Purpose: This study aimed to identify novel hypoxia-inducible and prognostic markers in vivo from hypoxic tumor cells.Experimental Design: Using carbonic anhydrase 9 and CD34 as a guide for hypoxic tumor cells, laser capture microdissection was used to isolate colorectal cancer (CRC) liver metastases. The samples were analyzed by microarray analysis, in parallel with five CRC cell lines cultured under hypoxic conditions. To evaluate the prognostic impact of the expression of certain genes, samples from a total of 356 CRC patients were analyzed by microarray or quantitative reverse transcription-PCR. In vitro mechanistic studies and in vivo therapeutic experiments were also done about a histone H3 Lys 9 demethylase, Jumonji domain containing 1A (JMJD1A).Results: Several candidate genes were identified by microarray analysis of liver metastases and culturing of CRC cells under hypoxic conditions. Among them, we found that JMJD1A was a novel independent prognostic factor for CRC (P = 0.013). In vitro assays revealed that loss of JMJD1A by small interfering RNA treatment was associated with a reduction of proliferative activity and decrease in invasion of CRC cell lines. Furthermore, treatment with an adenovirus system for antisense JMJD1A construct displayed prominent therapeutic effects when injected into established tumor xenografts of the CRC cell lines HCT116 and DLD1.Conclusions: JMJD1A is a useful biomarker for hypoxic tumor cells and a prognostic marker that could be a promising therapeutic target against CRC. Clin Cancer Res; 16(18); 4636-46. ©2010 AACR.

Immunohistochemical staining of E-cadherin and CD44 may help to identify a subgroup of high-risk patients with Stage II CRC, especially in colon cancer, who may need intensive follow-up and appropriate therapeutic strategy.

Abstract. Angiogenesis is a compensatory mechanism that enables malignant tumors to survive in an oxygen-deficient environment. To test our hypothesis that hypoxia stimulates the production of angiopoietin-2 (Ang-2) in colorectal cancer (CRC), we investigated the expression of Ang-2 in three cultured CRC cell lines, and in specimens from 11 CRC metastatic liver tumors. Hypoxia-induced Ang-2 mRNA expression was clearly evident in HCT116 cells that did not express Ang-2 under normoxic conditions. Ang-2 mRNA was detected only after 48 h in hypoxic serum-deprived cultures in a LoVo cell line, and under both normoxic and hypoxic conditions without any noticeable difference in the HT29 cells. There was a stepwise increase in Ang-2 expression from the periphery to the central part of the liver metastatic foci, whereas an inverse result was noted in tumor blood vessels, with a gradual decrease in CD31-positive ECs from the edge to the central region of the metastatic lesion. An expression pattern similar to Ang-2 was found in glucose transporter 1 (Glut-1), a known hypoxia-induced factor. These findings suggest that hypoxia plays an important role in inducing the expression of Ang-2 in CRC.

Esophageal endoscopic submucosal dissection (ESD) is technically difficult. To make it safer, we developed a novel method using overtube with a traction forceps (OTF) for countertraction during submucosal dissection. We conducted an ex vivo animal study and compared the clinical outcomes between OTF-ESD and conventional method (C-ESD). A total of 32 esophageal ESD procedures were performed by four beginner and expert endoscopists. After circumferential mucosal incision for the target lesion, structured as the isolated pig esophagus 3 cm long, either C-ESD or OTF-ESD was randomly selected for submucosal dissection. All the ESD procedures were completed as en bloc resections, while perforation only occurred in a beginner's C-ESD procedure. The dissection time for OTF-ESD was significantly shorter than that for C-ESD for both the beginner and expert endoscopists (22.8 ± 8.3 min versus 7.8 ± 4.5 min, P < 0.001, and 11.3 ± 4.4 min versus 5.9 ± 2.5 min, P = 0.01, resp.). The frequency and volume of the submucosal injections were significantly smaller for OTF-ESD than for C-ESD (1.3 ± 0.6 times versus 2.9 ± 1.5 times, P < 0.001, and 5.3 ± 2.8 mL versus 15.6 ± 7.3 mL, P < 0.001, resp.). Histologically, muscular injury was more common among the C-ESD procedures (80% versus 13%, P = 0.009). Our results indicated that the OTF-ESD technique is useful for the safe and easy completion of esophageal ESD.