Identification of Stage IIIC/IV EGFR-Mutated Non-Small Cell Lung Cancer Populations Sensitive to Targeted Therapy Based on a PET/CT Radiomics Risk Model

摘要: Background: To address the clinical challenges that progression risk of patients with stage IIIC-IV EGFR-mutated Non-small cell lung cancer (NSCLC) is difficult to accurately quantify and stratify after first-line EGFR-TKI targeted drug therapy, we aimed construct an individualized PET/CT prognostic biomarker identify treatment-sensitive population. Methods: 250 with stage NSCLC underwent therapy were included from two institutions (140 in training cohort; 60 internal validation cohort, 50 external cohort). 64 the chemotherapy group. 1037 3D radiomics features extracted phenotypic characteristics tumor region PET CT images. According median value signature score (Rad-score), divided into low- high-risk groups. The progression-free survival (PFS) behaviors subgroups compared the chemotherapy group by Kaplan–Meier analysis.  Findings: Higher Rad-scores significantly associated worse PFS ( p  < 0.0001),  = 0.0153), 0.0006) cohorts. Rad-score can effectively a high rapid progression, for these patients.  Interpretation: PET/CT-derived realize precise quantitative stratification populations sensitive therapy. Funding: National Natural Science Foundation China; Guangdong Basic Applied Research Foundation; Technology Planning Project Province. Declaration Interests: authors declare they have no competing interests. Ethics Approval Statement: Ethics Committee Provincial People’s Hospital Academy Medical Sciences approved this retrospective study waived requirement informed consent patients. approval ID was GDREC2019154H(R1).