We have read the report by Huh et al. [1] with interest, which was published in the recent issue of Journal of Surgical Oncology. The authors examined the CD133 expression in colorectal cancer to analyze its relationship with microsatellite instability (MSI) and the clinicopathological factors. They examined the CD133 mRNA levels in 61 primary colorectal adenocarcinomas by reverse transcriptasepolymerase chain reaction and showed that a CD133 mRNA expression was significantly associated with the depth of invasion, lymph node involvement, and lymphovascular invasion. Also, they showed that a CD133 expression was significantly correlated with the MSI status. In survival analysis, the 5-year disease-free survival rate of patients with a low CD133 mRNA expression was significantly higher than that of those patients with high levels of CD133 mRNA expression. Therefore, they concluded that elevated CD133 mRNA levels represent more aggressive tumor biology and poorer survival in patients with colorectal cancer, correlating with a high level of MSI status. We agree that evaluation of mRNA expression in search for prognostic factor is useful and cost-effective in clinical practice instead of other modalities focusing on thousands of genes such as microarray. In fact, we have also reported that the detection of CEA mRNA in peritoneal washings is a useful, rapid genetic diagnosis for the prediction of peritoneal recurrence and survival in patients with colorectal cancer [2]. However, in Huh's study, there is a major issue that needs to be discussed before drawing their conclusion.As Huh et al. stated in their report, there was a significant association between CD133 mRNA expression and the depth of invasion, lymph node involvement, and lymphovascular invasion.And lymphovascular invasion and perineural invasion were the only significant factor for survival by multivariate analysis. Furthermore, as the authors stated, a CD133 mRNA expression was not an independent predictor of disease-free survival on the multivariate analysis. These results do not seem to give rationale to conclude that CD133 mRNA levels represent more aggressive tumor biology and poorer survival in patients with colorectal cancer. Another important issue is the MSI status of the patients. MSI is a well-known prognostic marker in patients with colon cancer. We have also previously shown that MSI patients show better prognosis than patients without MSI in high-risk stage II and stage III colon cancer patients [3]. However, Huh et al. did not include MSI status in either univariate or multivariate analysis of risk factors for survival. We believe they need to evaluate each factor at least on univariate analysis by including MSI status.Although Huh's study presented important findings, we believe these issues need to be addressed sufficiently in order to draw their conclusion.