Abstract
Esophageal and gastric cancer incidence has been increasing worldwide. Most cases are diagnosed in advanced stages, and current therapy has not been able to improve the modest survival rates after diagnosis. Epidermal growth factor receptor expression has been found to correlate with poor prognosis and aggressive disease in esophagogastric cancers. Targeting these receptors through monoclonal antibody or downstream inhibition of tyrosine kinase has produced some encouraging results. The Trastuzumab for Gastric Cancer (ToGA) trial demonstrated that trastuzumab, a monoclonal antibody directed against the Her-2 receptor improves survival in Her-2 positive advanced gastric and gastro-esophageal cancers. Encouraging results have been reported in the ongoing clinical trials of EGFR-directed therapies in combination with concurrent chemoradiation for the locally advanced esophagogastric cancers. Identification of pertinent biomarkers of efficacy will likely lead to further optimization of EGFR-directed treatment. In the current article, we will be discussing the mechanisms of action, completed phase II/III trials and future of epidermal growth factor targeted-therapy in gastric and esophageal cancers.
Esophageal and gastric cancer incidence all over the world has been rising, and mortality from these cancers combined, is second only to lung cancer. These two cancers combined as per the GLOBOCAN series published by the International agency for Research on cancer in 2008, was estimated to have a cumulative incidence of 1.47 million and a mortality of about 1.14 million (1). Applying the same prediction model, by 2020, the mortality could rise up to 1.56 million. The incidence and mortality rates from these two cancers combined vary widely throughout the world, and current mortality estimates are highest in Eastern Asia, 56% of deaths were estimated from here (1). Incidence in squamous cell cancer of esophagus in USA has been steady or declining in last 3 decades (2). The recent increase in incidence of gastric cardia, gastro-esophageal junction and distal esophagus adenocarcinoma has brought interest in elucidating common etio-pathology for this cancer and they are thought to have different etiologic factors compared to distal gastric and proximal esophageal cancers (3, 4). Changes in lifestyle including obesity, chronic gastro-esophageal reflux disease, and Barrett's esophagus has been implicated in the resurgence of adenocarcinoma as leading esophageal cancer (5-8).
There have been recent advances in surgical techniques, chemotherapy and radiotherapy, both in the adjuvant and neoadjuvant setting but outcome for patients with esophagogastric cancers remains poor (9). In advanced malignancies, median survival beyond one year has not been achievable with any combination chemotherapy (10, 11). Advances in molecular biology have helped us formulate tumor-specific therapies by utilizing the biological differences between normal and cancer cells. Epidermal growth factor receptors (EGFR) are overexpressed in esophagogastric cancers and their inhibition as a potential therapeutic strategy is being explored. Recently, trastuzumab, a monoclonal antibody against human EGFR-2 (Her-2) increased survival of the Her-2-positive advanced gastric or gastro-esophageal junction cancer in combination with first-line systemic chemotherapy. In this review, an overview of EGFR signaling pathway in esophagogastric cancer will be provided, supported by available data from clinical trials.
EGFR Receptor and Signaling
The epidermal growth factor receptor (EGFR) family of receptors, also called HER family of tyrosine kinases, include ERBb1 or EGFR, ERBb2 or Her-2, ERBb3 and ERBb4, are encoded by erb oncogenes (12) and have been implicated in tumor cell growth and differentiation. EGFR is a receptor with an extracellular ligand binding domain (domains I-IV), a transmembrane region, an intracellular domain with tyrosine kinase activity (except in ErbB3), and a tail containing tyrosine residues, required for downstream signaling. Multiple ligands including epidermal growth factor (EGF), transforming growth factor-alpha, amphiregulin, heparin binding–EGF, epiregulin and betacellulin bind and activate the EGFR receptors (13). Ligand-binding induces interaction with a second receptor of same class (homodimerization) or with a different class receptor (heterodimerization). Dimerization is followed by internalization of receptors, which is followed by auto-phosphorylation of the intracellular tyrosine kinase domain resulting in a casade of signaling pathways (12).
The signaling pathway (Figure 1) is a busy roadmap, has multiple channels and includes the Ras/Raf/mitogen-activated protein kinase (MAPK)/cyclin-D1, PI3K/AKT pathway and signal transducers and activators of transcription (STAT) signaling pathways, which are involved in cell proliferation and differentiation (14). The EGFR nuclear signaling independently of the pathways described above, has been implicated in tumor progression (15). After activation by ligand binding and internalization, EGFR receptor binds to importin-beta in the cytoplasm which facilitates its nuclear transport, where it stimulates various gene promoters including cyclin-D1, iNOS, B-myb, Aurora kinase A and COX2 (16-20).
Despite having no soluble ligand, ERBb2 is important because it is the preferred heterodimerization partner of the other ligand-bound family members (21). Overexpression of ERBb2 can lead to spontaneous dimerization and auto-activation of the downstream signaling. It can also amplify the signaling by other ERBb heterodimers by strengthening the ligand binding, receptor cycling and stability (21-23).
Role of EGFR Receptor Family in Esophageal and Gastric Cancer
Esophageal squamous cell carcinomas have been known to have higher frequency of EGFR overexpression in the range of 60-70 % (Table I), particularly, with gene amplification noted to be 28% in a series of resected tumours (24, 25). They have been associated with poor response to chemo-radiotherapy and poor survival (26). The EGFR overexpression has been seen in high-grade dysplasia in Barrett's oesophagus and a progressive increase in percentage could represent a marker for progression to malignancy (27). Genetic mutations in the tyrosine residue of the cytoplasmic domain are uncommon. EGF and TGF-alpha overexpression has been seen in Barrett's esophagus, esophageal adenocarcinoma and esophageal squamous cell carcinoma (13). It is associated with poor survival in squamous cell carcinoma and in particular those with recurrent disease and those undergoing esophagectomy (15).
In gastric cancers, although EGFR amplification has been low, EGFR expression has been comparable to esophageal cancers and it correlates with a poor prognosis (28). In a recently published series, 69 gastric cancer specimens were evaluated for EGFR and HER-2 expression and clinicopathological correlation (29). Her-2 protein was expressed in 42 % of samples and gene amplification or mutation noted in 52 % samples. EGFR expression was seen in 52% of samples with gene alteration in 59%. Gene amplifications were associated with lymph node metastasis and depth of invasion of tumor, with no correlation to other variables evaluated.
In the Trastuzumab for Gastric cancer trial (TOGA) 3,280 patients were screened prior to therapy, 22.1% were found to be positive for Her-2 expression by Fluorescence in situ hybridization (FISH) or Immunohistochemistry (IHC) (30). The intestinal type sub-group of gastric cancers are found to have higher positivity compared to the diffuse type (32 vs. 20.4%). There appears to be discordance in the degree of positivity when evaluated by the IHC method. A number of explanations have been provided including discrepancies in laboratory procedures and heterogeneity of gastric cancers, compared to breast cancer where Her-2 testing is routinely performed. False-negatives observed were up to 50% if same principles as used in latter testing were applied (31, 32). It is now advised that screening for Her-2 in gastric cancers and scoring should be performed as per guidelines, based on the TOGA trial (31).
HER-2 overexpression has been identified in 9-60% of esophageal cancers (33-35). In a recent single-institution retrospective analysis, 15% of esophageal cancers and 28% of GEJ cancers were found to be Her-2 positive (36). HER-2 over expression did not correlate with age, race, gender, tumor location, tumor length, stage, or presence of Barrett's esophagus and treatment outcomes including response, patterns of recurrence, and survival were also not statistically different between the groups. There was an increase in poorly differentiated tumors in the HeR-2-negative group.
Targeting the EpidermalGrowth Factor Receptor (EGFR)
Targeting the EGFR in cancer therapy was initially proposed more than 20 years ago. Out of the various EGFR-targeted agents, the most promising and well-researched are the monoclonal antibodies and the small-molecule EGFR tyrosine kinase inhibitors (TKIs).
EGFR signaling: EGFR is activated by ligand binding and subsequent receptor heterodimerization or homodimerization which results in auto-phosphorylation of tyrosine residues and binding of adaptor molecules like shc, gab -1 to the cytoplasmic domain. Intracellular signaling pathways include RAS/ Raf-1/ MAPK, PI3K/AKT, PLC- γ /PKC pathways which require the adaptor molecules for signaling. Src and STAT pathways are directly activated by phosphorylated receptors. Alternately, the activated receptors can undergo endocytosis followed by importin-mediated translocation to the nucleus, and co-transcriptional activation of key genes like Cox-2, iNOS, aurora kinase-A and cyclin-D1. All pathways lead to changes in gene expression and stimulation of cell proliferation, survival, invasion and metastasis. EGFR manipulation can be approached extracellularly by monoclonal antibodies through inhibiton of ligand binding and intracellularly, by tyrosine kinase inhibitor (TKI), which compete with ATP binding to receptor kinase for activation. Shc, Src homology-2 domain containing transforming protein-1; Grb2, growth factor receptor bound protein-2; Sos son of sevenless; Ras, rat sarcoma; Raf-1, rapidly accelerated fibrosarcoma; MAPK, mitogen activated protein kinase; Gab-1, Grb2 associated binding protein-1; PI3K, phosphoinositide 3 kinase; AKT, protein kinase B; Src, sarcoma gene; FAK, focal adhesion kinase; STAT, signal transducer and activator of transcription; TKI, tyrosine kinase inhibitor; PLC-γ, phospholipase C-γ; phapatidylinositol 4,5-bisphosphate (PIP2); IP3, inositol 1,3,5–triphosphate; DAG, 1,2-diacylglycerol; PKC, protein kinase C; ER, endoplasmic reticulum.
Anti-EGFR monoclonal antibodies
Trastuzumab. Trastuzumab is a humanized monoclonal antibody that binds to the extracellular domain IV of the Her-2/neu receptor and prevents the activation of its intracellular tyrosine kinase (37). Several mechanisms of action have been proposed including antibody-dependent cell cytoxicity (ADCC) activity, activation of natural killer cells and destruction of cancer cells bound to Fc domain of trastuzumab (38, 39). Inhibition of Her-2 shedding in overexpressing breast cancer cells (40), inhibition of PI3K pathway (41), inhibition of angiogenesis (42) and G1 phase cell-cycle arrest (43) are other the postulated mechanisms. Trastuzumab is approved for use in combination chemotherapy as adjuvant for Her-2/neu and node positive breast cancer (44, 45).
The ToGA trial evaluated the role of trastuzumab in combination with chemotherapy in metastatic gastric or gastro-esophageal junction cancers. It was an open-label, international, phase3, randomized trial conducted in 122 centers in 24 countries (46). Out of 3,807 patients initially enrolled with gastric or gastro-esophageal junction cancers, twenty two percent (n=810) were found to be positive for Her-2 if they scored 3+ on IHC or if they were FISH-positive (Her-2:CEP17 ratio ≥2). A total of 290 and 294 patients were randomized to receive either 5-FU or capecitabine with cisplatin with or without trastuzumab respectively. Trastuzumab-exposed patients showed and improved overall response rate (47.3% vs. 34.5%) and median progression-free survival (PFS) (6.7 vs. 5.5 months) as well as median overall survival (OS) (13.8 vs. 11.1 months). Rates of overall grade 3 or 4 adverse events (except for diarrhea in the trastuzumab arm) and cardiac adverse events did not differ between groups. Patients assigned to the trastuzumab-plus-chemotherapy arm had slightly higher rates of diarrhea, stomatitis, anemia, thrombocytopenia and weight loss than did patients assigned to the chemotherapy arm alone. There was also a mild increase in incidence of asymptomatic left ventricular ejection fraction decrease in trastuzumab arm (4.6 vs. 1.1%). With these results, FDA and European medicines agency (EMA) have approved trastuzumab for advanced gastric or gastro-esophageal cancer therapy in previously untreated Her-2-positive patients in combination with chemotherapy.
EGFR receptor expression and gene ampliflication in esophaogastric cancer.
In the only phase I/II study so far conducted in locally advanced esophageal carcinoma, escalating doses of trastuzumab were added to a regimen of weekly cisplatin/paclitaxel in combination with radiation therapy (47). Fourteen out of nineteen patients (74%) were positive for 3+ HER-2 by IHC or gene amplification by FISH. There was 50% survival at 2 years and OS was similar to previous studies without trastuzumab with no increase in toxicity. There is an on going randomized phase III trial investigating efficacy of radiation therapy, paclitaxel, and carboplatin together with or without trastuzumab in treating patients with locally advanced esophageal cancer.
T-DM1, is an antibody-drug conjugate of trastuzumab with microtubule polymerization inhibitor, derivative of maytansine(48). It binds to Her-2, undergoes receptor mediated internalization and DM-1 is released into cytoplasm. It can induce apoptosis and ADCC mediated tumour cytoxicity. It has been shown to inhibit gastric cancer cells in vitro and in vivo in mouse xenograft models (49). There is an ongoing phase II study of T-DM 1, the trastuzumab-emtansine conjugate, in advanced gastric cancer.
Cetuximab. Cetuximab is a chimeric (mouse/human) EGFR-blocking monoclonal antibody, originally derived from a mouse myeloma cell line (50). It blocks the binding of EGF and TGF-alpha to EGFR and inhibits ligand-induced activation of tyrosine kinase receptor. It also stimulates EGFR internalization and effectively removes the receptor from ligand interaction (51). Cetuximab is approved for clinical use in metastatic colorectal cancers with wild-type KRAS status and in locally advanced as well as recurrent/metastatic squamous cell carcinoma of head and neck (52-54). Cetuximab's antitumour activity has been studied extensively in pre-clinical models of esophagogastric cancer. In the gastric carcinoma cell line, NCI-N87, cetuximab down-regulates EGF-induced EGFR and Her-2 phosphorylation, inhibits homodimerization of EGFR, heterodimerisation between EGFR and Her-2 and downstream signaling via the MAPK and AKT pathway (55). In a mouse xenograft model of gastric cancer, cetuximab inhibited the growth of cancer cells and induced apoptosis (56).
Cetuximab in combination with three standard chemotherapy regimens (ECF-C, IC-C, FOLFOX-C) was investigated in a randomized phase-II study performed by the Eastern Cooperative Oncology Group (ECOG1206) in metastatic esophageal and gastro-esophageal junction (GEJ) cancer (57). A total of 245 patients were randomized to receive either one of the regimens, and response rates with the ECF-C, IC-C and FOLFOX-C were 58%, 38% and 51% respectively. FOLFOX-C was found to be least toxic and IC-C least effective. In the only trial with squamous cell cancer of esophagus, combination of cetuximab with cisplatin/5-FU produced 19% overall response in comparison to 13% in the chemotherapy-alone arm with median survival of 9.5 months vs. 5.5 months (58). Cetuximab did not exacerbate grade 3/4 toxicity, except for rash and diarrhea.
Several phase II/III studies have investigated the role of cetuximab in combination with various chemotherapeutic regimens as first-line therapy in advanced gastric and GEJ cancers (Table II). Three of these trials showed better response with cetuximab combination compared to available triple-agent chemotherapy regimens. Cetuximab combined with FUFOX (5-FU, D-folinic acid, oxaliplatin), showed 65% response rate with OS of 9.5 months (59). In another trial, 35 patients received cetuximab with cisplatin, 5-fluoruracil combination, in which OS had improved to 14.5 months with 69% overall response rate (ORR) (60). Oxaliplatin and irinotecan bi-weekly was combined with weekly cetuximab therapy as first line therapy in advanced gastric cancer (61). Although, response rate and OS were comparable to aforementioned trials, toxicities related to therapy were of lower grade. Multiple phase II studies have examined the role of cetuximab in combination with either cisplatin or irinotecan for the second-line treatment of esophagogastric cancers (62-64). Cetuximab increased survival only modestly in this patient population and was well-tolerated in combination with irinotecan (65). Grade 1-2 skin rash was noted in 73% of patients and portended a prolonged OS of 7.1 months (62). In a recently completed study, cetuximab was combined with a modified FOLFIRI regimen (66), as second-line therapy in advanced gastric cancer subjects, ORR of 35 % was noted and median OS was 8.1 months.
Cetuximab as second line, single-agent therapy in metastatic esophagogastric cancers has good tolerability, but minimal clinical activity (67, 68). Chan et al. reported a partial response in 3%, stable disease in 6% of patients treated with single-agent cetuximab with median PFS and OS of 1.6 and 3.1 months, respectively (68). In a phase-II southwest oncology group study, 55 patients were treated with single-agent cetuximab after failure of first-line chemotherapy, the median OS was 4 months (95% CI= 3.2-5.9), and the median PFS was 1.8 months (95% CI= 1.7-1.9). Two patients experienced Grade 4 fatigue, and there was one treatment related death due to pneumonitis (59). In locally advanced esophagogastric cancers, cetuximab has been used in the neo-adjuvant setting along with different chemotherapy backbones such as cisplatin/irinotecan (69), carboplatin/ paclitaxel (70), docetaxel/cisplatin (71) and FOLFOX-4 (bi-weekly 5-fluorouracil/leucovorin/oxaliplatin) (69-73). Complete pathological response rate (pCR) in these studies ranged from 60-70%, with no significant increase in toxicities. Cetuximab combination with cisplatin/irinotecan had a lower complete response rate and higher overall toxicity (54). Results from the first North American cooperative group trial (SWOG 0414) evaluating cetuximab-plus-cisplatin, irinotecan, and thoracic radiotherapy as definitive treatment for locally advanced, unresectable esophageal cancer has been disappointing (74). The study was closed early due to slow accrual, with 21 eligible patients (11 squamous, 10 adenocarcinoma). Concurrent radiation was administered at 1.8 Gy in 28 daily fractions to a total dose of 50.4 Gy, to begin with on day 1 of cycle 3. Chemotherapy regimen consisted of four 21-day cycles of cetuximab on days 1, 8, and 15, cisplatin 30 mg/m2 (days 1 and 8) and irinotecan 65 mg/m2 (days 1 and 8). ORR was 17.6%, with two year OS of 33.3% and PFS of 23.8%. Two deaths occured from the treatment with an overall ten percent treatment-related mortality and 76% of patients developing grade 3/4 toxicities.
Cetuximab combination with definitive chemoradiation for locally advanced esophageal squamous cell cancer in chinese patients, showed good clinical response with an acceptable safety profile. In a single-arm phase II study, 31 patients were enrolled and treated with weekly cetuximab, paclitaxel and cisplatin for 8 weeks in combination with 59.4 Gy radiotherapy (75). Among the 29 response-evaluable patients, 20 (69.0%) had a complete response (CR) with the one and two year PFS rates of 85.5% and 75.1%, respectively.
Matuzumutab. This is a humanized IgG1 anti-EGFR monoclonal antibody that has effective ADCC activity. In a randomized phase 2 trial (MATRIX), patients with advanced esophagogastric cancer were randomly assigned to either matuzumab-plus-epirubicin, cisplatin and capecitabine (ECX) or the same ECX regimen alone (76). The ORR was inferior (31vs. 58%) in the matuzumutab combination group and median OS was worse (9.4 vs. 12.2 months). The maximum tolerated dose (MTD) for combination of matuzumab and ECX from a phase 1 study was 800 mg weekly which was significantly less compared to single agent MTD of 1,600 mg weekly. Optimal biological dose to achieve efficacy may not have been reached in the combination matuzumab and ECX study.
Panitumumab. This fully-humanized IgG2 monoclonal antibody has already been approved along with cetuximab for metastatic colorectal cancer as single agent therapy or in combination (77). It differs from cetuximab, by inducing minimal immunogenic response and inability to stimulate ADCC. Panitumumab is currently being evaluated in a randomized phase 3 trial (REAL-3), in combination with epirubicin, oxaliplatin and capecitabine (78). Patients with untreated advanced esophagogastric adenocarcinoma were randomly allocated to receive up to eight 21-day cycles of open-label EOC (epirubicin 50 mg/m2 and oxaliplatin 130 mg/m2 on day 1 and capecitabine 1,250 mg/m2 per day on days 1-21) or modified-dose EOC plus panitumumab (mEOC+P; epirubicin 50 mg/m2 and oxaliplatin 100 mg/m2 on day 1, capecitabine 1000 mg/m2 per day on days 1-21, and panitumumab 9 mg/kg on day 1) (79). A total of 553 patients were enrolled, and the experimental arm (mEOC+P) had a poorer OS of 8.8 months compared to the EOC arm 11.3 months, [Hazard ratio (HR)=1.37, 95% CI=1.07-1.76; p=0.013]. Median PFS was 7.4 and 6.0 months respectively (HR=1.22: 95% CI=0.98-1.52, p=0.068), and ORR of 42% compared to 46% (odds ratio=1.16, 95% CI=0.81-1.57, p=0.467) was observed (80). The panitumumab-arm patients also had higher incidence of grade 3-4 toxicities, diarrhea (17 vs. 11%), rash (11 vs. 1%), mucositis (5% vs. none), and hypomagnesaemia (5% vs. none) but reduced incidence of neutropenia (13 vs. 28%). Interestingly, a subgroup of patients with grade 1-3 rash (77%) showed an improved OS of 10.2 months vs. 4.3 months (p<0.001) with similar improvements in RR and PFS. As per the authors of the study, poorer outcomes in the panitumumab group was thought to be related to the lower dose of oxaliplatin and epirubicin in the mEOC+P group. KRAS, BRAF and PIK3CA mutations and PTEN expression were assessed in 175 pre-treatment biopsies and correlated with response to mEOC+P (81). Mutations in KRAS (5.7%), BRAF (0%), PIK3CA (2.5%) and loss of PTEN expression (15.0%) were noted, however, were found to be too low to have any prognostic or predictive effect.
Monoclonal antibodies in phase 2/3 trials for esophagogastric cancer.
Nimotuzumab. It is a humanized monoclonal antibody with weak ADCC activity, has a comparatively better adverse effect profile compared to cetuximab and panitumumab, with absence of skin, renal and gastrointestinal toxicities (82). A recent dose-escalation study of nimotuzumab in advanced esophageal squamous cell carcinoma in combination with cisplatin and 5-fluorouracil showed an excellent tolerability profile and good treatment effect with ORR of 42% (83). A randomized phase II trial of nimotuzumab in combination with cisplatin and s-1 in metastatic gastric cancer, showed comparable response rates and superior time-to-progression (5.5 vs. 3 months) (84). In a phase 2 trial with 63 advanced (locally advanced and metastatic) esophageal cancer patients, cisplatin, 5-FU and radiotherapy was given alone or in combination with six weekly infusions of nimotuzumab (85). An ORR of 47.8% in the nimotuzumab group in comparison to 15.4% in the control group was noted. EGFR-expressing tumors had an objective response rate of 60% and disease control rate of 80%. As first-line therapy in esophageal squamous cell cancers, 55 patients received cisplatin with paclitaxel and nimotuzumab (86). The ORR was 54.7% with median PFS of 10.5 months. In a single-center study, 62 metastatic gastric cancer patients were enrolled to receive s-1 (twice daily on days 1-14 every 3 weeks) and cisplatin on days 1 and 2 alone or in combination with nimotuzumab (days 1, 8, 15) as first-line therapy. The ORR achieved was 50% in the latter group (NCS) in comparison to 63% in the first group (CS) with PFS of 5 months and 3 months respectively (87). Incidence of adverse events was similar in both groups.
Pertuzumab. Pertuzumab is a new humanized monoclonal antibody that binds to domain II dimerization arm of HER-2 in contrast to trastuzumab which binds to domain IV. It inhibits the HER-2 dimerization with other EGFR family members and also exhibits antibody-dependent cell-mediated cytotoxicity (ADCC) related antitumor activity (88, 89). In NCI-N87, an Her-2-positive human gastric cancer xenograft model, combination of trastuzumab and pertuzumab achieved higher efficacy in causing tumor regression, compared to maximum effective dose with single agent alone (90). This combination inhibited EGFR-Her-2 heterodimerization and down-regulated phosphorylation of HER-1/Her-2 receptors as well as downstream effectors. Potentiation of the ADCC activity by the combination, and anti-angiogenetic activity was also observed. Clinical trials to investigate the activity of this combination in advanced esophagogastric cancers are being designed.
Tyrosine kinase inhibitors
Tyrosine kinase inhibitors repress the phosphorylation step that follows the EGFR receptor dimerization, inhibiting further downstream signaling proteins (91). Gefitinib and erlotinib, both EGFR tyrosine kinase inhibitors are approved for treatment of advanced NSCLC after failure of first line platinum-based chemotherapy (92). Erlotinib is also approved with gemcitabine as first-line treatment of pancreatic cancer (93). The various trials and clinical outcomes for TKIs in esophagogastric cancer are summarized in Table III .
Gefitinib. Gefitinib in esophagogastric cancers has been disappointing, but careful patient selection for future trials might improve its efficacy for these cancers. As first line therapy, gefitinib along with perioperative 5-FU/cisplatin and radiation was evaluated in 80 locally advanced esophageal cancer patients, including both adenocarcinoma and squamous cell carcinoma histology (94). The patients did not develop any significant toxicity compared to a cohort of patients similarly treated without gefitinib. In this single-arm phase II study, addition of gefitinib to chemotherapy resulted in OS of 40%, compared to 28% in the historical cohort suggesting a survival benefit.
Gefitinib has a modest activity in advanced esophageal cancer patients after failure of first line chemotherapy treatment. In a phase II study of 36 esophageal cancer patients treated with daily gefitinib, one patient had partial response and 10 patients had stable disease with OS of 5.5 months (95). Higher response rate was associated with female sex, squamous cell histology and higher EGFR expression.
Another phase II study of gefitinib in 58 recurrent and metastatic esophageal and gastro–esophageal cancers demonstrated partial response in 7% and stable disease in 17% patients with median survival of 5.5 month (96).Treatment with gefitinib has been well tolerated, but efficacy is limited. Further patient selection and combination with other chemotherapy agents in future trials may improve outcome in trials with gefitinib.
Erlotinib. Erlotinib was evaluated in a phase 2 study concurrently with chemotherapy and radiation for locally advanced esophageal carcinoma patients(97). Erlotinib at 150 mg daily dosing was given for the entire duration of chemo-radiotherapy, and chemotherapy (paclitaxel and cisplatin) was administered on Day 1 and Day 29 of the radiotherapy. Two-year OS, loco-regional control, and relapse-free survival were 70.1, 87.5 and 57.4% respectively with low incidence of grade 3 or greater toxicities. In previously-untreated metastatic and locally advanced esophageal and GEJ cancer, daily erlotinib in combination with a modified FOLFOX chemotherapy regimen every 14 days was evaluated with a primary objective of overall response rate and secondary objective including other clinical outcomes (98). Thirty three patients were included and 51% ORR with 5.5 months PFS and eleven months median OS was noted. Diarrhea nausea and vomiting were the commonest grade 3 toxicities. The Southwest Oncology Group (SWOG 0127) conducted a phase II trial of erlotinib in unresectable or metastatic adenocarcinoma of the gastroesophageal junction (GEJ) or stomach as a first line of treatment (99). ORR was 9%, all occurring in patients with GEJ. The median survival was 6.7 months in GEJ and 3.5 months in patients with stomach cancer. Molecular correlates (EGFR, phosphorylated AKT, TGF-alpha expression) were not predictive of response and no somatic mutation or amplification of EGFR was observed.
Tyrosine kinase inhibitors in esophagogastric cancer.
Lapatinib. Lapatinib is an oral, reversible dual tyrosine kinase inhibitor of EFGR and Her-2. Compared to erlotinib and gefitinib, it binds to inactive conformation of EGFR and has slower receptor dissociation rates. In comparison to trastuzumab, it produces anti-tumor effects through survivins (100), an apoptosis inhibitor protein. Other than causing the accumulation of inactive Her-2 receptors on the cell surface, it potentiates along with trastuzumab the ADCC activity (101). Currently, it has FDA approval for Her-2-positive breast cancer patients with prior progression on trastuzumab, a taxane and anthracycline. Lapatinib has been shown to induce growth inhibition in Her-2 amplified gastric cancer cell lines in vivo, SNU-216 and NCI-N87 (102). It also induces apoptosis in the NCI-N87 line which has a high Her-2 amplification ratio and in combination with 5-FU, cisplatin, oxaliplatin and paclitaxel showed a synergestic effect. In vivo and in vitro studies on esophageal and gastric cell lines, with trastuzumab in combination with lapatinib synergistically showed tumor regression in Her-2 amplified cell lines (103).
Lapatinib as a single, first-line agent was evaluated in a southwest oncology group (SWOG) trial in 47 patients with gastric adenocarcinoma. Partial response rate was seven percent with a median OS of 5 months (104). A random discontinuation study of lapatinib was conducted in patients with solid tumors, refractory to chemotherapy and with Her-2 amplification. The trial closed early secondary to slow accrual and poor response (105). Lapatinib is currently being evaluated as first line therapy along with capecitabine or oxaliplatin in LOGIC trial (106). Initial results reported have been disappointing, as the primary end-point was not reached with a hazard ratio of 0.91 (95% CI=0.73, 1.12, p=0.35) and median survival of 12.2 vs. 10.5 months, respectively.
Dacomitinib. Dacomitnib is a pan-HER tyrosine kinase inhibitor, currently being evaluated in advanced gastric cancer (107). Twenty seven patients with advanced Her-2-positive gastric cancer were treated with single-agent dacomitinib after failure of at least one prior chemotherapy regimen. The ORR was 7.4% and OS was 7.1 months. This modest clinical activity in heavily pre-treated patients is encouraging and merits further evaluation. Adverse effect profiles included rash, diarrhea and fatigue.
Targeting ERBb3 and ERBb4 Receptor
Her-2 which is the favored hetero-dimerization partner among the EGFR receptors, prefers to dimerize with Her-3 when signaling through the PI3K pathway. Her-3 can directly activate the PI3K-Akt pathway, mediating the proliferative effects of the Her-2/Her-3 dimer (108). Her-3 loss of function has been found to inhibit the growth of HER-2 overexpressing cells (109) and these cells were sensitive to PI3K inhibitors (110). HER-3 overexpression is an independent poor prognostic indicator in patients with gastric cancer who underwent gastrectomy (111). Anti-Her-3 monoclonal antibodies are currently under development. In comparison to other members of the EGFR family, EGFR-4 receptor's role in oncogenesis remains less well-defined and it is suggested that it might actually be involved in signaling towards cell growth inhibition (112). At present, there are no drugs in clinical development targeting the Her-4 receptor specifically and its expression did not correlate with prognosis in gastric cancer specimens (111).
Predictors of Response
Despite development of new molecular-level therapies for managing gastric and esophageal cancers, reliable markers of response are yet to be validated in large prospective studies. These would not only help advance our understanding of these complex interactions at the molecular level but also help select patient populations through clinical trials in a more meaningful way.
EGFR Expression
EGFR overexpression was initially studied as a possible marker for response to EGFR-directed therapy. In metastatic colorectal cancer, EGFR overexpression (detected by IHC staining) has not been predictive of response to monoclonal antibodies (113). Similar lack of correlation between EGFR expression and therapeutic benefit of EGFR inhibitors was seen in head and neck cancer and lung cancer studies (54, 114). EGFR status determined by IHC is not considered reliable as it does have multiple technical difficulties for reliable result reproduction. EGFR amplification (detected by FISH) might be a better alternative (115). Out of the four trials, comparing EGFR expression by IHC and cetuximab response in esophagogastric cancers, all but one showed no correlation (116-118). Han et al. reported, that EGFR expression in advanced gastric cancers can be an independent predictor of time-to-progression and also low levels of ligands (EGF and TGF-alpha) at baseline correlated with response to treatment (119). The role of IHC to assist response prediction to anti-EGF monoclonal antibodies remains unclear.
Oncogene Mutations
Oncogenic mutations in the KRAS, BRAF or PIK3CA genes lead to production of constitutively active signaling proteins downstream of EGFR pathway, which are not affected by inhibition of EGFR through monoclonal antibody or tyrosine kinase inhibitors (120). Mutation status of KRAS gene, which is about 30-40 % in colorectal cancers, helps to identify a distinct subtype of metastatic colorectal cancer which does not respond to cetuximab or panitumumab (120, 121). It is being widely used in clinical practice as a validated marker in treatment selection. Among tumors carrying wild-type KRAS, mutations of BRAF or PIK3CA or loss of PTEN expression may be associated with resistance to the anti-EGFR monoclonal antibody treatment (122). Incidence of KRAS and BRAF mutations in gastric cancer studies has been low (2-20%) (123-126), and sporadic simultaneous presence has been noted with aggressive behavior (123). Data regarding the role of these mutations and relation to anti-EGFR therapy have been obtained in a number of phase-2 studies with cetuximab. In the FUFOX study (59), KRAS mutation in exon 1 was found only in 1 of 32 samples (3%) and no correlation to EGFR therapy could be identified. In a retrospective study KRAS mutation was found in 13.3 % of metastatic gastric cancer patients treated with cetuximab and irinotecan, and response to cetuximab or OS did not correlate with KRAS mutation status (63). Forty-four patients with advanced gastric cancer, were evaluated from 2 consecutive phase 2 studies; FOLCETUX and DOCETUX. KRAS and BRAF mutations were detected in 11.4 and 2.3 % patients respectively (127). The KRAS and BRAF mutations were mutually-exclusive and the latter was the classic V600E substitution. There was no correlation to oncogene mutations and treatment response and OS. Current evidence for KRAS and BRAF mutations as potential predictors of outcome for gastric cancer is lacking.
Other Predictors of Response
Cetuximab-induced rash can be a potential predictor of reponse. In a phase-2 trial of esophageal squamous cell cancer, cetuximab induced rash (grade >2) had a better CR rate and PFS than those with no or grade 1 rash (p<0.05) (75). Potential biomarkers to predict response to Her-2-targeted treatment include loss of PTEN, PI3KCA gene mutation and increase in signaling through EGFR and IGF receptor (128). In HER-2-positive breast cancer, loss of PTEN (examined by IHC) correlated with poor response rates to trastuzumab and chemotherapy treatment (129). Similar correlation in esophagogastric cancers needs further evaluation. Increased levels of EGFR ligands, epiregulin and amphiregulin promote tumor growth via an autocrine loop. In colorectal cancer patients with wild-type KRAS, expression of epiregulin and amphiregulin mRNA in primary tumors correlates with response to cetuximab. In esophagogastric cancers, expression of EGFR ligands has not been evaluated. PET scanning was evaluated in the FOLCETUX study, in advanced gastric cancer patients treated with cetuximab in combination with irinotecan, folinic acid with 5-fluorouracil (130). Metabolic responders on the scan had significantly longer time-to-progression compared to non-responders. Further evaluation in bigger studies might prove this to be a useful resource in predicting response to cetuximab-based treatment.
Resistance to EGFR Therapy
After an initial response to EGFR-targeted therapy, progression or recurrence develops secondary to intrinsic or acquired resistance. Our understanding of the resistance to these monoclonal antibodies and tyrosine kinase inhibitors comes from breast and colorectal cancer data where they have been used extensively. The etiology for EGFR receptor antibody resistance appears to vary. Increased angiogenesis in response to treatment with cetuximab has been proposed as a mechanism of resistance to cetuximab (131). Higher levels of VEGFR-1, VEGF and placental growth factor were found in EGFR-resistant cell lines and VEGFR inhibition restored sensitivity to anti-EGFR drugs (132). An in vitro study on five gastric cancer cell lines showed correlation between positivity for MET amplification and KRAS mutations, and cetuximab resistance (133). Dysregulation of EGFR internalization and degradation (134) along with subcellular compartmentalization to nucleus could contribute to resistance (135). The nuclear EGFR could constitutively activate gene promoters including cyclin-D1, iNOS, B-myb leading to EGFR antibody resistance. Genetic mutations in signaling proteins downstream to EGFR pathway such as RAS, RAF, AKT and PI3K could lead to constitutive activation of growth promoting signaling and resistance to EGFR antibody. Mutant KRAS and BRAF are well-established biomarkers which predict resistance to cetuximab in colorectal cancers, however such mutations are rare in esophagogastric cancers and have not correlated with response to anti-EGFR treatment in the published studies so far (136).
Multiple mechanisms have been postulated for resistance to Her-2 receptor inhibition and targeting of these mechanisms along with Her-2 receptor inhibition would help to improve Her-2-targeted therapy. IGF-1 receptor overexpression and cross-talk with Her-2 receptor, leading to it's activation was seen in only trastuzumab-resistant breast cancer cell lines and not in the parental trastuzumab-sensitive cells (137). Genetic alterations in the PI3K/AKT pathway such as gain of function mutation in PIK3CA, loss of PTEN and AKT mutations have been correlated with resistance to trastuzumab (138-140). Another mechanism of resistance described in breast cancer but not known in esophagogastric cancers in expression of p95HER2, a truncated Her-2 receptor at the amino terminal, which cannot bind to trastuzumab but has kinase activity. Metastatic breast cancer patients expressing p95HER2 showed limited clinical response to trastuzumab compared to patients with tumors expressing full-length Her-2 receptor (154).
Conclusion and Future Directions
Treatment of esophagogastric cancers with traditional chemotherapeutic agents had reached a plateau and recent demonstration of increased survival with the addition of trastuzumab to the chemotherapy has generated much excitement and there has been a surge of clinical trials involving EGFR-targeted agents. Although, EGFR-targeted agents have produced only modest success in esophagogastric cancers compared to breast, colorectal or head and neck cancers, clinical studies evaluating newer more potent drugs in combination with different treatment modalities are in the anvil. Success of these trials depends on identifying the predictive biomarkers and to select the right patient population that is likely to benefit. On-going clinical trials will establish the role of EGFR inhibitors in neoadjuvant, adjuvant therapy and in combination with radiation treatment. In the metastatic setting, the role of EGFR-targeted agents as maintenance therapy after achieving response with first-line therapy and their use beyond first-line of treatment will need to be further explored.
Further studies are needed to better-understand the role of EGFR receptor signaling in esophagogastric cancers in relation to other signal transduction pathways. There is plenty of cross-talk between multiple signaling pathways and targeting one of the molecules alone, allows for cancer cells to find an escape route utilizing alternate pathways. Increased understanding of the involved pathways will help to develop rational strategies to combine agents targeted against different pathways to achieve maximal therapeutic benefit. Identification of mechanism for acquired resistance will lead to design of clinical trials with newer approaches to delay or reverse the same.
Footnotes
-
This article is freely accessible online.
- Received August 23, 2013.
- Revision received September 17, 2013.
- Accepted September 18, 2013.
- Copyright© 2013 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
- Article
- Abstract
- EGFR Receptor and Signaling
- Role of EGFR Receptor Family in Esophageal and Gastric Cancer
- Targeting the EpidermalGrowth Factor Receptor (EGFR)
- Targeting ERBb3 and ERBb4 Receptor
- Predictors of Response
- EGFR Expression
- Oncogene Mutations
- Other Predictors of Response
- Resistance to EGFR Therapy
- Conclusion and Future Directions
- Footnotes
- References
- Figures & Data
- Info & Metrics