Department of Gastrointestinal Surgery, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, China.
Pathol Oncol Res. 2023 Feb 24;29:1610874. doi: 10.3389/pore.2023.1610874. eCollection 2023.
We aimed to explore reasonable lymph node classification strategies for left-sided colon cancer (LCC) patients. 48,425 LCC patients from 2010 to 2015 were identified in the US Surveillance, Epidemiology, and End Results database. We proposed an innovative revised nodal (rN) staging of the 8th American Joint Committee on Cancer (AJCC) Tumor/Node/Metastasis (TNM) classification based on the cut-off value of retrieved lymph nodes and survival analyses in patients with LCC. Log odds of positive lymph nodes (LODDS) stage is a numerical classification strategy obtained by a formula that incorporates the numbers of retrieved and positive lymph nodes. To develop the TrN or TLODDS classification, patients with similar survival rates were grouped by combining T and rN or LODDS stage. The TrN or TLODDS classification was further evaluated in a validation set of 12,436 LCC patients from 2016 to 2017 in the same database and a Chinese application set of 958 LCC patients. We developed novel TrN and TLODDS classifications for LCC patients that incorporated 7 stages with reference to the AJCC staging system. In comparison to the 8th AJCC TNM and TrN classifications, TLODDS classification demonstrated significantly better discrimination (area under the receiver operating characteristic curve, 0.650 vs. 0.656 vs. 0.661, < 0.001), better model-fitting (Akaike information criteria, 309,287 vs. 308,767 vs. 308,467), and superior net benefits. The predictive performance of the TrN and TLODDS classifications was further verified in the validation and application sets. Both the TrN and TLODDS classifications have better discriminatory ability, model-fitting, and net benefits than the existing TNM classification, and represent an alternative to the current TNM classification for LCC patients.
我们旨在探索合理的左半结肠癌（LCC）患者淋巴结分类策略。在美国监测、流行病学和最终结果数据库中，我们确定了 2010 年至 2015 年期间的 48425 例 LCC 患者。我们基于 LCC 患者的淋巴结检出值和生存分析，提出了一种新的第 8 版美国癌症联合委员会（AJCC）肿瘤/淋巴结/转移（TNM）分类的改良淋巴结（rN）分期。对数阳性淋巴结（LODDS）分期是一种数值分类策略，通过一个公式获得，该公式结合了检出和阳性淋巴结的数量。为了开发 TrN 或 TLODDS 分类，我们通过组合 T 期和 rN 期或 LODDS 分期将生存率相似的患者分为一组。在同一数据库中，我们还在 2016 年至 2017 年期间对 12436 例 LCC 患者的验证集和中国的 958 例 LCC 患者应用集进一步评估了 TrN 或 TLODDS 分类。我们开发了新的 LCC 患者 TrN 和 TLODDS 分类，其中纳入了 7 个分期，参考了 AJCC 分期系统。与第 8 版 AJCC TNM 和 TrN 分类相比，TLODDS 分类显示出显著更好的区分能力（接受者操作特征曲线下面积，0.650 对 0.656 对 0.661，<0.001），更好的模型拟合（赤池信息量准则，309287 对 308767 对 308467），以及更高的净收益。TrN 和 TLODDS 分类的预测性能在验证集和应用集中得到了进一步验证。TrN 和 TLODDS 分类都比现有的 TNM 分类具有更好的区分能力、模型拟合和净收益，是 LCC 患者目前 TNM 分类的替代方法。
