Division of Biomarker Discovery (Tsukiji Campus)

Hiroki Sasaki, Rie Komatsuzaki, Fumiko Chiwaki, Tomoko Hiroki, Kazuaki Yamamoto, Varenik Pavel, Masayuki Komatsu, Akio Ashida, Kanako Nakamura, Daichi Inami, Tsukasa Fujimoto

Introduction

The two major research areas of the Division of Biomarker Discovery are 1) preclinical studies using newly established gastric, esophageal, pancreatic, and ovarian cancer cell lines for the derivation of industrial and academia seeds/drugs to the Exploratory Oncology Research & Clinical Trial Center (EPOC), and 2) basic research and development of personalized cancer diagnosis and treatment for gastric cancer (GC) and esophageal squamous cell carcinoma (ESCC).

Research activities

1.Preclinical Studies Using Newly Established Cell Lines from Common Cancers in Asia

Genome-wide genetic information in about 1,000 cancer cell lines is available on COSMIC DB (Sanger Center, UK) and on recent NGS analyses (Klijin C et al., Nat. Biotech. 2015); however, among them, only 28 cell lines are derived from GC. Since driver gene mutation frequency in a certain cancer is often less than 5%, the establishment of cell lines from each patient to be analyzed is desired for functional selection of driver gene mutations. Furthermore, almost all of the 28 GC cell lines were established many years ago, thereby, their clinical and pathological information is insufficient. The wait is on for the establishment of new GC cell lines, especially from metastatic sites after therapy. Peritoneal metastasis is most frequent in GCs, especially diffuse-type GCs. In 2016, we successfully established 13 diffuse-type GC cell lines from the cancer ascites of 12 patients. In collaboration with the Division of Genetics, in total, we have established 78 diffuse-type GC cell lines (National Cancer Center Stomach Cancer (NSC) series) from 46 patients, and also established 18 pancreatic and two ovarian cancer cell lines, and so on. We are conducting omics analyses for gene expression and copy number variation, and hot spot- and genome wide-gene alteration in these cell lines. Moreover, for in vivo preclinical study, their tumorigenicity and histopathological characteristics in the xenograft, such as fibroblast rich-, hypovascular-, and dormant-state, were evaluated. Through collaboration with five pharmaceutical industries, in vitro and in vivo preclinical studies were conducted to derivate clinical trials in the EPOC.

2.Development of Personalized Diagnosis and Treatment for GC and ESCC

The study for GC: GC is one of the leading causes of cancer-related deaths worldwide. Histopathologically, GCs can be divided into two major categories: intestinal-type and diffuse-type. Unlike the decreasing incidence of intestinal-type, the prevalence of the diffuse-type is reportedly increasing worldwide. Although therapeutic results for GC have recently been improved, the prognosis of patients with advanced diffuse-type GC still remains poor. Peritoneal dissemination is a frequent form of metastasis of diffuse-type GC. The survival rates of patients with peritoneal metastasis (P1) at three and five years are only 9.8% and 0%, respectively. Peritoneal lavage cytology (CY) provides important prognostic information for GC after surgery. CY positive (CY1) is well known as a poor prognostic factor in advanced GC patients. However, the optical therapeutic strategy for patients with CY1 has not yet been established. The two-, three-, and five-year survival rates of GC patients with no peritoneal metastasis (P0) but with CY1 are 25.3, 13.8, and 7.8%, respectively. Development of a new therapeutic modality for peritoneal metastasis of GC is very important for improving the outcome of those patients with P1/CY1 or P0/CY1. In 2016, we identified DDR2 as a potential regulator of peritoneal dissemination by collaborative integrated molecular profiling. For personalized medicine, we have developed mini DNA chips containing six markers and three control genes for predicting GC recurrence from peritoneal washings. The CY offers important prognostic information for GC after surgery, but has only a limited sensitivity and the task requires great skill. Our collaborating company continues to prepare a lot of supporting data for developing an in vitro diagnostic (IVD).

The study for ESCC: Definitive chemoradiotherapy (CRT) is a less invasive therapy for ESCC; however, the five-year survival rate of locally advanced ESCC patients was only 37%. Therefore, a prediction of CRT-responder is awaited. We have successfully identified five intrinsic subtypes (1a/M1, 2a/I, 3b, 5/M2, and 7/E) of ESCCs by gene expression profile-based unsupervised clustering of 274 biopsy samples obtained before treatment. For the cases treated with CRT, the five-year survival rate was 24% in subtype M2, whereas it was 74% in subtype E. Furthermore, we found transcriptional pathways activated characteristically in each subtype; subtype E showed a differentiation phenotype, while non-E subtypes including M1 and M2 showed an epithelial-mesenchymal transition phenotype. In 2016, we reported that SIX1 maintains tumor basal cells via TGF-s pathway and associates with poor prognosis in non-E subtypes, and are preparing the investigator-initiated clinical trial of the immune checkpoint inhibitor for subtype I that shows tumor-specific cytotoxic T-lymphocyte activation signatures by CRT. To develop an IVD for predicting subtype E, we are collaborating with a pharmaceutical company and ten institutes for the clinical investigation of 225 ESCC patients with support of the Japan Agency of Medical Research and Development.

Education

We supported a postdoctoral fellow through design of a research, writing of paper, and grant applications, we also nurtured five assisting scientists, a graduate school student, and two undergraduate school students though experiments, discussion, and writing of a report and paper.

Future prospects

To develop IVD and to derivate new drugs to EPOC, collaboration with pharmaceutical industries will be continued.