Liang Zhong-Guo, Zhang Fan, Yu Bin-Bin, Li Ling, Qu Song, Li Ye, Guan Ying, Liang Ren-Ba, Han Lu, Zhu Xiao-Dong
Department of Radiation Oncology, Guangxi Medical University Cancer Hospital, Nanning, People's Republic of China.
Microbiome Research Centre, St George and Sutherland Clinical School, The University of New South Wales Sydney, St George Hospital, Kogarah, NSW, Australia.
Cancer Manag Res. 2020 Feb 4;12:801-812. doi: 10.2147/CMAR.S236128. eCollection 2020.
To construct a prognostic index (PI) for overall survival (OS) to stratify nasopharyngeal carcinoma (NPC) into high-risk and low-risk groups. We also applied the model to investigate the role of the addition of adjuvant chemotherapy (AC) to concurrent chemoradiotherapy (CCRT) regimens for the treatment of NPC.
A prognostic model was established based on a retrospective study of 362 patients from January 2008 to June 2011. The discriminative and calibration abilities of the model were evaluated by Harrell's concordance index (C-index), and calibration curves. Bootstrapping was used to perform for internal validation. External validation was conducted using 324 patients diagnosed with NPC from July 2011 to December 2012 at the same institution. Survival analyses were performed between CCRT-AC and CCRT alone groups for the high-risk and low-risk groups.
The primary PI comprised covariates that were associated with OS in the training cohort, including T stage, N stage, age, and plasma alkaline phosphatase (ALP). Internal and external validation showed that the discrimination of the PI for OS was significantly better than that of the 8th edition AJCC staging system. Discretization by using a fixed PI score cut-off of 407.96 determined from the training data set yielded high- and low-risk subgroups with distinct OS outcomes in the validation cohort. Adjuvant chemotherapy improved OS in high-risk patients (HR 0.620, 95% CI 0.408 to 0.941; P = 0.023) but increased the risk of distant metastasis (HR, 4.222, 95% CI, 0.959 to 18.585; P = 0.038) in low-risk patients.
The proposed prognostic model achieved good prediction and calibration of OS for patients with NPC. The addition of adjuvant chemotherapy might be a double-edged sword, bringing survival benefit to high-risk patients but greater risk of distant metastasis to low-risk patients.
构建一个用于总生存期（OS）的预后指数（PI），以便将鼻咽癌（NPC）患者分为高风险组和低风险组。我们还应用该模型来研究在同期放化疗（CCRT）方案中加入辅助化疗（AC）对鼻咽癌治疗的作用。
基于对2008年1月至2011年6月期间362例患者的回顾性研究建立预后模型。通过Harrell一致性指数（C指数）和校准曲线评估模型的判别能力和校准能力。采用自助法进行内部验证。使用同一机构2011年7月至2012年12月诊断为NPC的324例患者进行外部验证。对高风险组和低风险组的CCRT-AC组和单纯CCRT组进行生存分析。
主要预后指数包含与训练队列中总生存期相关的协变量，包括T分期、N分期、年龄和血浆碱性磷酸酶（ALP）。内部和外部验证表明，该预后指数对总生存期的判别能力明显优于第8版美国癌症联合委员会（AJCC）分期系统。根据训练数据集确定的固定预后指数得分临界值407.96进行离散化处理，在验证队列中产生了总生存期结局不同的高风险和低风险亚组。辅助化疗改善了高风险患者的总生存期（风险比[HR]0.620，95%置信区间[CI]0.408至0.941；P = 0.023），但增加了低风险患者远处转移的风险（HR 4.222，95% CI 0.959至18.585；P = 0.038）。
所提出的预后模型对鼻咽癌患者的总生存期实现了良好的预测和校准。加入辅助化疗可能是一把双刃剑，给高风险患者带来生存益处，但给低风险患者带来更大的远处转移风险。
