Wang Sisheng, Zheng Shaoluan, Hu Kongzu, Sun Heyan, Zhang Jinling, Rong Genxiang, Gao Jie, Ding Nan, Gui Binjie
Department of Joint and Reconstructive Microsurgery, the First Affiliated Hospital of Anhui Medical University, He Fei Xia Men Hospital of Traditional Chinese Medicine, Department of Thoracic Surgery, Xia Men, China.
Medicine (Baltimore). 2017 Jan;96(3):e5909. doi: 10.1097/MD.0000000000005909.
Osteosarcomas (OSs) represent a huge challenge to improve the overall survival, especially in metastatic patients. Increasing evidence indicates that both tumor-associated elements but also on host-associated elements are under a remarkable effect on the prognosis of cancer patients, especially systemic inflammatory response. By analyzing a series prognosis of factors, including age, gender, primary tumor size, tumor location, tumor grade, and histological classification, monocyte ratio, and NLR ratio, a clinical predictive model was established by using stepwise logistic regression involved circulating leukocyte to compute the estimated probabilities of metastases for OS patients. The clinical predictive model was described by the following equations: probability of developing metastases = ex/(1 + ex), x = -2.150 + (1.680 × monocyte ratio) + (1.533 × NLR ratio), where is the base of the natural logarithm, the assignment to each of the 2 variables is 1 if the ratio >1 (otherwise 0). The calculated AUC of the receiver-operating characteristic curve as 0.793 revealed well accuracy of this model (95% CI, 0.740-0.845). The predicted probabilities that we generated with the cross-validation procedure had a similar AUC (0.743; 95% CI, 0.684-0.803). The present model could be used to improve the outcomes of the metastases by developing a predictive model considering circulating leukocyte influence to estimate the pretest probability of developing metastases in patients with OS.
骨肉瘤（OS）对提高总体生存率构成了巨大挑战，尤其是在转移性患者中。越来越多的证据表明，肿瘤相关因素以及宿主相关因素都对癌症患者的预后有显著影响，尤其是全身炎症反应。通过分析一系列预后因素，包括年龄、性别、原发肿瘤大小、肿瘤位置、肿瘤分级、组织学分类、单核细胞比率和中性粒细胞与淋巴细胞比率（NLR），使用逐步逻辑回归建立了一个临床预测模型，该模型涉及循环白细胞来计算骨肉瘤患者发生转移的估计概率。该临床预测模型由以下方程描述：发生转移的概率 = ex / (1 + ex)，x = -2.150 + (1.680 × 单核细胞比率) + (1.533 × NLR比率)，其中e是自然对数的底数，如果比率>1，则两个变量中的每一个赋值为1（否则为0）。计算得到的受试者工作特征曲线的AUC为0.793，表明该模型具有良好的准确性（95%CI，0.740 - 0.845）。我们通过交叉验证程序生成的预测概率具有相似的AUC（0.743；95%CI，0.684 - 0.803）。通过建立一个考虑循环白细胞影响的预测模型来估计骨肉瘤患者发生转移的预测试概率，本模型可用于改善转移患者的预后。
