Federica Morano, MB, BCh; Sing Yu Moorcraft, MB, BCh; and Ian Chau, MD
Gastric cancer is the fourth most common cause of cancer-related death and is responsible for approximately 107,000 deaths every year.1 Unfortunately, patients often present with advanced disease at diagnosis, and their prognosis remains poor despite improvements in treatment. Over the past few years, cancer immunotherapy has shown increasingly exciting results in melanoma, where it was first tested, as well as in breast, prostate, kidney, and lung cancers. For this reason, immunotherapy was selected as the 2013 breakthrough of the year by Science.2 There is increasing interest in developing immunotherapeutic strategies for the treatment of gastric cancer.
Dr. Federica Morano
Dr. Sing Yu Moorcraft
The Role of the Immune System in Cancer
The immune system has the ability to differentiate between “self” and “non-self.” The concept that the immune system is also capable of detecting and killing nascent malignant cells was first developed by Sir Frank Macfarlane Burnet, MD, PhD, and Lewis Thomas, MD, in their cancer immunosurveillance hypothesis.3-5 Initially abandoned, this concept gradually regained support as a result of strong experimental data, and it is now thought to be a component of cancer immunoediting.6
The immunoediting process has three main phases: elimination, equilibrium, and escape. During the elimination phase, the growth of neoplastic cells and their resulting size may cause changes in the microenvironment and local damage. Innate immune system cells, such as natural killer cells and natural killer T cells, secrete interferon-gamma and cytokines, as well as inhibiting angiogenesis and tumor cell proliferation. Tumor cell debris is then digested by dendritic cells (DC) and presented to T cells. Eventually, tumor-specific CD4+ and CD8+ T cells help to identify and kill the remaining antigen-bearing tumor cells. The equilibrium phase is usually the longest of the three phases. In this process, CD8+ T cells, DCs, and tumor cells are in a phase of dynamic balance, and the tumor cells are quiescent. Unfortunately, during this long process genetically unstable tumor cells may survive and acquire resistance, and this leads to the escape phase.
Another milestone was identification of the first human tumor antigen recognized by T cells by Van der Bruggen and colleagues.7 This resulted from cloning of the melanoma antigen-encoding gene (MAGE), which encodes antigens recognized by cytotoxic T cells. This discovery was not only new proof of the ability of our immune system to seek and destroy tumor cells but also the first identification of the molecular nature of these antigens.
Immunotherapy: Cancer Vaccines
Cancer vaccines are designed to enhance the ability of the human immune system to seek and destroy tumor cells by boosting tumor-specific T lymphocytes. In particular, tumor antigens must be presented to T cells by dedicated antigen-presenting cells (e.g., DCs). Usually, these antigens are small intracellular peptides presented by major histocompatibility complex (MHC) molecules on the surface of tumor cells to cytolytic T cells. Different MHC molecules present tumor antigens to different T cells; MHC class I is responsible for presenting the antigen to CD8+ T cells, whereas MHC class II presents antigens to CD4+ T cells.
There is increasing interest in developing immunotherapeutic strategies for the treatment of gastric cancer, which is the fourth most common cause of cancer-related death and is responsible for approximately 107,000 deaths every year.
Tumor antigens are being discovered and tested and may have an important role in the development of new cancer vaccines.
The Cancer Genome Atlas Research Network recently analyzed the molecular characteristics of gastric adenocarcinoma, identifying four tumor subtypes. The Epstein–Barr virus subgroup showed elevated PD-L1 expression, suggesting the robust presence of immune cells and supporting the use of immune checkpoint inhibitors in gastric cancer.
Pembrolizumab, a monoclonal antibody designed to block the interaction between PD-1 and its ligands PD-L1 and PD-L2, was recently tested in patients with gastric cancer in a phase Ib trial. Of the patients treated, 41% experienced a decrease in tumor burden. Furthermore, preliminary evidence of a relationship among progression-free survival, overall response rate, and PD-L1 expression was observed. Several trials of other PD-1/PDL-1 inhibitors alone or in association with other immunomodulating agents are now ongoing.
To date, thousands of tumor antigens have been discovered and tested. The MAGE gene, first identified in patients with melanoma, is variably expressed in different solid tumors. It is also present in normal cells, but it is usually silent. Other potential antigens identified are peptides deriving from mutations (e.g., KRAS and passenger-related mutations), differentiation antigens, overexpressed antigens (e.g., HER2/neu and carcinoembryonic antigens), and viral antigens (e.g., HPV).
Thirty-eight percent of gastric cancers express MAGE, and preclinical data also showed how Helicobacter pylori may induce the expression of MAGE-3.8,9 Yang et al. used a nanovaccine loaded with a MAGE-3 peptide to enhance the immune response in a mouse model of gastric cancer and achieved tumor regression.10 A phase I/II study from Japan used peptides derived from VEGF-R1 and VEGF-R2 in association with S-1 and cisplatin in 22 patients with advanced gastric cancer, resulting in a partial response in 55% of patients and prolonged overall survival.11 Furthermore, tumor antigens associated with HER2/neu, typically overexpressed in gastric cancer, may have an important role in the development of new immunotherapy and vaccine strategies.
Immunotherapy: Adoptive Cell Therapy
“Adoptive cell therapy” is another type of immunotherapy. Tumor-specific T cells are isolated from a patient and amplified in vitro. These cells are subsequently re-infused into the patient in large numbers. It is also now possible to genetically modify these cells in vitro before their transfer. Different types of cells may be used as activated killer cells, and several have been tested in gastric cancer, including tumor-infiltrating lymphocytes.12
In a recent trial, Kono et al. treated 44 patients with advanced gastric cancer with tumor-associated lymphocytes with or without chemotherapy.13 Their results showed that overall survival was improved in patients treated with chemotherapy and tumor-associated lymphocytes. Moreover, cytokine-induced killer cells have been tested in gastric cancer.14,15 Clinical trials confirmed that patients treated with chemotherapy combined with cytokine-induced killer cells, amplified in vitro with a combination of interleukin-2 and anti-CD3 antibodies, have a survival benefit compared to patients treated with chemotherapy alone. However, the high specificity of these types of transferred killer cells is of paramount importance in order to reduce the risk of life-threatening side effects related to a massive immune response.
The activity of T cells must be carefully regulated in order to avoid excessive proliferation and tissue damage. Several checkpoints are involved in this process. CTLA-4 and PD-1 are both inhibitory receptors expressed by T cells. These molecules usually appear on the surface of T cells after their activation and deliver a negative signal. The blockade of these receptors by antibodies causes increased activity of T cells and robust clinical activity in many tumor types (Fig. 1).
More recently, The Cancer Genome Atlas Research Network analyzed the molecular characteristics of gastric adenocarcinoma.16 Four tumor subtypes were identified: tumors positive for Epstein–Barr virus, microsatellite unstable tumors, genomically stable tumors, and tumors with chromosomal instability. The Epstein–Barr virus subgroup (15% of tumors) showed elevated PD-L1 expression, suggesting the robust presence of immune cells and supporting the use of immune checkpoint inhibitors in gastric cancer.
In a small phase II trial, tremelimumab, a fully humanized anti–CTLA-4 monoclonal antibody, was tested as a second-line therapy in 18 patients with gastric cancer.17 Although the objective response rate was 5%, the median survival was 4.8 months and, therefore, similar to that expected with other chemotherapies in gastric cancer. However, this led to the idea that the combination of two checkpoint inhibitors may offer clinical benefit in patients with advanced gastric cancer. MEDI4736 is a human IgG1 monoclonal antibody that binds to PD-L1 and prevents its binding to PD-1 and CD80. Preliminary data from a phase I study showed encouraging clinical activity of this drug in multiple tumor types.18 To date, several trials of MEDI4736 alone or in association with other immunomodulating agents, such as tremelimumab, are ongoing (NCT02340975).
Another antibody that blocks the interaction between PD-1 and the corresponding ligand PD-L1 is nivolumab. Nivolumab has shown encouraging efficacy in many tumor types. In December 2014, the U.S. Food and Drug Administration (FDA) approved the use of nivolumab for the treatment of patients with unresectable or metastatic melanoma whose disease progressed following treatment with ipilimumab (and a BRAF inhibitor if BRAFV600-mutation positive). More recently, nivolumab has been licensed as a second-line therapy in squamous non–small cell lung cancer based on an improvement in overall survival compared with docetaxel (9.2 months vs. 6.0 months, HR 0.59; p = 0.00025 [NCT01642004]). In gastric cancer, phase I trials of this anti–PD-L1 immunostimulating antibody alone or in combination are currently ongoing. Preclinical data showed that the dual blockade of PD-1 and CTLA-4 was associated with increased cytokine release and increased proliferation of CD8+ and CD4+ T cells when compared with single receptor blockade.19,20 An ongoing phase Ib/II trial is investigating the activity of single-agent nivolumab and nivolumab plus ipilimumab in patients with metastatic gastric cancer, pancreatic cancer, triple-negative breast cancer, and small cell lung cancer.21
Pembrolizumab is a highly specific, humanized monoclonal IgG4 antibody designed to block the interaction between PD-1 and its ligands PD-L1 and PD-L2. This antibody has shown promising results in 135 patients with metastatic melanoma in the large randomized phase I trial, KEYNOTE-001 (NCT01295827).22,23 The use of pembrolizumab resulted in an overall response rate (ORR) of 26% and had a manageable safety profile, leading to FDA approval for the treatment of melanoma. More recently, Muro et al. investigated the safety and activity of pembrolizumab in patients with gastric cancer in a phase Ib trial.24 One hundred sixty-five patients with advanced gastric or gastroesophageal junction tumors were screened. Forty percent of patients (65 patients) were PD-L1 positive (defined as PD-L1 staining in stroma or ≥ 1% of tumor cells), and 39 of these patients were treated with pembrolizumab (10 mg/kg) every 2 weeks. Forty-one percent of patients experienced a decrease in tumor burden. The ORR was 32% in Asian patients and 30% in non-Asian patients. The treatment was very well tolerated in this population of patients who were heavily pretreated. Furthermore, preliminary evidence of a relationship between progression-free survival, ORR, and PD-L1 expression was observed.
Cancer immunotherapy has now come of age. Astonishing results have been obtained, especially in melanoma, where this therapeutic approach has led to improvements in survival. In gastric cancer, several promising steps have been made, but more are necessary. Furthermore, the discovery of effective biomarkers would represent an important step in the identification of the optimal patients for immunotherapy. This would not only increase the proportion of patients who derive benefit, but would also spare patients who would not benefit from potentially substantial toxicities and also help make these therapeutic options more economically viable. Future clinical trials should, therefore, include a substantial translational research component, investigating potential biomarkers such as PD-L1 expression, immune infiltration (e.g., CD8/ regulatory T cell ratio and chemokines), and tumor mutational frequency.
About the Authors: Drs. Morano and Moorcraft are both clinical research fellows with the Royal Marsden NHS Foundation Trust. Dr. Chau is a consultant medical oncologist in the Gastrointestinal and Lymphoma Unit at the Royal Marsden and honorary senior lecturer at the Institute of Cancer Research.