Beryllium oxide (BeO) optically stimulated luminescent dosimeters were characterized across a broad range of energies and doses to evaluate their suitability for clinical dosimetry applications beyond manufacturer-specified limits. Seventy-eight unscreened BeO dosimeters were evaluated for sensitivity, stability, dose linearity, energy dependence, and signal depletion using clinical radiation sources spanning kilovoltage photons (100-320 kVp), megavoltage photons (6-18 MV), and electron beams (6-20 MeV). Dose response was assessed from 0.5 to 50 Gy, and signal depletion was quantified over ten successive readouts at multiple post-irradiation time points. Individual dosimeter sensitivities varied from 0.8 to 1.3 relative to the batch mean, while 75% demonstrated a coefficient of variation below 1% across repeated measurements, indicating good reproducibility. Dose response remained linear up to approximately 30 Gy, with saturation observed near 40 Gy. No degradation in dose calibration accuracy was observed for accumulated doses up to 50 Gy, and a small sensitivity shift at 100 Gy remained within predefined equivalence bounds. A pronounced under-response of 25-40% was observed at kilovoltage energies relative to megavoltage photons, whereas responses across megavoltage photon and electron beams were stable within +/- 2%. Signal depletion was consistent at approximately 1% per readout and independent of post-irradiation time. These results define the operational limits of unscreened BeO dosimeters and support their use in clinical dosimetry, provided that energy-specific calibration is applied for kilovoltage applications.

Additively manufactured (3D-printed) surgical reconstruction implants are increasingly encountered in head and neck oncology and may introduce density assignment challenges during radiotherapy treatment planning. This case report describes a clinically significant dosimetric discrepancy arising from the application of nominal bulk material density to an additively manufactured surgical reconstruction implant and to highlight workflow considerations for density assignment. A patient treated with helical tomotherapy following mandibular reconstruction with an additively manufactured titanium implant underwent kilovoltage computed tomography (kVCT) simulation. Saturation of Hounsfield units within the implant precluded reliable density estimation, necessitating a density override based on nominal bulk material properties. During treatment, megavoltage computed tomography (MVCT) imaging with adaptive dose recalculation was used to evaluate the dosimetric impact of this assumption. Replanning was performed using a revised density override informed by imaging-based evaluation while maintaining institutional planning criteria. Adaptive dose recalculation demonstrated a localized calculated dose increase of approximately 7% within portions of the clinical and planning target volumes adjacent to the reconstruction implant. Replanning with a revised density override reduced unintended target dose heterogeneity while preserving target coverage. Nominal bulk material density may not accurately represent TPS-relevant radiologic density for additively manufactured surgical reconstruction implants, and this discrepancy may not be apparent on kVCT imaging. Independent evaluation of density assumptions using alternative imaging modalities can identify clinically meaningful discrepancies and support patient-protective replanning decisions.

The purpose of this work was to evaluate the imaging dose for the Accuray Radixact ClearRT system. Low-contrast resolution and CT number consistency was evaluated as well. CTDI measurements were compared to vendor supplied values, and similar measurements were done on a Varian TrueBeam kV cone-beam CT (CBCT) and a Philips Big Bore CT scanner. In-field imaging doses were measured using various protocols in an anthropomorphic phantom, while out-of-field doses were measured 10 cm from the imaging field edge. Comparison of the CTDI and the in-field doses showed considerable disagreement when the patient anatomy size was not congruent with the size of the CTDI phantom. While that is an expected outcome, this work provides an estimate of the differences for a variety of ClearRT protocols when this situation arises. The CNR was measured for all combinations of ClearRT settings for comparisons within the system, as well as for a comparison with a CBCT and fan-beam CT system. The CNR and dose information provided in this work can be used to aid in selecting a ClearRT imaging protocol. The CT number stability was tracked over 27 months; two instances where the CT number constancy exceeded tolerance were observed after service.

Purpose/Objectives:Late renal toxicity can be impactful for stem cell transplant (SCT) patients with total body irradiation (TBI) as part of the conditioning regimen. This work describes a two component repair model to predict renal toxicity among pediatric patients. Materials/Methods:Systematic literature searches described in two other reports were used to identify articles describing late renal toxicity in children treated with TBI as part of SCT. Six studies were included with a combined sample size of 172 patients. Data pertaining to TBI and renal toxicity endpoints were extracted. Probability of renal toxicity was analyzed in relation to TBI dose, fractionation, and dose rate. A logistic regression model was developed using studies with sufficient TBI and pediatric outcome data. The toxicity was calculated from reported rates from each study and were primarily based on abnormal laboratory values. Results:Physical dose and a traditional method of calculating EQD2 were tested for a significant dose response relationship although none were found. Instead, a statistically significant relationship was found with a two component sublethal repair EQD2 calculation model using logistic regression (p = 0.03). The model predicted the odds ratio for renal toxicity to be 2.9 Gy-1. Predicted probabilities of 5 %, 10 %, and 50 % clinically significant renal toxicities are associated with EQD2 values of 10.4, 11.1, and 13.2 Gy, respectively. Conclusions:A model to predict the risk of renal toxicity was determined. This model indicates that dose rate and the time interval between fractions are important factors in renal toxicity for pediatric patients receiving SCT with TBI as part of the conditioning regimen.

Purpose: We present a patient-specific model to estimate tumor location in the thorax during radiation therapy using chest surface displacement as the surrogate signal.Methods: Two types of data are used for model construction: Four-dimensional computed tomography (4D-CT) images of the patient and the displacement of two points on the patient's skin on the thoracic area. Principal component analysis is used to fit the correspondence model. This model incorporates the recorded surrogate signals during radiation delivery as input and delivers the 3D trajectory of the tumor as output. We evaluated the accuracy of the proposed model on a respiratory phantom and five lung cancer patients.Results: For the respiratory phantom, the location of the center of the sphere during treatment was calculated in three directions: Left-Right (LR), Anterior-Posterior (AP) and, Superior-Inferior (SI). The error of localization was less than 1 mm in the LR and AP directions and less than 2 mm in the SI direction. The location of the tumor center for two of the patients, and the location of the apex of the diaphragm for the other three, was calculated in three directions. For all patients, the localization error in the LR and AP directions was less than 1.1 mm for two fractions and the maximum localization error in the SI direction was 6.4 mm.Conclusions: This work presents a feasibility study of utilizing surface displacement data to locate the tumor in the thorax during radiation treatment. Future work will validate the model on a larger patient population.

PurposePatient positioning and immobilization devices are commonly employed in radiation therapy. Unfortunately, cases can arise where the devices need to be reconstructed or improved. This work describes clinical processes to use a planning CT, to design and 3D print immobilization devices for reproducible patient positioning within a clinically feasible time frame when traditional methods can no longer be used or are insufficient. Materials/MethodsThree clinical cases required rapid 3D printing of an immobilization device mid-treatment due to the following: (1) a lost headrest cushion, (2) needed improvement in lumbar spine positioning, and (3) a partially deflated vacuum immobilization mattress. ResultsIn the three cases, the 3D printed immobilization devices were clinically implemented successfully; two of the devices were fully designed and printed in 1 day. The 3D printed immobilization devices achieved a positioning accuracy sufficient to avoid the necessity to repeat the simulation and planning process. ConclusionIf traditional immobilization devices fail or are misplaced, it is feasible to have a 3D printed replacement within the time span of 1 day. The design and fabrication methods, as well as the experiences gained, are described in detail to assist clinicians to implement 3D printing for similar situations.

PURPOSE:Pulmonary complications, especially idiopathic pneumonitis syndrome (IPS), are potentially life altering or fatal sequelae of hematopoietic cell transplantation (HCT). Total body irradiation (TBI) as part of the conditioning regimen has been implicated in IPS. A comprehensive PENTEC (Pediatric Normal Tissues in the Clinic) review was performed to increase our understanding of the role of TBI in the development of acute, noninfectious IPS. METHODS AND MATERIALS:A systematic literature search was conducted using the MEDLINE, PubMed, and Cochrane library databases for articles describing pulmonary toxicity in children treated with HCT. Data pertaining to TBI and pulmonary endpoints were extracted. Risk of IPS was analyzed in relation to patient age, TBI dose, fractionation, dose rate, lung shielding, timing, and type of transplant, with the goal to better understand factors associated with this complication in children undergoing HCT. A logistic regression model was developed using a subset of studies with comparable transplant regimens and sufficient TBI data. RESULTS:Six studies met criteria for modeling of the correlation of TBI parameters with IPS; all consisted of pediatric patients undergoing allogeneic HCT with a cyclophosphamide-based chemotherapy regimen. IPS was variably defined, but all studies that reported IPS were included in this analysis. The mean incidence of post-HCT IPS was 16% (range, 4%-41%). Mortality from IPS, when it occurred, was high (median, 50%; range, 45%-100%). Fractionated TBI prescription doses encompassed a narrow range of 9 to 14 Gy. Many differing TBI methods were reported, and there was an absence of 3-dimensional dose analysis of lung blocking techniques. Thus, a univariate correlation between IPS and total TBI dose, dose fractionation, dose rate, or TBI technique could not be made. However, a model, built from these studies based on prescribed dose using a normalized dose parameter of equivalent dose in 2-Gy fractions (EQD2), adjusted for dose rate, suggested correlation with the development of IPS (P = .0004). The model-predicted odds ratio for IPS was 24.3 Gy-1 (95% confidence interval, 7.0-84.3). Use of TBI lung dose metrics (eg, midlung point dose) could not be successfully modeled, potentially because of dosimetric uncertainties in the actual delivered volumetric lung dose and imperfections in our modeling process. CONCLUSIONS:This PENTEC report is a comprehensive review of IPS in pediatric patients receiving fractionated TBI regimens for allogenic HCT. IPS was not clearly associated with 1 single TBI factor. Modeling using dose-rate adjusted EQD2 showed a response with IPS for allogeneic HCT using a cyclophosphamide-based chemotherapy regimen. Therefore, this model suggests IPS mitigation strategies can focus on not just the dose and dose per fraction but also the dose rate used in TBI. More data are needed to confirm this model and to determine the influence of chemotherapy regimens and contribution from graft-versus-host disease. The presence of confounding variables (eg, systemic chemotherapies) that affect risk, the narrow range of fractionated TBI doses found in the literature, and limitations of other reported data (eg, lung point dose) may have prevented a more straightforward link between IPS and total dose from being observed.

AbstractPurposeTo evaluate the use of commercial‐grade strontium aluminate phosphorescent powder as a thermoluminescent (TL) dosimeter for clinical radiotherapy beams.Materials and MethodCommercially available Eu2+, Dy3+ co‐doped strontium aluminate powder (SrAl2O4:Eu, Dy) was annealed and then irradiated using 20 × 20 cm2 field size, with 6‐MV (PDD10 = 70.7) and 18‐MV (PDD10 = 79.4) photon beams and and 9‐MeV (R50 = 3.6), 15 MeV (R50 = 5.9) and 18‐MeV (R50 = 7.2) electron beams. To calibrate the relationship between the TL readings and the irradiated doses, TL glow curves were acquired for doses up to 600 cGy at all beam energies. For the percentage depth dose (PDD) measurement, the SrAl2O4:Eu, Dy powder was sandwiched by solid water phantoms, with varying thickness of solid water placed above to determine the depth. PDDs were measured at four representative depths and compared against the commissioning depth dose data for each beam energy.ResultsLinear dose response models of doses up to 200 cGy were created for all beam energies. Superlinearity was observed with doses greater than 200 cGy. The PDD measurement acquired experimentally agrees well with the commissioning data of the medical linear accelerator. Trapping parameters such as order of kinetics, activation energy and frequency factor have been obtained via TL glow curve analysis.ConclusionThe linear dose response demonstrates that SrAl2O4:Eu, Dy is a potential TLD dosimeter for both electron beams and photon beams at different beam energies. The PDD measurements further support its potential use in quality assurance and radiation dosimetry.

3D printing is a method of semi-automated custom fabrication. The advantages of this technology for medical physics is apparent; objects can be constructed for specific clinical needs and constructed with moderate expertise. The cost of 3D printing is cheaper compared to other custom fabrication methods or purchasing from established vendors. Phantoms can be constructed for specific quality control tasks and radiation dosimetry measurements, including patient specific measurements. Beyond phantoms, 3D printing can be used for procedure devices in the form of radiation bolus, immobilization devices, surgical planning models, and rudimentary radiation detectors. Real clinical uses of 3D printing will be presented as well.

PURPOSE:To characterize the dose distribution in water of a novel beta-emitting brachytherapy source for use in a Conformal Superficial Brachytherapy (CSBT) device. METHODS AND MATERIALS:Yttrium-90 (90Y) sources were designed for use with a uniquely designed CSBT device. Depth dose and planar dose measurements were performed for bare sources and sources housed within a 3D printed source holder. Monte Carlo simulated dose rate distributions were compared to film-based measurements. Gamma analysis was performed to compare simulated and measured dose rates from seven 90Y sources placed simultaneously using the CSBT device. RESULTS:The film-based maximum measured surface dose rate for a bare source in contact with the surface was 3.35 × 10-7 cGy s-1 Bq-1. When placed in the source holder, the maximum measured dose rate was 1.41 × 10-7 cGy s-1 Bq-1. The Monte Carlo simulated depth dose rates were within 10% or 0.02 cm of the measured dose rates for each depth of measurement. The maximum film surface dose rate measured using a seven-source configuration within the CSBT device was 1.78 × 10-7 cGy s-1 Bq-1. Measured and simulated dose rate distribution of the seven-source configuration were compared by gamma analysis and yielded a passing rate of 94.08%. The gamma criteria were 3% for dose-difference and 0.07056 cm for distance-to-agreement. The estimated measured dose rate uncertainty was 5.34%. CONCLUSIONS:90Y is a unique source that can be optimally designed for a customized CSBT device. The rapid dose falloff provided a high dose gradient, ideal for treatment of superficial lesions. The dose rate uncertainty of the 90Y-based CSBT device was within acceptable brachytherapy standards and warrants further investigation.

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.

The purpose of this work is to evaluate a commercially available copper-plastic composite material for use as a custom fit 3D printed bolus. Superficial dose under copper-plastic composite bolus was assessed for 0.4 mm, 0.6 mm, and 0.8 mm thicknesses. Superficial dose measurements were performed with an Affix parallel plate ionization chamber and radiochromic film. Additionally, a custom-fit bolus was designed for the temporal-frontal cranial region of an anthropomorphic phantom. A treatment plan with a tangential field arrangement was designed, and radiochromic film was used to measure the dose enhancement to the surface of the phantom from the bolus and compared to the calculated dose. It was shown that 3D printed copper-plastic composite bolus can provide the equivalent dose enhancement of thicker conventional bolus. Due to the limited thickness of the copper-plastic composite the bolus can remain flexible, which can aid in the placement of the bolus and improve patient comfort.

BACKGROUND:Skin tumors are the most predominant form of cancer in the United States. Radiation therapy, particularly high dose-rate (HDR) brachytherapy, provides an effective form of cancer control when surgery is not possible or when surgical margins are incomplete. The treatment of superficial skin cancers on irregular surfaces, such as the nose, lips or ears, present challenges for treatment. To address this issue, we designed and constructed a novel conformal superficial brachytherapy (CSBT) device prototype to improve patient-specific treatment for complex sites. The device is mounted on an automated remote after-loader, providing limited radiation exposure to operating personnel, is inexpensive to construct, and offers a unique method of conformal surface radiation therapy.RESULTS:A prototype of the CSBT device was successfully manufactured. A computed tomography (CT) scan of a Rando phantom was used to plan the target treatment area. The CSBT device has a hexagonal lattice array of retractable rods with radioactive seeds placed at the tip of each rod. A 3D-printed conformal shape insert with a hexagonal array of cylindrical projections of varying length is driven into the rods by a single linear actuator. The rods are displaced to conform to the patient's skin. This elegant device design permits the delivery of radiation to complex targets using readily available beta-emitting radionuclides, such as Yttrium-90 (Y-90) or Strontium-90 (Sr-90).CONCLUSION:A working prototype of a novel CSBT device was built using 3D-printing technology that provides a safe and economically attractive means of improving radiation delivery to complex treatment sites.

Introduction: To correlate dose from ionizing radiation with cell kill probability, viability and clonogenic assays are often performed following in vitro irradiation of cells in 96-well culture plates. The objective of this work was to examine how dose to adherent cells in 96-well culture plates depended on the conditions in which the irradiations were performed. The aim was to provide support for standardized irradiation setup conditions so that in vitro cell assays can be more precisely compared. Materials and Methods: 96-well culture plates were exposed to a 6 MV photon beam from a linear accelerator. Dose measurements were performed by securing a piece of EBT3 Gafchromic film to the underside of each plate. Each well contained either 200 mu l of tap water or canine osteosarcoma cells (OSCA 40) in 200 mu l of media. Exposures were performed to assess the dependence of dose to the depth of measurement, the thickness of the backscatter material, the field size, and the position of a plate within the treatment field. Results: Culture plate doses demonstrated a strong dependence on backscatter material thickness and field size. Providing less than 0.5 cm of water equivalent backscatter material resulted in a mean measured underdose of more than 3% for a 30 x 30 cm(2) field. Decreasing the field size to 13 x 9 cm(2) yielded a mean measured dose within 0.5% of the prescribed dose. At the central-axis depth where maximum dose occurs (d(max) = 1.5 cm), irradiation of four plates simultaneously resulted in a mean measured dose 3.5% higher than measurements of single plates placed in the center of the field. Conclusions: Inadequate consideration of in vitro irradiation setup conditions is likely to result in discrepancies between prescribed and delivered doses. Such discrepancies can be minimized by providing 5 cm of backscatter material and 10 cm of buildup material; and by exposing one centrally located plate at a time.

3D printing technology has allowed the creation of custom applicators for high dose rate (HDR) brachytherapy, especially for complex anatomy. With conformal therapy comes the need for advanced dosimetric verification. It is important to demonstrate how dose to 3D printed materials can be related to dose to water. This study aimed to determine dose differences and uncertainties using 3D printed PLA and ABS plastics for Radiochromic film calibration in HDR brachytherapy. Gafchromic EBT3 film pieces were irradiated in water with an Ir-192 source at calculated dose levels ranging from 0 to 800 cGy, to create the control calibration curve. Similarly, film was placed below 3D printed PLA and ABS blocks and irradiated at the same dose levels calculated for water, ranging from 0 to 800 cGy. After a 72-h development time, film pieces were scanned on a flatbed scanner and the median pixel value was recorded in the region of highest dose. This value was converted to net optical density (NOD). A rational function was used to fit a calibration curve in water that relates NOD to dose for red, green, and blue color channels. Based on this fitted curve, ABS and PLA NOD values were used to estimate dose in 3D printed plastics. From the fitted calibration curve, mean residual error between measured and planned dose to water was less than 1% for each color channel at high dose levels. At high dose levels, ABS and PLA mean residual errors were about 6.9 and 7.8% in the red channel, while 5.2 and 5.7% in the green channel. Combined uncertainties measured to be about 6.9% at high dose levels. This study demonstrated dose differences and uncertainties using 3D printed applicators for HDR Ir-192 brachytherapy.

Bolus is commonly used in radiation therapy to improve radiation dose distribution to the target volume, but commercially available products do not always conform well to the patient surface. Tumor control may be compromised, particularly for superficial tumors, if bolus does not conform well and air gaps exist between the patient surface and the bolus. Three-dimensional (3D) printing technology allows the creation of highly detailed, variable shaped objects, making it an attractive and affordable option for customized, patient-specific bolus creation. The use of 3D printing in the clinical setting remains limited. Therefore, the objective of this study was to assess the implications on time and clinical fit using a workflow for 3D printing of customized bolus in companion animals with spontaneous tumors treated with radiation therapy. The primary aim of this study was to evaluate the time required to create a clinical 3D printed bolus. The secondary aims were to evaluate the clinical fit of the bolus and to verify the skin surface dose. Time to segmentation and 3D printing were documented, while the clinical fit of the bolus was assessed in comparison to the bolus created in the treatment planner. The mean and median time from segmentation to generation of 3D printed boluses was 6.15 h and 5.25 h, respectively. The 3D printed bolus was significantly less deviated from the planned bolus compared to the conventional bolus (p = 0.0078) with measured dose under the bolus within 5% agreement of expected dose in 88% of the measurements. Clinically acceptable 3D printed customized bolus was successfully created for treatment within one working day. The most significant impact on time is the 3D printing itself, which therefore has minimal implications on personnel and staffing. Quality assurance steps are recommended when implementing a 3D printing workflow to the radiotherapy clinic.

3D printing is not a new concept. The recent advances in printing speed, technology, and material selection are promoting its significant impacts in several industries, including health care. For our medical physics field, researchers are also finding its applications in various clinical aspects. However, the interests still remain in a few academic centers who have the luxuries of owning such an unconventional device in the radiation oncology department, or collaborating with a local 3D printing lab. As the 3D printing technology is becoming an unstoppable driving force in manufacturing revolution, are we also envisioning a future that 3D printing will become as common as a block-cutting machine in a radiation oncology department? In this debate, we invited two researchers who are experienced in studying the clinical use of 3D printing in medical physics field. Dr. Eric Ehler is arguing for the proposition that “3D printing technology will eventually eliminate the need of purchasing commercial phantoms for clinical medical physics QA procedures” and Dr. Daniel Craft is arguing against. Dr. Eric Ehler is an Assistant Professor in the Department of Radiation Oncology at the University of Minnesota. He is the medical physics residency program director at the University of Minnesota Medical Center. His education and research interests are 3D printing, pediatric radiotherapy, radiation dosimetry, and machine learning. Dr. Daniel Craft is currently a medical physics resident at The Mayo Clinic in Phoenix, AZ. Prior to the beginning of his residency, Dr. Craft was a graduate research assistant and PhD student at the University of Texas MD Anderson Cancer Center in Houston Texas, where he studied techniques to deliver postmastectomy radiation therapy using 3D printed patient-specific tissue compensators. He completed his Ph.D. in Medical Physics in May, 2018, and also holds an undergraduate degree in Physics from Brigham Young University. Phantoms provide medical physicists a means to assess the performance of medical devices in imaging, nuclear medicine, and radiation therapy.1 Historically, phantoms were designed and constructed by clinical staff and/or hospital engineers using materials and formulations available to them at the time.2 Currently, many vendors in the medical physics market provide a wide array of phantoms for clinical use. The reason for this shift could reasonably be attributed to convenience and in the interest of standardization of quality check (QC) procedures and quality assurance (QA) programs. 3D printing has been around since 1980s.3, 4 The expiration of patents related to 3D printing has lowered the cost of 3D printers. 3D printing technology has been described as the democratization of manufacturing; 3D printing is shifting the means of manufacture from a centralized system to a distributed network. The impact of increased access to manufacturing capability will reduce the convenience factor of commercial phantoms as clinicians can custom design and print phantoms as needed. The argument “3D printing technology will eventually eliminate the need of purchasing commercial phantoms for clinical medical physics QA procedures” is already becoming reality. In most clinics, the Linac morning QA is performed with a commercial image guidance radiotherapy (IGRT) phantom, which is a cubic phantom with marks on the faces for laser alignment and embedded features for x-ray imaging. An IGRT phantom with submillimeter accuracy was fabricated and reported by Woods et al.5 using computer-aided design freeware and a relatively low cost 3D printer (commercially available for $3150 USD). In our clinic, rather than purchasing multiple identical IGRT phantoms, our team designed our own phantom in a similar manner as Woods et al. The phantom was 3D printed with PET-G plastic for a cost of $10, using a 3D printer in a cost range of $900. The 3D printed phantom did not have the full capabilities of our commercial IGRT phantom but it fits our clinical needs as we did not fully use the features of the commercial phantom during morning QA. Additionally, when compared to a commercial small animal PET/CT imaging phantom, the 3D printed phantom was described as “functionally equivalent to commercially available phantoms”.6 3D printed phantoms have also been described for MRI7 and PET/MRI6 systems. A feature of these phantoms is that they can be customized and produced by the end users at a low cost. 3D printed phantoms have been explored as patient specific phantoms for use in intensity modulated radiotherapy (IMRT) QA,8, 9 vascular imaging,10 and molecular imaging.11 For IMRT QA, 3D printing a patient specific phantom for every patient treated with IMRT is not currently clinically feasible, mostly due to time constraints. However, for commissioning new procedures or for a periodic QA schedule, using a 3D printed phantom is warranted. The use of patient specific phantoms allows for a true end-to-end test on a per-patient basis at reduced cost of commercial, nonpatient specific, anthropomorphic phantoms. Beyond phantoms, 3D printing has been investigated for radiation therapy immobilization devices,12 bolus,13-16 electron blocks,17 and other treatment aids. In fact, the strongest argument for clinical acquisition of 3D printing technology is for the fabrication of treatment devices due to the unique nature of patient anatomy and the high frequency of use of treatment devices. If clinics possess 3D printers for the purpose of treatment device fabrication, the convenience of 3D printing phantoms will increase greatly. A word of caution: 3D printing materials are not tightly controlled by all 3D printing material suppliers. For example, slight differences in formulation of 3D printing materials may affect the radiographic or other physical properties of the material. This variation could arise between one material supplier and another or even from batch to batch of the same supplier. Also, 3D printers can have defects in the printed object such as small unintended air voids or warping during printing. Air voids can occur from imperfect material deposition during the printing. Warping is an issue for fused deposition modeling (FDM) where a plastic filament is melted, extruded out of a nozzle, deposited, and then cools. Cooling can cause contraction, which may cause the FDM 3D printed object to warp. For charged particle radiation beams especially, this can negatively impact the performance of the 3D printed device or phantom.18 Therefore, QC of the manufacturing process will need to be performed by 3D printing staff or clinicians whereas for commercial phantoms, QC is performed by the vendor and verified by the clinicians. For example commercial water equivalent plastic blocks are usually supplied with a certificate stating the physical dimensional accuracy of the plastic, uniformity of the plastic, and the attenuation properties of the plastic. If the blocks are 3D printed by clinic staff, these tests will need to be performed in-house. In summary, I believe there is already a market advantage for the clinical use of 3D printed phantoms. As 3D printers gain use in routine clinical device fabrication, their utilization in other clinical areas, such as phantom fabrication, will expand. In the long term, as 3D printing capabilities increase and 3D printing materials are designed specifically for medical physics use, 3D printed phantoms will increasingly replace commercial phantoms for clinical QA procedures. 3D printing is a transformative technology that allows users to physically manufacture anything that they can model with a computer. Over the last several years there has been enthusiastic and rapid adoption of 3D printing technology in medical physics to create a wide spectrum of custom, patient-specific devices. 3D printers are well-suited to manufacture a number of devices that are currently much more expensive, or much more inconvenient to procure from commercial vendors. These include customized, patient-specific bolus and customized phantoms that may only be used once, or for a single patient. However, despite the interesting applications and enormous potential of 3D printing technology for some radiotherapy applications, presently, there are several limitations that will prevent it from being uniformly adopted as the preferred phantom fabrication technique in hospitals across the country. The first major limitation of 3D printing is the material properties of 3D printed parts. 3D printable materials must have some specific properties; they have to either be a thermoplastic with a glass transition temperature near 200°C, or a photopolymerizing resin. This effectively limits the number of potential materials to thermoplastics and things that can be mixed with them. If a material cannot be melted and turned into a filament, it generally cannot be 3D printed. There are some creative materials that mix in other substances — like wood shavings or copper powder — with thermoplastic bases to create materials with slightly different densities and HU values, but these material differences are mostly cosmetic and intended for hobbyist 3D printing. Importantly, there currently are no commercially available materials that can replicate either bone or lung tissues. Most current 3D printed phantoms either ignore bone entirely8, 19 or use custom in-house mixed materials to mimic bone that requires custom filament creating equipment.20 The first solution reduces the usefulness of the phantom, and the second solution dramatically reduces the convenience that 3D printing was supposed to provide in the first place. Similarly, the lungs are usually left open, or printed with “low infill” that matches lung density but is highly variable depending on the direction of an incident radiation beam.21, 22 Contrast this 3D printed phantom with a common commercial anthropomorphic phantom which comes with several different tissue types, including bone, cartilage, brain, soft tissue, and lung (Computerized Imaging Reference Systems, Inc. A Castleray company, Norfolk, VA). Additionally, these phantoms’ low density material properties do not depend on the direction of incident radiation like low density 3D printed phantoms. Even if a full range of perfectly matched 3D printable materials were to be found, there are still large variations between identical 3D printed parts. We have previously shown that identically printed blocks of material can vary in density from each other up to 7%,23 and that is using the same printer, the same model, and the same roll of filament. There are currently dozens of different kinds of 3D printers in use in clinics around the country using many different materials and printer settings. If 3D printing QA devices becomes commonplace, it will be difficult to make meaningful comparisons of measurements across institutions that are using different 3D printers to produce phantoms based on their own specific materials and printing protocols. Another problem with wide adoption of 3D printing is increased cost and complexity. To be clear, the actual material costs to 3D print a simple phantom are almost certainly less than the cost to purchase a similar commercial phantom. The cost of 3D printers, however, can range anywhere from several hundred dollars to several hundred thousand dollars, with a commensurately huge range in printer complexity, print quality, available features, material compatibility, and reliability. For example, the cheapest 3D printers available on Amazon.com cost less than $200, but can only print using PLA filament, have minimum layer resolutions of approximately 200 microns, and have a build volume of only a few centimeters in any direction. On the other end of the spectrum, the HP Jet Fusion 3D 3200 uses multi-jet fusion technology to dynamically blend plastics to create parts up to 30 cm in each dimension with multiple colors and material properties, and has a minimum layer resolution of 70 microns. However, its cost starts at $155,000. It is important to remember that in-house phantom production will require in-house 3D printing expertise, so will it be the medical physicist's responsibility to be proficient in 3D design as well as the mechanical operation and maintenance of a 3D printer? Whose responsibility will it be if the 3D printer jams during a print and patient QA cannot be performed? 3D printers mostly operate in the background, but they do require operators to plan and start models printing, as well as change out materials and occasionally replace parts. Especially with less expensive printers the user must be able to troubleshoot and fix errors. This may be feasible in larger academic centers, but I do not think it is a reasonable expectation that the many small or nonacademic clinics that make up the majority of cancer care will embrace this unnecessary increased workload. In conclusion, 3D printing is currently not a mature enough technology to become the primary technique for fabricating important QA devices in radiotherapy clinics. Conventionally fabricated commercial phantoms are more uniform, reliable, and simple than 3D printed ones. It is definitely true that 3D printing has a place in radiation oncology — and an exciting one at that! The question that 3D printing must address is: what additional value does it provide over conventional phantom fabrication that outweighs the previously mentioned limitations. In my opinion, that value is in creating highly customized or unique phantoms for research and development in major academic centers, not in creating routine QA phantoms that every clinic needs. I am confident that 3D printing will eventually replace some commercial phantoms for clinical medical physics QA procedures at some clinics, but definitely not for all, or even most of them. I agree with Dr. Craft that currently there are many difficulties to overcome. However, in the long-term view I maintain the argument that all QA phantoms will be fabricated with 3D printing. It is true that currently available 3D printing materials are not equivalent to human tissues. Attributable to the complexity in designing a material that is compatible with 3D printing and is tissue or water equivalent, materials science developments are needed. In the meantime, there is an alternative to fully 3D printing a phantom if it is desired to be tissue or water equivalent. That is to use 3D printing to create a mold to fill with an equivalent material(s); this strategy can be used for phantoms9 as well as radiotherapy bolus.15 This can reduce 3D printing times and bypass deficiencies in the radiologic properties of current 3D printing materials such as those demonstrated by Dr. Craft.23 Regarding 3D printer QA and additional workload, monitoring printers for jams or other print failures can be performed with a software packages such as OctoPrint. The software can be used to monitor printing progress via webcam and, if necessary, the print job can be aborted remotely. Updates on the printing progress can even be sent to mobile devices. To lend perspective on the frequency of print failures, one of our printers (Lulzbot Taz 6) has over 200 print hours with only one failed part in that time while a previously used printer failed quite regularly; thus the choice in the 3D printer is important. In addition, it is true that QA will be required for 3D printed phantoms or devices. However as physicists, we are responsible for the materials and devices used clinically. Regardless of whether a phantom is fabricated in-house or purchased from an established vendor, validation of the phantom and implementation into clinical use is required. There may be additional considerations in the QA of 3D printed phantoms or devices, but the advantages offset the additional workload. Finally, I contest the statement that 3D printing may be feasible for large academic centers but not for smaller clinics. In fact, I believe that the greatest benefit will be to smaller clinics. At a large academic center, there are likely engineers within the hospital and engineering machine shops nearby to fabricate phantoms and devices. Smaller clinics likely lack these resources and 3D printing can fill that gap at a reasonable cost. There are several points upon which Dr. Ehler and I agree. First, and most importantly, we share a concern for some of the variable material properties that 3D printed objects can have. As he notes, different material suppliers are not held to strict material standards, which can lead to various imperfections and inconsistencies in 3D printed parts. Objects printed from different suppliers using an equivalently labeled material could have different densities and radiological properties.23, 24 This is, however, not the only potential source of uncertainty. I would add that the quality of a printed object will depend equally as largely on the 3D printer used, and the model that has been designed. There are many 3D printers with slightly different properties that could affect print quality, such as how stable it can maintain the nozzle and bed temperature, how fast the extruder moves, and many more. Additionally, unless 3D models of useful phantoms are shared across all institutions there will be additional variation between clinics in the actual characteristics of phantoms used for QA. This leads to the second point on which we have common ground: if phantoms are printed in house, calibration and standardization tests into dimensional accuracy, material uniformity, and material attenuation properties will also have to be performed in house. As Dr. Ehler notes, these certifications currently come with phantoms from commercial suppliers. While larger research institutions may have additional resources and time to make this in house testing feasible, having to perform these tests for every printed object is an unnecessary workload for most smaller clinics. This increased workload for physicists in designing objects to be printed, maintaining a 3D printer, and validating 3D printed objects is in my opinion a major limiting factor in the widespread adoption of clinical 3D printing. As Dr. Ehler has mentioned, another use for 3D printing aside from creating clinical phantoms is the creation of patient-specific treatment devices. This is a very interesting application of 3D printing, because many of these devices are currently difficult, time-consuming, or expensive to acquire through conventional fabrication. With 3D printing, however, patient specific bolus13, 15, 25 can be rapidly and inexpensively produced that reduces air gaps and improves dosimetric plan characteristics relative to less conformal bolus. In fact, I agree with Dr. Ehler that “the strongest argument for clinical acquisition of 3D printing technology is for the fabrication of treatment devices.” I disagree, however, with his assertion that this technology can be applied equally to creating phantoms for every clinical need. Although 3D printed bolus is in many ways more convenient than and superior to conventional bolus, 3D printed phantoms are generally harder to manufacture and have inferior material properties relative to conventional phantoms. Ultimately, the debate around 3D printing taking over conventional commercial phantoms is an argument of magnitude. It is clear that 3D printing is currently being used in clinics around the country for a variety of interesting purposes including phantom development,9, 11 treatment device fabrication,13, 16 and more.7, 26, 27 As the technology matures and continues to develop I am sure that it will improve and more use cases will be found. However, it is my opinion that 3D printing will remain a supplemental technology to fabricate a few special things, and will not ever completely replace conventionally fabricated commercial phantoms.

The purpose of this study was to test the feasibility of a patient specific phantom for patient specific dosimetric verification.Using the head and neck region of an anthropomorphic phantom as a substitute for an actual patient, a soft-tissue equivalent model was constructed with the use of a 3D printer. Calculated and measured dose in the anthropomorphic phantom and the 3D printed phantom was compared for a parallel-opposed head and neck field geometry to establish tissue equivalence. A nine-field IMRT plan was constructed and dose verification measurements were performed for the 3D printed phantom as well as traditional standard phantoms.The maximum difference in calculated dose was 1.8% for the parallel-opposed configuration. Passing rates of various dosimetric parameters were compared for the IMRT plan measurements; the 3D printed phantom results showed greater disagreement at superficial depths than other methods.A custom phantom was created using a 3D printer. It was determined that the use of patient specific phantoms to perform dosimetric verification and estimate the dose in the patient is feasible. In addition, end-to-end testing on a per-patient basis was possible with the 3D printed phantom. Further refinement of the phantom construction process is needed for routine use.

We have studied the effect of target and lung density on block margin for small stereotactic body radiotherapy (SBRT) targets. A phantom (50 × 50 × 50cm(3)) was created in the Pinnacle (V9.2) planning system with a 23-cm diameter lung region of interest insert. Diameter targets of 1.6, 2.0, 3.0, and 4.0cm were placed in the lung region of interest and centered at a physical depth of 15cm. Target densities evaluated were 0.1 to 1.0g/cm(3), whereas the surrounding lung density was varied between 0.05 and 0.6g/cm(3). A dose of 100cGy was delivered to the isocenter via a single 6-MV field, and the ratio of the average dose to points defining the lateral edges of the target to the isocenter dose was recorded for each combination. Field margins were varied from none to 1.5cm in 0.25-cm steps. Data obtained in the phantom study were used to predict planning treatment volume (PTV) margins that would match the clinical PTV and isodose prescription for a clinical set of 39 SBRT cases. The average internal target volume (ITV) density was 0.73 ± 0.17, average local lung density was 0.33 ± 0.16, and average ITV diameter was 2.16 ± 0.8cm. The phantom results initially underpredicted PTV margins by 0.35cm. With this offset included in the model, the ratio of predicted-to-clinical PTVs was 1.05 ± 0.32. For a given target and lung density, it was found that treatment margin was insensitive to target diameter, except for the smallest (1.6-cm diameter) target, for which the treatment margin was more sensitive to density changes than the larger targets. We have developed a graphical relationship for block margin as a function of target and lung density, which should save time in the planning phase by shortening the design of PTV margins that can satisfy Radiation Therapy Oncology Group mandated treatment volume ratios.