Aim: To cross-sectionally and longitudinally investigate the association between tumor markers (Cancer embryonic antigen (CEA) Carbohydrate antigen 19-9 (CA19-9)) and malignancies in type 2 diabetes patients without evidence of malignancy.
Materials and methods: The study included 707 patients admitted for the treatment of diabetes from 1 August 2010 to 1 September 2018. Serum CEA and CA19-9 levels were measured for screening of malignancies at admission. Abdominal ultrasonography, computed tomography, and endoscopy were performed for close examination. The percentage of patients diagnosed with malignancy was calculated, and among those without malignancy, the incidence of malignancies was examined after discharge.
Results: A total of 26 patients (3.7%) were newly diagnosed with malignancy during hospitalization. The optimal cut-off value of CEA and CA19-9 by receiver operating characteristic analysis was 5.0 ng/mL and 75 U/mL, and their positive predictive values (PPV) were 8.7% and 22.5%, respectively. The addition of CA19-9 to age, smoking status, body mass index, and glycated hemoglobin significantly improved classification performance for malignancy using net reclassification improvement (0.682, 95% CI 0.256-1.107) and integrated discrimination improvement (0.150, 95% CI 0.007-0.294). Among 681 patients without malignancies during hospitalization, 30 patients (4.4%) developed malignancies during an average follow-up of 3.9 years. CA19-9 (hazard ratio: 1.005, 95% CI: 1.003-1.008) was associated with the development of malignancies.
Conclusions: PPV of serum CEA and CA19-9 for detecting malignancy was high in type 2 diabetes patients with poor glycemic control. Measuring CA19-9 was found to be valuable to cross-sectionally and longitudinally detect malignancies.
Supplementary information: The online version contains supplementary material available at 10.1007/s13340-022-00594-x.
Keywords: Glycemic control; Malignancy; Tumor marker; Type 2 diabetes.
© The Japan Diabetes Society 2022.