The poor long-term outcomes associated with current chemotherapy treatment of patients with advanced gastric cancer suggest a need for novel targeted agents that may confer a better survival benefit. Evidence of mammalian target of rapamycin (mTOR) activation has been demonstrated in patient-derived gastric cancer cells and tumors. This review explores the relevance of the mTOR pathway to gastric cancer pathogenesis and its potential as a therapeutic target in patients with gastric cancer as well as presenting the first available clinical data on mTOR inhibitors in this disease setting. Preclinical data suggest that suppression of the mTOR pathway inhibited the proliferation of gastric cancer cells and delayed tumor progression in in vitro and animal models. In the clinical setting, the mTOR inhibitor everolimus has been active and well tolerated in phase I/II studies of patients with chemotherapy-refractory metastatic gastric cancer. Based on these promising results, everolimus currently is being investigated as a monotherapy or in combination with chemotherapeutic agents in ongoing phase II/III clinical studies.
Gastric cancer is the fourth most common cancer worldwide, with 640,556 new cases among men and 349,042 new cases among women annually, and it is the second most common cause of cancer death, with ∼738,069 deaths annually worldwide.1 Current management of gastric cancer is based on surgical resection of the primary tumor.2 Although surgical resection can be curative for patients with earlier stages of disease, 60% of patients in the Western hemisphere eventually relapse.3 Furthermore, most patients with gastric cancer commonly present with advanced unresectable disease. For patients with advanced disease or for those developing recurrent disease after surgery, evidence supports the use of chemotherapy to prolong survival and maintain quality of life.2, 3
At present there is no globally accepted standard chemotherapeutic regimen for the treatment of patients with gastric cancer.2 Combination chemotherapy regimens, including 5-fluorouracil (5-FU), taxanes, and platinum derivatives have been shown to prolong median overall survival in patients with advanced gastric cancer, although the duration of survival is usually less than 1 year,4–9 except for some trial results in Japanese patients.10–13 These poor long-term outcomes associated with chemotherapy treatments suggest a need for novel targeted agents that may confer a better survival benefit in patients with advanced gastric cancer. Targets currently under investigation include the mammalian target of rapamycin (mTOR) and human epidermal growth factor receptor-2 (HER2), which engage multiple downstream signaling pathways including the phosphoinositol-3 kinase (PI3K)/Akt/mTOR pathway. Both mTOR and HER2 are dysregulated in many cancers, including gastric cancer,14, 15 and may represent relevant therapeutic targets in patients with this disease. This review will focus specifically on the relevance of the mTOR pathway to gastric cancer pathogenesis and its potential as a therapeutic target in patients with gastric cancer.
mTOR is a central regulatory kinase that increases the production of proteins that stimulate key cellular processes such as cell growth and proliferation, cell metabolism, and angiogenesis.16–18 mTOR increases translation of proteins that drive cell growth and cell division, such as cyclin D1, and decreases translation of negative regulators of cell cycle progression.19 mTOR plays a role in cellular metabolism by stimulating the surface expression of nutrient transporters,20 resulting in increased availability of nutrients to fulfill the energy and metabolic requirements for mTOR-activated cell growth and proliferation. mTOR also increases the translation of hypoxia-inducible factor-1α (HIF-1α), which drives the expression of angiogenic growth factors such as vascular endothelial growth factor (VEGF), resulting in new vasculature.18
mTOR integrates intracellular cues regarding nutrient and energy availability and signals from the growth factor pathways to directly mediate cellular processes via protein translation.21 Signaling to mTOR from growth factors has been well characterized and is mediated via the PI3K pathway through Akt and the tuberous sclerosis complex proteins TSC1 and TSC2, which form a dimer to regulate mTOR activity (Fig. 1). In physiologically normal cells, PI3K/Akt-mediated inactivation of TSC2 results in proteosomal degradation of the TSC1/TSC2 protein complex, thereby permitting mTOR activation.22, 23 Depending on available protein binding partners, mTOR can form one of two complexes: mTOR complex 1 (mTORC1), composed of mTOR plus FK506-binding protein 12 kDa (FKBP12), mammalian LST8 (mLST8), and the regulatory-associated protein of mTOR (raptor); or mTORC2, composed of mTOR plus the rapamycin-insensitive companion of mTOR (rictor), Sin1 and mLST8.24, 25 Activated mTORC1 phosphorylates two effector molecules, S6 kinase 1 (S6K1) and 4E-binding protein 1 (4E-BP1), to stimulate protein synthesis supporting cell growth and proliferation, cell metabolism and angiogenesis.16–18 Activated mTORC2 phosphorylates Akt leading to its full activation. The functions of mTORC2, which include regulating the actin cytoskeleton and cell survival, and its role in tumorigenesis are not yet well defined.24, 25 Pharmacologically, mTORC1 is sensitive to mTOR inhibition by rapamycin and rapamycin derivatives such as everolimus, an oral rapamycin derivative with improved pharmacokinetic properties.24, 26 mTORC2 is insensitive to rapamycin,24 although prolonged rapamycin treatment can result in mTORC2 inhibition potentially through irreversible mTOR sequestration, thus preventing mTORC2 formation.27 In addition, mTORC1-mediated S6K1 activation may inhibit mTORC2 via a negative feedback loop, resulting in suppressed Akt signaling.28, 29 This feedback loop was recently demonstrated in a preclinical study of gastrointestinal cancer cells.30 Akt activity is also attenuated by mTORC1-mediated S6K1 phosphorylation and subsequent proteolysis of insulin receptor substrate, which results in reduced PI3K/Akt signaling.28, 29
Mutations in upstream components of the mTOR signaling pathway result in inappropriate mTOR activation,15 thus promoting tumor progression. In a genetic analysis of patients with esophageal cancer, mutations in Akt were associated with an increased risk of recurrence after chemotherapy, and mutations in phosphatase and tensin homolog (PTEN), a negative regulator of PI3K/Akt, was associated with a decreased risk of recurrence, possibly due to increased PTEN expression.31 Mutations in PI3K also have been noted.32 Regardless of the mechanism of aberrant mTOR activation, preclinical studies suggest that tumors with mutations in the mTOR pathway are uniquely susceptible to mTOR inhibitor therapy.33, 34 However, not all tumors with mTOR activation respond to mTOR inhibitor therapy. Possible explanations for this include the presence of a negative feedback loop resulting in increased PI3K/Akt/mTOR signaling even in the presence of rapamycin, involvement of mTORC2, and incomplete mTOR inhibition.35 Moreover, there are several mechanisms of drug resistance downstream of mTOR, including downregulation of 4E-BP1 and S6K1 mutations resulting in constitutive activation.36 Markers of drug-target inhibition have been developed, including inhibition of S6K1 activity in peripheral blood mononuclear cells (PBMCs), which correlated well with mTOR inhibitor exposure in clinical studies and preclinical models of human cancer.26 Further studies are needed to determine whether such pharmacodynamic measurements are useful in predicting outcomes. Clinical studies to determine relevant prognostic biomarkers to predict patient response to mTOR inhibitor therapy are ongoing.34
Preclinical rationale for mTOR inhibition in gastric cancer
Several preclinical studies have indicated dysregulation of mTOR activity in gastric cancer cell models, suggesting mTOR as a potential therapeutic target. Mutations in upstream regulators of mTOR signaling pathway epithelial growth factor receptor (EGFR),37 PI3K37, 38 and PTEN39 have been observed in patient-derived gastric tumor samples. In addition, preclinical studies have demonstrated evidence of mTOR activation in gastric cancer cells and tumors. Patient-derived gastric cancer samples have been shown to express phosphorylated mTOR indicative of mTOR activation,40–43 which has been positively correlated with tumor progression and poor survival in patients with gastric cancer.41, 43, 44
Treatment of gastric cancer cell lines with the mTOR inhibitors sirolimus or everolimus resulted in a decrease in phosphorylation of S6K1 and 4E-BP1 as well as attenuation of HIF-1α and VEGF.40, 45–47 Everolimus treatment also resulted in G1 cell cycle arrest and inhibited the proliferation of gastric cancer cell lines.45, 46, 48 Moreover, everolimus in combination with 5-FU, cisplatin, oxaliplatin or docetaxel resulted in synergistic growth inhibition of 5-FU-resistant gastric cancer cell lines.48 Consistent with the antiproliferative effects observed in vitro, mTOR inhibitors alone or in combination with other agents significantly delayed tumor progression in xenograft models of gastric cancer.40, 45
Everolimus in patients with gastric cancer
Currently, everolimus is the only mTOR inhibitor that has been investigated in phase I/II clinical trials of patients with advanced gastric cancer. In a phase I study of everolimus in 9 Japanese patients with advanced solid tumors, everolimus 10 mg day−1 resulted in a partial response with a duration of >4 months in a heavily pretreated patient with gastric cancer and liver metastases.49 Similarly, a phase I dose-escalation trial of everolimus (5–10 mg day−1) plus mitomycin C in 15 patients with advanced gastric cancer previously treated with platinum-based chemotherapy conducted in Europe also demonstrated antitumor activity. Of 13 evaluable patients, 3 (23%) experienced a partial response, one of which lasted for 1 year, and three patients had stable disease.50 There were no dose-limiting toxicities reported. Frequent grade 1/2 toxicities related to treatment included mucositis (61%), thrombocytopenia (54%), nausea (54%), leukopenia (46%), fatigue (31%) and diarrhea (23%). Grade 3/4 toxicities included leukopenia (15%), mucositis (8%), neutropenia (8%), lymphopenia (8%) and increased bilirubin (8%).50
In a recent phase II trial conducted in Japan, everolimus 10 mg day−1 was administered to 53 patients with metastatic gastric cancer previously treated with one (n = 27) or two (n = 26) prior chemotherapy regimens.51 The primary study end point was disease control rate, which was defined as the percentage of patients with a complete response, partial response, or stable disease as best overall response.51 The study showed an overall disease control rate of 56% (all with stable disease), and subgroup analyses indicated that the disease control rate benefit was consistent in patients receiving everolimus as a second- or third-line therapy.51 Although no complete or partial responses were documented, 45% of patients had a decrease in tumor size from baseline by independent radiologic review.51 Median progression-free survival was 2.7 months,51 suggesting that the primary benefit of everolimus was disease stabilization. Median overall survival was 10.1 months,51 although it may have been confounded by the 85% of patients who received additional chemotherapy after discontinuation from everolimus.51 Exploratory overall and progression-free survival subgroup analyses demonstrated that the survival benefit was consistent regardless of the number of chemotherapy regimens previously received.51 Everolimus therapy was generally well tolerated, and adverse events typically were grade 1 or 2 in severity.51 The most common adverse events were stomatitis (74%), anorexia (53%), fatigue (51%), rash (45%), nausea (32%), peripheral edema (23%), diarrhea (21%) and pruritus (19%).51 Grade 3 adverse events occurred in 20 patients and included anemia (9%), hyponatremia (9%), increased gamma-glutamyltransferase (8%) and lymphopenia (8%).51 Grade 4 adverse events that were considered to be possibly related to everolimus treatment occurred in four patients (tumor hemorrhage, increased γ-glutamyltransferase, lymphopenia, and cerebral infarction).51
The efficacy of everolimus in patients with previously treated metastatic gastric cancer was supported by preliminary results from a phase II study conducted in Korea. Of 54 patients with chemotherapy-refractory advanced gastric cancer treated with everolimus 10 mg day−1, 2 (4%) achieved confirmed partial responses and 19 (35%) had stable disease, resulting in a disease control rate of 39%. The 4-month progression-free survival rate, the primary end point, was 18.4% with a median progression-free survival of 1.7 months. Median overall survival was 8.3 months. Treatment was generally well tolerated; the most common grade 3/4 adverse events included hepatic dysfunction (11%), anemia (9%) and thrombocytopenia (9%).52
Emerging treatment paradigms look promising for treating patients with advanced gastric cancer. However, additional work is needed to optimize treatment regimens and determine biomarkers predictive of therapeutic success. Available clinical trial data show that objective responses are rare in patients with gastric cancer receiving mTOR inhibitors. This raises questions regarding the prevalence of inappropriate mTOR activation in gastric cancer and whether the administration of mTOR inhibitors to unselected patient populations will be able to show expected results.
Patient selection based on molecular events known to be associated with mTOR activation such as the overexpression of PI3K/Akt and the growth factor receptors HER2 and insulin-like growth factor receptor (IGFR) as well as mutations in PI3K and mutations/amplifications of Akt or down regulation of PTEN may represent an appropriate way to identify populations that are more likely to show significant benefit from mTOR inhibitors. Unfortunately, little is known about the molecular profiles of patient who responded to everolimus in initial trials, although some clinical studies currently enrolling patients (NCT01049620, NCT00632268; Table 1) include biomarker analyses.
Table 1. Ongoing trials of everolimus in gastric cancer
Two phase III trials evaluating everolimus in gastric cancer are ongoing. The GRANITE-1 (Gastric antitumor trial with everolimus-1) compares everolimus/supportive care with placebo/best supportive care in patients with advanced gastric cancer after progression on one or two prior systemic chemotherapies, with a target enrollment of 633 patients (NCT00879333). The primary end point is overall survival. The AIO-STO-0111 (RADPAC) trial, on the other hand, evaluates paclitaxel monotherapy with or without everolimus in the second- or third-line setting. The study is conducted by the German Arbeitsgemeinschaft Internistische Onkologie, with a target enrollment of 480 patients and overall survival as the primary end point (NCT01248403). Both of these phase III studies involve administration of everolimus to unselected patient populations, and both are accompanied by exploratory biomarker research programs that will examine the predictive relationship of phosphorylated S6K1, HER2, phosphorylated Akt, HIF-2α, PTEN, cyclin D1, Ki-67 frequency, p53 and CC3, as well as the mutational status of PI3K catalytic subunit and PTEN, with efficacy endpoints. Irrespective of whether the studies will meet their primary end points, it will be very important to identify potential markers of response to everolimus and validate their role in future studies. However, the appropriate assessment of biomarkers remains challenging. The quantification of immunohistochemistry with cyclin D1, PTEN and phospho-specific antibodies in formalin-fixed tissues of solid tumors has not been fully standardized, and significant concerns exist about the stability of phospho-proteins, which are known to be influenced by type of sampling (e.g., biopsy versus surgical resection) and subsequent processing.
Another path to improve efficacy of mTOR inhibitors is combination with either cytotoxic agents or other targeted therapies. Based on preclinical studies showing synergistic effects with some cytotoxic agents,34 some ongoing clinical studies currently are evaluating everolimus in patients with gastric cancer in combination with chemotherapeutic agents (Table 1). The finding that mTOR inhibition has been shown to induce feedback activation of Akt provides a rationale for the combination of mTOR inhibitors with targeted drugs that can inhibit Akt activation such as small-molecule inhibitors of Akt, the anti-HER2 antibody trastuzumab and IGF-IR inhibitors. A recent phase III trial demonstrated that the addition of trastuzumab to chemotherapy improved outcomes in patients with metastatic gastric cancer who overexpressed HER2,53 a feature found in ∼20% of patients.54 Notably, loss of PTEN, a negative regulator of PI3K/Akt/mTOR, has been shown to mediate trastuzumab resistance.55, 56 Taken together, the data provide a foundation to evaluate the combination of mTOR inhibitors and trastuzumab in HER2-positive gastric cancer.
To more effectively inhibit the mTOR pathway, it will also be important to better understand the role of mTOR-dependent networks, its relationship to compensatory pathways, and the contribution of mTORC2 in the progression of gastric cancer. This increased knowledge can provide the foundation for generating novel strategies to improve therapeutic success. For example, it has been hypothesized that inhibition of both mTORC1 and mTORC2 would be a more effective way of treating tumors, as the presence of feedback loops (e.g., via Akt and RAS/MAPK signaling) and the possible contribution of mTORC2 to tumor progression may reduce the efficacy of mTORC1 inhibitors.27–29 To circumvent these potential limitations, novel inhibitors of both mTORC1 and mTORC2 (e.g., OSI-027, BEZ235 and XL765) are being currently evaluated in phase I/II trials of patients with solid tumors.57
The authors thank Scientific Connexions, Newtown, PA, USA for providing assistance with medical writing and editing. S.E. Al-Batran has received research funding from and serves in an advisory role for Novartis Pharmaceuticals Corporation. M. Ducreux has received research funding (French investigator for the GRANITE study) from and served as a speaker for Novartis Pharmaceuticals Corporation. A. Ohtsu declares no conflicts of interest.