APRT deficiency: the need for early diagnosis

Abstract

Adenine phosphoribosyltransferase (APRT) deficiency is a rare autosomal recessive disorder which leads to accumulation of poorly soluble 2,8-dihydroxyadenine in kidneys resulting in nephrolithiasis as well as chronic kidney disease from crystal nephropathy. This report describes a 55-year-old previously fit man who presented with shortness of breath and the investigative pathway that eventually led to a diagnosis of APRT deficiency. Early diagnosis has aided in timely institution of allopurinol, thereby improving his renal function and possibility of weaning off renal replacement therapy. Genetic testing has enabled early identification of other family members at risk and prevention of renal failure by commencing xanthine oxidoreductase (XOR) inhibitors. The issues surrounding kidney donation by a member of this family are also discussed. This case represents the importance of awareness and recognition of the signs and symptoms of this rare condition, complications of which can be easily prevented by early institution of XOR inhibitor therapy.

Background

Adenine phosphoribosyltransferase (APRT) deficiency (Online Mendelian Inheritance in Man number 614723) is a rare autosomal recessive disorder caused by absence of the enzyme APRT.^{1 2} APRT is an enzyme in the purine metabolism pathway which normally recycles adenine into soluble adenosine monophosphate (AMP) (see figure 1).³ Deficiency of APRT leads to excessive accumulation of adenine, which is converted by xanthine oxidoreductase (XOR; xanthine dehydrogenase/xanthine oxidase) to 2,8-dihydroxyadenine (2,8-DHA) a poorly soluble metabolite, which is excreted by the kidney in large amounts. This results in affected individuals developing nephrolithiasis and chronic kidney disease (CKD) from excessive crystal deposition in the renal parenchyma and resultant nephropathy (crystal nephropathy).³

Figure 1.

Purine metabolism pathway showing the function of the enzyme APRT (modified and adapted from Ceballos-Picot et al⁴). AMP, adenosine monophosphate; APRT, adenine phosphoribosyltransferase.

The following report describes a family where the consultand with APRT deficiency has developed end-stage renal failure due to crystal nephropathy. Early diagnosis and institution of allopurinol has resulted in improvement of his renal function and the potential to wean off peritoneal dialysis. Genetic testing has also enabled early identification of other family members at risk. The ethical issues surrounding kidney donation by a member of this family are also discussed. This case highlights the importance of early detection of APRT deficiency in optimising management and prevention of end-stage renal failure.

Case presentation

A 55-year-old man presented with a history of 2–3 months of shortness of breath on exertion. He was a previously fit, healthy man who was an avid water skier. He was born to non-consanguineous parents of English/Irish origin. He had a remote history of renal colic, haematuria and passing urinary stones in young adulthood. He remained asymptomatic subsequently and did not seek medical attention for any kidney-related problems. He was not obese and did not have metabolic syndrome. He did not report excessive calcium intake and did not consume any vitamins or herbal supplements.

Screening investigations revealed a serum creatinine (SCr) of 318 µmol/L and an eGFR of 18 mL/min/1.73 m². Urinalysis was normal. Renal ultrasound showed bilateral echogenic kidneys with increased resistive index consistent with medical renal disease; however, a CT scan of his kidneys showed no evidence of renal calculi.

During the course of his assessment, it became apparent that one of his brothers had kidney stones and was commenced ‘on a medication which almost cured him of kidney stones’ in Japan. While he was unsure, the possibility of APRT deficiency was mentioned. A second brother died of a ruptured cerebral aneurysm at 50 years of age. It was not known if he or any other family member had any renal disease, see figure 2.

Figure 2.

Pedigree. APRT, adenine phosphoribosyltransferase.

Investigations

As no further information regarding his brother’s diagnosis overseas was available at that stage, and decisions regarding renal replacement therapy had to be made expediently due to his deteriorating renal function, genetic testing for APRT was organised which revealed two pathogenic variants in the APRT gene. The first was a c.194A>T variant in exon 3 of one allele. It is a missense variant that changes aspartic acid, an acidic amino acid to the hydrophobic valine at position 65 (p.Asp65Val) and disrupts the phosphoribosyltransferase domain. This variant has been recorded as pathogenic in the ClinVar database. It occurs at a very low population frequency of 0.000009706 as described in the ExAC browser (http://exac.broadinstitute.org/gene/ENSG00000198931). All Icelandic cases of APRT deficiency harbour this mutation.² The second variant is a canonical splice site variant c.400+2 dup which is also described at a low frequency of 0.00008288 in ExAC. Both these variants occur frequently in Europeans with APRT deficiency and have been described as pathogenic in a large French cohort studied by Ceballos-Picot et al⁴ Subsequent RBC enzyme analysis in the consultand revealed APRTase activity below normal (0.11 nmol/min/mg; reference range 0.15–2 nmol/min/mg). As the consultand (II.3) had advanced CKD and as a diagnosis was reached by way of genetic and enzyme testing, no renal biopsy was done and urinalysis for crystals was not performed at this stage.

Treatment

The consultand was commenced on 50 mg of allopurinol immediately after the diagnosis. Subsequently, the dose was increased by 50 mg every week until he was established on 600 mg of allopurinol daily. We acknowledge that this is a very slow dose increase regimen and in the absence of frank intolerance, a starting dose of 300 mg/day (10 mg/kg/day for children) increasing rapidly to 400–600 mg/day is generally recommended. The consultand commenced peritoneal dialysis approximately 3 months after the initial diagnosis in addition to being active on a renal transplant waitlist. His SCr levels have improved from >1000 µmol/L at its peak to 354 µmol/L since commencement of allopurinol. His most recent urine examination did not demonstrate any 2,8-DHA crystals. If his renal function continues to improve on high dose allopurinol, he may be trialled off dialysis.

The immunosuppressive regimen that will be followed if he undergoes renal transplantation will be basiliximab for induction and tacrolimus, mycophenolate mofetil (MMF) and prednisolone for maintenance. Although in our department, we have used azathioprine in patients who developed side effects with MMF, especially if they became positive for John Cunningham virus in plasma, the consultand cannot receive azathioprine as it interacts with allopurinol and our patient will most likely require lifelong treatment with allopurinol.

Outcome and follow-up

The consultand’s asymptomatic sister underwent predictive genetic testing and she was found to have the same biallelic APRT variants initially found in her brother. 2,8-DHA crystals were found at three separate occasions in her urine sample. Since commencement of allopurinol, no crystals have been detected in her urine. She has not had any deterioration of her renal function with her SCr stable at 70 µmol/L and eGFR of 90 mL/min/1.73 m². She will continue on 600 mg/day allopurinol for maintenance.

It was subsequently followed up that the proband’s brother (II.4) in Japan was investigated for APRT deficiency due to incidental finding of elevated SCr with values around 177 µmol/L. As part of routine urinalysis 2,8-DHA like crystals were found in his urine. Soon after this finding, he was commenced on allopurinol 100 mg/day, after which no further urinary crystals were detected. SCr gradually improved since commencement of allopurinol with the latest readings being 114 µmol/L. He never required any form of renal replacement therapy. Due to facial flushing, thought to be a side effect of allopurinol, his medication was changed to febuxostat 10 mg/day initially and subsequently increased to 20 mg/day.

Discussion

Renal stone disease is not uncommon with dietary and lifestyle factors being the major contributors.⁵ Around 15% of cases are reported to be due to mutations in one of 30 known genes associated with renal stone disease.⁶ Recurrent or bilateral stones, onset in children, a family history of renal stones or history of parental consanguinity should alert the treating physician to a possible genetic aetiology.⁷

APRT deficiency is a rare autosomal recessive inborn error of metabolism caused by biallelic mutations in the gene APRT resulting in deficiency of APRT enzyme activity.⁸ APRT is an enzyme of the purine metabolic pathway that recycles adenine back to the soluble by-product AMP.⁹ In the absence of APRT, there is increased accumulation of adenine, which is converted to 2,8-DHA by XOR. 2,8-DHA is excreted in large amounts by the kidney and is poorly soluble in urine, resulting in heavy crystalluria. The disease is not in the kidney itself, but it results from ongoing 2,8-DHA excretion, leading to crystal deposition in renal tubules, interstitium and inside renal tubular epithelial cells in addition to interstitial inflammation and fibrosis induced by the crystals (crystal nephropathy) and eventual CKD.^{1 2}

The presentation can occur at any age from childhood^10–12 to adulthood.¹³ Initial presenting features are generally non-specific and can range from asymptomatic haematuria, dark diaper staining in infants, renal colic, acute kidney injury, urinary tract obstruction, passing of stones in the urine, to CKD from crystal nephropathy.^{8 14 15} Diagnosis of APRT deficiency can be delayed, particularly in asymptomatic patients leading to end stage renal failure.¹³ In some individuals, diagnosis is not made until after renal transplantation leading to poor transplant outcomes for the recipient and wasteful expenditure of the donor organ.^{16 17} No extrarenal symptoms have been described so far although eye discomfort possibly due to ocular deposition of the DHA crystals has been suspected.¹⁸

It should be noted that the DHA crystals are radiolucent and plain abdominal X-rays do not detect the stones, however, high-resolution CT scans may pick up evidence of renal stones. In patients who present due to recurrent passing of stones, analysis of the stone using infra-red spectroscopy is important to reach a conclusive diagnosis. Early in the disease, before the renal clearance of the crystals decreases, 2,8-DHA crystals may be detected in the urine under microscopy.^{8 13} Deficient erythrocyte APRT enzyme activity confirms the diagnosis. Genetic testing and finding mutations known to completely abolish APRT enzyme activity is also diagnostic of the disorder.¹

More than 40 pathogenic genetic variants have been described in the coding regions of the APRT gene.^{13 19} Biallelic loss of function variants in APRT abolish APRT enzyme activity.²⁰ Heterozygotes generally remain asymptomatic. In individuals with a clear diagnosis through enzymatic assay, if biallelic mutations are not identified by sequencing, deletion and/or duplication of the gene or part of the gene should be investigated through methods such as multiplex ligation probe amplification as current sequencing techniques cannot identify such deletions/duplications. It should be noted that in such cases, inability to identify a second mutation does not necessarily exclude presence of biallelic mutations.

Treatment with the XOR inhibitor allopurinol has been shown to prevent renal failure from crystal nephropathy and even dissolve the crystals and improve renal function.¹³ More recently another XOR inhibitor, febuxostat has also been shown to be effective.²¹ End-stage renal failure from APRT deficiency is treated in the same way as any other cause, with renal replacement therapy or renal transplant. Recurrence of the crystalline nephropathy in the transplanted kidney has been demonstrated in multiple cases and this can lead to rapid allograft failure.^22–24

If the disease is recognised prior to transplant, pretreatment with allopurinol may reduce the chances of allograft failure post-transplant. After transplant, urine may be monitored for appearance of 2,8-DHA crystals as this is the only available modality for monitoring of therapy. However, it is not a reliable method for surveillance of therapeutic effect of treatment with XOR inhibitors. A high-throughput ultra-performance liquid chromatography–electrospray tandem mass spectrometry assay for measurement of 2,8-DHA levels in urine samples was published by Thorsteinsdottir et al and shows promise for monitoring of pharmacotherapy in the future.²⁵ In fact, a recent pilot study comparing the effect of allopurinol 400 mg/day and febuxostat 80 mg/day in the treatment of APRT deficiency used this method to monitor the levels of urinary excretion of 2,8-DHA and found febuxostat to be more efficacious, pending confirmation in larger studies.²⁶ Crystal nephropathy can be identified on biopsy of the transplanted kidney.²²

Treatment with a XOR inhibitor should be continued lifelong to prevent recurrence.²² Prevention of dehydration is important. Although low purine diet is recommended, compliance is very poor and benefits not clear. Therefore, this diet is not generally prescribed, especially in patients taking XOR inhibitors. High-dose allopurinol, up to 600 mg/day is recommended to preserve renal function in transplant recipients.¹⁷ Selection of appropriate doses of XOR inhibitors and the best dose for maintenance are not clearly defined. Dose increase and maintenance may also depend on tolerance to the various side effects.

XOR normally inactivates 6-mercaptopurine (6-MP), which is an active form of the drug azathioprine. When XOR is inhibited (by using allopurinol or febuxostat), the elevated 6-MP is shunted into producing other active metabolites which can cause severe bone marrow suppression.²⁷ Therefore, extreme caution should be used if azathioprine is used in patients on high-dose allopurinol.

It is important to note that no clear guidelines currently exist with regard to the appropriate dosing of XOR inhibitors. The weekly dosage increment of allopurinol we followed was based on tolerability and lack of side effects in our patient, in whom we were able to achieve a maximum maintenance dose of 600 mg/day. This regimen may not suit all patients with APRT deficiency and therefore must be individualised. However, routine reduction in dose of allopurinol in CKD is not recommended as this may compromise the renal outcomes. We acknowledge that the febuxostat dose that the proband’s brother is maintained on is subtherapeutic, with a dose of 80 mg/day generally recommended if tolerated.

In addition, kidneys were donated for transplant by his elder brother, who died from a ruptured cerebral aneurysm, prior to knowledge of the mutation in the family. This brother may have carried one or both of the family mutations but may also have been mutation free. Since the age of onset is variable, we cannot infer any data from the fact that he had no known kidney disease at the age of 50 years. Furthermore, there is no known association reported in literature between APRT deficiency and cerebral aneurysms. While the status of these transplanted kidneys is thus far unknown, it is usual practice for kidneys in most units in Australia to be biopsied at 3 and 12 months. Assuming absence of APRT deficiency, post-transplant 2,8-DHA nephropathy is unlikely in the recipient as no new 2,8-DHA is being produced by the recipient and excreted by the allograft. Therefore, it is unlikely that the recipient will develop disease if the kidney was healthy at the time of donation.

Genetic testing of family members becomes important for their own screening, and to check suitability for kidney donation for a living-related transplant in other affected family members. In some instances, only one mutation is detectable in patients with APRT deficiency. This makes familial screening more difficult, especially if kidney donation by siblings is considered. In such situation, measurement of APRT activity, in addition to mutation analysis, may be helpful to exclude APRT deficiency.

In conclusion, early recognition of APRT deficiency provides an opportunity for targeted therapy, and possible avoidance of the need for renal replacement therapy. Absent APRT enzyme activity is diagnostic of the disorder. Detection of biallelic APRT mutations, known to abolish APRT enzyme activity is also diagnostic. Outcomes of living donor as well as cadaveric transplant can be favourable, provided adequate pretreatment and long-term treatment with a XOR inhibitor is used.

Patient’s perspective.

I believe it is important that the medical community are aware of the symptoms of this rare condition and are able to diagnose APRT deficiency early, so the complications can be prevented. Our family hopes that describing our own story helps detect others early.

Learning points.

• Deficiency of adenine phosphoribosyltransferase causes accumulation of 2,8-dihydroxyadenine, which is poorly soluble in urine and can lead to nephrolithiasis as well as chronic kidney disease from crystal nephropathy.

• These crystals are radiolucent and plain abdominal X-rays do not detect the stones. Hence, higher clinical suspicion is necessary for accurate and timely diagnosis.

• Early diagnosis and commencement of xanthine oxidoreductase inhibitors may possibly prevent the need for renal replacement therapy.

• Genetic testing may be able to identify asymptomatic family members and prevent future adverse outcomes from nephropathy.

• Family members should be carefully screened prior to being accepted as kidney donors.