Genetically proxied risk and protective factors for pancreatic cancer: a systematic review and meta-analysis of Mendelian randomization studies

Abstract

Background: Despite extensive observational evidence implicating lifestyle habits, metabolic factors, hormone levels, predisposing genetic conditions, medications, and other exposures, causal inference regarding pancreatic ductal adenocarcinoma (PDAC) risk factors remain contentious due to residual confounding. This Mendelian randomization (MR) meta-analysis was employed to address unresolved etiological controversies and systematically investigate causal relationships between these modifiable factors and clinical outcomes.

Methods: We conducted a systematic assessment and meta-analysis focusing on studies that employed MR methods to assess the correlation between impact factors and PDAC. We searched five databases, including PubMed, Embase, Web of Science, Scopus, and Ovid, and ultimately included 82 studies, with 292 independent findings. A meta-analysis was performed on 56 of these influencing factors.

Results: Our study found that several factors may be risk factors for PDAC, including body mass index (BMI) [odds ratio (OR): 1.23, 95% confidence interval (CI): 1.10-1.36], gut microbiota (OR: 1.25, 95% CI: 1.05-1.49), diabetes mellitus (OR: 1.07, 95% CI: 1.02-1.13), body size (OR: 1.72, 95% CI: 1.48-2.00), fasting insulin (OR: 2.23, 95% CI: 1.61-3.09), hip circumference (OR: 1.34, 95% CI: 1.11-1.61) and inflammatory bowel disease (IBD) (OR: 1.18, 95% CI: 1.04-1.34). Conversely, potential protective factors against PDAC included lycopene intake (OR: 0.87, 95% CI: 0.77-0.99) and cathepsin E levels (OR: 0.96, 95% CI: 0.94-0.99), and these factors should be further investigated.

Conclusions: This meta-analysis of MR studies identified several risk and protective factors for pancreatic cancer, with most showing low heterogeneity. However, these findings represent associations, not causality. These findings may inform clinical risk assessment while awaiting further validation. The study highlights the need for continued research focusing on diverse populations, mechanistic studies, longitudinal investigations, and translational work to advance understanding and develop effective strategies for pancreatic cancer prevention and treatment.

Keywords: Mendelian randomization (MR); Pancreatic neoplasms; meta-analysis; risk factors; systematic review.

Figure 1

PRISMA flow diagram of pancreatic cancer-related Mendelian randomization.

Figure 2

Distribution of bias risk across included studies.

Figure 3

Forest plots assessing the causal relationship between risk factors and pancreatic cancer using all OR values obtained via the IVW MR method. The results for BMI (A), diabetes (B), gut microbiota (C), body size (D), fasting insulin (E), hip circumference (F), IBD (G) and lycopene (H). BMI, body mass index; CI, confidence interval; IVW, inverse variance weighting; MR, Mendelian randomization; OR, odds ratio; BMI, body mass index; IBD, inflammatory bowel disease.

Figure 4

Forest plots assessing the causal relationship between cathepsin family and pancreatic cancer using all OR values obtained via the IVW MR method. The results of cathepsin E (A), cathepsin B (B), cathepsin F (C), cathepsin G (D), cathepsin H (E), cathepsin L2 (F), cathepsin O (G), cathepsin S (H), and cathepsin Z (I). CI, confidence interval; IVW, inverse variance weighting; MR, Mendelian randomization; OR, odds ratio.