Abstract
Background
Hypoxia, which affects the development, metastasis and prognosis of cancer, represents a key feature of cancer. This study describe a hypoxia risk factor model, with predicting the prognosis of cervical cancer.
Methods
Based on hypoxia pathway related genes, we divided cervical cancer samples into high and low expression groups. A cox analysis was then performed. Genes from these cervical cancer samples showing a significant impact on OS were selected for cluster analysis to obtain two subtypes. The TPM dataset of TCGA was divided into training and validation sets. For the training set, a lasso analysis was conducted as based on cox analysis of meaningful genes and a risk factor model was constructed. The constructed model was verified in internal and external data sets. Finally, RT-PCR, immunohistochemistry were used to detect the expression of relative genes or proteins and functional assays were used to evaluate the biological function of signature genes.
Results
Two molecular subtypes were obtained, Cluster2 vs Cluster1.These subtypes were obtained by clustering with a total of 149 DEGs (Differential expressed genes) being in line with this standard, of which 27 were up-regulated and 122 were down-regulated. The five genes with lambda = 0.0571 were selected to construct the model, the RiskScore = AK4*0.042 + HK2*0.021 + P4HA1*0.22 + TGFBI*0.1 + VEGFA*0.077. Further, in order to verify the signature, we used TCGA-test and GSE44001 chip datasets to test, and finally got a good risk prediction effect in those datasets. Moreover, the result of RT-PCR and immunohistochemistry demonstrated that AK4, HK2, P4HA1, TGFBI and VEGFA were all highly expressed in these cervical cancer tissue samples. The functional study shown that expression of AK4, HK2, P4HA1, TGFBI and VEGFA can regulate the proliferation, migration, and invasion ability of cervical cancer cells in vitro.
Conclusions
In summary, we developed a 5-gene signature prognostic hierarchical system based on the hypoxic pathway of cervical cancer, which is independent of clinical characteristics. And also conducted experimental verifications on these signature gene. Therefore, we propose that use of this classifier as a molecular diagnostic test can provide an effective means for evaluating the prognostic risk of cervical cancer patients, and provide potential targets for the treatment of cervical cancer patients.