背景:这项研究旨在开发和验证一种新颖的列线图,以预测手术后升结肠腺癌患者的癌症特异性生存率(CSS)。
方法:从监测中纳入升结肠腺癌患者,流行病学,和1973年至2015年的最终结果(SEER)数据库,随机分为训练集(5930)和验证集(2540)。年龄的截止值,通过X-tile软件计算肿瘤大小和淋巴结比率(LNR)。在训练集中,使用单变量和多变量Cox分析确定独立的预后因素,随后建立了包含这些因素的列线图。来自验证集的数据用于评估列线图的可靠性和准确性,然后与美国癌症联合委员会(AJCC)肿瘤淋巴结转移(TNM)分期系统的第8版进行比较。此外,从中国的单个机构进行外部验证.
结果:从SEER数据库中招募了8470名患者,5930名患者被分配到训练组,2540名患者被分配到内部验证组,并且473名患者的单独组被分配到外部验证组。年龄的最佳截止值,肿瘤大小和淋巴结比率分别为73和85、33和75、4.9和32.8。单因素和多因素Cox多元回归显示,年龄,AJCC第8版T,N级和M级,癌胚抗原(CEA),肿瘤分化,化疗,神经周浸润和LNR是患者CSS的独立危险因素。列线图显示出良好的预测能力,如辨别能力和校准所示,训练和验证集的C统计量为0.835(95%CI,0.823-0.847)和0.848(95%CI,0.830-0.866),外部验证集中为0.732(95%CI,0.664-0.799)。列线图显示出良好的辨别和校准能力,并且比AJCCTNM分期系统表现更好。
结论:新的验证列线图可以有效预测升结肠腺癌患者术后,这种预测能力可以指导临床医生准确的预后判断。
BACKGROUND: This study aimed to develop and validate a novel nomogram to predict the cancer-specific survival (CSS) of patients with ascending colon adenocarcinoma after surgery.
METHODS: Patients with ascending colon adenocarcinoma were enrolled from the Surveillance, Epidemiology, and End Results (SEER) database from 1973 to 2015 and randomly divided into a training set (5930) and a validation set (2540). The cut-off values for age, tumour size and lymph node ratio (LNR) were calculated via X-tile software. In the training set, independent prognostic factors were identified using univariate and multivariate Cox analyses, and a nomogram incorporating these factors was subsequently built. Data from the validation set were used to assess the reliability and accuracy of the nomogram and then compared with the 8th edition of the American Joint Committee on Cancer (AJCC) tumour-node-metastasis (TNM) staging system. Furthermore, external validation was performed from a single institution in China.
RESULTS: A total of 8470 patients were enrolled from the SEER database, 5930 patients were allocated to the training set, 2540 were allocated to the internal validation set and a separate set of 473 patients was allocated to the external validation set. The optimal cut-off values of age, tumour size and lymph node ratio were 73 and 85, 33 and 75 and 4.9 and 32.8, respectively. Univariate and multivariate Cox multivariate regression revealed that age, AJCC 8th edition T, N and M stage, carcinoembryonic antigen (CEA), tumour differentiation, chemotherapy, perineural invasion and LNR were independent risk factors for patient CSS. The nomogram showed good predictive ability, as indicated by discriminative ability and calibration, with C statistics of 0.835 (95% CI, 0.823-0.847) and 0.848 (95% CI, 0.830-0.866) in the training and validation sets and 0.732 (95% CI, 0.664-0.799) in the external validation set. The nomogram showed favourable discrimination and calibration abilities and performed better than the AJCC TNM staging system.
CONCLUSIONS: A novel validated nomogram could effectively predict patients with ascending colon adenocarcinoma after surgery, and this predictive power may guide clinicians in accurate prognostic judgement.