Three-Gene Expression Signature for Predicting Lymph Node Metastasis in Cervical Cancer

A retrospective biomarker study developed and validated a three-gene signature (LOC494141, GLOD5, GML) to predict lymph node metastasis in cervical squamous cell carcinoma. The signature was discovered using transcriptomic data from 116 TCGA samples and validated on a prospective cohort of 202 patients from Fudan University Shanghai Cancer Center. The dataset was last updated on 2026-05-13.

Use Cases

• Training a binary classifier for lymph node metastasis prediction based on the three-gene signature.
• Evaluating the incremental value of a molecular signature when combined with clinical features like tumor size.
• Benchmarking new predictive models against the reported AUC of 0.745.
• Conducting a meta-analysis on gene expression biomarkers for cervical cancer metastasis.

Strengths

• The signature was validated on an independent clinical cohort of 202 patients (101 LNM-positive, 101 LNM-negative).
• Performance metrics are reported, including an AUC of 0.745, sensitivity of 62.38%, and specificity of 64.36%.
• A combined model integrating the signature with clinical features achieved an AUC of 0.789.
• The dataset is licensed under CC-BY-4.0, allowing for open reuse.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small at 26.2 KB, indicating limited scope.

Provenance

Source

The Cancer Genome Atlas (TCGA) and Fudan University Shanghai Cancer Center.

Collection Method

Retrospective biomarker discovery using TCGA transcriptomic data and validation via qRT-PCR on a clinical cohort.

Freshness

Last updated 2026-05-13 05:56:58; freshness should be verified.