causal mediation effects in single case experimental designs

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

Estimating and Testing Causal Mediation Effects in Single-Case Experimental Designs Using State-Space Modeling

Affiliations.

• 1 9167Bielefeld University, Bielefeld, Germany.
• 2 1108Presbyterian Healthcare Services, Albuquerque, NM, USA.
• 3 8125Utrecht University, Utrecht, Netherlands.
• 4 26657KU Leuven, Leuven, Belgium.
• PMID: 35245983
• DOI: 10.1177/01632787211067533

In this article, we present single-case causal mediation analysis as the application of causal mediation analysis to data collected within a single-case experiment. This method combines the focus on the individual with the focus on mechanisms of change, rendering it a promising approach for both mediation and single-case researchers. For this purpose, we propose a new method based on time-discrete state-space modeling to estimate the direct and indirect treatment effects. We demonstrate how to estimate the model for a single-case experiment on stress and craving in a routine alcohol consumer before and after an imposed period of abstinence. Furthermore, we present a simulation study that examines the estimation and testing of the standardized indirect effect. All parameters used to generate the data were recovered with acceptable precision. We use maximum likelihood and permutation procedures to calculate p -values and standard errors of the parameters estimates. The new method is promising for testing mediated effects in single-case experimental designs. We further discuss limitations of the new method with respect to causal inference, as well as more technical concerns, such as the choice of the time lags between the measurements.

Keywords: experimental designs; mediator variables; permutation tests; person-centered approach; time series.

PubMed Disclaimer

Similar articles

• Causal mediation effects in single case experimental designs. Valente MJ, Rijnhart JJM, Miočević M. Valente MJ, et al. Psychol Methods. 2023 Apr;28(2):488-506. doi: 10.1037/met0000497. Epub 2022 May 12. Psychol Methods. 2023. PMID: 35549318 Free PMC article.
• Causal mediation analysis with mediator values below an assay limit. Chernofsky A, Bosch RJ, Lok JJ. Chernofsky A, et al. Stat Med. 2024 May 30;43(12):2299-2313. doi: 10.1002/sim.10065. Epub 2024 Mar 31. Stat Med. 2024. PMID: 38556761
• Selecting a Within- or Between-Subject Design for Mediation: Validity, Causality, and Statistical Power. Montoya AK. Montoya AK. Multivariate Behav Res. 2023 May-Jun;58(3):616-636. doi: 10.1080/00273171.2022.2077287. Epub 2022 Jun 9. Multivariate Behav Res. 2023. PMID: 35679239
• A brief primer on conducting regression-based causal mediation analysis. Li Y, Yoshida K, Kaufman JS, Mathur MB. Li Y, et al. Psychol Trauma. 2023 Sep;15(6):930-938. doi: 10.1037/tra0001421. Epub 2023 Jan 26. Psychol Trauma. 2023. PMID: 36701540 Free PMC article. Review.
• Advances in mediation analysis: a survey and synthesis of new developments. Preacher KJ. Preacher KJ. Annu Rev Psychol. 2015 Jan 3;66:825-52. doi: 10.1146/annurev-psych-010814-015258. Epub 2014 Aug 21. Annu Rev Psychol. 2015. PMID: 25148853 Review.
• Causal Mediation Analysis in Single Case Experimental Designs: Introduction to the Special Issue. Miočević M, Moeyaert M, Mayer A, Montoya AK. Miočević M, et al. Eval Health Prof. 2022 Mar;45(1):3-7. doi: 10.1177/01632787211073194. Epub 2022 Feb 3. Eval Health Prof. 2022. PMID: 35112918 Free PMC article.
• Search in MeSH

LinkOut - more resources

Full text sources.

• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Type your tag names separated by a space and hit enter

Causal mediation effects in single case experimental designs. Psychol Methods . 2023 Apr; 28(2):488-506. PM

Single case experimental designs (SCEDs) are used to test treatment effects in a wide range of fields and consist of repeated measurements for a single case throughout one or more baseline phases and throughout one or more treatment phases. Recently, mediation analysis has been applied to SCEDs. Mediation analysis decomposes the total treatment-outcome effect into a direct and indirect effect, and therefore aims to unravel the causal processes underlying treatment-outcome effects. The most recent methodological advancement for mediation analysis is the development of causal mediation analysis methodology which clarifies the necessary causal assumptions for mediation analysis. The goal of this article is to derive the causal mediation effects and corresponding standard errors based on piecewise linear regression models for the mediator and outcome and to evaluate the performance of these regression estimators and standard errors. Whereas previous studies estimated the direct and indirect effects as either the change in level or change in trend, we showed that the causal direct and indirect effects incorporate both the change in level and change in trend. Based on our simulation study we showed that for the causal indirect effects, Monte Carlo confidence intervals provided accurate (i.e., p = .05) Type I error rates and higher statistical power than normal theory confidence intervals. For the causal direct effects and total effect, normal theory confidence intervals provided accurate Type I error rates and higher statistical power than the Monte Carlo confidence intervals. Limitations and future directions are discussed. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Authors +Show Affiliations

Pub type(s).

• PMC Free PDF
• Rijnhart JJM
• Research Design
• Computer Simulation
• Linear Models
• Monte Carlo Method
• Models, Statistical

Recent advances in statistical mediation analysis with N = 1

Scientists and practitioners are aware that the effectiveness of a treatment varies between individuals. When studying effects on a group level, there is an implicit assumption that the treatment works the same way for all group members. Without examining effects at the level of the individual, we have no way of knowing the causal mechanism through which a treatment works (or does not work) for them, and we might reach the wrong conclusions if we try to generalize findings obtained from a group to an individual. The field of personalized medicine solves these issues by collecting detailed biological and health information from each patient and evaluating individual-specific treatment effects (Schleidgen, Klingler, Bertram, Rogowski, & Marckmann, 2013). 

In psychology, moderation analysis is often the method of choice for discovering conditions in which the treatment leads to desired changes in the outcome. This top-down approach for evaluating the characteristics of individuals for which the treatment is effective is not ideal because accounting for all relevant moderators leads to extremely complex models. A bottom-up approach might be preferable, and fortunately there is a vast literature on methods for evaluating the effectiveness of treatments for a single individual (see e.g. Manolov & Moeyaert, 2017).

Methods for Single Case Experimental Designs (SCEDs)

Single Case Experimental Designs (SCEDs) are research designs in which a single unit (participant) is measured repeatedly in several levels of the treatment (Onghena, 2005). SCEDs are akin to within-subjects experiments, but the analysis is performed at the level of the individual instead of the level of the group. Unlike most case studies and observational time series research, SCEDs are defined by the existence of an experimental manipulation (Onghena, 2005). Whereas SCEDs allow for testing hypotheses about changes in the outcome of interest caused by the treatment, almost all statistical methods for SCEDs are designed for univariate analyses. In order to answer important questions in clinical settings, statistical methodology for SCEDs ought to be extended to mediation analysis (Maric, Wiers, & Prins, 2012).

Causal mediation analysis with group-level data

Randomization of participants to treatment is often considered necessary in order to claim that changes in the outcome are due to treatment. In mediation analysis, it is possible to randomize participants to values of the independent variable, but usually not to values of the mediator. This obstacle to making causal claims about the effect of the mediator on the outcome has led to numerous methodological developments in the field of causal inference in mediation analysis. Some of these developments focus on necessary assumptions and procedures for causal interpretations of the indirect effect (Coffman & Zhong, 2012; Jo, Stuart, MacKinnon, & Vinokur, 2011; Mayer, Thoemmes, Rose, Steyer, & West, 2014; Valeri & VanderWeele, 2013), and others on quantifying the sensitivity of the observed indirect effect to unmeasured confounders (Cox, Kisbu-Sakarya, Miočević, & MacKinnon, 2013; Imai, Keele, & Tingley, 2010; VanderWeele, 2010).

In a clinical setting, it is often of interest to evaluate whether a treatment had a causal indirect effect on the outcome regardless of whether the analysis is conducted at the level of the group or the individual. Whereas there have been several proposed methods for mediation analysis in SCEDs, ranging from visual analysis supplemented by ipsative z-scores (Gaynor & Harris, 2008), to the SCED version of the joint significance test (Geuke, Maric, Miočević, Wolters, & de Haan, 2019) and Bayesian methods for parameter estimation and informative hypothesis testing for the mediated effect in SCEDs (Miočević, Klaassen, Geuke, Moeyaert, & Maric, 2019), none of the currently available methods allow for causal interpretations of the indirect effect.

Workshop on ‘Single Subject Causal Mediation Analysis’

Examining causal mechanisms through which treatments achieve effects is important for identifying the most potent program components and for conducting more evidence-based personalized (mental) health care (Maric, Wiers, & Prins, 2012). Coming up with methods for single subject causal mediation analysis was facilitated by bringing together clinical psychologists who collect repeated measures data in N = 1 trials (e.g., in therapy) and statisticians for social sciences working on mediation analysis, single subject data analysis, and causal inference. In August 2019, we organized a five-day meeting at the Lorentz Center in Leiden, Netherlands. The goal of the meeting was to brainstorm and test out ways to upgrade existing statistical methods for SCEDs with new developments in causal mediation analysis to allow for causal interpretations of the indirect effect.

The format of the meeting was a combination of a conference, workshop and hackathon. The number of invited talks and poster presentations was limited in order to allow for at least one hour of work in small groups each day. The groups were composed of four-five researchers from different fields. Each interdisciplinary group was tasked with coming up with ideas and examples of methods for causal mediation analysis that they presented on the last day of the workshop.

The diverse areas of expertise of the meeting attendees led to rich discussions on topics ranging from the required frequency of measurement of the mediator and outcome to potential for randomization in a mediation model and necessary assumptions regarding other covariates in the model and timing of effects. We aim to disseminate the ideas, proposed methods and considerations that emerged during the meeting through publications and presentations at the meeting of the European Methodology Association in Valencia, July 2020. This research is still in progress, and the meeting in Leiden was a productive step in the formation of a community of researchers interested in advancing methods for causal mediation analysis with N = 1.

I would like to thank the Lorentz center for hosting the meeting on Single Subject Causal Mediation Analysis, my co-organizers Marija Maric from University of Amsterdam, Mariola Moeyaert from University of Albany, New York, and Axel Mayer from RWTH Aachen University, and of course, all researchers who attended and contributed their ideas and who are continuing to work on new methods for causal mediation analysis in SCEDs.

Coffman, D. L., & Zhong, W. (2012). Assessing mediation using marginal structural models in the presence of confounding and moderation. Psychological Methods, 17 , 642–664.

Cox, M. G., Kisbu-Sakarya, Y., Miočević, M., & MacKinnon, D. P. (2013). Sensitivity plots for confounder bias in the single mediator model.  Evaluation Review ,  37 (5), 405-431.

Gaynor, S. T., & Harris, A. (2008). Single-participant assessment of treatment mediators: Strategy description and examples from a behavioral activation intervention for depressed adolescents. Behavior Modification, 32 (3), 372-402.

Geuke, G., Maric, M., Miočević, M., Wolters, L. H., & de Haan, E. (2019). Testing          mediators of youth intervention outcomes using Single-Case Experimental        Designs (SCEDs). New Directions for Child and Adolescent Development, 167 , 39-64.

Imai, K., Keele, L., & Tingley, D. (2010). A general approach to causal mediation            analysis. Psychological Methods, 15 , 309–334.

Jo, B., Stuart, E. A., MacKinnon, D. P., & Vinokur, A. D. (2011). The use of propensity   scores in mediation analysis. Multivariate Behavioral Research, 46 , 425–452.

Manolov, R., Moeyaert, M. (2017). Recommendations for choosing single-case data         analytical techniques. Behavior Therapy, 48 (1), 97-114.

Maric, M., Wiers, R. W., & Prins, P. J. (2012). Ten ways to improve the use of statistical mediation analysis in the practice of child and adolescent treatment       research.  Clinical child and family psychology review ,  15 (3), 177-191.

Mayer, A., Thoemmes, F., Rose, N., Steyer, R., & West, S. G. (2014). Theory and analysis of total, direct, and indirect causal effects.  Multivariate Behavioral Research ,  49 (5), 425-442.

Miočević, M., Klaassen, F., Geuke, G., Moeyaert, M. & Maric, M. (under review). Using             Bayesian Methods to test Mediators of Intervention Outcomes in Single Case            Experimental Designs (SCEDs).

Onghena, P. (2005). Single‐Case Designs.  Encyclopedia of Statistics in Behavioral           Science , 4, 1850–1854. Schleidgen, S., Klingler, C., Bertram, T., Rogowski, W. H., & Marckmann, G. (2013). What is personalized medicine: sharpening a vague term based on a systematic literature review.  BMC Medical Ethics ,  14 (1), 55.

Valeri, L., & VanderWeele, T. J. (2013). Mediation analysis allowing for exposure–          mediator interactions and causal interpretation: Theoretical assumptions and             implementation with SAS and SPSS macros. Psychological Methods, 18 , 137–150.

VanderWeele, T. J. (2010). Bias formulas for sensitivity analysis for direct and indirect effects. Epidemiology, 21 , 540–551.

Recommended Reading

• Find out more about the Amsterdam University Medical Centres

Causal Mediation Effects in Single Case Experimental Designs

• Epidemiology and Data Science (VUmc)
• APH - Methodology (VUmc)

Research output : Contribution to journal › Article › Academic › peer-review

• Causal inference
• Longitudinal data
• Single-case experimental design

Access to Document

• https://doi.org/10.1037/met0000497

Other files and links

• Link to publication in Scopus

T1 - Causal Mediation Effects in Single Case Experimental Designs

AU - Valente, Matthew J.

AU - Rijnhart, Judith J. M.

AU - Mioĉević, Milica

N1 - Funding Information: This research was supported in part by the National Institute on Drug Abuse (R37DA09757). Portions of this research were presented at the 2019 Single Subject Causal Mediation Analysis Workshop, Lorentz Center, Leiden, the Netherlands.We thank the Lorentz Center for hosting the workshop where this idea originated. We also thank Dave MacKinnon and Mariola Moeyaert for helpful feedback related to this research Publisher Copyright: © 2022 American Psychological Association

N2 - Single case experimental designs (SCEDs) are used to test treatment effects in a wide range of fields and consist of repeated measurements for a single case throughout one or more baseline phases and throughout one or more treatment phases. Recently, mediation analysis has been applied to SCEDs. Mediation analysis decomposes the total treatment-outcome effect into a direct and indirect effect, and therefore aims to unravel the causal processes underlying treatment-outcome effects. The most recent methodological advancement for mediation analysis is the development of causal mediation analysis methodology which clarifies the necessary causal assumptions for mediation analysis. The goal of this article is to derive the causal mediation effects and corresponding standard errors based on piecewise linear regression models for the mediator and outcome and to evaluate the performance of these regression estimators and standard errors. Whereas previous studies estimated the direct and indirect effects as either the change in level or change in trend, we showed that the causal direct and indirect effects incorporate both the change in level and change in trend. Based on our simulation study we showed that for the causal indirect effects, Monte Carlo confidence intervals provided accurate (i.e., p =.05) Type I error rates and higher statistical power than normal theory confidence intervals. For the causal direct effects and total effect, normal theory confidence intervals provided accurate Type I error rates and higher statistical power than the Monte Carlo confidence intervals. Limitations and future directions are discussed

AB - Single case experimental designs (SCEDs) are used to test treatment effects in a wide range of fields and consist of repeated measurements for a single case throughout one or more baseline phases and throughout one or more treatment phases. Recently, mediation analysis has been applied to SCEDs. Mediation analysis decomposes the total treatment-outcome effect into a direct and indirect effect, and therefore aims to unravel the causal processes underlying treatment-outcome effects. The most recent methodological advancement for mediation analysis is the development of causal mediation analysis methodology which clarifies the necessary causal assumptions for mediation analysis. The goal of this article is to derive the causal mediation effects and corresponding standard errors based on piecewise linear regression models for the mediator and outcome and to evaluate the performance of these regression estimators and standard errors. Whereas previous studies estimated the direct and indirect effects as either the change in level or change in trend, we showed that the causal direct and indirect effects incorporate both the change in level and change in trend. Based on our simulation study we showed that for the causal indirect effects, Monte Carlo confidence intervals provided accurate (i.e., p =.05) Type I error rates and higher statistical power than normal theory confidence intervals. For the causal direct effects and total effect, normal theory confidence intervals provided accurate Type I error rates and higher statistical power than the Monte Carlo confidence intervals. Limitations and future directions are discussed

KW - Causal inference

KW - Longitudinal data

KW - Mediation

KW - Single-case experimental design

UR - http://www.scopus.com/inward/record.url?scp=85130793295&partnerID=8YFLogxK

U2 - https://doi.org/10.1037/met0000497

DO - https://doi.org/10.1037/met0000497

M3 - Article

C2 - 35549318

SN - 1082-989X

JO - Psychological methods

JF - Psychological methods

• DOI: 10.1177/01632787211073194
• Corpus ID: 246487260

Causal Mediation Analysis in Single Case Experimental Designs: Introduction to the Special Issue

• Milica Miočević , Mariola Moeyaert , +1 author A. Montoya
• Published in Evaluation & the Health… 3 February 2022

4 Citations

Exploring human biology with n-of-1 clinical trials, ambivalent attitudes toward life and death and suicide ideation among adolescents-the mediating role of depression., assessing the criterion validity of curriculum-based measures for children with intellectual disabilities, estrés en deportistas universitarios: revisión de la eead en jóvenes mexicanos, 49 references, using bayesian methods to test mediators of intervention outcomes in single-case experimental designs, estimating and testing causal mediation effects in single-case experimental designs using state-space modeling, within-subject mediation analysis in ab/ba crossover designs, sensitivity plots for confounder bias in the single mediator model, the influence of the design matrix on treatment effect estimates in the quantitative analyses of single-subject experimental design research, theory and analysis of total, direct, and indirect causal effects, the measurement of the mediator and its influence on statistical mediation conclusions., identification, inference and sensitivity analysis for causal mediation effects, methods for modeling autocorrelation and handling missing data in mediation analysis in single case experimental designs (sceds), a state space modeling approach to mediation analysis, related papers.

Showing 1 through 3 of 0 Related Papers

Estimating and Testing Causal Mediation Effects in Single-Case Experimental Designs Using State-Space Modeling

• Evaluation & the Health Professions 45(1):1632787211067533
• 45(1):1632787211067533

• Maastricht University
• This person is not on ResearchGate, or hasn't claimed this research yet.

• Utrecht University

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations

No full-text available

To read the full-text of this research, you can request a copy directly from the authors.

• EVAL HEALTH PROF

• Amanda K. Montoya

• STRUCT EQU MODELING

• Lisa A. Marsch
• Russell A. Poldrack

• BEHAV RES METHODS

• Aaron Jason Fisher
• John D. Megdalia

• W. Jack Rejeski
• Meredith R Petrie
• Lise Gauvin

• Edward G. Carr

• Mark Wolery

• Kristopher J. Preacher

• J STAT SOFTW
• Yves Rosseel
• Aaron B. Taylor

• Michael Myrtek
• Kurt Pawlik

• J CLIN CHILD ADOLESC

• Ruud Wetzels

• PSYCHOL SCI

• Leif D. Nelson
• Uri Simonsohn
• William R. Shadish

• Kosuke Imai

• Chu Kim-Prieto

• Joshua M. Smyth

• Chondra M. Lockwood

• Angelina Davis

• David A. Cole
• Scott E. Maxwell
• Eugene S. Edgington
• Roos Rodenburg
• BEHAV MODIF

• Paul Verkempynck

• Alan E. Kazdin

• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

Europe PMC requires Javascript to function effectively.

Either your web browser doesn't support Javascript or it is currently turned off. In the latter case, please turn on Javascript support in your web browser and reload this page.

ORIGINAL RESEARCH article

Pharmacological targets of sglt2 inhibition on prostate cancer mediated by circulating metabolites: a drug-target mendelian randomization study.

• 1 Department of Breast Surgery, The First Affiliated Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen, China
• 2 School of Medicine, Xiamen University, Xiamen, China
• 3 Department of Hematology, Xiangya Hospital, Xiangya School of Medicine, Central South University, Changsha, China
• 4 The First Clinical College of Guangzhou Medical University, Guangzhou Medical University, Guangzhou, China
• 5 The School of Clinical Medicine, Fujian Medical University, Fuzhou, China

Background: The relationship between sodium-glucose cotransporter 2 (SGLT2) inhibitors and prostate cancer is still unknown. Although these inhibitors can influence tumor glycolysis, the underlying mechanism requires further exploration.

Methods: A two-sample two-step MR was used to determine 1) causal effects of SGLT2 inhibition on prostate cancer; 2) causal effects of 1,400 circulating metabolites or metabolite ratios on prostate cancer; and 3) mediation effects of these circulating metabolites. Genetic proxies for SGLT2 inhibition were identified as variants in the SLC5A2 gene and glycated hemoglobin level (HbA1c). Additionally, positive control analysis on type 2 diabetes mellitus (T2DM) was conducted to test the selection of genetic proxies. Phenome Wide Association Study (PheWAS) and MR-PheWAS analysis were used to explore potential treatable diseases and adverse outcomes of SGLT2 inhibitors.

Results: Genetically predicted SGLT2 inhibition (per 1 SD decrement in HbA1c) was associated with reduced risk of T2DM [odds ratio (OR) = 0.66 (95% CI 0.53, 0.82), P = 1.57 × 10 −4 ]; prostate cancer [0.34 (0.23, 0.49), P = 2.21 × 10 −8 ] and prostate-specific antigen [0.26 (0.08, 0.81), P = 2.07 × 10 −2 ]. The effect of SGLT2 inhibition on prostate cancer was mediated by uridine level, with a mediated proportion of 9.34% of the total effect. In MR-PheWAS, 65 traits were found to be associated with SLGT2 inhibitors ( P < 1.78 × 10 −5 ), and among them, 13 were related to diabetes.

Conclusion: Our study suggested that SGLT2 inhibition could lower prostate cancer risk through uridine mediation. More mechanistic and clinical research is necessary to explore how uridine mediates the link between SGLT2 inhibition and prostate cancer.

Introduction

Sodium-glucose cotransporter 2 (SGLT2) inhibitors are a type of oral antidiabetic medication that directly removes glucose of the systemic circulation ( Heerspink et al., 2016 ; Brown et al., 2021 ). Extensive clinical trials have offered strong evidence supporting the positive impact of these interventions on reducing glucose levels, as well as enhancing cardiovascular, heart failure, and renal outcomes ( Neal et al., 2017 ; Wiviott et al., 2019 ; Cannon et al., 2020 ; Bhatt et al., 2021 ). Additionally, emerging evidence indicated their potential of reducing the risk of cancers. A meta-analysis revealed that SGLT2 inhibitors provided a protective effect against cancers compared to a placebo group, with a notable effectiveness observed for dapaglifozin and ertuglifozin ( Benedetti et al., 2022 ). SGLT2 inhibitors were considered as potential anticancer drugs that could slow down the tumor growth of carcinoma expressing SGLT2 ( Koepsell, 2017 ). The expression of SGLT2 has been confirmed at the mRNA level and by immunohistochemistry in prostate cancer ( Scafoglio et al., 2015 ). However, the associations between SGLT2 inhibitors and prostate cancer are controversial and the underlying metabolic mechanism still need further exploring.

Uridine, a pyrimidine nucleoside, is characterized by its high abundance and solubility in the bloodstream. The bioavailability of uridine is crucial for the synthesis of RNA and cell metabolism. For RNA synthesis, uridine 5′-triphosphate (UTP) participates in the process, which is produced by adding a third phosphate group to uridine. For cell metabolism, uracil and beta-alanine engage in the tricarboxylic acid (TCA) cycle, which are the products of the catabolism of uridine ( Connolly and Duley, 1999 ). A study employed nutrient-sensitized genome-wide genetic screens and the Profiling Relative Inhibition Simultaneously in Mixtures (PRISM) growth assay on 482 cancer cell lines, to identify cells capable of tolerating complete glucose deprivation and to elucidate the underlying mechanisms ( Skinner et al., 2023 ). The research revealed a significant upsurge in uridine levels at various stages of prostate cancer, particularly in tissue samples. These findings suggested that cancer cells can harness uridine and its metabolic byproducts for glycolysis, thereby bolstering their growth metabolism, and this capability appears to be widespread. It also suggested that uridine may play a role in the pathogenesis of prostate cancer. As a SGLT2 inhibitor, empagliflozin has been reported to have the potential to upregulate uridine, thereby potentially protecting the kidney ( Bangarbale et al., 2022 ). However, the influence of SGLT2 inhibitors on uridine metabolism remains controversial.

Mendelian randomization (MR) is a relatively vigorous and widely used approach, which uses single nucleotide polymorphisms (SNPs) as genetic instruments to examine the causal association. In this study, genetic colocalization was employed to selected MR instrumental variables that were correlated with drug target mRNA expression ( Chen J. et al., 2023 ; Bakker et al., 2023 ). The method can be applied to investigate the biological mechanisms of how SGLT2 inhibition affects type 2 diabetes mellitus (T2DM) and prostate cancer ( Bouras et al., 2022 ; Lin et al., 2022 ). In earlier studies on cardiovascular-related diseases, similar methods have been employed to obtain instrumental variables for exposure and use metabolites as mediators ( Xu et al., 2022 ; Li et al., 2023 ). Although previous studies have examined the casual association between inflammation factors, lipids, amino acid alterations and prostate cancer, these studies were limited in sample size and the range of metabolites investigated. Therefore, more comprehensive research is needed to explore the relationship between the metabolome and prostate cancer.

In order to investigate the causal link between SGLT2 inhibitors and prostate cancer risk via circulating metabolites or metabolite ratios, we conducted a two-sample, two-step MR study to delve into the hypothetical metabolic pathways that might interconnect the pharmacological action of SGLT2 inhibitors with the development of prostate cancer. The finding would provide insights into the uridine metabolism linking the effect of SGLT2 inhibitors with prostate cancer.

Materials and methods

Study design.

The research aims to ascertain if metabolites or metabolite ratios (mediators) play a causal function in facilitating the impact of SGLT2 inhibition on prostate cancer. In this study, we performed a two-sample two-step MR design ( Figure 1 ): Initially, genetic variants representing the effects of SGLT2 inhibition were chosen, followed by the estimation of the causal effects of SGLT2 inhibition on T2DM. Based on that, we further selected prostate cancer as our outcome, and investigating the association between SGLT2 inhibitors and prostate cancer. Then we explored changes in metabolites associated with prostate cancer. In final, we investigated the casual impact of SGLT2 inhibition on prostate cancer related metabolites and calculated mediation effect ( Figure 1 ). The MR analyses fulfilled three core assumptions. First, genetic variants are strongly linked to the outcome. Second, genetic variants are not influenced by other confounding factors. Third, the impact of genetic variants on the outcome is solely driven by exposure. Our study was conducted following the Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization (STROBE-MR) guidelines ( Skrivankova et al., 2021 ).

Figure 1 . Overview of the study design. (A) The flowchart of evaluating the effects of circulating metabolites in mediating the effect of SGLT2 inhibition on prostate cancer. (B) The framework of the two-step MR. (C) Diagram of the two steps of MR models. HbA1c, glycated hemoglobin; SNP, single nucleotide polymorphism.

Data source and genetic instruments for SGLT2 inhibition

The selection of genetic variants serving as proxies for SGLT2 inhibition involved four steps ( Figure 1A ). 1) Gather GWAS data from the Genotype-Tissue Expression (GTEx) ( GTEx Consortium, 2020 ) and eQTLGen Consortium ( Võsa et al., 2021 ) to identify genetic variants linked to the mRNA expression of SLC5A2 ( Supplementary Table S1 ). 2) Determine the correlation between each SLC5A2 variant and HbA1c levels, which reflect the glucose-lowering impact through SGLT2 inhibition, and identify variants strongly linked to HbA1c ( P < 1 × 10 −4 ). The GWAS data for HbA1c level was obtained from UK Biobank (n = 344,182) ( Supplementary Table S1 ). 3) Investigate if SLC5A2 gene and HbA1c level share the same causal variant through colocalization analysis. Colocalization aims to estimate the posterior probability that SLC5A2 expression and circulating HbA1c levels share the same causal variant in specific region ( Zuber et al., 2022 ). Evidence for colocalization (a posterior probability >70%) was defined as significant ( Lin et al., 2023 ). 4) Conduct a standard clumping process, using the 1,000 Genomes European reference panel to select genetic variants in linkage disequilibrium (LD) with an r 2 < 0.8 within 1,000 kb. 5) Extract the instrumental variables from the GWAS of HbA1c and obtain the inverse beta values to represent the SGLT2 inhibition.

Data source and genetic instruments for circulating metabolites

In this study, we systematically obtained GWAS data of 1,091 metabolites and 309 metabolite ratios from 8,299 individuals from the Canadian Longitudinal Study on Aging (CLSA) cohort, and uridine was available, named as Uridine (1) ( Supplementary Table S1 ). The GWAS summary statistics of the 1,400 metabolites or metabolite ratios were publicly available through the GWAS Catalog study ( https://www.ebi.ac.uk/gwas/ ) and the accession numbers for European GWASs: GCST90199621-90201020 ( Chen Y. et al., 2023 ). We also gathered another data through the IEU Open-GWAS Project (ID: ebi-a-GCST90026036) ( Panyard et al., 2021 ) (Uridine2) ( Supplementary Table S1 ). Given the limited number of metabolite-associated SNPs, we select SNPs related to metabolites using a relaxed threshold ( P < 1 × 10 −5 , r 2 < 0.1, and a 500 kb distance) at first. To mitigate the effect of weak IV bias and sensitivity analysis, we replicated the MR analysis with a strict threshold. We selected SNPs at conventional GWAS thresholds ( P < 5 × 10 −8 ), and we excluded SNPs in LD, with an R 2 value greater than or equal to 0.001 and within a 10 mb distance.

Study outcomes

To ascertain the significance of the selected SGLT2 inhibitor, a positive control analysis was conducted using T2DM. We utilized the latest 2024 meta-analyses data on T2DM, from the DIAbetes Genetics Replication and Meta-analysis (DIAGRAM) Consortium. The data conclude all ancestry and we choose European individuals (36 GWAS, 242,283 T2DM cases and 1,569,734 controls) ( Suzuki et al., 2023 ) ( Supplementary Table S1 ). We obtained publicly available GWAS summary data for prostate cancer from the Prostate Cancer Association Group to Investigate Cancer-Associated Alterations in the Genome (PRACTICAL) Consortium ( Schumacher et al., 2018 ) (79,148 prostate cancer cases and 61,106 disease-free controls, European ancestry) ( Supplementary Table S1 ). Prostate-specific antigen (PSA) data were from INTERVAL study ( Sun et al., 2018 ) (2,994 plasma proteins in 3,301 healthy participants) ( Supplementary Table S1 ).

Statistical analyses

Mr estimates of sglt2 inhibition on t2dm and prostate cancer.

Two-sample univariable MR (UVMR) was used to assess the influence of SGLT2 inhibition on T2DM, prostate cancer, and PSA. The MR results demonstrated the impact of SGLT2 inhibition on outcomes. An odds ratio (OR) > 1 indicated an increased risk of outcome, while an OR < 1 signified a reduced risk of outcome occurrence. Mendelian Randomization Pleiotropy RESidual Sum and Outlier (MR-PRESSO) was employed to detect and correct potential horizontal pleiotropy and heterogeneity, with the objective of identifying and address outlier genetic variants ( Bowden et al., 2018 ; Verbanck et al., 2018 ). Firstly, we employed the inverse variance-weighted (IVW) approach as the main analytical method, as it can provide the most accurate and robust estimates, given the assurance that all genetic variants are valid after screening ( Bowden et al., 2016 ; Bowden et al., 2019 ). Then we utilized the weighted median and MR-PRESSO method to estimate the MR effects, improving numerical performance and stability.

Mediation MR analysis through metabolites or metabolite ratios

Two-step MR was performed to assess how metabolites or metabolite ratios mediated the association between SGLT2 inhibitor use and prostate cancer ( Figures 1B, C ). We first used UVMR to screen for metabolites significantly correlated with prostate cancer among 1,091 human circulating metabolites and 309 metabolite ratios (β2). Significant MR outputs (0.05/1,400 = 3.57 × 10 −5 ) were regarded based on a P -value <0.05 Bonferroni-corrected for the number of metabolites and metabolite ratios. Metabolite that remained significant after P -value adjustment under both lenient and strict criteria was considered as prostate cancer related metabolite. Then, we utilized UVMR to assess the impact of SGLT2 inhibition on prostate cancer related metabolites (β1). The mediation proportion of significant metabolite in the association between SGLT2 inhibition and prostate cancer was calculated as the product of β1 and β2 divided by the total effect of SGLT2 inhibition on prostate cancer (β3). Then, we further performed multivariable MR (MVMR) to evaluate the effect of uridine on prostate cancer after adjusting for the genetic effect of SGLT2 inhibition (β2′). The mediation proportion was calculated as the product of β1and β2′ divided by β3.

Sensitivity analysis

The MR- PRESSO method detects potential outliers and removing these outliers, thus quantifying the presence of pleiotropy during analysis process ( Verbanck et al., 2018 ). The weighted median approach can reliably estimate when over 50% of genetic instruments are valid ( Bowden et al., 2016 ). In MR analysis, we used weighted median and MR-PRESSO approaches to confirm the robustness and reliability of the IVW MR estimates. The effectiveness of the instrument variables was measured using F statistics, with instruments being considered weak if the F statistics were less than 10. The MR-Egger intercept test was used to detect horizontal pleiotropy for MR estimates ( Bowden et al., 2015 ). To assess the heterogeneity between the instruments, Cochrane’s Q test and the global test for MR-PRESSO were computed. The MR analysis and sensitivity analysis were performed utilizing the “TwoSampleMR” and “MRPRESSO” packages in R software (version 4.2.1). Meta-analysis was performed through the Reviewer Manager software (Version 5.3). To account for multiple testing, we considered a two-sided P -value that met the Bonferroni corrected P threshold as statistically significant. For the impact of SGLT2 inhibition on T2DM and prostate cancer-related outcomes, a P -value <0.05 was considered statistically significant.

PheWAS analysis

GeneATLAS ( http://geneatlas.roslin.ed.ac.uk/ ) is a comprehensive database utilizing the UK Biobank cohort, encompassing hundreds of traits and millions of variant associations ( Canela-Xandri et al., 2018 ). These associations were computed using 452,264 UK Biobank British white individuals. GeneATLAS includes information on 778 traits (118 quantitative traits and 660 binary traits) and the database allows for querying of GWAS results for 9,113,133 genetic variants and downloading GWAS summary statistics for over 30 million imputed genetic variants, which corresponds to more than 23 billion phenotype-genotype pairs ( Canela-Xandri et al., 2018 ). We searched the GeneATLAS for trait associated with SGLT2 inhibition SNPs. p values of traits for each SNP were corrected using the Bonferroni method, setting the threshold for P values at 6.43 × 10 −6 (0.05/7,780), where 7,780 represented the product of the SNP counts (10) and number of traits (778) queried for each SNP. In addition, we performed MR-PheWAS to explore the potential causal relationship between SGLT2 inhibitors and other diseases. The 2,803 outcomes were obtained from FinnGen consortium R5 release data, including 218,792 individuals and 16,962,023 variants ( Kurki et al., 2023 ). The FinnGen project is a large public-private collaboration that combines genome and health data from 500,000 Finnish biobank participants with digital health records from the Finnish National Health Registers ( https://www.finngen.fi/en ). IVW was used as the main analytical method and we also performed MR analysis by MR-Egger and weighted median method. Significant MR-PheWAS outputs were regarded based on a p -value <0.05 Bonferroni-corrected for the number of traits (0.05/2,803 = 1.78 × 10 −5 ). For Sensitivity analysis, the MR-Egger intercept test and Cochran’s Q test was employed to identify horizontal pleiotropy and heterogeneity.

Causal effects of SGLT2 inhibition on T2DM and prostate cancer

Altogether, 10 distinct SNPs were chosen as instrumental variables of SGLT2 inhibition. The F-statistics ranged from 15.897 to 97.893, which indicated sufficient instrument strength for univariable analyses ( Supplementary Table S2 ). Through two-sample MR analysis, we observed an association between SGLT2 inhibition and a reduced risk of T2DM [OR = 0.66 (95% CI 0.53, 0.82), P = 1.57 × 10 −4 ], prostate cancer [0.34 (0.23, 0.49), P = 2.21 × 10 −8 ] and PSA [0.26 (0.08, 0.81), P = 2.07 × 10 −2 ], for per 1 SD lowering of HbA1c via SGLT2 inhibition ( Figure 2 ). The weighted median and MR-PRESSO supported these results. No heterogeneity was detected for the impact of SGLT2 inhibition on T2DM, prostate cancer and PSA ( P = 0.653; P = 0.844; P = 0.386), and no evidence of horizontal pleiotropy was found ( P = 0.261; P = 0.384; P = 0.109) ( Figure 2 ).

Figure 2 . MR estimates of the effect of SGLT2 inhibition on T2DM, prostate cancer and prostate specific antigen. OR, odds ratio; SGLT2, sodium-glucose cotransporter 2.

Causal effects of metabolites or metabolite ratios on prostate cancer

We explored the influences of 1,091 metabolites and 309 metabolite ratios on the risk for prostate cancer under the relaxed threshold ( P < 1 × 10 −5 ) of instrument variables selection, and results demonstrated that total seven metabolites and one metabolite ratios were significantly linked to prostate cancer (Bonferroni-corrected p -value threshold = 0.05/1,400 = 3.57 × 10 −5 ) ( Figure 3 ). X-17690, X-12798, X-21312, and X-07765 were currently unidentified metabolites. The result revealed that cysteinylglycine disulfide [OR = 0.94 (95% CI 0.92, 0.96), P = 4.41 × 10 −7 , Figure 3 ] and cys-gly, oxidized [0.96 (0.94, 0.98), P = 1.44 × 10 −5 , Figure 3 ] could reduce the risk of prostate cancer. Uridine [1.12 (1.07, 1.18), P = 4.12 × 10 −6 , Figure 3 ] was identified as the only risk metabolite for prostate cancer that remains significant even after adjustment. Then, the strict threshold was used to choose instrument variables. Interestingly, only the uridine levels remained significant after Bonferroni corrected and it was considered prostate cancer related metabolite for further analysis.

Figure 3 . MR estimates of the effects of blood metabolites or metabolite ratios on prostate cancer. OR, odds ratio.

Causal effects of SGLT2 inhibition on prostate cancer related metabolite

The MR results revealed that SGLT2 inhibition decreased the total concentration of uridine (1) [ P = 0.018, β = −0.84 (95% CI −1.54, −0.14), Table 1 ]. Cochrane’s Q test implied no evidence of heterogeneity (Q = 7.352, P = 0.60, Table 1 ). To further validate the causal association between SGLT2 inhibition and uridine, we included another GWAS dataset of uridine level and conducted a meta-analysis of the MR results. The meta-analysis indicated that SGLT2 inhibition could decrease uridine levels significantly [OR = 0.58 (95% CI 0.39, 0.87), Figure 4 ]. Based on previous findings, we discovered that uridine level mediated the reduction in prostate cancer risk associated with SGLT2 inhibition, with the mediating effect accounting for 9.07% of the total effect. In MVMR analysis, SGLT2 inhibitor and uridine were still associated with prostate cancer [OR SGLT2 inhibitor = 0.43 (95% CI 0.24, 0.77), P SGLT2 inhibitor = 0.005, OR uridine = 1.13 (95% CI 1.08, 1.18), P uridine < 0.001]. And the uridine level had a mediated proportion of 9.34% of the total effect.

Table 1 . MR estimates of the effect of SGLT2 inhibition on uridine level.

Figure 4 . Meta-analysis result of the causal associations between SGLT2 inhibition and uridine levels. CLSA, Canadian Longitudinal Study on Aging; WADRC, Wisconsin Alzheimer’s Disease Research Center; WRAP, Wisconsin Registry for Alzheimer’s Prevention.

We obtained trait associations with prioritized druggable target genes from GeneATLAS, and then we set the threshold for P values at 6.43 × 10 −6 (0.05/7, 780). The top 15 traits with the highest number of SNPs associated with them were listed in Supplementary Table S3 . The traits most closely related to SGLT2 inhibitors included platelet count, volume, red blood cell distribution width, reticulocyte count, and systemic impedance. These traits can also serve as potential therapeutic indications and targets for SGLT2 inhibitors. In MR-PheWAS, 65 traits associated with SLGT2 inhibitor ( P < 1.78 × 10 −5 , Supplementary Table S4 ) were identified, with 13 of these traits being linked to diabetes. The use of SGLT2 inhibitors was associated with a reduced risk of benign neoplasms of the meninges, pediculosis, acariasis, scabies, strabismus, and venous thromboembolism. On the other hand, the administration of SGLT2 inhibitors might result in various drug side effects, including endometriosis, psoriasis vulgaris, alcoholic gastritis, and other alcohol-related diseases.

In this study, we focused on evaluating the causal associations between SGLT2 inhibition and two main outcomes: T2DM and prostate cancer. Additionally, we explored the metabolites or metabolite ratios related to prostate cancer risk. Our study revealed that genetic variation in SGLT2 inhibition targets was linked to a reduced risk of prostate cancer [0.34 (0.23, 0.49), P = 2.21 × 10 −8 ] and PSA [0.26 (0.08, 0.81), P = 2.07 × 10 −2 ]. The circulating uridine level might mediate 9.07% of the impact of SGLT2 inhibition on prostate cancer.

As a part of another glucose import system for cancer cells, SGLT2 is functionally expressed in several cancer types ( Scafoglio et al., 2015 ). Large random clinical trials and meta-analysis have indicated that the medicine which can silence or inhibit SGLT2, such like SGLT2 inhibitors, has the potential ability to suppress cancer cell growth. Research has demonstrated that SGLT2 inhibitors exhibit anti-proliferative effects on liver cancer, breast cancer ( Shiba et al., 2018 ; Komatsu et al., 2020 ). The mechanisms underlying the inhibition of SGLT2 in prostate cancer remain controversial. Glucose is an indispensable metabolic required for cancer cell survival and growth. SGLT2 inhibitors can block the uptake of glucose by cancer cells, thus whether their anti-proliferative effects on tumor cells are mediated through the reduction of glucose concentration requires further evidence for support. Besides, emerging evidence indicate that SGLT2 inhibitors effect the progression and metastasis of prostate cancer cells through metabolic dysregulation, especially mitochondria function. AMP-activated kinase (AMPK), a master regulator of metabolism, its activation led to reduced cancer cell proliferation ( Penfold et al., 2023 ). A random clinical trial indicated that, Canagliflozin, a kind of SGLT inhibitors, can leads to the activation of AMPK ( Hawley et al., 2016 ; Villani et al., 2016 ). Act as an attractive strategy based on targeting mitochondrial metabolism, Canagliflozin has expressed potential anti-cancer activity for prostate cancer ( Ali et al., 2023 ). Further research is needed to determine whether SGLT2 affects mitochondrial metabolism through the activation of the AMPK pathway, and more research to investigate whether SGLT2 has other potential effects on metabolic pathways that contribute to its anti-cancer effects.

Our study confirmed that SGLT2 inhibitors reduced the total content of uridine in the serum. This finding highlights the lack of research on the impact of SGLT2 inhibitors on pyrimidine metabolism, particularly regarding uridine levels. One study found no significant difference in uridine levels before and after treatment with canagliflozin on the metabolome of liver cancer cells ( P = 0.5648) ( Nakano et al., 2020 ). Other studies have mainly discussed the relationship between SGLT2 inhibitors and pyrimidine metabolism, and conclusion remains controversial. An in vitro study analyzed the effects of different SGLT2 inhibitors on pyrimidine metabolism ( Zügner et al., 2022 ). The results suggested that treatment with canagliflozin led to an increase in metabolites in the pyrimidine pathway, while the trend was opposite for dapagliflozin and ertugliflozin. An in vivo experiment selected 40 type 2 diabetes patients and measured the plasma and urine metabolites after 8 weeks of treatment with a 25 mg dose of the SGLT2 inhibitor empagliflozin. The study results showed that SGLT2 treatment influences pyrimidine metabolism, but changes in UTP levels were not detected ( Lu et al., 2022 ). Further experimental research is needed to investigate whether SGLT2 affects pyrimidine levels in prostate cancer cells and in the plasma metabolites of patients, particularly the influence of uridine.

Given our research findings that SGLT2 inhibitors could reduce uridine levels, and considering that uridine has a positive association with prostate cancer, we hypothesize that the anti-cancer ability of SGLT2 inhibitors may be mediated through uridine. Previous studies have described the characteristics of plasma metabolomics in prostate cancer patients, suggesting a correlation between uridine and PSA metabolism ( Markin et al., 2020 ). Research at the tissue level has confirmed that uridine levels are elevated in prostate tumor tissues compared to adjacent normal control tissues ( Ren et al., 2016 ). The above studies have supported our hypothesis to some extent.

The investigation of whether SGLT2 is worthwhile as a novel anti-pancreatic cancer drug is a promising area for further research. The methods of metabolomics have become more prominent focus for understanding the mechanisms of anti-pancreatic cancer drugs. Proxalutamide is a novel androgen receptor inhibitor currently used in the castration-resistant prostate cancer patients, and it has entered Phase III clinical trials. Research has been conducted to analyze the metabolomics of prostate cancer cells to elucidate the anti-tumor efficacy of Proxalutamide, and results demonstrated that it significantly decreased the intracellular levels of uridine ( Qu et al., 2020 ). Although the study observed significant differences in other metabolites of prostate cancer cells, such as glutamine (a kind of TCA cycle markers), after treatment with Proxalutamide, the results suggest that one of the potential mechanisms of anti-tumor activity of Proxalutamide is the inhibition of pyrimidine synthesis. The study provided support for the potential key role of uridine in the anti-tumor mechanism of SGLT2 inhibitors in prostate cancer. Our study proposes a novel perspective on the potential anti-prostate cancer mechanism of SGLT2 inhibitors, focus on the uridine, providing a theoretical basis for their potential clinical applications in oncology.

Given the potential of SGLT2 inhibitors in preventing prostate cancer, we should not ignore their side effects. A retrospective cohort study found that SGLT2 inhibitors was associated with an approximately three-fold increased risk of genital infections when compared to the use of Dipeptidyl peptidase-4 inhibitors and Glucagon-like peptide-1 receptor agonists ( Dave et al., 2019 ). In addition, SGLT2 inhibitors may lead to other concerns, including cardiovascular safety, acute renal failure, hypoglycemia, volume depletion, volume depletion, euglycemic ketoacidosis, and bone fractures ( Scheen, 2019 ). In our MR-PheWAS analysis, we found SGLT2 inhibitors may lead to some side effects, which provided cautionary notes for the use of SGLT2 inhibitors in the future.

Although previous studies have investigated the impact of SGLT2 inhibitors on prostate cancer using MR, these studies were incomplete and published as research letters. Interestingly, the two studies reported contradictory results: Lin et al. (2024) study suggested a protective effect of SGLT2 inhibitors on prostate cancer, whereas Han et al. (2024) study indicated an increased risk. The reason for the different results obtained from two studies is that the study by Han et al. did not further analyze the instrumental variables’ beta values by taking its opposite into consideration, or convert the OR into its reciprocal as the final outcome. Therefore, the conclusion drawn in Han et al. (2024) study is actually the association between SGLT2 activator and prostate cancer. In comparison to previous research, we used T2DM as a positive control, demonstrating the effectiveness of the obtained SGLT2 inhibitor instrumental variables. The T2DM and prostate cancer GWAS data were sourced from the largest genetic studies to date. What’s more, our study has the novel findings that SGLT2 was associated with prostate cancer mediated by uridine levels. In our investigation of the relationship between circulating metabolites and prostate cancer, we applied the Bonferroni method to correct P values, ensuring statistical robustness and the reliability of our research findings. Based on SGLT2 inhibitor’s instrument variables, we also conducted PheWAS analysis to explore the potential causal relationship between SGLT2 inhibitors and other diseases. What’s more, we employed colocalization analysis, a powerful tool in uncovering the pleiotropic effects of certain loci on multiple traits ( Zuber et al., 2022 ). This approach aids in a better understanding of the genetic architecture underlying complex traits. We have provided evidence for the potential anti-cancer mechanism of SGLT2 inhibitors, indicating that the overall concentration of uridine exerts anti-cancer effects on prostate cancer.

Nevertheless, our study is subject to certain limitations. Firstly, the genetic variations emulating SGLT2 inhibition demonstrate the long-term effects of SGLT2 inhibitors, whereas the SGLT2 inhibitors in prior studies are mainly intended for short-term use ( Li et al., 2023 ). Therefore, whether short-term use of SGLT2 inhibitors actually leads to the same results as lifelong effects needs further exploration in experimental and clinical trials. Although the variety of circulating metabolites data we used is comprehensive, due to the lack of available GWAS of pyrimidines in the cohort, limits our ability to further validate the results of our study ( Julkunen et al., 2023 ). Finally, further research is needed to generalize the results of this study to populations other than individuals of European descent.

In conclusion, this study suggested the causal associations between genetically predicted SGLT2 inhibition, uridine level and prostate cancer. Our findings provide novel insights into the mechanism by which SGLT2 inhibitors reduce the risk of cancer, particularly prostate cancer, and supports further basic and clinical trials of SGLT2 inhibitors in cancer therapy.

Data availability statement

The GWAS Summary statistics used in this study were publicly accessed from the IEU OpenGWAS project ( https://gwas.mrcieu.ac.uk/ ), the GWAS Catalog ( https://www.ebi.ac.uk/gwas/ ), the GTEx Portal ( https://www.gtexportal.org/ ), the eQTLGen Consortium ( https://eqtlgen.org/ ), DIAGRAM consortium ( https://diagram-consortium.org/ ) and PRACTICAL Consortium ( http://practical.icr.ac.uk/blog/ ). Further inquiries can be directed to the corresponding authors.

Author contributions

YL: Conceptualization, Data curation, Formal Analysis, Validation, Visualization, Writing–original draft, Writing–review and editing. YZ: Data curation, Writing–original draft. SW: Data curation, Formal Analysis, Investigation, Visualization, Writing–original draft. LC: Data curation, Writing–review and editing. RZ: Data curation, Writing–review and editing. XM: Data curation, Writing–review and editing. QLY: Data curation, Writing–review and editing. LZ: Conceptualization, Data curation, Writing–review and editing. QMY: Conceptualization, Data curation, Funding acquisition, Writing–review and editing.

The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. This study received funding from the National Natural Science Foundation of China (grant no. 82303078) and the Beijing Medical Award Foundation (grant no. YXJL-2020-0941-0746).

Acknowledgments

We thank the participants in all the GWASs used in this study and the investigators who made these GWAS data publicly available. We thank the contribution of the PRACTICAL consortium and DIAGRAM consortium for this study.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fphar.2024.1443045/full#supplementary-material

Abbreviations

AMPK, AMP-activated kinase; CLSA, Canadian Longitudinal Study on Aging; DIAGRAM, DIAbetes Genetics Replication and Meta-analysis; eQTLs, expression quantitative trait loci; GTEx, Genotype-Tissue Expression; GWAS, genome-wide association study; INTERVEL, Efficiency and safety of varying the frequency of whole blood donation; LD, linkage disequilibrium; MR, Mendelian randomization; MR-PRESSO, Mendelian Randomization Pleiotropy RESidual Sum and Outlier; MVMR, multivariable MR; PRACTICAL, Prostate Cancer Association Group to Investigate Cancer-Associated Alterations in the Genom; PRISM, Profiling Relative Inhibition Simultaneously in Mixtures; PSA, Prostate-Specific Antigen; RNA, Ribonucleic acid; SGLT2, sodium-glucose cotransporter 2; SNPs, single nucleotide polymorphisms; STROBE-MR, Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization; T2DM, type 2 diabetes mellitus; TCA, tricarboxylic acid; UTP, uridine 5′-triphosphate; UVMR, univariable Mendelian randomization.

Ali, A., Mekhaeil, B., Biziotis, O. D., Tsakiridis, E. E., Ahmadi, E., Wu, J., et al. (2023). The SGLT2 inhibitor canagliflozin suppresses growth and enhances prostate cancer response to radiotherapy. Commun. Biol. 6 (1), 919. doi:10.1038/s42003-023-05289-w

PubMed Abstract | CrossRef Full Text | Google Scholar

Bakker, M. K., van Straten, T., Chong, M., Paré, G., Gill, D., and Ruigrok, Y. M. (2023). Anti-epileptic drug target perturbation and intracranial aneurysm risk: mendelian randomization and colocalization study. Stroke 54 (1), 208–216. doi:10.1161/STROKEAHA.122.040598

Bangarbale, S., Shepard, B. D., Bansal, S., Jayatilake, M. M., Kurtz, R., Levi, M., et al. (2022). Renal metabolome in obese mice treated with empagliflozin suggests a reduction in cellular respiration. Biomolecules 12 (9), 1176. doi:10.3390/biom12091176

Benedetti, R., Benincasa, G., Glass, K., Chianese, U., Vietri, M. T., Congi, R., et al. (2022). Effects of novel SGLT2 inhibitors on cancer incidence in hyperglycemic patients: a meta-analysis of randomized clinical trials. Pharmacol. Res. 175, 106039. doi:10.1016/j.phrs.2021.106039

Bhatt, D. L., Szarek, M., Pitt, B., Cannon, C. P., Leiter, L. A., McGuire, D. K., et al. (2021). Sotagliflozin in patients with diabetes and chronic kidney disease. N. Engl. J. Med. 384 (2), 129–139. doi:10.1056/NEJMoa2030186

Bouras, E., Karhunen, V., Gill, D., Huang, J., Haycock, P. C., Gunter, M. J., et al. (2022). Circulating inflammatory cytokines and risk of five cancers: a Mendelian randomization analysis. BMC Med. 20 (1), 3. doi:10.1186/s12916-021-02193-0

Bowden, J., Davey Smith, G., and Burgess, S. (2015). Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int. J. Epidemiol. 44 (2), 512–525. doi:10.1093/ije/dyv080

Bowden, J., Del Greco, F. M., Minelli, C., Zhao, Q., Lawlor, D. A., Sheehan, N. A., et al. (2019). Improving the accuracy of two-sample summary-data Mendelian randomization: moving beyond the NOME assumption. Int. J. Epidemiol. 48 (3), 728–742. doi:10.1093/ije/dyy258

Bowden, J., Smith, G. D., Haycock, P. C., and Burgess, S. (2016). Consistent estimation in mendelian randomization with some invalid instruments using a weighted median estimator. Genet. Epidemiol. 40 (4), 304–314. doi:10.1002/gepi.21965

Bowden, J., Spiller, W., Del Greco, M. F., Sheehan, N., Thompson, J., Minelli, C., et al. (2018). Improving the visualization, interpretation and analysis of two-sample summary data Mendelian randomization via the Radial plot and Radial regression. Int. J. Epidemiol. 47 (4), 1264–1278. doi:10.1093/ije/dyy101

Brown, E., Heerspink, H. J. L., Cuthbertson, D. J., and Wilding, J. P. H. (2021). SGLT2 inhibitors and GLP-1 receptor agonists: established and emerging indications. Lancet 398 (10296), 262–276. doi:10.1016/S0140-6736(21)00536-5

Canela-Xandri, O., Rawlik, K., and Tenesa, A. (2018). An atlas of genetic associations in UK Biobank. Nat. Genet. 50 (11), 1593–1599. doi:10.1038/s41588-018-0248-z

Cannon, C. P., Pratley, R., Dagogo-Jack, S., Mancuso, J., Huyck, S., Masiukiewicz, U., et al. (2020). Cardiovascular outcomes with ertugliflozin in type 2 diabetes. N. Engl. J. Med. 383 (15), 1425–1435. doi:10.1056/NEJMoa2004967

Chen, J., Xu, F., Ruan, X., Sun, J., Zhang, Y., Zhang, H., et al. (2023a). Therapeutic targets for inflammatory bowel disease: proteome-wide Mendelian randomization and colocalization analyses. EBioMedicine 89, 104494. doi:10.1016/j.ebiom.2023.104494

Chen, Y., Lu, T., Pettersson-Kymmer, U., Stewart, I. D., Butler-Laporte, G., Nakanishi, T., et al. (2023b). Genomic atlas of the plasma metabolome prioritizes metabolites implicated in human diseases. Nat. Genet. 55 (1), 44–53. doi:10.1038/s41588-022-01270-1

Connolly, G. P., and Duley, J. A. (1999). Uridine and its nucleotides: biological actions, therapeutic potentials. Trends Pharmacol. Sci. 20 (5), 218–225. doi:10.1016/s0165-6147(99)01298-5

Dave, C. V., Schneeweiss, S., and Patorno, E. (2019). Comparative risk of genital infections associated with sodium-glucose co-transporter-2 inhibitors. Diabetes Obes. Metab. 21 (2), 434–438. doi:10.1111/dom.13531

GTEx Consortium (2020). The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 369 (6509), 1318–1330. doi:10.1126/science.aaz1776

Han, Z., He, Y., Li, X., Li, S., and Ai, J. (2024). Insights into the impact of sodium-glucose cotransporter 2 inhibition on urinary tract malignancy: a two-sample Mendelian randomization. Diabetes Obes. Metab. 26 (5), 1986–1989. doi:10.1111/dom.15490

Hawley, S. A., Ford, R. J., Smith, B. K., Gowans, G. J., Mancini, S. J., Pitt, R. D., et al. (2016). The Na+/Glucose cotransporter inhibitor canagliflozin activates AMPK by inhibiting mitochondrial function and increasing cellular AMP levels. Diabetes 65 (9), 2784–2794. doi:10.2337/db16-0058

Heerspink, H. J., Perkins, B. A., Fitchett, D. H., Husain, M., and Cherney, D. Z. (2016). Sodium glucose cotransporter 2 inhibitors in the treatment of diabetes mellitus: cardiovascular and kidney effects, potential mechanisms, and clinical applications. Circulation 134 (10), 752–772. doi:10.1161/CIRCULATIONAHA.116.021887

Julkunen, H., Cichońska, A., Tiainen, M., Koskela, H., Nybo, K., Mäkelä, V., et al. (2023). Atlas of plasma NMR biomarkers for health and disease in 118,461 individuals from the UK Biobank. Nat. Commun. 14 (1), 604. doi:10.1038/s41467-023-36231-7

Koepsell, H. (2017). The Na+-D-glucose cotransporters SGLT1 and SGLT2 are targets for the treatment of diabetes and cancer. Pharmacol. Ther. 170, 148–165. doi:10.1016/j.pharmthera.2016.10.017

Komatsu, S., Nomiyama, T., Numata, T., Kawanami, T., Hamaguchi, Y., Iwaya, C., et al. (2020). SGLT2 inhibitor ipragliflozin attenuates breast cancer cell proliferation. Endocr. J. 67 (1), 99–106. doi:10.1507/endocrj.EJ19-0428

Kurki, M. I., Karjalainen, J., Palta, P., Sipilä, T. P., Kristiansson, K., Donner, K. M., et al. (2023). FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 613 (7944), 508–518. doi:10.1038/s41586-022-05473-8

Li, J., Yu, Y., Sun, Y., Yu, B., Tan, X., Wang, B., et al. (2023). SGLT2 inhibition, circulating metabolites, and atrial fibrillation: a Mendelian randomization study. Cardiovasc. Diabetol. 22 (1), 278. doi:10.1186/s12933-023-02019-8

Lin, J., Zhou, J., and Xu, Y. (2023). Potential drug targets for multiple sclerosis identified through Mendelian randomization analysis. Brain 146 (8), 3364–3372. doi:10.1093/brain/awad070

Lin, L., Ning, K., Xiang, L., Peng, L., and Li, X. (2024). SGLT2 inhibition and three urological cancers: up-to-date results. Diabetes Metab. Res. Rev. 40 (3), e3797. doi:10.1002/dmrr.3797

Lin, Y., Yang, Z., Li, J., Sun, Y., Zhang, X., Qu, Z., et al. (2022). Effects of glutamate and aspartate on prostate cancer and breast cancer: a Mendelian randomization study. BMC Genomics 23 (1), 213. doi:10.1186/s12864-022-08442-7

Lu, Y. P., Zhang, Z. Y., Wu, H. W., Fang, L. J., Hu, B., Tang, C., et al. (2022). SGLT2 inhibitors improve kidney function and morphology by regulating renal metabolic reprogramming in mice with diabetic kidney disease. J. Transl. Med. 20 (1), 420. doi:10.1186/s12967-022-03629-8

Markin, P. A., Brito, A., Moskaleva, N., Lartsova, E. V., Shpot, Y. V., Lerner, Y. V., et al. (2020). Plasma metabolomic profile in prostatic intraepithelial neoplasia and prostate cancer and associations with the prostate-specific antigen and the Gleason score. Metabolomics 16 (7), 74. doi:10.1007/s11306-020-01694-y

Nakano, D., Kawaguchi, T., Iwamoto, H., Hayakawa, M., Koga, H., and Torimura, T. (2020). Effects of canagliflozin on growth and metabolic reprograming in hepatocellular carcinoma cells: multi-omics analysis of metabolomics and absolute quantification proteomics (iMPAQT). PLoS One 15 (4), e0232283. doi:10.1371/journal.pone.0232283

Neal, B., Perkovic, V., Mahaffey, K. W., de Zeeuw, D., Fulcher, G., Erondu, N., et al. (2017). Canagliflozin and cardiovascular and renal events in type 2 diabetes. N. Engl. J. Med. 377 (7), 644–657. doi:10.1056/NEJMoa1611925

Panyard, D. J., Kim, K. M., Darst, B. F., Deming, Y. K., Zhong, X., Wu, Y., et al. (2021). Cerebrospinal fluid metabolomics identifies 19 brain-related phenotype associations. Commun. Biol. 4 (1), 63. doi:10.1038/s42003-020-01583-z

Penfold, L., Woods, A., Pollard, A. E., Arizanova, J., Pascual-Navarro, E., Muckett, P. J., et al. (2023). AMPK activation protects against prostate cancer by inducing a catabolic cellular state. Cell Rep. 42 (4), 112396. doi:10.1016/j.celrep.2023.112396

Qu, F., Gu, Y., Wang, Q., He, M., Zhou, F., Sun, J., et al. (2020). Metabolomic profiling to evaluate the efficacy of proxalutamide, a novel androgen receptor antagonist, in prostate cancer cells. Invest. New Drugs 38 (5), 1292–1302. doi:10.1007/s10637-020-00901-w

Ren, S., Shao, Y., Zhao, X., Hong, C. S., Wang, F., Lu, X., et al. (2016). Integration of metabolomics and transcriptomics reveals major metabolic pathways and potential biomarker involved in prostate cancer. Mol. Cell Proteomics 15 (1), 154–163. doi:10.1074/mcp.M115.052381

Scafoglio, C., Hirayama, B. A., Kepe, V., Liu, J., Ghezzi, C., Satyamurthy, N., et al. (2015). Functional expression of sodium-glucose transporters in cancer. Proc. Natl. Acad. Sci. U. S. A. 112 (30), E4111–E4119. doi:10.1073/pnas.1511698112

Scheen, A. J. (2019). An update on the safety of SGLT2 inhibitors. Expert Opin. Drug Saf. 18 (4), 295–311. doi:10.1080/14740338.2019.1602116

Schumacher, F. R., Al Olama, A. A., Berndt, S. I., Benlloch, S., Ahmed, M., Saunders, E. J., et al. (2018). Association analyses of more than 140,000 men identify 63 new prostate cancer susceptibility loci. Nat. Genet. 50 (7), 928–936. doi:10.1038/s41588-018-0142-8

Shiba, K., Tsuchiya, K., Komiya, C., Miyachi, Y., Mori, K., Shimazu, N., et al. (2018). Canagliflozin, an SGLT2 inhibitor, attenuates the development of hepatocellular carcinoma in a mouse model of human NASH. Sci. Rep. 8 (1), 2362. doi:10.1038/s41598-018-19658-7

Skinner, O. S., Blanco-Fernández, J., Goodman, R. P., Kawakami, A., Shen, H., Kemény, L. V., et al. (2023). Salvage of ribose from uridine or RNA supports glycolysis in nutrient-limited conditions. Nat. Metab. 5 (5), 765–776. doi:10.1038/s42255-023-00774-2

Skrivankova, V. W., Richmond, R. C., Woolf, B. A. R., Yarmolinsky, J., Davies, N. M., Swanson, S. A., et al. (2021). Strengthening the reporting of observational studies in Epidemiology using mendelian randomization: the STROBE-MR statement. Jama 326 (16), 1614–1621. doi:10.1001/jama.2021.18236

Sun, B. B., Maranville, J. C., Peters, J. E., Stacey, D., Staley, J. R., Blackshaw, J., et al. (2018). Genomic atlas of the human plasma proteome. Nature 558 (7708), 73–79. doi:10.1038/s41586-018-0175-2

Suzuki, K., Hatzikotoulas, K., Southam, L., Taylor, H. J., Yin, X., Lorenz, K. M., et al. (2023). Multi-ancestry genome-wide study in >2.5 million individuals reveals heterogeneity in mechanistic pathways of type 2 diabetes and complications. medRxiv.

Google Scholar

Verbanck, M., Chen, C. Y., Neale, B., and Do, R. (2018). Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat. Genet. 50 (5), 693–698. doi:10.1038/s41588-018-0099-7

Villani, L. A., Smith, B. K., Marcinko, K., Ford, R. J., Broadfield, L. A., Green, A. E., et al. (2016). The diabetes medication Canagliflozin reduces cancer cell proliferation by inhibiting mitochondrial complex-I supported respiration. Mol. Metab. 5 (10), 1048–1056. doi:10.1016/j.molmet.2016.08.014

Võsa, U., Claringbould, A., Westra, H. J., Bonder, M. J., Deelen, P., Zeng, B., et al. (2021). Large-scale cis- and trans-eQTL analyses identify thousands of genetic loci and polygenic scores that regulate blood gene expression. Nat. Genet. 53 (9), 1300–1310. doi:10.1038/s41588-021-00913-z

Wiviott, S. D., Raz, I., Bonaca, M. P., Mosenzon, O., Kato, E. T., Cahn, A., et al. (2019). Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N. Engl. J. Med. 380 (4), 347–357. doi:10.1056/NEJMoa1812389

Xu, M., Zheng, J., Hou, T., Lin, H., Wang, T., Wang, S., et al. (2022). SGLT2 inhibition, choline metabolites, and cardiometabolic diseases: a mediation mendelian randomization study. Diabetes Care 45 (11), 2718–2728. doi:10.2337/dc22-0323

Zuber, V., Grinberg, N. F., Gill, D., Manipur, I., Slob, E. A. W., Patel, A., et al. (2022). Combining evidence from Mendelian randomization and colocalization: review and comparison of approaches. Am. J. Hum. Genet. 109 (5), 767–782. doi:10.1016/j.ajhg.2022.04.001

Zügner, E., Yang, H. C., Kotzbeck, P., Boulgaropoulos, B., Sourij, H., Hagvall, S., et al. (2022). Differential in vitro effects of SGLT2 inhibitors on mitochondrial oxidative phosphorylation, glucose uptake and cell metabolism. Int. J. Mol. Sci. 23 (14), 7966. doi:10.3390/ijms23147966

Keywords: SGLT2 inhibitor, prostate cancer, uridine, Mendelian randomization, PheWAS

Citation: Lin Y, Zhang Y, Wang S, Cao L, Zhao R, Ma X, Yang Q, Zhang L and Yang Q (2024) Pharmacological targets of SGLT2 inhibition on prostate cancer mediated by circulating metabolites: a drug-target Mendelian randomization study. Front. Pharmacol. 15:1443045. doi: 10.3389/fphar.2024.1443045

Received: 03 June 2024; Accepted: 19 July 2024; Published: 06 August 2024.

Reviewed by:

Copyright © 2024 Lin, Zhang, Wang, Cao, Zhao, Ma, Yang, Zhang and Yang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Yilong Lin, [email protected] ; Liyi Zhang, [email protected] ; Qingmo Yang, [email protected]

† These authors have contributed equally to this work

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

PUB - Publikationen an der Universität Bielefeld

Causal mediation analysis in single case experimental designs: introduction to the special issue.

Miocevic M, Moeyaert M, Mayer A, Montoya AK (2022) Evaluation and the Health Professions : 1632787211073194.

• Export / Suche
• APA (6th ed.)

Markieren/ Markierung löschen Markierte Publikationen

Open Data PUB

Web of science.