Adenine Phosphoribosyltransferase Deficiency

Pediatr Nephrol (2012) 27:571–579 DOI 10.1007/s00467-011-2037-0

ORIGINAL ARTICLE

Adenine phosphoribosyltransferase deficiency in children Jérôme Harambat & Guillaume Bollée & Michel Daudon & Irène Ceballos-Picot & Albert Bensman & APRT Study Group

Received: 26 July 2011 / Revised: 21 September 2011 / Accepted: 22 September 2011 / Published online: 3 January 2012 # IPNA 2011

Abstract Adenine phosphoribosyltransferase (APRT) deficiency is a rare autosomal recessive disorder characterized by 2,8-dihydroxyadenine (2,8-DHA) crystalluria that can cause nephrolithiasis and chronic kidney disease. The aim of our study was to assess the clinical presentation, diagnosis, and outcome of APRT deficiency in a large pediatric cohort. All pediatric cases of

APRT deficiency confirmed at the same French reference laboratories between 1978 and 2010 were retrospectively reviewed. Twenty-one patients from 18 families were identified. The median age at diagnosis was 3 years. Diagnosis was made after one or more episodes of nephrolithiasis (17 patients), after urinary tract infection (1 patient), and by family screening (3

Jérôme Harambat and Guillaume Bollée contributed equally to this study.

J. Harambat Service de Pédiatrie, Centre Hospitalier Universitaire de Bordeaux, Centre de référence Maladies Rénales Rares du Sud Ouest, Bordeaux, France

Irène Ceballos-Picot and Albert Bensman contributed equally to this study. The members and affiliations of the APRT Study Group are: Thierry Boussemart (Centre Hospitalier du Mans, Service de Pédiatrie, Le Mans, France); Gérard Champion (Centre Hospitalier Universitaire d’Angers, Service de Pédiatrie, Angers, France); Sophie Taque (Centre Hospitalier Universitaire de Rennes, Service de Pédiatrie, Rennes, France); Margarida Almeida (Hospital Universitário de Santa María, Serviço de Pediatria, Lisbon, Portugal); Didier Bernon (Centre Hospitalier de La Rochelle, Service de Néphrologie, La Rochelle, France); Céline Dheu (Centre Hospitalier Universitaire de Strasbourg, Service de Pédiatrie, Strasbourg, France); Susana Ferrando Monleón (Hospital de la Ribera, Servicio de Pediatría, Alzira, Valencia, Spain); Benjamin Horen (Centre Hospitalier de Bigorre, Service de Pédiatrie, Tarbes, France); Arnaud Garnier (Centre Hospitalier Universitaire de Toulouse, Service de Néphrologie Pédiatrique, Toulouse, France); Geneviève Guest (APHP, Service de Néphrologie Pédiatrique, Hôpital Necker-Enfants Malades, Paris, France); Astrid Godron (Centre Hospitalier Universitaire de Bordeaux, Service de Pédiatrie, Bordeaux, France); Bertrand Knebelmann (APHP, Service de Néphrologie, Hôpital Necker-Enfants Malades, Paris, France); Brigitte Llanas (Centre Hospitalier Universitaire de Bordeaux, Service de Pédiatrie, Bordeaux, France); Giuseppina Marra (Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Servizio di Nefrologia Pediatrica, Milan, Italy); Michel Rincé (Centre Hospitalier Universitaire de Limoges, Service de Néphrologie, Limoges, France); Jean-François Subra (Centre Hospitalier Universitaire d’Angers, Service de Néphrologie, Angers, France); François Vrtovsnik (APHP, Service de Néphrologie, Hôpital Bichat, Paris, France)

G. Bollée Service de Néphrologie, Hôpital Necker-Enfants Malades, Assistance Publique–Hôpitaux de Paris (APHP), Université Descartes, Sorbonne Paris Cité, Paris, France M. Daudon Laboratoire de Biochimie A, Hôpital Necker-Enfants Malades, APHP, Université Descartes, Sorbonne Paris Cité, Paris, France I. Ceballos-Picot Laboratoire de Biochimie Métabolique, Hôpital Necker-Enfants Malades, APHP, Université Descartes, Sorbonne Paris Cité, Paris, France A. Bensman Service de Néphrologie Pédiatrique, Hôpital Armand Trousseau, APHP, Université Pierre et Marie Curie, Paris, France J. Harambat (*) Hôpital Pellegrin-Enfants, place Amélie Raba Léon, 33076 Bordeaux Cedex, France e-mail: [email protected]

572

Pediatr Nephrol (2012) 27:571–579

patients). The diagnosis was based on stone analysis and microscopic examination of urine and/or enzymatic determination of APRT on red blood cells. All children had null APRT enzyme activity in erythrocytes. APRT gene sequencing was performed on 18 patients, revealing six homozygous and 12 compound heterozygous mutations. At diagnosis, half of the patients had decreased kidney function, and two children presented with acute renal failure. Allopurinol treatment was given to all patients at a median dose of 9 mg/kg/day. After a median follow-up of 5 years, all patients showed stabilization or improvement of kidney function, normal growth and development, and six patients had recurrence of nephrolithiasis. Based on these results, we conclude that an excellent outcome can be achieved in children with APRT deficiency who receive the proper treatment. Key words APRT deficiency . Inborn error of purine metabolism . Nephrolithiasis . Chronic kidney disease . Children

Abbreviations 2,8-DHA APRT ESRD eGFR IQR SDS UTI

2,8-dihydroxyadenine Adenine phosphoribosyltransferase End-stage renal disease Estimated glomerular filtration rate Interquartile range Standard Deviation Score Urinary tract infection

Introduction Adenine phosphoribosyltransferase (APRT) deficiency (OMIM no. 102600) is a rare autosomal recessive disorder of purine metabolism. APRT catalyzes the formation of adenosine monophosphate from adenine. In APRT deficiency, adenine is catabolized by xanthine dehydrogenase to 2,8-dihydroxyadenine (2,8-DHA). As 2,8-DHA is insoluble in the urine, APRT deficiency usually presents with symptoms referable to the kidney and urinary tract, such as hematuria, nephrolithiasis, urinary tract infections (UTI), acute kidney injury, and chronic kidney disease [1]. More rarely, 2,8-DHA deposition in the tubules and the interstitium may result in irreversible renal damage leading to end-stage renal disease (ESRD) [2]. In terms of radiological findings, 2,8DHA stones are characteristically radiolucent and are detectable by ultrasonography. Diagnostic procedures have improved over time and are currently based on different diagnostic tools, includ-

ing crystalluria, stone analysis, APRT enzyme activity in erythrocytes, and molecular analysis. It is likely that APRT deficiency is under-recognized by physicians, and diagnosis is often delayed for years [2]. A substantial proportion of patients (up to 15%) is only diagnosed at a late stage, once ESRD has occurred in adulthood [2], after disease recurrence post-transplantation [3, 4], or, even worse, after renal transplant failure [5, 6]. A timely diagnosis of APRT deficiency is crucial since early treatment with allopurinol can prevent further nephrolithiasis and may stabilize or even improve renal function [7, 8]. Given the rarity of the disease, only a few reports limited to single cases presenting the clinical and biological data of APRT deficiency in children are available, especially in the Caucasian population [8–11]. Long-term data on treatment outcome are scarce. To better delineate the scope of the disease in children, physicians require a broader collection of clinical, genetic, and therapeutic data from patients with APRT deficiency. In this regard, we report here on a large population of children with APRT deficiency, providing significant information on clinical features, genotype, management, and long-term outcome.

Methods Study population All pediatric cases (5

1

1 1 1 1 1 4 1 0 0

Episodes of nephrolithiasis

-

-

+

-

-

+

-

-

+ +

UTI

Signs and symptoms at diagnosis

+ +

+

Family screening

6.2 7.6

2.5

7.9 1.5

1.6 0.8 1.6

5.4

1.8

1.8

0.4

3.3 7.5 1.8 3.0 1.5 3.7 6.3 15.2 6.6

Age at diagnosis (years)

99 135

70

67 146

AKI 75 AKI

79

70

64

33

87 97 75 85 71 81 83 87 91

eGFR at diagnosis (ml/min/1.73 m2)

ND ND

+

ND ND

ND + ND

ND

+

+

+

ND ND + ND + ND + + +

Crystalluria

ND ND

+

+ +

+ + +

+

+

+

ND

+ + + + + + ND ND ND

Stone analysis

Tests leading to diagnosis

+ +

+

+ +

+ + +

+

+

+

+

+ + + + + + + + +

Null APRT activity

ND ND

+

ND +

+ + +

+

+

+

+

+ + + + + + + + +

Mutation analysis

Patients 7 and 8 are siblings. Patients 19, 20 and 21 are siblings

APRT, Adenine phosphoribosyltransferase; M, male; F, female; A, asymptomatic; eGFR, estimated glomerular filtration rate; UTI, urinary tract infection; ND, not determined; +, positive; -, negative; AKI, acute kidney injury

Age at onset of symptoms (years)

Patients (no./sex)

Table 1 Clinical presentation and diagnostic test results in 21 children with APRT deficiency

574 Pediatr Nephrol (2012) 27:571–579

Pediatr Nephrol (2012) 27:571–579 Fig. 1 2,8-Dihydroxyadenine (2,8-DHA) crystals and stones. a, b Crystalluria study by polarized microscopy (a) and light microscopy (b) revealing a round, brown crystal of 2,8-DHA with a characteristic central Maltese cross pattern. c Morphologic examination showing reddish-brown, friable stone, with a rough and bumpy surface

575

B

A

15 µm

15 µm

C

Treatment Urological treatment of stones was necessary in eight of the 17 (47%) patients with stone disease and included extracorporeal shock wave lithotripsy (n=4), stone removal by ureteroscopy (n=2), and open (n=1) and percutaneous surgery (n=1). A high fluid intake (median 2 l/1.73 m2 per day) and a low purine diet were recommended for nine and six patients, respectively. Treatment with allopurinol was given to all patients (median age 3.1; range 0.5–24 years) and was initiated soon after diagnosis in 19 patients. The treatment was introduced much later in two patients: 21 years after diagnosis in one patient when he presented with repeated renal colic requiring ureteroscopy and a ureteral JJ stent, and 16 years after diagnosis in a previously asymptomatic patient who was lost to follow-up. Allopurinol was given at a median dose of 9 mg/kg per day (IQR 5–10) and seemed to be well tolerated. Outcome The median duration of follow-up since diagnosis was 5.1 (IQR 1.6–10.5; mean 8.0) years. The median age at last

follow-up was 10.2 years (range 2.5–34.4; IQR 5.9–16.8). Renal function improved significantly in 14 of 19 patients with at least 6 months of follow-up and remained stable in five of these patients. The median increase in eGFR between diagnostic and last follow-up visit was 25 ml/min/1.73 m2 (IQR 2–36) (Fig. 2). At last follow-up, median eGFR was 102 ml/min/1.73 m2 (IQR 88–118). All patients showed normal growth and development. Stone recurrence occurred in six patients (29%), five of whom were receiving allopurinol therapy at the time of recurrence (allopurinol dose not different from that of patients who remained stone free). Four of these patients have subsequently been asymptomatic for years. Three additional patients presented residual stones formed before treatment. Four patients required repeated urological interventions. Serial determinations of crystalluria during follow-up were available for 12 patients. All but one showed the disappearance of crystalluria at the last examination, but seven patients experienced at least one positive test during treatment. The proportion of individuals with at least one positive crystalluria determination was higher in the group of patients with recurrent stones (4/4) than in the group of patients who remained stone-free (3/8). Patients who

Intron 4

Intron 4 Exon 3

Exon 3

Intron 4 Exon 1

Intron 4

Exon 5 Exon 3

ND Exon 2

Exon 1 ND ND

8/M

9/F 10/M

11/M

12/F 13/M

14/M

15/M 16/M

17/M 18/M

19/F 20/F 21/F

1A > G

84 C > A

532 C > T 194A>T

IVS4+2insT

IVS4+2insT 3G>A

286_287delAC

IVS4+2insT 311A > G

IVS4+2insT

287_288delCT or 289_290delCT IVS4+2insT IVS4+2insT 287_288delCT or 289_290delCT 194A>T 428 T > C IVS4+2insT

No protein

Asp28Glu

Gln178X Asp65Val

Ala108GluX3

Ala108GluX3 No protein

Thr96SerfsX13

Ala108GluX3 Glu104Gly

Ala108GluX3

Thr96ThrfsX13 or Leu97ValfsX12 Ala108GluX3 Ala108GluX3 Thr96ThrfsX13 or Leu97ValfsX12 Asp65Val Leu143Pro Ala108GluX3

Exon 5 ND ND

ND Exon 2-Intron 2

Not found Exon 4

Not found

Intron 4 Exon 1

Exon 3

Not found Exon 3

Exon 5

Intron 4 Intron 4 Exon 5 Intron 4 Exon 3 Intron 4 Exon 5

491 G > A

g.347_372del26

329 T>C

IVS4+2insT 3G>A

286_287delAC

311A > G

526 C > T

IVS4+2insT IVS4+2insT 524 C > T IVS4+2insT 194A>T IVS4+2insT 526 C > T

cDNA nucleotide sequence

Patients 7 and 8 are siblings. Patients 19, 20 and 21 are siblings. Mutations in patients 18 and 19 (in exon 5) have not been previously reported

Exon 3 Intron 4 Intron 4 Exon 3 Exon 3 Exon 5 Intron 4

Gene region

Protein sequence

Gene region

cDNA nucleotide sequence

Allele 2

Allele 1

1/M 2/F 3/F 4/M 5/M 6/M 7/M

Patients (no./sex)

Table 2 Mutation analysis of the APRT gene

Gly174Asp

Leu110Pro

Ala108GluX3 No protein

Thr96SerfsX13

Glu104Gly

Leu126Phe

Ala108GluX3 Ala108GluX3 Ser175Phe Ala108GluX3 Asp65Val Ala108GluX3 Leu126Phe

Protein sequence

576 Pediatr Nephrol (2012) 27:571–579

Pediatr Nephrol (2012) 27:571–579

577

200

eGFR (ml/min/1,73 m²)

175

150

125

100

75

50

25

eGFR at diagnosis

eGFR at last follow-up

Fig. 2 Change in estimated glomerular filtration rate (eGFR) between diagnostic and last follow-up in 19 patients with childhood onset adenine phosphoribosyltransferase (APRT) deficiency

experienced stone recurrence had more frequently positive crystalluria than those who remained stone-free (50 vs. 29% of urine samples).

Discussion Adenine phosphoribosyltransferase deficiency is a rare metabolic disease. The estimated prevalence of APRT homozygosity ranges from 1 to 2 in 100,000 so that the disease surely remains underreported, with a lower number of detected cases in Caucasian populations than expected [19]. The prevalence of the disease might be higher in some populations due to the high prevalence of a specific mutation, such as Asp65Val in the Icelandic population [7] and probably Arg108GluX3 in the French population [2]. The study reported here involves by far the largest number of pediatric patients reported to date. The main findings of the study are: (1) the disease often presents early in life (median age at onset 3 years); (2) diagnosis was delayed in 20% of cases; (3) at the time of diagnosis, half of patients already showed a mild decrease in eGFR; (4) allopurinol treatment may improve renal function; (5) excellent long-term outcome can be achieved in pediatric patients with adequate management. Recent publications pointed out that the disease is generally misdiagnosed too often, even though early diagnosis and initiation of treatment are crucial [2, 3, 7]. Among the patients enrolled in our study, however, diagnosis was made shortly after the first symptoms in 80% of children, and after a median delay of 1 year in the remaining cases. One possible reason for this improved

diagnosis may be that pediatricians are more highly aware of rare genetic kidney diseases related to inborn error of metabolism, including nephrolithiasis, than nephrologists or urologists. It is also likely that stone episodes are more systematically and extensively investigated in children than in adults. Our results and those of a previous study by our group [2] enable several clinical forms of APRT deficiency to be differentiated: (1) infanthood or childhood onset disease with early nephrolithiasis and/or UTI (approx. 40%); 2) late-onset disease with occasional stone passages in adulthood, possibly renal dysfunction and even ESRD (approx. 50%); (3) asymptomatic disease only diagnosed at family screening (approx. 10%). A reddish-brown diaper strain is also an important presentation in young children [7]. This classification implies that approximately half of patients with late diagnosis are at risk for progressive chronic kidney disease or ESRD and stresses the importance of an early and accurate diagnosis. The diagnosis of the disease is based on stone analysis by infrared spectroscopy or microscopic examination of urine, which reveals typical 2,8-DHA crystals in almost all cases. DHA stones were originally misdiagnosed as uric acid stones because of the identical reactivity of the two compounds in common biochemical tests [20]. Due to this limitation, biochemical analysis should be abandoned, and stone analysis should consist of combined morphologic examination by stereomicroscopy and infrared spectroscopy [17]. In other cases, crystals on renal biopsy were initially identified as calcium oxalate [3]. In our study, crystalluria revealed typical 2,8-DHA crystals in all tested children. We therefore suggest that, crystalluria, when performed by experienced laboratory physicians, is a major diagnostic tool for 2,8-DHA screening and should be proposed to any child with a history of kidney stone. In a second step, the diagnosis must be confirmed by enzyme activity measurements in erythrocyte lysates as these enable the identification of homozygotes (no detectable activity in Caucasians while Japanese patients have enzyme levels of 5–25% of normal activity [20]) and heterozygotes in the Caucasian population (≤40% of normal enzyme activity). Finally, a molecular approach can be used to identify the mutations in the APRT gene. The complete diagnostic procedure can now be based on reasonable recommendations [2], and the results of this study may increase awareness of this rare and underdiagnosed disease. The genetic basis for APRT deficiency has been identified [21]. The APRT gene is located on chromosome 16q24.3, and mutant alleles responsible for the disease have been designated as APRT*Q0 and APRT*J [22]. More than 30 mutations have been reported in APRT*Q0 as predominantly affecting the Caucasian population [2, 23]. The mutation in APRT*J is a single missense mutation exclusively observed in Japanese patients who harbor the genotype APRT*J/

578

APRT*J or more rarely APRT*J/APRT*Q0 [24–26]. Since our patients had complete enzyme deficiency in vitro (type I deficiency) and none exhibited the APRT*J allele, we considered that the novel mutations described in this study responsible for the APRT deficiency were in APRT*Q0. Whole APRT gene sequencing has become increasingly available and is now a valuable diagnostic tool [27]. Although all patients with a homozygous mutation have null APRT activity in vivo, the phenotype is very heterogeneous, even in patients carrying the same mutations. There are some reports of inter- and intra-familial heterogeneity, but currently no clear explanation has been put forward to explain this phenomenon [9, 28]. There is no evidence of genotype– phenotype associations. Environmental factors or modifiers might be responsible for this heterogeneity [29]. As in humans, the severity of the symptoms varies widely among APRT-deficient mice. Untreated animals develop recurrent nephrolithiasis and renal failure, and male mice are more severely affected than female mice [30, 31]. In humans, approximately two-thirds of patients are males. This gender difference may partially be explained by environmental factors, such as variations in dietary habits. The treatment of APRT deficiency is based on a xanthine dehydrogenase inhibitor, allopurinol, which makes adenine the major urinary component. High fluid intake and a low purine diet are usually encouraged. To date, no prospective studies have investigated the effect of allopurinol or other treatments on the clinical outcome of patients with APRT deficiency. However, retrospective studies clearly suggest that allopurinol prevents stone recurrence and even reverses established chronic kidney disease in properly treated patients [2, 7]. None of the patients in our pediatric population showed any side effects from allopurinol treatment, and the treatment appeared to be highly efficient since a majority of patients did not show stone recurrence and the eGFR improved over the long-term follow-up. No patient had an eGFR