Main

The discovery of the human epidermal growth factor receptor 2 (HER2) (King et al, 1985; Schechter et al, 1985), its association with poor prognosis in breast cancer (Slamon et al, 1987) and the potential of recombinant DNA technology to produce monoclonal antibodies generated great enthusiasm in academic laboratories and industries in the 1980s. The development of trastuzumab (herceptin, Genentech, South San Francisco, CA, USA) monoclonal antibody therapy represented a paradigm shift in oncology from non-specific chemotherapy to a molecularly targeted approach (Esteva, 2004). Trastuzumab binds domain IV of the extracellular component of the HER2 protein located close to the cell membrane, resulting in signal transduction blockade and prevention of HER2 cleavage (Figure 1). Addition of trastuzumab to conventional cytotoxic chemotherapy improved overall response rates (ORR), time to progression (TTP) and overall survival (OS) rates (Slamon et al, 2001). Lapatinib, a tyrosine kinase inhibitor of EGFR and HER2 was found to be effective in combination with capecitabine in patients whose metastatic tumours were progressing on trastuzumab-based chemotherapy (Geyer et al, 2006). Pertuzumab (Perjeta, Genentech) is a humanised monoclonal antibody directed against domain II of the extracellular component of HER2, which is where receptor dimerisation occurs (Adams et al, 2006). The development of antibody–drug conjugates is an area of high interest in oncology drug development. Ado-trastuzumab-DM1 (T-DM1, Kadcyla) combines the HER2-targeted antitumor properties of trastuzumab with the anti-microtubule agent DM1 (derived from maytansine) allowing preferential intracellular drug delivery to the HER2+ tumour cells. Trastuzumab, pertuzumab and T-DM1 can all induce antibody-dependent cellular cytotoxicity.

Figure 1
figure 1

Molecular approaches to HER2 targeted therapy.

PowerPoint slide

To date, HER2 remains to be the most important predictive factor of response to HER2-targeted therapies (Seidman et al, 2001). A tumour is considered HER2+ if the ratio of HER2/cep17 is at least 2.0 or if the HER2 gene copy number is >6 (independently of chromosome 17) (Wolff et al, 2013).

Despite improvements in progression-free survival (PFS) and OS rates, HER2+ MBC remains an incurable disease and clinical research remains as important as ever. In this article we discuss the optimal sequencing of HER2-targeted therapies in HER2+ MBC based on line of therapy. We discuss potential predictive markers of resistance to HER2-targeted therapies and review ongoing efforts to incorporate novel drugs and drug combinations, including the promise of immunotherapy.

Optimal sequencing of ANTI-HER2 therapies in MBC

First-line therapy of HER2+ MBC

In the trastuzumab pivotal phase III trial, 469 women with HER2+ metastatic breast cancer were randomized to standard chemotherapy alone (doxorubicin or epirubicin in combination with cyclophosphamide) vs chemotherapy plus trastuzumab. The addition of trastuzumab to chemotherapy was associated with longer TTP (7.4 months vs 4.6 months, P<0.001), a higher ORR (50% vs 32%, P<0.001), a longer duration of response (9.1 months vs 6.1 months, P<0.001), a lower rate of death at 1 year (22% vs 33%, P=0.008), a longer OS (25.1 months vs 20.3 months, P=0.01) and a 20% reduction in the risk of mortality (Slamon et al, 2001). Based on this trial, trastuzumab was approved in combination with paclitaxel for first-line treatment of HER2+ MBC in 1998. Other combinations shown to be effective in this setting include docetaxel plus trastuzumab (Esteva et al, 2002; Marty et al, 2005) and vinorelbine plus trastuzumab (Burstein et al, 2007). Although efficacious, anthracycline/trastuzumab combinations are not indicated outside clinical trials in MBC due to increased risk of cardiac toxicity (Slamon et al, 2001).

Based on strong preclinical evidence that platinum and trastuzumab have synergistic activity, the three-drug combination of trastuzumab, carboplatin and docetaxel was tested in the Breast Cancer International Research Group 007 (BCIRG007) trial without showing improvement in PFS, OS or ORR when compared with trastuzumab–docetaxel combination (Valero et al, 2011). In a separate open-label, randomized study, addition of platinum to trastuzumab and paclitaxel improved PFS and ORR but not OS (Robert et al, 2006). Carboplatin adds toxicity with no clear improvement in clinical outcomes, and therefore is not considered standard of care for patients with HER2+ MBC.

To evaluate if trastuzumab plus chemotherapy was a superior strategy in metastatic disease compared with trastuzumab alone, Inoue et al (2011) completed a randomized study in which they compared two arms: trastuzumab (H) alone followed by H+docetaxel (D) (H→H+D) upon progression vs H+D combination therapy from the onset for HER2+ MBC. Both PFS and OS were significantly prolonged in the H+D group. Based on the results of this trial, combination of trastuzumab with chemotherapeutic agent is considered a preferred first-line strategy compared with than trastuzumab alone (Inoue et al, 2011).

Preclinical studies showed that a combination of trastuzumab and pertuzumab can induce apoptosis in vitro (Nahta et al, 2004) and tumour regression in vivo (Lee-Hoeflich et al, 2008). The Clinical Evaluation of Pertuzumab and Trastuzumab (CLEOPATRA) trial was a large phase III randomized trial in which 808 patients with HER2+ MBC were randomized to trastuzumab plus docetaxel (TH) with or without pertuzumab in the first-line setting. The pertuzumab group showed significant improvement in PFS (18.5 months vs 12.4 months, 95% confidence interval (CI): 0.51–0.75, P<0.001) and OS rates (DeLong et al, 1988). Based on the results of this trial, FDA granted approval for the use of trastuzumab, pertuzumab and docetaxel combination as the first-line therapy in HER2+ MBC. This is currently the preferred first-line treatment for most patients with HER2+ MBC (Figure 2).

Figure 2
figure 2

Sequencing of targeted therapies in HER2-positive metastatic breast cancer.

PowerPoint slide

Hormone receptor positive and HER2+ MBC

About 50% of the HER2+ patients are also hormone receptor positive (HR+). There are important biological differences underlying pure HER2+/hormone receptor negative (HR−) tumours and HER2+/HR+ tumours. When HR is overexpressed, the molecular profile of these tumours resembles the so-called luminal B subtype of breast cancer (Perou et al, 2000). Cross-talk between HER2 and HR leads to resistance to hormonal agents, and resistance can be partially overcome by anti-HER2 therapies. In the phase III Trastuzumab and Anastrozole Directed Against oestrogen receptor (ER)-Positive HER2-Positive Mammary Carcinoma (TAnDEM) trial, 207 postmenopausal women with HER2+/HR+ disease were assigned to anastrozole alone or with trastuzumab. The combination arm showed significant improvement in PFS (HR: 0.63, 95% CI: 0.47–0.84, median: PFS 4.8 vs 2.4 months, P=0.0016) but not in the OS. However, there was a 70% cross-over rate to the combination arm in this study (Kaufman et al, 2009). Similarly, combination of letrozole with lapatinib was found to be more effective than letrozole alone in patients with HER2+/HR+ MBC (Johnston et al, 2009). It is an acceptable approach to start with a combination of an anti-oestrogen and an anti-HER2 therapy in the metastatic setting especially if the disease burden is limited without visceral crisis.

Clinical trials in the neoadjuvant setting have shown that HR status is a marker of sensitivity to anti-HER2-directed therapies, with higher pCR rates consistently observed in patients with HR-negative tumours. For example, in the NEO-SPHERE trial 553 patients with HER2+ breast cancer were randomized to one out of four neoadjuvant regimens prior to surgery as follows: TH, pertuzumab plus trastuzumab and docetaxel (THP), pertuzumab plus trastuzumab (HP) or pertuzumab plus docetaxel (TP). Randomisation was stratified by breast cancer type 556 (operable, locally advanced or inflammatory) and ER or progesterone 557 receptor positivity. The pCR rates were significantly higher for HR-negative tumours compared with HR-positive in all four groups (TH 20% HR+ vs 37% HR−; THP 26% HR+ vs 63 HR−; HP 6% vs 27%; TP 17% vs 30%) (Gianni et al, 2012). Similar findings have been reported from the NEO-ALTTO (Baselga et al, 2012a) and GEPARQUINTO (Untch et al, 2012) trials. This has important implications on future trial design and sample size calculations. Ideally, clinical trials should focus on either HER2+/HR− tumours or HER2+/HR+ tumours. If both tumour types are included, the sample size should be calculated separately for each subgroup.

Ongoing trials in first-line HER2+ MBC

Improving outcomes in the first-line setting is always important for patients with MBC, especially if survival rates are significantly improved. The MARIANNE trial is evaluating the role of first-line T-DM1 in HER2+ MBC. More than 1000 patients have been randomized to one of the three arms, taxane plus trastuzumab, T-DM1 alone or T-DM1 plus pertuzumab. Results of this trial are awaited with great interest because they may change clinical practice.

Second-line therapy and treatment of refractory HER2+ MBC

The FDA approved lapatinib for patients with HER2+ MBC who had received prior anthracycline and taxane-based chemotherapy and were progressing on trastuzumab-based therapy based on a phase III trial, comparing lapatinib/capecitabine combination with capecitabine alone. The addition of lapatinib to capecitabine was shown to prolong TTP (8.4 months vs 4.1 months, HR: 0.47, 95 CI: 0.33–0.67, P<0.001) (Geyer et al, 2006). This combination is now an accepted second-line approach for patients with HER2+ MBC who have progressed on prior trastuzumab containing regimen especially when T-DM1 is not available.

In the phase III EMILIA trial, 991 patients with HER2+ MBC who had previously been treated with trastuzumab and a taxane, were randomly assigned to T-DM1 or lapatinib/capecitabine combination. The T-DM1 group showed significant improvement in the median PFS (9.6 months vs 6.4 months, 95% CI: 0.55–0.77, P<0.001), median OS (30.9 months vs 25.1 months, HR for death from any cause: 0.68, 95%: CI 0.55–0.85, P<0.001) and ORR (43.6% vs 30.8%, P<0.001) with a lower incidence of grade 3 toxicity (57% vs 41%) (Verma et al, 2012). Based on the results of the EMILIA trial, the FDA approved T-DM1 as a second-line agent in HER2+ MBC.

In a phase II trial, Krop et al (2012) reported significant activity for T-DM1 in patients with HER2+ MBC who had received prior anthracycline, taxanes capecitabine, trastuzumab and lapatinib therapy. The phase III TH3RESA trial randomized ∼600 patients with advanced HER2-positive breast cancer, previously treated with at least two HER2-directed therapies (including trastuzumab and lapatinib) in a 2 : 1 ratio to T-DM1 or physician’s choice of treatment (HER2-targeted regimens for 83.2% and single-agent chemotherapy for 16.8%). In the initial results of this trial reported at the European Cancer Congress in September 2013, patients treated with T-DM1 had a significantly prolonged median PFS (6.2 months vs 3.3 months, HR: 0.528, 95% CI: 0.422–0.661, P<.0001). Therefore T-DM1 appears to be efficacious even in heavily pretreated patients who have not been exposed to T-DM1 as second-line therapy.

Combination of trastuzumab and pertuzumab has also demonstrated clinical efficacy in the second-line setting (Cortes et al, 2012). The combination of pertuzumab with trastuzumab is reasonable in patients who have not been exposed to first-line trastuzumab/pertuzumab/docetaxel as in CLEOPATRA trial.

Continuing trastuzumab beyond progression

Trastuzumab has demonstrated efficacy in heavily pretreated HER2+ MBC in combination with several other chemotherapeutic agents such as capecitabine, gemcitabine and vinorelbine. In a phase III randomized study, 156 patients with HER2+ MBC who had progressed on prior trastuzumab were randomized to capecitabine alone or capecitabine plus trastuzumab. The trastuzumab arm showed significant improvement in the TTP (5.6 months vs 8.2 months, HR: 0.69, 95% CI: 0.48–0.97, P=0.0338) and ORR (27% vs 48.1%, Odds Ratio: 2.5, P=0.0115) but not OS (20.6 months vs 24.9 months, 95% CI: 0.65–1.35, P=0.73) (von Minckwitz et al, 2009). This study was terminated prematurely because of the approval of lapatinib and capecitabine in this indication. Although it is not clear why patients would continue to benefit from continuation of trastuzumab in the setting of disease progression, it is possible that HER2 inhibition may continue to provide synergy with different chemotherapeutic agents administered sequentially.

Dual inhibition targeting both intracellular and extracellular domain of HER2 by combining lapatinib with trastuzumab has also been tested in metastatic HER2+ breast cancer. In the phase III EGF104900 trial, a heavily pretreated population which had progressed on prior trastuzumab-based regimens was randomly assigned to receive combination of lapatinib and trastuzumab or lapatinib monotherapy. In the final survival analysis, dual HER2 blockade led to significant 4.5-month improvement in OS (HR: 0.74, 95% CI: 0.57–0.97, P=0.026) in the HR+ group (Blackwell et al, 2010).

Molecular mechanisms of resistance and ongoing clinical trials in refractory HER2+ MBC

Human epidermal growth factor receptor 2 quantitative expression, either in terms of protein or mRNA levels within the clinically-defined HER2-positive tumours seems to predict higher or lower probability of response, as shown repeatedly from pre-specified analyses of very influential prospective trials (e.g., CLEOPATRA, EMILIA, NEO-SPHERE, NEO-ALTTO and TRYPHAENA) (Baselga et al, 2012c; Gianni et al, 2012; Schneeweiss et al, 2013; Verma et al, 2012).

Hyperactivation of the PI3K pathway by activating mutations or loss of PTEN expression has been associated with resistance to trastuzumab-based chemotherapy (Esteva et al, 2010; Nagata et al, 2004). In a prospective study of patients who had been previously treated with trastuzumab and subsequently developed metastatic breast cancer, the metastatic tumours expressed lower levels of PTEN compared with primary tumours, suggesting that the PTEN loss may be a marker of trastuzumab resistance (Chandarlapaty et al, 2012). However, PTEN expression in primary breast cancer was not predictive of disease-free survival (DFS) or OS in adjuvant trastuzumab trials. Other proposed markers of trastuzumab resistance include a truncated form of HER2 (p95), PIK3CA mutations (Berns et al, 2007; Esteva et al, 2010), HER2/IGF-IR dimerisation (Nahta et al, 2005) and Src activation (Zhang et al, 2011). Although none of these markers have been validated in prospective clinical trials to exclude patients from HER2-directed therapies, molecular understanding provides target for rational drug development.

A multitude of clinical trials are assessing the safety and efficacy of novel treatments in patients with refractory HER2+ MBC (Table 1). In this section we will discuss selected novel approaches including those based on proposed mechanisms of resistance to HER2-targeted therapy (Figure 1).

Table 1 Ongoing clinical trials in HER2-positive metastatic breast cancer

Tyrosine kinase inhibitors

Afatinib is an orally active irreversible dual inhibitor of EGFR and HER2 receptors. In a phase II study, afatinib monotherapy in heavily pretreated HER2+ MBC demonstrated partial response (PR) in 4 patients (10% of 41) and stable disease in 11 patients (37%of 41) (Lin et al, 2012).

Neratinib is an orally active irreversible inhibitor of EGFR, HER2 and HER4 receptors. In a phase II open-label clinical trial, 240 mg of oral neratinib was administered to trastuzumab pretreated (n=66) and a trastuzumab naïve cohort (n=70). The ORR was 24% and 56%, respectively, and the most common grade 3 toxicity was diarrhoea.

PI3K/Akt/mTOR pathway inhibitors

In preclinical models, mTOR inhibitors synergize with trastuzumab and have shown to cause complete regression of mouse HER2+ mammary tumours (Lu et al, 2007) In a phase I/II trial of trastuzumab combined with mTOR inhibitor everolimus for HER2+ MBC, PR was seen in 15% patients and s.d.>6 months in 19% (Morrow et al, 2011). BOLERO-3 was a phase III trial comparing vinorelbine and trastuzumab alone or in combination with everolimus in 569 patents with HER2+ MBC resistant to trastuzumab. The preliminary findings of this study show significant prolongation in TTP (5.8 months vs 7 months, HR: 0.78; 95% CI: 0.65–0.95; P<0.01) in the everolimus arm. The data on OS, the secondary endpoint of this study is not yet mature. Exploratory analysis of biomarkers in the BOLERO-3 trial suggests that the addition of everolimus to trastuzumab plus vinorelbine for HER2-positive advanced breast cancer may be most beneficial in patients with low PTEN or high pS6 levels (Jerusalem et al, 2013). No clear benefit of everolimus was observed in patients with normal PTEN or low pS6 levels. These data support the hypothesis that low PTEN expression is a marker of trastuzumab resistance (Nagata et al, 2004; Esteva et al, 2010). Subset analyses showed a larger benefit for the everolimus group in HER2+/HR− tumours, compared with HER2+/HR+ tumours. This seems counterintuitive in view of the results of everolimus and exemestane in HR+ tumours (BOLERO-2) (Baselga et al, 2012b). No treatment–biomarker interaction was reported between everolimus and PI3K mutations. Data from the Cancer Genome Atlas (TCGA-Network, 2012) suggest that despite a similar incidence of PI3K mutation in HER2 enriched and luminal tumours, markers of pathway activations were differently expressed in these two subtypes in the presence of PI3K mutations.

Several PI3 kinase inhibitors are under phase 1/2 stage of development. A phase 1/2 study of SAR245408 (S08) in combination with trastuzumab or paclitaxel and trastuzumab in patients with HER2+ MBC, who progressed on a previous trastuzumab-based regimen, has been completed. Other PI3 kinase inhibitors (e.g., BKM120 (NCT01589861)) are also under investigation.

Heat shock protein 90 (HSP90) inhibitors

Proteosomal degradation of oncoproteins such as HER2 is caused by compounds called HSP90 inhibitors. Moreover, p-95HER2, which is a truncated form of HER2 and also a major mechanism of trastuzumab resistance has been shown to undergo degradation by HSP90 inhibitors. Tanespimycin has been evaluated in HER2+ MBC that had previously progressed trastuzumab in the phase II setting. The ORR was 22%, the clinical benefit rate (ORR and SD for at least 6 months) was 59% and median PFS was 6 months (95% CI: 4–9 months). Another HSP90 inhibitor, ganetespib (STA-9090), was tested as monotherapy in a phase II setting (Jhaveri et al, 2014). While the study did not meet the primary endpoint of ORR in the first stage of the Simon-2 stage design, modest activity was noted in heavily pretreated, trastuzumab-refractory patients. Based on the preclinical data that shows synergistic activity for combining HSP90 inhibitors with taxanes, we are now conducting a phase I trial of ganetespib plus paclitaxel plus trastuzumab in trastuzumab refractory, HER2+ MBC (NCT02060253).

Other targeted strategies

KD019 is a small molecule that simultaneously blocks the tyrosine kinase of EGFR, HER2, Src and the vascular endothelial growth factor receptor 2 (VEGFR2), all of which are implicated in the processes of tumour cell growth, angiogenesis and metastases. A phase I study of KD019 plus trastuzumab in HER2 overexpressed or amplified MBC is ongoing for patients who have received two or more prior anti-HER2-directed therapies (Jhaveri et al (2014), Perlmutter Cancer Center at NYU Langone).

It has been demonstrated that cross-talk between IGF-1R and HER2 as well as IGF mediated phosphorylation of HER2 results in trastuzumab resistance (Nahta et al, 2005, 2006), providing merit to exploring IGF-1R inhibitors such as Cixotumumab in clinical trials. Preclinical studies have shown upregulation of VEGF in HER2 overexpressed breast cancers, and the phase 2 study of trastuzumab and bevacizumab combination in HER2+ MBC was promising. However, bevacizumab, a monoclonal antibody against VEGF-A receptor, failed to improve PFS when combined with trastuzumab and docetaxel in a phase 3 trial compared with the non-bevacizumab arm (Gianni et al, 2013).

One of the mechanisms of action of trastuzumab is thought to be induction of ADCC. It has been demonstrated that higher percentage of tumour infiltrating lymphocytes is associated with better response to trastuzumab in both adjuvant (Loi et al, 2013) and neoadjuvant setting. In correlative preclinical studies, higher PD-1 (programmed death-1), a T-cell checkpoint ligand expression, was associated with greater trastuzumab benefit. The negative regulator of T-cell mediated immune response is PD-1, so antibodies blocking PD-1 and its ligand PD-L1 enhance the T-cell mediated immune response. Trastuzumab may modulate tumour microenvironment by inhibiting tumour-mediated immunosuppression via factors like PD-1 (Stagg et al, 2011). Combining trastuzumab with anti-PD-1 and anti-PD-L1 antibodies showed greater tumour regression in mouse models of HER2+ mammary tumours. Therefore, there seems to be merit in exploring the impact of combining trastuzumab with inhibitors of negative T-cell regulation, such as anti-CTLA4 antibody, anti-PD-1 or anti-PDL-1, in HER2+ MBC.

The HER2 has been explored as an antigen for vaccine development in HER2+ breast cancer (Table 2). One peptide-based vaccine that merits special mention is the E75 vaccine derived from the extracellular domain of HER2 receptor. When compared with the unvaccinated arm, E75 was found to decrease recurrence rates when administered in the adjuvant setting for node positive HER2+ breast cancer (DFS rates at 22 months: 85.7% vs 59.8%) (Peoples et al, 2005).

Table 2 Vaccine types available for HER2 overexpressing breast cancer

Brain metastases (BM)

The incidence of BM in patients with HER2+ MBC increases over time and may be attributable partly to marked reduction in mortality as a result of HER2 inhibition and control of non-CNS metastatic progression and monoclonal antibodies’ inability to cross the blood brain barrier (BBB). Central nervous system involvement and its treatment remains one of the biggest challenges in HER2+ MBC. Current research is focused on various approaches including use of small molecule inhibitors that have the potential to cross the BBB (for e.g. Afatinib and everolimus), using molecules concurrently with radiation, and lastly utilising immunotherapy before and after radiotherapy based on the efficacy noted in melanoma patients with BM. In addition in the phase 2 LANDSCAPE study, 45 patients with untreated HER2+ BM received lapatinib and capecitabine in combination. After a median follow-up of 21.2 months, 66% of patients achieved a PR. The incidence of grade 3 or 4 adverse events was 49%. This study suggests that capecitabine and lapatinib combination may be an acceptable first-line regimen in the management of BM in HER2+ MBC (Bachelot et al, 2013). However, this regimen is yet to be compared with other treatment modalities such as whole brain radiation in a larger phase 3 trial.

Conclusions

Human epidermal growth factor receptor 2 is a validated therapeutic target that remains relevant throughout the disease process. Patients with HER2+ MBC can be treated safely with a variety of systemic therapies and survival rates are improving. The landscape of anti-HER2 therapy is changing very quickly. In 2012, pertuzumab, a humanised monoclonal antibody against HER2, was approved for first-line therapy of HER2+ MBC in combination with trastuzumab and docetaxel. In 2013, the antibody–drug conjugate T-DM1 was approved in the second-line setting, displacing lapatinib/capecitabine. The MARIANNE trial is evaluating the role of T-DM1 as front-line therapy and this study may change practice. Dual HER2 inhibition with trastuzumab and pertuzumab, trastuzumab and lapatinib or pertuzumab and T-DM1 is here to stay in the metastatic setting. In fact, in an attempt to improve disease-free and OS rates, pertuzumab and T-DM1 are now being evaluated in the neoadjuvant and adjuvant settings (Jhaveri and Esteva, 2014). How to use these agents after patients are exposed to them in the adjuvant setting is not known, but efficacy is likely to be lower than what has been reported to date (Murthy et al, 2014). In addition to specific targeted therapies, patients with HER2+ MBC are treated with chemotherapy (e.g., taxanes, capecitabine, vinorelbine, gemcitabine and platinum salts) and endocrine therapy. With so many treatment options available, it is increasingly important to develop evidence-based guidelines for the initial and subsequent treatment of HER2+ MBC. Furthermore, clinical trials should take into consideration the optimal sequence of HER2 targeted therapies in patients with HR+ and HR− HER2+ MBC. Understanding mechanisms of resistance in individual patients remains a challenge and trials that incorporate biopsies and biomarker analysis will be needed in the new world of personalised cancer therapy. Directing the immune system to eradicate cancer cells is exciting, either by using vaccines or immune checkpoint modulators.