Causal inference study of the effect of serum enzymes on breast cancer risk - Two-sample Mendelian randomization

Sep 22, 2026

Upload by Kurz

Breast cancer (BC) has the highest incidence and mortality among women worldwide, while serum enzyme associations in prior observational research remain inconsistent. This study applied two-sample Mendelian randomization to estimate causal effects of serum enzymes using 216,026 Finnish-heritage subjects as exposure and 212,402 UK Biobank subjects as outcome. Univariable and multivariable MR with multiple estimation methods, plus sensitivity checks, support a significant causal relationship for abnormal enzyme levels and show stability across models.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages8

Reading Time20 min

AccessFree to download

Abstract

Introduction

® Observational Study Causal inference study of the effect of serum enzymes on breast cancer risk Two-sample Mendelian randomization Liying Zhao, MMeda ,b ,c ,d ,e , Cheng Zeng, MDf , Xiaohui Jia, MMeda ,b ,c ,d ,e , Keliang Cui, MDa ,b ,c ,d ,e , Jiani Wang, MDf , Bing Xia, MDa ,b ,c ,d ,g ,* | Abstract
Breast cancer (BC) has the highest cancer incidence and mortality in women worldwide, and the relationship between serum enzymes and BC has been inconsistent in previous observational studies. In this paper, 2-sample Mendelian randomization (MR) is used to clarify the causal relationship between them. The serum enzymes in 216,026 study subjects from Finnish heritage subjects as the exposure data. The BC included 212,402 study subjects in United Kingdom Biobank as the result data. We performed univariable MR with 5 MR methods inverse variance weighted, MR-Egger regression, weighted median, simple mode, and weighted mode and multivariable MR to estimate the effect of serum enzymes independent of BC. Finally, we performed a series of sensitivity analyses as validation of primary MR results. The single nucleotide polymorphisms screened in this study were mostly strong effect instrumental variables. A total of 5 forest plot model results indicate that abnormal serum enzyme levels have a significant causal relationship with BC. MR analysis of different models of abnormal serum enzyme levels gave consistent direction estimates, and their slopes were relatively consistent. Serum enzyme indicators did not consider heterogeneity in the MR results of BC (Cochran Q P value > .05, I 2 < 50%) . MR-Egger’s results show that there is no effect of pleiotropy(P ≥ .05, intercepts were close to 0) . Sensitivity analysis showed that no significant change in the estimated effect value of abnormal serum enzyme levels was found, indicating the stability of the results. inverse variance weighted model results indicate that BC has no causal effect on abnormal serum enzyme levels (P value > .05) . The results of multivariate MR showed that serum enzyme indicators still have a significant direct effect on BC. Our findings suggest that a significant statistical causal association between abnormal serum enzyme levels and the risk of BC. Abnormal serum enzyme levels have important clinical implications and might be used as BC risk.
Abbreviations: ALT = alanine aminotransferase, BC = breast cancer, GWAS = genome wide association study, IVs = instrumental variables, IVW = inverse variance weighted, LDH = lactate dehydrogenase, LPL = lipoprotein lipase, MR = Mendelian randomization, SM = simple mode, SNP = single nucleotide polymorphism. | | --- | | Keywords: breast cancer, Mendelian randomization, serum enzymes, single nucleotide polymorphism | 1. Introduction Breast cancer (BC) has garnered global attention due to its high incidence worldwide. According to the Global Cancer Statistics 2020 report, issued by the International Agency for Research on Cancer in 2021, BC has surpassed lung cancer as the most frequently diagnosed cancer among women, accounting for 11.7% of all cancer cases globally. Additionally, BC remains a leading cause of cancer-related mortality in women, with a death rate of 15 per 10,000 . [1] Furthermore, the Global Cancer Observatory LZ, CZ, and XJ contributed to this article equally. JWand BX contributed to this article equally. This work has been funded by Tianjin Key Medical Discipline Construction Project (TJYXZDXK-3-003A). Thanks for the support from the National Natural Science Foundation of China (No. 81600163). The authors have no conflicts of interest to disclose. All data generated or analyzed during this study are included in this published article [and its supplementary information files]. This study focuses on Mendelian genetics and related classical inheritance research, with no involvement of human subjects, animal experiments, clinical data, or biological sample collection. No ethical ap

FAQs

What causal question does the study address?

Whether abnormal serum enzyme levels have a causal effect on breast cancer risk, rather than only an association observed in observational studies.

How were data sources defined for the two-sample Mendelian randomization?

Serum enzymes from 216,026 Finnish-heritage subjects were used as exposure data, and breast cancer cases from 212,402 UK Biobank participants were used as outcome data.

What analytical methods were used to estimate causal effects?

The study used univariable MR with inverse variance weighted, MR-Egger regression, weighted median, simple mode, and weighted mode, and also performed multivariable MR.

Sep 22, 2026

Upload by Kurz