LAUSR

Abstract Osteogenesis imperfecta (OI) is clinically and genetically heterogeneous. Defects in collagen type 1 are reportedly the main cause of OI (85-90%), but most available data has arisen from developed countries. Massively parallel sequencing (MPS) technologies now allow for systematic and comprehensive analysis of OI genes simultaneously. Our objective was to obtain the molecular diagnosis of OI through targeted MPS in a single tertiary center cohort of Brazilian adults with OI. After informed consent, DNA samples were obtained from 52 cases of OI (9 families and 43 sporadic; total of 64 sequenced individuals, 96% adults). Sixty-nine percent of the cohort had moderate to severe OI, and consanguinity was common (22%). Coding regions and 25-bp boundaries of 15 OI genes (COL1A1, COL1A2, IFITM5 [plus 5’UTR], SERPINF1, CRTAP, P3H1, PPIB, SERPINH1, FKBP10, PLOD2, BMP1, SP7, TMEM38B, WNT1, CREB3L1) were captured with Agilent SureSelectXT and sequenced in Illumina NextSeq. Identified variants were classified according to ACMG/AMP guidelines and those considered to be disease-causing were confirmed by Sanger sequencing. Segregation analysis was pursued when familial samples were available. A molecular diagnosis was obtained in 92% of cases. Altogether, 55 variants were identified, 21 of which had never been previously reported in international databases. Variants in COL1A1 or COL1A2 were identified in 77%, whereas 23% had variants in other candidate genes. Amongst these, variants in SERPINF1, FKBP10 and PLOD2 were more prevalent. Only one case had the previously described IFITM5 c.-14C>T variant. A peculiar combination of four heterozygous P3H1 and WNT1 variants was detected in a non-consanguineous case, where one variant in each gene was inherited from each parent. In two cases, potentially modifier variants in LRP5 were identified. Surprisingly, in four consanguineous families the molecular cause was still related to COL1A1 or COL1A2, and two non-consanguineous cases had compound heterozygous PLOD2 variants. In conclusion, targeted MPS has effectively allowed establishing the molecular basis of OI in this Brazilian cohort, unraveling novel disease-causing variants in 29% of cases, and potentially reflecting new aspects of OI pathogenesis in Brazil. Non-collagen defects were found in 23% of cases, with a higher prevalence of P3H1, FKBP10, PLOD2 and SERPINF1 defects, with potential digenic interactions, and a lower prevalence of IFITM5-related OI. Inferring the molecular diagnosis from a family history of consanguinity was misleading in this setting. Obtaining a precise diagnosis of OI in underrepresented populations allows expanding our understanding of its molecular landscape and may lead to improved personalized care in the future.