Predicting overall survival in synchronous metastatic nasopharyngeal carcinoma using a stacking ensemble machine learning model: a multicenter retrospective study - Takeaways - MDSpire

Predicting overall survival in synchronous metastatic nasopharyngeal carcinoma using a stacking ensemble machine learning model: a multicenter retrospective study

  • By

  • Canwen Che

  • Shanyue Lin

  • Jie Ma

  • Haibo Liu

  • Wentao Liu

  • Zhanhong Tan

  • Xin Chen

  • Xiaoyi Zeng

  • Qiwen Duan

  • Guanxun Cheng

  • July 1, 2026

  • 0 min

Share

  • 1

    A stacking ensemble machine learning model was developed to predict 3-year overall survival in patients with synchronous metastatic nasopharyngeal carcinoma.

  • 2

    The study included 413 patients, with 289 in the training set and 124 in the internal validation set, using various demographic and clinical predictors.

  • 3

    The stacking ensemble achieved an internal held-out validation AUC of 0.8358, outperforming several individual machine learning models.

  • 4

    Key predictors identified by SHAP included immunotherapy, first-line regimen, number of metastatic lesions, and number of metastatic organs.

  • 5

    The study highlights the potential of ensemble learning to improve prognostic assessment in heterogeneous clinical datasets.

Original Source(s)

Predicting overall survival in synchronous metastatic nasopharyngeal carcinoma using a stacking ensemble machine learning model: a multicenter retrospective study

  • by Canwen Che, Shanyue Lin, Jie Ma, Haibo Liu, Wentao Liu, Zhanhong Tan, Xin Chen, Xiaoyi Zeng, Qiwen Duan, Guanxun Cheng

  • July 1, 2026

Related Content

Systematic pan-cancer analysis reveals context-dependent prognostic and immunological roles of WDR1 with divergent effects in renal and gastric cancers

Ferroptosis in chemotherapy resistance and resensitization in breast cancer: a systematic review of preclinical evidence and translational implications

Therapeutic inhibition of RAS in non-small cell lung cancer