I. Okamoto, Department of Medical Oncology, Kinki University School of Medicine, 377-2 Ohno-higashi, Osaka-Sayama, Osaka 589-8511, Japan Tel: +81 72 366 0221 Fax: +81 72 360 5000 E-mail: firstname.lastname@example.org
The discovery that signaling by the epidermal growth factor receptor (EGFR) plays a key role in tumorigenesis prompted efforts to target this receptor in anticancer therapy. Two different types of EGFR-targeted therapeutic agents were subsequently developed: mAbs, such as cetuximab and panitumumab, which target the extracellular domain of the receptor, thereby inhibiting ligand-dependent EGFR signal transduction; and small-molecule tyrosine kinase inhibitors, such as gefitinib and erlotinib, which target the intracellular tyrosine kinase domain of the EGFR. Furthermore, recent clinical and laboratory studies have identified molecular markers that have the potential to improve the clinical effectiveness of EGFR-targeted therapies. This minireview summarizes the emerging role of molecular profiling in guiding the clinical use of anti-EGFR therapeutic agents.
KRAS mutations and sensitivity to therapy with mAb to epidermal growth factor receptor in colorectal cancer
Cetuximab is a chimeric mouse–human mAb that targets the extracellular domain of the epidermal growth factor receptor (EGFR) and thereby blocks downstream signal transduction via the phosphatidylinositol 3-kinase/Akt and Ras/Raf/mitogen-activated protein kinase pathways (Fig. 1). Because it is an antibody (IgG1 isotype), cetuximab may also induce antibody-dependent cell-mediated cytotoxicity, although the clinical relevance of antibody-dependent cell-mediated cytotoxicity with regard to the antitumor efficacy of cetuximab is likely to be relatively low .
Cetuximab exhibits single-agent activity against metastatic colorectal cancer (mCRC) refractory to previous chemotherapies . An analysis of 80 patients with mCRC, (who had previously undergone treatment) enrolled in a study of cetuximab monotherapy found a mutation rate of 38% for the proto-oncogene KRAS in tumor specimens and discovered that such mutations were associated with resistance to cetuximab, showing an overall response rate of 0 versus 10% for mutation-positive and mutation-negative patients, respectively . More recently, a trial comparing cetuximab + best supportive care (BSC) with BSC alone in 394 patients with mCRC after failure of prespecified chemotherapy found a KRAS mutation rate of 69% . Analysis of the cetuximab + BSC arm (n =198) of the trial, however, revealed that only 1.2% of the KRAS mutation-positive patients (n =81), compared with 12.8% of patients with wild-type KRAS (n =117), responded to cetuximab monotherapy (Table 1). Furthermore, KRAS mutations were significantly associated with a shorter progression-free survival (PFS) (7.2 versus 14.8 weeks) and a shorter overall survival (OS) (4.5 versus 9.5 months) among the cetuximab-treated patients (Table 1). No survival benefit was observed in patients whose tumors harbored wild-type KRAS compared with those whose tumors were positive for mutant KRAS in the BSC-only arm (OS of 4.8 versus 4.6 months, respectively), revealing a lack of prognostic value for KRAS status (Table 1). These data thus indicate that the prolonged survival of patients with tumors harboring wild-type KRAS was a result of the benefit from cetuximab monotherapy rather than of a more favorable prognosis for the subset of patients treated with cetuximab + BSC.
Table 1. Activity of therapy with monoclonal anti-EGFR in patients with mCRC, based on the KRAS mutation status. MT, mutant; RR, response rate; WT, wild-type.
Similar findings, in terms of clinical efficacy among patients with tumors harboring wild-type KRAS, were obtained in a retrospective analysis of a trial of panitumumab in patients with mCRC . Panitumumab, a fully human mAb targeted to the extracellular domain of EGFR, is of the IgG2 isotype, and its antitumor effects are probably attributable to inhibition of EGFR signaling rather than to antibody-dependent cell-mediated cytotoxicity. The KRAS status was assessed in 92% (n =427) of tumor samples from patients enrolled in the phase III registration trial of panitumumab versus BSC, and KRAS mutations were detected in 43% of the tested tumors. Furthermore, patients whose tumors harbored wild-type KRAS exhibited a 17% response rate in the panitumumab-monotherapy arm, whereas those with KRAS mutation–positive tumors failed to respond to panitumumab (Table 1). The median PFS time was significantly longer in panitumumab-treated patients with wild-type KRAS than in those with mutant KRAS (12.3 versus 7.4 weeks) (Table 1). The median OS time in panitumumab-treated patients with wild-type KRAS was also longer than that in those with mutant KRAS (8.1 versus 4.9 months) (Table 1). On the basis of these results, the European Medicines Agency approved the use of panitumumab only for mCRC patients with tumors possessing wild-type KRAS. This was the first approval of an agent for mCRC that was based on patient-specific molecular profiling, opening a new vista for genotype-directed therapy in this disease.
KRAS mutation as a mechanism of resistance to EGFR-targeted therapy
The KRAS protein is localized to the inner surface of the cell membrane. The binding of ligand to EGFR induces receptor dimerization and consequent conformational changes that result in activation of the intrinsic tyrosine kinase, receptor autophosphorylation and a transient activation of RAS GTPases (Fig. 2). Activated RAS targets various downstream effectors to exert pleiotropic cellular effects. KRAS is the most frequently mutated oncogene in several types of human cancer. These mutations, most of which are located in codons 12 and 13, occur in up to 40% of patients with mCRC . Activating mutations of KRAS result in activation of the mitogen-activated protein kinase signaling cascade, independently of EGFR activation. Mutation of KRAS thus bypasses the need for ligand binding to EGFR and results in constitutive activation of signaling downstream of the receptor, which, in turn, promotes cell proliferation and metastasis as well as inhibiting apoptosis. These effects of KRAS mutation support continued cancer cell survival, even in the presence of upstream EGFR inhibition [7,8].
EGFR mutations and sensitivity to EGFR-tyrosine kinase inhibitor therapy in non–small cell lung cancer
Imatinib was designed to compete with ATP at the ATP-binding site within the tyrosine kinase domain of ABL, which is activated as a result of the chromosomal translocation that gives rise to the BCR–ABL fusion gene in chronic myeloid leukemia (CML). The marked success of imatinib in the treatment of CML provided compelling evidence for the effectiveness of small-molecule tyrosine kinase inhibitors (TKIs) and triggered the development of this class of agents for targeting growth factor receptors frequently expressed in epithelial cancers . Two such inhibitors of the tyrosine kinase activity of EGFR (EGFR-TKIs), gefitinib and erlotinib, compete with ATP for binding to the tyrosine kinase pocket of the receptor, thereby inhibiting receptor tyrosine kinase activity and EGFR signaling pathways (Fig. 1). Early clinical studies showed that a subset of patients with non-small cell lung cancer (NSCLC) experienced a rapid, pronounced and durable response to single-agent therapy with EGFR-TKIs. Subsequent retrospective analysis of clinical data consistently demonstrated that a clinical response to these agents is more common in women than in men, in Japanese people than in individuals from Europe or the USA, in patients with adenocarcinoma than in those with other histological subtypes of cancer, and in individuals who have never smoked than in those with a history of smoking . These clinical observations paved the way for translational research that aimed to identify, at the molecular level, patients who might benefit from such therapy. In 2004, three groups in the USA made the landmark observation that NSCLC patients who experienced a dramatic response to gefitinib or erlotinib commonly harbored somatic mutations of the drug’s target, EGFR [11–13]. Indeed, EGFR mutations are present more frequently in women, in individuals of East Asian ethnicity, in patients with adenocarcinoma, and in never-smokers, the same groups identified clinically as most likely to respond to treatment with EGFR-TKIs.
Several prospective clinical trials of gefitinib or erlotinib for treatment of NSCLC patients with EGFR mutations have been performed to date, revealing radiographic response rates from 55 to 91% [14–21] (Table 2). These values are much higher than those historically observed with standard cytotoxic chemotherapy for advanced NSCLC. As the data accumulate, an improvement in OS, conferred by treatment with these drugs, is also expected in patients harboring EGFR mutations. It was not possible to evaluate OS in most of the clinical trials at the time of publication because the number of patients was not sufficiently large and the follow-up period was not long enough to obtain precise estimates of survival outcome. Our group has recently analyzed updated individual patient data from seven Japanese prospective phase II trials of gefitinib monotherapy, including a total of 148 EGFR mutation–positive individuals . The Iressa Combined Analysis of Mutation Positives study showed that gefitinib confers a highly favorable PFS (9.7 months) and OS (24.3 months) in such patients. The median survival time of approximately 2 years, achieved in patients with EGFR mutation-positive NSCLC by treatment with EGFR-TKIs, supports the notion that this group of patients constitutes a clinically distinct population. The substantial clinical benefits of treatment with EGFR-TKIs in EGFR mutation-positive NSCLC patients raise the question of whether first-line treatment with EGFR-TKIs might be more beneficial than standard cytotoxic chemotherapy in this genotype-defined population. In the Iressa Combined Analysis of Mutation Positives study, we performed an exploratory comparison between gefitinib and systemic chemotherapy in the first-line setting. We found that first-line gefitinib treatment yielded a significantly longer PFS than did systemic chemotherapy in EGFR mutation-positive NSCLC patients, supporting the use of gefitinib as an initial therapy in this patient population. This finding is consistent with a subset analysis of a recently completed randomized phase III study, known as the Iressa Pan-Asia Study, which showed that first-line treatment with gefitinib significantly improved the PFS of EGFR mutation-positive patients with advanced NSCLC compared to treatment with carboplatin and paclitaxel. We are currently performing phase III randomized studies comparing platinum-based chemotherapy with gefitinib in chemotherapy-naïve NSCLC patients with EGFR mutations. Such ongoing phase III clinical trials will help to determine whether gefitinib monotherapy becomes the standard of care for EGFR mutation-positive NSCLC.
Table 2. Prospective study of EGFR-TKI monotherapy for NSCLC patients with EGFR mutations. RR, response rate.
EGFR mutation as a mechanism underlying sensitivity to therapy with EGFR-TKIs
The discovery of EGFR mutations has led not only to the identification of a molecular predictor of sensitivity to EGFR-TKIs but also to examination of the biological effects of such mutations on EGFR function. Deletions in exon 19, and a point mutation (L858R) in exon 21, are the most common EGFR mutations as well as the most extensively evaluated to date. Initial studies, based on transient transfection of various cell types with vectors encoding wild-type or mutant versions of EGFR, showed that the extent of activation of mutant receptors by EGF is more pronounced and sustained than is that of the wild-type receptor . Subsequently, NSCLC cell lines with exon-19 deletions or the L858R point mutation were identified, and the EGFR mutations were found to confer ligand-independent activation of EGFR . We also found that the constitutive activation of endogenous mutant EGFR is attributable to the ability of the receptor to undergo ligand-independent dimerization (Fig. 2) . Introduction of the two most common EGFR mutants into transgenic mice was recently shown to result in the formation of lung adenocarcinomas, demonstrating that expression of these constitutively activated forms of EGFR is sufficient for transformation and required for maintenance of these tumors . These various observations indicate that EGFR mutation-positive tumors are dependent on, or ‘addicted’ to, EGFR signaling for their growth and survival. Similar addiction is evident in BCR/ABL-positive CML and in KIT mutation-positive gastrointestinal stromal tumors, both of which are highly sensitive to imatinib. Exposure of EGFR mutation-positive NSCLC tumors to EGFR-TKIs thus results in EGFR signaling pathways being turned off and the cancer cells undergoing apoptosis. Moreover, EGFR mutations result in repositioning of critical residues surrounding the ATP-binding cleft of the tyrosine kinase domain of the receptor and thereby stabilize the interaction with EGF-TKIs, leading to an increase of ∼ 100-fold in sensitivity to inhibition by EGFR-TKIs compared with that of the wild-type receptor (Fig. 2) [11,25]. These factors combine to render EGFR mutation-positive NSCLC more sensitive to EGFR-TKIs.
Molecular mechanisms associated with acquired resistance to therapy with EGFR-TKIs
Despite the great benefits of EGFR-TKIs in the treatment of NSCLC associated with EGFR mutations, most, if not all, patients ultimately develop resistance to these drugs. The first mechanism to be discovered of such acquired resistance is a secondary mutation, T790M, in the EGFR . To date, this mutation has been found in ∼ 50% of NSCLC tumors from patients who developed acquired resistance to EGFR-TKIs. The position of the T790M mutation within the EGFR is analogous to the positions of mutations in other tyrosine kinases known to result in resistance to imatinib (T315I in ABL, T764I in PDGFRA and T670I in KIT) [27–29]. The conserved threonine residues in these different kinases are located near the kinase active site and appear to be critical for the binding of ATP and the corresponding TKIs. Structural modeling suggests that the T790M mutation of EGFR creates steric hindrance that prevents EGFR-TKIs from interacting with the ATP-binding pocket of the receptor. Furthermore, biochemical analysis showed that, in cells expressing both T790M mutant and wild-type forms of EGFR, EGFR-TKIs are not able to inhibit the phosphorylation of either type of the receptor.
The T790M mutation of EGFR was initially thought to occur during treatment with EGFR-TKIs, given that it was initially identified only in tumor specimens from a patient with NSCLC who relapsed after 24 months of complete remission despite continued gefitinib therapy . However, subsequent development of a highly sensitive detection method, mutant-enriched PCR analysis, and its application to detect the T790M mutation in 280 NSCLC tumor specimens obtained from patients before treatment with EGFR-TKIs, revealed the presence of the mutation in a small proportion of tumor cells in 10 (3.6%) of these specimens . Similarly, a minor proportion of cells harboring a BCR/ABL mutation associated with imatinib resistance was detected in a patient with CML before treatment with this drug; the proportion of mutant cells was later found to have increased after treatment onset and the development of resistance . These observations suggest that a small fraction of NSCLC tumor cells may harbor the T790M mutation of EGFR before treatment with EGFR-TKIs and that these cells come to predominate as a result of their selective proliferation during such treatment, resulting in the development of clinical resistance.
NSCLC tumors that acquire resistance to gefitinib or erlotinib as a result of the EGFR T790M mutation remain dependent on EGFR signaling for their growth and survival. Alternative strategies for inhibiting the activity of the mutant receptors may thus be able to overcome the acquired resistance to EGFR-TKIs. This possibility has prompted the development of second-generation irreversible EGFR-TKIs. These agents are also ATP mimetics, similarly to the reversible EGFR-TKIs gefitinib and erlotinib, but they covalently bind cysteine 797 at the edge of the ATP-binding cleft of the EGFR . Some irreversible EGFR-TKIs have been shown to inhibit EGFR phosphorylation, as well as the growth of NSCLC cell lines harboring the T790M mutation of EGFR [32,33]. Future clinical trials of these irreversible EGFR-TKIs in NSCLC patients with the EGFR T790M mutation are warranted.
Amplification of the gene for the receptor tyrosine kinase MET has also recently been identified as a mechanism of EGFR-TKI resistance, being detected in 22% of tumor samples from NSCLC patients with EGFR mutations who acquired gefitinib resistance . MET amplification confers EGFR-TKI resistance by activating ERBB3 signaling in an EGFR-independent manner. This redundant activation of ERBB3 permits the cells to transmit the same downstream signaling in the presence of EGFR-TKIs. Exposure of EGFR-TKI-resistant NSCLC cells with MET amplification to MET-TKI or EGFR-TKI alone did not inhibit cell growth or survival signaling, given that both EGFR and MET signaling were found to be activated and to be mediated by ERBB3 (also known as HER3) in these cells. However, the combination of both types of TKI overcame resistance to EGFR-TKIs, attributable to MET amplification.
The EGFR T790M mutation and MET amplification account for ∼ 70% of all known causes of acquired resistance to EGFR-TKIs in NSCLC, indicating that other mechanisms of resistance await discovery. It is therefore important to continue to study preclinical models, with regard to which the collection of tumor specimens and establishment of cell lines from patients who have developed EGFR-TKI resistance is key.