Gout is a common inflammatory disease characterized by acute arthritis and hyperuricemia. A number of epidemiological studies have suggested the critical role of gout in carcinogenesis. The aim of this study was to estimate the association between gout and cancer risk by meta-analysis of all relevant studies published to date. A comprehensive literature search in PubMed and Embase databases from their inception up to July 1, 2014, was performed to identify eligible studies. The strength for relationship between gout and the risk of different cancers was evaluated by calculating pooled relative risks (RRs) with 95% confidence intervals (95% CIs). All analyses were carried out by STATA 12.0 software. Gout patients were at an increased risk of cancer, particularly urological cancers, digestive system cancers, and lung cancer. No such significant association between gout and the risk of breast or brain cancers was observed. Sensitivity analysis did not materially alter the pooled results. Gout is a risk factor of cancer, particularly that of urological cancers, digestive system cancers, and lung cancer. The pooled data further support the hypothesis of a link between gout and carcinogenesis.
Gout is a common inflammatory disease characterized by acute arthritis and hyperuricemia due to disorder of purine metabolism [
During the past few years, the association between cancer and gout as well as hyperuricemia has drawn a lot of attention. High cell turnover can lead to hyperuricemia and tumorigenesis, implicating an underlying link between purine metabolism disorders and cancer [
Although recent epidemiological studies have sparked interest in the hypothesis, the precise association between gout and cancer risk remains obscure. Our study was firstly aimed to investigate this association by meta-analysis of all currently available data.
We performed a comprehensive literature search in PubMed and Embase databases from their inception up to July 1, 2014, for studies on the association between gout and cancer risk. The following search terms were used: gout, or hyperuricemia; cancer, tumor, or carcinoma; risk, or incidence. The references of retrieved studies were also screened for additional publications. No language restriction was imposed.
The included studies must meet the inclusion criteria as follows: (1) studies concerning the association between gout and cancer risk; (2) studies in a cohort design; (3) studies with enough information for odds ratio (ORs), relative risks (RRs), or hazard ratios (HRs) with corresponding 95% confidence intervals (95% CIs).
Studies were excluded if they were (1) publications not related to the role of gout in cancer risk; (2) case-only study; (3) case report; (4) reviews; (5) animal studies; (6) studies with overlapping data.
Two investigators independently extracted available data from all studies as follows: first author’s name, year of publication, origins, diagnosis of patients, study design, type of cancers, total sample size, study period, adjusted factors, baseline time, follow-up duration, mean age of gout patients, and RRs or HRs or ORs with corresponding 95% CIs. Disagreements were resolved by consensus.
The strength for relationship between gout and cancer risk was estimated by calculating the pooled RRs with 95% CIs. The between-study heterogeneity was evaluated by Cochran’s Q-statistic test and
After a comprehensive literature search, three prospective cohort studies with a total of 50, 358 subjects were retrieved for meta-analysis [
Characteristics of all studies.
Study | Year | Origins | Number of subjects | Follow-up duration (years) | Baseline time | Mean age of patients | Cancer sites | Adjusted factors |
---|---|---|---|---|---|---|---|---|
Chen et al. [ |
2014 | Taiwan | 9,413 | 8 | 1998–2000 | 51.03 ± 14.52 | Prostate, bladder, kidney, colorectum, liver, gallbladder and bile duct, stomach, lung, esophageal, pancreas, nasopharyngeal, brain, and oral, and so forth | Age, sex, year, and month of first diagnosis |
Kuo et al. [ |
2012 | Taiwan | 24,088 | 7.8-7.9 | 2000–2008 | 42.3 ± 16.3 | Prostate, bladder, colon, liver, stomach, lung, and breast, and so forth | Age and sex |
Bofetta et al. [ |
2009 | Sweden | 10,500 | NR | 1965–1995 | NR | Prostate, bladder, kidney, rectum, colon, liver and bile duct, stomach, lung, pancreas, brain, and oral, and so forth | Sex, time since first hospitalization, and gout as primary or the only diagnosis |
Bofetta et al. [ |
2009 | Sweden | 6,357 | NR | 1965–1995 | NR | Bladder, kidney, rectum, colon, liver and bile duct, stomach, lung, pancreas, brain, oral, breast, and so forth | Sex, time since first hospitalization, and gout as primary or the only diagnosis |
Gout was related to a significantly increased risk of cancer as suggested by overall analysis of four independent cohort studies (RR = 1.42, 95% CI 1.09–1.84,
Summary of meta-analysis results.
Type of cancer | aRR [95% CI] |
|
|
|
---|---|---|---|---|
All | 1.42 [1.09–1.84] | 0.008 | 98.1 | <0.001 |
Males | 1.67 [0.93–3.01] | 0.087 | 99.0 | <0.001 |
Urological cancers | 1.72 [1.30–2.26] | <0.001 | 90.7 | <0.001 |
Digestive system cancers | 1.39 [1.23–1.56] | <0.001 | 68.2 | <0.001 |
Lung cancer | 1.29 [1.01–1.65] | 0.039 | 75.5 | 0.007 |
Breast cancer | 0.92 [0.77–1.09] | 0.336 | 45.9 | 0.174 |
Brain cancer | 1.24 [0.82–1.87] | 0.309 | 0.0 | 0.946 |
Forest plot for gout and cancer risk.
An elevated risk of urological cancers was observed to be associated with gout, although obvious heterogeneity existed among included studies (RR = 1.72, 95% CI 1.30–2.26,
For the risk of digestive system cancers, the pooled RRs suggested that gout exerted risk effect on the development of digestive system cancers (RR = 1.39, 95% CI 1.23–1.56,
A modest association between gout and lung cancer risk was demonstrated, which suggested a risk factor of gout for the development of lung cancer (RR = 1.29, 95% CI 1.01–1.65,
No significant relationship between gout and the risk of breast cancer or brain cancer was observed in our study (Table
Begg’s funnel plots and Egger’s test did not show any evidence for publication bias risk in the present meta-analysis (data not shown).
The present meta-analysis based on three prospective cohort studies shows the evidence that gout confers risk effect on carcinogenesis, particularly in urological cancers, digestive system cancers, and lung cancer. Besides, gout did not modify the risk of breast cancer or brain cancer or total cancer risk in males, as suggested by the pooled RRs, respectively. Nevertheless, the findings about breast and brain cancers risk must be interpreted with caution due to limited sample size. Sensitivity analysis did not materially alter the pooled results.
Gout is a disorder of purine metabolism, which has been suggested in relation to many kinds of diseases, such as end-stage renal disease, myocardial infarction, obesity, and diabetes, due to increased serum uric acid [
Up till now, three prospective cohort studies with a total of 50,358 subjects have been performed to evaluate the association between gout and cancer risk [
Some limitations must be seriously considered. Firstly, only three cohort studies were included into our study, but the statistical power was sufficient in estimating the role of gout in carcinogenesis. More relevant studies on the association between gout and individual cancers risk, especially brain and breast cancers, are warranted for further investigation. Secondly, adjusted factors of all included studies were not consistent. Thus, bias might be introduced in our study because of confounding factors such as age, sex, and ethnicity. Future studies based on adjusted analysis are encouraged to investigate the effect of gout disease on cancer risk. Thirdly, the influence of gout disease duration in tumorigenesis was elucidated in a recent study [
In summary, our study does show evidence for the hypothesis of a link between gout and carcinogenesis. It suggests that gout is an independent risk factor for the incidence of total cancer, particularly urological cancers, digestive system cancers, and lung cancer. The precise association between gout and individual cancers warrants further investigation by more epidemiological studies with high quality.
Relative risks
Odds ratio
Hazard ratios
95% confidence intervals.
The authors declare that there is no conflict of interests regarding the publication of this paper.
Weijie Wang, Donghua Xu, Xiaoyang Sun, Beicheng Sun, and Xuehao Wang conceived and designed the experiments. Bin Wang, Shushan Yan, and Xiaochen Wang performed the experiments and collected data. Weijie Wang, Donghua Xu, Bin Wang, and Shushan Yan analyzed the data. Xiaochen Wang and Yin Yin contributed to reagents/materials/analysis tools. Weijie Wang, Donghua Xu, Xiaoyang Sun, and Beicheng Sun wrote the paper. Weijie Wang, Donghua Xu, Bin Wang, Shushan Yan, and Xiaochen Wang equally contributed to this work.