Targeted therapy for esophagogastric cancers: a review.

The incidence of esophagogastric cancers is increasing rapidly in the Western population. Despite better understanding of the biology and intense research in the treatment of these cancers, the long-term survival remains poor both in the locally advanced and metastatic settings. The addition of combined modality strategies has resulted in modest improvement in 5-year survival rates. A number of biologic agents targeting epidermal-derived growth factor receptor, vascular endothelial derived growth factor and its receptor, and mammalian target of rapamycin (mTOR) are being currently evaluated in Phase II and III clinical trials. Some of these, like trastuzumab, cetuximab, and bevacizumab, have shown promising results. This review provides a brief overview of the recent developments in biologic agents for the treatment of esophagogastric cancers.

Introduction

Esophagogastric cancers (EGCs) include tumors involving the esophagus, gastroesophageal junction (GEJ), and stomach. While the tumors involving the esophagus can be of either adenocarcinoma (AC) or squamous cell carcinoma (SCC) histology, tumors involving the GEJ and stomach are almost exclusively of AC type. The EGCs constitute a significant problem worldwide. The incidence of EGC-AC is increasing rapidly in the Western population.1 There are demographic variations in the incidence of these tumors, and there has been a major change in the epidemiological trend towards AC histological subtype in the Western world over the last two decades.2–4 Globally, SCC is still the most common subtype of esophageal cancer, but AC is now the most prevalent in the West.

Despite greater understanding of the biology and intense research in the treatment of EGC, the prognosis and long-term survival remains poor for both locally advanced and metastatic disease, with a case–fatality ratio of 84% (esophageal) and 75% ( gastric), respectively.5 Current treatment options for locally advanced disease include surgery alone, combined modality treatment (perioperative chemoradiotherapy, or chemotherapy), or definitive chemoradiation therapy. The 5-year survival rates for locally advanced disease with surgery alone are only 20%–25%.6,7 Poor outcomes with surgery alone have led to the evaluation of combined modality approaches to improve the survival. Such therapy has now become the standard of care for locally advanced disease. However, the addition of combined modality strategies results in modest improvement in 5-year survival rates to only 30%–35%.7–10 Palliative chemotherapy is the current treatment approach practiced by medical oncologists worldwide for metastatic disease. The objective response rate (ORR) with palliative chemotherapy for metastatic disease is only 20%–40%, with a median overall survival (OS) of 8–10 months.11 Various combination chemotherapy regimens have been used with addition of a third drug but have achieved a modest benefit only at the cost of increased toxicity in this cohort of patients with increased comorbidites.12,13

The lack of any further significant benefit in OS and ORR with cytotoxic chemotherapy seems unlikely though not impossible and has lead investigators to believe that a plateau has been reached. Therefore, in the era of intense translational research and personalized therapy, the focus of current clinical trials has shifted to integrating molecular targeted therapies into current treatment strategies and to exploiting the molecular abnormalities that are involved in the etiology and pathogenesis of these tumors. The molecular targets currently being evaluated in various Phase II and III clinical trials include the epidermal-derived growth factor receptor (EGFR) with subtypes ERBB-1, ERBB-2 (Her2/neu), ERBB-3 and ERBB-4, vascular endothelial derived growth factor (VEGF) and its receptor (VEGFR), and mammalian target of rapamycin (mTOR). The focus of this review is to evaluate the current evidence for molecular-targeted agents in the management of EGC.

Potential targets and biologic agents

HER2/neu tyrosine kinase receptor

Her2/neu is a member of the ERBB tyrosine kinase (TK) receptor family. Peptide ligands bind to its extracellular domain, leading to receptor homo- and hetero-dimerization followed by auto-phosphorylation of the kinase. Her2/neu has been variably expressed in EGC (AC) (mean 23%; range 0%–43%).14,15 Approximately 25% of advanced gastric cancers (GCs) (similar to breast cancer) overexpress Her2/neu.16–18 This variability is presumably related to the differences in Her2/neu testing based on immunohistochemistry (IHC) or fluorescent in-situ hybridization (FISH) and different stages of disease. Her2 positivity is more common in the intestinal type and EGC rather than the diffuse type and pure GC.19–21 Her2 amplification on FISH is an independent prognostic factor which correlates with the depth of invasion, nodal/distant metastasis, and poor survival.21–23 There is high concordance observed between HER2 results obtained by both IHC and FISH on primary tumors and corresponding metastases.24

Anti-Her2/neu monoclonal antibody (trastuzumab)

Trastuzumab (Herceptin®, Genentech, South San Francisco, CA) is a humanized IgG1 monoclonal antibody targeting the Her2/neu receptor. It acts through several mechanisms, including inhibition of receptor dimerization, increasing receptor endocytosis and degradation, and inducing antibody dependent cytotoxicity.25 Trastuzumab is approved by the United States Food and Drug Administration (FDA) for the treatment of breast cancer, both in the adjuvant and metastatic settings.26–29

Safran et al30 did a Phase I/II study (n = 19) of trastuzumab in combination with cisplatin and paclitaxel in locally advanced, HER2 overexpressing, esophageal adenocarcinoma. The ORR was 43%, with an OS of 24 months. The 2-year survival was 50%. There was no significant cardiotoxicity related to trastuzumab.

Recently, Bang et al31 published the results of the Phase III ToGA (Trastuzumab for GC) trial of capecitabine/5-fluorouracil (5-FU) and cispaltin with or without trastuzumab. This was an open-label, international, Phase III, randomized controlled trial and included patients with Her2 overexpressing (by IHC or FISH) GC or GEJ tumors. Participants were randomly allocated in a 1:1 ratio to receive a chemotherapy regimen (capecitabine/5-FU plus cisplatin given every 3 weeks for six cycles) or the same chemotherapy in combination with intravenous trastuzumab. Out of a total of 3807 patients whose tumors were assessed for Her2 positivity, 22.1% were Her2 positive and 594 were randomized. Median follow-up was 18.6 months in the trastuzumab plus chemotherapy group and 17.1 months in the chemotherapy alone group. There was a statistically significant increase in ORR in the trastuzumab-containing arm compared with the chemotherapy alone arm (47.3% versus 34.5%; P = 0.0017). The median progression-free survival (PFS) (6.7 versus 5.5 months, hazard ratio [HR] 0.71, 95% confidence interval [CI] 0.59–0.85; P = 0.0002) and OS (13.8 months versus 11.1 months P = 0.0046, HR = 0.74, 95% CI 0.60–0.91) were also in favor of the trastuzumab-containing arm compared with the chemotherapy alone arm.

In an exploratory post-hoc analysis, the OS was longer in patients with high expression of Her2 compared with those with low Her2 expression.31 This analysis suggested that in patients with the highest levels of HER2 protein expression (HER2 2+ and FISH positive, HER2 3+ and FISH positive), trastuzumab conferred an even greater survival benefit than that observed for the intention-to-treat population (16.0 versus 11.8 months; HR 0.65). This treatment combination is the first to result in a median survival beyond 1 year and will change the standard of care for HER2-positive GC. Based on these data, trastuzumab has been approved for the treatment of HER2-positive advanced GC in combination with chemotherapy in several countries, including in Australia where it is approved (but not yet subsidized) for first-line HER2-positive advanced GC/GEJ cancers in combination with cisplatin, and either 5-FU or capecitabine. The data suggests that trastuzumab is more effective in the subgroup of patients with IHC 3+ tumors (HR 0.66, 95% CI 0.50–0.87) compared with patients with IHC 2+ tumors (HR 0.78, 95% CI 0.55–1.10). There was also no unexpected toxicity in the trastuzumab arm including symptomatic heart failure; however, there was an increased incidence of asymptomatic decrease in ejection fraction (4.6% versus 1.1%). The quality of life was not compromised in the trastuzumab arm in a recent analysis.32 Future research should focus on evaluating the role of trastuzumab beyond progression and locally advanced (neo-adjuvant/adjuvant) settings. The pattern of HER2 amplification/overexpression in GC tissue (heterogeneous and frequently focal), and the scoring system used to assess it, differ from that in breast cancer, and consequently, HER2 testing protocols used for breast cancer specimens require modification to be used for GC specimens.33,34

Anti-Her2/neu tyrosine kinase inhibitor (lapatinib)

Lapatinib (Tykerb®, GlaxoSmithKline, London, UK) is an orally active, dual TK inhibitor (TKI) with activity against both EGFR (ERBB1) and Her2 (ERBB2). Two Phase II trials have evaluated the role of lapatinib in EGC, but the results have been disappointing. The Southwest Oncology Group (SWOG) performed a Phase II study evaluating the role of lapatinib as monotherapy in the first-line setting in advanced GC patients (n = 47).35 The partial response rate was only 7%, with a median time to treatment failure and OS of 2 and 5 months respectively. In the second Phase II study, 25 patients with pretreated Her2-positive EGC (through IHC or FISH) were evaluated.36 The ORR was 0% in 21 evaluable patients, with two patients having stable disease for 5 and 9 months.

Despite the poor ORR from Phase II studies, two Phase III studies are evaluating the role of lapatinib in conjunction with chemotherapy: LOGIC Trial (lapatinib in combination with capecitabine and oxaliplatin as first line) and TYTAN trial (lapatinib in combination with weekly paclitaxel as second line). The clinical trials of anti-Her2/neu agents in EGC are summarized in Table 1.

Table 1

Clinical trials of anti-Her2/neu agents in EGC

Author	Phase	Histology	Stage	Treatment	No of patients	ORR or pCR rate	TTP/PFS (months)	OS (months)
Bang et al31	III	AC	LA (20)Met (584)	5-FU/capecitabine + cisplatinversus5-FU/capecitabine + cisplatin +trastuzumab	290versus294	34.5%versus47.3%	5.5versus6.7	11.1versus13.8
Safran et al30	I/II	AC	LA	Trastuzumab + cisplatin/paclitaxel/RT	19	43%	NS	24
Iqbal et al35	II	AC	Met	Lapatinib	47	7%	2	5
Hecht et al36	II	AC	Met	Lapatinib	25	0%	NS	NS

Abbreviations: 5-FU, 5-fluorouracil; EGC, esophagogastric cancer; NS, not stated; ORR, overall response rate; TTP, time to progression; PFS, progression-free survival; OS, overall survival; pCR, pathological complete response; AC, adenocarcinoma; LA, locally advanced; Met, metastatic; RT, radiotherapy.

EGFR

EGFR (or ERBB1) is a member of the ERBB TK receptors. Ligand binding to the extracellular domain of the TK receptor leads to its activation and subsequent homo-dimerization, followed by auto-phosphorylation of the intracellular signaling cascade including RAS/RAF/MAP kinase pathway. These pathways play an important role in angiogenesis, cell survival and proliferation, apoptosis, and metastasis.

Abnormal expression and activating mutations of EGFR have been reported in EGC, eg, EGFR overexpression by IHC/SISH occurs in 50%–63% of patients with GC.37 Overexpression of EGFR is a poor prognostic indicator in EGC with poorly differentiated histology, depth of invasion, and shorter survival.38 EGFR overexpression is more commonly associated with SCC histology than AC.

Another oncogene downstream of EGFR is KRAS. The predictive value of KRAS status to anti-EGFR therapy has been well validated in metastatic colorectal cancer.39,40 At present, none of the trials evaluating the EGFR-targeted therapies are restricted to patients with wild-type KRAS tumors, as the frequency of KRAS mutations in GC is expected to be low (around 5%),41 and there are currently no data to suggest that KRAS gene mutation is predictive of lack of response to EGFR-targeted monoclonal antibody therapy in this tumor type.

Anti-EGFR monoclonal antibodies

Cetuximab

Cetuximab (Erbitux®, ImClone Systems, New York, NY) is a partially humanized murine IgG1 monoclonal antibody that blocks the binding of protein ligands to the EGFR and its activation.42 Various other proposed mechanisms of action of cetuximab include receptor internalization through endocytosis and immune-mediated mechanisms including antibody-dependent cytotoxicity, complement-dependent cytotoxicity, and complement-dependent cell-mediated cytotoxicity.43,44 It is currently FDA approved in the management of metastatic colorectal cancer as monotherapy or combination therapy and in conjunction with radiotherapy (RT) in the management of head and neck SCC.45,46

Cetuximab is the most extensively studied anti-EGFR monoclonal antibody in the management of EGC, both in the locally advanced and metastatic settings as first- or second-line therapy. Cetuximab has been evaluated in the locally advanced setting in six clinical trials. It has been used in combination with various cytotoxic regimens and RT preoperatively (4/6),47–50 monotherapy with RT preoperatively (1/6),51 and in combination with cytotoxic therapy and RT both preoperatively and postoperatively (1/6).52 The results from these trials should be interpreted with caution as the findings are preliminary and time to progression (TTP) and OS data are awaited. The ORR/pathological complete response (pCR) rates range from 13%–40% across these trials. The results have been summarized in Table 2. With the exception of the trial by Ma et al52 (100% grade 3/4 serious adverse event rate and lower pCR), the toxicity profile in these trials have been consistent with other combined chemoradiotherapy trials.

Table 2

Clinical trials of cetuximab in EGC

Author	Phase	Histology	Stage	Treatment	No of patients	ORR or pCR rate	TTP/PFS (months)	OS (months)
Enzinger et al47	II	AC	LA	Cetuximab + cisplatin/irinotecan/RT	17	13%	NS	NS
Safran et al48	II	AC/SCC	LA	Cetuximab + carbolatin/paclitaxel/RT	60	27%	NS	NS
Ruhstaller et al50	II	AC/SCC	LA Cetuximab + cisplatin/docetaxel/RT	28	32%	NS	NS
De Vita et al49	II	AC/SCC	LA	Cetuximab + FOLFOX/RT	27	40%	NS	NS
Agrawala et al51	II	AC/SCC	LA	Cetuximab + RT	40	36%	NS	NS
Ma et al52	II	AC	LA	Cetuximab + cisplatin/irinotecan + surgery → Cetuximab + 5-FU/LV/RT	20	0%	NS	NS
Lordick et al59	II	AC	Met	Cetuximab + FUFOX	52	65 (of 46)	7.6	9.5
Pinto et al57	II	AC	Met	Cetuximab + cisplatin/docetaxel	48	41 (of 42)	NS	NS
Zhang et al56	II	AC	Met	Cetuximab + cisplatin/capecitabine	49	48 (of 47)	5.2	NS
Han et al61	II	AC	Met	Cetuximab + FOLFOX	40	50	5.5	9.9
Yeh et al55	II	AC	Met	Cetuximab + high dose 5-FU/LV/cisplatin	35	69	11	14
Woell et al58	II	AC	Met	Cetuximab + oxaliplatin/irinotecan	51	63% (of 35)	6.2	9.5
Lorenzen et al54	II	SCC	Met	Cetuximab + 5-FU/cisplatinversus5-FU/cisplatin	32versus30	19%versus13%	5.7versus3.6	9.5versus5.5
Pinto et al53	II	AC	Met/UR	Cetuximab + FOLFIRI (FOLCETUX)	38	44% (of 34)	8	16
Kanzler et al60	II	AC	Met	Cetuximab + FUFIRI	49	42% (of 48)	8.5	16.6

Abbreviations: EGC, esophagogastric cancer; 5-FU, 5-fluorouracil; FOLFIRI, two weekly bolus 5-FU/leucovorin, irinotecan, infusional 5-FU; FUFIRI, weekly irinotecan/leucovorin/infusional; 5-FU FUFOX, weekly oxaliplatin/leucovorin/infusional 5-FU; LV, leucovorin; RT, radiation therapy; NS, not stated; ORR, overall response rate; TTP, time to progression; PFS, progression-free survival; OS, overall survival; pCR, pathological complete response; AC, adenocarcinoma; SCC, squamous cell carcinoma; LA, locally advanced; Met, metastatic; UR, unresectable.

Cetuximab has been evaluated in the first-line metastatic setting in combination with other cytotoxic therapies in numerous trials. The chemotherapy backbone used in these trials includes FOLFIRI (two weekly bolus 5-FU/leucovorin/irinotecan and infusional 5-FU),53 5-FU/cisplatin,54 continuous infusion high-dose 5-FU/leucovorin/cisplatin,55 capecitabine/cisplatin,56 cisplatin/docetaxel,57 oxaliplatin/irinotecan,58 FUFOX (weekly oxaliplatin/leucovorin/infusional 5-FU),59 FUFIRI regimen (weekly irinotecan/infusional 5-FU/leucovorin),60 and mFOLFOX.61 The results of these trials are summarized in Table 3. These results are promising, with ORR of 40%–69%, TTP 5.0–8.5 months, and OS of 9.5–17.0 months across these eight trials. The toxicity in these trials has been consistent with known side-effect profiles of the chemotherapy backbone used, along with additive toxicity from cetuximab (grade 3/4: diarrhea 4%–33%, skin toxicity 6%–24%, infusion reactions/anaphylaxis < 5%).

Table 3

Clinical trials of anti-VEGF agents

Author	Phase	Histology	Stage	Treatment	No of patients	ORR or pCR rate	TTP/PFS (months)	OS (months)
Ohtsu et al88	II	AC	Met/UR	5-FU/capecitabine + cisplatin + placeboversus5-FU/capecitabine + cisplatin + bevacizumab	387versus387	29.5%versus38%	5.3versus6.7	10.1versus12.1
Shah et al84	I/II	AC	Met	Bevacizumab + cisplatin/irinotecan	47	65%	8.3	12.3
Kelsen et al86	II	AC	Met	Bevacizumab + docetaxel/cisplatin/5-FU	44	67% (of 37)	12	16.2
Enzinger et al87	II	AC/SCC	Met	Bevacizumab + docetaxel/cisplatin/irinotecan	32	63% (of 30)	NS	NS
Bang et al91	II	AC	Met	Sunitinib	42	2.6%	2.3	6.8
Sun et al95	II	AC	Met/UR	Sorafenib + docetaxel/cisplatin	44	38.5%	5.8	14.9

Abbreviations: 5-FU, 5-fluorouracil; NS, no stated; ORR, overall response rate; TTP, time to progression; PFS, progression-free survival; OS, overall survival; pCR, pathological complete response; AC, adenocarcinoma; SCC, squamous cell carcinoma; LA, locally advanced; Met, metastatic; UR, unresectable; VEGF, vascular endothelial derived growth factor.

Tumor overexpression of EGFR has been associated with a poorer prognosis in GC.62 Overexpression of EGFR by IHC/SISH was not a prerequisite for most of these trials, except the trial by Pinto et al.53 Therefore, the impact of EGFR positivity on response remains unclear. Even in the study by Pinto et al,53 evaluating the combination of FOLFIRI and cetuximab, there was no correlation between the degree of EGFR positivity and treatment response. Similar results were also noted by Lordick et al59 when they evaluated a combination of cetuximab with 5-FU/leucovorin/oxaliplatin (FUFOX). In contrast, Han et al61 in their trial of cetuximab in combination with modified 5-FU/leucovorin/oxaliplatin (mFOLFOX-6), noted that patients with EGFR-positive tumors and low EGF/TGF alpha levels (n = 11) had an ORR of 100% compared with only 37% in the remaining patients (n = 27) with EGFR-negative tumors (P ≤ 0.001).

The studies evaluating combination of cetuximab with other chemotherapeutic agents in previously treated EGC have been disappointing, with the best ORR of only 11% in the SWOG study.63–65 These studies are summarized in Table 3. Currently, cetuximab is being evaluated in an open-label Phase III study in combination with capecitabine and cisplatin versus capecitabine/cisplatin alone (EXPAND-NCT00678535; Phase III; CX +/− cetuximab) as a first-line therapy in the management of advanced GC/GEJ AC.

Panitumumab

Panitumumab (Vectibix®, Amgen, Thousand Oaks, CA) is a fully humanized IgG2 anti-EGFR monoclonal antibody. It is currently FDA approved for the management of metastatic colorectal cancer.66 It has not been evaluated as extensively as cetuximab in the management of EGC; however, a Phase III study is being conducted in the UK to determine whether adding panitumumab to epirubicin, oxalipaltin, and capecitabine (EOX) prolongs OS (REAL3 – NCT00824785; Phase III; EOX +/− panitumumab). Another study currently open to recruitment (NEOPECX) is an open-label randomized controlled Phase II trial of panitumumabin in combination with epirubicin, cisplatin, and capecitabine (ECX) versus chemotherapy alone in patients with locally advanced GC/GEJ tumors.

Matuzumab

Matuzumab (EMD 72000, Merck) is a humanized monoclonal antibody targeting the EGFR. It has been evaluated in a Phase I study in combination with epirubicin, cisplatin, and capecitabine (ECX) as a first-line therapy for patients with EGFR-positive EGC.67 The response to different doses of matuzumab was 57% (four partial responses and two stable disease) with 400 mg and 43% (three partial responses and two stable disease) with 800 mg. The major dose limiting toxicity was grade 3 fatigue. A Phase II study looking at a combination of matuzumab and ECX has recently completed recruitment (MATRIX EG).

Anti- EGFR tyrosine kinase inhibitors

Erlotinib

Erlotinib (Tarceva®, Genentech) is an oral small molecule TKI. It inhibits the binding of adenosine triphosphate to the TK domain of the EGFR, leading to cessation of downstream autophosphorylation and signal transduction. It is currently FDA approved for the management of advanced non-small cell lung cancer and pancreatic cancer.68,69

A Phase II study conducted by the SWOG evaluated erlotinib in the first-line setting in the management of GC (n = 26) and GEJ tumors (n = 44).70 There were no responses in the GC patients, and the ORR was only 9% (one complete response and two partial responses) in the GEJ tumors. The reason for the apparent differential sensitivity of GEJ and GC to EGFR blockade using erlotinib is unclear.

The TTF and OS were 2 months and 6.7 months respectively in the GEJ tumor group. No EGFR gene mutation was detected in the 54 samples analyzed in this trial.

Gefitinib

Gefitinib (Iressa®, AstraZeneca Pharmaceuticals, London, UK) is another oral small molecule TKI. It is currently approved in the management of advanced non-small cell lung cancer.71

Rodriguez et al84 assessed the efficacy of gefitinib in combination with cisplatin/5-FU and RT (n = 80) to chemo-RT alone (n = 93) in a Phase II trial of patients with EGC. There was a trend towards improved 3-year OS rates in favor of gefitinib arm (40% versus 28%; P = 0.07). Toxicity profiles were similar in both the arms.

In another Phase II study, gefitinib was evaluated in the second-line setting in 28 patients.72 The partial response rate was 3% and stable disease in 28%. The TTP and OS were 2.0 and 5.5 months respectively. There was a nonsignificant trend towards improved outcome in patients with high EGFR expression compared with low EGFR expression (TTP 5.1 months versus 1.8 months and OS 7.8 months versus 2.8 months).73

In all the trials of oral TKIs as monotherapy or combination therapy, there has been no significant additional toxicity associated with the use of TKIs. However, they have limited activity as monotherapy in both the first- and second-line setting in the management of EGC. From the available evidence, it seems that patients with squamous histology, high EGFR expression, and esophageal/GEJ tumors are more likely to have any meaningful response to TKIs, but these tumors are rarely of squamous type.

VEGFR

Our increasing understanding of the pathogenesis of tumor biology in most solid tumors has led us to believe that there is a strong link between tumor growth, distant metastasis, and angiogenesis.74 VEGF is the most potent and specific pro-angiogenic factor regulating endothelial cell mitogenesis and migration, induction of proteinases, increased vascular permeability, and maintaining survival of newly formed blood vessels.75 It exerts its angiogenic effect by binding to various transmembrane receptors, more specifically VEGFRs type 1 (VEGFR-1) and 2 (VEGFR-2).76

VEGF is overexpressed in 30%–60% of EGCs, and elevated levels of VEGF in the serum and tumor are associated with poor prognosis in EGCs.77–81 There is also evidence to support in esophageal AC that higher VEGF expression correlates with transition from Barrett’s esophagus to high grade dysplasia and from micro-invasive disease to locally advanced cancer.82,83

Anti-VEGF monoclonal antibodies

Bevacizumab

Bevacizumab (Avastin®, Genentech) is a humanized IgG1 monoclonal antibody-targeting VEGF. It is currently FDA approved for the management of advanced colorectal cancer, non-small cell lung cancer, glioblastoma multiforme, and metastatic renal cell carcinoma.

Multiple Phase II studies have looked at the role of bevacizumab in combination with chemotherapy as the initial treatment for patients with advanced GC/GEJ tumors. The early results from these studies have been encouraging. Shah et al85 from Memorial Sloan-Kettering Cancer Centre (MSKCC) looked at a combination of bevacizumab with cisplatin and irinotecan in the first-line management of advanced GC/GEJ AC (n = 47). The results were encouraging, with an ORR 65%, TTP 8.3 months (compared with TTP in historical controls of 5 months) and OS of 12.3 months. Toxicities related to the use of bevacizumab were: gastric perforation/near perforation 6%, myocardial infarction 2%, grade 3/4 thromboembolic events 25.5%.86 The same group from MSKCC also evaluated bevacizumab in combination with a modified regimen of docetaxel/cisplatin/5-FU (mDCF) in chemo-naïve patients (n = 44) and reported ORR of 67%, PFS of 12 months, and OS of 16.2 months in 39 patients with measurable disease.87 The toxicity profile of the mDCF was better than the original DCF,13 with febrile neutropenia rates of 4% versus 29% respectively. The side effects attributed to bevacizumab included: grade 3/4 thromboembolism 31% and gastric perforation/bleeding 2%. The studies evaluating bevacizumab in the first-line setting in combination with chemotherapy are summarized in Table 3.

Bevacizumab has also been tested in the second-line setting in a small study by Enzinger et al88 in combination with docetaxel. The ORR was 24% (17 patients evaluated). Bevacizumab-attributed grade 3/4 serious adverse events (SAEs) included: gastrointestinal bleeding (12%) and arterial thrombosis (8%).

Due to the encouraging results with bevacizumab in conjunction with chemotherapy, both in the first- and secondline settings, a confirmatory Phase III trial (AVAGAST) was conducted. 89 The AVAGAST trial was a multinational, randomized, placebo-controlled trial designed to evaluate the efficacy of bevacizumab in combination with capecitabine/cisplatin in the first-line treatment of advanced GC/GEJ tumors (n = 774). The OS (primary end point) was 12.1 months with bevacizumab plus capecitabine/5-FU-cisplatin and 10.1 months with placebo plus capecitabine/5-FU-cisplatin (HR 0.87; 95% CI, 0.73–1.03; P = 0.1002). Although the primary endpoint (OS) was not met, both ORR (46.0% versus 37.4%; P = 0.0315) and median PFS (6.7 versus 5.3 months; HR 0.80; 95% CI, 0.68–0.93; P = 0.0037) were significantly improved with bevacizumab versus placebo. The incidence of venous and arterial thrombosis did not differ between the two groups. Although the rates of hypertension and bleeding were slightly higher in the bevacizumab arm, most of the bleeding was grade 1 nasal bleeding.

There were significant regional differences in the outcome in an unplanned subset analysis. The Asian population had the longest OS (12.1 months), but the smallest impact from addition of bevacizumab (HR, 0.97; 95% CI, 0.75–1.25) compared with the Pan-American population (which contributed only one-fifth of patients to the whole group) who had the shortest OS (6.8 months) but the greatest impact from addition of bevacizumab (median, 11.5 versus 6.8 months for placebo-chemotherapy; HR, 0.63; 95% CI, 0.43–0.94). The potential explanations forwarded for these differences were that the Asian population had low incidence of junctional tumors, low frequency of liver metastases, and a higher proportion of them received second-line chemotherapy compared with the American group (66% versus 21%).

In the UK, another Phase III study [Medical Research Council Adjuvant Gastric Infusional Chemotherapy trial (MAGIC)-B] is currently recruiting patients to determine the efficacy of adding bevacizumab to chemotherapy (ECX) in the peri-operative setting.

Ramucirumab

Ramucirumab (IMC-1121B, ImClone Systems Corporation) is a fully humanized, IgG1 monoclonal antibody targeting VEGFR-2. It is currently being tested in a randomized, double-blind placebo-controlled trial in patients with metastatic GC/GEJ AC who have failed first-line chemotherapy with a platinum agent or fluoropyrimidine.

Anti-VEGF TKIs

Three anti-VEGF TKIs have been evaluated in EGC: sunitinib, sorafenib, telatinib.

Sunitinib

Sunitinib (Sutent®, Pfizer, New York, NY) is an oral multi-targeted TKI with activity against VEGFR, platelet derived growth factor (PDGFR), RET, c-kit, and Flt-3. It is currently FDA approved for the treatment of metastatic renal cell carcinoma and gastrointestinal stromal tumors.90,91

Sunitinib has been tested in two Phase II trials. Bang et al92 evaluated sunitinib as a second-line therapy in advanced GC/GEJ AC. Two patients had a partial response, and 25 had stable disease with PFS and OS of 2.3 and 6.8 months respectively. Moehler et al93 also examined sunitinib in previously treated GC and reported disease control in 5/14 patients. The toxicity with sunitinib was manageable in both these studies, with the common SAEs being thrombocytopenia, anorexia, nausea, and fatigue.

Sorafenib

Sorafenib (Nexavar®, Bayer, Leverkusen, Germany) is another oral multikinase inhibitor, with activity against both intracellular RAF-kinases (CRAF, BRAF) and cell surface kinase receptors (VEGFR-1 to 3, PDGFR-beta, RET, Flt-3, and c-kit). It is currently FDA approved for advanced renal cell carcinoma and hepato-cellular carcinoma.94,95

Sorafenib has been tested by the Eastern Cooperative Oncology group (ECOG) in a Phase II study in combination with cisplatin/docetaxel (n = 44) in the management of metastatic/unresectable GC/GEJ AC.96 The results were encouraging, with an ORR of 38.5%, TTP of 5.8 months, and OS of 14.9 months. Grade 3 or 4 neutropenia occurred in 49% of patients.

Telatinib

Telatinib is also an oral small molecule TKI with selective inhibitory effect against VEGFR and PDGFR. It is not currently FDA approved. Ko et al97 conducted a Phase II study of telatinib in combination with capecitabine/cisplatin as first-line treatment in patients with advanced GC/GEJ tumors. The best response in the seven patients evaluated included two with partial response and four with stable disease. One patient progressed while on treatment. Grade 3 SAEs that have been reported so far include hand-foot syndrome, fatigue, hypertension, anorexia, febrile neutropenia, and pulmonary embolism. No grade 4 SAEs have been recorded so far.

The clinical trials of anti-VEGF agents are summarized in Table 3.

Mammalian target of rapamycin (mTOR)

Mammalian target of rapamycin (mTOR) serinethreonine kinase is a downstream component of the phosphatidylinositol 3-kinase/Aktkinase signaling pathway. This pathway regulates cellular metabolism and growth by acting as a cellular sensor for nutrients and growth factors. Upregulation of this pathway has been associated with worse prognosis in GC98 and also in squamous cell esophageal cancer.99 It has been postulated to contribute to chemotherapy resistance.100

Everolimus (Afinitor®, Novartis, Basel, Switzerland) blocks the mTOR pathway by forming a complex with the immunophilin FK506-binding protein-12. It is currently FDA approved for use in progressive neuroendocrine tumors of pancreatic origin and advanced renal cell carcinoma after failure of first-line sunitinib or sorafenib. Phase I data exist showing marked tumor response to everolimus in esophageal carcinoma;101 however, as yet no Phase II data are available investigating its use in esophageal cancer.

More progress has been made investigating the use of everolimus in metastatic gastric cancer. Doi et al102 reported a Phase II study of everolimus in this population with promising results. Fifty-three patients who had previously received chemotherapy for metastatic GC received everolimus. There were no complete or partial responses in this trial; however, 45% of patients had a reduction in tumor size from baseline. Furthermore, OS was 10.1 months, which compares favorably with other single-agent trials,103,104 where OS ranges from 3.5 to 7.2 months and combination trials,105,106 where OS ranges from 6.0 to 10.7 months. Recently, the results of the Phase III GRANITE-1study evaluating the role of everolimus in previously treated advanced GC patients were released at the ASCO GI 2012 meeting, concluding that everolimus monotherapy does not significantly improve the OS and PFS.107 A total of 656 patients with advanced GC were randomized in 2:1 fashion to everolimus/best supportive care compared with placebo/best supportive care. The OS was 5.39 months with everolimus compared with 4.34 months in the placebo arm (HR, 0.90; 95% CI, 0.75–1.08; P = 0.1244). Median PFS per local investigator assessment was 1.68 and 1.41 months with everolimus and placebo respectively (HR, 0.66; 95% CI, 0.56–0.78; P < 0.0001). The ORR was also not very encouraging (4.5% with everolimus versus 2.1% with placebo).

Novel targets

Marimastat

Marimastat is an MMP inhibitor and has been tested in a Phase III study (compared with placebo) including 369 patients with inoperable/metastatic GC/GEJ AC.108 There was a small but statistically significant improvement in OS (160 versus 138 days; P = 0.02) and 2-year survival (9% versus 3%), favoring the marimastat arm. However, this drug’s further development has stopped.

Foretinib

Another novel target, c-MET, which is a receptor for hepatocyte growth factor is currently under evaluation.109 The c-MET oncogene is amplified in 10%–15% of GC, and its overexpression has been associated with poor prognosis.110,111 Foretinib is an oral TKI with activity against VEGFR-2 and c-MET. It was evaluated at two different doses in a Phase II study (n = 64) involving previously treated patients with EGC.108 The ORR was 0%, with stable disease in 21% and 25% of patients in each dose cohort. Amplification of c-MET was lower than expected (5%) in this study, and future studies are planned limiting the use of foretinib in patients with c-MET amplification only.

Bryostatin

Bryostatin-1 is an inhibitor of protein kinase C, which plays a role in mediation of anti-apoptotic signals.112 Two Phase II studies have evaluated sequential taxol and bryostatin in EGC. Grade 3/4 myalgias occurred in about half of the patients, leading to cessation of further development of this drug.113,114

Cyclo-oxygenase-2 inhibitors

Cyclo-oxygenase-2 (COX-2) is an enzyme involved in the prostaglandin synthesis and malignant transformation of Barrett’s esophagus. The use of aspirin and other nonsteroidal agents have been associated with lower esophageal cancer rates. In a meta-analysis including nine studies with more than 1800 patients, there was 43% risk reduction of developing esophageal cancer with these agents.115 However, the chemoprevention for Barrett’s esophagus trial did not find any difference in the primary outcome – change from baseline to 48 weeks of therapy in the proportion of biopsy specimens with dysplasia in the celecoxib and placebo arms.116 The criticism of this study was the short treatment duration and follow up, but the negative results in conjunction with the known cardiotoxicity of COX-2 inhibitors117 makes routine use of these agents as chemoprevention inappropriate.

Other novel agents

Studies are ongoing with other novel agents targeting insulinlike growth factor receptor (IGF), its ligand IGF-1, and telomerase enzyme.118

Conclusion

In conclusion, the prognosis associated with advanced EGC remains poor. Various molecular targeted therapies have been evaluated in clinical trials with trastuzumab now becoming a standard treatment option for patients with Her2-positive EGC. Further research in this field is urgently needed to improve outcomes for patients.