Annals of Surgical Oncology

, 16:1068

Challenges in Developing Robust Genetic Markers and Targets to Predict and Prevent Distant and Peritoneal Recurrence in Gastric Cancer


  • D. Ziogas
    • Biobank Personalized Cancer Biomedicine, Department of Surgery, School of MedicineUniversity of Ioannina
    • Biobank Personalized Cancer Biomedicine, Department of Surgery, School of MedicineUniversity of Ioannina
Gastrointestinal Oncology

DOI: 10.1245/s10434-008-0300-9

Cite this article as:
Ziogas, D. & Roukos, D.H. Ann Surg Oncol (2009) 16: 1068. doi:10.1245/s10434-008-0300-9

Mimori et al. in a recent issue of the Journal reported interesting data on membrane type 1 matrix metalloproteinase (MT1-MMP) expression in peripheral blood and bone marrow of 810 patients with gastric cancer.1 This large-scale study evaluated the potential utility of MT1-MMP as a marker to predict risk of distant metastasis and peritoneal dissemination in gastric cancer.

Key genes may drive the metastatic ability of an individual primary tumor and may govern metastasis or recurrence pattern in specific distant sites or organs. Identifying these genes, their genetic variants, and their deregulated signaling pathways might lead to the development of both markers to predict metastasis and combination of targeted agents to inhibit activated signaling pathways and prevent distant failures in gastric cancer and other solid tumors.24 However, despite major advances with high-throughput technologies, there are multiple challenges in translating basic science discoveries into personalized anticancer therapeutics and personalized risk-stratification-based preventive medicine.5

Recently, a research team led by Massagué identified a group of genes involved in angiogenesis that may be responsible for distant metastasis of breast cancer, particularly in the lung.6 The researchers found that four genes, the epiregulin (EREG), which is a ligand for the epidermal growth factor receptor, the cyclooxygenase COX2, and matrix metalloproteinase 1 (MMP1) and MMP2 genes, facilitated the formation of new tumor blood vessels, the release of cancer cells into the blood stream, and the penetration of tumor cells from the blood into the lung. The authors believe that these gene targets might be used in the future to stop breast cancer spread.6

Metastasis is the leading cause of death in cancer patients. Through numerous biological functions, it enables cancer cells to collectively migrate from the primary tumor, disseminate, and reach distant organs. Indeed, despite adequate locoregional tumor control, due surgery, and adjuvant chemoradiation, most patients with advanced, stage II or III resectable gastric cancer die because of peritoneal and distant recurrence.

Most recently, two high-quality randomized controlled trials from Japan testing the efficacy and extent of lymphadenectomy and adjuvant chemotherapy, respectively, have both demonstrated the problem of peritoneal recurrence.7,8 In the first, surgical trial, peritoneal recurrence rate (39.4%) 5-years after D2 gastrectomy was the most frequent failure among patients who had locoregional or systemic relapse, and an overall relapse rate of 16.4% (43/261) was recorded.7 In the second trial, although adjuvant S-1 chemotherapy after D2 surgery statistically significantly reduced rate of peritoneal relapse, it was 11% after short follow-up of 3 years.8 In the West, systemic and peritoneal recurrence rates are much higher, despite quality control of surgery for standardized D2 gastrectomy.9,10

How could distant and peritoneal events be accurately predicted? The tumor–node–staging (TNM) system and clinicopathologic factors are suboptimal. Several risk factors, including serosa invasion (T3/T4 tumor), linitis plastica, and diffuse-type cancer, have been reported to be associated with increased risk of peritoneal recurrence. Peritoneal cytology either at staging laparoscopy or at open laparotomy is a reliable test for positive result, but the negative result is of less clinical value.

Therefore, there is a need to develop markers to predict patients at high risk for distant metastasis and peritoneal dissemination. It should be emphasized that, in the absence of specific therapy, beyond systemic adjuvant chemotherapy, to prevent peritoneal dissemination, the identification of patients at high risk has limited applications. Indeed, despite some promising findings with prophylactic hyperthermic intraperitoneal chemotherapy (HIPEC), there is no high-quality evidence that it effectively reduces peritoneal relapse.9

Currently, there are two goals in the potentially curative treatment of gastric cancer: firstly, to develop markers to predict risk of distant and peritoneal recurrence; and secondly, to identify novel targets to stop dissemination and prevent recurrence. Tailoring a combination of adjuvant systemic and targeted therapies at high-risk individual patients with a high response probability to a specific combination therapy for these individual tumors could lead to a dramatic improvement of survival.2,3

To approach these goals, Mimori et al. used complementary DNA (cDNA) microarray analysis of total RNA from whole bone marrow blood from six cases with metastasis and three cases without metastasis. Among identified metastasis-related genes, the authors focused on MT1-MMP expression in bone marrow and peripheral blood from 810 patients with gastric cancer through a quantitative real-time reverse-transcriptase polymerase chain reaction (RT-PCR).1 Mimori et al. conclude that expression of MT1-MMP in peripheral blood from gastric cancer cases is a predictor of peritoneal dissemination, and that presence of MT1-MMP-expressing cells in bone marrow indicates higher risk for distant metastasis in gastric cancer cases. Further research on matrix metalloproteinases as potential targets might lead to the discovery of novel anticancer drugs which could be applied selectively only in gastric cancer patients with MMP-positive expression.

However, given the highly complex metastatic process of most solid tumors, including several biological steps, tumor heterogeneity, and microenvironment, beyond angiogenesis, there is skepticism about whether MMPs could be used in the clinic as a single marker to predict recurrence.11 Perhaps prospective validation of data in other patient population might provide useful data on the potential predictive role of MMPs in gastric cancer.

Development of robust biomarkers for recurrence risk stratification and response prediction has proved to be a very difficult approach. Despite efforts based on new technologies in the fields of gene expression profiling, pharmacogenomics, and most recently with genome-wide association studies there has been limited or no clinical success in tailoring the best treatment for patients with solid tumors, including gastric cancer.2,12 On the other hand, important advances have been taken in the field of personalized primary prevention of gastric cancer and breast cancer as well as in the treatment of breast cancer.1316

Major promise for personalized treatment approaches is provided by next-generation sequencing technology with newer genotyping platforms. Genome-wide association studies (GWAS) have already identified a numerous of novel genetic risk variants, single-nucleotide polymorphisms (SNPs), and copy-number variants (CNVs), which increase a person’s risk for the development of a complex disorder.17 Human genetic variation could also be used to predict a patient’s risk of recurrence. However, at present none of these risk variants, including SNPs and CNVs, can be used in the clinic as personal genetic tools for risk prediction and preventive or therapeutic medicine.18,19

Copyright information

© Society of Surgical Oncology 2009