Heterozygous variants in COL1A1 and COL1A2 account for approximately 80–85% of cases and underpin the classical Sillence types I through IV. COL1A1 encodes the pro-alpha1(I) chain and COL1A2 the pro-alpha2(I) chain, which assemble into triple-helical type I procollagen before being processed into collagen fibrils. Pathogenic variants are divided broadly into quantitative defects, or haploinsufficiency, and qualitative, dominant-negative defects. Nonsense or frameshift variants commonly trigger nonsense-mediated decay and reduce synthesis of structurally normal collagen. Missense variants, especially substitutions of critical glycine residues within the Gly-X-Y repeats, can disrupt triple-helix folding. Misfolded molecules then undergo prolonged enzymatic over-modification and accumulate in the endoplasmic reticulum, activating the unfolded-protein response, disturbing osteoblast function and contributing to abnormal matrix fibrils. Haploinsufficiency generally produces the mild type I phenotype, whereas dominant-negative defects more often accompany the more severe type II, type III and type IV forms.

The expanded classification shows why patients with similar fractures can require different management. Type V OI results from a recurrent c.-14C>T variant in the untranslated region of IFITM5. The variant creates an alternative translation start and adds five amino acids to the bone-restricted interferon-induced transmembrane protein-like protein. A 2025 single-cell RNA-sequencing study in mouse models identified “chondrocyte-like osteoblasts” expressing both osteogenic and chondrogenic markers, supporting a mechanism in which abnormal BRIL redirects skeletal progenitors toward ectopic cartilage and away from normal bone formation. Clinically, type V is marked by hypertrophic callus, forearm interosseous-membrane calcification and radial-head dislocation, although the first feature appears in only about 65% of cases. People carrying the same recurrent variant can range from near-normal stature and ambulation to severe growth impairment and lower bone density, underscoring the contribution of modifiers, biological sex and other factors not captured by the genotype alone.

Type VI OI has a different molecular basis, arising from biallelic loss-of-function variants in SERPINF1, which encodes pigment epithelium-derived factor, or PEDF. Most affected people have undetectable serum PEDF, making its absence a specific and sensitive biomarker for this subtype. The disorder causes a profound mineralization defect with abundant unmineralized osteoid and a characteristic “fish-scale” pattern in lamellar bone under polarized light microscopy. Fractures commonly begin after the first year of life and are followed by severe deformity, short stature and loss of independent walking. A retrospective analysis of non-COL1 OI found consistent fracture reductions during bisphosphonate treatment in type V, whereas responses in type VI were heterogeneous; some patients experienced more fractures despite improved bone mineral density. This discordance separates the quantity of mineral from the quality of the bone matrix and illustrates why a shared diagnostic label does not imply a shared drug response.

Other subtypes arise from failures in specific stages of collagen handling. Biallelic variants in CRTAP, P3H1 or PPIB destabilize the complex that performs 3-hydroxylation of procollagen. The resulting types VII, VIII and IX are severe or lethal and commonly feature fractures around birth, extreme osteoporosis, rhizomelic limb shortening and popcorn-like metaphyses, although the spectrum ranges from lethal to moderate. SERPINH1 and FKBP10 variants interfere with the chaperones HSP47 and FKBP65, affecting collagen folding, transport and cross-linking. Type X can include dentinogenesis imperfecta, chronic lung disease and renal abnormalities, whereas type XI characteristically includes congenital joint contractures and overlaps clinically with Bruck syndrome type I. BMP1 variants impair cleavage of the collagen C-propeptide and can produce severe fragility with normal or elevated bone mineral density. KDELR2 variants prevent efficient retrieval of HSP47 from the Golgi apparatus, disrupting fibril formation; type XXI ranges from prenatal fractures and progressive deformity to neurodevelopmental delay. These are distinct molecular diseases, not interchangeable presentations of one collagen defect.

Defects in osteoblast differentiation and signaling create another set of recognizable phenotypes. SP7 encodes Osterix, a zinc-finger transcription factor acting downstream of RUNX2 and required for osteoblast differentiation. Biallelic loss-of-function variants generally cause severe bone fragility, growth deficiency and recurrent fractures, while a de novo neomorphic missense variant produces a high-turnover disorder combining fragility, defective mineralization, cranial hyperostosis and craniosynostosis. WNT1-related type XV is caused by biallelic variants affecting canonical Wnt/beta-catenin signaling. Its manifestations range from moderate disease to progressive deformity, reduced calvarial mineralization, short stature, ptosis and brainstem-cerebellar hypoplasia. MESD defects impair the folding and trafficking of the Wnt co-receptors LRP5 and LRP6, producing type XX, which can include prenatal fractures, severe osteopenia, pulmonary hypoplasia and early respiratory failure. Other emerging types involve CREB3L1-mediated collagen transport, MBTPS2-dependent intracellular proteolysis, SPARC-associated extracellular-matrix regulation, TENT5A-dependent mRNA stability, CCDC134-linked RAS/MAPK signaling and PHLDB1-dependent AKT signaling.

Genetic patterns also vary across Asian populations, complicating any uniform interpretation of diagnostic findings. In Chinese cohorts, COL1A1 and COL1A2 variants account for approximately 76–90% of molecular diagnoses, with type I the most common clinical subtype. Among 74 Chinese families with recessive OI, WNT1 accounted for 40.54%, SERPINF1 for 29.7% and FKBP10 for 13.5%, a distribution different from that often reported in Western cohorts. In South Korea, collagen-gene variants account for about 87.4% of confirmed cases. Glycine substitutions are associated with a mean height Z-score of −5.94 rather than −1.66 and first fractures at 11.12 rather than 23.74 months; 62.5% of affected patients older than 14 had more than 10 lifetime fractures. In 91 unrelated Vietnamese patients, COL1A1 and COL1A2 variants accounted for 59.4%, while a P3H1 splice-site variant occurred in 85.7% of recessive families and 10.3% of the total sample. An Indian referral cohort attributed only 47.6% of cases to heterozygous collagen-gene variants and reported recessive OI in as many as 48% of cases, with recurrent WNT1 and FKBP10 variants suggesting founder effects in some subpopulations.

Diagnosis is becoming more technically refined as the genetic range expands. Gene panels or whole-exome sequencing are recommended first-line tools, but broad sequencing does not automatically detect every relevant variant. Dedicated copy-number and structural-variant analysis is needed to identify exon-level deletions and duplications that standard pipelines may miss. Karyotyping can detect balanced rearrangements, while long-read or whole-genome sequencing can resolve complex breakpoints. Sanger sequencing remains appropriate for testing a known familial variant or the characteristic IFITM5 mutation in suspected type V OI. Variant interpretation follows ACMG/AMP and ClinGen guidance, while RNA analysis, minigene reporters and collagen electrophoresis can supply functional evidence for uncertain findings. This layered approach combines sequencing, structural analysis and biochemical testing rather than relying on sequence data alone.

Bisphosphonates remain the best-established medical treatment for children with OI, although their effects vary by formulation and subtype. Intravenous pamidronate and neridronate generally increase bone mineral density and reduce fractures. Evidence for oral bisphosphonates is weaker: a large randomized trial of alendronate increased lumbar-spine mineral density but did not produce a statistically significant reduction in long-bone fractures compared with placebo. Common adverse effects include transient fever and myalgia after the first intravenous infusion and gastrointestinal discomfort with oral agents. Newer agents act through different pathways. Denosumab blocks RANKL and inhibits osteoclast activity. Small pediatric studies have reported density and growth gains, but withdrawal can produce rapid bone turnover and hypercalcemia; in one pediatric comparison, 31% of treated children developed rebound hypercalcemia, with nearly half of those cases progressing to hypercalcemic crisis. Romosozumab and setrusumab inhibit sclerostin to promote bone formation, but evidence in OI remains limited to adult case reports and early trials. Fresolimumab blocks transforming growth factor-beta: a phase I adult trial found dose-dependent effects on bone turnover and variable changes in lumbar-spine density, with no treatment-related serious adverse events. These results remain investigational in OI and do not establish efficacy for children.

Experimental therapies aim directly at the causal cells or genes. Mesenchymal stromal cells from bone marrow, umbilical cord blood or fetal liver can differentiate toward osteoblasts and influence their microenvironment through paracrine signals. Clinical studies of repeated HLA-haploidentical mesenchymal stromal-cell infusions have reported reduced fractures and improved quality of life, while a prenatal fetal-liver-cell transplant produced 7.4% engraftment, normalized bone morphology and a postnatal period with minimal fractures. The effect waned as donor cells were depleted, exposing the unresolved problem of durable engraftment. CRISPR-Cas9 editing has corrected COL1A1 and COL1A2 variants in patient-derived induced pluripotent stem cells, with improved collagen expression, alkaline phosphatase activity and mineralization in differentiated cells. In an OI mouse model, systemic delivery of Cas9 and a corrective donor sequence achieved 12% homology-directed repair in bone and was associated with improved bone mass, skeletal deformity, fracture healing and grip strength. These are laboratory or animal findings, and editing efficiency, off-target effects, delivery and durability remain major barriers to clinical use.

Precision management also depends on surgical care and multidisciplinary follow-up. Telescopic intramedullary rods, including the Fassier-Duval system, have improved ambulation and reduced revision surgery compared with non-telescopic rods. A reported 4-year rod-survival rate of 88% for telescopic rods, compared with 40% for non-telescopic rods, came with complications that can include migration, bending, failure to telescope and problems at distal fixation. The evidence therefore supports a combined approach involving surgery, antiresorptive or investigational therapy, rehabilitation and developmental support, but outcomes remain dependent on severity, anatomy and the timing of intervention. As genetic classification becomes more informative, its greatest immediate value lies in identifying distinct mechanisms, guiding appropriate surveillance and testing whether a particular molecular subtype predicts a different response to therapy. The emerging paradigm links each skeleton’s failure to a measurable molecular defect while retaining the broader care required for fractures, deformity, respiratory complications, hearing loss and dentinogenesis imperfecta.

Subject of Research: Genetic mechanisms, clinical heterogeneity and precision management of osteogenesis imperfecta

Article Title: Pathogenesis and precision management of osteogenesis imperfecta: from genetic mechanisms to novel therapies

Article References: Wang, X.-Q., Zhang, C., Li, X., Sun, Y., Chen, Q., Yang, B.-Y., Li, Y.-S., Jiang, X.-Y., Zhao, Z.-H., Wang, L., Sun, X.-H., Wang, H.-Y., Lu, W.-L., Wei, H.-Y., Su, Z., Ren, X.-Z., Cao, B.-Y., Wu, W., & Chen, T. (2026). Pathogenesis and precision management of osteogenesis imperfecta: from genetic mechanisms to novel therapies. World Journal of Pediatrics, 22(9), 982-1011. https://doi.org/10.1007/s12519-026-01086-7

Image Credits: AI Generated

DOI: 10.1007/s12519-026-01086-7

Keywords: osteogenesis imperfecta, bone fragility, genetic diagnosis, precision medicine, collagen biology, bisphosphonates, gene editing, mesenchymal stromal cells, Asian populations, skeletal dysplasia, Pathogenesis, precision