Using CT Scans to Predict Treatment Response in Esophageal Cancer
Author Information
Author(s): Kong Weibo, Xu Junrui, Huang Yunlong, Zhu Kun, Yao Long, Wu Kaiming, Wang Hanlin, Ma Yuhang, Zhang Qi, Zhang Renquan
Primary Institution: First Affiliated Hospital of Anhui Medical University
Hypothesis
Habitat analysis can predict the effect of neoadjuvant therapy for esophageal cancer.
Conclusion
The study found that a model based on habitat data outperforms traditional radiomic analysis models in predicting treatment response.
Supporting Evidence
- The model based on habitat data significantly improves predictive accuracy for treatment response.
- AUC of habitat models was 0.909 in training and 0.829 in external test cohorts.
- The nomogram model showed an AUC of 0.914 in training and 0.849 in test cohorts.
Takeaway
Doctors can use special scans to see how well treatment works for patients with esophageal cancer, helping them make better decisions.
Methodology
This was a two-center retrospective clinical study using pretreatment CT scans of patients with esophageal cancer treated with neoadjuvant chemoimmunotherapy.
Potential Biases
Manual tumor segmentation by different experts may affect the stability of radiomic features.
Limitations
The study is retrospective, which may introduce selection bias and affect sample representativeness.
Participant Demographics
Patients with biopsy-proven resectable esophageal cancer, treated with neoadjuvant chemoimmunotherapy.
Statistical Information
P-Value
0.014
Confidence Interval
0.8418–0.9758
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website