MET inhibitors for targeted therapy of EGFR TKI-resistant lung cancer

Lung cancer is the leading cause of cancer death among both men and women and accounts for one third of all cancer deaths worldwide. Non-small cell lung cancer (NSCLC) constitutes over 80% of lung cancer cases and has a low 5-year survival rate of about 18% [1], despite great efforts made worldwide over the past decades to combat lung cancer. The development of epidermal growth factor receptor (EGFR) tyrosin1e kinase inhibitors (EGFR-TKIs) based on the discovery of EGFR-activating mutations is an important milestone in the targeted therapy of NSCLC.

The majority of EGFR-activating mutations (~ 90%) primarily present as an exon 19 deletion (Del19; ~ 60%) or exon 21 point mutation L858R (~ 30%). The prevalence of these mutations is ~ 15% and ~ 40% in Western and Asian populations with NSCLC, respectively [2]. These EGFR mutations increase the affinity of EGFR-TKIs for the mutant receptor, thus conferring sensitivity to EGFR-TKI treatment. First-generation EGFR-TKIs, such as gefitinib and erlotinib, are competitive reversible inhibitors of ATP, thereby preventing autophosphorylation of the TK domain and blocking the activation of signaling downstream of EGFR [2]. First-generation EGFR-TKIs provide significant clinical benefit in patients with these mutations, representing the first successful targeted therapy against lung cancer. However, patients eventually develop disease progression because of acquired resistance, which limits the long-term efficacy of these agents [2‐4].

Acquired resistance to first-generation EGFR-TKIs is often caused by the acquisition of the T790 M mutation, which accounts for approximately 60% of resistant cases. In addition, MET (c-MET) gene amplification is another important mechanism and is detectable in approximately 5–22% of NSCLC patients with acquired resistance to first-generation EGFR-TKIs [2‐4]. Mechanistically, MET amplification causes EGFR-TKI resistance by activating EGFR-independent phosphorylation of ErbB3 and downstream activation of the PI3K/AKT pathway, providing a bypass pathway in the presence of an EGFR inhibitor. This redundant activation of ErbB3 permits cells to transmit the same downstream signaling in the presence of EGFR-TKIs. Thus, concomitant inhibition of both EGFR and MET would be required to overcome resistance to EGFR inhibitors by MET amplification [5]. Although MET amplification can occur with the EGFR T790 M mutation, about 60% of MET amplification is found without T790 M mutation. There is an inverse correlation between the presence of T790 M and MET gene copy number, suggesting a complementary or independent role of the two mechanisms in the acquisition of resistance [6].

Osimertinib (AZD9291 or TAGRISSO^TM), rociletinib (CO1686), olmutinib (HM61713), nazartinib (EGF816), naquotinib (ASP8273), mavelertinib (PF-0647775), and avitinib (AC0010) are examples of third-generation EGFR-TKIs, which selectively and irreversibly inhibit the common “sensitive” EGFR mutations, Del19 and L858R, and the resistant T790 M mutation while sparing wild-type (WT) EGFR (see their chemical structures in Fig. 1). Osimertinib is now an FDA-approved drug for treating patients with NSCLC that has become resistant to the first-generation EGFR-TKIs through the T790 M mutation and for EGFR mutation-positive advanced NSCLC as a first-line treatment. Although osimertinib has achieved great success in the clinic, all patients have eventually relapsed and developed resistance to the treatment, resulting in treatment failure. Unfortunately, the resistance mechanisms are largely unknown except for some related to C797S mutation and MET amplification.

To conquer resistance to EGFR TKIs, many clinical trials that test novel EGFR, MET, and VEGFR inhibitors have been designed and launched in China and all over the world [7‐9]. Toward C797S mutation, the fourth-generation EGFR-TKIs such as EAI045 has been developed and is under preclinical development [10]. This review will primarily focus on the role of MET amplification in mediating acquired resistance to osimertinib as well as other third-generation EGFR-TKIs.

MET proto-oncogene exists in the long arm of human chromosome 7 and encodes MET (c-MET) protein that is a membrane tyrosine kinase receptor. The initially encoded preproprotein is proteolytically processed to generate α and β subunits that are linked via disulfide bonds to form the mature receptor. The binding of MET to its ligand, hepatocyte growth factor (HGF) secreted by stromal cells, induces dimerization and activation of the receptor. Therefore, the activated MET is a heterodimer linked by an extracellular α chain and a transmembrane β chain that contains a SEMA (sema homology region) domain, a PSI (plexin-semaphorin-integrin) domain, four IPT (immunoglobulin-like regions in plexins and transcription factors) domains, a transmembrane domain, a juxtamembrane domain, a tyrosine kinase domain, and a C-terminal tail region. The SEMA domain is the site where HGF binds directly to MET, and PSI can stabilize this interaction. When HGF binds MET, autophosphorylation of Y1234 and Y1235 in the intracellular tyrosine kinase domain occurs, resulting in Y1349 and Y1356 autophosphorylation in the C-terminal multifunctional docking site. This induces the recruitment of several intracellular effector adaptor proteins such as growth factor receptor-bound protein 2 (GRB2), GAB1, SRC, and PI3K and consequently the activation of downstream signaling pathways (Fig. 2) [11, 12]. The HGF/MET signaling pathway is highly regulated and plays an important role in cell proliferation, survival, embryogenesis, and cellular migration and invasion [11‐13]. The main types of variation of HGF/MET signaling pathway in NSCLC patients are point mutations, amplification, exon 14 skipping mutations, and fusion [14].

The initial preclinical link between MET and resistance to third-generation EGFR-TKIs came from our observation that an EGFR mutant (EGFRm) NSCLC cell line with acquired resistance to erlotinib (HCC827/ER), which has amplified MET gene and hyperactivated MET protein, was cross-resistant to both osimertinib and rociletinib. Moreover, HCC827 cells with acquired resistance to osimertinib (HCC827/AR) also possessed amplified MET gene and hyperactivated MET protein, evidenced by increased p-MET levels in comparison with the parental cell line, and were resistant to not only rociletinib, but also erlotinib [15]. In agreement, inhibition of MET with either a small molecule MET inhibitor or a genetic knockdown of MET expression restored the ability of osimertinib to effectively inhibit the growth of both HCC827/ER and HCC827/AR cells in vitro and in vivo and to inactivate ErbB3 or suppress ErbB3 phosphorylation [15]. Our findings together suggest that MET gene amplification and protein hyperactivation are likely a common resistance mechanism to both first- and third-generation EGFR-TKIs. Moreover, our results also suggest that monotherapy with osimertinib or other third-generation EGFR-TKIs will likely be ineffective for the treatment of EGFRm NSCLCs with acquired resistance to first-generation EGFR-TKIs due to MET amplification and/or protein hyperactivation.

Consistently, several recent studies have generated similar observations. HCC827 cells resistant to erlotinib, which were established in a different laboratory, exhibited MET amplification with increased protein expression and were resistant to osimertinib [16]. Similarly, H1975-P1 cells resistant to AC0010 derived from resistant H1975 xenografts in nude mice after treatment for over 3 months or selection with AC0010 possessed overexpressed MET gene with increased levels of both MET protein and p-MET and were cross-resistant to afatinib, osimertinib, and rociletinib [17]. In a naquotinib-resistant clone (PC-9/NaqR2) derived from the EGFRm PC-9 cell line, MET amplification was also detected accompanied with elevated levels of both MET and p-MET. This resistant line was cross-resistant to gefitinib but sensitive to the combination of naquotinib with a MET inhibitor (crizotinib or SGX532) [18].

MET dysregulation in human cancer tissues can be detected at the gene level (e.g., amplification) and at the protein level as discussed below. Beyond, MET alterations in ctDNA should be another way for detecting the dysregulation (Table 1).

In line with our preclinical findings, clinical detection of MET amplification in EGFRm NSCLCs after relapse from osimertinib or other third-generation EGFR-TKIs was also reported. An early case report documented that a patient who developed resistance to osimertinib after a confirmed partial response for 9 months had a high level of MET amplification post-osimertinib treatment [21]. This observation has been subsequently confirmed by several clinical studies with different cohorts of patients although the frequencies of MET amplification have varied.

While only one case of MET amplification (4%) was detected among 25 NSCLC patients positive for EGFR T790 M that developed resistance to osimertinib [22], other studies have detected much higher frequencies of MET amplification. Le et al. [16] reported that 5 cases of MET amplification (14%) were detected among 42 cases of progression following treatment with osimertinib. Piotrowska et al. [23] analyzed tissue biopsies from 32 osimertinib-resistant EGFRm NSCLC patients and detected 7 (22%) carrying MET amplification, but only 6 patients (19%) with acquired EGFR C797S. Another study by Oxnard et al. [24] documented that among 41 patients who developed resistance to osimertinib and underwent biopsy after relapse, 4 cases (10%) of MET amplification were detected. Analysis of plasma samples from 73 patients with resistance to osimertinib second-line treatment in the large phase III clinical study AURA3 showed that MET amplification was the most common (19%) resistance mechanism, followed by EGFR C797 secondary mutation (15%), with 10 cases of C797S and 1 case of C797G [25].

In a cohort of Chinese NSCLC patients with T790 M enrolled in the AURA trial, 5 (50%) of 10 patients assessed for MET amplification were positive; however, C797S was detected only in two (17%) of 12 assessed patients [26]. In a different study with a cohort of 13 Chinese NSCLC patients who developed disease progression after osimertinib, 4 cases of MET amplification (31%) were detected [27]. Interestingly, no MET amplification was detected through core needle biopsy and next-generation sequencing (NGS) in another cohort of 9 Chinese patients after progression with osimertinib treatment although C797S or C797G was detected in 5 patients [28]. Another study with a cohort of 93 Chinese NSCLC patients relapsed from osimertinib treatment reported 5 cases of MET amplification (5.4%). Moreover, two other cases containing rare mutations of MET P97Q and I865F were also detected although the biological functions of these mutations are unknown [29].

Similar observations have been made in studies with other third-generation EGFR-TKIs. Although the EGFR C797S mutation was initially detected in ∼ 32% of patients after relapse from osimertinib treatment [30], analysis of circulating tumor DNA (ctDNA) in 43 EGFRm NSCLC patients resistant to rociletinib treatment revealed < 3% EGFR C797S mutation, but increased MET copy number in 11 patients (26%) [31]. In a more recent study, MET amplification was observed only in 7.6% (5/66) of patients with acquired resistance to rociletinib [32]. Consistently, a low percentage of C797S mutation (4.5%; 3/66) was detected in this study. In a study of 16 EGFRm NSCLC patients with development of resistance to AC0010, MET amplification was detected in only one case (6.25%), but EGFR C797S mutation was not detected [33].

The majority of studies reported so far have focused on the development of resistance to osimertinib or other third-generation EGFR-TKIs as second-line treatment. Information regarding MET amplification in resistance to first-line osimertinib treatment for EGFR mutation-positive advanced NSCLC is limited, largely due to its recent approval for this indication. In the first reported trial, MET amplification was detected in one case (5.3%) among 19 patients with detectable circulating plasma tDNA [34]. In a study analyzing 91 plasma samples by NGS from patients receiving first-line treatment with osimertinib in the phase III FLAURA clinical trial recently presented at the 2018 ESMO annual meeting, the most common acquired resistance mechanism was MET amplification (15%) followed by EGFR C797S mutation (7%) [35].

Our preclinical studies suggest that monotherapy with osimertinib or other third generation EGFR-TKIs will likely be ineffective for the treatment of EGFRm NSCLCs with acquired resistance to first- or second-generation EGFR-TKIs due to MET gene amplification and protein hyperactivation [15]. In the clinic, patients with multiple pre-existing mechanisms (T790 M and MET) experienced inferior responses [31]. Moreover, patients with MET amplification after osimertinib resistance tended to have inferior median progression-free survival (PFS) and median overall survival (OS) than patients without the appearance of or increase in MET amplification [27]. Therefore, we need effective strategies for the treatment of patients with MET-amplified NSCLC that has relapsed from first- or second-generation EGFR-TKI treatment or patients who develop acquired resistance to osimertinib due to MET amplification and protein hyperactivation.

The basic mechanism by which MET amplification causes EGFR-TKI resistance is associated with the activation of EGFR-independent phosphorylation of ErbB3 and downstream activation of the PI3K/AKT pathway, providing a bypass signaling pathway even in the presence of an EGFR-TKI (Fig. 3) [5]. Thus, co-targeting both EGFR and MET would be required to overcome resistance to EGFR-TKIs by MET amplification, as previously suggested [5]. Indeed, our preclinical studies have shown that inhibition of MET with either gene knockdown or small molecule MET inhibitor (e.g., crizotinib) combined with osimertinib very effectively inhibited the growth of HCC827/ER cells and HCC827/AR cells, which both have MET amplification, both in vitro and in vivo [15]. Similar results were also generated in different resistant models with MET amplification in different laboratories [17, 18, 31]. We found that ErbB3 phosphorylation in both HCC827/ER and HCC827/AR cell lines was minimally inhibited by osimertinib alone, but could be fully suppressed when combined with a MET inhibitor both in vitro and in vivo. This was also true for phosphorylation of other proteins including Akt, S6, and ERK1/2. Hence, full suppression of ErbB3 phosphorylation is tightly associated with the enhanced efficacy of osimertinib and its combination with MET inhibition against the growth of EGFR-TKI-resistant cell lines with MET amplification [15].

A similar attempt has been made in the clinic. It was reported that combinatorial treatment of a first/third-generation EGFR-TKI and crizotinib was tested in two patients with newly acquired MET amplification after osimertinib resistance. Partial responses were achieved both clinically and radiographically [27]. A recent case report also shows that a patient with NSCLC harboring EGFR L858R mutation had emergent MET amplification after disease progression on erlotinib and had a sustained partial response to a combination of full-dose osimertinib and crizotinib with excellent tolerance [36]. Therefore, the current preclinical and clinical studies warrant further investigation of MET inhibition combined with osimertinib or other third-generation EGFR-TKIs for the treatment of EGFRm NSCLCs with MET amplification caused by treatment with first- or second-generation EGFR-TKIs or with third-generation EGFR-TKIs (Fig. 4).

Beyond MET inhibition as we discussed above, our recent preclinical studies suggest that MEK inhibition with a small molecule MEK inhibitor such as trametinib (GSK1120212) is also a very effective strategy in overcoming MET-mediated acquired resistance to osimertinib [37]. Different MEK inhibitors including trametinib, selumetinib (AZD6244), and PD0325901 were all very effective when combined with osimertinib in inhibiting the growth of HCC827/AR cells in vitro or tumors in vivo including induction of apoptosis [37]. The advantage of this therapeutic regimen over MET inhibition is its potent efficacy against not only osimertinib-resistant cells with MET amplification, but also other resistant cell lines with different underlying mechanisms including C797S mutation, which are not responsive to the combination of osimertinib and MET inhibition based on our results [37]. This is important in the clinic if this therapeutic strategy is active against acquired resistance to third-generation EGFR-TKIs regardless of their underlying mechanisms (Fig. 4).

MET inhibitors can be divided into three categories: the small molecule MET receptor inhibitors (e.g., crizotinib, tivantinib, savolitinib, tepotinib, cabozantinib, and foretinib) (Fig. 5), the MET receptor monoclonal antibodies (e.g., onartuzumab), and antibodies against its ligand HGF (e.g., ficlatuzumab and rilotumumab) [38]. Some MET inhibitors have been tested in the clinic against NSCLCs combined with a first- or second-generation EGFR-TKI (Table 2). Mixed outcomes have been generated depending on whether patient populations were selected based on MET status. In general, these combinations did not show improved efficacies in the treatment of unselected NSCLC patient populations. However, some positive results were generated in patient populations selected for MET amplification or overexpression as highlighted below. Therefore, MET may still remain a rational target for therapy in patients with EGFR TKI-resistant and MET-amplified NSCLCs [39].

Osimertinib is now an FDA-approved drug for the treatment of EGFRm NSCLC with T790 mutation after relapse from first- or second-generation EGFR-TKI treatment (second line) and for the therapy of NSCLCs with activating EGFR mutations. However, about 20% of these patients do not respond well to osimertinib. Based on our preclinical findings, NSCLCs with MET amplification or protein overexpression/hyperactivation are unlikely to respond to osimertinib or other third-generation EGFRR-TKIs. We predict that most of these non-responders are likely to have MET amplification and/or protein hyperactivation. Therefore, it may be necessary to detect MET status before osimertinib treatment. MET-amplified EGFRm NSCLCs are likely to be insensitive to osimertinib or other third-generation EGFR-TKIs.

MET amplification and MET protein expression are usually detected in the clinic. However, there is no study that detects phosphorylated MET (p-MET), which represents activated MET protein, in EGFRm NSCLC tissues or those with acquired resistance to EGFR-TKIs and its impact on patient response to EGFR-TKIs. In our preclinical studies, MET-amplified EGFRm NSCLC cell lines possess not only high levels of MET, but also elevated levels of p-MET [15]. Therefore, the detection of p-MET and its impact as a predictive marker for osimertinib-based therapy against EGFRm NSCLCs should be explored.

In EGFRm NSCLC patients with MET amplification and/or hyperactivation or patients relapsed from osimertinib due to MET amplification and/or hyperactivation, combinatorial therapy with a MET or MEK inhibitor may be explored based on preclinical and some clinical pilot studies (Fig. 4). Currently, there is an ongoing clinical trial that tests the efficacy of osimertinib in combination with savolitinib in patients with EGFRm+ and MET+, locally advanced or metastatic NSCLC who have progressed following treatment with osimertinib (https://​clinicaltrials.​gov/​ct2/​show/​NCT03778229) (Table 2). In this trial, MET+ is defined as a high expression of MET (by IHC) and/or increased MET gene copy number (by FISH). We anticipate more similar trials coming in the near future.

The recent development of immunotherapies that target programmed death ligand-1(PD-L1) or programmed death-1 (PD-1) has shown dramatic success in some lung cancer patients [46‐48]. However, these immune checkpoint inhibitors were poorly effective in NSCLC patients with EGFR mutations [49]. Recent data show that treatment with MET inhibitors counteracts the induction of PD-1 ligands by interferon-γ in MET-amplified cancers [50]. Whether combining an anti-MET drug with a PD-1 or PD-L1 blockade is a potential strategy against EGFR-mutant NSCLCs relapsed from osimertinib due to MET amplification and/or hyperactivation needs further investigation.