Preliminary Investigation Into Construction Of Radiosensitivity Model To Predict Benefit From Adjuvant Radiotherapy In Cervical Cancer Based On Intratumoral Gene MRNA Expression

Objective Growing evidence has shown that intrinsic radiosensitivity-related genes expression levels in tumor tissue exhibits significant impact on radiosensitivity in rectal carcinoma,breast cancer and head and neck squamous cell carcinoma.It could be feasible that appropriate radiation dose for individuals were prescribed according to intrinsic expression levels of radiosensitivity-related genes in tumor tissue.However,there was no research focused on adjuvant radiotherapy for early stage cervical cancer with regard to radiosensitivity.This study was undertaken to investigate the feasibility of radiosensitivty model developed based on mRNA expression in tumor tissues and its capacity to predict benefit from adjuvant radiotherapy in cervical cancer.Method Baseline clinicopathological characteristics?TCGA.CESC.sampleMap/CESC_clinicalMatrix?,mRNA expression matrix?TCGA.CESC.sampleMap/HiSeqV2?and microRNA expression matrix?TCGA.CESC.sampleMap/miRNA_HiSeq_gene?were obtained from TCGA datasets?TCGA Cervical Cancer,CESC?.The information of age at diagnosis,Histological TNM stage,clinical stag e,differentiation grade,histological subtype,formation of vascular tumor thrombus,purpose of primary surgery,accomplishment of external radiotherapy,dose delivered,recurrence free survival,overall survival and censored status was extracted.Patients underwent radical hysterectomy alone or radical hysterectomy plus adjuvant radiotherapy were selected.The baseline clinicopathological characteristics and gene expression were merged with sample identifier provided by TCGA.Candidate radiosensitive genes were identified through multivariate Cox regression with recurrence free survival?RFS?as end-point events.The sum of products of regression coefficients and expression values of candidate genes were calculated as radiosensivity index?RI?.Time-dependent ROC analysis was used to compare predictive accuracy of RI and clinical prognostic factors for RFS.Kaplan-Meier plot coupled with log rank test was used to evaluate difference in RFS between subgroups with different radiosensitivity.Results A total number of 155 patients underwent radical hysterectomy were retrieved from the CESC phenotype dataset among which 18 patients were completely treated with planned adjuvant radiotherapy.The 2 year RFS in patients underwent radical hysterectomy alone and those treated with the radical hysterectomy and adjuvant radiotherapy was 90.17%and 72.13 respectively(P_{log rank}=0.063).Expression values of 17157 genes and 480 microRNAs after probes filtration were included into analysis.Twenty-three genes were identified consistent with statistical criterion.The annotated function of 11 genes among the 23 candidate genes were involved in mitochondrial autophagy,cell cycle,sphingomyelin synthesis,genomic methylation and hypoxia-inducible factor hydroxylation.Hierarchical cluster analysis revealed 7 subgroups existed across these 23 genes.The genes MAT2A?PIGY?SUSD1?ZNF341?TMEM85?HIF1AN and CLEC18A were finally selected to calculate RI.The median value and range of RI was 0.7258?range:-5.501-9.046?.In subset consist of patients underwent radical hysterectomy alone,adjusted for intravascular tumor thrombus,RI was independent prognostic factor?HR=1.456,95%CI:1.192-1.780,P<0.001?.Whereas,RI was remained significantly associated with RFS after adjustment for the number of positive nodes in subset comprised of patients underwent radical hysterectomy plus adjuvant radiotherapy?HR=0.573,95%CI:0.328-0.998,P=0.049?.Time-dependent ROC curve analysis yielded that the areas under curves?AUC?for predicting the 3-year and 5-year RFS of N-stage and intravascular tumor thrombus were 0.794 and 0.692 and the RIs of 7 genes were 0.848and 0.839 in the simple radical surgery subgroup,respectively.The combinations of RI and N-stage and intravascular tumor thrombus were 0.849 and 0.794,respectively.In the post-operative adjuvant radiotherapy subgroup,the AUC of N-stage and intravascular tumor thrombus for 3 and 5 years DFS prediction were 0.786 and 0.750,respectively.The RIs were 0.947 and 1.000,the combinations of RI,N staging and intravascular tumor thrombus were 0.849 and 0.932,respectively.RI exhibited more accuracy in prediction for 3or 5 years RFS compared with clinical prognostic factors N stage and intravascular tumor thrombus.Conclusion RI has potential value to select cervical cancer patients benefiting from adjuvant radiotherapy.This model should be validated using actual tumor samples in the future.