Osteogenesis imperfecta (OI), also called brittle bone disease, Lobstein disease, or Vrolik syndrome, is a rare genetic disorder of the connective tissues characterized by skeletal dysplasia with bone fragility, caused by an abnormality in the metabolism of type I collagen, with a reported incidence of 1 in 15,000 to 20,000 births ^[1][2][3][1,2,3].

The first case of OI was reported by the French priest Melabranche in the 17th century (1674): a 20-year-old male with an intellectual disability and multiple bone fractures. A Swedish surgeon, Olaus Jakob Ekman, was the first to scientifically describe OI in the 18th century (1788), but the diagnosis of osteomalacia was considered for all his patients. In the 19th century, more features of the disease were described. Edmond Axman was the first, in 1831, to highlight the four major characteristics of OI: frail body, fragility of the bones, hypermobility with easy dislocation of the joints, and blue sclerae. Johann Lobstein identified, in 1833, the hereditary nature of the disease, but the term “osteogenesis imperfecta” was first introduced in 1849 by the Dutch professor Willem Vrolik. In the 20th century (1979), Sillence, Senn, and Danks proposed the most widely used classification of the disease, the Sillence classification ^[1][4][1,4].

Most patients (85%) with OI carry an autosomal dominant (AD) mutation in the genes that encode the production of type I collagen (COL1A1 and COL1A2). There are two major mutations in these genes associated with either a structural or quantitative abnormality in the synthesis or processing of type I collagen. Also, autosomal recessive (AR) and X-linked variants of the disease have been described, affecting other genes ^{[2][3][5][6][7]}[2,3,5,6,7].

The first and most widely used classification of OI was published in 1978 by Sillence. Over the years, the nosology of this disease has evolved with the identification of new genes and new mutations. The latest accepted classification was proposed by Botor et al. in 2021 and adapted, as shown in Table 1 ^[4][7][8][4,7,8].

The current treatment of OI varies with the age, severity, and functional status of patients. Therapeutic options include medical management with several pharmacological resources and orthopedic therapy.

The current gold standard in OI treatment is considered the use of bisphosphonates. Bisphosphonates are anti-resorptive drugs that inhibit osteoclasts’ activity and increase bone volume. The goal of bisphosphonate therapy is to counteract the high cellular turnover status, and although new bone is still made of poor-quality collagen, the increase in bone volume may be beneficial despite its impaired quality. It has been proven that they increase BMD, but the effect on fracture reduction is inconclusive. Also, they are most effective in the first year of treatment ^{[3][5][6][7][9][10]}[3,5,6,7,9,10].

The most studied and used bisphosphonate in children is Pamidronate. Glorieux et al. published, in 1998, the first study with Pamidronate in children with OI, using a dose of 1.5–3 mg/kg for 3 consecutive days in 30 children, with repeated administrations at 4–6 months over 5 years (1992–1997). Ralston et al. published a systematic review in 2020, in which they claim to have found over 150 studies regarding the use of Pamidronate in OI that support Glorieux’s results of increasing BMD, improving symptoms, and reducing fracture rates. In theour review of the literature focusing on Pamidronate usage in children with OI, the researchers found different doses and intervals of administration without a clear consensus ^{[11][12][13][14][15]}[11,12,13,14,15].

Surgical treatment is indicated in the case of fractures and for surgical correction of long bone deformities, using telescopic rods that extend during growth [5].

2. Mesenchymal Stem Cells (MSCs)

Transplantation of bone marrow is being investigated as a promising therapy in OI in studies performed on mice and humans, both adults and children ^{[7][10][11][16]}[7,10,11,16].

2.1. Mechanism of Action

Adult stem cells are present in many tissues, but their number decreases with age, and the ones found in the bone marrow have the highest potential in terms of multi-lineage. MSCs are multi-potent stem cells that have the ability of self-renewal and proliferation, and can differentiate into multi-lineage cell types, including osteoblasts. These stem cells have the capacity to migrate into injury sites and secrete cytokines, chemokines, and growth factors useful for the regeneration of the tissue. For these reasons, MSCs become a source of continual renewal of the bone cells and production of healthy collagen and can attenuate genetic disorders of the bone. Immunomodulatory proprieties of the MSCs decrease the risk of immune rejection; therefore, allogenic transplants would not need immunosuppression of the host. Most MSCs used are from bone marrow cells; another source of the stem cells is fetal MSCs, which are less immunogenetic and have an increased proliferative, anti-inflammatory, and homing capacity than their adult counterparts. Guillot et al. published, in 2008, a study that compared human mesenchymal fetal stem cells (hfMSCs) and bone marrow mesenchymal stem cells (BMSCs) and noticed that the fetal cells had a higher level of osteogenic upregulation and produced more calcium, both in vitro and in vivo (murine) ^{[17][18][19][20][21][22][23][24][25][26]}[17,18,19,20,21,22,23,24,25,26].

2.2. Murine Studies

The studies performed on mice with OI are promising, with a significant increase in bone collagen and mineralization, improving the bone structure and reducing fracture incidence ^{[25][27][28][29][30]}[25,27,28,29,30]. Battle and Co. published a 2021 meta-analysis of 10 studies about the MSCs’ efficacy in mouse models with OI and noticed an increase in the mechanical and structural bone proprieties and a decrease in fracture rates. The same meta-analysis reviewed data about cell engraftment in five studies and reported a low benefit [22].

2.3. Human Studies

Bone Marrow MSCs (BMSCs)

The first trial of MCSs, which used BMSC therapy in children with OI, was published in 1999 by Horwitz et al. and reported an increase in BMD and growth velocity and the diminishing of fracture rates in three children with OI type III that received BMCSs from their siblings. Two years later, they published a follow-up study comparing the results for the three treated children with two control patients and reported a decline in the growth rate or a plateau phase in time; however, the BMD continued to increase at a similar rate in healthy children ^[31][32][31,32]. In 2021, Infante et al. published the TERCELOI clinical trial, conducted over a 2.5-year period, including two children with moderate and severe OI forms with repetitive infusions of their siblings’ BMSCs. Both patients showed an increase in BMD (bone volume to the total volume ratio) (BV/TV)), a decrease in fracture rates, and chronic pain. Also, the benefits were observed at a 2-year follow-up after halting the therapy [33].

Fetal MSCs

Götherström et al. published a 2013 study that included two infants with severe OI transplanted with hfMSCs pre- and postnatal, reporting promising results in their BMDs, motility, and fracture rates ^[34][35][34,35]. The trial, Boost Brittle Bones before birth (BOOSTB4), was conducted in Sweden at the Karolinska Institute between 1 January 2016 and 31 December 2022, with administration of hfMSCs pre- and postnatal in children with OI type III and type IV, without published results; however, Lindgren et al. reported no complications with the preliminary results on 17 participants receiving 1 to 4 doses of MSCs. The efficacy was yet to be evaluated ^[36][37][36,37]. Overall, the MSCs in OI look promising. These studies show potential in increasing the BMD and reducing the fracture rates in murine and human studies. Nevertheless, there are some limitations in these studies, like the fact that cells deteriorate under standard conditions, needing to be cryopreserved in special conditions, and the fact that some implications of MSCs in cancers and in worsening bacterial infections have been reported [27].

3. Anti-RANKL Antibody

Denosumab is a human monoclonal antibody with bone anti-resorptive effects, approved for osteoporosis in postmenopausal patients in 2010, as well as for other bone diseases such as giant cell tumors or bone metastases. In children, the pharmacological proprieties are unknown, but they could be used in some diseases, one of them being brittle bone disease ^[38][39][40][38,39,40].

3.1. Mechanism of Action

Denosumab is an IgG2 human monoclonal antibody that reduces bone resorption and increases bone mass. It binds to the receptor activator of NF kappa B ligand (RANKL) and blocks its binding to the receptor (RANK). By binding on RANK, RANKL promotes osteoclast activity and increases bone resorption. This binding is normally blocked by soluble osteoprotegerin (OPG), a member of the tumor necrosis factor (TNF) receptor family. Denosumab mimics the anti-resorptive effect of OPG by blocking RANKL ^[40][41][42][40,41,42].

3.2. Human Studies

In 2012, Semler et al. published the first study of Denosumab use in OI. Four children with OI type VI were included and received Denosumab for 24 months, with an increase in BMD and improvement in their pain. One patient had mild hypocalcemia [43]. In 2014, an article reported two cases of OI children treated with Denosumab with improvements in their metaphyseal density on X-rays [44]. Hoyer-Kuhn et al. published, in 2016, a 48-week prospective trial on 10 children with OI who received four doses of Denosumab. The authors reported four fractures within this period without any beneficial effect on bone pain but with improvements in their BMD and height during this time. The X-ray showed new bone formation. The fracture frequency was not analyzed. In 2019, they published a follow-up study, with only 8 of those 10 patients included, who continued to receive Denosumab. The BMD had a significant reduction during the first follow-up year, but at the end of the follow-up, it was still higher than at the start of the trial. Vertebral shape improved further during the follow-up. Growth was not influenced, and mobility was not significantly improved. The mean calcium levels decreased in all patients, and calcium excretion on urinary spots increased; however, only one patient presented symptomatic hypercalciuria with urolithiasis ^[45][46][45,46]. In 2016, Ward et al. reported a case of a 23-month-old male patient with OI type VI treated with Denosumab. The results were disappointing, with a high fracture rate, no change in mineralization on a bone sample, and a high number of osteoclasts in trabecular bone [47]. In 2017, Uehara et al. published a study that included three female patients with OI type I: two adults and one adolescent who received Denosumab. They all had an improvement in their BMD during treatment, resorption markers, and bone formation, and no fracture occurred in any of them [48]. Trejo et al. published an article in 2018 about four children with OI type VI who received treatment with Denosumab. Their spinal BMD increased, but after an interval between administrations was increased at 6 months, the results were not maintained. All of them developed hypercalciuria in time, and two of them presented hypercalcemia [49]. Another study, published in 2018 by Kobayashi et al., analyzed eight patients with OI type I (six adults and two children) receiving Denosumab. Their BMD generally increased in all patients and the fracture rates and bone turnover markers decreased in most of the patients. They reported no hypocalcemia [50]. In 2019, Maldonado et al. published a case report of a 9-year-old girl with OI type IV treated with Denosumab and with a history of cerebral palsy and epilepsy. She had a decrease in bone resorption, an increase in quality of life, and a decrease in the fracture rate. Hypercalcemia was reported during treatment administration [51]. The researchers only found case reports and two prospective cohort studies involving this therapy, therefore lacking high quality and strength of evidence. Of what the researchers found, publications generally reported an increase in the BMD. In almost all cases, hypocalcemia is reported—this side effect of Denosumab is known. Serum calcium levels must be monitored ^[38][40][52][38,40,52].

4. Sclerostin Inhibition

4.1. Mechanism of Action

Sclerostin is a glycoprotein secreted by osteocytes that acts as a negative regulator of osteoblast differentiation, bone formation, and mineralization, therefore blocking its activity. It may show promise in OI, its efficiency being demonstrated in other bone mineral deficiencies, such as osteoporosis in postmenopausal women. Some studies have also explored the use of sclerostin as a bone turnover marker in some metabolic bone diseases, including OI, with increased levels of sclerostin reported, while others reported normal levels of OI ^{[53][54][55][56][57]}[53,54,55,56,57].

4.2. Animal Studies

A study by Sinder et al., published in 2013, researched the use of sclerostin antibodies (Scl-Ab) in the Brtl/+ mouse model of OI. They reported an anabolic effect on osteoblasts, with an increase in mechanical bone proprieties [58]. Jacobsen et al. published a study in February 2014 using Scl-Ab therapy in an OI-type IV mouse model. They reported an increased bone mass and strength in the treated mice compared with the control group [59]. Another study that used Scl-Ab on adult mice with OI was published by Sinder et al. in May 2014, in which Scl-Ab was administrated for 6 months to Brtl/+ mice, and after therapy, an increase in the bone formation rate, trabecular cortical bone mass, and mechanical proprieties were reported [60]. In September 2014, Roschger et al. proposed a study about Scl-Ab in a model with severe OI, Col1a1Jrt/+ mice, including growing and adult mice. The authors found no significant changes in bone formation or resorption markers but reported a higher trabecular volume and cortical thickness in growing mice and no changes in the adult ones, concluding that this treatment was less effective in the severe OI mouse model [61]. Cardinal et al., in 2019, published a study on their OI type III mice model, in which Scl-Ab reduced long bone fractures and increased the BMD and biomechanical strength of bones ^[62][65]. In 2022, Wang et al. published a study that only used loop 3-specific antibodies of sclerostin (aptscl56) in mice with OI. Because romosozumab, an Scl-Ab used in postmenopausal women for osteoporosis, had a related increased cardiovascular risk, the authors theorized that the protective cardiovascular effect of sclerotin was particular to loop 3 inhibition, while the cardiovascular risk was increased by the inhibition of loop 1 and loop 2. A conjugation of aptck56 with PEG40k raises the stability and the half-life, resulting in PEG40k-aptscl56 (Apc001PE). Apc001PE promoted the formation of the bone without a higher cardiovascular risk ^[63][66].

4.3. Human Studies

Glorieux et al. reported, in 2017, a phase 2a trial of Scl-Ab (BPS804) in adults with moderate OI. The authors concluded that the antibody increased the BMD, reduced resorption, and stimulated bone formation [55]. In 2021, Uehara et al. published a case report of a 64-year-old severe osteoporotic man with OI type I, treated with romosozumab for 12 months with an increase in his BMD, bone formation markers, and a decrease in his resorption markers ^[64][67]. Another case report of romosozumab use in OI was published by Dattagupta et al. in 2023. The subject was a 52-year-old woman with type I OI who underwent alendronate therapy for almost 1 year—8 years before this article was published—without an increase in her BMD. She received romosozumab for 1 year, and an increase in BMD was reported ^[65][68]. The reports are generally favorable, with an increase in the BMD. As for limits, the researchers only found three studies that involve humans, none of these studies focusing on children, and two of them being case reports.

5. Recombinant Human Parathormone

(PTH) (teriparatide) is a potent osteoanabolic agent currently used in adult osteoporosis ^[66][69].

5.1. Mechanism of Action

Teriparatide is a recombinant human PTH that increases the osteoblasts’ survival and their number in three ways: (1) increasing pro-osteoblastogenic factors like fibroblast growth factor 2 (FGF2) and insulin-like growth factor 1 (IGF1), using the upregulation of transcription in these factors; (2) the downregulation of wnt-antagonist sclerostin; and (3) it increases the expression of Runx2 (a transcription factor that is involved in the differentiation of osteoblasts). Teriparatide has the same effect as endogen PTH, increasing serum calcium levels and decreasing serum phosphate ^[66][67][69,70].

5.2. Human Studies

In 2014, Orwoll et al. published a double-blind, placebo-controlled trial on 78 adults with OI over a period of 18 months. Forty of them were randomized in the placebo group and 38 in the teriparatide therapy one. The placebo group included patients with OI type I (27 patients), type III (7 patients) and type IV (5 patients), and the treatment group included patients with OI type I (24 patients), type III (7 patients), and type IV (7 patients). At the end of the study, 65 participants completed the protocol, but only 56 could be analyzed (27 in the placebo group and 29 in the treatment group). The treatment group received 20 μg of teriparatide daily for 18 months. The BMD change was higher in the treatment group, with a decline in the vertebral bone mineral density (vBMD) in the placebo group and a considerable increase in the teriparatide group (–4.7% ± 5.7% vs. 18.3% ± 5.9% change; p < 0.05). These changes were significant for OI type I (4.5% ± 7.3% vs. –5.5% ± 7.3% change; p = 0.008), but the authors did not find a statistically significant change in types III and IV compared with the placebo group. The changes in LsBMD were also significant in the type I treatment group (p < 0.001). There was no difference in the fracture rates reported by the patients. The study conclusions are that the better response occurred in patients with a less severe type. No complications were reported ^[68][71]. In November 2023, Hald et al. submitted a protocol for a randomized trial (ISRCTN15313991) of using teriparatide daily for 2 years, followed by a dose of zoledronic acid in adults with OI, but the results are not yet available ^[69][73]. The researchers found only three studies in the literature that included this therapy usage for OI. Despite good results, teriparatide was effective only for mild forms of OI and was tested only in adults.

6. Anti-Transforming Growth Factor βeta (TGF-β) Antibodies

6.1. Mechanism of Action

TGF-β acts as a coordinator of osteoclasts and osteoblasts, and its excessive signaling is associated with a decrease in bone mass and increased bone fragility in OI; this last statement has been demonstrated by Grafe et al. in OI mouse models, where they found excessive TGF-β signaling. They also used a murine monoclonal antibody, 1D11, as anti TGF-β treatment, resulting in a correction of the bone phenotype ^[70][71][72][74,75,76].

6.2. Human Studies

In 2022, Song et al. published a study about the use of TGB-β in OI. They first collected bone fragments from 10 children with OI type III and 4 non-OI children. The OI bone presented a disorganized Haversian system and increased TGF-β signaling. A phase 1 dose-escalating clinical trial (NCT03064074) included eight adults with OI for 6 months and evaluated the safety of an anti-TGF-β monoclonal human antibody (Fresolimumab), an antibody that neutralizes all three TGF-β homodimers. The patients were divided into two groups of four participants each, who received a different dose of Fresolimumab (a single dose of 1 mg/kg or 4 mg/kg). In the first group (1 mg/kg), one patient with OI type III, two patients with type IV, and one with type VIII OI were included. In the second group (4 mg/kg), one patient with OI type III and three with type IV OI were included. Both groups were followed for 6 months. A decrease in Osteocalcin (Ocn) was observed significantly more in the group with 4 mg/kg (p = 0.00045) without significant changes in CTX and Procollagen type 1 N-terminal propeptide (P1NP) between the two groups. In the low-dose group, the ones with type IV OI had a robust increase in LsBMD; the one with OI type III reported a drop in the BMD, while in the one with OI type VIII, no changes in the BMD were noticed. In high-dose group 2, of the ones with OI type IV, an increase in the BMD was shown, while the one with OI type III had a femur fracture, with difficulty in evaluation. Nine adverse events were reported, possibly related to the medication (two patients in the low-dose group and seven in the high-dose group). The conclusion was that this therapy was beneficial for those with OI type IV and could be a potential treatment ^[73][77]. Only one human phase 1 study is available to date, and it was conducted on the adult population. Beneficial effects on BMD were reported only for type IV OI.

7. Genes Therapy

7.1. Mechanism of Action

As OI is a connective tissue disorder very often caused by dominant mutations in the genes COL1A1 and COL1A2, gene silencing through RNA interference is a promising field. Interfering RNA (iRNA) was designed to target the allele carrying the mutations COL1A1 and COL1A2 in bone cells. But more than 800 different mutations are described, and it is quite impossible to create iRNA against each mutation; however, by developing small interfering RNAs (siRNA) against common polymorphic variations, it would be possible to silence the mutation wherever the mutation is located on the allele ^[74][78].

7.2. Animal Studies

Rousseau et al. published, in 2013, a study in which they used iRNA for successful Col1a1 silencing in an OI mouse model ^[75][79].

7.3. Cell Studies

COL1A1 and COL1A2 mutations represent most cases of OI, but more than 20 other genes are associated with OI. There are various gene-targeted therapies: suppression of harmful transcripts, increased expression of healthy alleles, or gene repair ^[76][80]. In 1996, Wang and Marini published a case of antisense oligodeoxynucleotides used to selectively suppress the mutant type I collagen in fibroblasts from a patient with OI type IV. Suppression of the mutant message was conducted, but the suppression achieved was insufficient for clinical intervention ^[77][81]. In 2004, Millington-Ward published a study in which iRNA was used to downgrade the expression of COL1A1 in mesenchymal progenitor cells (MPCs), with good results but allele specificity ^[78][82]. Chamberlain et al. published, in 2004, an article about the adeno-associated virus vectors (AAV-COLe1INpA gene targeting vector) in MSCs in two patients with OI with an improvement in collagen stability. In 2008, another study by Chamberlain was published about the treatment with MSCs to disrupt COL1A2 with AAV in patients with OI. The cells produced normal type I procollagen ^[79][80][83,84]. In 2008, Lindahl et al. used siRNA to successfully silence COL1A2 in bone cells from OI individuals. In 2013, Lindahl et al. published another study in which they used silencing by iRNAi in COL1A1 and COL1A2 genes in bone cells in patients with OI. The average mRNA levels from both genes were successfully significantly reduced ^[81][82][85,86]. The results look very promising in this kind of therapy, but all human studies are in vitro.

8. 4-Phenylbutiric Acid (4-PBA)

8.1. Mechanism of Action

The altered collagen products in OI are responsible for endoplasmic reticulum (ER) stress. The accumulation of misfolded proteins in the ER activates a specific response—the unfolded protein response (UPR)—that may initiate pro-apoptotic pathways. 4-PBA alleviates ER stress by helping to fold proteins into the ER and maintaining the homeostasis of the ER ^[83][84][87,88].

8.2. Animal Studies

In 2017, Gioia et al. published a study that used a Zebrafish larvae mutant model called “Chihuahua” (Chi/+) that carries a substitution (G574D) in the α1 chain of type I collagen, validated as a model for OI. 4-PBA is a drug that, through a reduction in ER stress, stimulates collagen secretion and ameliorates the OI phenotype (bone mineralization), making it a promising candidate for the treatment of OI ^[85][89]. In 2022, Duran et al. examined the effect of 4-PBA on mouse models Aga2+/−, a model that manifests moderate and severe OI, reporting improvements in growth and bone resistance ^[86][90]. In 2022, Scheiber et al. published an animal study on OI mice treated with 4-PBA, showing a reduction in growth deficiency but without ameliorating bone fragility ^[87][91]. Daponte et al., in a study published in 2023, used two different Zebrafish models, Chihuahua (Chi/+) (dominant model) and p3h1−/− (recessive model), that lacks an enzyme, prolyl 3-hydroxylase (P3h1), proving that 4-PBA enhanced bone formation only in the recessive model ^[88][92].

8.3. Cell Studies

Besio et al. conducted a study in 2018 treating human mutant fibroblasts from a skin biopsy of patients with OI with 4-PBA, showing a decrease in stress and apoptotic markers and an increase in general protein secretion in all treated cells ^[89][93]. In 2019, Takeyari et al. published a study using 4-PBA on human mutant fibroblasts, reporting ameliorated overglycosylation and the capability of calcification and diminishing the production of excessive collagen type I and its accumulation in fibroblasts ^[90][94]. Takeyari, in another study published in 2021 with 4-PBA in dermal fibroblasts from six patients with OI, noticed that the medication improves osteoblast mineralization, reduces ER stress, and normalizes the production of type I collagen ^[84][88]. These studies are reporting a beneficial effect on the production of collagen type I, but there are no in vivo studies on humans with OI.

9. Inhibition of Eukaryotic Translation Initiation Factor 2 (eIF2α) Phosphatase Enzymes (Salubrinal)

9.1. Mechanism of Action

Salubrinal is an agent that is a specific inhibitor of eIF2α phosphatase enzymes. Elevated phosphorylation of this factor stimulates bone formation and reduces bone resorption. eIF2α stimulates bone formation by activating transcription factor 4 (ATF4) and, by phosphorylation of eIF2α, reduces the translation efficiency of most proteins, except for a limited number, one of them being ATF4. Another effect of Salubrinal is the suppression of RANKL activation ^[91][92][93][95,96,97].

9.2. Murine Studies

The only study discovered was an animal study with a duration of 2 months, published by Takigawa in 2016, on 6-week-old Oim mice (þ/; B6C3Fe a/a-Col1a2OIM/J). The study had three groups: the control one with wild-type mice, group 2 (placebo) with Oim mice that received a placebo, and group 3 (treatment) with Oim mice that received daily Salubrinal at 2 mg/kg for 2 months and then followed for another 2 months. Compared with the placebo group, there was no difference in BMD, but they reported improvements in the mechanical proprieties compared with the placebo group. The femur stiffness (N/mm) and elastic module (GPa) were undistinguishable when compared with the control group. No complications were reported ^[94][98].

9.3. Human Studies

The researchers found no studies of Salubrinal and humans with OI. The researchers only found one study of this therapy in OI, and that study was on animals. More research is needed to analyze the effects of Salubrinal in OI.