REVIEW ARTICLE Trastuzumab cardiotoxicity: from clinical trials to experimental studies

Transcription

1 British Journal of Pharmacology British Journal of Pharmacology (2017) Themed Section: New Insights into Cardiotoxicity Caused by Chemotherapeutic Agents REVIEW ARTICLE Trastuzumab cardiotoxicity: from clinical trials to experimental studies Correspondence Pal Pacher MD, PhD, FAHA, FACC or Balazs Nemeth MD, Laboratory of Cardiovascular Physiology and Tissue Injury, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, 5625 Fishers Lane, MSC-9413, room 2N-17, Rockville, Maryland 20852, USA. Received 10 August 2016; Revised 21 September 2016; Accepted 24 September 2016 Balazs T Nemeth 1, Zoltan V Varga 1,WenJinWu 2 and Pal Pacher 1 1 Laboratory of Cardiovascular Physiology and Tissue Injury, National Institute on Alcohol Abuse and Alcoholism, Rockville, MD, USA, and 2 Division of Biotechnology Research and Review 1, Office of Biotechnology Products, Office of Pharmaceutical Quality, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Bethesda, MD, USA Epidermal growth factor receptor-2 (HER-2) is overexpressed in 20 to 25% of human breast cancers, which is associated with aggressive tumour growth and poor prognosis. Trastuzumab (Herceptin ) is a humanized monoclonal antibody directed against HER-2, the first highly selective form of therapy targeting HER-2 overexpressing tumours. Although initial trials indicated high efficacy and a favourable safety profile of the drug, the first large, randomized trial prompted a retrospective analysis of cardiac dysfunction in earlier trials utilizing trastuzumab. There has been ongoing debate on the cardiac safety of trastuzumab ever since, initiating numerous clinical and preclinical investigations to better understand the background of trastuzumab cardiotoxicity and evaluate its effects on patient morbidity. Here, we have given a comprehensive overview of our current knowledge on the cardiotoxicity of trastuzumab, primarily focusing on data from clinical trials and highlighting the main molecular mechanisms proposed. LINKED ARTICLES This article is part of a themed section on New Insights into Cardiotoxicity Caused by Chemotherapeutic Agents. To view the other articles in this section visit Abbreviations BCIRG, Breast Cancer International Research Group; CD, cardiac dysfunction; CHF, congestive heart failure; CI, confidence interval; CREC, Cardiac Review and Evaluation Committee; ErbB2, erythroblastic leukaemia viral oncogene homolog 2; FDA, Food and Drug Administration; FinHer, Finland Herceptin;HERA,HerceptinAdjuvant;HER-2,humanepidermal growth factor receptor-2; LVEF, left ventricular ejection fraction; mab, monoclonal antibody; MBC, metastatic breast cancer; NRG, neuregulin; RR, risk ratio Published This article is a U.S. Government work and is in the public domain in the USA. DOI: /bph.13643

2 B T Nemeth et al. Tables of Links TARGETS Enzymes a Akt (Protein kinase B) Endothelial NOS (NOS3) LIGANDS Anastrozole Capecitabine Cisplatin Cyclophosphamide Docetaxel Doxorubicin EGF Protein kinase G (PKG) SERCA2 (Ca 2+ -ATPases) Lapatinib Neuregulin-1 Paclitaxel Pertuzumab Trastuzumab Trastuzumab emtansine Vinorelbine Catalytic receptors b MAP kinases PI3K family HER2 HER4 These Tables list key protein targets and ligands in this article that are hyperlinked to corresponding entries in the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Southan et al., 2016), and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 ( a,b Alexander et al., 2015a,b). Introduction Trastuzumab (Herceptin ) is a humanized monoclonal antibody (mab) against the erythroblastic leukaemia viral oncogene homolog 2 (ErbB2), also known as the human epidermal growth factor receptor-2 (HER-2). Although this transmembrane receptor tyrosine kinase plays an important role in the normal growth and development of various cell types, its overexpression in 20 to 25% of human breast cancers is associated with aggressive growth and poor prognosis (Slamon et al., 1987). Trastuzumab became the first clinically used mab against HER-2. The drug blocks HER-2 signalling by multiple mechanisms, including disruption of normal ErbB signal transduction via physically disabling HER-2 molecules to form dimers with each other or further members of the HER/ErbB family (Figure 1). Being a highly selective therapeutic option, it promised a positive outlook for HER-2 positive breast cancer patients to finally be treated with a molecule generating fewer or none of the adverse effects associated with conventional chemotherapeutics. Although early phase II trials (Pegram et al., 1998; Baselga et al., 1999) indicated high efficacy and a favourable safety profile of trastuzumab, an unexpectedly high rate of adverse cardiac events during the first phase III trial (Slamon et al., 2001)promptedaretrospectiveanalysisofallcasesofcardiac dysfunction (CD) in earlier trials utilizing trastuzumab (Seidman et al., 2002) (Tables 1 and 3). It was noted during this analysis, however, that most of the patients analysed in that cohort had received earlier anthracycline therapy (Seidman et al., 2002) and that the observed CD in response to trastuzumab resembled the cardiomyopathy associated with previously described anthracycline cardiotoxicity (Singal and Iliskovic, 1998). Since then, there has been an ongoing debate on the cardiac safety of trastuzumab, initiating clinical and preclinical trials to better understand the background of trastuzumab cardiotoxicity and evaluate its effects on patient morbidity. Here, we have attempted to provide a comprehensive overview of our current knowledge of this effect, focusing primarily on data from clinical trials. We also wish to highlight the major molecular mechanisms uncovered hitherto in the background of trastuzumab cardiotoxicity. Two decades of clinical experience Historical background: the road to clinical use Trastuzumab is a humanized mab against the extracellular domain of HER-2/ErbB2, a member of the HER/ErbB group of transmembrane receptors with intracellular tyrosine kinase activity. After its discovery in 1985 (Coussens et al., 1985) and demonstration of its presence in human mammary carcinoma (King et al., 1985), HER-2 rapidly became the focus of intensive investigations. Its overexpression was associated with poorly differentiated, aggressively growing tumours with a high metastatic potential, comprising approximately 20 25% of all breast carcinomas (Slamon et al., 1987). MAbs against HER-2 proved to effectively inhibit proliferation of such breast cancer cells (Hudziak et al., 1989). Following humanization of the murine anti-her-2 mab 4D5 (Carter et al., 1992), this new type of anticancer drug was ready to be tested in the clinical practice (Figure 2). The initial, groundbreaking trials that supported the approval of trastuzumab by the Food and Drug Administration (FDA) for metastatic breast cancer (MBC) did not report any significant cardiotoxic effect of the drug (Pegram et al., 1998; Baselga et al., 1999; Shak, 1999; Baselga, 2001). Trastuzumab was found to be well tolerated in all doses from mg i.v. both alone and in combination with cisplatin in phase I trials (Shak, 1999). This was shortly followed by phase II trials, providing proof-of-principle that the agent might offer major benefit for some patients. The first such trial of trastuzumab used a loading dose of 250 mg followed by 100 mg every week for 10 weeks, and involved 46 women with immunohistochemically proven HER-2 positive MBC (Table 1., (Baselga et al., 1999)). A total of 39 patients with similar disease status were enrolled in a combination study with cisplatin (Pegram et al., 1998), where the same dose of trastuzumab was administered for 9 weeks, and gave an overall response rate of 11.6 and 24.3% respectively (Table 3). Patients in these trials could continue until toxic effects or disease progression occurred. None of these trials reported serious cardiac events that could be related to trastuzumab. Based on this safe and favourable preliminary efficacy, clinical trials of trastuzumab involving large numbers of patients were warranted British Journal of Pharmacology (2017)

3 Trastuzumab cardiotoxicity BJP Figure 1 Cardiotoxicity of doxorubicin and anti-her-2 in breast cancer. Anti-HER-2 results in diminished pro-survival signalling in cardiomyocytes. Gene expression, ATP- and nitric oxide production favouring survival and proliferation are attenuated. In parallel, without appropriate counterbalancing, pro-apoptotic and necrosis-inducing processes such as cytochrome-c release from mitochondria and an increase in intracellular calcium concentration can facilitate cell death. The addition of doxorubicin to the system boosts oxidative and nitrosative stress via increased mitochondrial and iron-dependent generation of superoxide, which then is transformed into hydrogen peroxide and peroxynitrite. These reactive oxygen and nitrogen species are capable of inducing oxidative DNA damage, further aggravating the imbalance between pro-survival and pro-apoptotic/necrotic signalling. Green lines represent pro-survival mechanisms, while red lines show pro-apoptotic/necrotic processes. Dashed lines indicate an indirect connection. Akt protein kinase B; Cyt-C cytochrome-c; Dox doxorubicin; MAPK mitogen activated protein kinase; NO nitric oxide; NOS3 endothelial nitric oxide synthase; NRG-1 neuregulin-1; ONOO peroxynitrite; PI3K - phosphatidylinositol-3-kinase; PKG protein kinase G; Pln phospholamban; SERCA2 sarcoplasmic/endoplasmic reticulum Ca 2+ -ATPase 2; SOD superoxide dismutase. The landmark phase II, multicentre trial conducted by Cobleigh et al., the first to utilize administration of trastuzumab corrected for body weight (Table 3), had similar response rates described in previous smaller trials (Cobleigh et al., 1999). This trial unveiled, however, that the extent of HER-2 positivity (grade 3 vs. grade 2) had a significant effect on time to disease progression (3.3 vs. 1.9 months respectively) and that patients with grade 3 positivity tended to have a higher response rate (18 vs. 6%, P = 0.06) to trastuzumab therapy. Dose escalation did not increase response rate in a trial in which 114 patients with MBC were administered either a loading dose of 4 mg kg 1 followed by 2 mg kg 1 trastuzumab each week or twice of this dose at the same frequency (Vogel et al., 2001). Due to the long serum half-life of trastuzumab, however, higher doses of the drug allow infrequent dosing. Specifically, when given in a loading dose of 8 mg kg 1 followed by 6 mg kg 1 every third week, trastuzumab showed similar serum levels to those seen in earlier trials with weekly dosing (Leyland-Jones et al., 2003). The pivotal, randomized, multicentre, phase III trial [Table 1, (Slamon et al., 2001)] conducted to compare chemotherapy in combination with trastuzumab versus chemotherapy alone in patients with HER-2 positive MBC showed that addition of the antibody to standard chemotherapy significantly increased time to disease progression (7.4 vs. 4.6 months), objective response rate (50 vs. 32%) and duration of response (9.1 vs. 6.1 months). Trastuzumab also decreased death rate at 1 year (22 vs. 33%) and reduced the risk of death by 20%. The unexpectedly high number of cardiac adverse events during these trials, however, triggered a retrospective investigation of all clinical trials that had been using trastuzumab by an independent Cardiac Review and Evaluation Committee (CREC, Tables 1 and 3), based on concerns with the cardiac safety of trastuzumab (Seidman et al., 2002). The CREC defined CD as one of the following: (1) cardiomyopathy British Journal of Pharmacology (2017)

4 B T Nemeth et al. Table 1 Randomized, phase III clinical trials with trastuzumab Study Phase Disease stage Enrolled patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Slamon et al., 2001 III MBC 143 Unspecified until PD Piccart-Gebhart et al., 2005 HERA Tan-Chiu et al., 2005 B31 Joensuu et al., 2006 FinHer 3 weeks (6 cycles) 27% (according to CREC criteria) 135 None As above 8% (according to CREC criteria) III I-III 1694 Anthracyclines in 94% of patients 91 Anthracyclines 4 mg kg 1 loading, until PD Paclitaxel 175 mg m 2 i.v. every 3 weeks (6 cycles) 13% (according to CREC criteria) 95 None As above 1% (according to CREC criteria) 8mg kg 1 loading, 6mg kg 1 every 3 weeks for 1 year None 2% CHF 7% (LVEF reduction >10% to 1693 None None <1% CHF2% (LVEF reduction >10% to III II-III 850 Unspecified 4 mg kg 1 loading, 2mg kg 1 starting with paclitaxel weekly for 1 year 3 weeks (4 cycles), then paclitaxel 175 mg m 2 i.v. every 3 weeks (4 cycles) 4% CHF 5% (asymptomatic LVEF drop; a total 19% discontinued trastuzumab due to LVEF drops) 814 None As above 1% CHF 1% (asymptomatic LVEF drop) III IB-III 390 Unspecified None (HER-2 negative) Docetaxel 100 mg m 2 i.v. every 3 weeks (3 cycles) + fluorouracil 600 mg m 2 i.v., epirubicin 60 mg m 2 i.v., cyclophosphamide 3 weeks (3 cycles) 388 Vinorelbine 25 mg m 2 0.3% CHF 0.3% (LVEF reduction >10% to i.v. 3 weeks on, 1 week off (3 cycles) + as above 58 None (HER-2 positive) Docetaxel + FEC as above 58 None (HER-2 positive) Vinorelbine + FEC as above 54 Docetaxel + FEC as above 0% 1 62 Vinorelbine + FEC as above (9 cycles) continues 3730 British Journal of Pharmacology (2017)

5 Trastuzumab cardiotoxicity BJP Table 1 (Continued) Study Phase Disease stage Enrolled patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Perez et al., 2008 N9831 Kaufman et al., 2009 TAnDEM Gianni et al., 2010 NOAH Slamon et al., 2011 BCIRG-006 III I-III 664 No anthracyclines or taxanes mg kg 1 loading, for 1 year following paclitaxel 570 As above, concurrently with paclitaxel III MBC 103 Anthracyclines in 46% of patients 104 Anthracyclines in 53% of patients III III 117 Neoadjuvant design, no prior None until PD 600 mg m 2 I.V. every 3 weeks (4 cycles), then paclitaxel 80 mg m 2 i.v. weekly (12 cycles) 0.3% (CHF or cardiac related death) As above 2.8% (CHF or cardiac related death) As above 3.3% (CHF or cardiac related death) Anastrozole 1 mg/day orally until PD none As above 0% CHF 8mg kg 1 loading, 6mg kg 1 every 3 weeks (10 cycles); after surgery, continued to reach 1year i.v. + paclitaxel 150 mg m 2 i.v. every 3 weeks (3 cycles), then paclitaxel 175 mg m 2 i.v. every 3 weeks (4 cycles), then cyclophosphamide 600 mg m 2 i.v. + fluorouracil 600 mg m 2 i.v. + methotrexate 40 mg m 2 i.v., 2 weeks on, 2 weeks off (3 cycles) 1% CHF 4% (NYHA I, asymptomatic CHF) 2% CHF 2% (LVEF reduction >20%, or to 23% (LVEF reduction >10%, but <20%) 118 None As above 0% CHF 1% (LVEF reduction >20%, or to 16% (LVEF reduction >10%, but <20%) 99 None (HER2 negative) 0% CHF 1% (LVEF reduction >20%, or to 13% (LVEF reduction >10%, but <20%) III I-III 1074 No chemotherapy 4 mg kg 1 loading, with docetaxel, then 6mg kg 1 every 3 weeks to reach 1 year 3 weeks (4 cycles) then docetaxel 100 mg m 2 2% CHF 19% (relative LVEF reduction >10%) continues British Journal of Pharmacology (2017)

6 B T Nemeth et al. Table 1 (Continued) Study Phase Disease stage Enrolled patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Slamon et al., 2011 BCIRG-006 (continued) Valero et al., 2011 BCIRG-007 i.v. every 2 weeks (4 cycles) 1073 None As above 0.7% CHF 11% (relative LVEF reduction >10%) mg kg 1 loading, with docetaxel, then 6 mg kg 1 III MBC 131 Anthracyclines in 33% of patients every 3 weeks to reach 1year (24 cycles), then 6 mg kg 1 every 3 weeks until PD Docetaxel 75 mg m 2 i.v. + carboplatin AUC = 6 mg ml 1 i.v. every 3 weeks (6 cycles) Docetaxel 100 mg m 2 i.v. every 3 weeks (8 cycles) 131 Docetaxel 75 mg m 2 i.v. + carboplatin AUC = 6 mg ml 1 0.4% CHF 9% (relative LVEF reduction >10%) 1% SCD, 1% CHF, 6% (LVEF reduction >15%) 7% (LVEF reduction >15%) Baselga et al., 2012b CLEOPATRA Baselga et al. 2012a NeoALTTO III MBC 406 Anthracyclines in 40% of patients 8mg kg 1 loading, 6mg kg 1 every 3 weeks until PD i.v. every 3 weeks (8 cycles) Docetaxel 75 mg m 2 i.v. every 3 weeks (6 cycles) 402 As above Pertuzumab 840 mg loading, 420 mg until PD + docetaxel 75 mg m 2 i.v. every 3 weeks (6 cycles) III II-III 154 Neoadjuvant design, no prior mg kg 1 loading, (18 cycles), then 8 mg kg 1 None Lapatinib 1500 mg kg 1 loading, 6 mg kg 1 every 3 weeks for 34 weeks after FEC p.o. daily for 18 weeks + as below, then continued after FEC for 34 weeks Paclitaxel 80 mg m 2 i.v. weekly following trastuzumab or lapatinib (12 cycles); after surgery cyclophosphamide 500 mg m 2 i.v. + fluorouracil 500 mg m 2 i.v. + epirubicin 100 mg m 2 i.v. every 3 weeks (3 cycles) 152 As above Lapatinib 1000 mg kg 1 p.o. daily for 18 weeks + as above, then continued after FEC for 34 weeks 3% CHF 5% (LVEF drop of grade 1 2) 1% CHF 4% (LVEF drop of grade 1 2) 1% (based on NCI-CTCAE grades) 1% (based on NCI-CTCAE grades) 3% (based on NCI-CTCAE grades) continues 3732 British Journal of Pharmacology (2017)

7 Trastuzumab cardiotoxicity BJP Table 1 (Continued) Study Phase Disease stage Enrolled patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Robidoux et al., 2013 B41 Pivot et al., 2013 PHARE Krop et al., 2014 TH3RESA III II-IIIA 174 Neoadjuvant design, no prior III I-III % AC, 100% trastuzumab up to 6 months with paclitaxel; after surgery, 6mg kg 1 every 3 weeks to reach 1year 3 weeks (4 cycles), then paclitaxel 80 mg m 2 i.v. 3 weeks on, 1 week off (4 cycles) + lapatinib 1000 mg daily p.o. 1% CHF 181 As above As above, without lapatinib 4% CHF 174 None; after surgery, 8mg kg 1 loading, 6mg kg 1 every 3 weeks to reach 1year 8mg kg 1 loading, 6mg kg 1 every 3 weeks for 1 year 3 weeks (4 cycles), then paclitaxel 80 mg m 2 i.v. 3 weeks on, 1 week off (4 cycles) + lapatinib 1500 mg daily p.o. 4% CHF None 6% (CHF, or LVEF reduction to below 50%, or >15% if above 50%, or >10% if 1693 As above for 6 months None 2% (as above) III MBC 404 Trastuzumab 100%, otherwise unspecified 198 Physician s choice: -chemotherapy (any single agent); -hormonal therapy for hormonereceptor-positive disease (single-agent or dual therapy); -HER2-directed therapy (single-agent, dual HER2-targeted therapy, or combination with either of the above) None Ado-trastuzumab emtansine 3.6 mg kg 1 every 3 weeks until PD 1% (LVEF reduction >15% to 1% (LVEF reduction >15% to Italics mark concurrent anthracycline and trastuzumab dosing. AUC area under the plasma concentration-time curve; FEC fluorouracil, epirubicin, and cyclophosphamide-containing chemotherapeutic regimen; MBC metastatic breast cancer; NCI-CTCAE National Cancer Institute Common Toxicity Criteria for Adverse Events; NYHA New York Heart Association classification; PD progressive disease. British Journal of Pharmacology (2017)

8 B T Nemeth et al. Figure 2 Timeline of development of anti-her-2. Discovered in 1985, HER-2 rapidly emerged into the centre of focus after its linkage to poor clinical outcomes in breast cancer patients was shown. Proliferation of breast cancer cells was successfully decreased using mouse mabs developed against HER-2, and humanization of the most active 4D5 enabled it to enter the clinical stage. Despite the unexpected adverse cardiac effect of trastuzumab, the drug was approved by the FDA to treat metastatic breast cancer in Since then, anti-her-2 options have broadened with the approval of lapatinib and pertuzumab in 2007 and 2012 respectively. FDA Food and Drug Administration; mab monoclonal antibody. characterized by a decrease in cardiac left ventricular ejection fraction (LVEF) that was either global or more severe in the septum; (2) symptoms of congestive heart failure (CHF); (3) associated signs of CHF, including but not limited to S3 gallop, tachycardia or both; and (4) decline in LVEF of at least 5 to less than 55% with accompanying signs or symptoms of CHF or a decline in LVEF of at least 10% to below 55% without accompanying signs or symptoms (Seidman et al., 2002). Of the 1219 treated patients (14% overall, range: 1 27%) analysed in the CREC report, 174 met at least one of the above criteria. Sixty-two (5% overall, range 1 16%) of these patients had New York Heart Association Functional Class III or IV CHF. Another important conclusion of this retrospective analysis was that all patients had been or were exposed to anthracyclines. Incidence rates were found to be greatest when trastuzumab and anthracyclines were used concomitantly (Tables 1 3) and diagnosed CD events began to increase at approximately 300 mg m 2 of cumulative doxorubicin dose (Seidman et al., 2002). The authors also concluded, however, that there might have been biased over-reporting of cardiac events in the trastuzumabcontaining arms of the trial; therefore, cautious interpretation of their results is necessary. The recognition that trastuzumab might have a significant adverse cardiac effect changed the design of subsequent trials. Concomitant administration of trastuzumab and anthracyclines in large clinical trials was replaced by sequential dosage, and patients with decreased baseline LVEF were ineligible for most of the trials initiated after the CREC report. Cardiac-related adverse events were prospectively defined and also advanced into becoming major dose-limiting toxicity criteria. Patients exposed to trastuzumab were carefully monitored not only for CHF but also for asymptomatic falls in LVEF in all studies. Temporary or permanent trastuzumab discontinuation and protocols of dose limitation based on cardiac events have also been incorporated into most trials conducted since. Tables 1 4 contain an overview of clinical trials with trastuzumab, with a focus on the drugs used and the consequential CD events (as defined in the appropriate trial). Trastuzumab in combination with chemotherapeutic agents Anthracyclines. Anthracyclines, despite their well-known cardiotoxicity, are widely used due to being among the most active chemotherapeutic agents in breast cancer. Most of the clinical trials in which trastuzumab and anthracyclines were concurrently administered (Bianchi et al., 2003; Untch et al., 2004, 2010; Buzdar et al., 2005; Gennari et al., 2009), reported even higher percentage (range: 7 75%) of any type of CD than Seidman et al. in their retrospective analysis (Table 2). Regardless of the criteria used to determine asymptomatic CD in these studies, there is a clear dependency of CD incidence on anthracycline dose, in the observed populations (Table 2). Investigators of the randomized NOAH trial (Gianni et al., 2010) reported very low occurrence of CHF in their patients despite the similar dose and timing of anthracyclines employed. It is noteworthy, however, that these patients did not receive anthracyclines earlier (Table 1). Another, single arm trial (Venturini et al., 2006), on the other hand, had to be prematurely terminated due to reaching five cases (11%) of symptomatic CHF. In this trial, five additional patients (11%) experienced asymptomatic LVEF drops below an absolute LVEF of 45%. CHF incidence, however, remained less than 8% in the rest of the trials, most of the investigators reporting no such events. Encapsulation of doxorubicin in liposomes was introduced in order to decrease cardiotoxicity of the drug. A randomized, phase III trial with pegylated liposomal doxorubicin (O Brien et al., 2004) showed that encapsulation of doxorubicin did not alter its efficacy, but significantly reduced CD events. Trials utilizing either pegylated (Chia et al., 2006; Christodoulou et al., 2009; Martin et al., 2011; Rayson et al., 2012) or non-pegylated (Venturini et al., 2010; Amadori et al., 2011; Anton et al., 2011) liposomal doxorubicin concurrently with trastuzumab showed similar results (Table 2). However, a recent phase III trial with non-pegylated liposomal doxorubicin (Baselga et al., 2014) reported that there was no significant improvement in progression free survival with the addition of this agent to the standard 3734 British Journal of Pharmacology (2017)

9 Trastuzumab cardiotoxicity BJP Table 2 Clinical trials of trastuzumab with anthracyclines Study Phase Disease stage Number of patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Burstein et al., 2003a Bianchi et al., 2003 Untch et al., 2004 Buzdar et al., 2005 Venturini et al., 2006 Chia et al., 2006 Kelly et al., 2006 II II-III 40 Neoadjuvant design, no prior (12 cycles) II IIIB/IV 16 Unspecified until PD or 1 year 16 As above, starting with paclitaxel I MBC 26 Unspecified until PD 25 until PD Paclitaxel 80 mg m 2 i.v. every 3 weeks (4 cycles); after surgery, doxorubicin 60 mg m 2 i.v. + cyclophosphamide 3 weeks (4 cycles) i.v. + paclitaxel 150 mg m 2 i.v. every 3 weeks (3 cycles); then paclitaxel 80 mg m 2 i.v. weekly (9 cycles) 28% (LVEF reduction >10%) 75% (LVEF reduction >10% to below 55% or LVEF reduction of any extent to As above 33% (as above) Epirubicin 60 mg m 2 i.v. + cyclophosphamide 600 mg m 2 i.v. every 3 weeks (6 cycles) Epirubicin 90 mg m 2 i.v. + cyclophosphamide 600 mg m 2 i.v. every 3 weeks (4 6 cycles) 4% CHF 48% (LVEF reduction >10%) 8% CHF 56% (LVEF reduction >10%) 23 None As above 24% (LVEF reduction >10%) II II-IIIA 23 Neoadjuvant design, no prior (24 cycles) Paclitaxel 225 mg m 2 i.v. every 3 weeks (4 cycles); after surgery, epirubicin 75 mg m 2 i.v. + cyclophosphamide 500 mg m 2 i.v. + fluorouracil 500 mg m 2 i.v. twice every 3 weeks (4 cycles) 30% (LVEF reduction >10%) 19 None As above 26% (LVEF reduction >10%) II MBC 45 No anthracyclines or taxanes II MBC 30 Anthracyclines in 43% of patients II II-IV 52 Neoadjuvant design, no prior until PD (24 cycles) (12 cycles); after surgery, 8 mg kg 1 Doxorubicin 75 mg m 2 i.v. + docetaxel 75 mg m 2 i.v. every 3 weeks (8 cycles) Pegylated liposomal doxorubicin 50 mg m 2 i.v. every 4 weeks (6 cycles) 3 weeks (4 cycles), then 11% CHF 11% (LVEF reduction to below 45%) 10% (LVEF reduction >15%) 4% CHF 21% (LVEF reduction to continues British Journal of Pharmacology (2017)

10 B T Nemeth et al. Table 2 (Continued) Study Phase Disease stage Number of patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Kelly et al., 2006 (continued) Dang et al., 2008 Gennari et al., 2009 Cortes et al., 2009 Untch et al., 2010 HERCULES Venturini et al., 2010 Anton et al., 2011 GEICAM Martin et al., 2011 GEICAM II I-IV 70 Unspecified, but all received chemotherapy II MBC 45 Anthracyclines in 51% of patients loading, 6 mg kg 1 every 3 week to complete 1 year with paclitaxel, then 6 mg kg 1 every 3 weeks (12 cycles) 2mg kg 1 loading, 1 mg kg 1 weekly for 1 year II LABC MBC 54 Unspecified 4 mg kg 1 loading, for 1 year following paclitaxel II MBC 60 22% of patients received chemotherapy (unspecified) until PD paclitaxel 90 mg m 2 i.v. weekly (12 cycles) 2 weeks (4 cycles) then paclitaxel 175 mg m 2 i.v. every 2 weeks (4 cycles) Epirubicin 90 mg m 2 i.v. every 3 weeks (6 8 cycles) Nonpegylated liposomal doxorubicin 50 mg m 2 i.v. every 3 weeks (6 cycles) + paclitaxel 80 mg m 2 i.v. weekly for up to 1 year Epirubicin 60 mg m 2 3 weeks (6 cycles) 60 33%, as above As above Epirubicin 90 mg m mg m 2 i.v. every 3 weeks (6 cycles) 60 28%, as above None As above 0% II MBC 31 94% of patients received chemotherapy (unspecified) II II-IIIA 59 Neoadjuvant design, no prior II MBC 48 Anthracyclines in 48% of patients for 1 year (24 cycles) (24 cycles) Nonpegylated liposomal doxorubicin 50 mg m 2 i.v. + docetaxel 75 mg m 2 i.v. every 3 weeks (8 cycles) Liposomal doxorubicin 50 mg m 2 i.v. + docetaxel 60 mg m 2 i.v. every 3 weeks (6 cycles) Pegylated liposomal doxorubicin 50 mg m 2 4 weeks (6 cycles) 1.4% CHF 4% (LVEF reduction >15% or to below 55%) 4.5% CHF 14% (LVEF reduction >15%) 15% (LVEF reduction >10% to below 50% or >20%) 7% (LVEF reduction >10% to 2% DLC 15% (LVEF reduction >10% to 5% DLC 3% CHF 7% (LVEF reduction >20% to 45% or below) 9% (LVEF reduction >20% or below LLN) 18% (LVEF reduction >10% but <20%) 17% (LVEF reduction >13%) continues 3736 British Journal of Pharmacology (2017)

11 Trastuzumab cardiotoxicity BJP Table 2 (Continued) Study Phase Disease stage Number of patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Amadori et al., 2011 Untch et al., 2011 TECHNO Rayson et al., 2012 Aogi et al., 2013 HER2NAT II LABC, MBC 45 Anthracyclines in 31% of patients II II-III 217 Neoadjuvant design, no prior II I-III 60 No prior (exclusion criterion) (24 cycles) 8mg kg 1 loading, 6mg kg 1 every 3 weeks (4 cycles) with paclitaxel; after surgery, continue to complete 1 year starting with paclitaxel, for 1 year 120 for 1 year II IIIB-IV 38 Neoadjuvant design, no prior with docetaxel (12 cycles); after surgery either continue, or 6 mg kg 1 every 3 weeks to complete 1year Nonpegylated liposomal doxorubicin 50 mg m 2 i.v. every 3 weeks + docetaxel 30 mg m 2 i.v. 2 weeks on, 1 week off (8 cycles) Epirubicin 90 mg m 2 3 weeks (4 cycles) + paclitaxel 175 mg m 2 i.v. every 3 weeks (4 cycles) 3 weeks (4 cycles), then paclitaxel 80 mg m 2 i.v. weekly (12 cycles) Pegylated liposomal doxorubicin 50 mg m 2 3 weeks (4 cycles), then as above Epirubicin 90 mg m 2 3 weeks (4 cycles); after surgery docetaxel 75 mg m 2 i.v. every 3 weeks (4 cycles) 8% (LVEF reduction >15% or to 0.5% CHF 3.5% (based on NCI-CTC 2.0 grades) 19% (LVEF reduction >10% to 4% (LVEF reduction >10% to 8% (LVEF reduction >10%) Italics mark concurrent anthracycline and trastuzumab dosing. DLC dose limiting cardiotoxicity; LABC locally advanced, non-operable breast cancer; LLN lower limit of normal; MBC metastatic breast cancer; NCI-CTC 2.0 National Cancer Institute Common Toxicity Criteria version 2.0; PD progressive disease. British Journal of Pharmacology (2017)

12 B T Nemeth et al. Table 3 Clinical trials of trastuzumab in combination with other chemotherapeutics Study Phase Disease stage Number of patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Pegram et al., 1998 Baselga et al., 1999 Cobleigh et al., 1999 Vogel et al., 2002 Esteva et al., Jahanzeb et al., 2002 Burstein et al., 2003b Leyland-Jones et al., 2003 Gori et al., 2004 Pegram et al., 2004 O Shaughnessy et al., 2004 II MBC 39 Unspecified 250 mg loading, 100 mg weekly (9 cycles) II MBC 46 98% received chemotherapy, otherwise unspecified II MBC 213 Anthracyclines in 94% of patients II MBC 114 Anthracyclines in 50% of patients II MBC 30 Anthracycline in 80% of patients (before metastasis) 250 mg loading, 100 mg weekly (10 cycles) 4 or 8 mg kg 1 loading, 2 or 4 mg kg 1 weekly until PD 2mg kg 1 3 weeks on, 1 week off until PD II MBC 40 Unspecified 4 mg kg 1 loading, until PD II MBC 54 Anthracyclines in 52% of patients II MBC 32 Anthracyclines in 70% of patients II MBC 25 Anthracyclines and taxanes in all cases II IIIB/IV 62(BCIRG-101) Anthracyclines in 34% of patients 62(UCLA-ORN) Anthracyclines in 45%, taxanes in 15% of patients II MBC 64 Anthracyclines in 95% of patients until AE 8mg kg 1 loading, 6mg kg 1 every 3weeks for 6 months for 1 year Cisplatin 75 mg m 2 i.v. every 4 weeks (3 cycles) 3% (according to CREC criteria) None 7% (according to CREC criteria) None 5% (according to CREC criteria) None 3% (according to CREC criteria) Docetaxel 35 mg m 2 i.v., 3 weeks on, 1 week off until PD Vinorelbine 30 mg m 2 i.v. weekly until PD Vinorelbine 25 mg m 2 i.v. weekly until AE Paclitaxel 175 mg m 2 i.v. every 3 weeks (8 cycles) Paclitaxel mg m 2 i.v. weekly for 6 months Docetaxel 75 mg m 2 i.v. + cisplatin 75 mg m 2 i.v. every 3 weeks (6 cycles) As above Docetaxel 75 mg m 2 i.v. + carboplatin AUC = 6 mg ml 1 i.v. every 3 weeks (6 cycles) Gemcitabine 1200 mg m 2 i.v. 2 weeks on, 1 week off until PD 10% (LVEF reduction to below 50%) 3% (LVEF reduction >15%) 4% CHF 15% (based on NCI grades) 6% CHF 25% (LVEF reduction >15%) 8% CHF 2% CHF 2% CHF No overt CHF continues 3738 British Journal of Pharmacology (2017)

13 Trastuzumab cardiotoxicity BJP Table 3 (Continued) Study Phase Disease stage Number of patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Marty et al., 2005 Coudert et al., 2006 Coudert et al., 2007 GETN(A)-1 Yamamoto et al., 2008 Wardley et al., 2010 CHAT Tolaney et al., 2015 II MBC 92 Anthracyclines in 64% of patients 94 Anthracyclines in 55% of patients II II-III 33 Neoadjuvant design, no prior II II-III 70 Neoadjuvant design, no prior II MBC 59 Anthracyclines in 79% of patients II LABC, MBC 112 Anthracyclines in 45% of patients 4 mg kg 1 loading, 2 mg kg 1 weekly until PD Docetaxel 100 mg m 2 i.v. every 3 weeks (6 cycles) 25 CHF 17% (LVEF reduction >15%) None As above 8% (LVEF reduction >15%) 4 mg kg 1 loading, 2 mg kg 1 weekly (18 cycles) 4 mg kg 1 loading, 2 mg kg 1 weekly (18 cycles); after surgery, 6 mg kg 1 every 3 weeks to complete 1 year 4 mg kg 1 loading, 2 mg kg 1 weekly until PD 8 mg kg 1 loading, 6 mg kg 1 every 3 weeks until PD Docetaxel 100 mg m 2 i.v. every 3 weeks (6 cycles) Docetaxel 75 mg m 2 i.v. + carboplatin AUC = 6 mg ml 1 i.v. every 3 weeks (6 cycles) Capecitabine 1657 mg m 2 orally twice daily, 3 weeks on, 1 week off until PD Docetaxel 100 mg m 2 i.v. every 3 weeks until PD 113 As above Docetaxel 75 mg m 2 III I-IIA 406 None 4 mg kg 1 loading, 2 mg kg 1 weekly for 1 year i.v. every 3 weeks + capecitabine 950 mg m 2 orally twice daily, 2 weeks on, 1 week off until PD Paclitaxel 80 mg m 2 i.v. weekly (12 cycles) 3% (LVEF reduction >20%, or to 3% (LVEF reduction to below 50%) 2% (LVEF reduction >15%, or to 19% (LVEF reduction >15%) 17% (LVEF reduction >15%) 0.5% CHF 3% (LVEF reduction 10 15% to below 50%, or LVEF reduction >16%) AUC area under the plasma concentration-time curve; LABC locally advanced, non-operable breast cancer; MBC metastatic breast cancer; NCI National Cancer Institute; PD progressive disease. British Journal of Pharmacology (2017)

14 B T Nemeth et al. Table 4 Clinical trials of trastuzumab in combination with further anti-her-2 drugs Study Phase Disease stage Number of patients Prior anticancer Trastuzumab dose Concomitant anticancer Portera et al., 2008 Dang et al., 2010 Burris et al., 2011 Krop et al., 2012 Gianni et al., 2012 NeoSphere Guarneri et al., 2012 CHER-LOB II MBC 11 All patients received anthracyclines and trastuzumab 8 mg kg 1 loading, 6 mg kg 1 every 3 weeks until PD II I-III 95 Unspecified 4 mg kg 1 loading, 2 mg kg 1 weekly with paclitaxel, then 6 mg kg 1 II MBC 112 Anthracyclines in 74% of patients II MBC 110 All patients received anthracyclines, taxanes, trastuzumab and lapatinib II II-III 107 Neoadjuvant design, no prior every 3 weeks to complete 1 year Pertuzumab 840 mg loading, then 420 mg every 3 weeks until PD 2 weeks (4 cycles) then paclitaxel 80 mg m 2 i.v. weekly (12 cycles) + lapatinib 1000 mg daily for 1 year None Ado-trastuzumab emtansine 3.6 mg kg 1 every 3 weeks for 1 year None Ado-trastuzumab emtansine 3.6 mg kg 1 every 3 weeks for 1 year 8mg kg 1 loading, 6mg kg 1 every 3 weeks (4 cycles); after surgery, continued with FEC for 1 year Docetaxel 75 mg m 2 i.v. every 3 weeks (4 cycles); after surgery, cyclophosphamide 500 mg m 2 i.v. + fluorouracil 500 mg m 2 i.v. + epirubicin 100 mg m 2 i.v. every 3 weeks (3 cycles) 107 As above Pertuzumab 840 mg loading, 420 mg + as above 107 As above Pertuzumab alone first, then as above 96 Initially none; after surgery as above IIb II-IIIA 36 No prior (exclusion criterion) for26weeks Pertuzumab with docetaxel, then FEC as above Paclitaxel 80 mg m 2 i.v. weekly (12 cycles), then cyclophosphamide 600 mg m 2 i.v. + fluorouracil 600 mg m 2 i.v. + epirubicin 75 mg m 2 i.v. every 3 weeks (4 cycles) 46 As above As above + lapatinib 1000 mg/day for 26 weeks Cardiac dysfunction observed (defined as) 9% CHF 45% (LVEF reduction >15%, or to 3% CHF 7% (significant, asymptomatic LVEF drop, otherwise unspecified) 2% (LVEF reduction to below 45%) 0% (LVEF reduction to below 45% or CHF) 1% (LVEF reduction to 3% (LVEF reduction to 0% 1% (LVEF reduction to 2.5% (LVEF reduction >10%) 0% continues 3740 British Journal of Pharmacology (2017)

15 Trastuzumab cardiotoxicity BJP Table 4 (Continued) Study Phase Disease stage Number of patients Prior anticancer Trastuzumab dose Concomitant anticancer Cardiac dysfunction observed (defined as) Guarneri et al., 2012 CHER-LOB (continued) 39 None As above, but lapatinib in 1000 mg/day dose 0% Hurvitz et al., 2013 II MBC 70 Anthracyclines in 50% of patients Moreno-Aspitia et al., 2013 RC0639 Schneeweiss et al., 2013 TRYPHAENA Yardley et al., 2015 T-PAS 8 mg kg 1 loading, 6 mg kg 1 every 3 weeks until PD Docetaxel 75 or 100 mg m 2 i.v. every 3 weeks until PD 67 None Ado-trastuzumab emtansine 3.6 mg kg 1 every 3 weeks for 1 year II I-III 109 Unspecified (neoadjuvant therapy possible) II II-III 73 Neoadjuvant design, no prior 4 mg kg 1 loading, 2 mg kg 1 weekly with paclitaxel (12 cycles); then 6 mg kg 1 every 3 weeks to complete 1year 8mg kg 1 loading, 6mg kg 1 every 3 weeks (6 cycles); after surgery, continue to complete 1 year 75 As above, except start with docetaxel (3 cycles) 77 8 mg kg 1 loading, 6 mg kg 1 every 3 weeks (3 cycles) II III-IV 215 Trastuzumab and lapatinib in 100%, anthracyclenes in 44% of patients 3 weeks (4 cycles), then paclitaxel 80 mg m 2 i.v. weekly (12 cycles) + lapatinib 1000 mg daily Pertuzumab 840 mg loading, 420 mg every 3 weeks (6 cycles) together with cyclophosphamide 600 mg m 2 i.v. + fluorouracil 500 mg m 2 i.v. + epirubicin 100 mg m 2 i.v. every 3 weeks (3 cycles), then with docetaxel 75 mg m 2 i.v. every 3 weeks (3 cycles) FEC as above, then pertuzumab together with docetaxel (3 cycles) Pertuzumab 840 mg loading, 420 mg + docetaxel 75 mg m 2 i.v. + carboplatin AUC = 6 mg ml 1 i.v. every 3 weeks (6 cycles) None Ado-trastuzumab emtansine 3.6 mg kg 1 every 3 weeks for 1 year 1.5% CHF 3% LVEF drop 1.5% CHF 3% LVEF drop 1% CHF 12% (LVEF reduction to 0% CHF 3% CHF 1.5% CHF 0.5% CHF 6% (LVEF decrease of any grade) Italics mark concurrent anthracycline and trastuzumab dosing. AUC area under the plasma concentration-time curve; FEC fluorouracil, epirubicin, and cyclophosphamide-containing chemotherapeutic regimen; MBC metastatic breast cancer; PD progressive disease. British Journal of Pharmacology (2017)

16 B T Nemeth et al. paclitaxel-trastuzumab first-line therapy. The occurrence of CD also decreased by administering trastuzumab following completion of anthracycline dosing both in nonencapsulated (Dang et al., 2008; Aogi et al., 2013) and liposomal (Cortes et al., 2009) forms of doxorubicin (Table 2). Taxanes. Single-arm (Esteva et al., 2002; Leyland-Jones et al., 2003; Burstein et al., 2003a; Gori et al., 2004; Coudert et al., 2006; Kelly et al., 2006; Dang et al., 2008; Untch et al., 2011; Aogi et al., 2013; Tolaney et al., 2015) or randomized (Marty et al., 2005; Joensuu et al., 2006; Wardley et al., 2010; Valero et al., 2011; Gianni et al., 2012; Baselga et al., 2012b) trials utilizing taxanes concurrently with trastuzumab in the adjuvant setting reported an increasing number of CD events (LVEF drop: 3 25%, CHF: 0 8%) with increasing prior exposure of patients to anthracyclines (Table 3). Combination of taxanes with platinum salts (Pegram et al., 2004; Coudert et al., 2007; Slamon et al., 2011; Valero et al., 2011) or capecitabine (Wardley et al., 2010) did not affect the incidence of CD. Other agents or administration protocols. Trastuzumab administered alone (Piccart-Gebhart et al., 2005) or together with vinorelbine (Jahanzeb et al., 2002; Burstein et al., 2003b), nucleoside analogues (O Shaughnessy et al., 2004; Yamamoto et al., 2008) or hormone therapy with anastrazole (Kaufman et al., 2009) resulted in a similar, relatively safe cardiac profile, keeping LVEF drops and CHF below 15 and 4% respectively (Tables 1 and 3). In a randomized, phase III, non-inferiority trial [PHARE (Pivot et al., 2013)], conducted in the hope of decreasing trastuzumab cardiotoxicity by administering the drug for 6 months instead of the standard 1 year, significantly fewer cardiac events occurred in the 6 month arm (Table 1). Despite the favourable cardiac safety, investigators failed to show non-inferiority of the shorter regimen. In line with the results from PHARE, a retrospective analysis on a large cohort of early breast cancer patients requiring discontinuation of trastuzumab due to cardiotoxic effects, conducted recently by Gong et al., showed that early discontinuation of the drug was a strong independent predictor of cardiac events and clinically significant relapse (Gong et al., 2016). Conversely, extending the period to 2 years (Goldhirsch et al., 2013) or even longer (Pistilli et al., 2015) is not associated with better outcomes. It must also be noted, however, that trastuzumab cardiotoxicity did not seem to increase with sustained dosing in either studies. Altogether, these results provide further support for the 1 year regime as the standard of care. Although cyclophosphamide is known to have cardiotoxic effects (Gottdiener et al., 1981), its contribution to cardiotoxicity in this setting is indistinguishable from that of anthracyclines, as they were administered in combination in all of the reviewed trials (Tables 2 and 4). Experience from large, randomized trials knowledge from the past, outlook to the future Large, multicentre, randomized trials such as National Surgical Adjuvant Breast and Bowel project B-31 (Tan-Chiu et al., 2005), The Finland Herceptin (FinHer) (Joensuu et al., 2006), Herceptin Adjuvant (HERA) (Piccart-Gebhart et al., 2005), North Central Cancer Treatment Group N 9831 (Perez et al., 2008) and Breast Cancer International Research Group (BCIRG) 006 (Slamon et al., 2011) showed that the addition of trastuzumab to anthracycline containing regimens in a sequential manner maintained CHF events below 4%, which was sustained over long-term follow-up (Gianni et al., 2011; Romond et al., 2012; de Azambuja et al., 2014). Asymptomatic LVEF drops, however, were more prevalent in patients receiving trastuzumab (Table 1), forcing the investigators to temporarily or permanently discontinue the drug in many cases. Furthermore, investigators of the BCIRG-006 trial reported that a subclinical toxic effect persisted for several years in patients who developed CD in the study arm receiving trastuzumab and anthracyclines (Slamon et al., 2011). A meta-analysis of the above five and two further randomized trials (Buzdar et al., 2005; Spielmann et al., 2009) by Moja et al. reported that there was a fivefold risk of developing CHF with trastuzumab compared with chemotherapy [risk ratio (RR) 5.11; 90% confidence interval (CI): 3 to 8.72, P < (Moja et al., 2012)] in a population of patients with early or locally invasive breast cancer. They concluded that although there is a benefit oftrastuzumabin both overall survival and disease-free survival, patients need to be carefully selected for trastuzumab as its cardiotoxic effects may erode the benefit observed in survival. Of the analysed trials, FinHer was the sole study in which RR for CHF favoured trastuzumab over chemotherapy (RR: 0.5; 90% CI: 0.07 to 3.74). Its design was different from the others in terms of administering trastuzumab before the anthracycline (in this case, epirubicin), resulting in a lower incidence of CD events than in any prior large randomized trial (Joensuu et al., 2006). Furthermore, this observed low incidence of CD was lasting (Joensuu et al., 2009). NeoALTTO (Baselga et al., 2012a) and NeoSphere (Gianni et al., 2012), two randomized studies using trastuzumab and epirubicin in a similar temporal sequence, provided further evidence of the feasibility of this regimen by reporting 1% of CD in the relevant arms of both cohorts. One previous, single-arm trial (Burstein et al., 2003a) that used a similar protocol reported worse cardiac outcome (Table 2). The investigators in this trial, however, did not define any cardiac eligibility criteria; therefore, there is no information as to how many patients might have had anamnestic or concomitant cardiac-related disease in that cohort. New drugs for HER-2 antagonism pertuzumab, lapatinib and ado-trastuzumab emtansine Despite the significance of trastuzumab in up-to-date breast cancer, concerns in connection with its cardiotoxicity and resistant tumours warranted development of other HER-2 signalling blocking agents. Trials with pertuzumab (Portera et al., 2008; Gianni et al., 2012; Baselga et al., 2012b; Schneeweiss et al., 2013), lapatinib (Geyer et al., 2006; Dang et al., 2010; Guarneri et al., 2012; Baselga et al., 2012a; Moreno-Aspitia et al., 2013; Robidoux et al., 2013) and ado-trastuzumab emtansine [T-DM1, an antibody-drug conjugate, (Burris et al., 2011; Krop et al., 2012; Verma et al., 2012; Hurvitz et al., 2013; Krop et al., 2014; Yardley et al., 2015)] have reported promising results both in terms of cardiac safety (Table 4) and anti-tumour 3742 British Journal of Pharmacology (2017)

17 Trastuzumab cardiotoxicity BJP efficacy. Interestingly, further blockade of HER-2 signalling by adding these new molecules to trastuzumab-containing regimens does not seem to increase cardiac morbidity compared with trastuzumab alone (Table 4). A possible explanation for this might be that in the heart as it will be discussed in more detail in the experimental background section HER-2 appears to play a role only as a dimerization partner for other HER types. Consequently, new anti-her-2 signalling drugs that exert their effect on different locations on HER-2 might not inhibit heterodimer formation with other HERs to the extent trastuzumab does (Figure 1). Although all of these drugs are now approved by the FDA to treat breast cancer (Figure 2), real-life data and further larger trials with longer follow-up will determine whether their cardiac safety is truly better than that of trastuzumab. Overall, there has been a significant evolution of trastuzumab administration in relation to both the type and timing of chemotherapeutics with which it could safely be coadministered. Clinical data from the last almost two decades indicate that (1) anthracyclines should be used sequentially, preferably following trastuzumab ; (2) anthracyclines might be more feasible to be used in a liposome-encapsulated form; (3) taxanes and trastuzumab given concurrently are relatively safe and effective; and (4) emerging new anti-her-2 therapies are showing similar anticancer efficacy to trastuzumab, with a potentially better cardiac safety profile. Experimental background: what we know HER-2/ErbB signalling and trastuzumab The mechanism of action and cardiotoxicity of trastuzumab is still not completely defined despite the considerable efforts that have been made to reveal the underlying mechanisms (Hudis, 2007). Although early studies using HER-2/ErbB2 knockout mice have emphasized the importance of its downstream signalling in cardiovascular development, its role in the adult heart has only come into focus when patients receiving trastuzumab showed an increased incidence of heart failure and systolic dysfunction, as outlined above. Permanent loss of the function of HER-2 is incompatible with life and leads to death in utero (Lee et al., 1995; Chan et al., 2002). Subsequent investigations utilizing conditional cardiac disruption of the receptor in adult mice resulted in the development of spontaneous dilated cardiomyopathy (Crone et al., 2002; Ozcelik et al., 2002). HER receptors can be activated by numerous ligands in vivo, including EGF (HER-1) or neuregulins (NRGs, HER-3 and -4)(Yarden and Sliwkowski, 2001). Although to our current knowledge HER-2 itself is an orphan receptor, it is essential in the formation of heterodimers with other types of ErbB receptors, thereby increasing their activity (Karunagaran et al., 1996). Moreover, HER-2 homodimers seem to be constitutively active (Kraus et al., 1987) and are more commonly found on the surface of cells overexpressing HER-2, such as breast cancer cells. ErbB downstream signalling includes activation of several important pathways such as phosphatidylinositol-3-kinase/akt, MAPK and endothelial nitric oxide synthase, which are all major contributors in cell survival, mitochondrial function, sarcoplasmic reticulum calcium uptake, growth or proliferation [Figure 1, (Odiete et al., 2012; Varga et al., 2015)]. In the heart, these pathways are important mostly in homeostatic processes and are activated predominantly through HER-4. As HER-2 is a transmembrane protein, it is a potential target for proteolysis. p95-her-2, the smaller degradation product of this process, remains embedded in the plasma membrane in an active state. Furthermore, many breast cancers express p95-her-2 via alternative translation of the HER-2 mrna (Arribas et al., 2011). Interestingly, this constitutively active fragment regulates several genes involved in developing and maintaining metastatic potential that are not influenced by the full-length receptor (Pedersen et al., 2009). Also, tumours expressing p95-her-2 tend to be resistant to trastuzumab but have a favourable response rate to the tyrosine kinase inhibitor lapatinib (Scaltriti et al., 2010; Arribas et al., 2011). Inactivation of HER-2 signalling by trastuzumab possibly comprises multiple effects. It appears that even though anti- HER-2mAbsinduceHER-2homodimerization,thisdoesnot result in increased downstream signalling. Instead, the amount of HER-2 receptors on the cell surface was found to be reduced in response to trastuzumab, albeit via an uncertain mechanism (Hudziak et al., 1989; Sliwkowski et al., 1999). Trastuzumab was also shown to decrease cell proliferation by inhibiting the cell cycle (Sliwkowski et al., 1999), thus being more cytostatic than cytotoxic. Antibody-dependent cellular cytotoxicity is efficiently induced by trastuzumab as well (Carter et al., 1992). The most likely mechanism involved in the cardiotoxicity of trastuzumab is the consequence of its interference with NRG/ErbB signalling (Pentassuglia et al., 2007), as activity of both HER-3 and HER-4 is impaired when HER-2 is not available for formation of heterodimers (Graus-Porta et al., 1997). Thus, the important cellular defensive and energy-generating systems of cardiomyocytes outlined above might not function properly in the presence of trastuzumab (Figure 1). Although this cardiotoxic effect was initially deemed reversible upon the discontinuation of the drug, experimental results imply that there might be lasting effects as a result of ultrastructural changes observed in rat ventricular myocytes (Sawyer et al., 2002) and in mice (ElZarrad et al., 2013) treatedwithtrastuzumab. Thedrugalso changed expression of genes involved in DNA repair and adaptation to stress (ElZarrad et al., 2013). Therefore, further investigations to uncover the precise mechanisms of trastuzumab-induced effects in cardiomyocytes are needed. Doxorubicin and trastuzumab synergy in cardiotoxicity A highly possible explanation for the additive cardiotoxic effect of doxorubicin and trastuzumab is that while doxorubicin increases the production of reactive oxygen and nitrogen species (ROS/RNS) (Doroshow and Davies, 1986; Pacher et al., 2002, 2003; Mukhopadhyay et al., 2009; Zhao et al., 2010), blockade of HER-2 signalling results in decreased activation of survival pathways and exacerbates oxidative and nitrative stress [Figure 1, (Milano et al., 2014)]. This leads to intracellular ROS accumulation, resulting in CD and cardiomyocyte apoptosis. NRG-1/ErbB signalling was shown to protect cardiac myocytes from anthracycline-induced apoptosis (Fukazawa et al., 2003). Furthermore, calcium dysregulation and British Journal of Pharmacology (2017)