Genomic and clinical profiling of a national ... - Kidney International

0 downloads 0 Views 583KB Size Report
Jan 20, 2017 - Agnieszka Bierzynska1, Hugh J. McCarthy1, Katrina Soderquest2, Ethan S. Sen1, ... Rodney D. Gilbert13, Manish D. Sinha14, Graham M. Lord2, Michael Simpson15, Ania B. Koziell2,. Gavin I. Welsh1 and Moin A. Saleem1.
clinical investigation

www.kidney-international.org

Genomic and clinical profiling of a national nephrotic syndrome cohort advocates a precision medicine approach to disease management

see commentary on page 781

Agnieszka Bierzynska1, Hugh J. McCarthy1, Katrina Soderquest2, Ethan S. Sen1, Elizabeth Colby1, Wen Y. Ding1, Marwa M. Nabhan3, Larissa Kerecuk4, Shivram Hegde5, David Hughes6, Stephen Marks7, Sally Feather8, Caroline Jones9, Nicholas J.A. Webb10, Milos Ognjanovic11, Martin Christian12, Rodney D. Gilbert13, Manish D. Sinha14, Graham M. Lord2, Michael Simpson15, Ania B. Koziell2, Gavin I. Welsh1 and Moin A. Saleem1 1

Bristol Renal and Children’s Renal Unit, School of Clinical Sciences, University of Bristol, UK; 2Division of Transplantation Immunology and Mucosal Biology, Department of Experimental Immunobiology, Faculty of Life Sciences and Medicine, King’s College London, London, UK; 3Egyptian group for orphan renal diseases (EGORD), Department of paediatrics, Kasr Al Ainy School of Medicine, Cairo University, Cairo, Egypt; 4Birmingham Children’s Hospital, Birmingham, UK; 5University Hospital of Wales, Cardiff, UK; 6Royal Hospital for Children, Glasgow, UK; 7Great Ormond Street Hospital, London, UK; 8St James’s University Hospital, Leeds, UK; 9Alder Hey Children’s Hospital, Liverpool, UK; 10Department of Paediatric Nephrology and NIHR/Wellcome Trust Clinical Research Facility, University of Manchester, Manchester Academic Health Science Centre, Royal Manchester Children’s Hospital, Manchester, UK; 11Newcastle-upon-Tyne Hospitals NHS Foundation Trust, Newcastle-upon-Tyne, UK; 12Nottingham Children’s Hospital at the QMC, UK; 13Southampton Children’s Hospital and University of Southampton School of Medicine, Southampton, UK; 14Guy’s and St. Thomas’ Hospital, London, UK; and 15 Division of Genetics and Molecular Medicine, Faculty of Life Sciences and Medicine, King’s College London, London, UK

Steroid Resistant Nephrotic Syndrome (SRNS) in children and young adults has differing etiologies with monogenic disease accounting for 2.9–30% in selected series. Using whole exome sequencing we sought to stratify a national population of children with SRNS into monogenic and nonmonogenic forms, and further define those groups by detailed phenotypic analysis. Pediatric patients with SRNS were identified via a national United Kingdom Renal Registry. Whole exome sequencing was performed on 187 patients, of which 12% have a positive family history with a focus on the 53 genes currently known to be associated with nephrotic syndrome. Genetic findings were correlated with individual case disease characteristics. Disease causing variants were detected in 26.2% of patients. Most often this occurred in the three most common SRNS-associated genes: NPHS1, NPHS2, and WT1 but also in 14 other genes. The genotype did not always correlate with expected phenotype since mutations in OCRL, COL4A3, and DGKE associated with specific syndromes were detected in patients with isolated renal disease. Analysis by primary/ presumed compared with secondary steroid resistance found 30.8% monogenic disease in primary compared with none in secondary SRNS permitting further mechanistic stratification. Genetic SRNS progressed faster to end stage renal failure, with no documented disease recurrence Correspondence: Moin A. Saleem, University of Bristol, Children’s Renal Unit and Bristol Renal, Dorothy Hodgkin Building, Whitson Street, Bristol BS1 3NY, United Kingdom. E-mail: [email protected] Received 11 June 2016; revised 8 September 2016; accepted 6 October 2016; published online 20 January 2017 Kidney International (2017) 91, 937–947

post-transplantation within this cohort. Primary steroid resistance in which no gene mutation was identified had a 47.8% risk of recurrence. In this unbiased pediatric population, whole exome sequencing allowed screening of all current candidate genes. Thus, deep phenotyping combined with whole exome sequencing is an effective tool for early identification of SRNS etiology, yielding an evidence-based algorithm for clinical management. Kidney International (2017) 91, 937–947; http://dx.doi.org/10.1016/ j.kint.2016.10.013 KEYWORDS: cytoskeleton; focal segmental glomerulosclerosis; nephrotic syndrome; pediatric nephrology; podocyte; proteinuria Copyright ª 2016, International Society of Nephrology. Published by Elsevier Inc. All rights reserved.

N

ephrotic syndrome is a heterogeneous entity only recently divided into mechanistic categories. Although population analyses have been limited in pediatric cohorts,1–3 idiopathic nephrotic syndrome (INS) has an estimated incidence of approximately 2 to 5 per 100,000 children per year depending on ethnic background.2 INS is currently classified into steroid sensitive (SSNS) or steroid resistant (SRNS), with at least 2.9% to 30% of SRNS cases (in series variably enriched for consanguineous disease or other phenotypes) now known to have an underlying Mendelian, genetic cause. A challenging subset of cases, considered to be immunologically mediated and caused by an as yet unidentified circulating factor(s) can present as secondary SRNS after initial steroid sensitivity.4 The most dramatic evidence of the presence of a circulating factor is rapid recurrence of 937

clinical investigation

A Bierzynska et al.: Genotype/phenotype correlations in SRNS

Table 1 | Fifty-three genes associated with steroid-resistant nephrotic syndrome (SRNS) of congenital, childhood, or adult onset5–11 Gene

Inheritance

Accession #

Disease

AD AR AR AD AD AR AR

NM_004924 NM_024876 NM_019109 NM_018685 NM_001025616 NM_001185078 NM_004357

AD/AR AR AR AR X-linked AR AR AR AR AR AR AD AR AD

NM_012120 NM_000186 NM_000091 NM_000092 NM_000495 NM_015697 NM_182476 NM_173689 NM_001081 NM_003647 NM_001949 NM_001424 NM_022489

AR

NM_002204

ITGB4 KANK1 KANK2 KANK4 LAMB2* LMNA LMX1B* MYO1E* NUP93* NUP107* NUP205 NPHS1* NPHS2* NXF5 OCRL* PAX2 PDSS2 PLCe1 PMM2 PODXL* PTPRO SCARB2 SMARCAL1 SYNPO TRPC6* TTC21B WDR73 WT1*

AR AR AR AR AR AD AD AR AR AR AR AR AR X-linked recessive X-linked recessive AD AR AR AR AD AR AR AR AD AD AR AR AD

NM_000213 NM_015158 NM_015493 NM_181712 NM_002292 NM_170707 NM_002316 NM_004998 NM_014669 NM_020401 NM_015135 NM_004646 NM_014625 NM_032946 NM_000276 NM_003987 NM_020381 NM_016341 NM_000303 NM_005397 NM_030667 NM_005506 NM_014140 NM_007286 NM_004621 NM_024753 NM_032856 NM_024426

XPO5 ZMPSTE24 MYH9 APOL1*

AR AR AD/assoc. G1, G2 risk alleles

NM_020750 NM_005857 NM_002473 NM_003661

Familial and sporadic SRNS (usually adult) SRNS Congenital disorder of glycosylation FSGS (mainly adult) FSGS CNS NS, pretibial bullous skin lesions, neurosensory deafness, bilateral lacrimal duct stenosis, nail dystrophy, and thalassemia minor FSGS/SRNS MPGN type II þ NS Alport’s disease/FSGS Alport’s disease/FSGS Alport’s disease/FSGS Mitochondrial disease/isolated nephropathy NS  sensorineural deafness; DMS SRNS Intermittent nephrotic range proteinuria  with epilepsy Hemolytic-uremic syndrome, SRNS FSGS þ mental retardation (whole gene deletion) Childhood-onset SRNS and SSNS Familial and sporadic SRNS, FSGS-associated Charcot-Marie-Tooth neuropathy Congenital interstitial lung disease, nephrotic syndrome, and mild epidermolysis bullosa Epidermolysis bullosa and pyloric atresia þ FSGS SSNS SSNS/SDNS  hematuria SRNS þ hematuria Pierson syndrome Familial partial lipodystrophy þ FSGS Nail patella syndrome; also FSGS without extrarenal involvement Familial SRNS Childhood SRNS Childhood SRNS Childhood SRNS CNS/SRNS CNS, SRNS FSGS with co-segregating heart block disorder Dent’s disease-2, Lowe syndrome,  FSGS,  nephrotic range proteinuria Adult-onset FSGS without extrarenal manifestations Leigh syndrome CNS/SRNS Congenital disorder of glycosylation FSGS NS Action myoclonus renal failure syndrome  hearing loss Schimke immuno-osseous dysplasia Sporadic FSGS (promoter mutations) Familial and sporadic SRNS (mainly adult) FSGS with tubulointerstitial involvement Galloway-Mowat syndrome (microcephaly and SRNS) Sporadic SRNS (children: may be associated with abnormal genitalia); Denys-Drash and Frasier syndrome Childhood SRNS Mandibuloacral dysplasia with FSGS MYH9-related disease; Epstein and Fechtner syndromes Increased susceptibility to FSGS and ESRD in African Americans, Hispanic Americans and in individuals of African descent

ACTN4* ADCK4* ALG1 ANLN ARHGAP24 ARHGDIA CD151 CD2AP CFH COL4A3* COL4A4 COL4A5* COQ2 COQ6 CRB2* CUBN DGKE* E2F3 EMP2 INF2 ITGA3

AD, autosomal dominant; AR, autosomal recessive; CNS, congenital nephrotic syndrome; DMS, diffuse mesangial sclerosis; ESRD, end-stage renal disease; FSGS, focal segmental glomerulosclerosis; MPGN, membranoproliferative glomerulonephritis; NS, nephrotic syndrome; SDNS, steroid-dependent nephrotic syndrome; SRNS, steroid resistant nephrotic syndrome; SSNS, steroid sensitive nephrotic syndrome. *Genes with a likely or known mutation, or a risk allele, in this cohort.

938

Kidney International (2017) 91, 937–947

clinical investigation

A Bierzynska et al.: Genotype/phenotype correlations in SRNS

Table 2 | Clinical criteria for inclusion Inclusion criteria at age T c.320C>T:p.Ala107Val c.2600G>A:p.Gly867Asp c.500C>T:p.Pro167Leu (H) c.2335-1G>A c.3442C>T:p.Gln1148* c.2335-1G>A (p) c.2491C>T:p.Arg831Cys (m) c.2944dupA:p.Thr982Asnfs*36 (H) c.3250dupG:p.Val1084Glyfs*12 c.3481þ1G>T c.3478C>T:p.Arg1160* (H) c.413G>A:p.Arg138Gln (H) c.413G>A:p.Arg138Gln (H) c.413G>A:p.Arg138Gln c.643C>T:p.Gln215* c.413G>A:p.Arg138Gln c.643C>T:p.Gln215* c.413G>A:p.Arg138Gln c.868G>A:p.Val290Met c.686G>A:p.Arg229Gln c.871C>T:p.Arg291Trp c.686G>A:p.Arg229Gln c.871C>T:p.Arg291Trp c.705_713delTCTAGAGAG:p. Leu236_Arg238del (H) c.1301G>A:p.Arg434His

RS121907901

M

Y

17*

WT1

c.1384C>T:p.Arg462Trp

RS121907900

M

18 19

WT1 WT1 TRPC6pos

RS587776576 RS587776576 RS778276152

20 21

LAMB2 LMX1B

c.1432þ5G>A c.1432þ5G>A c.253_264dupGCATACATGTTT: p.Ala85_Phe88dupAYMF c.736C>T:p.Arg246Trp (H) c.737G>A:p.Arg246Gln (m)

Denys-Drash syndrome Denys-Drash syndrome Frasier syndrome Frasier syndrome No

2 3 4 5 6 7 8 9 10 11 12 13 14

Kidney International (2017) 91, 937–947

RS121912488 No

Pierson syndrome Nail patella–like renal disease

Splicing predictions-ALAMUT

Alternative allele count

Allele number

No. of h/hemi

Allele frequency

Likely pathogenic



4 8 82 82 82 — 82 — 82 9 3526 2 3526 2 —

— 115096 92032 — 121044 120996 121044 120918 — — 121270 121256 121298 121298 121298 — 121298 — 121298 114830 119108 113818 119108 113818 —

— 0 0 — 0 0 0 0 — — 0 0 0 0 0 — 0 — 0 0 69 0 69 0 —

— 1.74E-05 2.17E-05 — 5.78E-05 1.65E-05 5.78E-05 2.48E-05 — — 3.30E-05 6.60E-05 0.000676 0.000676 0.000676 — 0.000676 — 0.000676 7.84E-05 0.0296 1.76E-05 0.0296 1.76E-05 —

None









Y

None









Na Na NA

Na NA NA

Na NA NA

— — —

— — —

— — —

— — —

H M

Y Y

Y Y

1

119680 —

0 —

8.36E-06 —

— 8

— 121256

— 0

— 6.60E-05

MutPred

SIFT

MT

2 2 — Likely pathogenic

7 2 7 3

Likely pathogenic — — Likely pathogenic

Likely pathogenic Likely pathogenic



Probably 22

NPHS1 NPHS1

Deletion of exon 8 c.3478C>T:p.Arg1160*

Novel RS267606919

No CNS

Na Na

Na Na

Na Na

A Bierzynska et al.: Genotype/phenotype correlations in SRNS

Definitely

HGMD phenotype (professional 2015.2)

clinical investigation

940

Table 4 | Known and potential mutations in the 49 patients with monogenic nephrotic syndrome

Probably 23

NPHS1 NPHS1

c.2227C>T:p.Arg743Cys c.2387G>A:p.Gly796Glu

RS386833909 Novel

24

26

NPHS1 NPHS1 NPHS2 NPHS2 NPHS2

c.2335-1G>A c.3061dupG:p.Asp1021Glyfs*6 c.156delG:p.T53Pfs*46 c.413G>A:p.Arg138Gln c.378þ1_378þ2delinsTG

RS150038620 RS34124941 Novel RS74315342 Novel

27

NPHS2 NPHS2

c.413G>A:p.Arg138Gln c.378þ1_378þ2delinsTG

RS74315342 Novel

29

NPHS2 NPHS2 NPHS2 ACTN4

RS74315342 RS771280420 RS74315342 Novel

30

TRPC6

31 32

MYO1E COL4A3

33

COL4A3 DGKE DGKE

c.413G>A:p.Arg138Gln c.378þ5G>A c.413G>A:p.Arg138Gln c.778_786delTATGTGTCC: p.Tyr260_Ser262del (not m) c.523C>T:p.Arg175Trp (not m) c.2094T>A:p.Tyr698* (H) c.2621_2622delGAinsT: p.Gly874Valfs*9 (m) c.4803delT:p.Gly1602Alafs*13 (p) c.966G>A:p.Trp322* c.1303C>T:p.Arg435*

25

28†

CNS if p.G796R -> DMS CNS No No SRNS if c.378þ1G>A -> SRNS SRNS If c.378þ1G>A ->SRNS SRNS SRNS SRNS If S262F or S262P -> FSGS

H L

Y Y

Y Y

Na Na Na H Na

Na NA Na Y Na

Na NA Na Y Na

H Na

Y Na

Y Na

H Na H Na

Y Na Y Na

Y Na Y Na

2 —

121386 —

0 —

1.65E-05 —

Likely pathogenic

7 — — 82 —

121044 — — 121298 —

0 — — 0 —

5.78E-05 — — 0.000676 —

Likely pathogenic

82 —

121298 —

0 —

0.000676 —

82 2 82 —

121298 121194 121298 —

0 0 0 —

0.000676 1.65E-05 0.000676 —

Likely pathogenic

Likely pathogenic

Novel

If R175Q ->FSGS

M

Y

Y









Novel No

No Alport Syndrome

Na Na

Na Na

Na Na

— —

— —

— —

— —

RS760846085 RS138924661 Novel

No aHUS No

Na Na Na

Na Na Na

Na Na Na

1 7 —

120346 121398 —

0 0 —

8.31E-06 5.77E-05 —

Novel Novel RS747849728 Novel RS147641617 RS763162233 Novel RS200319221 Novel Novel No RS147596471 Novel Novel RS202128397 Novel

No No No if p.C528F -> CNS No NS (?) No No No No Nephropathy (?) Nephropathy (?) SRNS SRNS SRNS SRNS

Y Y Y Na Y No Na Y Y Y Y Y Na Na Y Na

— — 1 — 12 — — — — — — 180 — — 2 —

— — 115602 — 112690 — — — — — — 117904 — — 119188 —

— — 0 — 0 — — — — — — 0 — — 0 —

— — 8.65E-06 — 0.0001065 — — — — — — 0.001527 — — 1.68E-05 —

A Bierzynska et al.: Genotype/phenotype correlations in SRNS

Kidney International (2017) 91, 937–947

Table 4 | (Continued)

Possibly novel and potentially pathogenic 34 35 36

41§ 42¶

c.136G>T:p.Gly46Trp c.925G>A:p.Glu309Lys c.325T>C:p.Tyr109His c.1584C>T:p.¼ (p.Cys528Cys) c.1747G>A:p.Glu583Lys c.2734G>A:p.Ala912Thr c.1910_1912delTCT: p.Phe637del (H) c.2345A>T:p.Lys782Ile c.676C>T:p.Leu226Phe c.3097G>C:p.Gly1033Arg (Hemi) (m) c.352A>G:p.Lys118Glu (H) (m, p) c.2627C>G:p.Thr876Arg (H) (m, p) c.101G>A:p.Trp34* (p) c.954_956dupGAC:p.Thr319dup (m) c.1882C>T:p.Arg628Cys (m) c.3089_3104dupGGCCCGGCGCGGCCCC: p.Gly1036Alafs*43 (de novo)

H M M Na M L Na M L H M H Na Na M Na

Y No Y Na No No Na Y Y No Y Y Na Na Y Na

Likely pathogenic

(Continued on next page) 941

clinical investigation

37 38 39 40‡

NPHS1 NPHS1 NPHS1 NPHS1 NPHS1 NPHS1 NPHS1 TRPC6 LMX1B COL4A5 MYO1E MYO1E ADCK4 ADCK4 CRB2 CRB2

942

Using Alamut Visual 2.7 (SpliceSiteFinder-like, MaxEntScan, NNSPLICE, GeneSplicer, Human Splicing Finder) variants were considered likely pathogenic where there was a consistent predicted splice effect across most of the tools. aHUS, atypical hemolytic uremic syndrome; DP, disease-associated polymorphism; ExAC, The Exome Aggregation Consortium (http://exac.broadinstitute.org); F, familial; H, homozygote; Hemi, hemizygote; m, maternal; MT, MutationTaster; MutPred predictions: L = low risk, M = medium risk, H = high risk; Na, not applicable; NS, nephrotic syndrome; p, paternal; pos, this TRPC6 variant is considered possibly pathogenic here; SIFT, MT, and splicing predictions from Alamut; SRNS, steroid resistant nephrotic syndrome; Y, deleterious. Splicing predictions for the synonymous variant in NPHS1 found in patient 35 are presented in Supplementary Figure S1. *Patient included in Tanteles et al.12 † c.378þ5G>,13 patients 26 and 27 are siblings, patients 45, 46, and 47 are siblings. ‡ Patient included in Lennon et al.14 § Patient included in Ashraf et al.15 ¶ Patient included in Ebarasi et al.7

Y Na Y Y Y Y Y Y Y Na Y Y Y Y Y Y M Likely L L L H Y Y No No No No No No No No Novel Novel RS730882216 RS730882216 RS730882216 RS745342141 RS764134014 RS748221967 c.1427A>T:p.His476Leu c.1467-2A>G (Hemi) (m) c.303G>A:p.Met101Ile (H) c.303G>A:p.Met101Ile (H) c.303G>A:p.Met101Ile (H) c.1325G>A:p.Cys442Tyr (H) c.2084T>C:p.Leu695Ser c.2267T>C:p.Leu756Ser PODXL OCRL NUP107 NUP107 NUP107 NUP107 NUP93 NUP93 43 44 45 46 47 48 49

Possibly novel and potentially pathogenic

Table 4 | (Continued) Known and potential mutations in the 49 patients with monogenic nephrotic syndrome

Likely pathogenic Likely pathogenic Likely pathogenic

— — 1 1 1 1 3 —

— — 120758 120758 120758 121196 121398 —

— — 0 0 0 0 0 —

— — 8.28E-06 8.28E-06 8.28E-06 8.25E-06 2.47E-05 —

clinical investigation

A Bierzynska et al.: Genotype/phenotype correlations in SRNS

whether this was linked to the clinical predictability of disease progression, recurrence risk after transplantation, and mechanistically targeted disease management. RESULTS Cohort characteristics and incidence of genetic disease

The clinical details of the 187 cases screened are presented in Table 3, with more detailed phenotyping shown in Supplementary Table S1. Twenty-two of 187 patients (12%) had familial disease, 164 patients were sporadic, and 1 patient was unknown (presumed sporadic). Fifty-nine percent (13 of 22) of patients in the familial group and 22% (36 of 164) in the sporadic group had an identified mutation or a variant likely to be diseasecausing. Most familial patients had a mutation in recessive or dominant negative (WT1) genes (n ¼ 39), and this group of patients had a trend toward an earlier age of onset compared with patients with a dominant gene mutation (n ¼ 10) (unpaired t test; 2-tailed; P ¼ 0.07). Sixty-nine of 187 patients (36.9%) developed end-stage renal failure. Fifty-four cases underwent renal transplantation, and the other 15 cases were on peritoneal dialysis or hemodialysis at last follow-up. The subgroup that had reached chronic kidney disease (CKD) stage 5, and that had a genetic etiology of disease, had an earlier age of onset compared with CKD stage 5 cases with SRNS in whom gene mutations had not been identified (4.75 years vs. 6.28 years; P ¼ 0.0082; unpaired t test). No such difference was observed between patients with primary and/or presumed and secondary SRNS (4.86 years vs. 5.64 years; P ¼ 0.6321; unpaired t test). Recurrence of the disease was noted in 27.8% of the patients who had transplantations. The rate of recurrence in those with a genetic form was 0 per 25 transplantations, and with nongenetic and/or unknown, the recurrence rate was 15 of 29 transplantations (51.7%) (Fisher exact test; 2-sided; P < 0.0001). This phenotypic analysis confirms that patients with an identified gene mutation are more likely to progress faster to end-stage renal failure (Figure 1). In this cohort, 59.8% of genetic and 19.4% nongenetic and/or unknown SRNS patients reached CKD stage 5 within 4 years from diagnosis. The mean length of follow-up for both patient groups was 7.6 years (range, 1.0018.03) and 6.2 years (range, 0.5816.55), respectively. Patients with a mutation in NPHS1 had an earlier mean age of onset than those with a mutation in NPHS2 (0.72 and 5.94 years, respectively). Thirteen patients with congenital nephrotic syndrome (CNS) were found to have known or potential mutations in NPHS1, LAMB2, and WT1. Histopathological findings

A renal biopsy was undertaken in 181 of 187 (96.8%) patients within 3 months of their date of diagnosis (3 patients with CNS were not biopsied, and 3 patients’ results were not available). Most patients had FSGS (54.1%; >98 of 181) or minimal change disease (MCD) (23.8%; >43 of 181) on biopsy; however, other variations (Table 3) were also Kidney International (2017) 91, 937–947

clinical investigation

A Bierzynska et al.: Genotype/phenotype correlations in SRNS

Age at Onset

Number of patients

60

Genetic

50

Number:

Number:

Genetic

Genetic testing

Cause

negative

Diagnosed