Journal of Clinical & Experimental Pathology
Open Access

Abstract

Background: As one of the most common malignancy, colorectal cancer (CRC) poses a serious threat to human health. Both CRC incidence and mortality rates continue to increase worldwide. The establishment of recurrence prediction model of CRC has a very important practical significance. Methods: 37 gene mRNA expression levels in 473 CRC tissues were detected using nCounter analysis system. Differentially expressed genes were screened out between the recurrent group and the non-recurrent group. The recurrence prediction model of CRC was established by using the binary Logistic regression analysis. Results: A 37 gene prediction model was generated by using the binary logistic regression analysis. The univariate analysis (Log- Rank) indicated that DFS was significantly worse in the recurrent group than in the non-recurrent group both in training group and testing group (P<0.05). The univariate analysis also indicated that DEPDC1 mRNA expression were significantly higher in nonrecurrent group than those in recurrent group (P<0.05). DEPDC1 and TNM stage had significant correlation with DFS (P<0.05). The Cox proportional hazards regression indicated that DEPDC1 mRNA expression was an independent factor for CRC patients. Conclusion: In this part, we have validated the expression of differentially expressed genes and established the recurrence prediction model of CRC using nCounter analysis system which is a high-throughput platform, further study will be done to verify and optimize the prediction model in the future.

Biography

Email: 12111230001@fudan.edu.cn