Additive Association of ABCG2 rs4148155 and SLC22A12 rs75786299 Polymorphisms with Hyperuricemia, Gout, and Nephrolithiasis: A Hospital-Based, Case-Control Study

Abstract

Introduction

This study aimed to investigate the interaction between the ABCG2 rs4148155 and SLC22A12 rs75786299 variants and their association with incident gout and nephrolithiasis in the Taiwanese population to better understand the genetic loci regulating hyperuricemia and their contribution to nephrolithiasis development.

Methods

This retrospective case-control study involved 35,280 adults from the Taiwan Precise Medicine Initiative database. We examined the prevalence of gout and ultrasound confirmed nephrolithiasis as the primary and secondary outcome. Logistic regression models were used to explore the associations between genetic variants, serum uric acid levels, incident gout, and nephrolithiasis.

Results

The frequencies of the rs4148155 variant and the rs75786299 variant were 63.2% and 3.7%, respectively. Among participants, 7,056 were gout, and 4,110 had nephrolithiasis. Multivariable odds ratios (ORs) for gout were 1.67 and 1.96 among rs4148155 and rs75786299 carriers, respectively (p = 0.01 and p < 0.001). For nephrolithiasis, the multivariable ORs were 1.1 and 1.11 for rs4148155 and rs75786299 carriers, respectively (p = 0.004 and p = 0.32). Sex-stratified analysis revealed an additive risk of gout and nephrolithiasis among carriers of these genetic variants, regardless of gender. Independent risk factors for nephrolithiasis included higher age, male gender, and the presence of gout, hypertension, and hyperlipidemia.

Conclusion

The study highlights a significant association between the rs4148155, rs75786299 variants and the development of gout and nephrolithiasis, indicating an additive risk among carriers. These findings support precision healthcare approaches for individuals with risk genetic variants to target hyperuricemia, gout, and systemic comorbidities, ultimately preventing nephrolithiasis.

Introduction

Nephrolithiasis is clinically characterized by acute flank colicky pain, abdominal pain, nausea, vomiting, and hematuria. Its prevalence in the USA has significantly risen from 3.2% in 1980 to 10.6% in 2018, resulting in a substantial increase in associated healthcare costs [1–4]. The etiology of nephrolithiasis is multifactorial, influenced by lifestyle, genetics, diet, environmental factors, and systemic comorbidities [3, 5].

In patients with gout, urate stones account for a major proportion of urolithiasis up to 52.2% of cases [6]. Risk factors for urate stone formation include hyperuricemia and gout, which are influenced by low urinary pH levels, hypovolemia, and hyperuricosuria [7]. Additionally, patients with hyperuricemia and gout may develop calcium nephrolithiasis due to calcium oxalate precipitation under the conditions of hyperuricosuria and hypercalciuria [7]. Genetic studies, including genome-wide association studies, have identified loci associated with hyperuricemia and gout across different populations [8]. Among these, the ABCG2 (adenosine triphosphatase-binding cassette sub-family G member 2) gene and the SLC22A12 (solute carrier family 22 member 12) gene play pivotal roles in urate transport and elimination [9, 10]. The ABCG2 gene encodes the breast cancer resistance protein (BCRP), a transporter implicated in urate excretion through renal tubules and the gastrointestinal tract [11, 12]. Polymorphisms in ABCG2 are strongly associated with hyperuricemia, early-onset gout, and tophi formation [8, 11–16]. Notably, the ABCG2 rs2231142 polymorphism has been linked to an increased risk of nephrolithiasis in men with hyperuricemia [17]. However, the role of the ABCG2 rs4148155 variant in hyperuricemia and gout risk remains uncertain among Taiwanese population.

SLC22A12, which encodes urate transporter 1 (URAT1), is implicated in both hyperuricemia and hereditary renal hypouricemia type 1 (RHUC1) [8, 10, 18]. Loss-of-function mutations cause excessive renal urate excretion, low serum urate, and susceptibility to exercise-induced kidney injury and urolithiasis, whereas gain-of-function variants reduce urate clearance, leading to hyperuricemia and gout [10, 18]. This bidirectional role highlights the gene’s importance in urate homeostasis and its relevance to gout and nephrolithiasis risk in different populations [18–20]. The rs75786299 (IVS3+11A/G) of SLC22A12 has shown a strong association with hyperuricemia, with an odds ratio (OR) of 32.05 [21], suggesting a potential regulatory effect on URAT1 expression or splicing. Despite this, it remains unclear whether this genetic polymorphism contributes to hyperuricemia and gout in the Taiwanese population.

ABCG2 knockout in mice reduces intestinal urate excretion and renal URAT1 expression, increasing the risk of renal overload hyperuricemia [10, 12]. The ABCG2 rs4148155 variant, although noncoding, is in strong linkage disequilibrium with the functional rs2231142 (Q141K) variant, which reduces urate transport activity [12, 22]. Elevated uricosuria may promote urate and calcium stone formation, particularly at low urine pH [7, 23], suggesting that ABCG2 rs4148155 and SLC22A12 rs75786299 variants may increase susceptibility to gout and nephrolithiasis.

Given their previously reported associations and relatively high allele frequencies in East Asian populations, we selected these two noncoding variants to clarify their potential roles in hyperuricemia, gout, and nephrolithiasis in the Taiwanese population. Accordingly, this study aims to investigate the association of ABCG2 rs4148155 and SLC22A12 rs75786299 genetic variants with the risk of hyperuricemia and gout in a community-based Taiwanese population and further examines the relationship between these variants and nephrolithiasis among patients affected by these conditions.

Methods

Data Sources and Ethics Statement

This retrospective case-control study utilized data from the Taiwan Precision Medicine Initiative (TPMI), a nationwide genetic program coordinated by Academia Sinica in collaboration with partner hospitals. At Taichung Veterans General Hospital (TCVGH), the TPMI enrolled 58,091 participants aged 18 years and older who visited 28 surgical and medical outpatient clinics between June 2019 and December 2021. Comprehensive data, including electronic health records and biological specimens, were collected. The study protocol was reviewed and approved by the Ethics Committee of the Institutional Review Board of TCVGH, Taiwan (IRB No. CE222205A-2). Written informed consent was obtained from all participants following principles of Declaration of Helsinki.

Participants

Of the total participants, 7,056 patients with a clinical diagnosis of gout (ICD-9-CM code 274.0, 274.81, 274.82, 274.9/ICD-10-CM code M10.00, M10.9) or hyperuricemia (ICD-9-CM code 790.6/ICD-10-CM code E79.0, M10.9) during hospitalization or outpatient visit were designed as the case group. A total of 28,224 individuals without a medical history of gout or hyperuricemia were matched to the case group by sex and age at a ratio of 1:4, forming the control group. Genetic profiles of the participants were linked to their medical claims data at TCVGH. The dataset included demographic information (e.g., sex and age), lifestyle factors (e.g., alcohol consumption, smoking and physical activity), physical examination metrics (e.g., body mass index [kg/m²]), diagnoses, medical procedures, surgeries, prescribed medications, and biochemical data (e.g., serum uric acid [UA] levels and estimated glomerular filtration rate [eGFR; mL/min/1.73 m²]).

Genotyping and Quality Controls

Genomic DNA was isolated from 58,091 participants enrolled in the TPMI program using DNA isolation kits from TIANGEN Biotech (Beijing, China). DNA concentrations were measured using a NanoDrop 2000 Spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). Genotyping was performed with the Axiom Genome-Wide TWB 2.0 Array Plate (Affymetrix, Santa Clara, CA, USA), comprising 714,431 single-nucleotide polymorphisms (SNPs) specifically tailored for Taiwan’s Han Chinese population [24]. The use of high-coverage SNP data from large-scale Han Chinese ancestry in Taiwan using custom arrays has been previously detailed [25].

Genotyping quality control procedures were implemented, which involved the exclusion of SNPs displaying a sole allele occurrence within the cohort, as well as those with a total call rate below 95% or a total minor allele frequency (MAF) below 0.01. Furthermore, SNPs deviating significantly from the Hardy-Weinberg equilibrium (p < 1 × 10⁻⁵) were also excluded. Genetic data quality control and statistical analyses were conducted using the PLINK 1.9 software package. Following quality control, 591,048 SNPs were retained for subsequent analyses.

To provide a comprehensive assessment of the genetic associations within the ABCG2 and SLC22A12 loci, all variants located within the gene boundaries and their flanking 10-kb regions were extracted from the TWB 2.0 array data after quality control. Each variant was tested for association with gout or hyperuricemia using a logistic regression model, adjusting for age and sex. The resulting ORs, 95% confidence intervals (CIs), and p values are summarized in online supplementary Table 1 (for all online suppl. material, see https://doi.org/10.1159/000550471) (ABCG2) and online supplementary Table 2 (SLC22A12).

Participants with available genotyping data for ABCG2 rs4148155 and SLC22A12 rs75786299 were selected for further analysis. Among these, 2,595 individuals with the AA genotype of ABCG2 rs4148155 were classified as noncarriers, while 6,775 individuals with the GG genotype of SLC22A12 rs75786299 were classified as noncarriers. Additionally, 3,379 individuals with the AG genotype of ABCG2 rs4148155 were identified as heterozygous carriers, and 1,081 individuals with the GG genotype were identified as homozygous carriers. Regarding SLC22A12 rs75786299, 264 individuals with the GA genotype were classified as heterozygous carriers. Notably, no homozygous carriers of the SLC22A12 rs75786299 variant were identified in this study.

Covariates

Nephrolithiasis was defined as the presence of kidney stones confirmed through ultrasonography and a clinical diagnosis coded as ICD-9-CM 592.0, 594.1, V13.0, V18.69 or ICD-10-CM N20.0, N21.0, Z84.1, Z87.442. Specifically, nephrolithiasis was identified by the presence of echogenic foci measuring ≥3 mm with acoustic shadowing and twinkle artifact on color Doppler imaging within the renal sinus [26]. Renal calculi associated with conditions such as double-layer vessel calcification, cystic calcification, and mucous calcification were excluded through urinary tract ultrasonography [27]. Comorbidities, including hyperlipidemia (ICD-9-CM code 272, ICD-10-CM code E78.1-E78.5), hypertension (ICD-9-CM code 401–405, ICD-10-CM code I10-I15), diabetes mellitus (ICD-9-CM code 250, ICD-10-CM code E10.9 and E11.9), and chronic kidney disease (CKD) (ICD-9-CM code 585.9, ICD-10-CM code N18.9) were identified if diagnostic codes were recorded at least once during hospitalization or twice in the outpatient services. Patients receiving urate-lowering therapy (e.g., allopurinol, benzbromarone, febuxostat) or undergoing procedures such as arthrocentesis, intra-articular injection, or ultrasound-guided aspiration or injection for gout were identified via medical record review.

Statistical Analysis

The data analysis in this study was conducted using SAS software version 9.4 (SAS Institute Inc., Cary, NC). Descriptive statistics, including mean, standard deviations, and percentages, were utilized to describe the basic characteristics of participants across different groups. The chi-square test was employed to evaluate the statistical significance of categorical variables, while continuous variables were presented as mean ± standard deviation and analyzed using Student’s t test for normally distributed data. Univariable and multivariable logistic regression analyses were performed to investigate the association between ABCG2 rs4148155, and SLC22A12 rs75786299 variants and the risk of gout and nephrolithiasis.

Both additive [28, 29] and multiplicative interaction analyses [29] were performed to evaluate gene-gene interactions between ABCG2 rs4148155 and SLC22A12 rs75786299. Additive interactions were assessed using the relative excess risk due to interaction (RERI), attributable proportion due to interaction (AP), and synergy index (SI). The RERI was calculated on the additive scale using the ORs obtained from the fully adjusted multivariable logistic regression model, according to the formula:

$$\text{RERI}=\text{OR}\_\left\{11\right\}-\text{OR}\_\left\{10\right\}-\text{OR}\_\left\{01\right\}+1$$

RERI = 0, AP = 0, or SI = 1 indicates no interaction (exact additivity); positive values (RERI > 0, AP > 0, or SI > 1) indicate positive interaction or synergistic effects beyond additivity; negative values indicate negative interaction or less than additivity. The 95% CIs for these measures were estimated using the method described by Andersson et al. [30], implemented with the “epi.interaction” function in the epiR R package (https://cran.r-project.org/package=epiR). Multiplicative interactions were evaluated by including an interaction term (ABCG2 rs4148155 × SLC22A12 rs75786299) in the same multivariable logistic regression model. The final regression model incorporated significant covariates. A p value <0.05 was considered indicative of statistical significance.

Results

Baseline Demographics between Participants with or without Gout

In this study, a cohort of 35,280 individuals, comprising 9,725 women and 25,555 men, was analyzed. The study population was categorized into a case group of 7,056 participants diagnosed with gout (5,199 men and 1,857 women) and a control group of 28,224 participants without gout (20,356 men and 7,868 women). The prevalence of gout in the cohort was 19.10% among women and 20.34% among men. Baseline characteristics of the participants are detailed in Table 1. Gout cases were older (64.40 ± 14.62 vs. 62.37 ± 14.72, p < 0.0001), had lower eGFR (71 ± 32.27 vs. 90 ± 27.6, p < 0.0001), and included a higher proportion of men compared with women (73.7% vs. 26.3%, p = 0.009). Lifestyle factors such as regular physical exercise (33.6% vs. 19.1%, p = 0.0001) and being overweight (69.5% vs. 59.9%, p < 0.0001) were more prevalent in gout cases. Furthermore, comorbidities including hyperlipidemia (64.3% vs. 34.5%, p < 0.0001), hypertension (68.9% vs. 38.9%, p < 0.0001), diabetes mellitus (44.2% vs. 30.8%, p < 0.0001), and CKD (62.02% vs. 20.43%, p < 0.0001) were significantly more common in individuals with gout. The prevalence of ABCG2 rs4148155 carriers (AG + GG, dominant model) was significantly higher in gout cases than controls (63.2% vs. 51.7%, p < 0.0001). Similarly, SLC22A12 rs75786299 carriers (GA + AA, dominant model) were more frequent among gout cases (3.8% vs. 2.1%, p < 0.0001), indicating a dominant genetic effect on gout susceptibility.

Table 1.

Baseline demographics, comorbidities, and genetic variants between participants with or without gout

Variables	With gout (n = 7,056)		Without gout (n = 28,224)		p value
	n	%	n	%
Age (mean ± SD)a, years	64.40±14.62		62.37±14.72		<0.0001
eGFR (mean ± SD)a, mL/min/1.73 m2	71±32.27		90±27.6		<0.0001
Gender	​	​	​	​	0.009
Female	1,857	26.3	7,868	27.9	​
Male	5,199	73.7	20,356	72.1	​
Overweight (BMI ≥24 kg/m2)b	​	​	​	​	<0.0001
No	2,017	30.5	10,026	40.1	​
Yes	4,592	69.5	14,978	59.9	​
Smokingb	​	​	​	​	0.37
Never	5,605	79.5	22,358	79.9	​
Ever or current	1,450	20.6	5,616	20.1	​
Alcohol consumptionb	​	​	​	​	0.76
Former drinker	3,297	92.1	12,019	92.2	​
Current drinker	284	7.9	1,013	7.8	​
Regular physical exerciseb	​	​	​	​	0.0001
No	89	66.4	736	80.9	​
Yes	45	33.6	174	19.1	​
Hyperlipidemiab	​	​	​	​	<0.0001
No	2,516	35.7	18,499	65.5	​
Yes	4,540	64.3	9,725	34.5	​
Hypertensionb	​	​	​	​	<0.0001
No	2,198	31.2	17,257	61.1	​
Yes	4,858	68.9	10,967	38.9	​
Diabetes mellitusb	​	​	​	​	<0.0001
No	3,938	55.8	19,530	69.2	​
Yes	3,118	44.2	8,694	30.8	​
CKDb	​	​	​	​	<0.0001
No	2,680	37.98	22,459	79.57	​
Yes	4,376	62.02	5,765	20.43	​
ABCG2 rs4148155b	​	​	​	​	<0.0001
AA (noncarrier)	2,595	36.8	13,634	48.3	​
AG + GG (carrier)	4,460	63.2	14,576	51.7	​
SLC22A12 rs75786299b
GG (noncarrier)	6,775	96.2	27,569	97.9	<0.0001
GA + AA (carrier)	264	3.8	583	2.1	​

BMI, body mass index; eGFR, estimated glomerular filtration rate.

^aContinuous variables were expressed as mean ± SD and were analyzed using Student’s t test for normal data distributions.

^bCategorical variables were expressed as numbers (percent) and analyzed using the chi-square test.

Baseline Demographics between Participants with or without Nephrolithiasis

Table 2 presents the baseline characteristics and the incidence of nephrolithiasis among 35,280 participants (9,725 women and 25,555 men). Nephrolithiasis was diagnosed in 4, 110 individuals, comprising 3,170 men and 940 women, corresponding to prevalence rates of 12.40% in men and 9.67% in women. Individuals with nephrolithiasis were older (64.61 ± 12.68 vs. 62.53 ± 14.96, p < 0.0001) and had lower eGFR (83 ± 29.45 vs. 86 ± 29.67, p < 0.0001) compared with those without nephrolithiasis. Furthermore, men had a significantly higher risk of nephrolithiasis compared to women (77.1% vs. 22.9%, p < 0.0001).

Table 2.

Baseline demographics, comorbidities, and genetic variants between participants with or without nephrolithiasis

Variable	With nephrolithiasis (n = 4,110)		Without nephrolithiasis (n = 31,170)		p value
	n	%	n	%
Age (mean ± SD)a, years	64.61±12.68		62.53±14.96		<0.0001
eGFR (mean ± SD)a, mL/min/1.73 m2	83±29.45		86±29.67		<0.0001
Gender
Female	940	22.9	8,785	28.2	<0.0001
Male	3,170	77.1	22,385	71.8	​
Comorbidityb
Hyperlipidemia	2,044	49.73	12,221	39.21	<0.0001
Hypertension	2,332	56.74	13,493	43.29	<0.0001
Diabetes mellitus	1,563	38.03	10,249	32.88	<0.0001
CKD	1,759	42.8	8,382	26.89	<0.0001
Uric acid (highest)a
<5	248	7.6	2,273	12.0	<0.0001
5–7	1,066	32.6	6,823	35.9	​
>7	1,960	59.9	9,919	52.2	​
Goutb
No	2,754	67.0	25,470	81.7	<0.0001
Yes	1,356	33.0	5,700	18.3	​
ABCG2 rs4148155b
AA (noncarrier)	1,757	42.7	14,472	46.5	<0.0001
AG + GG (carrier)	2,353	57.3	16,683	53.5	​
SLC22A12 rs75786299b
GG (noncarrier)	3,980	97.2	30,364	97.6	0.060
GA + AA (carrier)	116	2.8	731	2.4	​
Urate-lowering therapyb
Allopurinol	524	12.75	1,661	5.33	<0.0001
Benzbromarone	567	13.8	2,417	7.75	<0.0001
Febuxostat	835	20.75	3,414	10.95	<0.0001
Treatmentb
Arthrocentesis	86	2.09	256	0.82	<0.0001
Intra-articular injection	164	3.99	539	1.73	<0.0001
Ultrasound-guided aspiration or injection	370	9.00	1,753	5.62	<0.0001
MSK ultrasound	1,011	24.6	3,100	9.95	<0.0001

MSK, musculoskeletal; eGFR, estimated glomerular filtration rate.

^aContinuous variables were expressed as mean ± SD and were analyzed using Student’s t test for normal data distributions.

^bCategorical variables were expressed as numbers (percent) and were analyzed using the chi-square test.

Individuals with nephrolithiasis had a higher frequency of ABCG2 rs4148155 carriers (AG + GG) compared with controls (57.3% vs. 53.5%, p < 0.0001), whereas no significant difference was observed for SLC22A12 rs75786299 carriers (2.8% vs. 2.4%, p = 0.060). Nephrolithiasis was also associated with higher prevalences of gout (33.0% vs. 18.3%, p < 0.001), hyperuricemia (59.9% vs. 52.2%, p < 0.0001), hyperlipidemia (49.7% vs. 39.2%, p < 0.0001), hypertension (56.7% vs. 43.3%, p < 0.0001), diabetes mellitus (38.0% vs. 32.9%, p < 0.0001), and CKD (42.8% vs. 26.9%, p < 0.0001) compared to controls. Correspondingly, more frequent use of urate-lowering therapy (allopurinol, benzbromarone, or febuxostat; p < 0.0001), arthrocentesis (2.1% vs. 0.8%, p < 0.001), intra-articular injection (4.0% vs. 1.7%, p < 0.001), and ultrasound-guided aspiration or injection (9.0% vs. 5.6%, p < 0.001) were observed in these individuals.

Association of Gout and Nephrolithiasis Risk with Demographic Factors, Systemic Comorbidities, and Genotypic Variants

Figure 1 demonstrates the univariable OR for gout and nephrolithiasis risk associated with rs4148155 and rs75786299 polymorphisms. To further explore these relationships, a multivariable-adjusted logistic regression model was developed (Table 3), incorporating demographic factors, systemic comorbidities, and the rs4148155(G) and rs75786299(A) variants. Adjustments were made for age, gender, gout, diabetes, hypertension, hyperlipidemia, CKD, eGFR and medication use. Compared to noncarriers, ABCG2 rs4148155 and SLC22A12 rs75786299 carriers had significantly increased risks of gout with ORs of 1.67 (95% CI: 1.57 to 1.76, p < 0.0001) and 1.96 (95% CI: 1.67 to 2.30, p < 0.0001), respectively. Male sex, hypertension, and hyperlipidemia were additional significant risk factors for gout with ORs (95% CI) of 1.15 (1.08 to 1.23), 2.73 (2.56 to 2.91), and 2.57 (2.42 to 2.73), respectively (all p < 0.0001). ABCG2 rs4148155 carriers also had an increased risk of nephrolithiasis (OR = 1.10, 95% CI: 1.03 to 1.18, p = 0.004), whereas SLC22A12 rs75786299 carriers showed no significant association (OR = 1.11, 95% CI: 0.91 to 1.36, p = 0.32). Beyond genetic factors, higher age, male sex, gout, hyperlipidemia, and hypertension were identified as significant risk factors, with ORs (95% CI) of 1.01 (1.00 to 1.01), 1.39 (1.29 to 1.50), 1.87 (1.74 to 2.02), 1.19 (1.10 to 1.28), and 1.37 (1.27 to 1.47), respectively (all p < 0.0001) (Table 3).

Fig. 1.

Risk of gout (left panel) and nephrolithiasis (right panel) among the carriers of ABCG2 rs4148155 and SLC22A12 rs75786299 genotypic variants. Error bars represent the 95% confidence intervals (CIs) of the odds ratios (ORs).

Table 3.

Association of gout and nephrolithiasis risk with demographic factors, systemic comorbidities, and ABCG2 rs4148155, SLC22A12 rs75786299 variants

Gout risk	OR	95% CI		p valuea
Age	0.99	0.99	1.00	<0.0001
Sex	1.15	1.08	1.23	<0.0001
Diabetes	0.99	0.93	1.05	0.79
Hypertension	2.73	2.56	2.91	<0.0001
Hyperlipidemia	2.57	2.42	2.73	<0.0001
ABCG2 rs4148155	1.67	1.57	1.76	<0.0001
SLC22A12 rs75786299	1.96	1.67	2.3	<0.0001

Nephrolithiasis risk	OR	95% CI		p valueb
Age	1.01	1.00	1.01	<0.0001
Sex	1.39	1.29	1.50	<0.0001
Gout	1.87	1.74	2.02	<0.0001
Hyperlipidemia	1.19	1.10	1.28	<0.0001
Hypertension	1.37	1.26	1.47	<0.0001
Diabetes	0.96	0.89	1.03	0.27
ABCG2 rs4148155	1.10	1.03	1.18	0.004
SLC22A12 rs75786299	1.11	0.91	1.36	0.32

Results obtained from multivariable logistic regression analyses.

CI, confidence interval.

^aOR adjusted for all variables in the table by age, gender, diabetes, hypertension, hyperlipidemia, CKD, eGFR, and medication use.

^bOR adjusted for all variables in the table by age, gender, gout, hyperlipidemia, hypertension, diabetes, CKD, eGFR, and medication use.

To further clarify the individual contribution of each variant and assess cumulative effects, participants were stratified into four mutually exclusive genotype groups (online suppl. Table 3). Compared to individuals carrying neither variant, the multivariable-adjusted ORs (95% CI) for gout risk were 1.61 (1.52–1.70) for ABCG2 rs4148155 carriers only, 1.84 (1.59–2.14) for SLC22A12 rs75786299 carriers only, and 3.59 (2.92–4.41) for carriers of both variants (all p < 0.0001). These findings demonstrate that both variants independently contribute to gout risk, with a significant dose-response relationship (p for trend <0.0001) and the highest risk observed in double carriers. For nephrolithiasis risk, ABCG2 rs4148155 carrier had a significant association (OR = 1.16, 95% CI: 1.09 to 1.24, p < 0.0001), whereas SLC22A12 rs75786299 carrier showed a marginally nonsignificant trend (OR = 1.21, 95% CI: 0.99 to 1.48, p = 0.06), likely due to the limited number of carriers. Nevertheless, individuals carrying both variants demonstrated significantly elevated risk (OR = 1.40, 95% CI: 1.09 to 1.81, p = 0.009), suggesting a cumulative effect of these genetic variants on nephrolithiasis susceptibility.

Gender Differences in Hyperuricemia, Gout, and Nephrolithiasis across the Additive Composite of the ABCG2 rs4148155(G) and SLG22A12 rs75786299(A) Genotypes

Sex-specific frequency of the ABCG2 rs4148155 and SLC22A12 rs75786299 variants were analyzed (Table 4) in relation to hyperuricemia status (defined as serum uric acid level above 7 mg/dL), gout risk, and nephrolithiasis risk using multiple logistic regression. Among females with hyperuricemia, carrier frequencies were significantly higher compared to non-hyperuricemia (59.5% vs. 51.84%, p < 0.0001) for ABCG2 rs4148155 AG + GG genotype and (2.97% vs. 2.14%, p = 0.03) for SLC22A12 rs75786299 GA + AA genotype. In males with hyperuricemia, the corresponding frequencies for ABCG2 rs4148155 carrier and SLC22A12 rs75786299 carrier were 64.54% vs. 51.6% and 4.03% vs. 2.04%, respectively (all p < 0.0001).

Table 4.

Association of ABCG2 rs4148155, SLC22A12 rs75786299 variants, hyperuricemia gout, and nephrolithiasis with sex stratification

Variables	Female				p value	Female			p valuea	Female			p valueb
	with HUA (n = 1,857)		without HUA (n = 7,868)			risk of gout				risk of nephrolithiasis
	n	%	n	%		OR	95% CI			OR	95% CI
ABCG2 rs4148155	​	​	​	​	<0.0001	​	​	​	​	​	​	​	​
AA (noncarrier)	752	40.5	3,788	48.16	​	1	–	–	​	1	–	–	​
AG + GG (carrier)	1,105	59.5	4,078	51.84	​	1.4	1.26	1.56	<0.0001	1.06	0.93	1.22	0.4
SLC22A12 rs75786299	​	​	​	​	0.03	​	​	​	​	​	​	​	​
GG (noncarrier)	1,798	97.03	7,684	97.86	​	1	–	–	​	1	–	–	​
GA + AA (carrier)	55	2.97	168	2.14	​	1.49	1.07	2.07	0.02	1.36	0.91	2.03	0.14

Variables	Male				p value	Male			p valuea	Male			p valueb
	with HUA (n = 5,199)		without HUA (n = 20,356)			risk of gout				risk of nephrolithiasis
	n	%	n	%		OR	95% CI			OR	95% CI
ABCG2 rs4148155	​	​	​	​	<0.0001	​	​	​	​	​	​	​	​
AA (noncarrier)	1,843	35.46	9,846	48.4	​	1	–	–	​	1	–	–	​
AG + GG (carrier)	3,355	64.54	10,498	51.6	​	1.78	1.67	1.91	<0.0001	1.2	1.11	1.29	<0.0001
SLC22A12 rs75786299	​	​	​	​	<0.0001	​	​	​	​	​	​	​	​
GG (noncarrier)	4,977	95.97	19,885	97.96	​	1	–	–	​	1	–	–	​
GA + AA (carrier)	209	4.03	415	2.04	​	2.17	1.81	2.6	<0.0001	1.16	0.92	1.46	0.2

Data are expressed as number (percentage).

The relationship between categorical variables was ascertained by the chi-square test.

CI, confidence interval; HUA, hyperuricemia.

^aOR adjusted for all variables (age, diabetes mellitus, hypertension, hyperlipidemia, CKD, eGFR, and medication use).

^bOR adjusted for all variables in the table by age, gout, hyperlipidemia, hypertension, diabetes, CKD, eGFR, and medication use.

Regarding gout risk, ABCG2 rs4148155 carriers and SLC22A12 rs75786299 carriers showed significant association in both sexes, with ORs of 1.40 (95% CI: 1.26 to 1.56, p < 0.0001) and 1.49 (95% CI: 1.07 to 2.07, p = 0.02) in females and stronger effects observed in males (OR = 1.78, 95% CI: 1.67 to 1.91, p < 0.0001, and OR = 2.17, 95% CI: 1.81 to 2.60, p < 0.0001, respectively). For nephrolithiasis risk, the ABCG2 rs4548155 variant showed no association in females (OR = 1.06, 95% CI: 0.93 to 1.22, p = 0.4) but was associated with increased risk in males (OR = 1.2, 95% CI: 1.11 to 1.29, p < 0.0001), while SLC22A12 rs75786299 showed no significant associations with nephrolithiasis in either sex.

Interaction Analyses of ABCG2 rs4148155 and SLC22A12 rs75786299

Interaction analysis between ABCG2 rs4148155 and SLC22A12 rs75786299 was conducted using a multiplicative interaction term in multivariable logistic regression models (online suppl. Table 4). Both SNPs showed significant independent associations with gout risk (rs4148155: OR = 1.65, 95% CI: 1.56 to 1.75, p < 0.0001; rs75786299: OR = 1.68, 95% CI: 1.3 to 2.17, p < 0.0001).

For gout risk, no significant multiplicative interaction was observed between ABCG2 rs4148155 and SLC22A12 rs75786299 (OR = 1.29, 95% CI: 0.93 to 1.79, p = 0.12). In contrast, additive interaction analysis demonstrated a significant synergistic effect. Individuals carrying both risk alleles exhibited a positive additive interaction, with an RERI of 1.26 (95% CI: 0.42 to 2.09), indicating that the combined effect exceeded the sum of their individual effects. The AP was 0.35 (95% CI: 0.17 to 0.53), indicating that 35% of the risk in individuals with both risk alleles could be attributed to the interaction itself. Furthermore, the SI was 1.95 (95% CI: 1.27 to 2.98), confirming significant departure from additivity.

For nephrolithiasis risk, no significant multiplicative interaction was observed between both risk alleles (OR = 1.46, 95% CI: 0.95 to 2.22, p = 0.08). Consistently, additive interaction analysis showed no evidence of synergistic effects, with an RERI of 0.43 (95% CI: −0.03 to 0.89). Although the AP was 0.31 (95% CI: 0.03 to 0.58), suggesting a potential trend toward interaction, the weak statistical support and the inability to calculate the SI limit definitive conclusions. In summary, the interaction between ABCG2 rs4148155 and SLC22A12 rs75786299 was disease specific, with a significant additive synergistic effect observed for gout risk but not for nephrolithiasis risk, for which only rs4148155 showed a modest independent association (OR = 1.09, p = 0.01).

Discussion

This case-control study, conducted from TPMI database, aimed to evaluate the combined effect of ABCG2 rs4148155 and SLC22A12 rs75786299 polymorphisms on the risk of nephrolithiasis in individuals with hyperuricemia and gout. Our findings indicate that these genetic polymorphisms, when considered collectively, significantly modulate the risk of nephrolithiasis within Taiwanese population. These results underscore the potential clinical utility of genetic profiling in the management of nephrolithiasis and suggest that personalized approaches may be necessary for individuals with hyperuricemia and gout.

Numerous genetic variants of the SLC22A12 and ABCG2 gene have shown a significant association with hyperuricemia and gout [8]. Despite the low prevalence of SLC22A12 rs75786299(A) carrier in the Korea population, a recent study identified a robust association between this variant in URAT1 gene and hyperuricemia with an OR of 32.05 [21]. Through this study, we have demonstrated a significant association between rs75786299(A) allele and gout within the Taiwanese population, with an OR of 1.84.

In the Taiwanese population (TWB2 array, imputed to GRCh38), the MAFs of ABCG2 rs4148155 and SLC22A12 rs75786299 are 0.3146 and 0.0126, respectively. Previous studies in East Asian populations reported an MAF of approximately 0.30 for rs4148155 and identified rs75786299 as a rare but strongly associated variant (OR ≈32.07) among Koreans [21, 22]. Strong ethnic differences in allele frequencies of both SLC22A12 and ABCG2 have been well documented. For example, the SLC22A12 c.1400C>T (p.T467M) variant is present at ∼5.56% in the Roma population, while other founder mutations such as p.W258X (∼2.3%) occur in Japanese cohorts [31–33]. Similarly, the ABCG2 Q141K variant is considerably more prevalent in East Asian populations (32% in Japanese and 34% in Han Chinese) compared to Europeans and European Americans (11–12%) [34]. These observations underscore the importance of interpreting our Taiwanese allele frequencies in a broader population-genetic context.

Meta-analyses of genome-wide association studies in Asian populations have revealed a significant association between the ABCG2 rs4148155(G) allele and gout risk [22, 35]. Consistent with these findings, our study identified a strong association between the rs4148155(G) allele and gout, with an OR = 1.61 (95% CI: 1.52 to 1.70, p < 0.0001). Furthermore, we demonstrated that both SLC22A12 rs75786299(A) and ABCG2 rs4148155(G) alleles are independent risk factors for gout. Importantly, we observed an additive risk for gout among individuals carrying both variants, regardless of gender. Taken together, these observations underscore marked differences in allele prevalence across populations and emphasize the importance of interpreting our Taiwanese allele frequencies in a broader population-genetic framework. This context is particularly relevant when considering the impact of these variants on hyperuricemia, gout, and nephrolithiasis risk.

Previous researches have demonstrated that polymorphisms in ABCG2 are strongly associated with an increased risk of hyperuricemia and gout [8, 11–15, 17, 22, 35]. Notably, in addition to rs2231142(T) allele, the rs4148155(G) allele has been identified as a significant gout-risk locus in the Japanese population within the renal overload subtype (OR = 2.79) [22]. ABCG2 dysfunction in both mice and humans reduces intestinal urate excretion shifting uric acid elimination to the kidneys, where compensatory downregulation of URAT1 and GLUT9 activity increases urinary excretion [12, 36]. Reductions in ABCG2 function greater than 50% are linked to CKD due to hyperuricosuria, urinary crystal formation, and inflammasome activation in renal tubular cells [23, 34, 37]. Additionally, ABCG2 dysfunction is associated with early-onset and tophaceous gout, where chronic inflammation and oxidative stress promote osteoclastogenesis, increasing the risk of osteoporosis and nephrolithiasis [13, 15, 16, 38–42]. Moreover, the elevated prevalence of nephrolithiasis among individuals with gout may be contributed to hyperuricosuria which serves as a critical driver for precipitation of calcium stones from native urine [2, 23]. In the present study, we observed that the rs4148155(G) allele independently confers a significant risk for incident nephrolithiasis with an OR of 1.16 (95% CI: 1.09 to 1.24, p < 0.0001). Consistent with findings from prior study utilizing the Taiwan Biobank database, our results reaffirm that genetic variations in ABCG2 represent a significant independent risk for nephrolithiasis [17]. Moreover, we identified an additive risk for incident nephrolithiasis among individuals carrying both the SLC22A12 rs75786299(A) and ABCG2 rs4148155(G) alleles, regardless of gender. Building upon these findings, this study provides further evidence supporting the notion that distinct SNPs in the SLC22A12 and ABCG2 collectively contribute to an elevated risk of both gout and nephrolithiasis.

Most epidemiological studies determine nephrolithiasis as a clinical endpoint based on self-reported patient’s history or treatment associated with nephrolithiasis [1, 2, 43]. However, the prevalence of self-reported nephrolithiasis tends to be higher compared to cases identified through on-site urinary tract ultrasonography (11.7% vs. 6.4%) [44]. Although ultrasonography traditionally exhibits lower sensitivity and specificity compared to non-contrast computed tomography, it offers the distinct advantage of avoiding radiation exposure in the evaluation of nephrolithiasis [26]. Notably, ultrasound screenings conducted in rural areas of China revealed that 33% of patients with nephrolithiasis harbored asymptomatic stones [44]. In clinical practice, both imaging modalities have demonstrated equivalent diagnostic accuracy in identifying obstructive uropathy caused by nephrolithiasis [45]. In light of these considerations, utilizing ultrasound to confirm nephrolithiasis in this study offers enhanced reliability and validity.

This study has several limitations that should be acknowledged. First, this study utilized a cross-sectional design based on data from the TPMI database. Although the risk factors for nephrolithiasis were prospectively collected, the cross-sectional nature of the study precludes establishment of causal relationships. Nonetheless, we propose that the genotypes of ABCG2 rs4148155 and SLC22A12 rs75786299 play a significant role in the development of gout and nephrolithiasis. Second, this study population included 1,044 participants who reported engaging in regular physical exercise and were specifically analyzed for their relationship with gout. The limited sample size may have affected the statistical power of the findings, and caution is warranted when extrapolating these results to broader populations. Third, nephrolithiasis was defined in this study based on the identification of renal stones exceeding 3 mm using ultrasonography. However, detailed assessments of urine chemistry and stone composition were not available, which could introduce potential information bias. Despite this limitation, our findings provide compelling evidence of an association between genetic variations in ABCG2 rs4148155 and SLC22A12 rs75786299 and an increased risk of gout and nephrolithiasis. Furthermore, the results suggest an additive risk pattern among carriers of these genetic variants.

In conclusion, this study demonstrates the additive association of the ABCG2 rs4148155 and SLC22A12 rs75786299 variants with the risk of gout and nephrolithiasis. Incorporating genetic testing for these variants may serve as a valuable tool in precision healthcare, offering additional insights into nephrolithiasis risk among patients with hyperuricemia and gout.

Acknowledgments

We thank all the participants and investigators from Taiwan Precision Medicine Initiative.

Statement of Ethics

This study was reviewed and approved by the Institutional Review Board of Taichung Veterans General Hospital (IRB No. CE222205A-2). All participants completed a written inform consent form according to the Helsinki declaration.

Conflict of Interest Statement

All the authors declare that they have no relevant financial interests relevant to this work and manuscript.

Funding Sources

This work was supported by Academia Sinica (40-05-GMM, AS-GC-110-MD02, and 236e-1100202), National Development Fund, Executive Yuan (NSTC 111-3114-Y-001-001), National Science and Technology Council, Taiwan (NSTC-111-2634-F-A49-014, NSTC-111-2218-E-039-001, and NSTC-111-2314-B-075A-003-MY3), and Taichung Veterans General Hospital, Taiwan (TCVGH-1127301C, TCVGH-1127302D, TCVGH-YM1120110, TCVGH-1137310C, TCVGH-1137319C, TCVGH-1137302D, TCVGH-1127304B, TCVGH-1137302B, TCVGH-1123803B, and TCVGH-1133802B).

Author Contributions

C-.T.L. conceived the study, drafted, and revised the manuscript. C-.T.L., I.C.C., Y.-J.C., Y.-C.L., C.-H.L., and Y.-M.C. verified the analytical methods. J.-C.C., T.-J.W., W.-N.H., Y.-H.C., and C.-Y.W. helped supervise the project. I.-C.C. and Y.-M.C. formed the original hypothesis, designed the study, drafted, and revised the manuscript. All authors approved the final version of the manuscript.

Funding Statement

This work was supported by Academia Sinica (40-05-GMM, AS-GC-110-MD02, and 236e-1100202), National Development Fund, Executive Yuan (NSTC 111-3114-Y-001-001), National Science and Technology Council, Taiwan (NSTC-111-2634-F-A49-014, NSTC-111-2218-E-039-001, and NSTC-111-2314-B-075A-003-MY3), and Taichung Veterans General Hospital, Taiwan (TCVGH-1127301C, TCVGH-1127302D, TCVGH-YM1120110, TCVGH-1137310C, TCVGH-1137319C, TCVGH-1137302D, TCVGH-1127304B, TCVGH-1137302B, TCVGH-1123803B, and TCVGH-1133802B).

Data Availability Statement

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author upon reasonable request.