HbA1c as a predictor of pancreatic ductal adenocarcinoma

Abstract

UK adults aged 60 or over with a new onset diabetes (NODM) diagnosis and recent weight loss are now being recommended for urgent investigations to rule out possible underlying pancreatic ductal adenocarcinoma (PDAC). Differently measured HbA1c concentration levels over time with repeat sampling may be a key approach to triggering referral pathways for further investigation. The aim of this study was to investigate the relationship between HbA1c interpreted as single values and as grouped values according to clinical criteria for pre-diabetes and diabetes, and occurrence of PDAC in individuals diagnosed with NODM. Methods In this matched case-control study, 502,459 UK Biobank dataset participants were screened for incidental PDAC and HbA1c measurements at study baseline. Cox proportional hazards regression modelling with univariate and multivariate analysis generated hazard ratios (HR) for each of our chosen variables. Log-rank tests were performed for statistical significance. Receiver-operating characteristic (ROC) analysis and Youden index calculations were used to evaluate the performance of HbA1c as a test for detecting PDAC. HbA1c was not a useful standalone predictive marker for PDAC. However, newly elevated HbA1c in participants without a history of diabetes suggestive of emerging new pre-diabetes or new diabetes diagnosis was associated with a significantly higher risk of PDAC. Interpretation of HbA1c measurements benefits from knowledge of prior diabetes status when predicting risk of PDAC. The limitations of our study with single HbA1c measurements suggest that to enhance our understanding of the relationship between HbA1c and PDAC, future studies may benefit from examining repeat HbA1c measurements over time.

More information

Identifiers

ORCID for Chris Byrne: orcid.org/0000-0001-6322-7753

ORCID for Zaed Z R Hamady: orcid.org/0000-0002-4591-5226

Catalogue record

Export record