Multidimensional Model To Predict The Efficacy And Prognosis Of Hypopharyngeal Carcinoma Based On Clinical Data,radiomics And Biomarker

Objiective:1.To explore the factors of induction chemotherapy effectiveness in hypopharyngeal carcinoma base on clinical data.2.We aimed to use a new imaging data extraction method,radiomic to predict the efficacy of induction chemotherapy in patients with locally advanced hypopharyngeal cancer before treatment.3.Lymphocyte telomere length(LTL)plays critical roles in maintaining genomic stability and integrity.Reduced TL leads to greater cell sensitivity to radiotherapy,a central component of SCCOP treatment.LTL thus may influence the etiology,diagnosis,and prognosis of SCCOP patient survival after radiotherapy,independent of HPV status.Materials and Methods:1.A total of 81 patients with confirmed squamous cell carcinoma of hypopharynx by pathological tests were analyzed,who recieved 2 cycles of induction chemotherapy by TPF.MRI scans were performed before and after induction chemotherapy.Gender,age,primary site,tumor stage,histological differentiation,size of positive neck node,tumor invasion site and distant metastasis were compared between effective and ineffective group of induction chemotherapy patients.We used univariate and multivariable logistic regression to find the risk factors for induction chemotherapy effectiveness,and calculated the Odds Ratio(OR).Area under the receiver operator characteristic curves(AUC),along with sensibility,specificity,and Youden’^index were calculated from the model.We used leave-one-out cross-validation to assess the predictive error of this model.2.One hundred and fifty three patients who were newly diagnosed as hypopharyngeal cancer and received induction chemotherapy were included(primary cohort:n=77;validation cohort:n=76).Radiomic features were extracted in each volume of interest of non-contrast CT(NCCT)and contrast-enhanced CT(CECT).Of these,the minimum redundancy maximum relevance(MRMR)algorithm was used to weight and se.,ct imaging features.Support vector machine support vector machine(SVM),random forest(RP)and artificial neural network(ANN)classification algorithms were compared through internal validation(100 times 10-fold cross-validation)in the primary cohort,and then the model with the best performance was chosen.Clinical characteristics were analyzed by multivariate analysis and the clinical model was selected by minimum Akaike’s information criterion(AIC).The performance of an overall nomogram integrated the radiomic signatures and remarkable clinical data was evaluated with receiver operating characteristics curves(ROC).3.The LTL and tumor HPV16 status were measured in 564 incident SCCOP patients with radiotherapy or chemoradiotherapy.Both univariate and multivariable Cox analyses were applied to estimate the associations between LTL and prognosis.Results:1.Size of positive neck node(OR=1.661,95%CI:1.069-2.583)and invasion of esophagus and thyroid cartilage simultaneously(OR = 15.803,95%CI:1.439~173.509)were the factors of induction chemotherapy effectiveness,while primary site,tumor stage,histological differentiation,distant metastasis were irrelevant factors(p = 0.059～0.364).AUC of this model was 0.786,sensitivity was 0.816,specificity was 0.698,and Youden’s index was 0.514.Predictive error of this model was 0.206.2.The radiomic signatures were built by SVM from NCCT and CECT.SVM-NCCT and SVM-CECT were significant in primary cohort and validation cohort for predicting treatment response of induction chemotherapy(AUC:0.892 0.838).Meanwhile,age and short axis of maximum enlarged lymph node(LN size)were selected among the clinical data.Radiomic signatures model and clinical model were integrated to develop a nomogram.At last,the overall nomogram has shown the best performance in primary and validation cohort with good calibration(AUC:0.913 0.864).Decision curve analysis had also proved the clinical value of this comprehensive nomogram.3.Approximately 85%of SCCOP patients were tumor HPV16-positive.The patients with shorter LTL had significantly better overall(OS),disease-specific(DSS),and disease-free survival(DFS)compared with those with the longer LTL(all log-rank:P<0.001).Moreover,compared to those with longer LTL,we found that patients with shorter LTL had statistically significantly better survivals,after adjusting other important prognostic confounders,independent of HPV status.Finally,the more pronounced significant effects of LTL on survival were found among HPV16-positive SCCOP patients only after stratified analyses by tumor HPV status.Conclusion1.For those hypoparyngeal carcinoma patients which neck node size exceed 2 cm and tumor invaded esophagus and thyroid cartilage simultaneously,selection of surgery may be better than induction chemotherapy.2.Pretreatment CT-based radiomic nomogram provided an innovative method to predict the response of induction chemotherapy in patients with locally advanced hypopharyngeal cancer,which help select non-response patients for surgery or concurrent systemic chemoradiotherapy before giving induction chemotherapy.3.Our findings indicate that the LTL plays a significant role in the survival of SCCOP,especially in HPV 16-positive SCCOP patients treated with definitive radiotherapy,supporting that the pretreatment LTL may be an independent prognostic biomarker for HPV16-positive SCCOP.