Haddad Ahmed Q, Luo Jun-Hang, Krabbe Laura-Maria, Darwish Oussama, Gayed Bishoy, Youssef Ramy, Kapur Payal, Rakheja Dinesh, Lotan Yair, Sagalowsky Arthur, Margulis Vitaly
Department of Urology, University of Louisville, Louisville, TX, USA.
Department of Urology, University of Texas Southwestern Medical Center, Dallas, TX, USA.
BJU Int. 2017 May;119(5):741-747. doi: 10.1111/bju.13776. Epub 2017 Feb 9.
To improve risk stratification for recurrence prognostication in patients with localised clear cell renal cell carcinoma (ccRCC).
In all, 367 patients with non-metastatic ccRCC were included. The cohort was divided into a training and validation set. Using tissue microarrays, immunostaining was performed for 24 biomarkers representative of key pathways in ccRCC. Using Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression, we identified several markers that were used to construct a risk classifier for risk of disease recurrence.
The median (interquartile range) follow-up was 63.5 (24.0-85.3) months. Five out of 24 markers were selected by LASSO Cox regression for the risk classifier: N-cadherin, E-cadherin, Ki67, cyclin D1 and phosphorylated eukaryotic initiation factor 4E binding protein-1 (p-4EBP1). Patients were classified as either low, intermediate or high risk of disease recurrence by tertiles of risk score. The 5-year recurrence-free survival (RFS) was 93.8%, 87.7% and 70% for patients with low-, intermediate- and high-risk scores, respectively (P < 0.001). Patients with a high marker score had worse RFS on multivariate analysis adjusted for age, gender, race and the Mayo Clinic Stage, Size, Grade, and Necrosis (SSIGN) score (hazard ratio 3.66, 95% confidence interval 1.58-8.49, P = 0.003 for high vs low marker score in the overall cohort). The five-marker classifier increased the concordance index of the clinical model in both the training and validation sets.
We developed a five-marker-based prognostic tool that can effectively classify patients with ccRCC according to risk of disease recurrence after surgery. This tool, if prospectively validated, could provide individualised risk estimation for patients with ccRCC.
改善局限性透明细胞肾细胞癌（ccRCC）患者复发预后的风险分层。
共纳入367例非转移性ccRCC患者。该队列被分为训练集和验证集。使用组织微阵列，对代表ccRCC关键通路的24种生物标志物进行免疫染色。使用最小绝对收缩和选择算子（LASSO）Cox回归，我们确定了几种用于构建疾病复发风险分类器的标志物。
中位（四分位间距）随访时间为63.5（24.0 - 85.3）个月。LASSO Cox回归为风险分类器选择了24种标志物中的5种：N - 钙黏蛋白、E - 钙黏蛋白、Ki67、细胞周期蛋白D1和磷酸化真核起始因子4E结合蛋白 - 1（p - 4EBP1）。根据风险评分三分位数，患者被分为疾病复发低、中或高风险组。低、中、高风险评分患者的5年无复发生存率（RFS）分别为93.8%、87.7%和70%（P < 0.001）。在对年龄、性别、种族和梅奥诊所分期、大小、分级及坏死（SSIGN）评分进行多变量分析调整后，标志物评分高的患者RFS更差（总体队列中，高标志物评分与低标志物评分相比，风险比为3.66，95%置信区间为1.58 - 8.49，P = 0.003）。五标志物分类器提高了训练集和验证集中临床模型的一致性指数。
我们开发了一种基于五种标志物的预后工具，可根据术后疾病复发风险有效对ccRCC患者进行分类。该工具若经前瞻性验证，可为ccRCC患者提供个体化风险评估。
