Cancer Medicine (Mar 2024)

Prognostic significance of glucose‐lipid metabolic index in pancreatic cancer patients with diabetes mellitus

  • Hailiang Wang,
  • Shiye Ruan,
  • Zelong Wu,
  • Qian Yan,
  • Yubin Chen,
  • Jinwei Cui,
  • Zhongyan Zhang,
  • Shanzhou Huang,
  • Baohua Hou,
  • Chuanzhao Zhang

DOI
https://doi.org/10.1002/cam4.7108
Journal volume & issue
Vol. 13, no. 6
pp. n/a – n/a

Abstract

Read online

Abstract Background The incidence of pancreatic cancer (PC) is higher in diabetic patients due to disturbances in glucose and lipid metabolism caused by insulin resistance (IR). However, the effect of diabetes as well as IR on the prognosis of PC patients remains inconclusive. Our study aims to assess the impact of IR on the prognosis of PC patients with diabetes. Methods We conducted a retrospective analysis of 172 PC patients with diabetes in our institute from 2015 to 2021. Prognostic assessment was performed using univariate/multifactorial analysis and survival analysis. The predictive efficacy of metabolic indices was compared using receiver operator characteristic (ROC) curve analysis. Results One hundred twenty‐one of 172 patients died during follow‐up, with a median follow‐up of 477 days and a median overall survival (OS) of 270 days. Survival analysis showed a significant difference in OS by IR related parameters, which were triglyceride‐glucose index (TyG), triglyceride‐glucose index‐body mass index (TyG‐BMI), and triglyceride/high‐density lipoprotein cholesterol ratio (TG/HDL‐c). The ROC curve indicated that TyG, TyG‐BMI, and TG/HDL‐c had prognostic efficacy for PC with diabetes. We next optimized TyG‐BMI and obtained a new parameter, namely glucose‐lipid metabolism index (GLMI), and the patients were classified into GLMI low group and high group based on the calculated cutoff value. The GLMI high group had higher TyG, TyG‐BMI, TyG/HDL‐c, BMI, TG, total cholesterol (TC), TC/HDL‐c, fasting plasma glucose, CA199, and more advanced tumor stage compared to low group. Univariate and multivariate analyses showed that GLMI was an independent prognostic factor. Furthermore, the patients of GLMI high group had worse OS compared to low group and the ROC curves showed GLMI had better predictive ability than TyG and TyG‐BMI. Conclusions IR is associated with the outcome of PC patients with diabetes and higher level of IR indicates worse prognosis. GLMI has a good predictive value for PC with diabetes.

Keywords