基于淋巴结阳性率的胃印戒细胞癌预后模型的建立与验证

Establishment and validation of a prognostic model for gastric signet ring cell carcinoma based on positive lymph node ratio

目的探讨淋巴结阳性率(LNR)在胃印戒细胞癌(GSRCC)预后中的价值,并构建基于LNR的预后列线图进行验证。方法收集并整理了SEER数据库和江苏大学附属宜兴医院GSRCC患者数据,分别作为训练集和验证集。利用Kaplan-Meier法绘制生存曲线,通过单因素和多因素Cox模型进行预后分析。构建基于LNR的预后列线图,并在验证集中进行验证。结果与低LNR患者相比,高LNR患者肿瘤分期更晚,中LNR患者化疗比例更高(P<0.05)。训练集低、中、高LNR患者的5年癌症特异性生存率分别为71.1%、30.7%、8.9%,差异有统计学意义(P<0.05)。验证集低、中、高LNR患者的5年癌症特异性生存率为85.0%、53.2%、17.9%,差异有统计学意义(P<0.05)。多因素回归分析显示,年龄、肿瘤大小、T分期、LNR、Stage分期、化疗是影响GSRCC预后的独立预后因素。进一步构建列线图模型,该模型在内部抽取和外部验证方面都具有较好的预测一致性。在内部验证中,1、3、5年生存率的曲线下面积(AUC)分别为0.787、0.848、0.855,外部验证的AUC分别为0.724、0.729、0.737。结论LNR对GSRCC具有独立的预后价值,以LNR为基础的列线图模型可被视为GSRCC预后的重要预测指标。

Objective To investigate the prognostic value of positive lymph node ratio(LNR)in gastric signet ring cell carcinoma(GSRCC)and to construct a LNR-based prognostic nomogram for validation.Methods Data from GSRCC patients in the SEER database and Yixing Hospital Affiliated to Jiangsu University were collected and categorized as training and validation sets,respectively.Kaplan-Meier method was used to plot survival curves,and univariate and multivariate Cox models were applied for prognostic analysis.A LNR-based prognostic nomogram was constructed and validated using the validation set.Results Compared with patients with a low LNR,those with a high LNR had a more advanced tumor stage,and patients with a medium LNR had a higher proportion of chemotherapy(P<0.05).In the training set,the 5-year cancer-specific survival(CSS)rates for the low,medium,and high LNR groups were 71.1%,30.7%,and 8.9%,respectively,with statistical differences(P<0.05).In the validation set,the 5-year CSS rates for the low,medium,and high LNR groups were 85.0%,53.2%,and 17.9%,respectively,with statistical differences(P<0.05).Multivariate regression analysis showed that age,tumor size,T stage,LNR,overall Stage,and chemotherapy were independent prognostic factors for GSRCC.Furthermore,a nomogram model was constructed,demonstrating good predictive consistency in both internal and external validation.In internal validation,the area under the curve(AUC)for 1-,3-,and 5-year survival rates were 0.787,0.848,and 0.855,respectively.In external validation,the corresponding AUC values were 0.724,0.729,and 0.737.Conclusions LNR has independent prognostic value in GSRCC patients.The LNR-based nomogram model can be considered an important prognostic indicator for GSRCC.