3D-printed guiding templates for improved osteosarcoma resection

Osteosarcoma resection is challenging due to the variable location of tumors and their proximity with surrounding tissues. It also carries a high risk of postoperative complications. To overcome the challenge in precise osteosarcoma resection, computer-aided design (CAD) was used to design patient-specific guiding templates for osteosarcoma resection on the basis of the computer tomography (CT) scan and magnetic resonance imaging (MRI) of the osteosarcoma of human patients. Then 3D printing technique was used to fabricate the guiding templates. The guiding templates were used to guide the osteosarcoma surgery, leading to more precise resection of the tumorous bone and the implantation of the bone implants, less blood loss, shorter operation time and reduced radiation exposure during the operation. Follow-up studies show that the patients recovered well to reach a mean Musculoskeletal Tumor Society score of 27.125.

with scanning parameters set at 120 kV and 205.50 mAs. The scanning matrix size was 512 × 512. The knee joint region was continuously scanned to reach 266 layers. Images were processed into data files in Digital Imaging and Communications in Medicine (DICOM) format, and transferred to a workstation running Simpleware 6.0 software (Simpleware Ltd, Exeter, United Kingdom) to generate a 3D reconstruction model for the desired femur bone model (Figs 1B and 2).
Preoperative CT and MRI examinations of each patient were performed. One week prior to surgery, CT and MRI examinations were carried out to obtain two-dimensional (2D) CT (Fig. 3A) and MRI data of the lesion (Fig. 3B), respectively. MRI images of the corresponding region were acquired using a 1.5-T unit (Siemens Sonata; Siemens Medical Solutions, Erlangen, Germany). Postcontrast T1-weighted axial images (TR, 512 milliseconds; TE, 13 milliseconds; 2-mmthick slices) were merged with CT images for better bone-soft tissue contrast. The merging of the CT and MRI images (Fig. 3C) guided the preoperative simulation processes including CAD design of the template assisted tumor resection and 3D reconstruction of the femur with the tumor profile highlighted in red (Fig. 3D).
Manufacture of individual guiding templates using 3D printing. The 3D femur model was exported in stereolithographic (STL) format and opened in a workstation running Reverse Engineering (RE) software UG image-ware 12.0 (EDS Co., USA) in order to determine the optimal position for tumor resection. The resection margin was extended 3-5 cm beyond the perimeter of the actual tumors in bone tissue. The design of the guiding CT scan is first performed on the tumorous bone (A). The CT scan image is then used to reconstruct a 3D model of the tumorous bone using Simpleware software (B). The 3D model is then used to design a guiding template using reverse engineering software (C). The model of the guiding template is then used to fabricate the template by 3D printing (D). The 3D printed template is characterized by SEM imaging (E) and biomechanical testing (F). The template is further used to guide the surgery of osteosarcoma (G). The patient is followed after surgery to test the efficacy of the surgery (H). template (Figs 1C and 4) was based on the findings of previous studies 2, 8 . The guiding template was fabricated by 3D printing of the commercial MED610 purchased from Objet Ltd, Israel (Figs 1D and 5). The guiding template used for the three-dimensional (3D) test had an incision (25 mm long, 6 mm thick, and 5 mm wide).  Determination of compressive properties. A mechanical testing machine (Instron5967, INSTRON, USA) was used to test samples. Prior to each test, the samples were fabricated into a cylinder about 1 cm in diameter and 2 cm in height. The loading speed was set as 1 mm/min (Fig. 1F). When the load pressure declined, the compression stopped and the maximum compression pressure was determined by the peak value of compressive curve where the pressure reduction was more than 5%.
Cytocompatibility. We followed our published protocol to evaluate the cytocompatibility of the guiding templates 13,14 . Specifically, MC3T3-E1 (ATCC) cells were cultured on a small piece of the guiding templates in α -modified Eagle's medium (α -MEM) supplemented with 10% fetal bovine serum (FBS) in a humidified atmosphere of 5% CO 2 at 37 °C. Cell proliferation was tested by cell counting kit-8 (CCK-8, Dojindo, Japan). At the indicated time points (1d, 3d, 5d), 300 μl of 10% CCK-8 solution was added to the plates and incubated for 2 h at 37 °C. The optical density was read on an ELISA plate reader at 450 nm. The experiments where cells were not cultured on the guiding templates were used as control.
Surgery using 3D printed guiding templates. Research ethics approval. This study was conducted in accordance with the principles outlined in the Declaration of Helsinki and was approved by the Ethics Committee of Guangzhou General Hospital of Guangzhou Military Command. Written informed consent was obtained from all patients involved in the study (or from their parents or guardians if they were aged less than 18 years). Patient data were kept anonymous to ensure confidentiality and privacy. Surgery. After induction of general anesthesia, the surgical area was prepared and draped, and the tumor was sufficiently exposed. After appropriate preparation, the 90° flexion of the left knee was located. During the  surgery, the sterile model and guiding template were used to determine the range of the osteotomy. The tumor was surgically removed with the use of guiding templates. Using the template, a large allogeneic bone obtained before surgery was trimmed into a 3D shape matching the bone defect after tumor resection. The guiding template was used to perform distal transverse osteotomy and prune the allografted bone (Figs 1G and 6). The allograft was implanted into the bone defect and fixed with an intramedullary nail and 4 lock nails. In each case, the duration of surgery and intraoperative blood loss were recorded.
Postoperative rehabilitation. In each case, ankle extension and flexion exercises were started on postoperative day 1, hip flexion and knee flexion exercises were begun on postoperative day 3, and straight leg raises were conducted on postoperative day 7. The patient was encouraged to walk with crutches 10 days after surgery. The patient was able to walk independently 3 months after surgery.

Results
Characterization of 3D guiding template. As auxiliary medical devices, the guiding templates will be in contact with the human body such as the skin when being used clinically. Therefore, it is better for the templates to have good biocompatibility. The template should also bear good mechanical property because surgeons exert vertical force on the templates during operation. Hence, the templates need to have good strength and tenacity to avoid fracture. Therefore, we characterized the mechanical properties and biocompatibility. Figure 7 showed the surface morphology of the 3D guiding template, which indicated that a compact scaffold had been successfully fabricated. The mechanical properties of the guiding template were further characterized. As can be seen from the results (Fig. 8), the maximal compressive strength of the template (60.01 ± 2.52 MPa).
Cell proliferation during the first 5 days of incubation with the guiding templates was assessed as showed in Fig. 9. On day 1, cells cultured in the presence or absence of the guiding plates showed a relatively lower absorbance in comparison with those at later time points. After 3 days, the absorbance measurements of all groups increased drastically, indicating that the number of living cells was increased. However, during the first 5 days, the cell numbers in the presence of the guiding plates showed no statistically significant differences compared with the control group (p > 0.05) (Fig. 9).
Clinical study. In all cases, the optimal position for resection was determined by the corresponding 3D printed guiding template. Each guiding template fitted its corresponding femur model perfectly. The entry point and direction for the osteosarcoma resection were therefore chosen based on the guiding templates, providing accuracy and convenience during the procedure. In conventional surgery without the use of the guiding template 15 , the average operation time is 217 min (300-420 min in our hospital), the average volume of blood loss is 1025 ml (1000-2000 ml in our hospital), and the size of the surgical incision in our hospital varies between 250 and 350 mm. In contrast, in the surgery guided by the 3D printed guiding template, the operation time, blood loss and incision size ranged 180-250 min, 560-900 ml, and 115-180 mm, respectively ( Table 1).
All patients were followed up from 25 to 42 months (mean, 33 months), and all 8 patients were alive at the time of reporting. The Musculoskeletal Tumor Society (MSTS) score was between 21 and 30 at the last follow-up 16 , and the mean knee flexion was 112.5° (range, 90°-130°, Table 2). It should be noted that the MSTS score is a common method for evaluating the postoperative evaluation of osteosarcoma surgery 16 . At the 2-year follow-up check, the patient was alive and living well without the evidence of recurrence. A plain X-ray film revealed that the bone defect was healed and no bone tumor was identified (Figs 1H and 10). The X-ray scan showed that using the individual templates resulted in the surgery with a high degree of precision, but without the cases of intramedullary nail misplacement (Fig. 10). The implants were stabilized and could bear sufficient weight to allow patients to eventually recover normal physical activity.

Discussion
Guiding templates have been applied in the field of orthopaedic oncology in recent years. It may facilitate resection and reconstruction in patients with complex bone tumor 8,11,12 . Inadequate resection margins are associated with higher risk of local tumor recurrence and poorer patient survival 17  technique, but the local recurrence rate was 13% at a mean follow-up of 13.1 months 18 . In other studies on the computer-assisted tumor surgery 19,20 , the local recurrence rates were reported to be 20-25% at a minimum of 3 years of follow-up.   In order to achieve safe and accurate tumor resections, it is necessary to establish the profile of the tumor in the bone and soft tissues. Due to the complexity associated with the musculoskeletal bone tumor resections and reconstructions, tumor surgeons usually used 2D CT and MRI images for surgical planning. Such planning is difficult without the help of a computer due to the more complex regional anatomy in osteosarcoma. In recent years, surgery using navigation systems has helped to facilitate resection and reconstruction in patients with complex osteosarcoma, and several preliminary reports have described its application in bone tumor surgery 17,[20][21][22][23][24] . These reports involve the resection of pelvic and sacral tumors and joint preserving limb salvage surgery 20,23,24 . The surgical navigation systems greatly improved the accuracy and safety of implantation and reduced neural, visceral, and vascular complications. Nevertheless, these systems still have a number of disadvantages including increased cost, extended surgical time, additional radiation exposure, cumbersome surgical procedures, and significant learning curves. Registration-related errors and changes in the patient's position can also have a negative impact on the accuracy of surgical navigation systems.
In this study, we used 3D printing technique to manufacture the guiding templates for high grade osteosarcoma. We used a combination of CT scan and MRI imaging to obtain the model of the tumors and then employed the model to produce the guiding templates by 3D printing technique. The cell proliferation studies clearly demonstrated that the guiding templates are biocompatible. The characterization of the mechanical properties also showed that the guiding templates have good compressive strength to sustain the surgery process. It was reported that using surgical guides could improve the accuracy of resection of malignant bone tumors 8,11 . It was also suggested that the osteotomy guides achieved by novel 3D printing technologies could reduce the treatment costs and production time 12 . Compared to these studies, our study not only characterized the biological  and materials properties of the guiding templates but also resulted in shorter operation time, less blood loss and smaller incision size.
We then used 3D printed guiding templates to assist the surgery of osteosarcoma and yielded satisfactory results. Specifically, the use of the 3D guiding templates has led to precise resection of the bone and implantation of allografts, less blood loss, shorter surgery time and reduced radiation (Tables 1 and 2). In addition, it took about 10 h to manufacture a 3D printed guiding template and femur model, and the cost of such manufacture was estimated to be about $30. Namely, the use of the guiding templates is cost-efficient. Our study shows that the patients subjected to the surgery guided by the guiding templates recovered in 3 months.
Although our studies show the advantages of the guiding templates, there are still some limitations of our study. First, the primary limitations of this study are the small number of cases and the relatively short follow-up time. Second, the potential benefits of the surgery guided technique in improving surgical accuracy may help reduce the risk of local recurrence but may not translate into better patient survival for metastatic disease. Third, the control study without using the guiding templates could not be performed on the same patients.

Conclusion
In conclusion, personalized guiding templates for osteosarcoma resection provide surgeons with a novel tool for optimizing allograft bone construction and thus minimizing overall tissue trauma and reducing the risk of damaging nervous and vascular structures. The guiding templates, manufactured by 3D printing technique, are biocompatible and have good compressive strength. The current study demonstrates that the application of such guiding templates can result in shorter surgical duration, lower radiation exposure and less blood loss. It also makes the surgery easier to perform. With its wide applicability, high accuracy, and cost-effectiveness, the use of this technology is expected to become widespread in the future.