Genotype-phenotype relationship in a large cohort of osteogenesis imperfecta patients with COL1A1 mutations revealed by a new scoring system

BACKGROUND: Osteogenesis imperfecta (OI), a heritable bone fragility disorder, is mainly caused by mutations in COL1A1 gene encoding α1 chain of type I collagen. This study aimed to investigate the COL1A1 mutation spectrum and quantitatively assess the genotype-phenotype relationship in a large cohort of Chinese patients with OI. METHODS: A total of 161 patients who were diagnosed as OI in Department of Endocrinology of Peking Union Medical College Hospital from January 2010 to December 2017 were included in the study. The COL1A1 mutation spectrum was identified by next generation sequencing and confirmed by Sanger sequencing. A new clinical scoring system was developed to quantitatively assess the clinical severity of OI and the genotype-phenotype relationship was analyzed. The independent sample t-test, analysis of variance, Mann-Whitney U-test, Chi-squared test, Pearson correlation, and multiple linear regression were applied for statistical analyses. RESULTS: Among 161 patients with OI, 32.9% missense mutations, 16.8% non-sense mutations, 24.2% splice-site mutations, 24.8% frameshift mutations, and 1.2% whole-gene deletions were identified, of which 38 variations were novel. These mutations led to 53 patients carrying qualitative defects and 67 patients carrying quantitative defects in type I collagen. Compared to patients with quantitative mutations, patients with qualitative mutations had lower alkaline phosphatase level (296 [132, 346] U/L vs. 218 [136, 284] U/L, P = 0.009) and higher clinical score (12.2 ± 5.3 vs. 7.4 ± 2.4, P < 0.001), denoting more severe phenotypes including shorter stature, lower bone mineral density, higher fracture frequency, more bone deformity, vertebral compressive fractures, limited movement, and dentinogenesis imperfecta (DI). Patients would not present with DI if the glycine substitutions happened before the 79th amino acid in triple helix of α1 chains. CONCLUSIONS: This presented distinctive COL1A1 mutation spectrum in a large cohort of Chinese patients with OI. This new quantitative analysis of genotype-phenotype correlation would be helpful to predict the prognosis of OI and genetic counseling.