Department of Ultrasound Imaging, Renmin Hospital of Wuhan University, Wuhan, China; 430061.
Department of Gastrointestinal Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China; 430022.
Int J Med Sci. 2022 May 29;19(6):1003-1012. doi: 10.7150/ijms.71971. eCollection 2022.
We aimed to determine if lactate dehydrogenase to albumin ratio (LAR) might play a prognostic role for patients with operable colorectal cancer (CRC). 1334 operable CRC patients in Wuhan Union Hospital Between July 2013 and September 2017 were enrolled in this study and were randomly appointed them into training (n=954) and validation (n=380) sets. The relationship between LAR and overall survival (OS) and disease-free survival (DFS) were determined by restricted cubic splines (RCS) with Cox regression models. LAR was then divided into three categories based on the RCS and compared to the well-known TNM stage system. Finally, survival nomograms were developed by compounding the LAR and other clinical factors. Baseline LAR values and the all-cause mortality were U shaped, which slowly decreased until around 4.50 and then started to increase rapidly when the LAR ranged from 4.50-6.68 and then became flat thereafter (P for non-linearity <0.001). LAR was superior to TNM stage for OS as well as DFS and LAR plus TNM stage could add more net benefit than clinical model alone. Moreover, the survival nomograms based on LAR achieved great predictive ability for OS and DFS in operable CRC patients. : LAR could be served as a reliable prognostic factor for OS as well as DFS, with more accurate prognostic prediction than current TNM stage for patients with operable CRC.
我们旨在确定乳酸脱氢酶与白蛋白比值(LAR)是否可能对可手术结直肠癌(CRC)患者的预后起作用。 本研究纳入了 2013 年 7 月至 2017 年 9 月期间武汉协和医院的 1334 例可手术 CRC 患者,并将其随机分为训练组(n=954)和验证组(n=380)。 通过 Cox 回归模型,采用受限立方样条(RCS)确定 LAR 与总生存期(OS)和无病生存期(DFS)之间的关系。 然后,根据 RCS 将 LAR 分为三个类别,并将其与著名的 TNM 分期系统进行比较。 最后,通过组合 LAR 和其他临床因素,制定了生存列线图。 基线 LAR 值和全因死亡率呈 U 形,在 LAR 范围为 4.50-6.68 时,LAR 缓慢下降至 4.50 左右,然后迅速上升,之后趋于平坦(P<0.001)。 LAR 在 OS 和 DFS 方面优于 TNM 分期,LAR 加 TNM 分期比临床模型单独使用能带来更多的净获益。 此外,基于 LAR 的生存列线图在可手术 CRC 患者中具有很好的 OS 和 DFS 预测能力。 LAR 可作为 OS 和 DFS 的可靠预后因素,在预测可手术 CRC 患者的预后方面比当前的 TNM 分期更准确。