A clinicopathological model to predict bone metastasis in hepatocellular carcinoma
A clinicopathological model to predict bone metastasis in hepatocellular carcinoma
复制标题
预测肝细胞癌骨转移的临床病理模型
DOI：
10.1007/s00432-011-1060-7
复制
发表时间：
2011-12-01
影响因子：
3.6
通讯作者：
Tang, Zhao-You
中科院分区：
文献类型：
--
作者：
Xiang, Zuo-Lin;Zeng, Zhao-Chong;Tang, Zhao-You
目的建立预测肝细胞癌骨转移风险的临床病理模型。方法我们首先评估了201例接受肝切除术的肝细胞癌患者的训练队列，发现以下因素独立地预测了骨转移的发生：血管侵犯、肿瘤转移分期、CXCR4、结缔组织生长因子和白介素11。这些变量被用来构建一个临床病理预测模型，得分从0到19。该模型的预测价值在179名肝切除术后肝癌患者的验证队列中得到了验证。结果在训练队列和验证队列的中位随访期分别为54.3个月和52.5个月，前者有23名患者(11.4%)发生BM，后者有19名患者(10.6%)发生BM。截断值9.4最好地区分了BM风险，并能够在验证队列中以高精度排除未来的BM发展。该方法的灵敏度为73.7%，特异度为78.7%，阳性预测值为29.2%，阴性预测值为96.2%。高危组1年和2年累积骨髓率分别为10.8%和27.4%，低危组分别为2.4%和4.3%。高危组与低危组发生骨质疏松症的风险比为9.240(95%CI：3.319~25.722)。结论根据临床病理参数建立的简单预测模型能准确预测肝癌患者骨质疏松症的发生。
BackgroundWe aimed to develop a clinicopathological model that would predict the risk of bone metastasis (BM) in hepatocellular carcinoma (HCC).MethodsWe first evaluated a training cohort of 201 HCC patients who had undergone hepatectomy and found that the following factors independently predicted BM development: vascular invasion, tumor-node-metastasis stage, CXCR4, connective tissue growth factor, and interleukin-11. These variables were used to construct a clinicopathological prediction model that may be scored from 0 to 19. The predictive value of the model was demonstrated in a validation cohort of 179 post-hepatectomy HCC patients.ResultsDuring a median follow-up of 54.3 months for the training cohort and 52.5 months for the validation cohort, 23 patients (11.4%) in the former and 19 patients (10.6%) in the latter developed BM. A cutoff value of 9.4 best discriminated BM risk and was able to exclude future BM development with high accuracy in the validation cohort. The sensitivity and specificity of the method were 73.7 and 78.7%, respectively, the positive predictive value was 29.2%, and the negative predictive value 96.2%. The 1- and 2-year cumulative BM rates were (respectively) 10.8% and 27.4% in the high-risk group and 2.4 and 4.3% in the low-risk group. The hazard ratio for BM of the high- versus low-risk group was 9.240 (95% CI: 3.319–25.722).ConclusionThe simple prediction model constructed from clinicopathological parameters is accurate in predicting BM development in HCC patients.