| PART ? Radiomics Signature Extracted from Diffusion-weighted Magnetic Resonance Imaging Predicts Outcomes in Osteosarcoma Purpose and BackgroundsOsteosarcoma usually arises in the metaphysis of a long bone,most commonly around the knee.Involvement of the axial skeleton and craniofacial bones is primarily observed in adults.Patients with clinically detectable lung metastasis account for about 20%-25%of total osteosarcomas diagnosed.Despite significant advances in the detection and treatment of osteosarcoma over the past two decades,the 5-year survival of patients with a metastatic disease is around 20%.Osteosarcoma often requires multidisciplinary treatment including surgery,chemotherapy and radiotherapy.Although the survival of patients with operable osteosarcoma is improved by chemotherapy,however,tumor behavior can vary widely among patients and selection of appropriate therapies in any individual patient,especially in children and adolescents,remains a critical challenge.An important component of management involves the assessment of treatment response to multidisciplinary treatment in ameliorating recurrence risk.Therefore,the identification of novel biomarkers of a metastatic and prognostic phenotype is essential for osteosarcoma therapy.Although magnetic resonance imaging(MRI)is considered the best technique for the local staging of musculoskeletal neoplasms,it is relatively little used in the primary diagnosis of bone tumors and its capacity to evaluate the prognostic might be underestimated.Presently,prognosis of non-metastatic osteosarcoma is determined largely based on histological response and completeness of resection;magnetic resonance imaging serves an adjunct role in assessing treatment response,generally on the basis of imaging appearance on fluid sensitive and post-contrast imaging sequences with a focus on global tumor volume reduction.Diffusion-weighted imaging(DWI)MRI can capture changes at the cellular level thanks to differences in movement of water protons in the different tissue regions.Therefore,DWI-MRI can provide information regarding tumor cellularity as a surrogate indicator of treatment response on the basis of a quantitative value.For the past several years,as an emerging individualized precision medical technology,radiomics has applied advanced computational methodologies to transform the image data of the regions of interest into high dimensional feature data.Next,quantitative and high-throughput analysis of feature data is completed to probe tumor phenotype.Radiomics utilizes noninvasive imaging to provide more comprehensive information about the entire tumor and can be used in diagnosis,prognosis and prediction.Thus,in this study,we developed and validated multiparametric DWI-MRI based radiomics as a novel approach for providing individualized,pretreatment evaluation of overall survival in patients with osteosarcoma.In addition,we sought to reveal association between radiomics features and clinical data.Materials and MethodsPre-treatment DWI-MRI were collected from 123 patients(9-67 years of age)with histological-proven osteosarcoma that were treated with curative intent.The entire dataset was divided in two subsets:the training and validation cohorts containing 76 and 24%of the data respectively.Clinical data were extracted from our medical record.Two experienced radiotherapists evaluated DWI-MRIs for quality and segmented the tumor.A total of 103 radiomic features were calculated for each image.Least absolute shrinkage and selection operator(LASSO)regression was applied to select features.Association between the radiomics signature and OS was explored.Further validation of the radiomics signature as an independent biomarker was performed by using multivariate Cox regression.The Cox proportional-hazard regression model was also used to analyze the correlation between the prognostic factor and the survival for the clinical(C)model after the univariate analysis.Radiomics(R)model identified radiomics signature,which is the best predictor from the radiomic variable classes based on LASSO regression.Harrell's C-index was used to demonstrate the incremental value of the radiomics signature to the traditional clinical risk factors for the individualized prediction performance.ResultsThe radiomics signature was associated with the OS in the training cohort(P=0.0452;HR=5.35,95%confidence interval[CI]:3.22,8.87),and this finding was confirmed in the validation cohort(P<0.001;HR=5.09;95%CI:3.12,7.91).Cox proportional-hazard regression model shows that:Tumor size,ALP status before treatment and distance metastasis were proven as the dependent clinical prognostic factors of osteosarcoma's overall survival time.The radiomics signature was significantly associated with OS,independent of clinical risk factors(radiomics signature:HR:5.11,95%CI:2.85,9.18,P<0.001).Incorporating the radiomics signature into the coalition(C+R)model resulted in better performance(P<.001)for the estimation of OS(C-index:0.811;95%CI:0.74,0.89)than with the clinical(C)model(C-index:0.774;95%CI:0.70,0.85),or the single radiomics(R)model(C-index:0.71;95%CI:0.65,0.77).ConclusionThis study shows that the radiomics signature extracted from pre-treatment DWI-MRIs improve prediction of OS over clinical features alone.Combination of the radiomics signature and the traditional clinical risk factors performed better for individualized OS estimation in patients with osteosarcoma,which might enable a step forward precise medicine.This technique may help better select patients most likely to benefit from intensified multimodality diagnosis and therapies.Future studies will focus on multi-center validation of an optimized model.PART II Prediction and Evaluation of Neoadjuvant Chemotherapy using Radiomics Signature Extracted from Diffusion-weighted Magnetic Resonance Imaging in Patients with Osteosarcoma Purpose and BackgroundsThe diagnosis,treatment and prognosis of osteosarcoma have always been a hot point concerned by clinicians.Clinical studies have found that tumor response to chemotherapy was closely related to the prognosis of patients,and timely monitoring the tumor response to chemotherapy is of great significance for the selection of reasonable treatment and the evaluation of prognosis.Diffusion-weighted imaging(DWI)can capture changes at the cellular level thanks to differences in movement of water protons in the different tissue regions.Therefore,DWI-MRI can provide information regarding tumor cellularity as a surrogate indicator of treatment response on the basis of a quantitative value.For the past several years,radiomics has applied advanced computational methodologies to transform the image data of the regions of interest into high dimensional feature data.Next,quantitative and high-throughput analysis of feature data is completed to probe tumor phenotype.The purpose of this prospective study is to investigate the feasibility of applying the radiomics signature extracted from diffusion-weighted magnetic resonance imaging(DWI-MRI)in predicting and evaluating the response to neoadjuvant chemotherapy in patients with osteosarcoma.Materials and MethodsThirty-three patients with osteosarcoma who underwent both pre-and post-chemotherapy DWI-MRI were enrolled in the study.In each patient,DWI-MRI feature data(radiomics signature)were collected before and after chemotherapy,respectively.A radiomics score(Rad-score)was computed for each patient through a linear combination of selected features weighted by their respective coefficients.Based on the tumor necrosis rate histologically confirmed by tumor resection after chemotherapy,the diagnostic performance of DWI-MRI radiomics signature was assessed in predicting tumor necrosis response with the pre-and post-chemotherapy imaging,as well as evaluating survival time performance.ResultsOn pre-chemotherapy DWI-MRI,the Rad-score in patients with good response was significantly lower.Tumor rad-score was negatively correlated with the tumor necrosis rate(r=-0.777,P<0.001).ROC curve analysis based on tumor necrosis rate showed that when Rad-score<-2.794 was taken as the prediction threshold for good chemotherapy response,a sensitivity of 85.7%,a specificity of 80.0%would be yielded.On post-chemotherapy imaging,pearson correlation analysis showed that the Rad-score was also correlated negatively with the tumor necrosis rate(r=-0.736,P<0.001).When Rad-score ?-4.15 was taken as the prediction threshold for good chemotherapy response,the predictive sensitivity,specificity were 100%,80.0%,respectively.Kaplan-Meier survival curve analysis showed that the patients with tumor necrosis rate>90%had significantly higher disease progression-free survival rate;meanwhile,patients with Rad-score less than or equal to-2.794 before chemotherapy or less than or equal to-4.15 after chemotherapy also had a better prognosis.ConclusionDWI-MRI radiomics signature is a useful approach for the evaluation of neoadjuvant chemotherapy in patients with osteosarcoma,and its mechanism could be simultaneously used in predicting and evaluating tumor response to chemotherapy. |
PDF Full Text Request