In recent years, the strategy in cancer therapy in general and breast cancer in particular has shifted from the use of high doses of toxic, nonspecific agents to a range of novel agents that target specific molecular lesions found in tumor cells. Advances in molecular biology have allowed the isolation of novel interactions and downstream targets, driving the development of rationally designed targeted therapies. The success of trastuzumab(Drug information on trastuzumab) (Herceptin) in breast cancer and imatinib(Drug information on imatinib) mesylate (Gleevec) in chronic myelogenous leukemia and gastrointestinal stromal tumors provides proof of principle that such an approach can have a marked impact when the mechanism of growth of a particular cancer is understood and specifically interrupted. This article will focus on new, molecular-targeted approaches to the treatment of breast cancer. Of particular interest are classes of drugs that target the tyrosine kinase signal transduction pathways, block tumor angiogenesis, modulate apoptosis, and inhibit histone deacetylation. Targeting the erbB1 Receptor The erbB family consists of four closely related transmembrane receptors: erbB1 (also termed epidermal growth factor receptor [EGFR] or HER1), erbB2 (also termed HER2 or neu), erbB3 (HER3), and erbB4 (HER4). All four erbB receptors share a common molecular architecture composed of three distinct regions: an extracellular ligand-binding domain, a transmembrane region, and an intracellular tyrosine kinase-containing domain that is responsible for the generation and regulation of intracellular signaling (Figure 1). The formation of erbB homodimers and heterodimers following ligand binding and receptor aggregation activates the intrinsic receptor kinase activity via intramolecular phosphorylation and generates a cascade of downstream chemical reactions that transmit a wide variety of cellular effects.[1] The rationale for and development of therapeutics targeting erbB2, particularly trastuzumab, have been reviewed elsewhere,[1] and this section will be limited to a discussion of therapeutics targeting erbB1. The erbB1 receptor is overexpressed in about 40% of breast cancers.[2,3] The frequency of overexpression varies depending on the evaluation method used and whether the truncated EGFRvIII form-a constitutively activated erbB1 variant expressed in a large proportion of breast cancers-is included.[3] The overexpression of erbB1 has been associated with increased proliferation, disease progression, and a poor prognosis in breast cancer.[3,4] ErbB1 expression has also been correlated with decreased estrogen-receptor expression and increased resistance to endocrine therapy.[2,3,5,6] ErbB2 and erbB1 are commonly (10%-36%) coexpressed, and such coexpression has been correlated with a less favorable prognosis.[7,8] Given the wide expression of erbB1 in breast cancer and the important role this receptor plays in signal transduction, the use of erbB1 inhibitors in the treatment of breast cancer has generated considerable interest. The aberrant signaling that occurs through the erbB1 pathway can be caused by high expression of erbB1, mutation of erbB1 (eg, EGFRvIII), decreased phosphatase levels, or heterodimerization of erbB1 with other members of the erbB receptor family (such as HER2).[3] Several different strategies have been used to downregulate signaling through this pathway (Table 1). These include monoclonal antibodies directed against erbB1 such as cetuximab(Drug information on cetuximab) (IMC-C225, Erbitux) and ABX-EGF, and small-molecule inhibitors of erbB1 tyrosine kinase such as gefitinib(Drug information on gefitinib) (ZD1839, Iressa) and erlotinib (OSI 774, Tarceva). Small Molecules Targeting erbB1 Tyrosine Kinase
Small-molecule inhibitors of erbB1 receptor tyrosine kinase prevent receptor dimerization, autophosphorylation, and the resulting downstream signaling. Hypothetically, this approach could inhibit signaling mediated by ligands as well as signaling that is independent of growth factors. In contrast to monoclonal antibodies, such agents may also inhibit ligandindependent signaling due to constitutively active mutant receptors (eg, EGFRvIII). Several erbB1 tyrosine kinase inhibitors are under evaluation, but the anilinoquinazolines, gefitinib and erlotinib, are in the most advanced stages of development.

• Gefitinib-In preclinical studies, gefitinib has demonstrated broad antitumor activity in lung, breast, ovarian, and other tumors.[9] Cell lines that overexpress erbB2 appear to be particularly sensitive to gefitinib, and preclinical data suggest a synergistic inhibitory effect when the agent is combined with trastuzumab in cell lines that coexpress erbB1 and erbB2.[10,11] These observations support the use of erbB1 inhibitors such as gefitinib in combination with therapies that target erbB2. In addition, preclinical data suggest that resistance to endocrine therapy in estrogen-dependent tumors may be modulated through erbB1, which may be thwarted by gefitinib.[6,12] This phenomenon was examined in a recent study in which nude mice bearing erbB2-expressing breast cancer cells (MCF-7/HER2-18) were treated with estrogen, tamoxifen(Drug information on tamoxifen), or estrogen-deprivation alone or together with gefitinib.[12] In this study, erbB2 overexpression increased the agonist properties of tamoxifen, resulting in stimulated growth. However, tamoxifen-stimulated MCF-7/HER2-18 tumor growth was completely blocked in mice treated with gefitinib. In mice treated with gefitinib and estrogen deprivation, the erbB1 tyrosine kinase inhibitor delayed the development of acquired resistance to estrogen deprivation. These observations support the concept that crosstalk between estrogen receptor and erbB1/erbB2-related pathways can modulate resistance to endocrine therapies and suggest that combination therapy may be useful in maintaining estrogen sensitivity following the development of hormone resistance. Additional potential benefits of gefitinib and other therapeutic agents targeting erbB1 stem from their favorable interaction with cytotoxic drugs (eg, paclitaxel, docetaxel(Drug information on docetaxel) [Taxotere], carboplatin(Drug information on carboplatin) [Paraplatin], cisplatin(Drug information on cisplatin), topotecan(Drug information on topotecan) [Hycamtin], and raltitrexed) in human tumor xenograft models and restoration of taxane sensitivity in multidrug-resistant cell lines.[1,13] In phase I trials conducted in patients with advanced breast cancer, gefitinib has demonstrated a favorable tolerability and predictable pharmacokinetic profile when given orally.[14] The clinical benefit and safety profiles of gefitinib were evaluated in a recently reported multicenter phase II study in patients with metastatic breast cancer.[15] Gefitinib was administered at a dose of 500 mg once daily until disease progression, intolerable toxicity, or consent withdrawal. Notably, there were no previous treatment restrictions, and study participants were not screened for the target or target aberrations. The study end point was the clinical benefit rate, defined as the sum of the response rate and the rate of stable disease for 6 months. Of the 63 patients in the trial, 27 (43%) had tumors that were estrogen-dependent, and 17 (27%) had tumors that demonstrated erbB2 over- expression by immunohistochemistry staining. Treatment was discontinued in 5% of patients because of treatmentrelated side effects, and four patients were able to continue treatment after a dose reduction to 250 mg daily. Grade 3/4 toxicity, mainly grade 3 diarrhea, rash, or nausea and vomiting developed in approximately 25% of the patients. One patient achieved a partial response, and two patients had stable disease for an excess of 6 months, yielding a clinical benefit rate of 4.8%. An additional six patients had stable disease for up to 6 months. The median time to progression was 57 days, and about 42% of patients reported diminished pain during therapy. Objective evidence of activity using a rigid definition was low in this heavily pretreated population. However, a considerable proportion of patients (14.3%) achieved a partial response or maintained stable disease for up to 6 months, and therefore, may have derived benefit from this therapy.
• Erlotinib-Another agent that has been studied in women with advanced breast cancer is erlotinib. Much like gefitinib, erlotinib is orally active and was well tolerated in phase I trials.[ 1,16] An open-label phase II trial of erlotinib in metastatic breast cancer was recently completed.[17] Two cohorts of patients were accrued to this study. The first cohort of 47 patients was required to have received prior therapy with an anthracycline, a taxane, and capecitabine(Drug information on capecitabine) (Xeloda). The second cohort of 22 patients merely had to have had tumor progression during chemotherapy. Again, study participants were not prospectively screened for erbB1 overexpression. Erlotinib was administered at 150 mg once daily until tumor progression with dose reduction permitted for treatment-related side effects. In the first cohort, one patient achieved a partial response, and two additional patients had stable disease. In the second cohort, no objective responses were observed, but one patient exhibited stable disease. Treatment- related side effects included acneiform rash, diarrhea, asthenia, and nausea. Correlative studies demonstrated that only 12% of patients had overexpression of erbB1. This suggests that an insufficient number of patients may have had the target to validly test this agent.
• Study Validity-The modest clinical benefit seen in these phase II stud- ies of the erbB1 tyrosine kinase inhibitors likely reflects the indiscriminate treatment of unscreened tumors that may or may not possess the appropriate target or determinants for response. The importance of appropriate identification of patients who are most likely to respond to a targeted approach is well illustrated in the success of trastuzumab in breast cancer. The survival benefits seen with trastuzumab therapy would not have been appreciated if patients had not been screened before treatment for overexpression of erbB2, the principal target of the drug. Equally important is the appropriate selection of end points for phase II studies, ie, those that will allow the appreciation and quantification of tumor growth delay, the predominant benefit of erbB-targeted therapeutics noted in preclinical studies. Therefore, both the identification of predictive biomarkers and a careful trial design are needed to ensure that the usefulness of erbB-targeted therapy is correctly assessed.
• New Directions in Research- More recently, attention has focused on evaluating the feasibility and efficacy of a multitargeted approach. The combination of trastuzumab and erbB1 inhibitors and the dual administration of endocrine therapy and erbB1 inhibitors are subjects of ongoing clinical trials in breast cancer. In addition, the irreversible, pan-erbB tyrosine kinase inhibitor CI-1033, the irreversible erbB1/erbB2 tyrosine kinase inhibitor EKB-569, and the reversible erbB1/erbB2 tyrosine kinase inhibitor GW572016 are undergoing clinical evaluation.[18-24] The relative merits of these mechanisms will be better understood following trials of CI-1033, EKB-569, and GW572016 in relevant tumor types. The rationale for the development of irreversible tyrosine kinase inhibitors such as CI-1033 and EKB-569 was, in part, the higher concentrations of erbB inhibitors required to continuously block erbB1 phosphorylation in intact cells where intracellular adenosine(Drug information on adenosine) triphosphate (ATP) concentrations are higher. The approximately 80% homology between the erbB1 and erbB2 tyrosine kinase has allowed the generation of these receptor tyrosine kinase inhibitors with activity in multiple erbB receptor families. Such agents have potential in patients who are resistant to trastuzumab, as compensatory signaling by other erbB receptors may contribute to trastuzumab resistance. CI-1033 and EKB-569 are comprised of chemical moieties that form covalent bonds with the receptor tyrosine kinase domain, resulting in irreversible receptor binding and sustained inhibition of tyrosine kinase in vitro. This feature may also circumvent drug-binding competition due to high intracellular ATP concentrations. In addition, irreversible compounds require that plasma concentration be attained only long enough to briefly expose the receptors to drug, which would then permanently suppress kinase activity. This process is in contrast to reversible erbB tyrosine kinase inhibitors that require adequate plasma concentrations and/or agents with relatively long half-lives to keep the target suppressed.[1] CI-1033 binds irreversibly within the ATP-binding pocket of erbB tyrosine kinase and inhibits both activation and downstream signaling emanating from erbB1, erbB2, erbB3, and erbB4. In preclinical models, CI- 1033 has been shown to inhibit erbB1 phosphorylation in A341 carcinoma and MDA-MB-453 human breast carcinoma cells and the growth of several human tumor xenografts.[1,18,19] The results of studies of long-term administration of CI-1033 indicate that it maintains tumor suppression for extended periods without the emergence of drug resistance. Like other erbB1 inhibitors, CI- 1033 has demonstrated synergy with other therapeutic modalities. For example, it enhances the cytotoxic effects of the topoisomerase inhibitors, SN-38 and topotecan (Hycamtin) in vitro, possibly interfering with a relevant drug-resistance mechanism.[1] Synergistic in vitro growth inhibition of the erbB1-overexpressing cell line A341 has also been demonstrated with CI-1033 and cisplatin.[19,20] This enhanced chemosensitivity was shown not to be the result of inhibition of DNA repair of cisplatin-DNA adducts, and it has been proposed that blockage of erbB signaling by CI-1033 enables cisplatin to inhibit key genes required for cell survival. In phase I studies, when CI-1033 was administered as a single oral dose weekly for 3 out of 4 weeks and daily for 7 days every 3 weeks, the most common toxicities were mild-to-moderate vomiting, diarrhea, and acneiform rash.[21,22] Antitumor activity has also been observed, with one partial response and stable disease in 30% of patients including one with heavily pretreated breast cancer.[22] Further clinical development of this agent is ongoing for patients with erbB-overexpressing advanced breast cancer. EKB-569 also binds covalently and irreversibly to erbB1. Consistent with its ability to irreversibly bind to erbB1 and erbB2, inhibition of receptor phosphorylation is sustained far longer than are plasma levels of the compound.[ 1,23] Phase I evaluations of EKB-569 administered continuously once daily and for 3 weeks every 4 weeks have been completed, and phase II studies of this agent are ongoing. The agent GW572016 inhibits erbB1 and erbB2 tyrosine kinase in a reversible manner. This drug has demonstrated potent inhibition of tumor growth in vitro and appears selective for tumor cells relative to normal cells. In vivo, GW572106 has antitumor activity against erbB2-overexpressing breast carcinoma xenografts.[24] Clinical evaluation of GW572016 administered on a once-daily continuous schedule is ongoing in breast cancer. In addition, combination studies with other cytotoxic agents (such as capecitabine) are in progress.

Targeting the Ras/Raf/MAPK Pathway The Ras proteins are guanine nucleotide- binding proteins that play a pivotal role in the control of normal and transformed cell growth. Following stimulation by several growth factors and cytokines, Ras activates multiple downstream effectors. The Ras/mitogen-activated protein kinase (Ras/MAPK) pathway plays an important role in breast cancer (Figure 2).[25] Although ras is functionally mutated in < 5% of breast cancers, an upregulation of the classic mitogenic Ras/Raf/MAPK cascade occurs, stimulated by overexpression or amplification of oncogenic protein tyrosine kinase activity (eg, erbB2 or erbB1).[26] Phospholipase-C, one of the signaling proteins activated by receptor dimerization of activated erbB1 and erbB2 enhances Ras activity through its SH3 domain.[27] In addition, the adaptor protein Grb2 that links protein tyrosine kinases to Ras and is overexpressed in breast cancer, may amplify signaling through the Ras pathway in response to growth factors.[28] The amplification of Ras signaling as a result of overexpression of these oncogenes and intermediate signaling molecules leads to increased stimulation of downstream effector molecules including phosphatidylinositol 3-kinase (PI3K) and protein kinase B (Akt). Such oncogenic activation not only confers a proliferative and survival advantage to cancer cells but also supports tumor growth through its proangiogenic effect. Farnesyl Transferase Inhibitors
The Ras pathway may be targeted through the inhibition of farnesylation. This key step in the posttranslational modification of Ras is necessary for membrane localization and function. Initial studies of farnesyl transferase inhibitors (FTIs) suggested that these agents selectively inhibit the anchorage-independent growth of rastransformed cells and reverse the transformational phenotype of rasmutated cells.[26] Recently, the role of Ras proteins in mediating the antitumor effects of FTIs has become less certain. FTIs have demonstrated insufficient activity in tumors with K-ras mutations such as pancreas and colorectal cancers, presumably because another prenylating enzyme, geranylgeranyl transferase, can alternatively prenylate or activate K-ras. In addition, FTIs have demonstrated antiproliferative activity in tumor cell lines with wild-type Ras, suggesting that mechanisms other than inhibition of Ras farnesylation may be involved.[29] The prevailing explanation for the activity of FTIs in tumors such as breast cancer-which rarely involves ras mutations- includes the fact that FTIs prevent signaling through wild-type Ras caused by upstream aberrations (eg, erbB1, erbB2) or that they inhibit farnesylation (activation) of other critical proteins.

• Clinical Trials-Various farnesyl transferase inhibitors have been evaluated in phase I/II clinical trials. These include R115777, SCH66336, and BMS 214662.[26,30-34] In addition, interest has been generated in optimizing the use of FTIs by combining them with cytotoxic agents. Certainly the synergy between cytotoxic agents (particulary taxanes) and FTIs observed in breast cancer cell lines with wild-type Ras supports this approach.[ 30] The prinicipal toxicities encountered with FTIs include schedule-dependent myelosuppression, gastrointestinal effects, and fatigue. Although many of the observed toxicities are common, certain side effects are unique and may be structurally related. Peripheral neuropathy is unique to R115777, whereas transaminitis appears to be encountered more often with BMS 214662. The first phase II study of an FTI in breast cancer was conducted using R115777.[32] Preliminary results indicate that R115777 has single-agent activity in advanced breast cancer, with a clinical benefit rate of 25%. It has also been evaluated in combination with chemotherapy. In a phase I study in patients with solid tumors, R115777 was combined with docetaxel.[ 33] Of 15 patients with breast cancer, 1 achieved a complete response, and 2 achieved partial responses. The dose-limiting toxicity was mostly febrile neutropenia, and the nonhematologic toxicities were diarrhea, fatigue, and vomiting. No discernable pharmacokinetic interaction between the two drugs was documented. The combination of R115777 and capecitabine has also been evaluated in a phase I trial.[34] Diarrhea and handfoot syndrome were the dose-limiting toxicities, and partial responses were seen in various malignancies including breast cancer. More recently, the concurrent inhibition of both erbB2 and Ras signaling is being studied in breast cancer. The rationale for the use of this combination is that inhibition of abnormal Ras expression and normal Ras signaling may enhance the growth inhibitory effects of trastuzumab in erbB2-expressing tumor cells.