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

Source: https://onco.cc/key-papers/paper-oncoseek-npj-precis-oncol-2025/  
OnCo record `paper-oncoseek-npj-precis-oncol-2025` (Key paper). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Pooling seven centres in three countries and 15,122 people, the OncoSeek protein test found about six in ten cancers with about eight false positives per hundred people without cancer.

## Summary

Shen and colleagues combined four new cohorts with the earlier training and validation sets: 15,122 participants (3,029 with cancer) from seven centres in three countries, on four platforms and two sample types. OncoSeek had an area under the curve of 0.829, 58.4% sensitivity, 92.0% specificity and 70.6% tissue-of-origin accuracy in true positives, across 14 cancer types with sensitivities of 38.9% to 83.3%. In a symptomatic cohort it had 73.1% sensitivity at 90.6% specificity.

## Fields

- Kind: Key paper
- Last checked: 2026-10-29
- Journal: npj Precision Oncology
- Year: 2025
- DOI: 10.1038/s41698-025-01105-2
- Authors: Shen Y, Xia Y, Chang Y, et al.
- Findings: 58.4% sensitivity, 92.0% specificity across 14 cancers.; Symptomatic cohort: 73.1% sensitivity at 90.6% specificity.
- What it means: Consistent across sites and platforms, with the same limit as before: specificity too low for population screening without a second test.
- Caveats: Mostly retrospective and case-control cohorts.; Authors include SeekIn staff.

## Sources

- npj Precision Oncology 2025: https://doi.org/10.1038/s41698-025-01105-2
- PubMed: https://pubmed.ncbi.nlm.nih.gov/41062647/

## Connected records

- fronts: [Early Detection & Screening](https://onco.cc/fronts/early-detection/)
- technologies: [Multi-cancer early detection (MCED)](https://onco.cc/technologies/mced/), [Proteomics & phosphoproteomics](https://onco.cc/technologies/proteomics/)
- drugs: [OncoSeek](https://onco.cc/drugs/oncoseek/)
- companies: [SeekIn](https://onco.cc/companies/seekin/)

---
JSON: https://onco.cc/api/v1/entities/paper-oncoseek-npj-precis-oncol-2025.json