Non-invasive decision support for NSCLC treatment using PET/CT radiomics

Abstract

Two major treatment strategies employed in non-small cell lung cancer, NSCLC, are tyrosine kinase inhibitors, TKIs, and immune checkpoint inhibitors, ICIs. The choice of strategy is based on heterogeneous biomarkers that can dynamically change during therapy. Thus, there is a compelling need to identify comprehensive biomarkers that can be used longitudinally to help guide therapy choice. Herein, we report a 18F-FDG-PET/CT-based deep learning model, which demonstrates high accuracy in EGFR mutation status prediction across patient cohorts from different institutions. A deep learning score (EGFR-DLS) was significantly and positively associated with longer progression free survival (PFS) in patients treated with EGFR-TKIs, while EGFR-DLS is significantly and negatively associated with higher durable clinical benefit, reduced hyperprogression, and longer PFS among patients treated with ICIs. Thus, the EGFR-DLS provides a non-invasive method for precise quantification of EGFR mutation status in NSCLC patients, which is promising to identify NSCLC patients sensitive to EGFR-TKI or ICI-treatments.

Structural and functional radiomics for lung cancer

“…Similar performance has also been reported in another retrospective study [106]. In addition, for patients with EGFR mutation, a deep radiomic score was a non-invasive tool to identify NSCLC patients susceptible to tyrosine kinase or immune checkpoint inhibitors [107].…”

Combining radiomic phenotypes of non-small cell lung cancer with liquid biopsy data may improve prediction of response to EGFR inhibitors

“…Radiomics analysis enables the detection of tumor imaging features and patterns of intra-tumor heterogeneity not appreciable by the human eye, increasing the wealth of information from radiological imaging. Studies specifically suggest that radiomic analysis may provide novel prognostic markers related to gene-expression patterns and responder signatures for NSCLC patients receiving targeted therapy 18 – 31 . Most studies to date have focused on using radiomic analysis on computed tomography (CT) and/or positron emission tomography (PET)/CT data to predict EGFR mutation status, using statistical modeling or machine learning approaches for reducing the high dimensionality of radiomic features 19 , 21 – 29 , 32 .…”

“…Most studies to date have focused on using radiomic analysis on computed tomography (CT) and/or positron emission tomography (PET)/CT data to predict EGFR mutation status, using statistical modeling or machine learning approaches for reducing the high dimensionality of radiomic features 19 , 21 – 29 , 32 . More recently deep learning approaches have also been used to predict outcomes after TKI therapy for NSCLC 31 , 33 . While this field is rapidly developing, a question still remains as to which extent radiomic analysis can complement established prognostic markers for TKIs, as most studies have either evaluated radiomic features in the absence of established prognostic biomarkers or have only examined surrogate endpoints, such as EGFR mutation status, rather than actual patient outcomes.…”

Imaging-based patient inclusion model improves clinical trial performance

“…Patient selection into the trial based on Dox group survival was executed according to the following method: firstly (1) radiomic and clinical features associated in training cohort with survival in Dox but not Dox+Evo treatment group were included in a multivariable Cox regression model (2), trained on Dox treated patients. The risk score assigned by the model to each training set patient was then used as a biomarker for inclusion into analysis, iteratively calculating the p value and hazard ratio for survival comparison between treatment arms depending on minimum risk score threshold (3). If available, threshold corresponding to significant (p value<0.05) treatment benefit of Dox+Evo at highest percentage of patients included was chosen (4), and subsequently tested in the test cohort (5), with risk scores assigned by the multivariable Cox model developed in step (2).…”