REVIEWS. Treatment of HER2-positive breast cancer: current status and future perspectives

Transcription

1 Treatment of HER2-positive breast cancer: current status and future perspectives Carlos L. Arteaga, Mark X. Sliwkowski, C. Kent Osborne, Edith A. Perez, Fabio Puglisi and Luca Gianni Abstract The advent of HER2-directed therapies has significantly improved the outlook for patients with HER2-positive early stage breast cancer. However, a significant proportion of these patients still relapse and die of breast cancer. Trials to define, refine and optimize the use of the two approved HER2-targeted agents (trastuzumab and lapatinib) in patients with HER2-positive early stage breast cancer are ongoing. In addition, promising new approaches are being developed including monoclonal antibodies and small-molecule tyrosine kinase inhibitors targeting HER2 or other HER family members, antibodies linked to cytotoxic moieties or modified to improve their immunological function, immunostimulatory peptides, and targeting the PI3K and IGF-1R pathways. Improved understanding of the HER2 signaling pathway, its relationship with other signaling pathways and mechanisms of resistance has also led to the development of rational combination therapies and to a greater insight into treatment response in patients with HER2-positive breast cancer. Based on promising results with new agents in HER2-positive advanced-stage disease, a series of large trials in the adjuvant and neoadjuvant settings are planned or ongoing. This Review focuses on current treatment for patients with HER2-positive breast cancer and aims to update practicing clinicians on likely future developments in the treatment for this disease according to ongoing clinical trials and translational research. Arteaga, C. L. et al. Nat. Rev. Clin. Oncol. 9, (2012); published online 29 November 2011; doi: /nrclinonc Breast Cancer Research Program, Vanderbilt Ingram Comprehensive Cancer Center, Vanderbilt University School of Medicine, th Avenue South, Nashville, TN , USA (C. L. Arteaga). Genentech Research Oncology, 1 DNA Way, South San Francisco, CA 94080, USA (M. X. Sliwkowski). Dan L. Duncan Cancer Center and Lester and Sue Smith Breast Center, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA (C. K. Osbourne). Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224, USA (E. A. Perez). Department of Oncology, University Hospital, Piazzale Santa Maria della Misericordia 15, Udine, Italy (F. Puglisi). Fondazione Centro San Raffaele del Monte Tabor, Via Olgettina 60, Milan, Italy (L. Gianni). Correspondence to: L. Gianni gianni.luca@hsr.it Introduction HER2 is a transmembrane receptor with tyrosine kinase activity but without a known ligand that was initially identified in a rat glioblastoma model. 1 It belongs to a family of four receptors (EGFR/HER1, HER2, HER3, HER4) that are involved in regulating cell growth, survival and differentiation through interlinked signal transduction involving activation of the PI3K/Akt and the Ras/Raf/MEK/ MAPK pathways (Figure 1). 2 When highly expressed, an excess of HER2 at the cell membrane results in constitutive signaling of downstream pathways. 2 Structural studies revealed that HER2 is always in an active conformation and ready to interact with the ligand-activated HER receptors, 3 and a dominant role has been proposed for HER3 in HER2 signaling. 4 Amplification of the HER2 gene and/ or overexpression at the messenger RNA or protein level occurs in about 20% of patients with early stage breast cancer. 5 Before the advent of HER2-directed therapies, this increased level of HER2 was associated with high recurrence rates and increased mortality in patients with node-positive and node-negative disease. 5,6 Competing interests M. X. Sliwkowski declares an association with the following company: Genentech. C. K. Osborne declares an association with the following companies: Astra Zeneca, Genentech, GlaxoSmithKline and Novartis. L. Gianni declares an association with the following companies: Astra Zeneca, Biogen Idec, Boehringer Ingelheim, Celgene, Eisai, Genentech, Genomic Health, GlaxoSmithKline, Millenium Takeda, Pfizer, Roche and Sanofi Aventis. See the article online for full details of the relationships. The other authors declare no competing interests. HER2-positive breast cancer therapy Trastuzumab The monoclonal antibody trastuzumab is currently the only approved adjuvant treatment specifically for patients with HER2-positive early stage breast cancer. Its anti tumor action is not completely understood but is thought to be mediated by several mechanisms following binding of the antibody to the extracellular domain (ECD) of the HER2 receptor; these mechanisms include antibody-dependent cell-mediated cytotoxicity (ADCC), inhibition of cleavage of the ECD of the HER2 receptor (preventing formation of a residual truncated but constitutively active form), 7 inhibition of ligand-independent HER2 receptor dimerization, inhibition of downstream signal transduction pathways, induction of cell-cycle arrest, induction of apoptosis, inhibition of angio genesis, and interference with DNA repair. 8,9 Other potential mechanisms of action have been proposed, but data from preclinical and translational studies are conflicting. These mechanisms include downregulation of HER2 through endocytosis and trastuzumab-induced internalization of HER2, with consequent increased intracellular degradation, and potential immunological mechanisms such as elimination of tumor-specific CD4 + CD25 bright regulatory T cells resulting in an improved immune response against HER2-positive tumors. 10 In patients, trastuzumab seems to induce tumor cell apoptosis, whereas in culture, anti proliferative effects predominate. 11 In the adjuvant setting, trastuzumab is recommended by both US (National Comprehensive Cancer Network [NCCN]) and European (St Gallen) guidelines for use as 16 JANUARY 2012 VOLUME 9

2 monotherapy after completion of chemotherapy, and in combination with paclitaxel or docetaxel after completion of doxorubicin plus cyclophosphamide, or given concurrently with carboplatin and docetaxel. 12,13 These recommendations are based on results of four large ongoing trials, in addition to several smaller trials (Table 1). Results of individual studies are supported by a recent meta-analysis that included six randomized clinical trials and showed that the combination of trastuzumab with adjuvant chemotherapy produced a significant benefit in disease-free survival (DFS; odds ratio [OR] = 0.69), overall survival (OR = 0.78), locoregional recurrence (OR = 0.53), and distant recurrence (OR = 0.62), as compared to chemotherapy alone. 14 Of note, the statistically significant advantage in overall survival initially observed with the addition of trastuzumab to chemotherapy in the HERA trial 15,16 was not evident at the most recent analysis. 16 However, this finding is probably because 65% of patients in the observation arm crossed over to trastuzumab after the release of positive trial results in Although small tumors (<1 cm) were not included in the majority of randomized trials that assessed trastuzumab, women with node-negative HER2-positive tumors that are cm in size are thought to benefit from adjuvant trastuzumab therapy, on the basis of the demonstration of higher risk than previously appreciated in this population. In addition, subgroup analyses from several of the randomized trials have shown consistent benefit of trastuzumab irrespective of tumor size. 17 Despite the wealth of clinical data available on adjuvant trastuzumab therapy, a number of important questions are still unanswered, including the optimal duration of treatment and how best to combine trastuzumab with cytotoxic and other agents so as to maximize efficacy, but minimize toxicity. Several trials are addressing the question of the optimal duration of trastuzumab therapy (Table 2). Data from the 2 year arm of the ongoing HERA study expected in 2012 should indicate whether longer trastuzumab therapy is better than the current standard duration of 1 year, and several other trials are comparing short durations (9 weeks or 6 months) with the standard treatment. The FinHER study has already provided evidence that 9 weeks of adjuvant trastuzumab (given concurrently with chemotherapy) has improved efficacy in terms of distant relapse-free survival compared with chemotherapy alone, but this was not compared with the standard 1-year duration of trastuzumab treatment. 18,19 Data from the NCCTG N9831 trial (Table 1) suggest that trastuzumab can be more effective if started concurrently with the taxane component of adjuvant chemotherapy than if started after chemotherapy has finished. 20 This approach also reduces the duration of intravenous therapy by approximately 3 months, which might improve convenience. 21 According to the results of large clinical trials, trastuzumab is generally well tolerated when added to, or administered following, a chemotherapy regimen. 22 Although potential cardiotoxicity is a concern, the majority of trastuzumab-related cardiac events are asymptomatic decreases in left ventricular ejection fraction Key points HER2 gene amplification and/or overexpression occurs in about 20% of breast cancers and is associated with more-aggressive disease and, until the advent of HER2-targeted agents, a worse outcome The monoclonal antibody, trastuzumab (which targets HER2), and the smallmolecule tyrosine kinase inhibitor, lapatinib (which targets HER1 and HER2), have considerable efficacy in HER2-positive breast cancer New agents in development include vaccines, modified antibodies and derivatives, tyrosine kinase inhibitors and other agents directed against HER2, other HER family members, and downstream and/or resistance pathways Targets in downstream and/or resistance pathways of particular interest in HER2-positive breast cancer include mtor, PI3K, IGF-1R, Akt, HSP90 and VEGF In advanced-stage disease, randomized trials suggest that the antibody drug conjugate, trastuzumab-dm1, and the dimerization inhibitor, pertuzumab, may have superior efficacy or add to the efficacy of trastuzumab-based therapy Lapatinib, bevacizumab (which targets VEGF), neratinib (a dual HER1 HER2 inhibitor), and the peptide vaccines, GP2 and AE37, are all in adjuvant trials for HER2-positive early stage breast cancer (LVEF). On the basis of available data on the use of trastuzumab in the adjuvant setting, both asymptomatic and symptomatic (including congestive heart failure [CHF]) side effects seem to be treatable and mostly reversible. 22 The risk of severe CHF or cardiac events in patients who received anthracyclines as treatment for breast cancer before receiving trastuzumab ranges from 0.6% to 3.9%, on the basis of the data from the main adjuvant trials. 22 In addition, a meta-analysis including 10,955 patients from adjuvant trials showed that the risk of clinically significant cardiac events (grade 3 or 4) related to chemotherapy and 1 year of trastuzumab therapy was 1.9% versus 0.3% in patients who did not receive the antibody. 23 The BCIRG 006 trial compared the efficacy and safety of a non-anthracycline regimen (docetaxel, carboplatin and trastuzumab) with doxorubicin and cyclophosphamide; both arms were followed by docetaxel treatment. 24 In addition, another arm evaluated doxorubicin and cyclophosphamide followed by docetaxel and trastuzumab. A higher incidence of symptomatic CHF was observed when trastuzumab was added to the anthra cycline-based regimen than when added to the non-anthracycline regimen (2% versus 0.4%; P <0.001). 24 Clinical evaluation before treatment with trastuzumab should include careful screening for cardiac risk factors (that is, baseline LVEF 50 55% in patients >65 years with hypertension, diabetes, or smoking habit, and with BMI >25) and consequent close cardiac monitoring according to specific circumstances. Trastuzumab must be avoided if baseline LVEF is <50%. Data from two major randomized trials in the neoadjuvant setting, 25,26 as well as many nonrandomized trials, 27 indicate that the addition of trastuzumab to neoadjuvant chemotherapy also significantly improves pathological complete response (pcr) rates and event-free survival in patients with locally advanced-stage or early stage breast cancer suitable for preoperative (primary systemic) therapy. In the two randomized neoadjuvant trials, 25,26 trastuzumab was administered concomitantly with anthracyclines and, although a decrease in LVEF was observed in 27% of patients, 25 symptomatic cardiac NATURE REVIEWS CLINICAL ONCOLOGY VOLUME 9 JANUARY

3 Ligand binding domain EGF HB-EGF TGF-α EPR BTC AR EPG NRG1 HB-EGF NRG4 NRG2 EPG EPR NRG1 NRG3 BTC HER family dimerization Extracellular space Kinase domain P P Cytoplasm EGFR/HER1 HER2 HER3 HER4 PI3K SOS Nucleus Akt P Ras HER1 ligand HER3 ligand HER4 ligand Proliferation Motility Invasiveness Resistance to apoptosis Angiogenesis Raf MEK MAPK P Figure 1 Heterodimer formation of members of the HER family and downstream signaling. Signaling downstream of HER family activation is dependent on heterodimerization of the HER family member triggered by ligand binding to the extracellular ligand-binding domain (with the exception of HER2, which has no identified ligand and is always in an open conformation that allows dimerization). Phosphorylation of the HER kinase domains (with the exception of HER3, which does not have a kinase domain) initiates a downstream cascade resulting in VEGF transcription and other physiological responses required for carcinogenesis. Abbreviations: AR, amphiregulin; BTC, betacellulin; EPG, epigen; EPR, epiregulin; HB EFG, heparin-binding EGF-like ligand; NRG, neuregulin. events occurred in only 2% of patients given trastuzumab concurrently with doxorubicin. 25 Trastuzumab is also approved and widely used for patients with metastatic HER2-positive disease in combination with paclitaxel 28 or docetaxel, 29 or as monotherapy. 30 On the basis of the preclinical data on the chemotherapy-sensitizing effect of trastuzumab, the benefit of continuing anti-her2 treatment beyond progression was examined in several retrospective trials and these studies supported this strategy. 31 In addition, although prematurely closed, a randomized phase III trial demonstrated that the combination of trastuzumab plus capecitabine resulted in a significant improvement in overall response and time to progression compared with capecitabine alone in patients with HER 2-positive breast cancer who experienced progression during trastuzumab treatment. 32 Trastuzumab can be combined with a range of other cytotoxic agents, including anthracyclines, although concurrent administration with anthracyclines is associated with an increased risk of cardiotoxicity. 28 This risk can be manageable if the cumulative dose of anthracycline is kept low and/or less-cardiotoxic anthra cyclines are used. 25,37,38 Importantly, the cardiac dysfunction associated with trastuzumab use, which is thought to be mediated via inhibition of HER2 signaling in cardiac myocytes, seems to be largely reversible Lapatinib Lapatinib is the only treatment, other than trastuzumab, approved specifically for patients with HER2-positive advanced-stage breast cancer. Lapatinib reversibly inhibits the intracellular tyrosine kinase activity of both HER2 and EGFR (also known as HER1), suppressing tyrosine autophosphorylation and thereby downstream pathways, such as the MAPK/Erk1/2 and PI3K/Akt pathways Importantly, preclinical studies showed that lapatinib could inhibit the growth of HER2-positive breast cancer cells that were resistant to trastuzumab (including those with truncated HER2 receptors), 45 and that lapatinib could enhance the apoptotic effect of anti-her2 antibodies. 44,46,47 These findings suggested that lapatinib might have additive or even synergistic activity if combined with trastuzumab, and that it might have activity in patients with disease resistant to trastuzumab. Better central nervous system (CNS) penetration was also predicted (since lapatinib is a small molecule), potentially leading to improved control of CNS disease by lapatinib compared with trastuzumab. Early clinical trials indicated modest clinical activity (response rates <10%) for single-agent lapatinib in patients whose disease had progressed when receiving trastuzumab, 48 but the combination of lapatinib and capecitabine showed significantly superior efficacy compared to capecitabine alone in such patients JANUARY 2012 VOLUME 9

4 Table 1 Phase III data for adjuvant trastuzumab in patients with HER2-postive early stage breast cancer Trial name Patients (n) Median follow-up (months) HERA 15,16,146 Node-positive or high-risk node-negative EBC having completed standard adjuvant chemotherapy (5,090) Treatment DFS* (P value) OS* (P value) 48.4 No additional therapy Trastuzumab for 1 year Trastuzumab for 2 years 78.6% 72.2% (P <0.0001) NR NSABP Node-positive EBC (2,101) 46.8 Doxorubicin cyclophosphamide paclitaxel 85.8% B ,148 Doxorubicin cyclophosphamide paclitaxel trastuzumab 75.8% (P <0.001) NCCTG N ,148,149 Node-positive or high-risk node-negative EBC (1,944) BCIRG Node-positive or high-risk ,150 node-negative EBC (3,222) PACS Operable node-positive EBC (528 # ) 63.6 Doxorubicin cyclophosphamide paclitaxel trastuzumab Doxorubicin cyclophosphamide paclitaxel trastuzumab 84% (5 years) 80% (P = 0.019) 65 Doxorubicin cyclophosphamide docetaxel Doxorubicin cyclophosphamide docetaxel trastuzumab Docetaxel carboplatin trastuzumab 47 5-FU cyclophosphamide epirubicin or epirubicin docetaxel 5-FU cyclophosphamide epirubicin or epirubicin docetaxel trastuzumab for 1 year ECOG Stage II BC (234) 64 Paclitaxel trastuzumab doxorubicin cyclophosphamide E ,153** Paclitaxel trastuzumab doxorubicin cyclophosphamide trastuzumab FinHER 18,19 Node-positive or high-risk node-negative EBC (232 # ) 62 Docetaxel or vinorelbine 5 FU cyclophosphamide epirubicin Docetaxel or vinorelbine 5 FU cyclophosphamide epirubicin trastuzumab 75% 84% (P <0.001 vs chemotherapy) 81% (P = 0.04 vs chemotherapy) 78% (3 years) 81% (P = 0.41) 76% (5 years) 73% (P = 0.55) 73.3% 83% (P = 0.12) 89.3% 87.7% (P = 0.11) NR 93.5% 89.4% (P <0.001) NR NR 87% 92% (P <0.001 vs chemotherapy) 87% (P = vs chemotherapy) 96% (3 years) 95% (P = 2.38) 88% (5 years) 83% (P = 0.29) 82.3% 91.3% (5 years) (P = 0.094) *Medians provided when available (generally not estimable). 8 mg/kg trastuzumab initially, then 6 mg/kg every 3 weeks. Based on the first joint analysis of NSABP B 31 and NCCTG N9831 studies (doxorubicin cyclophosphamide paclitaxel vs doxorubicin cyclophosphamide paclitaxel trastuzumab). 147 Protocol amended to allow paclitaxel trastuzumab in mg/kg trastuzumab initially, then 2 mg/kg weekly. # Subset of main study. **Pilot randomized safety study. Abbreviations:, followed by; 5 FU, 5 fluorouracil; BC, breast cancer; DFS, disease-free survival; EBC, early stage BC; NR, not reported; OS, overall survival. A subsequent randomized trial also showed that the combination of lapatinib and trastuzumab had better efficacy than lapatinib alone in patients whose disease had progressed on trastuzumab. 51,52 The combination of trastuzumab and lapatinib was well tolerated in this study, with a low incidence of symptomatic (2%) and asymptomatic cardiac events (3.4%). Overall, despite targeting the same pathway, the incidence of cardiac toxicity seems to be lower with lapatinib than with trastuzumab, 53 possibly owing to different effects on cardiomyocyte mitochondrial ATP stores, 54 other differences in the mechanism of action, or to less-sustained inhibition of HER2 by lapatinib compared with trastuzumab. Randomized trials (Table 3) have shown that CNS involvement might be reduced by lapatinib co-administration with chemotherapy, 49 but results of definitive head-to-head comparisons with trastuzumab are awaited (Table 4). The efficacy of lapatinib seems to be confined to patients with strong HER2 overexpression, as with trastuzumab, although some trials are still ongoing in patients with HER2-negative disease (Tables 3 and 4). 55 Two major adjuvant trials of lapatinib are now ongoing (TEACH and ALTTO), in addition to several trials in the neoadjuvant setting and in patients with advancedstage disease. These trials should establish in the next few years whether lapatinib and trastuzumab should be used together or sequentially, and which settings are optimal for the two agents. Initial results of the NEO-ALTTO trial have been presented. 56 The study randomized 455 patients with >2 cm HER2-positive breast cancer tumors suitable for neoadjuvant therapy to receive lapatinib, trastuzumab or the combination of both. Anti-HER2 agents were given without chemotherapy for 6 weeks, and then weekly paclitaxel was added for 12 weeks before surgery. Adjuvant therapy consisted of three cycles of 5 fluorouracil epirubicin cyclophosphamide, and the same anti-her2 therapy administered in the preoperative phase was continued up to a year. The rate of pcr was significantly higher with the combination of lapatinib and trastuzumab compared with trastuzumab alone (51.3% versus 29.5%; P = ); the pcr rate was 24.7% with lapatinib alone. In terms of toxic effects, patients in the lapatinib arm experienced more grade 3 diarrhea (23%), hepato toxicity (13%), neutropenia (16%) and skin disorders (7%) than patients in other arms. 56 Another phase III neoadjuvant trial (GeparQuinto) enrolled 620 patients with HER2-positive disease to receive trastuzumab and chemotherapy or lapatinib and chemotherapy. 57 A higher rate of pcr was observed in patients treated with trastuzumab (31.7%) than in patients treated with lapatinib (21.7%). Of note, in the lapatinib-treated arm 3.4% of patients discontinued treatment because of toxic effects. In the randomized phase II CHERLOB trial, the activity of preoperative taxane anthracycline chemotherapy in combination with trastuzumab, lapatinib, or combined treatment of trastuzumab and lapatinib was evaluated in patients with HER2-positive, stage II IIIA breast cancer. 58 For all arms, chemotherapy consisted of weekly paclitaxel NATURE REVIEWS CLINICAL ONCOLOGY VOLUME 9 JANUARY

5 Table 2 Ongoing phase III trials of adjuvant trastuzumab in patients with HER2+ early stage breast cancer Trial name (clinicaltrials.gov identifier) Patient characteristics (target accrual) Treatment PHARE (NCT ) Resectable BC (3,400) Standard chemotherapy + trastuzumab for 1 year (concurrent or sequential) vs standard chemotherapy + trastuzumab for 6 months (concurrent or sequential) NCT Node-positive EBC (478) 5 FU cyclophosphamide docetaxel* trastuzumab trastuzumab* for 1 year vs 5 FU cyclophosphamide docetaxel* trastuzumab trastuzumab for 6 months NSABP B 43 (NCT ) PERSEPHONE (NCT ) HER2+ DCIS resected by lumpectomy (2,000) WBI over 5 weeks vs WBI over 5 6 weeks + trastuzumab in week 1 and 3 EBC (4,000) Standard adjuvant or neoadjuvant chemotherapy + trastuzumab for 1 year (concurrent or sequential) vs standard adjuvant or neoadjuvant chemotherapy + trastuzumab for 1 year (concurrent or sequential) SHORT-HER (NCT ) Stage I IIIA BC (2,500) Doxorubicin cyclophosphamide or epirubicin cyclophosphamide paclitaxel or docetaxel + trastuzmab trastuzumab vs docetaxel trastuzmab 5 FU cyclophosphamide epirubicin NCT SOLD (NCT ) RESPECT (NCT ) Stage I IIIC HER2+ BC with measurable disease (552) Node-positive or high-risk node-negative EBC (3,000) Stage I IIIA HER2+ BC, age years (300) Docetaxel 5 FU cyclophosphamide epirubicin + intravenous trastuzumab vs docetaxel 5 FU cyclophosphamide epirubicin + subcutaneous trastuzumab Docetaxel trastuzumab 5 FU cyclophosphamide epirubicin vs docetaxel trastuzumab 5 FU cyclophosphamide epirubicin trastuzumab Trastuzumab plus chemotherapy (choice of 5 regimens) vs trastuzumab Primary end point Data expected TTR 2010 DFS 2010 Time to ipsilateral DCIS recurrence or invasive BC 2011 DFS 2011 DFS 2012 pcr and serum trastuzumab concentrations 2014 DFS 2015 DFS 2016 *Dose-dense docetaxel (every 2 weeks with granulocyte-colony stimulating factor). 6 mg/kg trastuzumab every 2 weeks. 8 mg/kg trastuzumab initially, then 6 mg/kg every 3 weeks. 4 mg/kg trastuzumab initially, then 2 mg/kg weekly. Abbreviations:, followed by; 5 FU, 5 fluorouracil; BC, breast cancer; EBC, early stage BC; DCIS, ductal carcinoma in situ; DFS, disease-free survival; HER2+, HER2-positive; pcr: pathological complete response; TTR, time to recurrence; WBI, whole breast irradiation. followed by 5 fluorouracil epirubicin cyclophosphamide. The pcr rate was 28% in the trastuzumab arm, 32% in the lapatinib arm, and 48% in the combination (trastuzumab lapatinib) arm. No patient had symptomatic cardiac events, including CHF. Diarrhea, rash, and hepatic disorders occurred more frequently in lapatinib-containing arms than in the trastuzumab arm. Overall, these data suggest that the combined use of trastuzumab and lapatinib might provide superior efficacy to either agent used alone, with manageable toxic effects. The recent premature closure of the lapatinib-alone arm of the ALTTO study supports this view. Mechanisms of resistance Although trial results anticipated in the next few years will help to optimize the adjuvant trastuzumab therapy and establish the role of lapatinib, it is likely that inherent and acquired resistance to trastuzumab and lapatinib will still result in relapse and progression of HER2-positive disease. At the moment, approximately 5,000 patients with HER2-positive breast cancer die from this disease each year in the USA, despite the availability of trastuzumab and lapatinib (M. Sliwkowski, personal communication). Furthermore, the risk of cardiotoxicity currently precludes certain patients from trastuzumab treatment, limits the choice of agents that can be used concurrently with trastuzumab, and necessitates careful cardiac monitoring during HER2-directed therapy. As a result, there is still a real need for new therapies for patients with HER2-positive breast cancer. Potential mechanisms of resistance to trastuzumab include factors related to HER2 interactions with other members of the HER family or trastuzumab, including the loss of, 59 or increased, HER2 expression; 60 increased HER1 or HER3 expression; 61 increased TGF α expression (a ligand for EGFR/HER1); steric hindrance of HER2-antibody interaction by membrane-associated glycoproteins; 62 and inhibition of trastuzumab binding by HER2 ECD fragments cleaved from the HER2 receptor. 61 Incomplete HER family blockade might be an important resistance mechanism, since it could allow another HER receptor to compensate when one receptor is blocked. 63 Resistance to trastuzumab might also arise through alternative signaling pathways or through constitutive activation of the PI3K/Akt signaling pathway, which is activated by HER2 signaling (and therefore suppressed by HER2 inhibitors such as trastuzumab). Constitutive activation might occur, for example, due to mutations in the PIK3CA gene and/or loss of PTEN. Similar to HER2, the IGF-1R, which can form heterodimers or hetero trimers with HER2, 64 activates the PI3K/Akt pathway and this mechanism is thought to be an important source of trastuzumab resistance. 61,62,65 Conversely, PTEN suppresses the activation of the PI3K/Akt pathway and loss of PTEN activity results in increased Akt activity and resistance to trastuzumab. 61,62,65 In addition, down regulation of the cyclin-dependent kinase p27 kip1, 62 increased acti vity of the GTPase p21-rac1, 66 and increased Met receptor tyrosine kinase activity, 67 have all been implicated in trastuzumab resistance, at least in 20 JANUARY 2012 VOLUME 9

6 Table 3 Selected phase III data for lapatinib in patients with breast cancer Trial name Patients (n) Treatment Primary end point Other outcomes EGF ,50 HER2+ ABC* (399) Capecitabine vs capecitabine lapatinib Median TTP 4.3 months ORR 14% vs 24% (P = 0.017); vs 6.2 months (P <0.001) median OS 15.3 months vs 15.6 months (P = 0.177) EGF EGF ,156 Previously untreated MBC (86 HER2+, 493 HER2 ) Postmenopausal women with HR+ ABC (219 HER2+, 1,067 HER2 ) Paclitaxel placebo vs paclitaxel lapatinib Letrozole placebo vs letrozole lapatinib TTP 22.9 weeks vs 29 weeks (P = 0.142) In HER2+ subset: 25.1 weeks vs 36.4 weeks (P = 0.005) In HER2+ subset: median PFS 3.0 months vs 8.2 months (P = 0.019) EGF HER2+ MBC (296) Lapatinib vs lapatinib trastuzumab Median PFS 8.1 weeks vs 12.0 weeks (P = 0.008) NEO-ALTTO 56 GeparQuinto #,57 HER2+ tumors >2 cm (stage II IIIA) (target accrual 455) HER2+ primary BC requiring adjuvant therapy (target accrual 2,547) Lapatinib lapatinib paclitaxel surgery FEC lapatinib vs trastuzumab trastuzumab paclitaxel surgery FEC trastuzumab vs lapatinib trastuzumab lapatinib trastuzumab paclitaxel surgery FEC lapatinib trastuzumab Epirubicin cyclophosphamide docetaxel trastuzumab vs epirubicin cyclophosphamide docetaxel lapatinib pcr rate 24.7% vs 29.5% vs 51.3% (P = favoring the combination of lapatinib and trastuzumab) pcr rate 31.7% vs 21.7% (P <0.05) ORR 25% vs 35% (P = 0.008); EFS or OS not significant; in HER2+ subset: EFS, ORR, and CBR significantly better with paclitaxel lapatinib In HER2+ subset: ORR 15% vs 28% (P = 0.021), CBR 29% vs 48% (P = 0.003), OS 32.3 months vs 33.3 months (P = 0.113) CBR 12.4% vs 24.7% (P = 0.01), OS 39.0 weeks vs 51.6 weeks (P = 0.106), ORR 6.9% vs 10.3% (P = 0.46) ORR, percentage of node-negative disease at surgery, rate of breast conserving surgery, DFS, OS to be reported Breast conservation rate, DFS, OS, cerebral DFS to be reported *Progressed on anthracycline, taxane, and trastuzumab. Results are based on the most recent data. 35 Not selected for HER2. Progressed on prior trastuzumab. 4 mg/kg trastuzumab initially, then 2 mg/kg every week. # HER2+ part (trial included four other arms for patients with HER disease). Abbreviations:, followed by; ABC, advanced-stage BC; BC, breast cancer; CBR, clinical benefit rate; DFS, disease-free survival; EFS, event-free survival; FEC, 5 fluorouracil epirubicin cyclophosphamide; HER2+, HER2-positive; HER2, HER2-negative; HR, hormone receptor; MBC, metastatic breast cancer; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; pcr, pathological complete response; TTP, time-to-tumor progression. in vitro studies. It is thought that multiple mechanisms of resistance may coexist in trastuzumab-resistant cells. 59 Many of the mechanisms of resistance to trastuzumab would not be expected to interfere with the activity of lapatinib, notably those involving interactions with the ECD of the HER2 receptor, ligand binding or dimerization. Resistance due to PI3KCA mutations and a low PTEN expression would be expected to affect the sensitivity of tumor cells to both trastuzumab and lapatinib. 59 However, despite some recent conflicting results, data suggest that PI3K amino acid substitutions and PTEN loss may be less important in resistance to lapatinib compared with trastuzumab. 61,68,69 Indeed, resistance to lapatinib has been attributed to redundant survival pathways that could be induced as a consequence of a marked inhibition of HER2 kinase activity. For example, prolonged inhibition of the PI3K/Akt pathway in lapatinib-exposed cells might result in upregulation of the transcription factor FOXO3A, which in turn leads to increased estrogen receptor signaling. 70 Similar to the derepression of estrogen receptor, increased phosphorylation of RelA, the pro-survival subunit of NFκB, has been proposed as a potential estrogen receptor-independent mechanism of acquired autoresistance to lapatinib. 71 Preclinical data suggest that activation of RelA by persistent exposure to lapatinib might promote cell survival and, in turn, cause resistance by competing with the drug-induced proapoptotic effects. A number of therapeutic strategies have been devised to overcome or avoid resistance to trastuzumab and lapatinib (Table 5). Strategies that are not likely to specifically benefit patients with HER2-positive disease or do not clearly involve the HER2 pathway (for example, new cytotoxic agents) will not be considered further here, although such agents can logically be ideal candidates for combination with HER2-targeted therapies owing to non-cross resistance and non-overlapping toxicity. Antibody modification ADCC involves an interaction between the constant region (Fc) of an antibody and leukocyte (Fcγ) receptors (FcγRs). The genotype of the FcγRIIIa 158 correlates with response and progression-free survival (PFS) in patients receiving trastuzumab-based therapy for metastatic breast cancer, suggesting that ADCC is important in the antitumor activity of trastuzumab. 72 Attempts to improve the immune effector function of therapeutic antibodies include synthesis of an IgE homolog of trastuzumab, 73 and modification of the Fc region by amino acid substitution or depletion of fucose from the oligosaccharide moiety. 74 Recombinant antibodies lacking fucose show enhanced FcγR binding and ADCC, 75 and in mice, nonfucosylated trastuzumab was more effective against tumor xenografts than unmodified trastuzumab. 76 Interestingly, ADCC of a fucose-negative version of trastuzumab and ADCC of commercial trastuzumab (fucosylated) were analyzed using peripheral blood mononuclear cells (PBMC) from 30 volunteers including 20 patients with breast cancer. 77 PBMC were used as effector cells and HER2-positive breast cancer cell lines as target cells. The study showed a significantly enhanced ADCC with the fucose-negative version of trastuzumab, NATURE REVIEWS CLINICAL ONCOLOGY VOLUME 9 JANUARY

7 Table 4 Ongoing randomized trials with lapatinib in patients with HER2-positive breast cancer Trial (ClinicalTrials.gov) Adjuvant setting TEACH (NCT ) ALTTO, BIG 2 06/N063D (NCT ) 157 Neoadjuvant setting LETLOB (NCT ) TRIO-TORI B-07 (NCT ) ELATE (NCT ) Phase III III II II II Patient characteristics* (target accrual) HER2+ stage I IIIC BC having completed standard adjuvant/ neoadjuvant chemotherapy (3,000) HER2+ BC having completed standard adjuvant/neoadjuvant chemotherapy (8,000) PW with HER2, HR+ tumors >2 cm (stage II IIIA) (91) HER2+ operable stage I III BC (140) HER2+ operable stage I IIIA BC (164) CHERLOB II HER2+, tumors >2 cm below (NCT ) 58,159 stage IIIB (120) GEICAM/ (NCT ) CALGB (NCT ) NSABP B 41 (NCT ) EPHOS B (NCT ) Advanced-stage disease HERLAP (NCT ) II III III III HER2+ resectable BC or LABC (stage I IIIB) (102) HER2+, operable, measurable stage II III BC (400) HER2+ tumors >2 cm diameter (522) Treatment Primary end point Data expected Placebo (1 year) vs lapatinib (1 year) DFS 2012 Trastuzumab trastuzumab (total 52 weeks) vs lapatinib (total 52 weeks) vs trastuzumab washout lapatinib vs lapatinib trastuzumab vs alternative design : same arms as above plus concurrent docetaxel paclitaxel Letrozole placebo vs letrozole lapatinib Trastuzumab docetaxel carboplatin vs lapatinib docetaxel carboplatin vs lapatinib trastuzumab docetaxel carboplatin lapatinib trastuzumab Lapatinib epirubicin cyclophosphamide paclitaxel lapatinib vs epirubicin cyclophosphamide paclitaxel trastuzumab DFS 2013 Response rate by ultrasound 2009 pcr rate 2011 pcr in breast 2012 pcr rate after Paclitaxel FEC trastuzumab vs paclitaxel FEC lapatinib vs paclitaxel FEC lapatinib trastuzumab 24 weeks 28% vs 32% vs 48% Epirubicin cyclophosphamide docetaxel trastuzumab vs epirubicin cyclophosphamide docetaxel lapatinib Paclitaxel trastuzumab surgery vs paclitaxel lapatinib surgery vs paclitaxel lapatinib trastuzumab surgery Doxorubicin cyclophosphamide paclitaxel trastuzumab surgery trastuzumab vs doxorubicin cyclophosphamide paclitaxel lapatinib surgery trastuzumab vs doxorubicin cyclophosphamide paclitaxel lapatinib trastuzumab surgery trastuzumab III HER2+ operable BC (250) Surgery only vs trastuzumab surgery trastuzumab vs lapatinib surgery lapatinib II Untreated patients with measurable HER2+ MBC (120) NCT II Untreated patients with HER2+, p95her2+ MBC (300) CALGB (NCT ) CAN-NCIC-MA31 (NCT ) CEREBEL (NCT ) III PW with measurable HR+ ABC (324) 2011 pcr rate 2011 pcr 2010 pcr rate 2012 Apoptosis, proliferation, RFS 2012 Trastuzumab vs lapatinib ORR 2011 Trastuzumab docetaxel/paclitaxel/vinorelbine vs lapatinib docetaxel/paclitaxel/vinorelbine III Untreated HER2+ MBC (600) Paclitaxel/docetaxel trastuzumab vs paclitaxel/docetaxel lapatinib PFS 2015 Fulvestrant vs fulvestrant lapatinib PFS 2008 # PFS 2011 III HER2+ MBC (650) Capecitabine trastuzumab vs capecitabine lapatinib Incidence of CNS metastases as site of first relapse NCT III HER2+ MBC (276) Trastuzumab vs lapatinib trastuzumab PFS 2014 NCT III PW with HR+, HER2+ MBC (525) AI trastuzumab vs AI trastuzumab lapatinib vs AI lapatinib OS 2017 *See the trial on ClinicalTrials.gov for full details of patient inclusion criteria. 4 mg/kg trastuzumab initially, then 2 mg/kg every week. 8 mg/kg trastuzumab initially, then 6 mg/kg every 3 weeks. Preliminary (blinded) results presented ASCO Data expected 2010 (recruiting in April 2011). # Data expected 2008 (active, but not recruiting in July 2011). Abbreviations:, followed by; ABC, advanced-stage BC; AI, aromatase inhibitor; BC, breast cancer; CNS, central nervous system; DFS, disease-free survival; FEC, 5 fluorouracil epirubicin cyclophosphamide; HER2+, HER2-positive; HER2, HER2-negative; HR, hormone receptor; LABC, locally advanced BC; MBC, metastatic BC; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; pcr, pathological complete response; PW, postmenopausal women; RFS, recurrence-free survival suggesting that removal of a fucose from the antibody structure could result in improved efficacy (and, possibly, a low dose of the drug required). Other ways of enhancing the immunological function of trastuzumab include construction of bispecific or trispecific antibodies, antibody fragments, or single-chain derivatives that bind to specific FcγRs or CD3 on the surface of immune effector cells, as well as to HER2. Single-chain antibodies can also be manipulated to reduce unwanted immunological effects, such as cytokine release. 78 Most single-chain antibodies have not progressed beyond preclinical evaluation, although ertumaxomab reached phase II clinical assessment. Ertumaxomab is a trifunctional, hybrid monoclonal 22 JANUARY 2012 VOLUME 9

8 antibody that binds to HER2, CD3, and the FcγR type I/III. Thus, it linked T lymphocytes and macrophages to HER2- expressing cancer cells, leading to their destruction by phagocytosis. 79 In vitro studies indicated that ertumaxomab could destroy cells with low levels of HER2 expression, as well as those with high HER2 overexpression. 80 A phase I study in patients with HER2-positive breast cancer showed antitumor responses in five of 15 patients, in addition to strong immunological responses in almost all patients. 81 Toxicity was mainly related to cytokine release, and systemic inflammatory response syndrome was the dose-limiting toxicity. Unfortunately, the development of ertumaxomab in breast cancer seems to have been terminated, although apparently not owing to safety concerns (NCT , NCT and NCT ). Arming HER2 targeting agents Trastuzumab, or derivatives of trastuzumab, have also been used as a means of delivering a range of toxins or drugs to HER2-expressing cells. However, toxicity (for example hepatotoxicity with pseudomomas exotoxin conjugates 82 ) can be problematic. The most advanced compound in development is trastuzumab-dm1, a conjugate of trastuzumab (one molecule) with an average of 3.5 molecules of the microtubule polymerization inhibitor DM1 (a derivate of maytansine), which retains the known mechanisms of action of trastuzumab, despite conjugation. 83 Thrombocytopenia was a dose-limiting toxicity in a phase I study. 84 In two phase II studies in patients with heavily pretreated HER2-positive cancers who had progressed on trastuzumab and lapatinib in the metastatic setting, trastuzumab-dm1 produced response rates between 33.8% and 41%. 85,86 Higher response rates were seen in patients with centrally-confirmed HER2- positive disease, reinforcing the importance of accurate HER2 assessment in patients receiving HER2-targeted therapies. 85,86 Trastuzumab-DM1 also produced higher response rates and had a favorable toxicity profile compared with trastuzumab plus docetaxel, in a randomized study in previously untreated patients with HER2- positive breast cancer. 88 In particular, the incidence of grade 3 adverse events in the trastuzumab-dm1 arm was half that in the trastuzumab plus docetaxel arm (37% versus 75%), no grade 3 neutropenia was observed with trastuzumab-dm1, and only 1.5% of patients experienced alopecia. Importantly, trastuzumab-dm1 was not associated with an increased risk of cardiotoxicity compared with trastuzumab plus docetaxel. 88 Currently, randomized studies are ongoing comparing trastuzumab- DM1 with capecitabine plus lapatinib, and combining trastuzumab-dm1 with pertuzumab, in patients with HER2-positive advanced-stage disease (Table 6). Inhibition of HER2 dimerization Although active against HER2 homodimers, trastuzumab is not effective against ligand-induced HER2 heterodimers. HER2 EGFR interactions and, particularly, HER2 HER3 interactions are important in driving HER2-positive breast cancer cells, and also in bypassing trastuzumab-mediated inhibition of cell growth and proliferation. 3,4 The monoclonal antibody, pertuzumab, binds to the HER2 ECD but at a different site to trastuzumab, and is able to inhibit ligand-induced dimerization of HER2 with its receptor partners. 89,90 Preclinical experiments showed that pertuzumab and trastuzumab produced a more-complete blockade of the HER signaling network when combined, and were more effective in HER2-positive tumor xenografts, than either antibody alone. 91 However, cetuximab, a monoclonal antibody targeting EGFR, did not increase the antiproliferative effects of either trastuzumab or pertuzumab when administered concurrently. 92 In a phase II clinical trial, treatment with pertuzumab and trastuzumab together resulted in a 24% overall response rate and a 50% clinical benefit rate, in patients with HER2-positive metastatic breast cancer that had previously progressed on trastuzumab. 93 However, efficacy in patients with HER2-negative breast cancer was disappointingly low (response rates <5%). 94 Currently, the efficacy and tolerability of pertuzumab in combination with trastuzumab are being evaluated in several randomized trials in patients with HER2-positive breast cancer. In the NEOSPHERE neoadjuvant trial, patients with operable, locally advanced or inflammatory HER2-positive breast cancer were randomized to receive one of four combination treatments: docetaxel plus trastuzumab and pertuzumab, docetaxel plus trastuzumab, docetaxel plus pertuzumab, or pertuzumab plus trastuzumab (without chemotherapy). 95 A statistically significant increase in pcr rate was seen when pertuzumab was combined with docetaxel and trastuzumab as compared with the docetaxel and trastuzumab combination (45.8% versus 29%; P = 0.014; Table 6). Interestingly, a pcr rate of 16.8% was observed in patients who did not receive chemotherapy. Although promising, these results are not considered to be practice changing because the study was not designed to test long-term outcomes and pcr is not unanimously accepted as a surrogate for disease-free survival and overall survival. However, a preliminary announcement of positive data from the CLEOPATRA study (in which patients with metastatic breast cancer were randomized to received pertuzumab plus trastuzumab and docetaxel, or placebo plus trastuzumab and docetaxel) suggest that the findings of the NEOSPHERE study may be validated in this larger and more-definitive trial. Since trastuzumab and pertuzumab both target the HER2 receptor and are structurally very similar, additive toxicity might be anticipated when the two drugs are administered concurrently. However, as seen with concurrent administration of trastuzumab and lapatinib, cardiac toxicity does not seem to be increased when pertuzumab is given with trastuzumab. A pooled analysis of cardiac safety in 598 patients participating in pertuzumab clinical trials showed no apparent increase in cardiac dysfunction when pertuzumab was given concurrently with trastuzumab. 96 Of the patients treated with pertuzumab alone, pertuzumab in combination with a non-anthracyclinecontaining cytotoxic, or pertuzumab with trastuzumab, 6.9%, 3.4%, and 6.5%, respectively, developed asymptomatic reduction in LVEF. In addition, 0.3%, 1.1%, and 1.1%, NATURE REVIEWS CLINICAL ONCOLOGY VOLUME 9 JANUARY

9 Table 5 New approaches in the therapy of patients with HER2-positive breast cancer* Strategies and drugs Stage of development Reference or ClinicalTrials.gov identifier Optimization of trastuzumab antibody structure Ertumaxomab (trifunctional, bispecific mab targeting HER2 and CD3) Conjugation of HER2-targeted agents with toxins Phase II (terminated) Kiew et al. (2008), 79 Jäger et al. (2009), 80 Kiewe et al. (2006) 81 Trastuzumab DM1 (trastuzumab conjugated to the maytansine derivative DM1) Phase III Krop et al. (2009) 160 Targeting HER1 Pelitinib (irreversible HER1 TKI) Phase I II (suspended) Ocaña et al. (2009) 107 Targeting HER3 MM 121 (HER3-targeted mab) Phase I II Schoeberl et al. (2010), 161 NCT , NCT MM 111 (HER2/HER3 bispecific antibody) Phase I II Huhalov et al. (2010), 162 NCT , NCT Targeting HER2 Pertuzumab (mab, HER2 dimerization inhibitor) Phase III Baselga et al. (2010) 93 Broad-spectrum TKIs Neratinib (irreversible HER1/HER2 TKI) Phase III Burstein et al. (2010) 108 BIBW 2992 (irreversible HER1/HER2 TKI) Phase II Hickish et al. (2009) 163 Inhibition of PI3K (class I) XL147 (pan-pi3k inhibitor [all class I isoforms]) Phase I II Shapiro et al. (2009), 115 NCT , NCT BGT226 (p110α-selective PI3K inhibitor) Phase I II NCT , NCT Inhibition of mtor Everolimus Phase III Ellard et al. (2009), 129 Baselga et al. (2009), 164 Inhibition of IGF 1R pathway Figitumumab (mab against IGF 1R) Phase II Gualberto (2010), 165 NCT Cixutumumab (mab against IGF 1R) Phase II McKian & Haluska (2009), 166 NCT , NCT AVE1642 (mab against IGF 1R) Phase II (terminated, company decision) NCT Dalotuzumab (mab against IGF 1R) Phase I II (completed) NCT AMG479 (mab targeting IGF 1R) Phase II NCT OSI 906 (IGF 1R inhibitor) Phase I II NCT Inhibition of HSP90 Alvespimycin Phase I II (phase II in HER2+ BC completed) Miller et al. (2007) 137 Retaspimycin Phase II Hanson et al. (2009), 138 NCT BIIB021 Phase I II Lundgren et al. (2009), 140 NCT AUY922 Phase I II NCT Vaccines and immunotherapy E75 (peptide vaccine based on extracellular domain of HER2) Phase II Peoples et al. (2008), 167 Mittendorf et al. (2008), 168 Patil et al. (2010), 169 Holmes et al. (2008) 170 GP2 (peptide vaccine based on transmembrane domain of HER2) Phase II Carmichael et al. (2010) 171 AE37 (Ii-key hybrid HER2 peptide vaccine) Phase II Holmes et al. (2008) 172 HER2 intracellular-domain peptide vaccine Phase I II NCT , NCT , NCT HER2 protein AUTOVAC (PX ) Phase I II (discontinued) NCT dher2 (a modified HER2 protein) with AS15 adjuvant Phase I II NCT , NCT Allogeneic GM CSF-secreting whole-cell breast-cancer vaccine Phase II NCT , NCT , NCT , NCT , NCT Autologous dendritic-cell vaccines (dendritic cells are loaded with HER2 peptides or genetically manipulated to express HER2) PG13-4D5 D12 (anti-her2 CAR; autologous peripheral blood lymphocytes transduced with a retroviral vector) Phase I II Phase I II Morse et al. (2007), 173 NCT , NCT NCT JANUARY 2012 VOLUME 9

10 Table 5 (Cont.) New approaches in the therapy of patients with HER2-positive breast cancer* Strategies and drugs Stage of development Reference or clinicaltrials.gov identifier Multitarget kinase and angiogenesis inhibitors Bevacizumab (mab against VEGF A) Phase III Pegram et al. (2006) 144 Pazopanib (inhibitor of VEGFR, PDGFR and c kit; inhibits cross-talk between HER2 and VEGFR pathways) Phase II Slamon et al. (2008) 174 Sunitinib (inhibitor of VEGFR, PDGFR, c Kit, RET, FLT3, and CSF-1R) Phase II Burstein et al. (2008) 175 *In clinical trials at phase I II or higher. As adjuvant and for patients with advanced-stage disease. Two negative phase III trials in HER2-negative disease. Abbreviations: AS15, antigenspecific cancer immunotherapy; BC, breast cancer; CAR, coxsackievirus and adenovirus receptor; CSF-1R, macrophage colony-stimulating factor 1 receptor; FLT3, Fms-like tyrosine kinase 3; GM CSF, granulocyte macrophage-colony stimulating factor; HSP90, heat-shock protein 90; IGF 1R, insulin-like growth factor 1 receptor; mab, monoclonal antibody; TKI, tyrosine kinase inhibitor. respectively, developed symptomatic CHF. However, the data on cardiac safety with novel anti-her2 agents need to be interpreted with caution because the trials are conducted in carefully selected populations of patients who tolerated prior trastuzumab treatment. Selective HER1 or HER3 inhibition Preclinical data indicate that overexpression of HER2 in breast cancer is frequently associated with over expression of HER1, and that inhibition of HER1 enhances the response to trastuzumab in HER1 HER2 co-expressing cells. 47,97 The potential utility of simultaneous HER1 and HER2 inhibition is supported by the positive findings of lapatinib and pertuzumab trials. However, despite these observations, clinical activity of selective HER1 inhibitors in patients with breast cancer has been disappointing, either as single agents, 98,99 or in combination with chemotherapy (in patients unselected for HER2 status), 100,101 or in combination with trastuzumab in patients with HER2-positive breast cancer. 102,103 As a result, attention has shifted to other members of the HER family, particularly HER3. Although HER3 has only weak intrinsic tyrosine kinase activity, 104 HER2 HER3 heterodimers form the most potent mitogenic signaling pair in the HER family, 105 and HER3 is now recognized as having a critical role as a co-receptor for amplified HER Accordingly, HER3 targeting agents are now in development, including several antibodies (Table 5). Novel tyrosine kinase inhibitors New tyrosine kinase inhibitors (TKIs) in development for patients with HER2-positive breast cancer include irreversible TKIs, and TKIs with a broader spectrum of activity than lapatinib (Table 5). Irreversible inhibitors have been shown to be more potent and to prolong target inhibition compared with lapatinib, 107 as well as potentially bypassing pathways involved in resistance to HER2-targeting agents. Neratinib is the most advanced irreversible EGFR HER2 TKI in development for breast cancer. A phase II study of neratinib in 136 patients with HER2-positive metastatic breast cancer showed a 24% response rate in women previously treated with trastuzumab, and a 56% response rate in trastuzumab-naive patients. PFS at 16 weeks was 59% and 78%, respectively results that compare favorably with other single-agent anti-her2 therapies. 108 No grade 3 or 4 cardio toxicity related to neratinib was reported, but grade 3 and 4 diarrhea was the most frequently occurring adverse effect. Neratinib is now being studied in various combinations and in head-to-head comparisons with trastuzumab, lapatinib and new targeted agents. A phase III trial of adjuvant neratinib has also started (Table 6). Inhibition of the PI3K pathway The PI3K family is complex, consisting of multiple members, divided into three main classes. 109 Class IA PI3Ks are activated by growth factors via tyrosine kinase receptors (including HER family members) and are most clearly involved in malignant diseases. Deregulation of this pathway is thought to be a cause of resistance to HER2-targeted therapies, as well as resistance to cytotoxics and hormonal therapies PI3K pathway inhibition would be expected to restore sensitivity to trastuzumab and/or lapatinib in patients with HER2-positive breast cancer, as well as being inherently antiproliferative and proapoptotic. However, multiple PI3K isoforms are expressed in the heart, where they are involved in hypertrophy and cardiac failure, 113 so PI3K inhibitors have the potential to exacerbate the cardiac toxicity of HER2-targeted agents. PI3K also has an important role in cellular responses to insulin, and inhibition of PI3K (particularly the p110α catalytic isoform) can potentially cause insulin resistance. Although this mechanism has not been a major problem in clinical studies so far, hyperglycemia has been observed in phase I studies Activation of PI3K results in stimulation of Akt and mtor kinases, leading to the promotion of tumor cell proliferation and survival. Inhibitors of the PI3K pathway can be categorized into four main groups: PI3K inhibitors (isoform-specific or pan-class I PI3K inhibitors), dual PI3K mtor inhibitors, Akt inhibitors, and mtor inhibitors. All seem to have only modest antitumor activity when used alone in patients with breast cancer and are likely to need co-administration of other agents for optimal activity. 120 GDC is one of several pan-class I PI3K inhibitors currently in phase I II clinical development (Table 5). In preclinical experiments, GDC 0941 was efficacious in trastuzumab-resistant cells, 111 was able to suppress the growth of >70% of breast cancer cell lines when used in combination with trastuzumab, pertuzumab or lapatinib, 110 and was able to render HER2-amplified cells and tumor xenografts more sensitive to docetaxel. 110 In phase I, there was evidence of antitumor activity in three NATURE REVIEWS CLINICAL ONCOLOGY VOLUME 9 JANUARY