Homozygous or compound heterozygous mutations in solute carrier family 34, member 3 (SLC34A3), encoding the sodium-dependent inorganic phosphate cotransport proteins 2c, cause hereditary hypophosphatemic rickets with hypercalciuria,1,2 characterized by hypophosphatemia; elevated 1,25-dihydroxyvitamin D; hypercalciuria; and rickets/osteomalacia. Prior studies on patients heterozygous for SLC34A3 pathogenic variants have been limited to relatives of patients with hypophosphatemic rickets with hypercalciuria.3 However, the phenotypic spectrum of these mutations and modifiable risk factors remain incompletely understood, largely because of a paucity of large, unselected cohort studies using a genotype-first approach. Using exome sequencing and electronic health record (EHR) data from a health system-based cohort (n=174,417) in central and northeast Pennsylvania, we attempted to bridge this knowledge gap, focusing on Ser192Leu, the most common pathogenic SLC34A3 variant in non-Finnish European people.4,5 We hypothesized that SLC34A3 carriers would be at increased risk of nephrolithiasis and hypophosphatemia. We also examined risks of eGFR <60 ml/min per 1.73 m2 and differences by sex, given higher exposure to calcium and vitamin D supplementation for women.
Of 238 participants heterozygous for SLC34A3, we excluded those with an additional SLC34A3 variant of uncertain significance (n=3) and those missing EHR data (n=17). The mean age was 58.6 (SD 18.6) years (Table 1). In analyses adjusted for age, sex, and genetic ancestry and clustered by family network, SLC34A3 Ser192Leu heterozygotes were at increased risk of nephrolithiasis International Classification of Diseases diagnosis (13% versus 6%; P < 0.001), minimum phosphate <2.5 mg/dl (32% versus 16%; P < 0.001), median phosphate <3.0 mg/dl (28% versus 15%; P < 0.001), and eGFR <60 ml/min per 1.73 m2 (27% versus 20%; P = 0.04); had lower eGFR (−4.43 ml/min per 1.73 m2, 95% confidence interval [CI], −7.03 to −1.83; P = 0.001); and tended to have a higher prevalence of kidney/liver cyst ICD codes (5% versus 3%; P = 0.09), compared with controls.
Table 1.
Characteristics of SLC34A3 Ser192Leu heterozygotes and controls without SLC34A3 variants
| Characteristic | SLC34A3 Ser192Leu heterozygotes (n=218) | Controls (n=168,882) | aORa (95% CI) |
|---|---|---|---|
| Follow-up time, yr (SD) | 14.72 (5.49) | 13.85 (6.00) | |
| Age, yr (SD) | 58.64 (18.56) | 57.36 (18.91) | |
| Female, % | 136 (62) | 101,952 (61) | |
| European ancestry, % | 213 (99) | 155,440 (94) | |
| eGFR available, % | 190 (87) | 148,628 (88) | |
| eGFR, ml/min per 1.73 m2 | 74.23 (23.55) | 80.91 (26.70) | |
| CKD eGFR stage | |||
| ≥90 ml/min per 1.73 m2 | 52 (27) | 52,698 (35) | |
| 60–89 ml/min per 1.73 m2 | 86 (45) | 65,831 (44) | |
| 30–59 ml/min per 1.73 m2 | 46 (24) | 24,897 (17) | |
| 15–29 ml/min per 1.73 m2 | 3 (2) | 2396 (2) | |
| <15 ml/min per 1.73 m2 | 3 (2) | 2962 (2) | |
| Serum phosphate available, % | 96 (44) | 39,636 (23) | |
| Minimum serum phosphate, mg/dl | 2.76 (0.73) | 3.08 (0.69) | |
| Median serum phosphate, mg/dl | 3.17 (0.66) | 3.43 (0.61) | |
| Nephrolithiasis ICD code, % | 28 (13) | 9767 (6) | 2.38 (1.59 to 3.56); P < 0.001 |
| Minimum phosphate <2.5 mg/dl, % | 31/96 (32) | 6226/39,636 (16) | 2.67 (1.74 to 4.11); P < 0.001 |
| Median phosphate <3.0 mg/dl, % | 27/96 (28) | 5768/39,636 (15) | 2.42 (1.52 to 3.86); P < 0.001 |
| Both minimum phosphate <2.5 and median phosphate <3.0 mg/dl, % | 21/96 (22) | 3125/39,636 (8) | 3.41 (2.05 to 5.68); P < 0.001 |
| eGFR <60, ml/min per 1.73 m2 | 52/190 (27) | 30,255/148,784 (20) | 1.45 (1.01 to 2.07); P = 0.04 |
| Kidney/Liver cysts, % | 11/218 (5) | 5084/168,882 (3) | 1.68 (0.92 to 3.08); P = 0.09 |
ICD 9/10 codes used were as follows: nephrolithiasis (N20, 592.0, 592.1, 592.9, 594.0, 594.1, 594.2, 594.8, 594.9, and 274.11) and kidney/liver cysts (Q61.2, Q61.3, 753.13, 753.12, 753.10, Q61.5, Q61.8, Q61.9, Q61.00, Q61.01, Q61.02, 753.11, 753.19, Q44.6, 573.8, and 751.62). aOR, adjusted odds ratio; CI, confidence interval; SLC34A3, solute carrier family 34, member 3; ICD, International Classification of Diseases.
Analyses are adjusted for age, sex, and genetic ancestry and clustered by family network.
Among SLC34A3 heterozygotes, female patients had greater risks of nephrolithiasis (adjusted odds ratio 2.64, 95% CI, 1.03 to 6.72; P = 0.04) and eGFR <60 ml/min per 1.73 m2 (adjusted odds ratio 3.97, 95% CI, 1.08 to 1.17; P < 0.001) than male patients. Use of supplements was greater for female patients than male patients for both calcium (43.4% versus 15.9%) and vitamin D (64.7% versus 34.2%), although neither calcium nor vitamin D supplementation were significantly associated with nephrolithiasis or eGFR <60 ml/min per 1.73 m2 (P > 0.05 for all comparisons; data not shown).
In addition to confirming that SLC34A3 Ser192Leu heterozygotes are at two to three-fold higher risk of nephrolithiasis and hypophosphatemia,3 we found that they also were at risk of decreased eGFR <60 ml/min per 1.73 m2, an association not previously described to our knowledge. Furthermore, we show that risks are particularly elevated for women with SLC34A3, which we speculate could be because of higher exposure to calcium and vitamin D for bone health. Nephrolithiasis findings are consistent with a study of 120 (61 monoallelic, 43 biallelic, 16 wild-type) individuals from 27 families with SLC34A3 pathogenic variants3; individuals with biallelic SLC34A3 mutations had ∼ten-fold higher risk of nephrolithiasis and nephrocalcinosis, whereas heterozygotes had approximately three-fold higher risk of nephrolithiasis.3 A prior case series of 12 patients with SLC34A3 pathogenic variants (7 monoallelic, 5 biallelic) described a higher preponderance for kidney cysts than the general population.6 SLC34A3 Ser192Leu heterozygotes had numerically more kidney/liver cyst ICD codes, although this was not statistically significant. Although we were limited by the lack of standardized laboratory testing, such as 24-hour kidney stone risk profiles, strengths of our study included a largely unselected cohort with long follow-up, allowing us to better capture nephrolithiasis burden.
Our study has important implications as it raises the question as to whether knowledge of this genetic risk could affect care for this population at high risk of nephrolithiasis. Few received 24-hour urine stone risk profile testing, and many patients were receiving calcium and vitamin D supplements, which could worsen nephrolithiasis risk. Oral phosphorus supplementation normalizes metabolic abnormalities of HRHH1,7 and could be used to treat the borderline-low serum phosphate levels in heterozygotes, offsetting the stimulus for increased calcitriol synthesis, and resultant hypercalciuria.3 Further genotype-first studies are needed to determine whether a precision medicine approach can prevent and reduce kidney stone burden and eGFR decline in SLC34A3 carriers.
Brief Methods: The study population was the one used for the MyCode/DiscovEHR study,5 an exome sequencing study linked to EHR data (demographics, laboratory test results, ICD diagnosis codes). The primary outcome was nephrolithiasis; secondary outcomes included eGFR <60 ml/min per 1.73 m2, outpatient hypophosphatemia (lowest phosphate <2.5 mg/dl; median phosphate <3.0 mg/dl; meeting both criteria), mean phosphate, mean eGFR, and any kidney/liver cyst (ICD codes for autosomal dominant polycystic kidney disease, cystic kidney diseases, congenital kidney cyst, or liver cystic disease).6,7 Chart reviews were performed by a nephrology fellow. EHR-based renal phenotypes were compared between SLC34A3 Ser192Leu heterozygotes and controls (without any SLC34A3 variants) using logistic and linear regression for categorical and continuous outcomes, respectively, adjusted for age, sex, and genetic ancestry and clustered by family network. An exploratory analysis among SLC34A3 Ser192Leu heterozygotes examined associations between sex, calcium and vitamin D supplementation with risks of nephrolithiasis, and eGFR <60 ml/min per 1.73 m2. We used STATA/MP 15.1 for analyses.
Acknowledgments
We thank the participants of the MyCode Community Health Initiative and the Geisinger-Regeneron DiscovEHR Collaboration for use of the genomic and electronic health information. We are also grateful to Thomas Jones for assistance with extracting EHR data and Tooraj Mirshahi and Jonathan Luo for the use of the genetic database.
Disclosures
I.D. Bucaloiu reports Employer: Geisinger Medical Center. A.R. Chang reports Employer: Geisinger Health System; Consultancy: Amgen, Medscape, Novartis, and Reata; Research Funding: Bayer, Novartis, and Novo Nordisk; Advisory or Leadership Role: Amgen and Reata; and Other Interests or Relationships: National Kidney Foundation—grant support. G. Singh reports Employer: Geisinger Health. N. Strande reports Employer: Mayo Clinic; Geisinger; and Research Funding: NIH. All remaining authors have nothing to disclose.
Funding
The patient enrollment and exome sequencing were funded by the Regeneron Genetics Center.
Author Contributions
Conceptualization: Ion D. Bucaloiu, Alexander R. Chang, Gurmukteshwar Singh.
Data curation: Alexander R. Chang, Bryn Moore, Chinedu Nwachukwu.
Formal analysis: Alexander R. Chang.
Investigation: Alexander R. Chang, Chinedu Nwachukwu.
Methodology: Ion D. Bucaloiu, Alexander R. Chang, Natasha T. Strande.
Resources: Bryn Moore.
Supervision: Ion D. Bucaloiu, Alexander R. Chang.
Visualization: Alexander R. Chang.
Writing – original draft: Ion D. Bucaloiu.
Writing – review & editing: Ion D. Bucaloiu, Alexander R. Chang, Bryn Moore, Chinedu Nwachukwu, Gurmukteshwar Singh, Natasha T. Strande.
Data Sharing Statement
All data produced in the present work are available upon reasonable request to the authors and subject to a data use agreement.
References
- 1.Schönauer R Petzold F Lucinescu W, et al. Evaluating pathogenicity of SLC34A3-Ser192Leu, a frequent European missense variant in disorders of renal phosphate wasting. Urolithiasis. 2019;47(6):511–519. doi: 10.1007/s00240-019-01116-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bergwitz C, Miyamoto K. Hereditary hypophosphatemic rickets with hypercalciuria: pathophysiology, clinical presentation, diagnosis and therapy. Pflugers Arch. 2019;471(1):149–163. doi: 10.1007/s00424-018-2184-2 [DOI] [PubMed] [Google Scholar]
- 3.Dasgupta D Wee MJ Reyes M, et al. Mutations in SLC34A3/NPT2c are associated with kidney stones and nephrocalcinosis. J Am Soc Nephrol. 2014;25(10):2366–2375. doi: 10.1681/ASN.2013101085 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gudmundsson S Karczewski KJ Francioli LC, et al. Addendum: the mutational constraint spectrum quantified from variation in 141,456 humans. Nature. 2021;597(7874):E3–E4. doi: 10.1038/s41586-021-03758-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Carey DJ Fetterolf SN Davis FD, et al. The Geisinger MyCode community health initiative: an electronic health record–linked biobank for precision medicine research. Genet Med. 2016;18(9):906–913. doi: 10.1038/gim.2015.187 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hanna C Potretzke TA Chedid M, et al. Kidney cysts in hypophosphatemic rickets with hypercalciuria: a case series. Kidney Med. 2022;4(3):100419. doi: 10.1016/j.xkme.2022.100419 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tieder M Modai D Shaked U, et al. Idiopathic hypercalciuria and hereditary hypophosphatemic rickets. N Engl J Med. 1987;316(3):125–129. doi: 10.1056/nejm198701153160302 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data produced in the present work are available upon reasonable request to the authors and subject to a data use agreement.