EGFR inhibition in non-small cell lung cancer: current evidence and future directions

The epidermal growth factor receptor (EGFR), a 170 kD transmembrane protein consisted of a N-terminus extracellular ligand-binding site, a hydrophobic transmembrane domain, and a C-terminus intracellular region with tyrosine kinase activity, is the first of the ErbB family of receptor tyrosine kinases (RTKs) [1, 2]. The other members include ErbB2 (HER2/neu), ErbB3 (HER3), and ErbB4 (HER4). These receptors trigger downstream signaling pathways which lead to multilayered, complex interactions resulting in combinatorial responses. Disruption of these pathways was found to cause malignant transformation [1]. The EGFR is activated through ligand-induced homo or heterodimerization of the receptor with other receptors of the ErbB family under physiologic conditions, but can also be activated due to receptor over-expression, increase of EGFR gene copy number, and activating mutations [3]. EGFR activation has been shown to play a key role in tumor cell proliferation, apoptosis, tumor-induced angiogenesis, metastasis, and DNA damage repair after cytotoxic insults [1, 4]. This makes it an attractive target in cancer therapy; and its inhibition a strategy for augmentation of the efficacy of chemotherapy and radiotherapy. Since the initial discovery of the EGFR in 1962, a class of antibodies blocking the EGFR’s extra-cellular ligand binding site to prevent receptor activation, and to down-regulate EGFR expression at the cell surface through antibody mediated receptor dimerization; and low molecular weight tyrosine kinase inhibitors, which competes with ATP to bind to the intra-cellular tyrosine kinase portion of the receptor to abrogate the receptor’s catalytic activity to activate downstream signaling pathways, have been developed [1, 4–6]. Among them, reversible small molecule (SM) tyrosine kinase inhibitors (TKIs), Gefitinib and Erlotinib, and the mono-clonal antibody (mAb) against EGFR, Cetuximab, have been the most thoroughly investigated. Both Cetuximab and the SM TKIs have been shown to have significant antitumor activity in various EGFR over-expressing cancers, and to enhance the potency of chemotherapy or radiotherapy in various pre-clinical and early phase I or II clinical studies [1, 6–8].

As the leading cause of cancer mortality worldwide, 50-80% of non-small cell lung cancer (NSCLC) is associated with EGFR overexpression; with 65% of them also found to have an increased EGFR gene copy number [9–11]. Given the high percentage of NSCLC being diagnosed in the advanced stage with a poor survival outcome, EGFR inhibition alone or combined with other approaches in the treatment of NSCLC become a very attractive strategy which has been validated in pre-clinical and preliminary clinical trials [12–14]. This led to further phase III randomized studies assessing the efficacy of combining a SM TKI or Cetuximab with chemotherapy as the first line treatment of stage IIIB-IV NSCLC; and some feasibility/phase I & II studies on their combination with radiotherapy, or chemo-radiation in the treatment of local-regional disease. In the following sections, we will describe the current clinical evidence on the utility of TKIs and Cetuximab with chemotherapy in the treatment of advanced stage NSCLC, their role in the treatment of loco-regionally confined NSCLC, such as radiosensitization; patient selection, and potential strategies of further enhancing their efficacy.

Numerous phase III studies on TKI alone or combined with chemotherapy as first line treatment for stage IIIB-IV NSCLC have been conducted since Gefitinib has been approved by the Food and Drug Administration (FDA) in 2003, and Erlotinib has been approved by the FDA in 2004 for chemotherapy-resistant stage IIIB-IV NSCLC [14–23]. In 2004, two large phase 3 randomized studies evaluating the benefit of adding Gefinitib to chemotherapy as first line treatment for stage IIIB-IV NSCLC have been reported [15, 16]. In INTACT 1, 1093 patients were randomized to placebo, Gefitinib 250 mg/day, and Gefitinib 500 mg/day after up to six cycles of Cisplatin and Gemcitabine. No differences in objective response rate (ORR), progression free survival (PFS), or overall survival (OS) was found between the 3 arms after median follow up of 15.9 months. The treatment was very well tolerated with no difference in the incidence of interstitial lung disease (ILD) reported and an overall incidence of ILD <1% [15]. In a parallel study, INTACT 2, similar findings were obtained after randomizing 1037 patients to placebo, Gefitinib 250 mg/day, and Gefitinib 500 mg/day after up to 6 cycles of Carboplatin and Paclitaxel in as first line treatment for stage IIIB/IV patients [16]. The addition of Gefitinib was again very well tolerated with only 0.9%, 2.1%, and 1.5% of ILD found in the three arms, respectively. In the subsequent year, a randomized trial (TRIBUTE) comparing first line Carboplatin and Paclitaxel up to 6 cycles and the same regimen combined with concurrent and maintenance Erlotinib (150 mg/day) in 1079 patients [17]. In this study, no difference in ORR, PFS, or OS was found between the two arms. However, a survival benefit from maintenance Erlotinib was found in patients who lived longer than 4 months, which suggested a benefit from maintenance therapy with Erlotinib. There were 5 cases of ILD-like in the Erlotinib arm (1.0%) and 1 on the chemotherapy arm (0.2%). All cases were fatal. Despite the overall negative findings in this study, never smokers, who tend to be younger, female, and have adenocarcinomas when compared with prior/current smokers, were found to have a median survival of 22.5 months with Erlotinib treatment and only 10.1 months when treated with placebo.

Despite the disappointment with the INTACT and TRIBUT studies, the importance of selecting patients who will respond to TKIs has been recognized, which is suggested by the prolonged survival observed in never smokers [16, 17]. This concept is also supported by the findings of in-frame deletions in exon 19 and missense mutations in exon 21 of the EGFR gene which lead to increased signaling through the EGFR pathway, and increased drug affinity at the EGFR’s intra-cellular tyrosine kinase ATP-binding site [24, 25]. Furthermore, the above mentioned EGFR mutations are more often found in never-smokers as well [17].

While effective only in EGFR mutation positive patients as first line treatment, both Gefitinib and Erlotinib have been shown to achieve similar ORR, PFS, and OS when compared with 2^nd line chemotherapy in randomized phase 3 studies [28, 29]. EGFR FISH-negative tumors were found to have a higher risk of disease progression after Erlotinib in one study [29]. When compared with placebo as maintenance therapy, Erlotinib treatment led to increased PFS, and EGFR expression was significantly correlated with PFS [30]. In addition, EGFR mutation s were associated a large improvement in PFS with the addition of Erlotinib as maintenance therapy.

While a large amount of clinical evidence on how to most effectively incorporate EGFR TKIs into the management of advanced NSCLC has been generated, the clinical experience with mono-clonal antibodies in advanced NSCLC is just emerging in recent years. Cetuximab as a single agent generated minimal response in advanced NSCLC [31]. However, its addition to first line Cisplatin and Vinorelbine was shown to increase tumor response in stage IIIB-IV NSCLC when compared with the same chemotherapy alone in a phase II study [32]. Two randomized phase III studies on the combination of Cetuximab with chemotherapy as first line treatment have been reported subsequently [33, 34]. In the FLEX study, which randomized 1125 patients with stage IIIB-IV NSCLC to Cetuximab delivered concurrently with up to 6 cycles of Cisplatin and Vinorelbine and until disease progression and the same chemotherapy alone, the addition of Cetuximab led to significantly increased ORR (35% vs. 29%, p = 0.010) and median survival (11.3 vs. 10.1 months, p = 0.044). Only 10% of the patients experienced an acne-like rash on the Cetuximab arm, and no >grade 3 acne-like rash was observed in this study. The superior response associated with the addition of Cetuximab to first line chemotherapy was confirmed in another phase III randomized study (BMS 099) randomizing 676 patients to Carboplatin, plus a taxane or Caboplatin-based chemotherapy combined with Cetuximab, which is given till disease progression. However, no survival benefit was demonstrated in that study (Table 1). Only 10.5% grade 3 or 4 acne-like rash was observed in this study.

Among phase II-III studies evaluating the efficacy of Cetuximab, only the EGFR gene copy number and EGFR mutation were found to be potential biomarkers for this mAb’s efficacy (Table 2). Among the two large phase 3 randomized studies (BMS099 & FLEX trials), the EGFR mutation was the only biomarker found to be predictive of superior survival after either chemotherapy combined with Cetuximab or chemotherapy alone in the FLEX trial [44, 45].

Based on the evidence generated from the studies described thus far, the EGFR mutation status appear to be the most consistently found predictive biomarker for EGFR TKI or mAb efficacy in advanced NSCLC while other potential markers, such as KRAS mutation and EGFR gene copy number, need to be further evaluated in larger cohort of patients in the future.

The EGFR has been known to play a key role in the activation of DNA repair and anti-apoptotic proteins, as well as tumor cell repopulation after irradiation [8, 46]. Upon irradiation, increased levels of EGFR auto-phosphorylation has been observed, leading to the activation of the EGFR and its downstream signaling pathways [8]. Furthermore, its expression has been shown to be correlated with radio-resistance and poor outcome after radiotherapy in pre-clinical and clinical studies [8, 47]. Thus, adding an EGFR TKI or mAb to inhibit EGFR activation induced by irradiation become a very sound strategy in enhancing the radiocurability of EGFR over-expressing cancers. This has been successfully validated in the treatment of locally advanced squamous cell carcinoma of the head and neck. In a randomized phase 3 study comparing concurrent Cetuximab plus radiotherapy and radiotherapy alone, a 9.2% 5 year overall survival benefit (45.6% vs. 36.4%, p <0.05) was observed with the addition of cetuximab [48]. As an EGFR over-expression cancer, locally advanced NSCLC has been treated with concurrent chemo-radiation with a 5-year survival of approximately 15%, and 5 year loco-regional control of approximately 70% based on major randomized studies [49]. Despite the poor outcome observed, concurrent chemoradiation is often associated with significant toxicity, which prevents further radiation dose escalation. Therefore, combining an EGFR TKI or mAb with radiotherapy or chemoradiation can potentially be a more effective strategy in the treatment of locally advanced NSCLC.

A limited number of studies investigating the radiosensitizing effects of a mAb have been reported as well (Table 3). Among them, several have translated into median survival rates comparable to what has observed after concurrent chemoradiation [64, 65, 69–72]. The addition of Cetuximab or Nimotuzumab to radiotherapy or chemoradiation appear to be well tolerated with >grade 3 pneumonitis reported only in cases which had excessive amount of radiation dose to the normal lung tissue [66–72]. The results from RTOG 0617, a confirmatory intergroup trial evaluating the addition of cetuximab to chemoradiation in stage III NSCLC are eagerly anticipated.

The findings from large phase III randomized studies demonstrated that patient selection is essential in the application of EGFR TKI in the treatment of stage IIIB-IV NSCLC. Patients who harbor a activating EGFR mutation usually responds very well to treatments including EGFR TKI or mAb alone or combined with chemotherapy and/or radiotherapy [35, 36, 38–42, 45, 55]. This further confirmed previous findings in phase II studies, which demonstrated higher response rates to EGFR TKIs in patients with activating EGFR mutations than that found after platinum-based chemotherapy, but very poor response in patients with only wild type EGFR expression [73]. The response associated with EGFR mutations is thought to be due mainly to an “addictive” dependence of tumor cells on the EGFR signaling pathway for survival [73]. However, response to the EGFR TKIs may be short lived due possibly to a secondary mutation in the EGFR (T790M), or increased signaling through pathways downstream to the EGFR without its activation [26, 27, 74]. T790M may increase the GTP affinity in the tyrosine kinase domain of the EGFR, or hindrance to TKI binding to the tyrosine kinase domain [75, 76]. As shown by Su et al., T790M appears to be more prevalent in patients who has not had any TKI treatment (25.2% - 31.5%), and it becomes even more prevalent after TKI treatment (83.3%) [76]. This finding suggests that TKI treatment selects for NSCLC cells with T790M mutation, and new strategy needs to be developed to overcome T790M related resistance. Further shown by Su et al., NSCLC patients who harbor mutations in exons 19, 20, and 21 can still benefit from TKI, with superior PFS to that in patients without any of these mutations [76]. Therefore, TKI is still indicated in patients with activating EGFR mutations on exons 19, and 21 with or without T790M identified while additional treatment should be added to overcome resistance due to T790M. Resistance due to T790M can potentially overcome by second generation irreversible EGFR TKIs, which appear to be well tolerated in early clinical studies [5]. Although did not demonstrated an overall survival benefit, Afatinib, an irreversible EGFR, Her-2, and ErbB4 blocker has led to significant PFS in patients who progressed after a 12 week course of Gefitinib or Erlotinib when compared to placebo in a phase 2b/3 trial [77]. This agent has been shown to be highly effective in adenocarcinoma harboring activating mutations on exon 19 and 21 when compared with Cisplatin and Pemetrexed (CP) approximately doubling the response rate and PFS achieved with CP, which makes first line treatment of choice in this cohort of patients [78].

Increased signaling through the PI3K pathway through MET-trigger ErbB3 signaling in TKI-resistant NSCLC cells overexpressing MET oncogene has been shown to be a classic example of acquired resistance through bypassing the EGFR, which can be overcome through dual EGFR and MET inhibition [74]. This strategy has been shown to be effective in pre-clinical studies. As shown by Nakagawa et al., WZ4002, a mutant EGFR-TKI, and E7050, a mutant selective dual inhibitor of Met and VEGFR-2, were able to inhibit tumor growth in Erlotinib resistant NSCLC cells in vitro and in vivo when given together with successful inhibition of the EGFR, Met, and their downstream PI3K-AKT pathway [79]. The combination of 2^nd generation TKIs and MET inhibitors with radiotherapy and/or chemotherapy still need to be further investigated in future studies. In addition to MET signaling stimulated PI3K signaling, BRAF mutation in the RAS-RAF-MEK-MAPK pathway, and loss of PTEN expression may also lead to acquired resistance to an EGFR TKI [80, 81].

As a another potential predictive biomarker for a lack of response to EGFR TKIs in NSCLC, KRAS mutation was only shown to be associated with treatment outcome in selected studies [38, 41]. It is observed in approximately 30% of adenocarcinomas [82]. Its mutated form often leads to a gain of function in its downstream signaling pathway, in particular, the Raf-MEK-ERK pathway. This subsequently increases MAPK signaling and tumor cell proliferation. In addition, the mutated KRAS also leads to increased TGF-α expression, which further stimulates the EGFR [82]. Although not consistently demonstrated in major randomized phase III studies, KRAS mutations was found to be significantly associated with an absence of response to TKIs in NSCLC in a meta-analysis of 17 studies [83]. The role of KRAS mutation in major studies evaluating the efficacy of Cetuximab has not been clearly shown. Thus, the role of KRAS mutation as a predictive biomarker for poor response to EGFR inhibition needs to be further elucidated.

Due to differences of the EGFR inhibition mechanism by the mAb, which can generate antibody dependent cellular cytotoxicity; and TKIs ability to inhibit multiple types of ErbB receptors, the role of dual blockade should be explored further in the clinical as they have been shown to increase the potency of EGFR inhibition in vitro and in vivo [1, 84]. Targeting downstream signaling pathways to the EGFR directly can also be a potential approach of augmenting the antitumor activity of radiotherapy and chemotherapy with EGFR inhibitors [85, 86]. At last, simultaneously blocking the EGFR, its downstream signaling pathway, such as the PI3K pathway, and parallel signaling pathways alone, or combined with chemotherapy and radiotherapy, may lead to maximal antitumor activity and warrants further investigation in the future.