陈思皓, 何淑娅, 邵世龙, 何姗珊, 王东生, 杨牧, 张石川. 基于外周血免疫学特征的鼻咽癌初诊远处转移风险评估模型构建[J]. 中国肿瘤临床, 2022, 49(8): 401-406. DOI: 10.12354/j.issn.1000-8179.2022.20211827
引用本文: 陈思皓, 何淑娅, 邵世龙, 何姗珊, 王东生, 杨牧, 张石川. 基于外周血免疫学特征的鼻咽癌初诊远处转移风险评估模型构建[J]. 中国肿瘤临床, 2022, 49(8): 401-406. DOI: 10.12354/j.issn.1000-8179.2022.20211827
shu
Sihao Chen, Shuya He, Shilong Shao, Shanshan He, Dongsheng Wang, Mu Yang, Shichuan Zhang. Construction of risk assessment model for distant metastasis of de novo nasopharyngeal carcinoma based on peripheral blood immunological characteristics[J]. CHINESE JOURNAL OF CLINICAL ONCOLOGY, 2022, 49(8): 401-406. DOI: 10.12354/j.issn.1000-8179.2022.20211827
Citation: Sihao Chen, Shuya He, Shilong Shao, Shanshan He, Dongsheng Wang, Mu Yang, Shichuan Zhang. Construction of risk assessment model for distant metastasis of de novo nasopharyngeal carcinoma based on peripheral blood immunological characteristics[J]. CHINESE JOURNAL OF CLINICAL ONCOLOGY, 2022, 49(8): 401-406. DOI: 10.12354/j.issn.1000-8179.2022.20211827
shu

基于外周血免疫学特征的鼻咽癌初诊远处转移风险评估模型构建

Construction of risk assessment model for distant metastasis of de novo nasopharyngeal carcinoma based on peripheral blood immunological characteristics

    摘要
  • 摘要:
      目的  对比分析初诊鼻咽癌有远处转移和无远处转移患者治疗前外周血免疫指标及其他临床特征，建立风险评估模型早期鉴别远处转移高风险患者。
      方法  回顾性分析2019年1月至2021年6月于四川省肿瘤医院231例初诊鼻咽癌患者临床病例资料及外周血免疫学指标，采用单因素和多因素Logistic回归分析与远处转移相关的临床指标，筛选独立相关因素，绘制列线图风险评估模型并进行内部验证。
      结果  231例初诊患者中共48例（20.8%）患者伴随远处器官转移。多因素回归分析显示，N分期、PNI指数、LDH水平、CD4/CD8比值以及NK细胞绝对数是初诊远处转移的独立相关因素（均P<0.05），据此构建列线图预测模型。采用Bootstrap重抽样法进行1 000次内部验证，该模型一致性指数（C-index）及受试者工作特征曲线下面积（AUC）值为0.826。Calibration校正曲线拟合良好，预测概率与实际概率之间具有较好的符合度。临床决策曲线（DCA）和临床影响曲线（CIC）进一步表明，该模型具有相对稳定的临床适用性。
      结论  鼻咽癌初诊远处转移患者具有显著的外周血免疫学指标变化。根据CD4/CD8比值和NK细胞计数，结合其他临床指标，可以较好地区分转移患者和非转移患者。据此建立的风险预测模型或有助于辅助临床鉴别初诊有隐匿性远处转移患者，实现远处转移的早期干预。

     

    Abstract:
      Objective  To compare the immunological factors of peripheral blood and other clinical characteristics between newly diagnosed nasopharyngeal cancer patients, with and without distant metastasis, and to establish a risk assessment model for early identification of patients with distant metastasis.
      Methods  The clinical data and pretreatment peripheral blood immunological indices of 231 patients with nasopharyngeal carcinoma, diagnosed between January 2019 and June 2021 at Sichuan Cancer Hospital & Institute, were retrospectively analyzed. Factors associated with distant metastasis were analyzed using univariate and multivariate Logistic regression. Independent correlates of distant metastasis were identified, and the corresponding nomogram models were generated. Internal validation was then performed.
      Results  Among the 231 patients enrolled, a total of 48 patients (20.8%) were clinically diagnosed with distant metastasis. Multifactorial regression analysis showed that N stage, prognostic nutritional index (PNI), lactate dehydrogenase (LDH) level, CD4/CD8 ratio, and absolute number of NK cells were independent correlates of distant metastasis. The risk assessment model was constructed accordingly. Using the bootstrap resampling method for 1, 000 internal validations, the concordance index (C-index) and area under the receiver operating characteristic curve (AUC) values of the model were both 0.826. The calibration curve was well fitted, and the predicted probability and the actual probability had a good degree of agreement. The clinical decision curve (DCA) and clinical impact curve (CIC) further indicated that the model has relatively stable clinical applicability.
      Conclusions  Patients with primary distant metastases of nasopharyngeal carcinoma had significant immunological changes in the peripheral blood. Combining the CD4/8 ratio, NK cell count, and several other clinical indicators could help in distinguishing metastatic and non-metastatic patients. The prediction model established may be useful for the screening of patients with early and occult metastasis to enable early intervention.

     

    • HTML全文
    • 参考文献(15)
    • 相关文章
    • 施引文献
  • 资源附件(0)

/

返回文章
返回