Establishment and evaluation of a nomogram for predicting the survival outcomes of patients with diffuse large B-cell lymphoma based on International Prognostic Index scores and clinical indicators.
Establishment and evaluation of a nomogram for predicting the survival outcomes of patients with diffuse large B-cell lymphoma based on International Prognostic Index scores and clinical indicators.
复制标题
基于国际预后指数评分和临床指标预测弥漫性大B细胞淋巴瘤患者生存结果的列线图的建立和评估
DOI：
10.21037/atm-22-6023
复制
发表时间：
2023-01-31
影响因子：
--
通讯作者：
Liu, Aichun
中科院分区：
文献类型：
--
作者：
Chen, Yao;Xu, Jiaqin;Meng, Jiao;Ding, Mingshuang;Guo, Yiwei;Fu, Dongwei;Liu, Aichun
背景弥漫性大B细胞淋巴瘤（DLBCL）是最常见的侵袭性淋巴瘤，患者的治疗结果差异很大。目前的国际预后指数（IPI）不足以区分预后不良的患者，而且基因检测非常昂贵，因此应开发一种廉价的风险预测工具，供临床医生快速识别DLBCL患者的预后不良。方法选择2008年至2017年在我院接受环磷酰胺、阿霉素、长春新碱和泼尼松联合治疗（CHOP）或不联合利妥昔单抗（R-CHOP）的DLBCL患者420例（18-80岁）。通过单变量和多变量考克斯回归分析确定潜在的生存预测因子，并使用显著变量构建预测列线图。新的预测模型进行了评估，使用一致性指数（C指数），校准曲线，并评估其临床效用的决策曲线分析（DCA）。结果全组5年总生存率（OS）为70.62%，5年无进展生存率（PFS）为59.02%。多变量考克斯分析表明，IPI，Ki-67，淋巴细胞/单核细胞比率，利妥昔单抗一线治疗与生存率显著相关。C指数结果表明，包括这些变量的预测模型对OS（0.73 vs. 0.67）和PFS（0.68 vs. 0.63）的区分度优于基于IPI的模型。校准图显示出良好的协议与观察和诺模图预测。DCA证明了列线图的临床价值。结论我们的研究确定了影响新诊断DLBCL患者预后的因素，并结合IPI和常见临床指标构建了个体化风险预测模型。我们的模型可能是一个有价值的工具，可用于预测接受标准一线治疗方案的DLBCL患者的预后。它使临床医生能够快速识别一些可能预后不良的患者，并为患者选择更积极的治疗，如嵌合抗原受体T细胞（CART）免疫治疗和其他新药治疗，从而延长患者的PFS和OS。
Background Diffuse large B-cell lymphoma (DLBCL) is the most common aggressive lymphoma, treatment outcomes of patients vary greatly. The current International Prognostic Index (IPI) is not enough to distinguish patients with poor prognosis, and genetic testing is very expensive, so a inexpensive risk prediction tool should be developed for clinicians to quickly identify the poor prognosis of DLBCL patients. Methods DLBCL patients (n=420; 18–80 years old) who received a combination of cyclophosphamide, adriamycin, vincristine, and prednisone (CHOP) with or without rituximab (R-CHOP) at our hospital between 2008 and 2017 were included in the study. Potential predictors of survival were determined by univariate and multivariate Cox regression analyses, and significant variables were used to construct predictive nomograms. The new prediction models were assessed using concordance indexes (C-indexes), calibration curves, and their clinical utility was assessed by decision curve analyses (DCAs). Results The 5-year overall survival (OS) rate was 70.62% and the 5-year progression-free survival (PFS) rate was 59.02%. The multivariate Cox analysis indicated that IPI, Ki-67, the lymphocyte/monocyte ratio, and first-line treatment with rituximab were significantly associated with survival. The C-index results indicated that a predictive model that included these variables had better discriminability for OS (0.73 vs. 0.67) and PFS (0.68 vs. 0.63) than the IPI-based model. The calibration plots showed good agreement with observations and nomogram predictions. The DCAs demonstrated the clinical value of the nomograms. Conclusions Our study identified prognostic factors in patients who were newly diagnosed with DLBCL to construct an individualized risk prediction model, combined IPI with common clinical indicators. Our model might be a valuable tool that could be used to predict the prognosis of DLBCL patients who receive standard first-line treatment regimens. It enables clinicians to quickly identify some patients with possible poor prognosis and choose more active treatment for patients, such as chimeric antigen receptor T-cell (CART) Immunotherapy and other new drugs therapy, so as to prolong the PFS and OS of patients.