Defining a Radiomic Response Phenotype: A Pilot Study using targeted therapy in NSCLC

Abstract

Medical imaging plays a fundamental role in oncology and drug development, by providing a non-invasive method to visualize tumor phenotype. Radiomics can quantify this phenotype comprehensively by applying image-characterization algorithms, and may provide important information beyond tumor size or burden. In this study, we investigated if radiomics can identify a gefitinib response-phenotype, studying high-resolution computed-tomography (CT) imaging of forty-seven patients with early-stage non-small cell lung cancer before and after three weeks of therapy. On the baseline-scan, radiomic-feature Laws-Energy was significantly predictive for EGFR-mutation status (AUC = 0.67, p = 0.03), while volume (AUC = 0.59, p = 0.27) and diameter (AUC = 0.56, p = 0.46) were not. Although no features were predictive on the post-treatment scan (p > 0.08), the change in features between the two scans was strongly predictive (significant feature AUC-range = 0.74–0.91). A technical validation revealed that the associated features were also highly stable for test-retest (mean ± std: ICC = 0.96 ± 0.06). This pilot study shows that radiomic data before treatment is able to predict mutation status and associated gefitinib response non-invasively, demonstrating the potential of radiomics-based phenotyping to improve the stratification and response assessment between tyrosine kinase inhibitors (TKIs) sensitive and resistant patient populations.

Data Science in Radiology: A Path Forward

“…[13][14][15], others the automated quantification of diseases (also known as CADx; refs. [16][17][18][19], and still others that link radiologic data with genomic data (also known as imaging-genomics or radiogenomics) to define genotype-phenotype interactions (6,(20)(21)(22)(23)(24). With promising results from these early studies and the increasing availability of imaging data (including large retrospective datasets with clinical endpoints), we expect radiomic investigations to continue to grow rapidly in number and complexity in the coming years.…”

Defining the biological and clinical basis of radiomics

“…It remains a difficulty to hypothesize why FLAIR imaging is less prognostic; however, previous studies have documented that tumor volume was not prognostic under FLAIR in recurrent glioblastoma and bevacizumab treatment. [41][42][43] Overall, a general tendency toward higher predictability of PFS was observed, which could be explained by a more clearly defined endpoint than OS, as patients who progressed were discontinued from study treatment and observed for OS. Differences between treatment arms in the BRAIN cohort were not expected and hence data have been pooled, as published previously.…”

“…Imaging criteria for disease assessment: Several imaging based criteria have been developed to assess characteristics, such as tumor stage or response to treatment, for example the Tumor Nodule Metastasis (TNM) system [39], the Response Evaluation Criteria in Solid Tumors (RECIST) [40], or the Response Assessment in Neuro-Oncology Criteria (RANO) [41,42]. While these criteria are widely implemented in hospitals treating patients with cancer, performance of these metrics is limited due to inherent oversimplifications.…”

“…As such imaging biomarkers do not require an invasive procedure as traditional biological biomarkers, these 192 biomarkers can be utilized over time to consistently monitor treatment response. To this end, in Chapters 5 and 6 [40,41]we aimed at defining radiomic response phenotypes in both brain and lung cancer treated with bevacizumab (Avastin TM ) and gefitinib (Iressa TM ), respectively.…”

“…To investigate whether radiomic approaches could identify patients who are sensitive to gefinitib using CT imaging pre (baseline) or post (follow-up) administration, we analyzed data from a clinical phase II trial in Chapter 6 [41]. Gefitinib is a specific inhibitor of EGFR signaling [58], implying that gefitinib will have high effectiveness for patients whose tumors contain a EGFR sensitizing mutation and no effectiveness otherwise.…”