Table 1_Serum metabolomic signatures predict clinical outcomes in advanced non-small cell

216 longitudinal serum samples from 36 patients with advanced non-small cell lung cancer receiving first-line pembrolizumab plus chemotherapy were analyzed via nuclear magnetic resonance spectroscopy. The data, authored by Peter May and last updated in May 2026, links metabolite levels to clinical outcomes like disease progression within 60 days and overall survival.

Use Cases

• Predicting disease progression within 60 days based on branched-chain amino acid levels.
• Modeling overall survival using lipid and inflammatory marker signatures.
• Training machine learning models like Random Forest for survival status prediction.
• Investigating host-tumor metabolic interactions through longitudinal metabolite profiles.

Strengths

• Includes 216 longitudinal serum samples, enabling analysis of dynamic changes.
• Random Forest model achieved high predictive accuracy (AUC = 0.93) for survival status.
• Specific key metabolites are identified, including valine, isoleucine, sphingomyelin, and apolipoproteins.

Limitations

• Row count and column-level documentation are unknown, limiting suitability assessment.
• The dataset is relatively small, with samples from 36 patients.
• Data may reflect bias inherent to the specific clinical cohort studied.

Provenance

Source

figshare

Collection Method

Nuclear magnetic resonance (NMR) spectroscopy-based metabolomic analysis.

Freshness

Last updated 2026-05-18 05:41:05