Clinical and Molecular Spectrum of Nonsyndromic Early-Onset Osteoarthritis

Texte intégral

(1)Clinical and Molecular Spectrum of Nonsyndromic Early-Onset Osteoarthritis V. Ruault, K. Yauy, Aude Fabre, M. Fradin, J. van Gils, C. Angelini, G Baujat, J. P. Blanchet, S. Cuinat, B. Isidor, et al.. To cite this version: V. Ruault, K. Yauy, Aude Fabre, M. Fradin, J. van Gils, et al.. Clinical and Molecular Spectrum of Nonsyndromic Early-Onset Osteoarthritis. Arthritis & rheumatology, Wiley, 2020, 72 (10), pp.16891693. �10.1002/art.41387�. �hal-02934316�. HAL Id: hal-02934316 https://hal.archives-ouvertes.fr/hal-02934316 Submitted on 9 Sep 2020. HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés..

(2) Accepted Article. MR. VALENTIN RUAULT (Orcid ID : 0000-0003-0562-1706). Article type. : Brief Report. Clinical and molecular spectrum of non-syndromic early-onset osteoarthritis. Clinical Trial: ID NCT04267510. Authors: Valentin Ruault (MS)1, Kevin Yauy (MD)1,2,3, Aurélie Fabre (MLT)1,4,MélanieFradin (MD)5, Julien Van-Gils (MD)6, ChloéAngelini (MS)6, Geneviève Baujat (MD)7, Patricia Blanchet (MD)1, Silvestre Cuinat (MS)8, Bertrand Isidor (MD, PhD)8, Christian Jorgensen (MD, PhD)9,10, Didier Lacombe (MD, PhD)6, SébastienMoutton (MD)11, Sylvie Odent (MD, PhD)5, Elodie Sanchez (MLT)1,10, Sabine Sigaudy (MD)12, Isabelle Touitou (MD, PhD)1,4,10,Marjolaine Willems (MD)1, Florence Apparailly (PhD)1,10, David Geneviève (MD,. PhD)1,10,$*, Mouna Barat-Houari (PharmD, PhD)1,4,$. $ Both authors contributed equally to this work. * Corresponding Author Pr David Geneviève Département de Génétique Médicale, CHU de Montpellier, Arnaud de Villeneuve 371 avenue du doyen Gaston Giraud, 34000 Montpellier, France.

(3) Accepted Article. Tel: +33 4 67 33 65 64 Fax: +33 4 67 33 60 52 Email: d-genevieve@chu-montpellier.fr. This work was supported by the medical genetics department of Montpellier’s UHC.. 1- Département de Génétique Médicale, Maladies rares et médecine personnalisée, Centre de compétence maladies osseuses, Univ Montpellier, CHU Montpellier, CLAD Sud Languedoc-Roussillon, Montpellier, France 2- Institute of Advanced Biosciences, Centre de recherche UGA, Inserm U 1209, CNRS UMR 5309, Grenoble, France 3- SeqOne, Montpellier, France 4- Autoinflammatorydiseases unit, Univ Montpellier, CHU Montpellier, Montpellier, France 5- Service de Génétique Clinique, Hôpital Sud, CLAD Ouest, CNRS UMR6290 Génétique et Pathologies du Développement, Université de Rennes, Rennes, France 6- Service de Génétique, Hôpital Pellegrin, CLAD Sud-Ouest, Bordeaux, France 7- Service de Génétique Clinique, Hôpital Necker-Enfants Malades, Paris, France 8- Service de Génétique, CHU Nantes, CLAD Ouest, Nantes, France 9- Département de rhumatologie, Service d’immuno-rhumatologie, Univ Montpellier, CHU Montpellier, Montpellier, France 10- IRMB, Univ Montpellier, CHU Montpellier, INSERM, Montpellier, France.

(4) Accepted Article. 11- Centre Pluridisciplinaire de Diagnostic Prénatal, Pôle Mère-Enfant, Maison de Santé Protestante de Bordeaux-Bagatelle, Talence, France 12- Département de génétique médicale, CHU de Marseille - Hôpital de la Timone, Marseille, France. Word count (excluding title page, abstract, references, figures and tables): 1670.

(5) Accepted Article. ABSTRACT Objective. Osteoarthritis (OA) is the most common joint disease worldwide. The etiology of OA is varied, ranging from multifactorial, environmental to monogenic. OA can occur earlier in some individuals than in the general population, called early-onset osteoarthritis (EOOA). To our knowledge, we lack large-scale genetic studies of individuals with EO-OA. Here we aimed to study monogenic osteoarthritis causes in individuals with nonsyndromic EO-OA. Methods. From 2013 to 2019, skeletal dysplasia experts referred non-syndromic EO-OA probands to our skeletal disease center. Criteria for EO-OA are based on: X-Ray evidence, BMI ≤ 30, age of onset ≤ 50 and ≥ 1 joint site involved. Molecular analyses involved nextgeneration sequencing panel approach. Results. We identified a pathogenic variant in 13/45 (29%) probands: 11 COL2A1, 1 ACAN and 1 SLC26A2. After familial segregation analysis, 20 additional individuals were included. Mean age of onset of joint pain was 19.5 ±3.9 (95% CI) years. Eighteen out of 33 (55%) individuals presenting non-syndromic EO-OA and carrying a pathogenic variant had at least one joint prosthesis (mean age: 41 years ±4.2 (95% CI); mean number of prosthesis: 2.6 per individual), and 21 (45%) of the joint replacement surgeries were performed before age 45 years. For the 20 individuals > 40 years old, 17 (85%) had at least one prosthesis. Conclusion. We confirmed that COL2A1 is the main monogenic cause of non-syndromic EO-OA. However, on the basis of genetic heterogeneity of EO-OA, we recommend next-.

(6) generation sequencing for all individuals with joint replacement before age 45 years. Lifestyle recommendations for prevenion should be implemented.. Accepted Article. INTRODUCTION Osteoarthritis (OA) is the most common degenerative joint disorder worldwide, one of the main sources of morbidity in developed countries and a major socioeconomic burden for healthcare systems. OA is characterized by joint cartilage damage, subchondral sclerosis and the formation of osteophytes, leading to severe pain and loss of joint function. Treatments are symptomatic and usually consist of joint replacement surgery. Unlike secondary OA, which is caused by another disease or trauma, infection or metabolic disorders, primary OA is considered a complex multifactorial disease according to previous epidemiological and molecular data(1). The estimated heritability (i.e., the part of phenotypic variance of genetic origin) of OA is about 50%(2). The common forms of OA are multifactorial, whereas rare forms of earlyonset OA (EO-OA) are dominantly inherited Mendelian diseases. At least 13 genes have been reported for non-syndromic forms of EO-OA: ACAN, COL2A1, COL9A3, COL11A2, COMP, MATN3, PRG4, RIPK2, SMAD3, TNFRSF11B, TRAPPC2, TRPV4, TUFT1 (MIM *155760, +120140, *120270, *120290, *600310, *602109, *604283, *603455, *603109, *602643, *300202, *605427, *600087, respectively)(3–13). To our knowledge, we lack a large-scale genetic overview of non-syndromic EO-OA or clinical follow-up of EO-OA patients with a known molecular characterization. Therefore, we aimed to study the clinical and molecular spectrum of non-syndromic EO-OA in a cohort of French individuals referred for molecular analysis. PATIENTS AND METHODS. Participants. From 2013 to 2019, EO-OA probands were referred to the Montpellier skeletal disease center by geneticist experts in skeletal dysplasia from 8 French hospitals who determined that affected individuals had EO-OA without chondrodysplasia. Inclusion criteria for primary EO-OA were based on Aury-Landas et al, (2016): no known cause of secondary.

(7) 2. Accepted Article. OA, body mass index (BMI) ≤ 30 kg/m before the first symptoms of OA, radiologically proven OA and age at OA onset ≤ 50 years (Supplementary Table 1). We systematically requested skeletal X-rays of affected joints, and extensive skeletal survey that have been review by bone dysplasia experts (competence or reference centre for skeletal dysplasia) during dedicated bone dysplasia meetings. To exclude syndromic EO-OA, we specifically checked for reported palatal or subtle cleft, ocular and hearing features in all individuals. We collected data regarding the age for the first joint pain and first joint replacement surgery, BMI, metabolic disease, profession, and sports practice. Genetic testing. We. used. the. Agilent. online. SureDesign. tool. (https://earray.chem.agilent.com/suredesign/) for designing the panel, targeting 49 genes associated with congenital skeletal disorders including OA (Supplementary Table 2). We prepared Illumina multiplexed paired-end libraries with the SureSelectQXT Target Enrichment system according to the manufacturer’s protocol, from genomic DNA for all probands. We then indexed, pooled and sequenced captured libraries by using the NextSeq® 500/550 MidOutput Kit v2 (300 cycles), according to standard protocols. We used SeqNext NGS analysis software (JSI) for read alignments, variant calling and annotation and classified variants in 5 pathogenicity groups according to the American College of Medical Genetics 2015 guidelines, then performed confirmation analysis of all variants of interest using conventional ABI Sanger sequencing restricted to exons of interest.. We. used. GnomAd. (https://gnomad.broadinstitute.org/). and. Varsome. (https://varsome.com/) to help variant classification. Segregation of the genetic variants. identified in index cases was performed using Sanger sequencing in all family members (Supplementary Figures 1-13). Consent. This study was register in Clinical Trial (ID NCT04267510). All individuals, parents or legal guardians gave their written informed consent for genetic diagnosis procedures.

(8) Accepted Article. according to the ethics committee of our institution. This study have been performed in accordance with the highest ethical standards of our national research committee and with the Helsinki Declaration. Ethic approval was given by Montpellier University Hospital Institutional Review Board (IRB-MPT_2020_03_202000373)..

(9) Accepted Article. RESULTS Cohort constitution. The 45 EO-OA probands were from 3 to 45 years old, with mean and median age of onset of joint pain of 19.2 and 16, respectively. We found a genetic variant in 13 (29%) probands who were then included in this cohort. After a familial cascade screening, the final cohort consisted of 33 EO-OA individuals with a genetic variant (Figure 1). Clinical findings. We report clinical characteristics of individuals in Table 1 and Supplementary Table 3. In this cohort, 18/33 (55%) individuals had at least one joint prosthesis, at mean and median age 41 ±4.2 (95% CI) and 42 years, respectively. These 18 individuals had a total of 47 joint prostheses (mean 2.6 per person); 21 (45%) joint replacement surgeries were performed before age 45 years. For the 20 individuals > 40 years old, 17 (85%) had at least one joint prosthesis. Hip was by far the most often affected joint: 15 individuals had at least one total hip replacement (THR) (Supplementary Figure 14). None of these individuals had any confounding diagnosis that are common reason for early hip replacement such as avascular necrosis or Perthes disease. Although none of these individuals had BMI > 30 kg/m2 at their first articular pain, 16% were obese at the last examination (Supplementary Table 3), a proportion not significantly different from the 17.2% obesity rate in the French general population(14). Supplementary Figures 15a and 15b are representative of abnormal X-rays in individuals from this cohort. Molecular findings. We identified pathogenic or likely pathogenic variants in 13/45 (29%) probands. All variants characteristics are shown in Supplementary Table 4. Eleven (85%) carried COL2A1 variants: 6 an arginine to cysteine substitution, 4 a glycine. substitution and 1 a frameshift variant localized in the chondrocalcin domain and probably altered the stability of the procollagen (Figure 2 and Supplementary Figure 16). After.

(10) Accepted Article. familial segregation analysis, we included 19 additional individuals carrying a COL2A1 pathogenic or likely pathogenic variant in this cohort. Also, we identified a heterozygous pathogenic nonsense variant in ACAN [NM_013227; c.2441C>G; p.(Ser814*)] in a mother–son dyad and a likely pathogenic homozygous variant in SLC26A2 [(MIM *606718): NM_000112; c.835C>T; p.(Arg279Trp)] in individual 31.. The final cohort consisted of 33 EO-OA individuals with a genetic variant (Figure 1). Although dominant negative variants in COL2A1 seem to cause a more severe OA phenotype than variants in ACAN or SLC26A2, we found no statistical difference between individuals on the basis of the involved gene or the nature of the genetic variant (Supplementary Tables 4-6)..

(11) Accepted Article. DISCUSSION We report clinical and molecular data for the first and largest cohort presenting nonsyndromic EO-OA. Clinical exploration of this non-syndromic EO-OA cohort led us to highlight diagnostic criterion, comorbidity prognostic and lifestyle recommendations. Although height is not a diagnostic criterion for primary EO-OA according to Aury-Landas et al. (2016), 3 individuals from our cohort could have had a syndromic form of EO-OA because they were at < -2 SD height. Nevertheless, they don’t show any other signs (normal sight and hearing, normal skeletal survey) that would suggest a more generalised skeletal dysplasia. Normal stature could be an additional criterion for nonsyndromic EO-OA. According to the literature, OA is the most frequent condition for THR and total knee replacement (TKR)(15). Mean age for THR and TKR in individuals from this series was significantly lower than in the French general population: 42 and 43.5 versus 72.8 and 71 years old (p=0.007)(16,17). In the general population, only 3.2% and 1.2% of THR and TKR, respectively, involved individuals < 45 years old(15), as compared with 45% of joint replacement surgery for individuals < 45 years old in our series (p=0.0001). Furthermore, joint surgery in a context of primary OA of articulations such as the wrist, ankle or shoulder is rare in the general population but concerned 5/33 (15%) individuals with a genetic variant in our series. We suggest referring all individuals who fit the EO-OA criteria from Aury-Landas et al (2016) for a genetic analysis as well as those who underwent joint replacement before age 45 in a context of primary EO-OA, especially for an unusual articulation. We also suggest referring first-degree relatives of individuals with EO-OA, even if the proband could not have genetic testing. An early diagnosis must lead to strict lifestyle interventions. Indeed, obesity aggravates the vicious cycle: pain, sedentary lifestyle, obesity, OA, pain. Regular non-traumatic sports activity, dietary instructions, professional reorientation and improved workstation ergonomics if necessary should be proposed to all patients. These precautions would help delay the onset of OA as well as the joint replacement surgery..

(12) Accepted Article. The most frequently involved gene is COL2A1, known to cause a wide spectrum of genetic disorders gathered under the name “COL2A1-related disorders” (COL2A1-RD). In most cases of COL2A1-RD, individuals have OA(11). Collagens are the major structural components of the extracellular matrix; type II collagen is the major collagen synthesized by chondrocytes. COL2A1 protein consists of 3 alpha-1 chains; glycine is the most important amino acid of this triple helical domain with a Gly– X–Y repeat motif. As expected, all COL2A1 variants we identified in our EO-OA series but one were glycine or arginine substitutions, located in the triple-helical domain of the protein. Nevertheless, because of the small size of this cohort, we identified neither obvious variant hotspots nor clear genotype–phenotype correlations. The limitations of this study include the question of updating the panel with recently published genes such as TNFRSF11B,(7) (mostly known for severe forms of Paget’s. disease), RIPK2(12) and TUFT1 (only reported once in the literature)(9). For a maximized cost-efficiency of the molecular analysis, exome or genome sequencing should be preferred because they have become the gold standard for sequencing analyses. To conclude, this comprehensive study of 45 probands emphasizes the high rate (29%) of monogenic etiology of EO-OA. COL2A1 is by far the most-often involved gene in EOOA (85% of monogenic cases). Among the 33 individuals with a genetic variant, the mean age at first joint replacement surgery was 41 years. The first joint replacement surgery was before age 45 for 61% of individuals. Moreover, 85% of individuals over 40 years had at least one joint prosthesis. We recommend genetic testing in individuals presenting EO-OA and those who undergo joint replacement surgery before age 45..

(13) Accepted Article. REFERENCES 1. Allen KD, Golightly YM. State of the evidence. Curr Opin Rheumatol 2015;27:276–283. 2. Zengini E, Hatzikotoulas K, Tachmazidou I, Steinberg J, Hartwig FP, Southam L, et al. Genome-wide analyses using UK Biobank data provide insights into the genetic architecture of osteoarthritis. Nat Genet 2018;50:549–558. 3. Jakkula E, Melkoniemi M, Kiviranta I, Lohiniva J, Räinä SS, Perälä M, et al. The role of sequence variations within the genes encoding collagen II, IX and XI in non-syndromic, early-onset osteoarthritis. Osteoarthritis and Cartilage 2005;13:497–507. Available at: http://dx.doi.org/10.1016/j.joca.2005.02.005. 4. Mu S-C, Lin Y-J, Liu H-C, Wu J-Y, Li S-C, Michael Lee M-T, et al. A mutation in cartilage oligomeric matrix protein (COMP) causes early-onset osteoarthritis in a large kindred study. Ann Hum Genet 2011;75:575–583. 5. Gu J, Rong J, Guan F, Jiang L, Tao S, Guan G, et al. MATN3 gene polymorphism is associated with osteoarthritis in Chinese Han population: a community-based casecontrol study. ScientificWorldJournal 2012;2012:656084. 6. Jeong C, Lee JY, Kim J, Chae H, Park H-I, Kim M, et al. Novel COL9A3 mutation in a family diagnosed with multiple epiphyseal dysplasia: a case report. BMC Musculoskeletal Disorders 2014;15. Available at: http://dx.doi.org/10.1186/1471-2474-15-371. 7. Ramos YFM, Bos SD, Breggen R van der, Kloppenburg M, Ye K, Lameijer E-WEMW, et al. A gain of function mutation in TNFRSF11B encoding osteoprotegerin causes osteoarthritis with chondrocalcinosis. Ann Rheum Dis 2015;74:1756–1762. 8. Aury-Landas J, Marcelli C, Leclercq S, Boumédiene K, Baugé C. Genetic Determinism of Primary Early-Onset Osteoarthritis. Trends Mol Med 2016;22:38–52. 9. Sliz E, Taipale M, Welling M, Skarp S, Alaraudanjoki V, Ignatius J, et al. TUFT1, a novel candidate gene for metatarsophalangeal osteoarthritis, plays a role in chondrogenesis on a calcium-related pathway. PLoS One 2017;12:e0175474..

(14) Accepted Article. 10. Dateki S, Nakatomi A, Watanabe S, Shimizu H, Inoue Y, Baba H, et al. Identification of a novel heterozygous mutation of the Aggrecan gene in a family with idiopathic short stature and multiple intervertebral disc herniation. Journal of Human Genetics 2017;62:717–721. Available at: http://dx.doi.org/10.1038/jhg.2017.33. 11. Kannu P, Bateman J, Savarirayan R. Clinical phenotypes associated with type II collagen mutations. Journal of Paediatrics and Child Health 2012;48:E38–E43. Available at: http://dx.doi.org/10.1111/j.1440-1754.2010.01979.x. 12. Jurynec MJ, Sawitzke AD, Beals TC, Redd MJ, Stevens J, Otterud B, et al. A hyperactivating proinflammatory RIPK2 allele associated with early-onset osteoarthritis. Hum Mol Genet 2018;27:2383–2391. 13. Kannu P, Bateman JF, Randle S, Cowie S, Sart D du, McGrath S, et al. Premature arthritis is a distinct type II collagen phenotype. Arthritis Rheum 2010;62:1421–1430. 14. Matta J, Carette C, Rives Lange C, Czernichow S. [French and worldwide epidemiology of obesity]. Presse Med 2018;47:434–438. 15. Lenza M, Ferraz S de B, Viola DCM, Garcia Filho RJ, Cendoroglo Neto M, Ferretti M. Epidemiology of total hip and knee replacement: a cross-sectional study. Einstein 2013;11:197–202. 16. Putman S, Girier N, Girard J, Pasquier G, Migaud H, Chazard E. Épidémiologie des prothèses de hanche en France : analyse de la base nationale du PMSI de 2008 à 2014. Revue de Chirurgie Orthopédique et Traumatologique 2017;103:S90. 17. Colas S, Occean B-V, Rudnichi A, Dray-Spira R, Zureik M. Étude d’utilisation des prothèses articulaires de genou en France entre 2008 et 2013. Revue d’Épidémiologie et de Santé Publique 2016;64:S23.. FIGURES.

(15) Accepted Article. Figure 1: Cohort flowchart and genes involved for individuals with early-onset osteoarthritis (EO-OA) Figure 2: Schematic representation of COL2A1 pathogenic variants found in this series (NM_001844).

(16) Accepted Article. TABLES Table 1: Mean clinical data for 33 individuals with non-syndromic early-onset osteoarthritis (EO-OA) and with a genetic variant. Age at last examination. Males. Females. Total. n=8. n=25. n=33. 43.0 [13-68]. 40.2 [7-70]. 40.9 [7-70]. 175.1 [164.5-181]. 161.1 [140-175]. 164.5 [140-181]. -0.13. -0.31. -0.27. 80.1 [66-94]. 60.2 [41-95]. 65.2 [41-95]. +1.93. +1.07. +1.33. 25.1 [19.5-30.3]. 22.9 [15.3-36]. 23.5 [15.3-36]. 27 [19-47]. 17.4 [3-45]. 19.5 ±3.854. (years), [range]. Height (only >18 years old) (cm), [range]. Height (SD). Weight (only >18 years old) (kg), [range]. Weight (SD). BMI (kg/m2), [range]. Age at first articular pain (years), [range]. (95% CI) [3-47].

(17) Accepted Article. Location of articular pain – n of individuals. Shoulder. 3/8 (38%). 8/25 (32%). 11/33 (33%). Elbow. 2/8 (25%). 4/25 (16%). 6/33 (18%). Wrist. 3/8 (38%). 10/25 (40%). 13/33 (39%). Hand. 4/8(50%). 14/25 (56%). 18/33 (55%). Vertebral column. 6/8 (75%). 21/25 (84%). 27/33 (82%). Hip. 6/8 (75%). 20/25 (80%). 26/33 (79%). Knee. 4/8 (50%). 17/25 (68%). 21/33 (64%). Ankle. 3/8 (38%). 8/25 (32%). 11/33 (33%). Foot. -. 3/25 (12%). 3/33 (9%). Age at first joint replacement surgery (years), [range].

(18) Accepted Article. Shoulder. -. 50.3 [49-51]. 50.3 [49-51]. Elbow. -. -. -. Wrist. 34 [-]. -. 34 [-]. Hand. -. -. -. Vertebral column. -. 45 [-]. 45 [-]. Hip. 40 [23-54]. 42.8 [30-59]. 42 [23-59]. Knee. 45 [40-50]. 42.8 [36-48]. 43.5 [36-50]. Ankle. -. 41 [40-42]. 41 [40-42]. Foot. -. 35 [-]. 35 [-]. BMI: body mass index.

(19) Accepted Article. KEY MESSAGES. What is already known about this subject? ● Non-syndromic. early-onset. osteoarthritis. (EO-OA). etiologies. are. considered mainly multifactorial but could also be caused by Mendelian monogenic disorders.. What does this study add? ● 85% of individuals > 40 years old with a molecular cause of EO-OA had at least one joint replacement surgery, mostly hip: the mean age for the first joint replacement surgery was 41 years. ●. COL2A1 was the major gene involved in non-syndromic EO-OA in the general population.. How might this impact on clinical practice or future developments? ●. Routine genetic testing is recommended for individuals presenting EOOA or who have undergone early joint replacement surgery before age 45 years.. ●. Lifestyle recommendations for individuals with EO-OA: non-traumatic sport, workstation ergonomics and analgesic monitoring..

(20) Accepted Article. Acknowledgements We thank all families involved in this study Competing interests. None. Funding. This work was supported by the medical genetics department of Montpellier’s UHC. FIGURES. Figure 1: Cohort flowchart and genes involved for individuals with early-onset osteoarthritis (EO-OA) Figure 2: Schematic representation of COL2A1 pathogenic variants found in this series (NM_001844).

(21) Accepted Article. art_41387_f1.tif.

(22) Accepted Article. art_41387_f2.tif.

(23)