Department of Urology, HELIOS Hospital, Bad Saarow, Germany.
Department of Urology, Campus Benjamin Franklin, University Hospital Charité, Germany.
Cancer Med. 2017 Oct;6(10):2252-2262. doi: 10.1002/cam4.1161. Epub 2017 Sep 6.
To improve the clinical decision-making regarding further treatment management and follow-up scheduling for patients with muscle-invasive bladder cancer (MIBC) after radical cystectomy (RC), a better prediction accuracy of prognosis for these patients is urgently needed. The objective of this study was to evaluate the validity of differentially expressed microRNAs (miRNAs) based on a previous study as prognostic markers for overall survival (OS) after RC in models combined with clinicopathological data. The expression of six miRNAs (miR-100-5p, miR-130b-3p, miR-141-3p, miR-199a-3p, miR-205-5p, and miR-214-3p) was measured by RT-qPCR in formalin-fixed, paraffin-embedded tissue samples from 156 MIBC patients who received RC in three urological centers. Samples from 2000 to 2013 were used according to their tissue availability, with follow-up until June 2016. The patient cohort was randomly divided into a training (n = 100) and test set (n = 56). Seventy-three samples from adjacent normal tissue were used as controls. Kaplan-Meier, univariate and multivariate Cox regression, and decision curve analyses were carried out to assess the association of clinicopathological variables and miRNAs to OS. Both increased (miR-130b-3p and miR-141-3p) and reduced (miR-100-5p, miR-199a-3p, and miR-214-3p) miRNA expressions were found in MIBC samples in comparison to nonmalignant tissue samples (P < 0.0001). miR-199a-3p and miR-214-3p were independent markers of OS in Cox regression models with the significant clinicopathological variables age, tumor status, and lymph node status. The prediction model with the clinicopathological variables was improved by these two miRNAs in both sets. The predictive benefit was confirmed by decision curve analysis. In conclusion, the inclusion of both miRNAs into models based on clinical data for the outcome prediction of MIBC patients after RC could be a valuable approach to improve prognostic accuracy.
为了改善接受根治性膀胱切除术 (RC) 后的肌层浸润性膀胱癌 (MIBC) 患者的进一步治疗管理和随访计划的临床决策，迫切需要更准确地预测这些患者的预后。本研究的目的是评估基于先前研究的差异表达 microRNAs (miRNAs) 作为 RC 后总生存 (OS) 的预后标志物的有效性，这些 miRNA 结合了临床病理数据建立的模型。使用 RT-qPCR 测量了 156 例 MIBC 患者在三个泌尿科中心接受 RC 后福尔马林固定、石蜡包埋组织样本中六种 miRNA (miR-100-5p、miR-130b-3p、miR-141-3p、miR-199a-3p、miR-205-5p 和 miR-214-3p) 的表达。根据组织可用性，使用了 2000 年至 2013 年的样本，并随访至 2016 年 6 月。患者队列被随机分为训练集 (n = 100) 和测试集 (n = 56)。使用 73 例相邻正常组织样本作为对照。进行 Kaplan-Meier、单变量和多变量 Cox 回归以及决策曲线分析，以评估临床病理变量和 miRNAs 与 OS 的关联。与非恶性组织样本相比，MIBC 样本中发现 miRNA 表达升高 (miR-130b-3p 和 miR-141-3p) 和降低 (miR-100-5p、miR-199a-3p 和 miR-214-3p) (P < 0.0001)。miR-199a-3p 和 miR-214-3p 是 Cox 回归模型中年龄、肿瘤状态和淋巴结状态等显著临床病理变量的 OS 独立标志物。在两个队列中，将这两个 miRNA 纳入基于临床数据的模型可以改善 MIBC 患者 RC 后结局预测的准确性。决策曲线分析证实了该预测模型的获益。总之，将这两个 miRNA 纳入基于临床数据的模型，以预测 MIBC 患者 RC 后的结局，可能是提高预后准确性的一种有价值的方法。
