Evidence Hub

Distinct prognostic impact of diabetes subtypes, metformin use, and glycemic control in pancreatic cancer: a systematic review and meta-analysis of cohort studies

Li G, Peng Z, Wang Y, Zhu Z, Huang S, Yu J, Huang H

Frontiers in oncology, 2026

Abstract

Background: The prognostic significance of glycemic status in pancreatic ductal adenocarcinoma (PDAC) remains controversial, largely due to heterogeneity in the exposure definitions (new-onset vs. long-standing diabetes, metformin use, and single fasting glucose measurement). This meta-analysis aimed to clarify the association of four distinct glycemic exposures with overall survival (OS) in patients with PDAC.

Methods: We systematically searched Web of Science, PubMed, and EMBASE from 1 January 2010 to 1 June 2026. Cohort studies reporting hazard ratios (HRs) for OS were included and categorized into four subgroups: 1) new-onset diabetes (NOD) vs. non-diabetes; 2) long-standing diabetes (LSD) vs. non-diabetes; 3) metformin use vs. non-use; and 4) hyperglycemia (elevated preoperative fasting glucose or poor glycemic control) vs. normoglycemia. Random-effects meta-analyses were performed. Heterogeneity was assessed using I 2, and publication bias was evaluated via funnel plots and the trim-and-fill method.

Results: A total of 21 studies (six for NOD, four for LSD, six for metformin, and eight for hyperglycemia) were included. NOD was significantly associated with worse OS (pooled HR = 1.33, 95%CI = 1.13-1.57, I 2 = 72.5%). LSD showed no significant association (HR = 1.21, 95%CI = 0.98-1.49, I 2 = 70.1%). Metformin use was associated with a 33% reduction in mortality (HR = 0.67, 95%CI = 0.53-0.84, I 2 = 70.5%). Preoperative hyperglycemia or poor glycemic control consistently…

How strong is the evidence?

Evidence Hub reads this as a meta-analysis, done in people. Its evidence level is 1 of 5 (Highest · pooled evidence). Pooled evidence from several studies sits at the top of the evidence pyramid.

How the five levels of evidence work. The level is assigned automatically from the abstract and can be wrong, so check the paper.

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free