Predicting Tertiary Lymphoid Structures in Lung Cancer Using CT Imaging
Author Information
Author(s): Zhao Xiaojiang, Wang Yuhang, Xue Mengli, Ding Yun, Zhang Han, Wang Kai, Ren Jie, Li Xin, Xu Meilin, Lv Jun, Wang Zixiao, Sun Daqiang
Primary Institution: Tianjin Chest Hospital, Tianjin University, Tianjin, China
Hypothesis
Can a multimodal predictive model based on preoperative CT radiomic features identify tertiary lymphoid structures (TLSs) in stage I lung adenocarcinoma patients?
Conclusion
The RAITS model effectively identifies TLSs in stage I lung adenocarcinoma and may improve clinical decision-making.
Supporting Evidence
- RAITS achieved an AUC of 0.78 in the validation cohort.
- TLS-positive patients had significantly higher recurrence-free survival compared to TLS-negative patients.
- RAITS outperformed traditional unimodal models in both training and validation cohorts.
Takeaway
Doctors can use a special computer model to see if certain structures in lung cancer patients might help them choose better treatments.
Methodology
The study used preoperative CT scans and pathology sections to develop and validate a predictive model for identifying TLSs.
Potential Biases
Potential selection bias due to the retrospective nature of the study.
Limitations
The study is retrospective, which may introduce selection bias, and the sample size may need to be expanded for better validation.
Participant Demographics
Patients included were diagnosed with stage I lung adenocarcinoma, with a mix of genders and ages.
Statistical Information
P-Value
p<0.01 for training cohort, p=0.02 for validation cohort
Confidence Interval
95% CI, 0.69–0.88
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website