Novel clinical and radiomic predictors of rapid disease progression phenotypes among lung cancer patients treated with immunotherapy: An early report

Abstract

Objectives-Immune-checkpoint blockades have exhibited durable responses and improved long-term survival in a subset of advanced non-small-cell lung cancer (NSCLC) patients. However, highly predictive markers of positive and negative responses to immunotherapy are a significant unmet clinical need. The objective of this study was to identify clinical and computational image-based predictors of rapid disease progression phenotypes in NSCLC patients treated with immune-checkpoint blockades. Materials and Methods-Using time-to-progression (TTP) and/or tumor growth rates, rapid disease progression phenotypes were developed including hyperprogressive disease. The pretreatment baseline predictors that were used to identify these phenotypes included patient demographics, clinical data, driver mutations, hematology data, and computational image-based features (radiomics) that were extracted from pre-treatment computed tomography scans. Synthetic Minority Oversampling Technique (SMOTE) was used to subsample minority groups to *

Role of medical imaging for immune checkpoint blockade therapy: From response assessment to prognosis prediction

“…Several studies have demonstrated that CT‐based radiomics features might effectively predict responses and prognosis in patients treated with immunotherapies . Sun et al analyzed the data from four independent multicenter cohorts of 491 patients with advanced solid tumors who received anti‐PD‐1/PD‐L1 monotherapy and established a CT‐based radiomics signature of tumor‐infiltrating CD8 cells .…”

“…They observed that a higher radiomic score at baseline was significantly associated with objective response, controlled disease, and favorable OS. More recently, Tunali and colleagues established clinical‐radiomic models to predict two rapid disease progression phenotypes (time to progression [TTP] <2 months and HPD) in a cohort of 228 NSCLC patients . In this study, radiomic features were extracted from both the intratumoral and peritumoral regions based on baseline contrast‐enhanced CT images within 1 month prior to the initiation of ICB therapy.…”

“…In terms of TTP <2 months, the prediction model incorporated four clinical covariates (hepatic and bone metastasis, previous lines of systemic therapies, and neutrophils to lymphocytes ratio) and four radiomic features (radial gradient border SD‐2D, border quartile coefficient of dispersion, border 3D Laws E5E5L5, and 3D Laws E5L5E5) and revealed an AUROC of 0.804, specificity of 83.4%, sensitivity of 63.4%, and accuracy of 73.4%. Notably, three of the four radiomic features were extracted from the border regions of tumors, where appeared to be more informative than intratumoral regions in reflection of the heterogeneity of tumor microenvironment . With regard to HPD, the prediction model was based on Royal Marsden Hospital prognostic score and one radiomic feature derived from the tumor border (border neighborhood gray‐tone difference matrix strength) with an AUROC of 0.865, specificity of 92.9%, sensitivity of 74.0%, and accuracy of 82.3%.…”

“…66 Several studies have demonstrated that CT-based radiomics features might effectively predict responses and prognosis in patients treated with immunotherapies. 15,67,68 Sun et al analyzed the data from four independent multicenter cohorts of 491 patients with advanced solid tumors who received anti-PD-1/PD-L1 monotherapy and established a CTbased radiomics signature of tumor-infiltrating CD8 cells. 15 The radiomics signature was comprised of eight variables and demonstrated good prediction performance for the gene expression signature of CD8 cells (area under the receiveroperator characteristic [AUROC]: 0.67) and tumor immune phenotypes (AUROC: 0.76).…”

“…More recently, Tunali and colleagues established clinical-radiomic models to predict two rapid disease progression phenotypes (time to progression [TTP] <2 months and HPD) in a cohort of 228 NSCLC patients. 68 In this study, radiomic features were extracted from both the intratumoral and peritumoral regions based on baseline contrast-enhanced CT images within 1 month prior to the initiation of ICB therapy. In terms of TTP <2 months, the prediction model incorporated four clinical covariates (hepatic and bone metastasis, previous lines of systemic therapies, and neutrophils to lymphocytes ratio) and four radiomic features (radial gradient border SD-2D, border quartile coefficient of dispersion, border 3D Laws E5E5L5, and 3D Laws E5L5E5) and revealed an AUROC of 0.804, specificity of 83.4%, sensitivity of 63.4%, and accuracy of 73.4%.…”