hypertension are rare.[45] Because of its generally favorable tolerability, it has been readily combined with other chemotherapies in hopes of augmenting antitumor response, including a phase II trial in patients with previously treated metastatic breast cancer where using bevacizumab with vinorelbine resulted in a 31% objective response rate.[46] In a recent study in previously untreated women with inflammatory breast cancer treated with bevacizumab and vinorelbine, reduced VEGF levels were demonstrated, as well as a decrease in vascular permeability and endothelial cell proliferation as seen on dynamic contrast enhanced MRI.[47] A large prospective randomized phase III trial recently compared capecitabine(Drug information on capecitabine) (Xeloda) alone or in combination with bevacizumab in anthracycline- and taxane-pretreated metastatic breast cancer. The combination increased the response rate (19.8% vs 9.1%, P = .001), yet there was no improvement in median time to progression or survival.[ 48] In May 2004, the Eastern Cooperative Oncology Group completed accrual for a phase III trial (study E2100) examining the addition of bevacizumab to weekly paclitaxel, as compared to paclitaxel alone for women with metastatic breast cancer. It is hoped that the results will demonstrate that this standard breast cancer therapy can be further augmented with the addition of bevacizumab. Recent evidence suggests that HER2 and VEGF signaling pathways also rely on "cross-talk" phosphorylation in human breast cancers in much the same way as the EGFR pathway has been linked to ER steroid hormone molecular signaling.[49] Gefitinib has been shown to inhibit production of tumor necrosis factor- alpha, bFGF, and VEGF in several human epithelial cancer cell lines.[50] Therefore, inhibition of both pathways is hypothesized to result in synergistic antitumor effects. A recently completed phase II trial at MSKCC and University of California-San Francisco examined the use of bevacizumab and erlotinib for their combined effects in metastatic breast cancer. Targeting Cyclooxygenase-2 Cyclooxygenase-2 (COX-2) is an inducible enzyme responsible for the rate-limiting conversion of arachidonic acid to prostaglandins in a variety of cellular perturbations such as inflammation and tissue damage.[51] Deregulation of COX-2 and downstream prostaglandins have also been demonstrated in tumorigenesis, and elevated COX-2 has been measured in a variety of epithelial carcinomas including colorectal,[ 52] lung,[53] esophageal,[54] and breast tumors.[55] Elevated levels of COX-2 mRNA, COX-2 protein, and PGE2 in colorectal carcinoma cells was linked to the activation of the HER2/ HER3 pathway in colorectal carcinoma cells-an EGFR family pathway discussed above for its association with breast cancer pathogenesis.[56] Once again, "cross-talk" between the HER2 and COX-2 pathways appears to be associated with the development of breast cancer tumors expressing the HER2-neu oncogene. COX-2-selective inhibitors, such as celecoxib(Drug information on celecoxib) (Celebrex), have been shown to delay incidence of mammary tumors in transgenic mice overexpressing HER-2.[57] The mechanism of action by which COX-2 inhibitors slow tumor progression is still not well understood, but recently Chang et al proposed that by blocking PGE2, a potent inducer of angiogenesis and angiogenic regulatory proteins such as VEGF, COX-2 inhibitors cause apoptosis of the tumor microvasculature as well as tumor mass reduction. Chang et al suggested that, along with VEGF, COX-2 may therefore be another crucial component of the "angiogenic switch," necessary for tumor vessel formation.[ 56] Preclinical studies using a transgenic mouse model demonstrate a decrease in primary mammary tumor burden due to decreased tumor proliferation, as well as an increase in the induction of apoptosis when treated with COX-2 inhibitors.[ 57] Oral administration of celecoxib also demonstrated decreased expression of proangiogenic VEGF. Based on these in vivo studies, an exploratory open phase I clinical trial using celecoxib (either 200 or 400 mg twice daily, orally) in breast cancer patients prior to surgery are currently under way (MSKCC 03-027). COX- 2 expression, downstream signaling, as well as aromatase activity are all being examined (Figure 4). In another preliminary phase II trial at Ohio State University, celecoxib at two different doses is being studied as remission maintenance therapy after four to six cycles of chemotherapy (Cancer and Leukemia Group B 40105).