Department of General Surgery, Guangdong Provincial Key Laboratory of Precision Medicine for Gastrointestinal Tumor, Nanfang Hospital, The First School of Clinical Medicine, Southern Medical University, Guangzhou, P.R. China.
Key Laboratory for Biorheological Science and Technology of Ministry of Education (Chongqing University), Chongqing University Cancer Hospital & Chongqing Cancer Institute & Chongqing Cancer Hospital, Chongqing, P.R. China.
BJS Open. 2024 Mar 1;8(2). doi: 10.1093/bjsopen/zrae014.
This study aimed to develop and validate a model based on the collagen signature and systemic immune-inflammation index to predict prognosis in rectal cancer patients who underwent neoadjuvant treatment.
Patients with rectal cancer who had residual disease after neoadjuvant treatment at two Chinese institutions between 2010 and 2018 were selected, one used as a training cohort and the other as a validation cohort. In total, 142 fully quantitative collagen features were extracted using multiphoton imaging, and a collagen signature was generated by least absolute shrinkage and selection operator Cox regression. Nomograms were developed by multivariable Cox regression. The performance of the nomograms was assessed via calibration, discrimination and clinical usefulness. The outcomes of interest were overall survival and disease-free survival calculated at 1, 2 and 3 years.
Of 559 eligible patients, 421 were selected (238 for the training cohort and 183 for the validation cohort). The eight-collagen-features collagen signature was built and multivariable Cox analysis demonstrated that it was an independent prognostic factor of prognosis along with the systemic immune-inflammation index, lymph node status after neoadjuvant treatment stage and tumour regression grade. Then, two nomograms that included the four predictors were computed for disease-free survival and overall survival. The nomograms showed satisfactory discrimination and calibration with a C-index of 0.792 for disease-free survival and 0.788 for overall survival in the training cohort and 0.793 for disease-free survival and 0.802 for overall survival in the validation cohort. Decision curve analysis revealed that the nomograms could add more net benefit than the traditional clinical-pathological variables.
The study found that the collagen signature, systemic immune-inflammation index and nomograms were significantly associated with prognosis.
本研究旨在建立并验证一个基于胶原特征和全身免疫炎症指数的模型，以预测接受新辅助治疗的直肠癌患者的预后。
本研究选取了 2010 年至 2018 年在中国的两家机构接受新辅助治疗后仍有残留病灶的直肠癌患者，其中一家机构的数据用于训练队列，另一家用于验证队列。本研究共提取了 142 个完全定量的胶原特征，并通过最小绝对收缩和选择算子 Cox 回归生成胶原特征。通过多变量 Cox 回归建立列线图。通过校准、判别和临床实用性评估列线图的性能。本研究的主要观察终点为 1、2、3 年的总生存率和无病生存率。
在 559 例符合条件的患者中，有 421 例患者入选（训练队列 238 例，验证队列 183 例）。构建了包含 8 个胶原特征的胶原特征signature，并通过多变量 Cox 分析表明，它是预后的独立预测因素，与全身免疫炎症指数、新辅助治疗后淋巴结状态和肿瘤消退分级一起。然后，计算了包含这四个预测因子的两个用于无病生存率和总生存率的列线图。列线图在训练队列中的无病生存率和总生存率的 C 指数分别为 0.792 和 0.788，在验证队列中的无病生存率和总生存率的 C 指数分别为 0.793 和 0.802，具有较好的判别和校准能力。决策曲线分析表明，列线图比传统的临床病理变量能带来更多的净效益。
本研究发现胶原特征、全身免疫炎症指数和列线图与预后显著相关。
