Three‐dimensional printing in radiation oncology: A systematic review of the literature

Abstract Purpose/objectives Three‐dimensional (3D) printing is recognized as an effective clinical and educational tool in procedurally intensive specialties. However, it has a nascent role in radiation oncology. The goal of this investigation is to clarify the extent to which 3D printing applications are currently being used in radiation oncology through a systematic review of the literature. Materials/methods A search protocol was defined according to preferred reporting items for systematic reviews and meta‐analyses (PRISMA) guidelines. Included articles were evaluated using parameters of interest including: year and country of publication, experimental design, sample size for clinical studies, radiation oncology topic, reported outcomes, and implementation barriers or safety concerns. Results One hundred and three publications from 2012 to 2019 met inclusion criteria. The most commonly described 3D printing applications included quality assurance phantoms (26%), brachytherapy applicators (20%), bolus (17%), preclinical animal irradiation (10%), compensators (7%), and immobilization devices (5%). Most studies were preclinical feasibility studies (63%), with few clinical investigations such as case reports or series (13%) or cohort studies (11%). The most common applications evaluated within clinical settings included brachytherapy applicators (44%) and bolus (28%). Sample sizes for clinical investigations were small (median 10, range 1–42). A minority of articles described basic or translational research (11%) and workflow or cost evaluation studies (3%). The number of articles increased over time (P < 0.0001). While outcomes were heterogeneous, most studies reported successful implementation of accurate and cost‐effective 3D printing methods. Conclusions Three‐dimensional printing is rapidly growing in radiation oncology and has been implemented effectively in a diverse array of applications. Although the number of 3D printing publications has steadily risen, the majority of current reports are preclinical in nature and the few clinical studies that do exist report on small sample sizes. Further dissemination of ongoing investigations describing the clinical application of developed 3D printing technologies in larger cohorts is warranted.


| INTRODUCTION
Three-dimensional (3D) printing is an additive manufacturing technique used to generate customizable 3D objects using a variety of stock materials. Given its versatility and commercial availability, 3D printing is increasingly being incorporated into medical practice and innovation. 1 It has been used in a diverse array of applications ranging from training and education 2 to therapeutic medical devices. 3 Three-dimensional printing often represents a low-cost alternative to traditional material fabrication methods and can also be used to generate patient-specific models which might not otherwise be readily obtainable. 4,5 Furthermore, 3D printing allows for fabrication of certain complex geometries that is not possible by other techniques such as milling or injection molding.
Although 3D printing is recognized as an effective clinical and teaching tool in procedurally intensive medical specialties such as oral maxillofacial surgery and orthopedics, 6 it has a nascent role in radiation oncology. Therefore, the goal of this investigation is to clarify the extent to which 3D printing applications are currently being used in radiation oncology through a systematic review of the literature. The specific aims are to comprehensively characterize current 3D printing applications in the field, identify possible areas of growth, and create a framework for encouraging safe and effective implementation of such technologies for current and future radiation oncology practitioners.

| MATERIALS AND METHODS
A search protocol was defined according to preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. 7 PubMed was searched by two independent reviewers (MKR and DMR) using combinations of predetermined medical subject headings (MeSH) and generic search terms. Included MeSH terms were radiation oncology, radiotherapy, radiotherapy planning, computer-assisted, printing, three-dimensional, and neoplasms. Generic search terms included three-dimensional printing, 3D printing, 3D printable, 3D printer, radiotherapy, radiation oncology, dosimetry, additive manufacturing, fused deposition modeling (FDM), and stereolithography (SLA). Fused deposition modeling and SLA are two of the popular additive manufacturing techniques performed by 3D printers. In FDM, a material is melted and extruded through a small nozzle, depositing the material layer by layer to construct the 3D-printed object. By contrast, SLA uses a curable photopolymer material to construct objects typically in a top-to-bottom fashion. The material is hardened through the application of focused or UV light to generate the final product.
Search queries with less than 200 results were manually reviewed (Table S1). In-press publications in radiation oncology and medical phy- Eligibility for inclusion of articles was determined using the population, intervention, control, outcomes (PICO) framework. 8 The population included any peer-reviewed publication after January 1, 2000 available in English language. This date was selected because it represents the approximate time when 3D printing was introduced and then widely adopted in medicine. 9 Interventions included any use of 3D printing technology in relation to the study or delivery of radiotherapy. The controls, when defined, were conventional procedures as compared to 3D printing interventions, as specified in each study.
Outcomes included qualitative and quantitative results described in articles, if applicable. There were no specific exclusion criteria. Any disagreements regarding inclusion of articles were discussed with the study team until consensus was reached.
Included studies were coded into a database and evaluated according to predetermined parameters of interest. End points included in this study were (a) year and country of publication, (b) description of the technology or intervention, (c) study type (randomized controlled trial, cohort study, case series or report, review or meta-analysis, translational or basic science research, preclinical feasibility study, cost or workflow evaluation study), (d) sample size for clinical intervention studies, (e) type of radiation oncology application, (f) reported outcomes, and (g) safety concerns or implementation barriers. Studies were considered clinical intervention studies if they evaluated or tested a 3D-printed application directly on patients. Given unknown trends in types of published 3D printing applications, the "type of radiation oncology application" end point was coded and categorized in a post hoc fashion after qualitative review of all studies. Outcomes were heterogeneous and general trends were qualitatively assessed after finalization of data collection. Coding of outcomes was performed with particular attention to (a) accuracy, (b) precision, (c) workflow efficiency, (d) cost, (e) patient experience, and (f) educational value.
Linear regression was used to analyze the rate of publication, and all statistical calculations were performed using R 3.3.2 (R Foundation, Vienna, Austria). PRISMA guidelines also recommend FUNDING INFORMATION The authors did not receive any financial support for this investigation. additional analyses such as sensitivity or subgroup analyses, as appropriate. However, such recommendations are more applicable for meta-analyses and thus were prospectively discarded for the present investigation. Institutional Review Board approval was not required due to the nature of the study.
Reported outcomes are summarized in Table 3. Although end points were heterogeneous, the majority of studies described effective implementation of 3D printing technology. The most commonly reported outcome was dosimetric evaluation of printed interventions (50.5%). Other common end points were related to the 3D printing F I G . 1. Search protocol for identification of eligible articles. *Medical Subject Headings (MeSH) terms included: radiation oncology; radiotherapy; radiotherapy planning, computer-assisted; printing, three-dimensional; neoplasms. Generic search terms included: three-dimensional printing; 3D printing; 3D printable; 3D printer; radiotherapy; radiation oncology; dosimetry; additive manufacturing; fused deposition modeling (FDM); stereolithography (SLA). **Inclusion criteria were defined according to the PICO (population, intervention, control, outcomes) framework. The population included any peer-reviewed publication after January 1, 1990 available in English. Interventions included any uses of 3D printing technology specifically in the context of radiation oncology or the delivery of radiotherapy. The control, if defined and applicable, was standard of care interventions or procedures compared to a 3D printing-related intervention as specified in each study. Outcomes included qualitative and quantitative results described in articles, if applicable process itself, such as printing accuracy, time, and cost. Furthermore, many studies evaluated the radiological properties (12.6%) and durability or deformation (2.9%) of printed materials. Studies describing the use of 3D printing for preclinical animal irradiation (9.7%), such as animal immobilizers, often described animal positional accuracy (3.9%), or histologic confirmation of targeted radiotherapy delivery (2.9%). Clinical or disease-related outcomes such as treatment toxicity (3.9%) and disease control or overall survival (1.9%) were rarely reported, reflecting the preclinical nature of many of the identified investigations.
The majority of articles described at least one safety concern or implementation barrier for the use of 3D printing applications in radiation oncology (69.9%, Table 4). The most commonly described barriers were related to the 3D printing process, including limitations in time/workflow (12.6%), printing accuracy (10.7%), cost (8.7%), limited print volume (6.8%), and a requirement for printing space (1%). Additionally, 3D printing filaments (materials that are directly printed) were routinely cited sources for safety concerns and implementation barriers. Variable radiological properties (8.7%), limited biocompatibility and sterilization capacity (7.8%), dosimetric inconsistency or inaccuracy (3.9%), hardness impacting patient comfort or tissue simulation (2.9%), durability or stability (2.9%), and limited color (1.9%), flexibility (1.9%), and number of simultaneously printed filaments (1.9%) were the most commonly described concerns with current 3D printing filaments or materials. Additionally, the need for clinical validation prior to routine use (6.8%) and limited generalizability of 3D-printed interventions for individual patient characteristics (5.8%) were commonly described barriers. One article described that there is a significant learning curve for the 3D printing fabrication process which acts as a barrier to routine clinical implementation.

| DISCUSSION
To our knowledge, this investigation represents the most comprehensive systematic review describing 3D printing applications in radiation oncology. Through formal characterization of currently reported 3D printing interventions and identification of common barriers to safe and effective implementation, this study will serve as a guide for practitioners and researchers considering introducing 3D printing technology in their routine practice. Many exciting opportunities exist for the implementation of 3D printing in the radiation oncology clinic or laboratory and the majority of identified interventions appeared to successfully improve treatment delivery according to dosimetric analysis or small-sample clinical evaluation. However, further work will be needed to confirm the efficacy of such interventions in larger clinical settings.
The most commonly described applications were quality assurance phantoms, brachytherapy applicators, and bolus (Table 1). This trend likely reflects the ability of 3D printers to fabricate relatively low-cost patient-specific models, which serves as one of the central advantages of 3D printing compared to traditional material fabrication methods such as casting or molding. 35 For quality assurance phantoms, brachytherapy applicators, and bolus, the use of individualized models can improve treatment delivery and minimize unnecessary toxicity. 11,24,36 For example, 3D-printed bolus can conform more closely to patient skin compared to traditional bolus and thereby minimize inaccuracies in treatment delivery. 12,13 Such patient-specific models are not easily created with traditional techniques or may have high cost of production and thus may not be readily available for routine use. With 3D printing T A B L E 1 Evaluation summary of identified articles.  Fig. 2). Furthermore, studies rarely reported objective clinical outcomes such as disease control or survival (Table 3). This This investigation also identifies important gaps in current 3D printing research in radiation oncology and should help guide future academic efforts in the field. For example, 3D printing has gained prominence as an educational tool in a number of procedurally intensive medical specialties. 2 However, this review identified only one study in radiation oncology which used 3D printing for educational purposes, specifically for the development of a cervical cancer brachytherapy training simulator. 37 In surgical and procedural settings, 3D printing is used to generate models on which learners can practice complex surgical or manual skills. 3,5,38 This approach offers a number of advantages over traditional teaching methods, including minimizing potential harm to patients and increasing overall training exposure, particularly in rarely encountered clinical situations. [39][40][41][42] Given the proven efficacy of 3D printing in other areas of medical education, this discrepancy represents an important opportunity for future innovation.
Although not typically considered a procedurally-intensive specialty, the field of radiation oncology could benefit from incorporation of 3D printing into training. For instance, 3D-printed patientspecific models have been used as tools to help visualize complex   Kuijten 30 Printed conformer with mask to aid in the treatment of contracted sockets after radiation damage in the treatment of retinoblastoma Case report/ series 1 Mask for treatment of radiationrelated tissue contraction As a result of the treatment, the patient can now wear a cosmetic prosthesis on average for 3 hours a day. The conjunctival lining was expanded compared to baseline. The mask and conformer were well tolerated, and no adverse effects were encountered. However, surgical intervention was still eventually required 2017 Zhao 31 Clinical application of 3D printing for bolus fabrication and brachytherapy applicators Case report/ series 5

Multidisciplinary
Based on the planning CT, the size of the largest air gap at the interface of the 3D-printed structure was 3 mm, 3 mm, 2 mm, and 2 mm for four cases using printed bolus. The surface brachytherapy plan for the adequate coverage (95% isodose to 95.6% of CTV]), but a relatively high dose to the left eye, owing to its proximity to the tumor 2016 Briggs 32 Personalized radiotherapy facial protective masks using a facial scanner.
Case report/ series 1 Protective shielding The lead mask fitted comfortably and was confirmed by the lead clinician to be safe to use for treatment. The final print time was 30 hours and a total of 200 g of print material was used at an approximate cost of £8.00 (Continues) arteriovenous malformations (ATMs) during radiosurgical treatment planning. 34 Expanding on this concept, 3D printing patient-specific organs or tumors might also help trainees visualize complex anatomical relationships when learning how to contour.
Furthermore, although not formally considered a component of graduate medical education, 3D printing could also be used as a tool to enhance patient education in radiation oncology. As an example, prior to surgical consultations, 3D-printed models of patient anatomy and pathology have been used as tools to guide conversations and educate patients. [43][44][45] Such interventions could similarly be used to help patients understand important details of radiotherapy, such as describing the use of radiosurgery for the treatment of brain metastases.
Another significant gap identified in this investigation is a lack of literature describing effective methods to implement 3D printing techniques in the laboratory and clinic. While many studies acknowledged the need for 3D printing expertise for consistent fabrication of accurate interventions, few reports explicitly described methods to teach current providers how to effectively use, design, and evaluate 3D printing applications. 46,47 Additionally, there were few studies describing the workflow systems for efficiently incorporating 3D printing into the clinic or laboratory. 15,48,49 Expansion of such work would be a pivotal step in encouraging widespread dissemination of this technology.
Although this investigation highlights exciting opportunities for the application of various 3D printing interventions in radiation oncology, it is important to carefully consider common safety concerns and implementation barriers prior to employment of such technologies ( Table 4). The most frequently described of such barriers involved various aspects of the 3D printing process which can require significant experience before consistently accurate performance can be achieved. 50 Additionally, the financial and time burden can be large, including not only the cost of printers and materials themselves, but also the need for a physical printing space and training of staff. Encouragingly, the cost of printers has consistently decreased over recent years and many highly accurate printers are now available at reasonable prices. 51 Printing times and maximum print volumes can also be restrictive, as the majority of commercially available printers can take up to 12 or more hours for a single print and may only be able to fabricate a cubic foot of material in one session. For some radiation oncology applications, these printing times and volumes may be acceptable, but for others such as full thoracic phantoms, such issues may be significantly limiting.
Similarly, although advances have been made to improve avail- Sharma 33 Clinical evaluation of lead face shielding created using an optical scanner and 3D printer Case report/ series 10 Protective shielding Lead shields created using this approach were accurate and wellfitting. The process added to patient convenience and addressed potential claustrophobia and medical inability to lie supine 2016 Conti 34 Creation of patient specific models for AVM lesions as tools to aid in radiosurgery treatment planning Cohort study 10 Radiosurgical treatment planning Contouring time was shorter when using 3D-printed model of the AVM than without (p = 0.001). The average volume contoured without the 3D model was 5.6 ± 3 mL whereas it was 5.2 ± 2.9 mL with the 3D-printed model (p = 0.003). Surgeons were absolutely confident or very confident in all cases that the volume contoured Again, researchers are currently developing methods to overcome such material limitations, such as the use of flexible printing filaments which would improve patient comfort when using 3D-printed bolus. Lastly, safety assessment of 3D printing may be useful to guide future applications and standardization of best practices. 56,57 Despite the use of a structured protocol according to PRISMA guidelines, this study has methodological limitations. First, it is possible that some publications were not identified using the predeter- printing and traditional approaches. Lastly, for the sake of simplicity, safety concerns and implementation barriers were coded as a single end point; however, in reality these represent two distinct although related entities. For example, while biocompatibility of materials is both a safety concern and a barrier to implementation, long printing time would be more accurately categorized as an implementation barrier alone. This distinction should be considered when interpreting aggregate data reported in this study.

| CONCLUSION
This systematic review comprehensively characterizes current uses of 3D printing in radiation oncology and identifies common barriers Animal positional accuracy 4 3.9 Histologic confirmation of accurate radiotherapy delivery (preclinical animal irradiation) 3 2.9 Material durability or deformation 3 2.9 Perceived utility by providers 2 1.9 Disease-related clinical outcome 2 1.9 Learner comfort with brachytherapy 1 1.0 Contouring time 1 1.0 T A B L E 4 Implementation barriers and safety concerns when using 3D printing technology in radiation oncology.
Implementation barrier or safety concern Count (n = 103) % At least one barrier or safety concern described 72 69.9 3D printing process Hardness (for patient comfort and/or tissue simulation) 3 2.9 Stability/durability 3 2.9 Limited color availability 2 1.9 Limited range of flexibility 2 1.9 Limited number of materials printed at a time 2 1.9 Requires clinical validation before implementation Therefore, this study provides a resource for radiotherapy practitioners considering introducing 3D printing in their practice and could guide further research efforts to expand the role of 3D printing in radiation oncology.

CONF LICTS OF INTEREST
Dr. Golden reports having a financial interest in RadOnc Questions, LLC.

SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of the article. Table S1. Search query results using combinations of predetermined search terms* Table S2. List of all identified articles describing 3D printing applications in radiation oncology