Bone fusion materials: past, present, and future

Autologous bone grafts

Autologous bone graft (autograft) is considered the gold standard due to its osteogenic, osteoinductive, and osteoconductive properties [2]. Common donor sites include the distal femur, proximal tibia, fibula, distal radius, and ribs, with the iliac crest bone graft (ICBG) being the most frequently used. Since autografts are harvested from the same patient, the risk of graft rejection or disease transmission is eliminated. However, autografts also have drawbacks. Concerns exist regarding complications associated with harvesting autografts from the iliac crest, such as donor site pain, blood loss, infection, nerve injuries, and iliac fractures. Although some studies have shown that postoperative pain related to harvesting is neither clinically nor statistically significant [11,12], the high incidence of these complications has driven the demand for reducing autograft usage and exploring alternative graft material sources.

Allogeneic bone grafts

Allogeneic bone graft (allograft) can be sourced from either human cadavers or living donors. To ensure sterilization, allografts undergo gamma radiation, a mandatory process that can also damage embedded natural proteins such as growth factors and cytokines, weakening the biomechanical properties of the grafts [13]. Allografts typically function as an osteoconductive scaffold with limited osteoinductive potential and lack osteogenic properties due to the loss of viable cells during the sterilization process.

Despite their limitations, previous studies have suggested that allografts are comparable to autografts in terms of efficacy. A randomized controlled trial (RCT) involving 40 patients undergoing segmental lumbar fusion showed an inferior fusion rate in the allograft group at 6 months, but a comparable fusion rate after 12 months [14]. Suchomel et al. [15] reported similar results in a prospective study of 79 patients undergoing anterior cervical fusion (ACF), demonstrating no significant difference in fusion rate between autograft and allograft groups, although the allograft group exhibited a longer time to fusion. The use of allografts to complement autografts can yield outcomes comparable to autografts [15]. Therefore, allografts are typically employed as extenders for autografts, rather than being used in isolation, particularly in posterior spinal fusion procedures.

Demineralized bone matrix

Demineralized bone matrix (DBM) is an allogeneic bone-derived material, processed to remove its mineral content while preserving the organic matrix. This processing typically involves acid extraction, which retains bone collagen and non-collagenous proteins, including growth factors such as BMPs. These embedded proteins are crucial for promoting bone regeneration and healing. Due to its osteoinductive properties, DBM is widely used in bone fusion procedures [2].

Clinical studies have demonstrated that DBM can be as effective as autografts in some spinal fusion procedures, especially when used in conjunction with other materials. An RCT by Kang et al. [16] reported fusion rates of 92% for autografts and 86% for DBM after a 2-year follow-up. Similarly, Kim et al. [17] found 2-year fusion rates of 62.2% with autografts and 52% with DBM. In posterior lumbar fusion, DBM is frequently used as a graft extender to fill defects, support bone healing, and reduce the need for harvesting large quantities of autograft from the patient’s iliac crest, thereby reducing the risk of pain and complications. However, the performance of DBM is less predictable due to variability in growth factor concentrations [18]. Some studies have also suggested that DBM’s osteoinductive potential can decrease with age, as protein content in bone tissues tends to decline over time [18–20]. While DBM serves as an effective autograft extender in spinal fusion, there is a lack of evidence supporting its effectiveness as a stand-alone osteobiologic material.

Ceramics

Ceramic materials are frequently employed in bone fusion due to their biological inertness, osteoconductive properties, and relatively lower cost. Although they lack mechanical strength, ceramics provide a scaffold for new bone growth and some types exhibit bioactive properties, facilitating integration with natural bone. The disadvantages of ceramics include the lack of osteogenic and osteoinductive potential [21]. Several types of ceramics are utilized in orthopedic and spinal fusion procedures, each possessing unique properties that can facilitate bone healing. These include hydroxyapatite (HA), tricalcium phosphate (TCP), and bioactive glass (BAG).

HA is one of the most commonly used ceramics due to its chemical similarity to the bone mineral component, high osteoconductivity, and bioactive properties. Multiple studies have reported its clinical effectiveness in spinal fusion surgeries. When combined with autograft, HA has demonstrated no significant difference in fusion rates compared to ICBG in both cervical and lumbar spine [22–24]. TCP is another calcium phosphate ceramic that exhibits osteoconductivity comparable to other ceramics. However, it undergoes rapid resorption, typically within 6 weeks. Despite this, previous studies have shown that TCP can serve as a successful bone graft substitute and extender. An RCT by Dai and Jiang [25] involved 62 patients undergoing instrumented posterolateral lumbar fusion (PLF) for degenerative lumbar stenosis. The patients were divided into two groups: 32 patients received β-TCP and 30 patients received ICBG [25]. At the 3-year follow-up, fusion rates were found to be comparable in the two groups. Recent studies have evaluated the effectiveness of TCP in minimally invasive spine surgeries. In patients undergoing lateral lumbar interbody fusion supplemented by instrumented fusion, the combination of TCP with BMA or HA yielded promising results, with fusion rates ranging from 96% to 100% [26,27].

BAG is a type of ceramic that forms a chemical bond with surrounding bone tissues, promoting osteogenesis. It releases ions such as silica, calcium, and phosphorus, which can stimulate new bone formation and facilitate graft integration with the host bone. A recent study reported a fusion rate of 93% at 1-year follow-up after posterior lumbar and cervical fusion surgery for degenerative and traumatic conditions [28]. BAG is a highly versatile and innovative material used in bone fusion procedures, primarily due to its osteoconductive and osteoinductive properties [21]. Composed of silica, calcium, phosphorus, and other oxides, BAG has revolutionized the field of bone repair and regeneration. BAG interacts with biological tissues through a series of surface reactions, leading to the formation of an HA layer that closely resembles the mineral component of natural bone [21,29]. It promotes bone cell activity, blood vessel growth, and exhibits antimicrobial properties due to its ionic release properties [29–31]. Known for its biocompatibility and versatility, BAG is used in spinal fusion, orthopedic repairs, and dental implants. In a recent clinical study, 30 patients who underwent cervical or lumbar fusion surgery with BAG for degenerative and traumatic conditions showed an overall fusion rate of 93% at postoperative 1-year [28]. Furthermore, BAG remains a focus of ongoing innovation, with advances in three-dimensional (3D) printing and composite forms aimed at enhancing its applications [32].

Bone morphogenetic proteins

Recombinant human bone morphogenetic proteins (rhBMPs) are synthetic versions of naturally occurring proteins in the body that play a pivotal role in bone formation and healing. BMPs belong to the transforming growth factor-beta superfamily and are renowned for their ability to induce differentiation of mesenchymal stem cells (MSCs) into osteoblasts, which are responsible for new bone formation [2,33]. Among the various BMPs, BMP-2 and BMP-7 (also known as osteogenic protein-1 or OP-1) are the most extensively used in clinical applications, particularly for spinal fusion and bone grafts. The United States Food and Drug Administration (FDA)-approved the use of rhBMP-2 for anterior lumbar interbody fusion (ALIF) with a titanium cage in 2002. Several RCTs have demonstrated the efficacy of rhBMP-2 in enhancing fusion rates in both cervical and lumbar spine fusion surgeries compared to autografts [34–36]. Consequently, rhBMP-2 has been widely utilized both on-label and off-label in various fusion surgeries to promote fusion. In 2015, BMP was used in 15% of all spinal fusions in the United States [37]. More recently, it has been reported that off-label use of BMP accounted for as high as 85% of all its applications [38–40]. A comprehensive analysis of 40 studies conducted by Parajon et al. [41] revealed that the fusion rate achieved with the use of rhBMP was higher than that obtained without rhBMP (96.6% versus 92.5%). A systematic review further demonstrated that the use of rhBMP-2 significantly enhanced fusion rates in ALIF and PLF procedures, but not in posterior or transforaminal lumbar interbody fusion (TLIF) [42]. However, the authors cautioned that their results might be biased due to variations in dosing and surgical techniques employed in the studies [42].

Despite its potential benefits, the use of rhBMP-2 for fusion has been linked to several complications. Inflammatory edema following rhBMP-2 use in cervical spine procedures can lead to airway obstruction and dysphagia [43,44]. In lumbar spine surgeries, other reported complications include heterotopic ossification, radiculitis, osteolysis, and urogenital events [44]. Another critical concern is that the use of rhBMP-2 could possibly increase the risk of cancer, although a meta-analysis failed to draw conclusive evidence on this issue [45–47].

rhBMP-2 inevitably requires a scaffold to ensure localized delivery, structural support, and sustained release. The most commonly used carrier is the absorbable collagen sponge (ACS), which is FDA-approved for human clinical application. However, ACS necessitates high doses of rhBMP-2 to induce human osteogenesis, which may be associated with adverse complications, including rapid protein degradation and diffusion, leading to soft tissue inflammation and ectopic bone formation, and potential carcinogenic risk [48,49]. To mitigate these risks, the development of spatiotemporal delivery systems is essential to enhance the efficacy and safety of rhBMP-2 delivery [50]. Various carriers composed of different materials, including ceramics (e.g., HA and TCP), natural polymers (e.g., collagen, chitosan, fibrin, and gelatin), and synthetic polymers (e.g., polyglycolic acid, poly-lactic-co-glycolic acid, and polylactic acid-polyethylene glycol), have been investigated for controlled and sustained release of rhBMP-2 [51]. Although calcium phosphates are inherently brittle with very low fracture toughness, they exhibit a high affinity for binding with rhBMP-2 [52]. Recently developed HA/polymer composites have shown promise in enabling sustained release of rhBMP-2 due to the polymer component. Preclinical and clinical studies have demonstrated that these composites exhibit better bone regeneration potential than ACS, even with low doses of rhBMP-2, through sustained release of rhBMP-2 [53–56]. Additionally, these composites may help avoid side effects associated with high doses of rhBMP-2 [53–55].

Synthetic peptides

P-15 is a synthetic 15-amino acid polypeptide that mimics the cell-binding domain of type I collagen. This domain plays a crucial role in promoting cell adhesion and migration, two critical steps in the bone healing process. When incorporated into bone graft substitutes, P-15 enhances the attachment of osteogenic cells (such as osteoblasts) to the graft material, leading to improved bone growth at the fusion site [57]. This binding can also stimulate the proliferation and differentiation of these bone-forming cells, promoting a more robust and rapid fusion process [58]. Furthermore, P-15 has been shown to enhance the production of extracellular matrix, which facilitates mineralization and maturation of the newly formed bone [57].

Due to its osteoinductive properties, P-15 can be an effective alternative to traditional bone grafting materials in spinal fusion surgeries. An RCT by Arnold et al. [59] demonstrated that i-Factor, a bone graft substitute containing P-15, was not inferior to autografts in single-level anterior cervical spine fusion, exhibiting similar clinical and functional outcomes at 6 years. Although high-quality clinical data on lumbar fusion are limited, Mobbs et al. [60] reported promising results, with 97.5%, 81%, and 100% of patients who underwent single-, double-, and triple-level surgery, respectively, showing fusion at all treated levels at a mean follow-up of 24 months (range, 15–43 months).

Cell-based allografts

Cell-based allografts (CBAs) represent an innovative type of bone graft material that incorporates living cells, typically MSCs, seeded on osteoconductive scaffolds. By introducing viable cells capable of differentiating into bone-forming osteoblasts, CBAs can provide osteoinductive properties. This dual function may enhance the biological environment at the fusion site and improve the chances of a successful spinal fusion [61].

One of the primary concerns regarding CBAs is whether viable MSCs exist within these grafts. Neman et al. [62] investigated the presence of viable cells within Osteocel Plus (NuVasive, San Diego, CA, USA) and found that this particular CBA contains viable cells capable of proliferating into osteogenic cells. However, it remains unclear whether MSCs within CBA can survive in the recipient’s tissue, and genuinely promote fusion. Lina et al. [63] examined the survival of bone marrow cells (BMCs) transplanted into immune-intact mice for lumbar fusion. They found that BMCs survived and proliferated over the first 2 weeks following transplantation and were still detectable even after 8 weeks. In contrast, Abedi et al. [64] reported skeptical results regarding CBAs in an in vivo animal study. They compared fusion rates between viable CBAs, devitalized CBAs, and cell-free DBM and found no additional benefits of CBAs in terms of spinal fusion. The authors concluded that the DBM component was likely the key determinant of fusion, as groups with a fiber-based DBM, regardless of viability or CBA product, exhibited higher fusion rates compared to groups with particulate DBM.

Several commercially available CBAs are present in the market, including Osteocel Plus (NuVasive, San Diego, CA, USA), Trinity Evolution (Orthofix, Lewisville, TX, USA), ViviGen (Depuy Synthes, Raynham, MA, USA), PrimaGen Advanced Allograft (Zimmer Biomet, Warsaw, IN, USA), among others [61]. However, there is a dearth of high-quality studies supporting their effectiveness, with only a few retrospective case series having been published. Kerr et al. [65] reported a fusion rate of 92.3% in 52 patients undergoing lumbar interbody fusion with Osteocel Plus. Another study investigated the use of Osteocel Plus in ACF in 182 patients (249 levels) and found an overall fusion rate of 87% and a fusion rate of 92% in a sub-group with single-level fusion by 24 months post-surgery [66].

Multiple studies investigating Trinity have reported similar fusion rates. Vanichkachorn et al. [67] prospectively observed 31 patients undergoing ACF using Trinity Evolution placed with a polyetheretherketone interbody cage and reported a fusion rate of 93.5% at 12 months. A recent study comparing Trinity versus local bone in 39 patients (81 levels) undergoing PLIF demonstrated no significant difference in fusion rates between CBA and local bone (79.0% versus 76.54% of levels, respectively) [68]. A systematic review and meta-analysis from Europe analyzed 10 studies comparing CBA and ICBG in 465 patients and found that fusion rates, clinical outcomes, and complications were not significantly different [69]. The authors concluded that CBAs may serve as an alternative bone substitute option, potentially offering fewer complications and reduced postoperative pain. However, previous studies are limited by heterogeneity in materials and grafting procedures, and most of those studies were industry-sponsored.

Autologous stem cell therapy

Cell therapy in bone regeneration harnesses the regenerative potential of cells, particularly MSCs, to enhance bone healing and fusion success. Unlike traditional grafting methods, cell therapy actively contributes to bone formation by supplying cells that can directly differentiate into bone tissue or secrete osteoinductive factors that stimulate bone growth. Researchers have explored various cell sources for this purpose, including MSCs, embryonic stem cells, amniotic fluid stem cells, and induced pluripotent stem cells [70,71].

MSCs are the most extensively researched cell type for bone regeneration. Sourced from bone marrow (BM), adipose tissue, or umbilical cord blood, MSCs possess the ability to differentiate into osteoblasts, chondrocytes, and adipocytes. Moreover, MSCs promote a favorable environment for bone healing by releasing cytokines and growth factors [72]. Common sources of MSCs include BM, adipose tissue, dental pulp, and umbilical cord blood. Several preclinical and clinical studies have demonstrated the potential of MSC-based therapy for spinal fusion [73–75]. In a recent RCT comparing autologous BM-MSCs combined with allografts to autogenous iliac bone grafts for L4–5 TLIF, the MSC group achieved significantly higher posterior spinal fusion rates at 6 and 12 months post-surgery [76]. However, Buser et al. [77] noted substantial heterogeneity across previous clinical studies, making direct comparison difficult.

Gene therapy

Gene therapy is an emerging area of research in bone fusion, focused on enhancing bone healing by leveraging genetic modifications to promote bone growth. Unlike traditional bone grafts or cell-based therapies, gene therapy involves introducing osteoinductive and osteogenic genes into cells at the fusion site to stimulate osteogenesis and improve the chances of successful fusion. There are two primary approaches to gene therapy: in vivo and ex vivo methods. The in vivo method involves directly injecting vectors carrying osteogenic genes into the fusion site [78]. Cells at the fusion site then take up the genes, producing proteins that stimulate bone growth. Commonly used vectors for in vivo gene therapy include adenovirus, retrovirus, lentivirus, nanoparticles, and plasmids. The ex vivo approach involves harvesting cells from the patient, genetically modifying them in a laboratory to overexpress osteogenic proteins, and re-implanting them at the fusion site [78]. These genetically modified cells enhance fusion by releasing growth factors that promote bone healing. In both methods, a commonly studied target gene for bone fusion is BMP-2, which plays a pivotal role in bone regeneration [78,79]. Despite its promise, gene therapy faces various challenges, including optimizing viral/non-viral vectors, delivery methods, scaffolds, and stem cell sources. Nonetheless, preclinical studies have shown encouraging results of gene therapy for bone regeneration [80–82].

Novel biomaterials

Peptide amphiphiles (PAs) are engineered molecules that self-assemble into nanoscale structures due to the unique combination of a hydrophobic (water-repelling) tail and a hydrophilic (water-attracting) peptide segment. These amphiphilic properties enable PAs to form various nanostructures, such as micelles, nanofibers, and vesicles, making them valuable in regenerative medicine and drug delivery applications [83]. In bone tissue engineering, PAs can be designed to incorporate osteogenic peptides that mimic BMPs, promoting osteoblast differentiation and mineralization [84]. The nanofibrous scaffolds created by these PAs provide a supportive matrix for new bone tissue formation, offering potential applications in spinal fusion and other bone repair surgeries [85]. Furthermore, PAs can serve as carriers for growth factors including BMPs, releasing them gradually at the fusion site to enhance osteogenesis and reduce the required dose for successful fusion. Lee et al. [84] demonstrated that PAs could deliver BMP-2 locally in a rodent PLF model, achieving excellent fusion rates with a BMP-2 dose 10-fold lower than the therapeutic dose.

3D printing, also known as additive manufacturing, is a transformative technology in bone regeneration. It entails the creation of 3D objects by adding materials layer by layer according to digital models, enabling the production of customized bone scaffolds that mimic the structure and function of native bone [86]. This approach, also known as “bioprinting,” combines the precision of 3D printing with bioactive materials to support cell attachment, proliferation, and differentiation, ultimately facilitating new bone growth. However, clinical applications of 3D bioprinting in bone tissues still face significant limitations such as lack of neurovascular structures and slow printing process for large or complex tissues [87]. Moreover, 3D printing can only yield static objects, while bone tissue is dynamic in the host. To overcome this limitation, a new concept called four-dimensional (4D) bioprinting has been introduced. This new generation of tissue engineering that integrates the concept of time into 3D bioprinting, was introduced in 2014 [88]. 4D bioprinting enables the biomaterials to change their properties over time in response to environmental conditions, mimicking natural tissue growth and adaptation [88]. This technology is poised to transform the field of bone tissue engineering, providing an adaptable approach that aligns with natural bone healing processes. An overview of all bone graft materials is provided in Table 1.

Table 1

Currently available bone graft options