A case report of autogenous bone grafting using fibula regenerated from beta-tricalcium phosphate for a bone tumor: application of the in vivo bioreactor principle

Abstract

Background

Reconstruction of large bone defects after tumor resection remains a major challenge in orthopedic surgery. Although autologous bone grafts have excellent osteogenic potential, their application is limited by the small volume available for harvest. Conversely, artificial bone substitutes, such as β-tricalcium phosphate (β-TCP), can be produced in large quantities but possess limited osteoinductive capacity. We present a case in which hybrid grafting using irradiated bone and β-TCP–derived regenerated fibula was performed to reconstruct a large segmental defect following resection of a recurrent bone tumor, achieving successful bone union.

Case presentation

A 10-year-old boy with fibrous dysplasia of the proximal femur initially underwent curettage, fibular strut grafting, and β-TCP implantation, including cylindrical β-TCP placement at the fibular donor site. At age 28, local recurrence was suspected, and the possibility of an intermediate malignant bone tumor could not be excluded. Therefore, a wide resection was selected after careful clinical consideration. The resulting segmental defect was reconstructed using an intercalary extracorporeally irradiated autograft, augmented with the regenerated fibula derived from the previous β-TCP implantation. At 1 year postoperatively, radiographic findings demonstrated bone union not only at the femoral osteotomy sites but also between the regenerated fibula and irradiated bone. Objective imaging analyses were performed to evaluate bone regeneration around the reconstructed segment using postoperative radiographs and CT as part of postoperative follow-up. Grayscale values at the osteotomy gaps reached levels comparable to cortical bone on radiographs. Computed tomography further revealed a significant increase in Hounsfield unit values at the bridging region between the regenerated fibula and irradiated bone.

Conclusion

Hybrid grafting using a β-TCP–derived regenerated fibula and irradiated bone successfully reconstructed a large femoral defect, highlighting the biological potential of regenerated bone. This reconstruction was based on the in vivo bioreactor concept, and objective imaging analysis proved useful for evaluating bone regeneration adjacent to the recycled bone.

Background

Bone grafting, commonly used in orthopedic surgery to reconstruct bone defects after tumor resection [1], is performed using various materials, including allografts, recycled bone, and synthetic bone substitutes in recent years [2–5]. Autogenous bone grafts have been the standard treatment for bone loss owing to their biocompatibility and bone-healing properties [1, 6]. The fibula, frequently used as a donor site for autologous bone grafting, can be harvested as either a free or vascularized graft and may be used alone or in combination with recycled bone grafts or synthetic substitutes [7, 8].

Recent reviews have highlighted that, beyond traditional grafting techniques, current reconstructive strategies for critical-sized bone defects increasingly include synthetic bone substitutes, composite scaffolds, and tissue-engineered constructs designed to enhance biological integration and bone regeneration [9–11]. In oncologic reconstruction, extracorporeal irradiation and replantation of the resected bone segment has also been established as a limb-salvage technique, allowing anatomical reconstruction using the patient’s own bone while achieving local tumor control [4, 12].

Synthetic bone substitutes are commonly used to treat bone defects located in areas with sufficient osteoconductive potential [13]. Beta-tricalcium phosphate (β-TCP), a biocompatible artificial bone substitute, is widely recognized for its ability to promote bone regeneration, as demonstrated by imaging and histological findings [2, 14, 15]. In contrast, artificial bone substitutes exhibit poor osteoinductivity [16]. In regions where the surrounding normal bone is limited, successful regeneration depends largely on osteoinductive mechanisms; thus, sufficient bone regeneration is often challenging to achieve. Therefore, multimodal bone grafting strategies are required to treat large bone defects.

In recent years, the concept of the in vivo bioreactor has emerged as a promising strategy in regenerative medicine, aiming to harness the host biological environment to generate vascularized and functional tissue constructs [17]. In this report, we present a case in which hybrid grafting using irradiated bone and β-TCP-derived regenerated fibula was performed to reconstruct a large bone defect following resection of a recurrent bone tumor, resulting in successful bone union. Although not originally designed based on this theory, the approach may be interpreted as consistent with the concept of the in vivo bioreactor. Quantitative evaluation of bone formation around the irradiated bone, including the bridging region to the regenerated fibula, was conducted.

Case presentation

A schematic timeline summarizing the clinical course is presented in Fig. 1. A 10-year-old boy was referred to our hospital for evaluation of a bone tumor in the left proximal femur. Figure 2 illustrates the clinical course and treatment progress after the initial surgery. Based on imaging studies and clinical findings, fibrous dysplasia was suspected. Given the potential risk of microscopic tumor persistence and subsequent progressive deformity typical of fibrous dysplasia, synthetic bone grafting alone was considered inadequate. Therefore, a fibular strut graft was selected to ensure sufficient mechanical strength for reconstruction of the defect. Accordingly, curettage with fibular strut grafting and β-TCP granule implantation was planned. There were no absolute or relative contraindications to a free vascularized fibular graft. However, at the time of the initial surgery, the lesion was considered benign, and preservation of the cortical bone was deemed feasible. Therefore, free vascularized fibular grafting was not considered as a primary reconstructive option. The lesion was located in the diaphysis of the femur without involvement of the physis. Therefore, the potential impact of surgery on longitudinal growth was considered to be minimal. A cortical window measuring approximately 1 cm in the transverse direction and 10 cm in the longitudinal direction was created on the lateral aspect of the femur. Intralesional curettage of the tumor was performed through this window. An 11-cm segment of fibula was harvested from the left lower leg and transplanted into the resulting bone defect. The surrounding residual cavity was subsequently filled with β-TCP granules. Cylindrical β-TCP blocks were implanted at the fibular donor site. Postoperatively, neither internal nor external fixation was used. The affected limb was maintained under strict non–weight-bearing conditions until bone union was confirmed radiographically. The postoperative pathological diagnosis was fibrous dysplasia. At 3 months after surgery, the transplanted fibular segment at the recipient site was maintained without radiographic signs of resorption. Imaging at 1.5 years postoperatively demonstrated that the cylindrical β-TCP blocks at the fibular donor site had been largely replaced by autologous bone, although regeneration in the distal portion remained incomplete. Radiological evaluation at 5 years postoperatively showed complete autologous bone replacement at the femoral defect and near-complete union of the distal fibula. Follow-up was completed 13 years postoperatively.

Fig. 1.

Schematic timeline of the clinical course

Fig. 2.

Clinical course after the initial surgery. Initial radiograph showing a bulging proximal femoral diaphyseal lesion with a radiolucent area A Curettage and reconstruction with a fibular strut and β-TCP granules B and a cylindrical β-TCP block at the donor site C The fibula remained intact (D) and the β-TCP at the donor site was gradually resorbed (E) at 3 months. At 1.5 years postoperatively, the transplanted fibula achieved union with the femur F and the β-TCP at the donor site was replaced by autologous bone except at the distal end (arrow) G At 5 years, the reconstructed femoral segment was fully remodeled into autologous bone H and the fibula, including the distal portion (arrow), had achieved nearly complete union I β-TCP, β-tricalcium phosphate

Eighteen years after the initial surgery, at age 28, the patient was referred again because of suspected recurrence. Radiographs showed that the transplanted fibula and artificial bone had almost completely disappeared (Fig. 2A). Recurrence was strongly suspected owing to the irregular osteolytic appearance on radiographs and the presence of a mass lesion on magnetic resonance imaging (Fig. 2C, D, and E). The grafted β-TCP blocks had been entirely replaced by autologous bone radiologically, consistent with full fibular regeneration (Fig. 2B). An open biopsy was performed, and histology again suggested fibrous dysplasia; however, reverse transcription polymerase chain reaction did not detect a GNAS gene mutation (data not shown). Despite benign pathological diagnoses in both the primary and recurrent lesions, gradual but definite local recurrence, and the lack of a GNAS mutation, raised concerns of possible low-grade malignancy. After a thorough discussion with the patient regarding the clinical course and potential risk of malignant behavior, the lesion was managed surgically with oncological margins due to concern for malignant potential, and a wide resection was proposed. The patient fully understood the rationale and consented to the surgical plan; wide resection was subsequently performed.

Reconstruction of the massive bone defect following wide resection was a major consideration in planning the surgery. In musculoskeletal oncology, reconstruction of large bone defects generally involves either prosthetic replacement or biological reconstruction using treated autografts. Given the patient’s young age, biological reconstruction using the patient’s own bone was considered preferable. In addition, considering the advantage of anatomical fitting, reconstruction centered on extracorporeally irradiated autograft was selected. After extensive discussion, an intercalary resection of the femoral shaft together with the tumor was performed, followed by extracorporeal irradiation of the resected bone segment and reimplantation to its original site. A regenerated fibula was used as an augmentation material to enhance mechanical stability and biological incorporation. Figure 3 shows the preoperative reconstruction plan. Preoperative planning was conducted using AquariusNET Viewer (TeraRecon Inc., Foster City, CA, USA) derived from plain radiographs, computed tomography (CT), and magnetic resonance imaging (MRI). A 2-cm oncological margin was set from the tumor boundary. Given that the tumor measured 17 cm in longitudinal extent, resection of a 21-cm segment of the femoral diaphysis was planned. To sufficiently exceed the resection length and allow stable reconstruction, a 25-cm fibular graft was planned for harvest. The tumor was excised en bloc with a wide margin, including the surrounding soft tissue (Fig. 4A). Following removal, all soft tissues were carefully stripped away, and the regenerated fibula, along with both native fibular ends, was isolated subperiosteally (Fig. 4C). The regenerated fibula appeared well-remodeled macroscopically, and its structural strength seemed sufficient. Extracorporeal irradiation of the resected bone was performed at 80 Gy (Fig. 4B), followed by intramedullary nail fixation. The harvested fibular segment was augmented as an onlay graft to reinforce the irradiated bone. (Fig. 4D). Bone union was evaluated using serial radiographs and CT examinations performed at 1 and 2 years postoperatively. Radiographic evaluation (Fig. 5) revealed that the proximal osteotomy gap became indistinct soon after surgery, indicating early union. The distal gap remained clearly visible at 6 months; however, radiographic union was observed at 1 year, and the gap had resolved completely by 1.5 years postoperatively. Figure 6A illustrates the 3D-reconstructed CT image obtained 1 year postoperatively. Bridging bone formation was observed between the augmented fibula and both proximal and distal femurs (Fig. 6B and D). In addition, a newly formed bridging bone was identified between the regenerated fibula and irradiated bone (Fig. 6C). These findings were maintained at 2 years postoperatively. At the 3-year follow-up after revision surgery, the patient showed no evidence of local recurrence or metastasis. He had returned to his previous work and achieved a Musculoskeletal Tumor Society score of 27.

Fig. 3.

Radiographic findings at recurrence. Radiographs showing that the transplanted fibula and β-TCP had almost completely disappeared A β-TCP at the donor site had been completely replaced by autologous fibula B MRI demonstrates a mass in the proximal femur on T2-fat-suppressed coronal imaging C isointense to muscle on T1-weighted axial imaging D and heterogeneously hyperintense on T2-fat-suppressed axial imaging E β-TCP, β-tricalcium phosphate; MRI, magnetic resonance imaging.

Fig. 4.

Preoperative reconstruction planning. Intercalary resection (dotted line) was planned, followed by extracorporeal irradiation of the resected bone (①). The native fibula (②) and regenerated fibula derived from the β-TCP graft (③) were harvested simultaneously. Subsequently, the irradiated bone and regenerated fibula were transplanted to reconstruct the segmental defects. β-TCP, β-tricalcium phosphate

Fig. 5.

Intraoperative findings. The tumor was excised en bloc with a wide margin, including the surrounding soft tissues (A). The resected bone was treated with extracorporeal irradiation at 80 Gy (B). The regenerated fibula is isolated subperiosteally together with both native fibular ends (C). Intramedullary nail fixation was performed, and the harvested fibula was applied to reinforce the irradiated bone (D)

Fig. 6.

Postoperative radiographs demonstrated the chronological progression of bone healing. At 1 week postoperatively, distinct osteotomy gaps were visible at both the proximal and distal sites (A). At 6 months, the proximal gap had become indistinct, although the distal gap remained clearly identifiable (B). At 1 year, progressive bridging bone formation was observed at the distal osteotomy site (C). At 1.5 years, the gaps at both osteotomy sites had completely disappeared, indicating radiographic union (D)

As part of postoperative follow-up in this case, objective imaging analyses were performed to evaluate bone regeneration around the reconstructed segment using postoperative radiographs and CT. In total, 12 postoperative radiographs and two CT scans obtained at 1 and 2 years after surgery were reviewed. Since radiographs were performed more frequently than CT scans, objective evaluation of new bone formation at the osteotomy gaps was conducted primarily using serial digital radiographs. Based on a report by Ebrakhim et al., who demonstrated that grayscale values on digital radiographs are proportional to bone density, the grayscale value was adopted as an indicator of bone formation within the osteotomy gap in this study, with some methodological modifications [18]. Grayscale values were measured from digital radiographs using ImageJ software (National Institutes of Health, Bethesda, MD) (Fig. 7A). For each radiograph, four regions of interest (ROIs) were manually defined: (1) Calcar cortical bone ROI (reference for full-intensity cortical bone) (2) External marker ROI (reference for zero intensity) (3) Osteotomy gap ROI at the proximal margin (4) Osteotomy gap ROIs at the distal margin. To maintain measurement consistency, the four ROIs determined on the initial postoperative radiograph were transferred to the corresponding anatomical positions on all serial radiographs, and the grayscale values of each ROI were obtained. As pixel values in digital radiographs do not carry absolute physical units, the measurements were normalized using the external and cortical reference ROIs. The external marker and cortical bone values were defined as 0% and 100%, respectively. The relative grey scale intensity of the osteotomy gap was calculated using the following formula:

Fig. 7.

CT findings at 1 year postoperatively. Three-dimensional reconstructed CT at 1 year postoperatively (A). Bridging bone formation between the augmented fibula and both the proximal and distal femurs (B, D), with additional bridging (white arrow) identified between the regenerated fibula and irradiated bone (C). CT, computed tomography

This formula was applied separately to the proximal and distal gap regions to quantify their progression toward cortical bone density over time. A value of 0% indicates an intensity equivalent to that of the external marker, whereas 100% represents complete approximation to cortical bone. Serial values were plotted over time to assess the temporal patterns of bone healing at each osteotomy gap. Figure 7B illustrates the time-dependent changes in grayscale percentage at the osteotomy gaps. At the ROI of the proximal osteotomy gap, relative intensity values were 95.5% and 98.1% at 1 and 2 weeks postoperatively, initially lower than those of the cortical bone, but subsequently exceeding the cortical (calcar) reference and remaining higher thereafter. The ROI of the distal osteotomy gap demonstrated a different pattern, maintaining approximately 75% until 6 months, followed by a steady increase. After 1 year, the values reached and maintained levels comparable to those of the cortical bone.

Since bone formation between the regenerated fibula and irradiated bone could not be adequately assessed using only frontal and lateral radiographs, CT imaging was used for evaluation. Based on previous reports, including a study by Tanaka et al. [19], which demonstrated a proportional relationship between Hounsfield unit (HU) values and bone formation, HU values were assumed to reflect new bone formation in our analysis. Objective analysis of CT imaging was performed using Horos (version 3.3.6; www.horosproject.org). (Fig. 8A) Two postoperative CT datasets obtained 1 and 2 years after surgery were displayed side-by-side. The images were manually synchronized by matching the slice position of the femur with the reference coronal section. Regions showing bridging bone formation between the regenerated fibula and irradiated bone were identified, and ROIs were manually drawn to encompass the newly formed bone. The surface area (mm²) and mean HU values within each ROI were calculated. CT slices in which screw artifacts interfered with accurate ROI placement were excluded, and eight slices were analyzed using the same procedure. Statistical comparison of the surface area and HU values between the 1- and 2-year CT images was performed statistically. Statistical analysis was performed with EZR (Jichi Medical University, Tochigi, Japan), which was a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria) [20]. More precisely, it was a modified version of R commander designed to add statistical functions frequently used in biostatistics. Figure 8B shows the objective evaluation of the bridging region between the regenerated fibula and irradiated bone. The bridging area increased from 10.491 mm² (range, 1.65–36.4) at 1 year postoperatively to 13.012 mm² (range, 3.72–36.1) at 2 years, although the difference was not statistically significant. In contrast, the HU value increased significantly from 703 (range, 578–812) at 1 year to 827 (range, 685–880) at 2 years (Fig. 9).

Fig. 8.

Objective radiographic evaluation of bone regeneration at the osteotomy sites. A ROIs were defined on the postoperative week-1 radiograph at four sites: normal cortical bone of the calcar, external marker, proximal osteotomy gap, and distal osteotomy gap. The same ROIs were applied to subsequent radiographs to extract gray-scale values B Temporal changes in grayscale percentages at the osteotomy gaps. At the proximal site, values remained below the cortical bone level until 2 weeks postoperatively, after which they exceeded the cortical (Calcar) reference and continued to increase over time. The distal gap remained at approximately 75% until 6 months, then gradually increased, reaching levels comparable to those of the cortical bone after 1 year. ROI: region of interest

Fig. 9.

Objective CT-based evaluation of bone regeneration between the irradiated bone and the regenerated fibula A CT scans obtained at 1 and 2 years postoperatively were synchronized to identical axial levels, and ROIs were placed across multiple slices at the bridging site to measure the surface area and mean HU values B The bridging area showed an increasing trend from 1 to 2 years postoperatively, whereas the HU values increased significantly. CT, computed tomography; HU, Hounsfield unit; ROI: region of interest

Discussion

Autologous bone grafts are limited by the need to harvest bone from the patient and the restricted volume that can be obtained, making the reconstruction of large defects challenging. These limitations have driven the development of synthetic bone substitutes that can be reliably supplied and customized in both shape and quantity. However, the osteoinductive potential of artificial bone remains modest [16]. Therefore, the concept of the in vivo bioreactor was proposed [17]. This concept uses the living body as a bioreactor and promotes tissue regeneration by harnessing endogenous nutrition, blood supply, and stem cell recruitment. The application of this principle has been explored in the fields of bone and cartilage regeneration [21–23]. A scaffold is implanted within the host body to induce cellular infiltration and growth factor activity, generating a tissue that closely resembles the autologous bone or cartilage, which is subsequently transplanted into the target defect site. In the present case, β-TCP placed at the fibular donor site was maintained in the host environment, where it matured into biologically viable regenerated bone. This regenerated bone was later harvested and reused as a graft for reconstruction. Although not originally designed for this purpose, the current surgical procedure is based on the in vivo bioreactor principle. Although clinical applications of this approach have been reported in the dental and maxillofacial fields [24–26], to the best of our knowledge, no prior studies have applied the in vivo bioreactor concept to limb salvage surgery in the musculoskeletal system. During preoperative planning, the use of an intact contralateral fibula as a graft was a reasonable option. Meanwhile, our previous clinical studies demonstrated that β-TCP implantation can achieve adequate bone regeneration, as verified by both radiological and histological evaluations [2, 14, 27]. Notably, one study demonstrated successful replacement of β-TCP implanted after fibular harvest, with radiographic restoration of both cortical and medullary structures in some cases [28]. Histological examination further revealed robust new bone formation around the β-TCP scaffold, with proliferation of osteoblasts and osteoclasts indicating active bone remodeling. Experimental studies have also confirmed histologically that new bone forms around β-TCP and the regenerated bone can function as living tissue over the long term [29–31]. Based on these findings, the regenerated fibula was considered sufficiently robust to serve as a graft material, and reconstruction using this strategy was undertaken in the current case.

Postoperative CT demonstrated newly formed bridging bone between the regenerated fibula and irradiated bone. In bone grafts treated with high-dose irradiation or other inactivation methods, the resident cells are eliminated after processing, and the graft becomes devitalized bone [32–34]. Therefore, bone regeneration occurs through contributions from adjacent viable bone and stem cells derived from surrounding soft tissues. Recycled bone serves primarily as a structural scaffold, and its intrinsic ability to generate new bone is inherently weak. The bridging observed on CT suggests that the regenerated fibula preserved functional bone remodeling capacity. The result indicates that bone grafting based on the in vivo bioreactor principle could be a feasible and clinically meaningful reconstructive approach for limb salvage surgery. On the other hand, this report describes a single case in an adolescent patient; therefore, the applicability of this approach to skeletally mature individuals remains uncertain. In adult patients, alternative reconstructive options such as synthetic bone substitutes and vascularized bone grafts may be considered. In addition, the 18-year interval between the initial implantation and graft harvest in this case does not define the minimal duration required for sufficient bone regeneration based on the in vivo bioreactor principle. The optimal timing for graft harvest remains to be determined. Given that this report describes a single case, it is not possible to determine the minimal duration required to achieve sufficient bone regeneration suitable for grafting based on the in vivo bioreactor principle. However, radiographic evaluation demonstrated cortical and medullary bone formation at the donor site 1.5 years after the initial surgery. Based on this observation, a period of at least 1.5 years may be considered necessary to allow the regenerated bone to function as a biologically viable graft, although further studies are required to confirm this assumption.

In this study, the objective imaging analyses were performed as an exploratory quantitative evaluation in a single case and were intended to describe postoperative bone regeneration. Therefore, this evaluation does not aim to establish generalizable thresholds. In contrast, most previous studies have assessed the regeneration of processed bone primarily through subjective observations [4, 35]. Bone regeneration at the osteotomy gap was evaluated using grayscale analysis of serial radiographs. At both the proximal and distal osteotomy gaps, grayscale values progressively increased toward the level of the cortical bone, consistent with the subjective assessment of bone union on plain radiographs. Therefore, this approach may serve as a simple and objective method for assessing bone healing using plain radiographs alone. However, as the evaluation was based solely on grayscale values, whether this method accurately reflects bone quality or mechanical strength remains unclear, which is a limitation of this study. Further analyses of multiple cases of recycled bone grafts are required to validate the reliability and applicability of this approach. CT-based methods for evaluating bone regeneration have been attempted in several fields outside musculoskeletal oncology [19, 36], and these approaches may be applicable for assessing bone formation around recycled bone grafts. In this study, the surface area of the bridging region showed a progressive increase over time, and the internal HU values demonstrated a significant elevation. These findings suggest that the newly formed bone gradually expanded, accompanied by an increase in bone density. Although only two time points were evaluated, and more frequent CT examinations could provide a more detailed assessment, repeated CT imaging is limited by radiation exposure and cannot be performed as often as plain radiographs. Nevertheless, these two quantitative approaches—the radiographic grayscale analysis and CT-based evaluation—may offer meaningful, objective tools for assessing bone regeneration following recycled bone grafting. Despite the inherent limitations, accumulating additional clinical cases is essential to validate the feasibility and clinical relevance of this approach. Such efforts may ultimately contribute to establishing a more standardized and quantitative method for evaluating bone regeneration in treated or recycled bones.

Conclusion

We report a case in which hybrid grafting using irradiated bone and β-TCP-derived regenerated fibula successfully reconstructed a large bone defect, consistent with the in vivo bioreactor principle. The regenerated bone demonstrated grafting potential comparable to autologous bone; however, sufficient maturation required a prolonged period.

The proposed quantitative radiological assessment enabled objective evaluation of bone remodeling beyond conventional qualitative assessment and may contribute to standardized analysis in bone reconstruction, particularly for assessing bone formation after recycled autograft or synthetic bone grafting.

Abbreviations

β-TCP

Beta-tricalcium phosphate

CT

Computed tomography

MRI

Magnetic resonance imaging

ROI

Region of interest

HU

Hounsfield unit

Authors’ contributions

TA conceptualized the article, drafted the manuscript, collected the data, and edited the images. HK and AO were involved in the management of the patient. YM and TM critically revised and gave final approval of the manuscript. NO is the corresponding author, guarantor, and takes full responsibility for the article. All authors read and approved the final manuscript.

Funding

This research received no external funding.

Data availability

The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable. Ethical approval was waived by the institutional review board because this study was a single case report.

Consent for publication

Written informed consent was obtained from the patient to publish this case report and any accompanying images. A copy of the written consent form is available for review by the Editor-in-Chief of this journal.

Competing interests

The authors declare no competing interests.