Department of Thoracic Surgery, Zhongshan Hospital of Fudan University, Shanghai 200032, China.
Br J Surg. 2018 Oct;105(11):1464-1470. doi: 10.1002/bjs.10882. Epub 2018 Jun 4.
Lymph node status is crucial in determining the prognosis for early oesophageal squamous cell carcinoma (SCC). This study aimed to develop and validate a nomogram for the prediction of lymph node metastasis in patients with early SCC.
A prediction model was developed in a derivation cohort of patients with clinicopathologically confirmed early SCC. Patients who underwent oesophagectomy for pT1 SCC between January 2010 and December 2013 were identified from an institutional database. Risk factors for lymph node metastasis were assessed using a binary logistic regression modelling technique. A nomogram for the prediction of lymph node metastasis was constructed using the results of multivariable analyses. For internal validation, bootstraps with 1000 resamples were performed. The predictive performance of the nomogram was measured by Harrell's concordance index (C-index). An independent cohort from the same hospital was used to validate the nomogram. This cohort included consecutive patients with early SCC who underwent oesophagectomy from January 2014 to December 2015.
The derivation cohort included 281 patients. Four variables associated with lymph node metastasis were included in the model: depth of tumour invasion (odds ratio (OR) 4·37, 95 per cent c.i. 1·59 to 12·03; P = 0·004), grade of differentiation (OR 4·47, 1·02 to 19·70; P = 0·048), tumour size (OR 2·52, 1·11 to 5·75; P = 0·028) and lymphovascular invasion (OR 6·58, 2·54 to 17·05; P < 0·001). The C-index was 0·790 (95 per cent c.i. 0·717 to 0·864) in the derivation cohort and 0·789 (0·709 to 0·869) for the validation cohort (198 patients).
A validated nomogram for patients with early oesophageal SCC can predict the risk of lymph node metastasis.
淋巴结状态对早期食管鳞癌（SCC）的预后至关重要。本研究旨在开发和验证一种预测早期 SCC 患者淋巴结转移的列线图。
在一个临床病理确诊为早期 SCC 的患者队列中建立预测模型。从机构数据库中确定 2010 年 1 月至 2013 年 12 月期间接受 T1 期 SCC 食管切除术的患者。使用二元逻辑回归模型技术评估淋巴结转移的危险因素。使用多变量分析的结果构建用于预测淋巴结转移的列线图。通过 1000 次重采样进行bootstrap 进行内部验证。列线图的预测性能通过 Harrell 的一致性指数（C-index）进行测量。使用来自同一医院的独立队列验证该列线图。该队列包括 2014 年 1 月至 2015 年 12 月期间接受食管切除术的连续早期 SCC 患者。
该推导队列包括 281 名患者。纳入模型的与淋巴结转移相关的四个变量包括：肿瘤浸润深度（优势比（OR）4.37，95%置信区间（CI）1.59 至 12.03；P=0.004）、分化程度（OR 4.47，1.02 至 19.70；P=0.048）、肿瘤大小（OR 2.52，1.11 至 5.75；P=0.028）和淋巴管侵犯（OR 6.58，2.54 至 17.05；P<0.001）。该推导队列的 C 指数为 0.790（95%CI 为 0.717 至 0.864），验证队列的 C 指数为 0.789（0.709 至 0.869）（198 名患者）。
用于早期食管 SCC 患者的验证列线图可以预测淋巴结转移的风险。
