A large-scale, multi-centre validation study of an AI-empowered blood-based test for multi-cancer early detection.

Journal: NPJ precision oncology

This study evaluated OncoSeek, an AI-powered blood-based multi-cancer early detection (MCED) test, across 15,122 participants—including 3,029 cancer patients—using data from seven centers in three countries, four platforms, and two sample types.

OncoSeek demonstrated strong performance with:

  • AUC: 0.829
  • Sensitivity: 58.4%
  • Specificity: 92.0%
  • Accuracy in predicting tissue of origin for true positives: 70.6%

The test detected 14 common cancer types responsible for 72% of global cancer deaths, with sensitivities ranging from 38.9% to 83.3%.

In symptomatic patients, sensitivity increased to 73.1% at 90.6% specificity, highlighting its potential for early diagnosis.

Consistent results across diverse populations and settings suggest OncoSeek is a promising, affordable, and accessible MCED tool, especially valuable for low- and middle-income countries.

Leave a Reply