Reproducibility of Histologic Classification of Gastric Cancer
Author Information
Author(s): D. Pallil, S. Bianchi, F. Cipriani, P. Duca, A. Amorosi, C. Avellini, A. Russo, A. Saragoni, P. Todde, E. Valdes, C. Vindigni, W.J. Blot, J.F. Fraumeni, Jr., E. Buiattil
Hypothesis
How reproducible are histologic classifications of gastric cancer among different pathologists?
Conclusion
The study found that while there is good overall concordance in histologic classification of gastric cancer, standardization is necessary for multi-centre studies.
Supporting Evidence
- Concordance in histologic classification was about 70-80% among pathologists.
- Intra-observer agreement was very high at 95% for one pathologist.
- Using the Lauren classification system showed a kappa value range of 0.38-0.70.
- Agreement for the Ming classification ranged from 57-73% with kappa values from 0.31-0.55.
- WHO classification agreement ranged from 68-79% with kappa values from 0.34-0.64.
- Exact agreement in diagnosing early vs advanced gastric cancer ranged from 87-96%.
Takeaway
Doctors looked at samples of stomach cancer to see if they agreed on how to classify them, and they mostly did, but sometimes they didn't.
Methodology
A panel review of histologic specimens was conducted by six pathologists to assess inter and intra-observer variability in the classification of gastric cancer.
Potential Biases
Variability in the availability of surgical versus biopsy material across centres could influence classification.
Limitations
The study did not provide information about the age and sex of the patients, which could be relevant for understanding the results.
Participant Demographics
Patients were residents aged 75 or less with histologically confirmed gastric cancer.
Statistical Information
P-Value
p<0.01
Statistical Significance
p<0.01
Want to read the original?
Access the complete publication on the publisher's website