Three-Tiered Risk Stratification Model for Barrett's Esophagus
Author Information
Author(s): Sato Fumiaki, Jin Zhe, Schulmann Karsten, Wang Jean, Greenwald Bruce D., Ito Tetsuo, Kan Takatsugu, Hamilton James P., Yang Jian, Paun Bogdan, David Stefan, Olaru Alexandru, Cheng Yulan, Mori Yuriko, Abraham John M., Yfantis Harris G., Wu Tsung-Teh, Fredericksen Mary B., Wang Kenneth K., Canto Marcia, Romero Yvonne, Feng Ziding, Meltzer Stephen J.
Primary Institution: Johns Hopkins University School of Medicine
Hypothesis
Can a three-tiered risk stratification model based on epigenetic and clinical features improve the prediction of progression in Barrett's esophagus?
Conclusion
The three-tiered risk stratification strategy can significantly enhance the accuracy and efficiency of Barrett's esophagus surveillance.
Supporting Evidence
- The study defined high-grade dysplasia as the endpoint of progression.
- Progression-free survivals differed significantly among the three risk groups.
- The three-tiered stratification method retained high specificity and sensitivity.
- Incremental value analyses showed that the number of methylated genes was the most influential factor for prediction accuracy.
Takeaway
This study created a way to sort patients with Barrett's esophagus into three risk groups to help doctors know who needs more check-ups to prevent cancer.
Methodology
The study analyzed 118 Barrett's esophagus tissue samples from 62 patients using epigenetic and clinical parameters to create prediction models.
Potential Biases
Potential sampling bias due to the heterogeneity of methylation status in Barrett's esophagus tissues.
Limitations
The number of progressor patients and specimens was relatively small, which may affect the generalizability of the findings.
Participant Demographics
Patients included 35 progressor and 27 non-progressor Barrett's esophagus patients.
Statistical Information
P-Value
p=0.0022
Confidence Interval
95% C.I.: 0.7273–0.9501
Statistical Significance
p<0.0001
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website