Ma Wenfang, Chang Zukuan, Li Shixing, Wang Xiuhua, Cao Guangshao, Fan Youjie
Department of Interventional Therapy, Xinxiang Central Hospital (the Fourth Clinical College of Xinxiang Medical College), Xinxiang, China.
Department of Comprehensive Intervention, Henan Provincial People's Hospital, Zhengzhou, China.
Front Oncol. 2025 Jul 10;15:1604994. doi: 10.3389/fonc.2025.1604994. eCollection 2025.
To explore the predictive value of serum miR-21 and miR-122 expressions on the efficacy of capecitabine combined with TACE for the treatment of postoperative liver metastasis in colorectal cancer patients, and to construct a nomogram model for verification.
A total of 252 patients who received this treatment from January 2021 to December 2023 were included in the study. The dataset was randomly split at a 7:3 ratio into a training set (n=181) and a validation set (n=71). Serum levels of miR-21 and miR-122 before treatment were detected and the relationship with clinical pathological characteristics was analyzed. Independent risk factors were screened by multivariate Logistic regression, and a nomogram model was constructed to evaluate efficacy.
In the training set, there were 86 cases with effective treatment and 95 cases with ineffective treatment after operation. Multivariate analysis showed that CEA, high serum miR-21 expression, low miR-122 expression, tumor size, BMI, and age were the independent risk factors for efficacy (<0.05). The nomogram model exhibited C-indexes of 0.809 (training set) and 0.732 (validation set). Additionally, the average absolute errors of the calibration curves were 0.178 and 0.210, respectively. The Hosmer-Lemeshow test result was good. The Receiver operating characteristic (ROC) curve showed that the area under the curve (AUC) of the model in predicting the efficacy was 0.810 (95% : 0.734-0.885) and 0.731 (95% : 0.597-0.866) in the training set and the verification set, respectively. The sensitivities and specificities were 0.820, 0.716 and 0.600 and 0.714, respectively.
The expression levels of serum miR-21 and miR-122 have predictive value for the efficacy of liver metastasis after colorectal cancer treatment. The nomogram model has good predictive performance, which can provide a reference for clinical decision-making. Furthermore, the identified predictive value of miR-21 and miR-122 provides a basis for exploring personalized combination therapies with targeted agents in future studies, which may help overcome the limitations of conventional chemotherapy.
探讨血清miR-21和miR-122表达对结直肠癌患者术后肝转移卡培他滨联合经动脉化疗栓塞术（TACE）疗效的预测价值，并构建列线图模型进行验证。
纳入2021年1月至2023年12月接受该治疗的252例患者。数据集按7:3的比例随机分为训练集（n = 181）和验证集（n = 71）。检测治疗前血清miR-21和miR-122水平，并分析其与临床病理特征的关系。通过多因素Logistic回归筛选独立危险因素，构建列线图模型评估疗效。
在训练集中，术后治疗有效的有86例，无效的有95例。多因素分析显示，癌胚抗原（CEA）、血清miR-21高表达、miR-122低表达、肿瘤大小、体重指数（BMI）和年龄是疗效的独立危险因素（<0.05）。列线图模型在训练集和验证集的C指数分别为0.809和0.732。此外，校准曲线的平均绝对误差分别为0.178和0.210。Hosmer-Lemeshow检验结果良好。受试者工作特征（ROC）曲线显示，该模型在训练集和验证集中预测疗效的曲线下面积（AUC）分别为0.810（95%：0.734 - 0.885）和0.731（95%：0.597 - 0.866）。敏感性和特异性分别为0.820、0.716和0.600、0.714。
血清miR-21和miR-122表达水平对结直肠癌治疗后肝转移疗效具有预测价值。列线图模型具有良好的预测性能，可为临床决策提供参考。此外，所确定的miR-21和miR-122的预测价值为未来研究探索靶向药物的个性化联合治疗提供了依据，这可能有助于克服传统化疗的局限性。
