CT-based Radiomics To Predict Response To Immunotherapy In NSCLC

Part1 Changes in CT radiomics features before and after treatment predict response to immunotherapy in advanced non-small-cell lung cancerPurpose: Immunotherapy has substantially changed therapeutic strategies for non-smallcell lung cancer(NSCLC);however,clinical responses and survival benefits are not universal among all treated patients.The study aimed to identify imaging biomarkers to assess predictive capacity of radiomics nomogram regarding treatment response status(responder/non-responder)in patients with advanced NSCLC undergoing antiPD1 immunotherapy.Materials and Methods: 197 eligible patients(median age,63 years;range,29-84 years;165 men [median age,64 years;range,29-84 years] and 32 women [median age,63 years;range,37-79 years])with histologically confirmed NSCLC between August 2016 and February 2019 were retrospectively enrolled from 9 hospitals.We carried out a radiomics characterization from target lesions(TL)approach and largest target lesion(LL)approach on baseline and first follow-up(TP1)CT imaging data.delta-radiomics feature was calculated as the relative net change in radiomics feature between baseline and TP1.Results: TP0-radiomics signature at baseline did not show significant predictive value regarding response status for LL approach(AUC = 0.59,P= 0.10),nor in terms of TL approach(AUC = 0.56,P = 0.27).In the sub-cohort of 161 patients,a combined deltaradiomics nomogram for target lesions that could reflect tumor burden had satisfactory performance in distinguishing responders from non-responders with AUCs of 0.83(95% CI: 0.75-0.91)and 0.81(95% CI: 0.68-0.95)in the training and test sets respectively,which was comparable with that from LL approach(P = 0.92,P= 0.97).Among a subset of those patients with available pretreatment PD-L1 expression status(n = 66),models that incorporating delta-radiomics features showed superior predictive accuracy than that of PD-L1 expression status alone(P < 0.001).Conclusion: Early response assessment using combined delta-radiomics nomograms have potential advantages to identify patients that were more likely to benefit from immunotherapy,and help oncologists modify treatments tailored individually to each patient under therapy.Part2 A combined-radiomics approach with plain and contrast enhanced CT images to predict response to immunotherapy in NSCLC: a retrospective multicenter studyPurpose: We aimed to identify a combined-radiomics model based on both non contrast enhanced CT(NCE-CT)and contrast enhanced CT images(CE-CT)and assess its predictive capacity regarding treatment response in patients with NSCLC undergoing immunotherapy.Materials and Methods: 131 eligible patients with histologically confirmed NSCLC were retrospectively enrolled from 7 institutions.We carried out a radiomics characterization from largest lesion(LL)approach and target lesions(TL)approach on pretreatment NCE-CT and CE-CT images respectively.Meanwhile,a combined-radiomics nomogram incorporating NCE/CE-CT radscores with clinical factor for LL approach was constructed,and the receiver operating characteristic curve and area under the curve (AUC)were used to evaluate the predictive performance of the nomogram in the training and validation queues.A De Long test was employed to compare the differences between different models.Results: For LL approach,radiomics nomograms incorporating NCE-CT or CE-CT radiomics signature with clinical factor of distant metastasis all had satisfactory performance in distinguishing responders from non-responders with AUCs of 0.84(95% CI: 0.75-0.92)and 0.77(95% CI: 0.67-0.87)in the training and 0.78(95% CI: 0.64-0.92)and 0.73(95% CI: 0.57-0.88)in testing sets,respectively.For TL approach,radiomics models based on NCE-CT or CE-CT images all did not show higher predictive value.A combined-radiomics nomogram based on NCE-CT and CE-CT images for LL had more satisfactory performance in distinguishing responders from non-responders with AUCs of 0.85(95% CI: 0.77-0.92)and 0.81(95% CI: 0.67-0.94)in the training and testing sets respectively.Conclusion: Early response assessment using combined-radiomics nomogram based on NCE/CE-CT images from LL approach have potential advantages to identify patients that were more likely to benefit from immunotherapy,and help oncologists modify treatments tailored individually to each patient under therapy.