淋巴轉移之局部晚期食道鱗狀細胞癌預後評分 系統之建立：結合病理、影像與臨床因子的多變項模型（PRIME評分系統）

Abstract

食道鱗狀細胞癌在東亞地區相當常見，診斷時多已進展至局部晚期且合併淋巴結轉移。雖然術前輔助化學放射線治療合併手術已成為標準治療，並能改善部分患者預後，但術後復發風險仍高，臨床上極需更準確的預後評估工具，以輔助術後管理與追蹤決策。 方法： 本研究回顧分析接受術前輔助化學放射線治療與手術的臨床淋巴結陽性局部晚期食道鱗狀細胞癌患者之臨床、影像、治療及病理資料。透過單變項 Cox 回歸分析篩選預後因子，建立一套整合14項顯著預測變數的新型預後評分系統，並進行驗證分析以預測復發與死亡風險。 結果： 此預後評分模型具良好之區辨力與校正能力。預測總生存期方面，其準確率為76.9%，曲線下面積（AUC）為0.734；預測無惡化存活期方面，準確率為71.4%，AUC 為0.721。相較於傳統 AJCC 分期與僅使用病理因子的模型，本評分系統展現出更佳的預測表現。 結論： 本研究提出一套結合臨床、影像、治療與病理資訊的預後評估工具，不僅優於現行標準分期系統，亦具臨床實用性，可協助辨識高風險患者，進一步強化術後照護與追蹤策略。

Background: Esophageal squamous cell carcinoma is a common malignancy in East Asia, frequently diagnosed at a locally advanced stage with lymph node involvement. Although chemoradiotherapy followed by surgery has become the standard treatment and improves outcomes in selected patients, recurrence remains a significant concern. There is a need for accurate prognostic tools to guide postoperative management and follow-up strategies. Methods: This study analyzed clinical, imaging, treatment, and pathological data from patients with clinically node-positive, locally advanced esophageal squamous cell carcinoma who underwent neoadjuvant chemoradiotherapy and surgery. Univariable Cox regression was used to identify prognostic factors. A novel scoring system, integrating 14 significant predictors, was developed and validated to estimate survival and recurrence risk. Results: The prognostic scoring model showed good discrimination and calibration. For predicting overall survival, the scoring system achieved an accuracy of 76.9% and an area under the curve of 0.734. For progression-free survival, the accuracy was 71.4% with an area under the curve of 0.721. Compared with traditional staging methods and pathologic-only systems, the new complete scoring model demonstrated superior predictive performance. Conclusions: This study presents a new risk prediction tool that integrates clinical, imaging, treatment, and pathological data. It outperforms conventional staging and offers a practical approach for identifying high-risk patients who may benefit from more intensive postoperative care and monitoring.