EUVIMEDEUVIMED entdecken
Crossref

PD09.08. Preoperative Nutritional–Inflammatory Composite Scores Derived via Principal Component Analysis for Survival Prediction in Esophageal Cancer: Multicenter Evidence

Ruichen Cui, Xiang Li, Shenglu Lian, Yushang Yang, Longqi Chen

Diseases of the Esophagus · 22.08.2026 · 39 (Supplement_1)

DOI: 10.1093/dote/doag077.147 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Diseases of the Esophagus, online erschienen am 22.08.2026. Verfasst von Ruichen Cui und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Topic Esophageal Cancer: Surgical Treatment of Esophageal Cancer – long term outcomes Background Multiple preoperative nutritional and inflammatory indices have been shown to be associated with cancer prognosis, but they are often used individually, resulting in fragmented information. In esophageal cancer, no effective approach has been established to integrate these indices into composite prognostic tools.

Methods

This multicenter, prospectively maintained, retrospective cohort study included 7,569 patients with pathologically confirmed esophageal cancer who underwent curative esophagectomy (2008–2022). Ten preoperative indices were standardized, and principal component analysis (PCA) derived two components (RC1: inflammatory burden; RC2: nutritional impairment). Kaplan–Meier, multivariable Cox regression, and an XGBoost-Cox model evaluated prognostic value for overall survival (OS) and disease-free survival (DFS).

Results

RC1 and RC2 explained 67% of variance and were significantly associated with poorer OS and DFS (all p < 0.01). The XGBoost-Cox model outperformed the traditional Cox model for OS (AUC: 0.85 [95% CI: 0.80–0.90], 0.83 [0.80–0.87], 0.86 [0.83–0.88] vs. 0.82 [0.80–0.85], 0.79 [0.77–0.81], 0.81 [0.79–0.82] at 1-, 3-, and 5-year) and DFS (AUC: 0.77 [0.69–0.85], 0.79 [0.75–0.83], 0.81 [0.78–0.85] vs. 0.78 [0.75–0.80], 0.77 [0.75–0.79], 0.81 [0.79–0.82]). Shapley additive explanations analysis indicated that RC1 and RC2 were high-impact variables in survival risk prediction.

Conclusion

Nutritional–inflammatory composite indices derived via PCA offer clinically relevant prognostic insights in esophageal cancer. Incorporating these indices into machine learning models may enhance survival prediction and support their application in individualized risk stratification and perioperative management.

Zitieren

APA

Cui, R., Li, X., Lian, S., Yang, Y., & Chen, L. (2026). PD09.08. Preoperative Nutritional–Inflammatory Composite Scores Derived via Principal Component Analysis for Survival Prediction in Esophageal Cancer: Multicenter Evidence. Diseases of the Esophagus, 39(Supplement_1). https://doi.org/10.1093/dote/doag077.147

Vancouver

Cui R, Li X, Lian S, Yang Y, Chen L. PD09.08. Preoperative Nutritional–Inflammatory Composite Scores Derived via Principal Component Analysis for Survival Prediction in Esophageal Cancer: Multicenter Evidence. Diseases of the Esophagus. 2026;39(Supplement_1). doi:10.1093/dote/doag077.147

BibTeX
@article{Cui2026,
  author = {Cui, Ruichen and Li, Xiang and Lian, Shenglu and Yang, Yushang and Chen, Longqi},
  title = {PD09.08. Preoperative Nutritional–Inflammatory Composite Scores Derived via Principal Component Analysis for Survival Prediction in Esophageal Cancer: Multicenter Evidence},
  journal = {Diseases of the Esophagus},
  year = {2026},
  volume = {39},
  number = {Supplement_1},
  doi = {10.1093/dote/doag077.147}
}
Kostenlos starten →