Radiomic phenotypes of oligometastatic non-small cell lung cancer

Le résumé fourni par la source

Background: Identifying biomarkers that optimize patient selection for ablative therapy in metastatic non-small cell lung cancer (NSCLC) remains an unmet need. In this study, we identified radiomic features associated with sustained oligometastatic and indolent NSCLC. Methods: We curated a retrospective cohort of 167 patients (132 training/validation, 35 test) with treatment-naïve, metastatic NSCLC and pre-treatment positron emission tomography/computed tomography (PET/CT). We extracted radiomic features from the primary lesion and used Pearson's correlation to identify those associated with extent of metastatic disease. For a subset (n=83), we also identified features associated with overall survival (OS) using Cox proportional hazards models. Results: We found that the inverse flatness was lower (r=-0.23; q=0.04) and surface area-to-volume ratio (SVR) was higher (r=0.24; q=0.04) in primary lesions for cases with more metastases. We further found that higher gray level variance on CT portends shorter OS [hazard ratio (HR) =1.84, 95% confidence interval (CI): 1.32-2.56, q=0.004; test c=0.74, 95% CI: 0.52-0.90]. Risk groups derived from these scores exhibited median survival times in the training set of 12.6 and >50 months (P=0.001) and in the test set of 13.8 and 32.7 months (P=0.06). This feature was determined primarily by inclusion of low-intensity voxels. Conclusions: This study identifies radiomic features of indolent NSCLC, demonstrating that flatter, higher surface area primary disease is associated with a greater number of metastases, and primary lesions with lower density representing lung parenchyma, airspace, or lower-density tumor portend shorter OS. These findings lay the groundwork toward identifying patients suitable for ablative radiotherapy despite the presence of metastatic disease.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.