CT-based Immune Radiomic Signature for Prognosis and Drug Response Prediction in NSCLC

A radiomic signature dataset for non-small cell lung cancer (NSCLC) developed from CT scans across multiple cohorts, including TCIA, TCGA, and an in-house hospital dataset. The dataset contains a CT-RadScore composed of 12 radiomic features, validated for prognostic performance and prediction of immunotherapy and anticancer drug response. It was created by Jianying Ma and last updated on April 29, 2026.

Use Cases

• Predict overall survival in NSCLC patients based on the CT-RadScore.
• Predict response to immune checkpoint inhibitor therapy based on the radiomic signature.
• Predict sensitivity to anticancer drugs like paclitaxel, gefitinib, and carboplatin based on the CT-RadScore.
• Stratify patients into immune-hot and immune-cold subtypes for treatment planning.
• Validate radiomic models against immune deconvolution algorithms and gene expression modules.

Strengths

• The CT-RadScore demonstrated validated prognostic performance with C-indexes of 0.791, 0.729, and 0.844 across three independent cohorts.
• The signature is composed of 12 non-redundant radiomic features derived from CT scans.
• High CT-RadScore was associated with significantly worse overall survival (all log-rank P<0.0017).
• Low CT-RadScore predicted improved immunotherapy response with response rates of 54.7–75% versus 30–34.4% in high-score groups.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect geographic or temporal bias inherent to the specific hospital and public cohorts used.

Provenance

Source

Multiple cohorts including TCIA NSCLC Radiogenomics, TCGA LUAD+LUSC, and an in-house cohort from Nanyang Central Hospital.

Collection Method

CT radiomic features were extracted using PyRadiomics, and machine-learning survival models (random survival forest, LASSO–Cox, Elastic Net–Cox) were compared.

Freshness

Last updated 2026-04-29 05:57:11; freshness should be verified.