Association between serum urate levels, gout and breast cancer: observational and Mendelian randomization analyses

Introduction

Breast cancer is the most frequent type of cancer in women globally and the primary cause of cancer-related death in women, with 2,261,419 new cases reported in 2020 (1). Women have a high lifetime risk of being diagnosed with breast cancer (2). Identifying risk factors associated with breast cancer is critical to the prevention, treatment, and management of breast cancer.

Uric acid is the metabolic final product of purines through xanthine oxidoreductase and is present in the blood in the form of urate. It is primarily eliminated via the kidneys and intestines (3). Obesity, alcoholism, and a diet high in purines are risk factors for increased serum uric acid, which can lead to hyperuricemia and gout (4). Hyperuricemia is defined as a serum uric acid concentration over 6.8 mg/dL (5), whereas gout is a progressive metabolic and arthropathic disease resulting from excessive urate deposits and characterized by symptomatic hyperuricemia (6). Clinically, many cancer treatment factors including radiotherapy, chemotherapy, targeted therapy can lead to hyperuricemia, and some patients even appear in the form of acute tumor lysis syndrome, because after the application of antitumor therapy, the tumor cells are destroyed, lysed and release a large amount of nucleic acid, which leads to an increase in uric acid by degradation. In addition, some chemotherapeutic drugs themselves or through their impact on renal function can also lead to an increase in the level of serum uric acid (7).

Uric acid is reported to be a potent antioxidant that may prevent cancer (8). However, current observational epidemiologic studies on uric acid, gout, and cancer associations provide inconsistent evidence (9-12). Decreased xanthine oxidoreductase activity in mouse and human mammary epithelial cells is associated with increased invasiveness, and uric acid increases the migration rate of human breast cancer cells and mouse mammary epithelial cells, suggesting a potential link between uric acid and breast cancer (13). But the relationship between serum urate levels, gout, and breast cancer remains controversial. There is evidence that urate levels and gout are negatively correlated with breast cancer (14,15), In particular, a multicenter cohort study showed that higher levels of uric acid were associated with a lower risk of breast cancer and cancer death (14). In addition, a large cohort study of 493,281 individuals aged 20 years and older also showed that serum uric acid levels were negatively associated with breast cancer risk (15). However, some studies have suggested otherwise (16-18). Observational studies are vulnerable to many confounders that can bias results. Mendelian randomization (MR) uses genetic variants strongly related to exposure as instrumental variables (IVs) to study the relationship between exposure and outcome. Due to the random assignment and temporal prioritization of genetic variants, MR can effectively overcome the effects of reverse causation and residual confounding that are difficult to avoid in observational studies (19,20). Therefore, we first explored the observational associations based on the National Health and Nutrition Examination Survey (NHANES) 2009–2018, and then further assessed the causality using MR. In addition, studies have shown that risk factors have different effects on estrogen receptor (ER)-positive and negative breast cancers (21), so we additionally assessed the causal associations between uric acid and gout with these two subtypes of breast cancer. We present this article in accordance with the STROBE-MR reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-1141/rc).

Methods

Overall study design

This study is divided into two sections. First, we utilized data from NHANES to explore the observational link between urate, gout and breast cancer. Then, we carried out MR analyses utilizing summary statistics of genome-wide association studies (GWAS) to evaluate the causal effects of serum urate levels and gout on breast cancer. The study design and flow chart are presented in Figure 1.

Figure 1 The study design and flow chart. (A) The flowchart of NHANES participants. (B) The study design and flow chart for MR analysis. NHANES, National Health and Nutrition Examination Survey; MR, Mendelian randomization; SNP, single nucleotide polymorphism; ER+, estrogen receptor-positive; ER−, estrogen receptor-negative.

Observational analysis

Participants in NHANES

NHANES is a cross-sectional survey performed by the National Center for Health Statistics (NCHS), part of the Centers for Disease Control and Prevention (CDC), to evaluate the health and nutritious condition of the American population (22). This survey has received ethical approval from the NCHS Ethics Review Board (ERB), and all subjects have provided informed consent. For more details, see this website (https://www.cdc.gov/nchs/nhanes/irba98.htm). To investigate the relationship between uric acid, gout, and breast cancer, we performed a retrospective analysis using the NHANES dataset from 5 cycles (2009–2018). The participant exclusion criteria are as follows: (I) male patients (N=24,533), (II) participants under the age of 20 years (N=10,278), (III) participants with missing uric acid data (N=1,490) or missing gout data (N=6), and (IV) participants with missing cancer data (N=7) or with other cancers (N=928). Figure 1 illustrates the entire recruiting process.

Assessment of uric acid and gout

Serum uric acid levels were measured by the Beckman Unicel DxC 800 Synchron. Information on gout was obtained from medical conditions in the questionnaire data. Based on participants’ self-reported responses to the question “Has a medical professional ever told you that you have gout?”, participants were defined as having gout or not, and if the answer was “Don’t know/Refused”, they were regarded as patients with missing gout data.

Assessment of breast cancer

Information on breast cancer was also obtained from medical conditions in the questionnaire data. According to participants’ self-reported answers to the question “Have you ever been told by a medical professional that you have cancer or any type of malignancy? Participants were defined as having cancer or not, and if the answer was “don’t know/refused”, they were regarded as participants with missing cancer data. Then based on the self-reported answer, “What type of cancer is it?”, participants were defined as having breast cancer if the answer was “breast cancer”, and participants who answered with other types of cancer were excluded from the study.

Covariate

This study included the following covariates: age, race, education level, marital status, cholesterol, triglycerides, fasting blood glucose, body mass index (BMI), age at menarche, age at menopause, ever been pregnant, smoking status, and alcohol use. Individuals who had consumed 100 cigarettes or more over their lives were regarded as smokers. Individuals who reported consuming a minimum of 12 alcoholic beverages annually were regarded as alcohol drinkers.

Statistical analysis

This study considered the sampling weights of NHANES. The continuous variables were represented as mean ± standard deviation (SD), and the differences between the non-breast cancer group and the breast cancer group were derived by weighted linear regression models. The categorical variables were represented as percentages, and the differences between the non-breast cancer group and the breast cancer group were derived by weighted chi-square test. The link between uric acid, gout and breast cancer was estimated using a multivariate logistic regression model. Model 1 was unadjusted, Model 2 was adjusted for race and age, and Model 3 was adjusted for race, age, education level, BMI, marital status, fasting blood glucose, triglycerides, cholesterol, smoking status, and alcohol use, age at first menstruation, age at menopause, ever been pregnant. All analyses were carried out using R 4.3.2 or Empowerstats 2.0. The statistical significance threshold was P<0.05.

MR

Data sources

GWAS data regarding urate levels and gout exposures in the study were acquired from the UK Biobank. The Urate Levels Cohort is a 2021 study that included 437,354 samples and 4,231,909 single nucleotide polymorphisms (SNPs), and the Gout Cohort is also a 2021 study that included 484,598 samples and 9,587,836 SNPs. Meanwhile, we got GWAS data regarding breast cancer outcomes from the Breast Cancer Association Consortium (BCAC). The overall breast cancer dataset included 106,776 samples and 10,680,257 SNPs, the ER+ breast cancer dataset included 83,691 samples and 10,680,257 SNPs, and the ER− breast cancer dataset included 55,149 samples and 10, 680,257 SNPs. In addition, we acquired datasets on smoking status, frequency of alcohol intake, and BMI for multivariate MR analysis. All data are from European populations and are available in the database of the IEU Open GWAS project (https://gwas.mrcieu.ac.uk/datasets/). Specific GWAS cohort characteristics are shown in Table S1.

IVs extraction

The selection of SNPs as IVs to assess the causal effects of urate levels and gout on breast cancer incidence required the fulfillment of three assumptions of MR as shown in Figure 1. The following were the specific SNP inclusion criteria: (I) SNPs linked to exposure with genome-wide significance (P<5×10⁻⁸). (II) there was no linkage disequilibrium between SNPs (r²<0.001 and within 10,000 kb window). (III) We searched for each included SNP using the PhenoScannerV2 website (http://www.phenoscanner.medschl.cam.ac.uk/), and SNPs closely related to confounders should be eliminated prior to MR analysis. Meanwhile, SNPs with frequencies of palindromic intermediate alleles were excluded. Because the SNPs for breast cancer outcomes were from studies at the same institution, we ultimately obtained 24 SNPs as IVs for gout and 300 SNPs as IVs for urate levels, regardless of whether the outcome was overall, ER+, or ER− breast cancer. The main data are available at https://cdn.amegroups.cn/static/public/TCR-24-1141-1.xlsx.

Statistical analysis

Two-sample MR analysis

Three methods were used for MR analysis: inverse variance weighted (IVW) random effects model (23), weighted median (24), and MR Egger (25), with the IVW as the primary analytical approach.

Sensitivity analysis

Heterogeneity was evaluated using Cochran’s Q test (26) and horizontal pleiotropy was mainly identified using the MR-Egger intercept test (25). MR-PRESSO was also performed to detect horizontal polytropy and to test for differences between the results of the MR analysis before and after outlier correction (27). In addition, the Steiger directionality test was implemented to avoid reverse causality. Finally, the leave-one-out plot and funnel plot were utilized to perform additional sensitivity analysis.

Multivariate MR analysis

We searched for SNPs obtained from two-sample MR analysis using the PhenoScannerV2 website (http://www.phenoscanner.medschl.cam.ac.uk/) and found that some of them were strongly related to smoking status, alcohol intake frequency, and BMI, regardless of whether the exposure was urate levels or gout. To avoid bias, we adjusted for these three confounders using multivariate MR analysis (28).

All analyses were conducted utilizing version 4.2.3 of the R software and the TwoSampleMR R package. The statistical significance threshold was P<0.05.

Results

Observational analysis

Participants characteristics of NHANES

This study ultimately included 12,451 participants, 358 participants with breast cancer, and 12,093 without breast cancer. Table 1 summarizes the characteristics of NHANES participants. The mean weighted ages of the breast cancer and control groups were 65.61±11.88 and 46.55±16.68 years, respectively. Individuals with and without breast cancer differed significantly (P<0.05) in age, race, education level, marital status, ever been pregnant, alcohol consumption, presence of gout, uric acid, cholesterol, triglycerides, fasting blood sugar, and age at menopause. Breast cancer patients, compared to patients who did not have breast cancer, were more likely to be older, non-Hispanic White, have a higher education level, be widowed/divorced/separated, have been pregnant, have gout, not drink alcohol, have a later age of menopause, and have higher uric acid, cholesterol, triglycerides, and fasting glucose (P<0.05).

Association between urate, gout and breast cancer

The findings of the multivariate logistic regression analyses of the relationship between urate, gout and breast cancer are shown in Table 2. In the unadjusted model (Model 1), there was a positive correlation between serum urate levels and the risk of breast cancer [odds ratio (OR) 1.29, 95% confidence interval (CI): 1.20–1.38, P<0.001]. However, in model 3, which adjusted for multiple covariates, there was a negative correlation between urate levels and the risk of breast cancer, despite the lack of statistical significance in the P value (OR 0.99, 95% CI: 0.89–1.09, P=0.80). Similarly, in the unadjusted model (Model 1), gout was linked to an elevated risk of breast cancer (OR 2.40, 95% CI: 1.52–3.78, P<0.001). However, in Model 3, which controlled for confounding, gout was observed to be linked to a lower incidence of breast cancer, despite the lack of statistical significance in the P value (OR 0.96, 95% CI: 0.53–1.76, P=0.90).

MR study

Two-sample MR analysis

Two-sample MR analysis suggested that elevated serum urate levels were related to decreased risk of ER− breast cancer (IVW, OR 0.916, 95% CI: 0.848–0.989, P=0.03), but not to overall and ER+ breast cancer risk (IVW, both P>0.05). Gout was negatively associated with overall (IVW, OR 0.07, 95% CI: 0.008–0.594, P=0.02), ER+ (IVW, OR 0.062, 95% CI: 0.005–0.742, P=0.03), and ER-breast cancer risk (IVW, OR 0.041, 95% CI: 0.004–0.472, P=0.01). The results in detail are illustrated in Figures 2,3.

Figure 2 Forest plot of the causal effect of urate levels on breast cancer and its subtypes. nSNP, number of single nucleotide polymorphism; ER+, estrogen receptor-positive; ER−, estrogen receptor-negative; OR, odds ratio; CI, confidence interval; MR, Mendelian randomization.

Figure 3 Forest plot of the causal effect of gout on breast cancer and its subtypes. nSNP, number of single nucleotide polymorphism; ER+, estrogen receptor-positive; ER−, estrogen receptor-negative; OR, odds ratio; CI, confidence interval; MR, Mendelian randomization.

Sensitivity analysis

The sensitivity analysis findings are presented in Table 3. Cochran’s Q test shows that there is heterogeneity in most of our results, but we used random effects IVW in our study, which effectively mitigates the effects of heterogeneity, so it does not affect the reliability of our results. The findings of the MR-Egger intercept test revealed no horizontal pleiotropy (P all >0.05). The results were not significantly changed before or after MR-PRESSO correction for outliers. The Steiger directionality test also shows that there is no reverse causality in our analyses, as detailed in Table S2. Figure 4 and Figure S1 illustrate the scatter plots and forest plots of the MR analysis. As shown in Figure S2, the funnel plot indicates that there is no significant heterogeneity in the results. As shown in Figure S3, the leave-one-out analyses indicated that no individual SNP had a significant impact on the MR estimate results, as the OR values were all on the one side of the zero-line.

Figure 4 The MR analysis scatter plots. The estimated effect of MR for each method is denoted by the slope of each line. (A) Gout on overall breast cancer. (B) Gout on ER+ breast cancer. (C) Gout on ER− breast cancer. (D) Urate levels on overall breast cancer. (E) Urate levels on ER+ breast cancer. (F) Urate levels on ER− breast cancer. MR, Mendelian randomization; ER, estrogen receptor; SNP, single nucleotide polymorphism.

Multivariate MR analysis

As shown in Table 4, after adjusting for smoking status, alcohol intake frequency, and BMI using multivariate MR, urate levels continued to have a negative correlation with ER− breast cancer (OR 0.899, 95% CI: 0.826–0.977, P=0.02). In addition, gout was still negatively associated with overall (OR 0.069, 95% CI: 0.012–0.385, P=0.002), ER+ (OR 0.069, 95% CI: 0.010–0.476, P=0.007), and ER− breast cancer risk (OR 0.021, 95% CI: 0.002–0.306, P=0.005).

Discussion

This is the first study to examine the relationship between uric acid, gout and breast cancer using NHANES and MR, to our knowledge. We comprehensively examined the relationships between urate levels, gout and breast cancer by combining an observational analysis using data from NHANES 2009–2018 and an MR analysis using data from the large-scale GWAS. In the present study, cross-sectional observational analysis found that urate levels and gout were negatively related to breast cancer risk, even though the P value was not significant. Two-sample MR analysis showed that serum urate levels were negatively related to the risk of ER− breast cancer, but not significantly related to overall or ER+ breast cancer. In addition, gout was negatively associated with the risk of overall, ER+, and ER− breast cancer. The results of the sensitivity analysis also showed the reliability of our MR results. These associations persisted after multivariate MR adjustment for smoking status, alcohol intake frequency, and BMI. Our study elucidated the link between uric acid, gout and breast cancer. In summary, our findings support the antioxidant role of uric acid in inhibiting breast carcinogenesis, and provide some reference for the management of uric acid levels in hyperuricemic patients who have high risk factors for breast cancer. However, the use of high uric acid levels or gout to combat breast cancer in the general population is not advocated until the mechanisms involved are clear, and further studies could explore the mechanisms and targets to develop relevant antitumor drugs.

It is speculated that uric acid, as a potent antioxidant, may be able to prevent cancer (8). However, there is still controversy about the connection between serum urate levels, gout and cancer. Similarly, the connection between urate levels, gout and breast cancer is inconclusive. A cohort study of 4,350 men followed for 38 years showed that higher serum uric acid levels were linked to a reduced risk of death from any malignancy [hazard ratio (HR) (95% CI): 0.85 (0.73–0.97)] (9). Horsfall et al. (10) discovered that smokers with low serum uric acid levels had a higher chance of getting lung cancer. This suggests that uric acid may help protect against cancer. In line with our findings, a population-based prospective study found that uric acid levels were inversely linked with breast cancer risk and death from cancer, but not significantly connected with prostate, lung, and colorectal cancer risk (14). In addition, Yiu et al. also revealed a negative link between serum uric acid levels and breast cancer in the AMORIS study (15). However, in a cross-sectional study including 1,050 women, Fan et al. discovered a J-shaped correlation between uric acid and the risk of breast cancer, but the sample size was small and there was no record of dietary habits associated with elevated uric acid (16). In a bidirectional MR analysis and prospective cohort study, Feng et al. found that uric acid mediated the link between BMI and increased incidence of postmenopausal breast cancer, but the study crossed MR and cohort studies rather than using both to mutually validate the association (17). Gremke et al. found gout to be related to an increased risk of breast cancer in a retrospective cohort study including 67,598 German cases, but the authors did not consider any external confounders that might have influenced the results of the study (18). Moreover, urate levels and the risk of cancer have been positively correlated in some studies (11,12). In general, the evidence regarding the connection between uric acid and the risk of cancer seems to be fairly inconsistent. Observational studies are vulnerable to many confounders that can bias results. We combined the NHANES cross-sectional study with MR analysis for a comprehensive investigation of the relationship between urate levels, gout and risk of breast cancer, with more reliable results.

It was initially reported that uric acid acts as an antioxidant, a scavenger of unilinear oxygen and radicals, and may be able to protect against cancer (8). Uric acid was found to reduce reactive oxygen species (ROS) production by systemic administration of uric acid to healthy volunteers (29). Some evidence also supports the antioxidant effects of uric acid (30,31). The uric acid transporter SLC2A9 has been demonstrated to be a key downstream target of p53, and the p53-SLC2A9 pathway has been found to prevent the accumulation of ROS-associated damage that may lead to cancer development (32). Conversely, cells exposed to uric acid develop oxidative stress (33), and ROS activates signaling pathways associated with proliferation, survival, angiogenesis, and metastasis, promoting tumorigenesis, progression, and metastasis (34). The dual role of uric acid as an antioxidant and pro-oxidant (35), as well as the dual role of uric acid in cancer, may explain the different associations between uric acid and different cancers. However, we have not yet found strong experimental evidence that uric acid plays a specific negative role in breast carcinogenesis, and further studies are still needed in the future to explore the mechanisms involved.

The primary advantage of the present study is the combination of a large cross-sectional observational study and MR analysis to more fully investigate the link between urate levels, gout and breast cancer, as conclusions were drawn from multiple lines of evidence. In addition, we applied various sensitivity analyses in the MR analysis to ensure the reliability and robustness of our results. Despite this, there are some limitations in this study. In the observational study, the number of cases of breast cancer and gout was too small, even though we used a large NHANES sample. Second, although we adjusted for multiple confounders, residual confounding may still exist when examining observational associations. In addition, we did not perform stratified analyses. In the MR analysis, horizontal pleiotropy could not be completely avoided due to the complexity of genetic variation, despite a series of sensitivity analyses. Finally, the participants in our cross-sectional study were from the U.S. population, whereas the population in the MR analysis were all from the European population, and there was demographic heterogeneity.

Conclusions

Our study elucidated the link between uric acid, gout and breast cancer. In summary, our findings support the antioxidant role of uric acid in inhibiting breast carcinogenesis, and provide some reference for the management of uric acid levels in hyperuricemic patients who have high risk factors for breast cancer. However, the use of high uric acid levels or gout to combat breast cancer in the general population is not advocated until the mechanisms involved are clear, and further studies could explore the mechanisms and targets to develop relevant antitumor drugs.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE-MR reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-1141/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-1141/prf

Funding: This study was financially supported by the National Natural Science Foundation of China (No. 82102937), Henan Provincial Health Commission (No. YXKC2021026), Henan Provincial Breast Cancer Precision Prevention and Control Engineering Research Center (ZC20220050), Henan Provincial People’s Hospital—23456 Talent Engineering Project (ZC23456026) and the study on the Effect and Mechanism of Macrophage-Derived Exosomes Loaded with Doxorubicin for Targeted Therapy of Triple-Negative Breast Cancer (ZC20220114), and Henan Provincial Department of Science and Technology (No. 222301420058).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-1141/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This survey has received ethical approval from the NCHS Ethics Review Board (ERB), and all subjects have provided informed consent. For more details, see this website (https://www.cdc.gov/nchs/nhanes/irba98.htm).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin 2020;70:7-30. [Crossref] [PubMed]
2. Miller KD, Nogueira L, Mariotto AB, et al. Cancer treatment and survivorship statistics, 2019. CA Cancer J Clin 2019;69:363-85. [Crossref] [PubMed]
3. Battelli MG, Bortolotti M, Polito L, et al. The role of xanthine oxidoreductase and uric acid in metabolic syndrome. Biochim Biophys Acta Mol Basis Dis 2018;1864:2557-65. [Crossref] [PubMed]
4. Roddy E, Choi HK. Epidemiology of gout. Rheum Dis Clin North Am 2014;40:155-75. [Crossref] [PubMed]
5. Ben Salem C, Slim R, Fathallah N, et al. Drug-induced hyperuricaemia and gout. Rheumatology (Oxford) 2017;56:679-88. [PubMed]
6. Dalbeth N, Gosling AL, Gaffo A, et al. Gout. Lancet 2021;397:1843-55. [Crossref] [PubMed]
7. Mahmoud HH, Leverger G, Patte C, et al. Advances in the management of malignancy-associated hyperuricaemia. Br J Cancer 1998;77:18-20. [Crossref] [PubMed]
8. Ames BN, Cathcart R, Schwiers E, et al. Uric acid provides an antioxidant defense in humans against oxidant- and radical-caused aging and cancer: a hypothesis. Proc Natl Acad Sci U S A 1981;78:6858-62. [Crossref] [PubMed]
9. Taghizadeh N, Vonk JM, Boezen HM. Serum uric acid levels and cancer mortality risk among males in a large general population-based cohort study. Cancer Causes Control 2014;25:1075-80. [Crossref] [PubMed]
10. Horsfall LJ, Nazareth I, Petersen I. Serum uric acid and the risk of respiratory disease: a population-based cohort study. Thorax 2014;69:1021-6. [Crossref] [PubMed]
11. Strasak AM, Rapp K, Hilbe W, et al. The role of serum uric acid as an antioxidant protecting against cancer: prospective study in more than 28 000 older Austrian women. Ann Oncol 2007;18:1893-7. [Crossref] [PubMed]
12. Wang W, Xu D, Wang B, et al. Increased Risk of Cancer in relation to Gout: A Review of Three Prospective Cohort Studies with 50,358 Subjects. Mediators Inflamm 2015;2015:680853. [Crossref] [PubMed]
13. Fini MA, Orchard-Webb D, Kosmider B, et al. Migratory activity of human breast cancer cells is modulated by differential expression of xanthine oxidoreductase. J Cell Biochem 2008;105:1008-26. [Crossref] [PubMed]
14. Kühn T, Sookthai D, Graf ME, et al. Albumin, bilirubin, uric acid and cancer risk: results from a prospective population-based study. Br J Cancer 2017;117:1572-9. [Crossref] [PubMed]
15. Yiu A, Van Hemelrijck M, Garmo H, et al. Circulating uric acid levels and subsequent development of cancer in 493,281 individuals: findings from the AMORIS Study. Oncotarget 2017;8:42332-42. [Crossref] [PubMed]
16. Fan K, Sun T, Yin F. J-shaped association between uric acid and breast cancer risk: a prospective case-control study. J Cancer Res Clin Oncol 2023;149:7629-36. [Crossref] [PubMed]
17. Feng Y, Fu M, Guan X, et al. Uric Acid Mediated the Association Between BMI and Postmenopausal Breast Cancer Incidence: A Bidirectional Mendelian Randomization Analysis and Prospective Cohort Study. Front Endocrinol (Lausanne) 2022;12:742411. [Crossref] [PubMed]
18. Gremke N, Griewing S, Kostev K, et al. Association between gout and subsequent breast cancer: a retrospective cohort study including 67,598 primary care patients in Germany. Breast Cancer Res Treat 2023;199:545-52. [Crossref] [PubMed]
19. Smith GD, Ebrahim S. 'Mendelian randomization': can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol 2003;32:1-22. [Crossref] [PubMed]
20. Skrivankova VW, Richmond RC, Woolf BAR, et al. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. JAMA 2021;326:1614-21. [Crossref] [PubMed]
21. Kerlikowske K, Gard CC, Tice JA, et al. Risk Factors That Increase Risk of Estrogen Receptor-Positive and -Negative Breast Cancer. J Natl Cancer Inst 2016;109:djw276. [Crossref] [PubMed]
22. Li W, Zheng Q, Xu M, et al. Association between circulating 25-hydroxyvitamin D metabolites and periodontitis: Results from the NHANES 2009-2012 and Mendelian randomization study. J Clin Periodontol 2023;50:252-64. [Crossref] [PubMed]
23. Burgess S, Scott RA, Timpson NJ, et al. Using published data in Mendelian randomization: a blueprint for efficient identification of causal risk factors. Eur J Epidemiol 2015;30:543-52. [Crossref] [PubMed]
24. Bowden J, Davey Smith G, Haycock PC, et al. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol 2016;40:304-14. [Crossref] [PubMed]
25. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015;44:512-25. [Crossref] [PubMed]
26. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol 2013;37:658-65. [Crossref] [PubMed]
27. Verbanck M, Chen CY, Neale B, et al. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet 2018;50:693-8. [Crossref] [PubMed]
28. Sanderson E, Davey Smith G, Windmeijer F, et al. An examination of multivariable Mendelian randomization in the single-sample and two-sample summary data settings. Int J Epidemiol 2019;48:713-27. [Crossref] [PubMed]
29. Waring WS, Webb DJ, Maxwell SR. Systemic uric acid administration increases serum antioxidant capacity in healthy volunteers. J Cardiovasc Pharmacol 2001;38:365-71. [Crossref] [PubMed]
30. Peden DB, Hohman R, Brown ME, et al. Uric acid is a major antioxidant in human nasal airway secretions. Proc Natl Acad Sci U S A 1990;87:7638-42. [Crossref] [PubMed]
31. Becker BF. Towards the physiological function of uric acid. Free Radic Biol Med 1993;14:615-31. [Crossref] [PubMed]
32. Itahana Y, Han R, Barbier S, et al. The uric acid transporter SLC2A9 is a direct target gene of the tumor suppressor p53 contributing to antioxidant defense. Oncogene 2015;34:1799-810. [Crossref] [PubMed]
33. Kang DH, Park SK, Lee IK, et al. Uric acid-induced C-reactive protein expression: implication on cell proliferation and nitric oxide production of human vascular cells. J Am Soc Nephrol 2005;16:3553-62. [Crossref] [PubMed]
34. Fini MA, Elias A, Johnson RJ, et al. Contribution of uric acid to cancer risk, recurrence, and mortality. Clin Transl Med 2012;1:16. [Crossref] [PubMed]
35. Kang DH, Ha SK. Uric Acid Puzzle: Dual Role as Anti-oxidantand Pro-oxidant. Electrolyte Blood Press 2014;12:1-6. [Crossref] [PubMed]