增强CT、PET/CT及放射组学特征在非鳞非小细胞肺癌中的应用OA

Objective:To evaluate the predictive value of contrast-enhanced CT,PET/CT metabolic parameters,and radiomics features for treatment response and survival in patients with non-squamous non-small cell lung cancer(NSCLC)receiving tislelizumab plus bevacizumab.Methods:A total of 120 patients with non squamous NSCLC were retrospectively included.Baseline and on-treatment contrast-enhanced chest CT and PET/CT were performed.CT volumetric indicators,PET metabolic parameters,and radiomics features were extracted.Survival analyses,multivariable Cox regression,and radiomics-based machine learning models were used to assess predictive performance.Results:Objective response rate(ORR)was 43.3％and disease control rate(DCR)was 81.7％;median PFS and OS were 13.54 and 19.36 months,respectively.Base-line CT_vol,MTV,and TLG were significantly associated with treatment outcomes,and≥30％decline in PET metabolic parameters at 6～9 weeks predicted prolonged PFS.A radiomics-clinical nomogram achieved an AUC of 0.784 and outper-formed conventional imaging metrics.Treatment-related adverse events were manageable.Conclusion:PET metabolic parameters,CT volumetrics,and radiomics features provide significant predictive value for treatment response and survival in non-squamous NSCLC,supporting individualized treatment decision-making for tislelizumab plus bevacizumab therapy.