Linear accelerator maintenance cost analysis.

PURPOSE: Medical linear accelerators are the most costly standard equipment used in radiation oncology, however the service costs for these machines are not well understood. With an increasing demand for linear accelerators due to a global increase in cancer incidence, it is important to understand the expected maintenance costs of a larger global installed base so that these costs can be incorporated into budgeting. The purpose of this investigation is to analyze the costs for medical linear accelerator service and maintenance at our institution, in order to estimate the service cost ratio. METHODS: We collected the costs of parts used for all service work done on 32 medical linear accelerators over a two year period. The data was segregated by center, machine, linear accelerator type, and failure area in the machine. RESULTS: We found the service cost ratio (excluding software support expenses) to be 3.13% [2.74%, 3.52%,]. We observed a variability of parts costs, and overall variability of the service cost ratio to be between 2.14% and 5.25%. This result is lower than other estimates for service costs for medical equipment in general and medical linear accelerators specifically. Two-thirds of the service costs were due to labor costs, which indicate the importance of a well-trained service technician workforce. CONCLUSIONS: We estimated the service cost ratio for medical linear accelerators to be 3.13% [3.52%, 2.74%] of the initial capital cost. This result was lower than other estimates of the service cost ratio.

Measurement of neutron yield for a medical linear accelerator below 10 MV.

BACKGROUND: The recent trend toward 10 MV for volumetric radiotherapy treatment such as volumetric modulated arc therapy (VMAT), stereotactic radiosurgery (SRS), and stereotactic ablative body radiotherapy (SABR) introduces photoneutron production, with implications for non-therapeutic patient dose and additional shielding requirements for treatment room design. The sharply nonlinear drop-off in photoneutron production below 10 MV to negligible at 6 MV has scarcely been characterized quantitatively, yet can elucidate important practical insights. PURPOSE: To measure photoneutron yields in a medical linac at 8 MV, which may strike a reasonable balance between usefully increased beam penetration and dose rate as compared to 6 MV while reducing photoneutron production which is present at 10 MV. METHODS: A Varian iX linear accelerator undergoing decommissioning at our clinic was made to operate over a range of photon energies between 6 and 15 MV by calibrating the bending magnet and adjusting other beam generation parameters. Neutron dose within the treatment room was measured using an Anderson-Braun type detector over a continuum of intermediate energies. RESULTS: The photoneutron production for energies below 10 MV was measured, adding to data that is otherwise scarce in the literature. Our results are consistent with previously published results for neutron yield. We found that the photoneutron production at 8 MV was about 1/10 of the value at 10 MV, and about 10 times higher than detector background at 6 MV. CONCLUSIONS: Photoneutron production drops off below 10 MV, but is still present at 8 MV. An 8 MV beam is more penetrating than a 6 MV beam, and may offer a suitable tradeoff for modern radiotherapy techniques such as VMAT, SRS, and SABR. Further studies are needed to better understand the impact on treatment plan quality between 8 and 10 MV beams considering the benefits to facility requirements and non-therapeutic patient dose.

Topic modeling of maintenance logs for linac failure modes and trends identification.

PURPOSE: Medical linear accelerators (linacs) can fail in a multitude of different manners due to complex structures. An unclear identification of failure modes occurring constantly is a major obstacle to maintenance arrangements, thereby may increasing downtime. This study aims to use natural language processing techniques to deal with the unformatted maintenance logs to identify the linac failure modes and trends over time. MATERIALS AND METHODS: The data used in our study are unformatted narrative maintenance logs recording linac conditions and repair actions. The latent Dirichlet allocation-based topic modeling method was used to identify topics and keywords regarding the failure modes. The temporal analysis method was applied to examine the variation of failure modes over 20 years. RESULTS: Based on the output of the topic modeling, 28 topics and keywords with frequency ranking were generated automatically. The latent failure modes in topics were identified and classified into six main subsystems of linacs. Furthermore, by using the temporal analysis method, the trends of all failure modes over 20 years were illustrated. Half of the topics demonstrated variations with three different patterns, namely periodic, increasing, and decreasing. CONCLUSIONS: The results of our study validated the effectiveness of using the topic modeling method to automatically analyze narrative maintenance logs. With domain knowledge, failure modes of linacs can be identified and categorized quantitatively.

Assessment of the Sun Nuclear ArcCHECK to detect errors in 6MV FFF VMAT delivery of brain SABR using ROC analysis.

Institutions use a range of different detector systems for patient-specific quality assurance (QA) measurements conducted to assure that the dose delivered by a patient's radiotherapy treatment plan matches the calculated dose distribution. However, the ability of different detectors to detect errors from different sources is often unreported. This study contains a systematic evaluation of Sun Nuclear's ArcCHECK in terms of the detectability of potential machine-related treatment errors. The five investigated sources of error were multileaf collimator (MLC) leaf positions, gantry angle, collimator angle, jaw positions, and dose output. The study encompassed the clinical treatment plans of 29 brain cancer patients who received stereotactic ablative radiotherapy (SABR). Six error magnitudes were investigated per source of error. In addition, the Eclipse AAA beam model dosimetric leaf gap (DLG) parameter was varied with four error magnitudes. Error detectability was determined based on the area under the receiver operating characteristic (ROC) curve (AUC). Detectability of DLG errors was good or excellent (AUC >0.8) at an error magnitude of at least ±0.4 mm, while MLC leaf position and gantry angle errors reached good or excellent detectability at error magnitudes of at least 1.0 mm and 0.6°, respectively. Ideal thresholds, that is, gamma passing rates, to maximize sensitivity and specificity ranged from 79.1% to 98.7%. The detectability of collimator angle, jaw position, and dose output errors was poor for all investigated error magnitudes, with an AUC between 0.5 and 0.6. The ArcCHECK device's ability to detect errors from treatment machine-related sources was evaluated, and ideal gamma passing rate thresholds were determined for each source of error. The ArcCHECK was able to detect errors in DLG value, MLC leaf positions, and gantry angle. The ArcCHECK was unable to detect the studied errors in collimator angle, jaw positions, and dose output.

Technique adaptation, strategic replanning, and team learning during implementation of MR-guided brachytherapy for cervical cancer.

PURPOSE: MR-guided brachytherapy (MRgBT) with interstitial needles is associated with improved outcomes in cervical cancer patients. However, there are implementation barriers, including magnetic resonance (MR) access, practitioner familiarity/comfort, and efficiency. This study explores a graded MRgBT implementation strategy that included the adaptive use of needles, strategic use of MR imaging/planning, and team learning. METHODS AND MATERIALS: Twenty patients with cervical cancer were treated with high-dose-rate MRgBT (28 Gy in four fractions, two insertions, daily MR imaging/planning). A tandem/ring applicator alone was used for the first insertion in most patients. Needles were added for the second insertion based on evaluation of the initial dosimetry. An interdisciplinary expert team reviewed and discussed the MR images and treatment plans. RESULTS: Dosimetry-trigger technique adaptation with the addition of needles for the second insertion improved target coverage in all patients with suboptimal dosimetry initially without compromising organ-at-risk (OAR) sparing. Target and OAR planning objectives were achieved in most patients. There were small or no systematic differences in tumor or OAR dosimetry between imaging/planning once per insertion vs. daily and only small random variations. Peer review and discussion of images, contours, and plans promoted learning and process development. CONCLUSIONS: Technique adaptation based on the initial dosimetry is an efficient approach to implementing MRgBT while gaining comfort with the use of needles. MR imaging and planning once per insertion is safe in most patients as long as applicator shifts, and large anatomical changes are excluded. Team learning is essential to building individual and programmatic competencies.

Intermediate dose-volume parameters and the development of late rectal toxicity after MRI-guided brachytherapy for locally advanced cervix cancer.

PURPOSE: The dose delivered to the most exposed 2 cm3 [Formula: see text] of the rectum and bladder is predictive of late rectal and bladder toxicity in cervix cancer patients. We investigated the predictive value of intermediate doses to the rectum and bladder for late rectal/bladder toxicity after MRI-guided brachytherapy for patients with locally advanced cervix cancer. METHODS AND MATERIALS: Toxicity was prospectively graded using Common Toxicity Criteria for Adverse Events v4.0 and retrospectively verified for women with Stage IB-IVA cervix cancer treated consecutively with MRI-guided brachytherapy between 2008 and 2013. The minimum equivalent dose in 2 Gy fractions delivered to 0.1, 1, 2, 5, and 10 cm3 of the rectum and bladder and the absolute volume of the rectum receiving 55, 60, 65, 70, and 75 Gy3 (V55-75) were collected. The association between dose-volume parameters and Grade 2+ rectal/bladder toxicity was examined using logistic regression. RESULTS: With a median followup of 44 months, cumulative incidences of Grade 2+ rectal and bladder toxicity among the 106 patients were 29% and 15% at 3 years, respectively. All the dose-volume parameters were significantly associated with late Grade 2+ rectal and bladder toxicity (p < 0.05), except for bladder [Formula: see text] . On multivariable logistic regression, both [Formula: see text]  > 70 Gy3 and V55 > 11 cm3 (p < 0.05) were predictive of late Grade 2+ rectal toxicity, with improved model fitting and higher area under the curve compared with the model with [Formula: see text]  > 70 Gy3 alone. CONCLUSIONS: In this study, V55 was predictive of late Grade 2+ rectal toxicity. Adding V55 to  [Formula: see text] improved prediction accuracy.

Dosimetric feasibility of ablative dose escalated focal monotherapy with MRI-guided high-dose-rate (HDR) brachytherapy for prostate cancer.

PURPOSE: To determine the dosimetric feasibility of dose-escalated MRI-guided high-dose-rate brachytherapy (HDR-BT) focal monotherapy for prostate cancer (PCa). METHODS: In all patients, GTV was defined with mpMRI, and deformably registered onto post-catheter insertion planning MRI. PTV included the GTV plus 9mm craniocaudal and 5mm in every other direction. In discovery-cohort, plans were obtained for each PTV independently aiming to deliver ⩾16.5Gy/fraction (two fraction schedule) while respecting predefined organs-at-risk (OAR) constraints or halted when achieved equivalent single-dose plan (24Gy). Dosimetric results of original and focal HDR-BT plans were evaluated to develop a planning protocol for the validation-cohort. RESULTS: In discovery-cohort (20-patients, 32-GTVs): PTV D95% ⩾16.5Gy could not be reached in a single plan (3%) and was accomplished (range 16.5-23.8Gy) in 15 GTVs (47%). Single-dose schedule was feasible in 16 (50%) plans. In the validation-cohort (10-patients, 10-GTVs, two separate implants each): plans met acceptable and ideal criteria in 100% and 43-100% respectively. Migration to single-dose treatment schedule was feasible in 7 implants (35%), without relaxing OAR's constraints or increasing the dose (D100% and D35%) to mpMRI-normal prostate (p>0.05). CONCLUSION: Focal ablative dose-escalated radiation is feasible with the proposed protocol. Prospective studies are warranted to determine the clinical outcomes.

Technical Note: Enhancing the surface dose using a weak longitudinal magnetic field.

PURPOSE: The surface dose in radiotherapy is subject to the physical properties of the radiation beam and collimator. The purpose of this work is to investigate the manipulation of surface dose using magnetic fields produced with a resistive magnet. Better understanding of the feasibility and mechanisms of altered surface dose could have important clinical applications where the surface dose must be increased for therapeutic goals, or reduced to enhance the therapeutic benefit. METHODS: A resistive magnet capable of generating a peak magnetic field up to 0.24 T was integrated with a cobalt treatment unit. The magnetic fringe field of the magnet was small due to the self-shielding built within the magnet. The magnetic field at the beam collimation jaws of the cobalt irradiator was less than 10 G. The surface dose and depth dose were measured for varying magnetic field strengths. RESULTS: The resistive magnet was able to alter the dose in the buildup region of the (60)Co depth dose significantly, and the magnitude of dose enhancement was directly related to the strength of the longitudinal magnetic field. Peak magnetic fields as low as 0.08 T were able to affect the surface dose. At a peak field of 0.24 T, the authors measured a surface dose enhancement of 2.8-fold. CONCLUSIONS: Surface dose enhancement using resistive magnets is feasible. Further experimental study is needed to understand the origin of the scattered electrons that contribute to the increase in surface dose.

Simulation of a medical linear accelerator for teaching purposes.

Simulation software for medical linear accelerators that can be used in a teaching environment was developed. The components of linear accelerators were modeled to first order accuracy using analytical expressions taken from the literature. The expressions used constants that were empirically set such that realistic response could be expected. These expressions were programmed in a MATLAB environment with a graphical user interface in order to produce an environment similar to that of linear accelerator service mode. The program was evaluated in a systematic fashion, where parameters affecting the clinical properties of medical linear accelerator beams were adjusted independently, and the effects on beam energy and dose rate recorded. These results confirmed that beam tuning adjustments could be simulated in a simple environment. Further, adjustment of service parameters over a large range was possible, and this allows the demonstration of linear accelerator physics in an environment accessible to both medical physicists and linear accelerator service engineers. In conclusion, a software tool, named SIMAC, was developed to improve the teaching of linear accelerator physics in a simulated environment. SIMAC performed in a similar manner to medical linear accelerators. The authors hope that this tool will be valuable as a teaching tool for medical physicists and linear accelerator service engineers.

A facility for magnetic resonance-guided radiation therapy.

Magnetic resonance (MR) imaging is routinely employed in the design of radiotherapy (RT) treatment plans for many disease sites. It is evident that tighter integration of MR imaging into the RT process would increase confidence in dose placement and facilitate the integration of new MR imaging information (including anatomical and functional imaging) into the therapy process. To this end, a dedicated MR-guided RT (MRgRT) facility has been created that integrates a state-of-the-art linear accelerator delivery system, high-dose rate brachytherapy afterloader, and superconducting MR scanner to allow MR-based online treatment guidance, adaptive replanning, and response monitoring while maintaining the clinical functionality of the existing delivery systems. This system is housed within a dedicated MRgRT suite and operates in a coordinated fashion to assure safe and efficient MRgRT treatments.

ROC analysis in patient specific quality assurance.

PURPOSE: This work investigates the use of receiver operating characteristic (ROC) methods in patient specific IMRT quality assurance (QA) in order to determine unbiased methods to set threshold criteria for γ-distance to agreement measurements. METHODS: A group of 17 prostate plans was delivered as planned while a second group of 17 prostate plans was modified with the introduction of random multileaf collimator (MLC) position errors that are normally distributed with σ ≈ ± 0.5, ± 1.0, ± 2.0, and ± 3.0 mm (a total of 68 modified plans were created). All plans were evaluated using five different γ-criteria. ROC methodology was applied by quantifying the fraction of modified plans reported as "fail" and unmodified plans reported as "pass." RESULTS: γ-based criteria were able to attain nearly 100% sensitivity/specificity in the detection of large random errors (σ > 3 mm). Sensitivity and specificity decrease rapidly for all γ-criteria as the size of error to be detected decreases below 2 mm. Predictive power is null with all criteria used in the detection of small MLC errors (σ < 0.5 mm). Optimal threshold values were established by determining which criteria maximized sensitivity and specificity. For 3%/3 mm γ-criteria, optimal threshold values range from 92% to 99%, whereas for 2%/2 mm, the range was from 77% to 94%. CONCLUSIONS: The optimal threshold values that were determined represent a maximized test sensitivity and specificity and are not subject to any user bias. When applied to the datasets that we studied, our results suggest the use of patient specific QA as a safety tool that can effectively prevent large errors (e.g., σ > 3 mm) as opposed to a tool to improve the quality of IMRT delivery.

A Practice-based Taxonomy for Radiation Treatment Errors.

PURPOSE: An absence of a common language for incident classification limits knowledge sharing within and between organizations in the radiotherapy community. This challenge provided the motivation to develop a clinically relevant taxonomy for radiotherapy errors. MATERIALS AND METHODS: This was a multicenter, prospective study that consisted of three phases: (1) an initial version of the taxonomy was developed based on the World Health Organization Conceptual Framework for the International Classification for Patient Safety and taxonomy models from radiotherapy and other industries; (2) the taxonomy was evaluated using actual incident data from a single practitioner and revised; and (3) face validity testing of the taxonomy was performed by two additional practitioners from different radiotherapy centers using simulated incident cases. RESULTS: The taxonomy consisted of seven classes: incident nature, impact, incident type, stage of origin, stage of discovery, contributing factors, and preventative strategies. Each class was divided into subcategories containing increasingly detailed information. A total of 191 consecutive incidents were classified in phase 2 to ensure no further revision to the taxonomy was required. In phase 3, low interobserver agreement (<60%) was obtained for most classes of the taxonomy in the first face validity test. After revisions were made to the taxonomy based on practitioners' feedback, a second face validity test yielded a high degree of agreement (70%-93%) for all classes. CONCLUSIONS: Our multiphase, iterative approach has yielded a workable and multidimensional set of incident classifiers that can be scaled to accommodate local, regional and discipline-specific requirements. Opportunities exist to implement this taxonomy in institutional and national incident databases to facilitate incident learning within and between institutions.

Compensation of missed fractions without knowledge of radiobiological parameters.

The purpose of this work was to develop simple formulas that can be used to estimate the biologic effect of missed radiotherapy fractions independently of radiobiological parameters. This is achieved by expressing the limits in biologically effective dose for very low or very high radiobiological parameter ratios. Worked examples are given.

Mechanistic formulation of a lineal-quadratic-linear (LQL) model: split-dose experiments and exponentially decaying sources.

PURPOSE: In recent years, several models were proposed that modify the standard linear-quadratic (LQ) model to make the predicted survival curve linear at high doses. Most of these models are purely phenomenological and can only be applied in the particular case of acute doses per fraction. The authors consider a mechanistic formulation of a linear-quadratic-linear (LQL) model in the case of split-dose experiments and exponentially decaying sources. This model provides a comprehensive description of radiation response for arbitrary dose rate and fractionation with only one additional parameter. METHODS: The authors use a compartmental formulation of the LQL model from the literature. They analytically solve the model's differential equations for the case of a split-dose experiment and for an exponentially decaying source. They compare the solutions of the survival fraction with the standard LQ equations and with the lethal-potentially lethal (LPL) model. RESULTS: In the case of the split-dose experiment, the LQL model predicts a recovery ratio as a function of dose per fraction that deviates from the square law of the standard LQ. The survival fraction as a function of time between fractions follows a similar exponential law as the LQ but adds a multiplicative factor to the LQ parameter beta. The LQL solution for the split-dose experiment is very close to the LPL prediction. For the decaying source, the differences between the LQL and the LQ solutions are negligible when the half-life of the source is much larger than the characteristic repair time, which is the clinically relevant case. CONCLUSIONS: The compartmental formulation of the LQL model can be used for arbitrary dose rates and provides a comprehensive description of dose response. When the survival fraction for acute doses is linear for high dose, a deviation of the square law formula of the recovery ratio for split doses is also predicted.

Population TCP estimators in case of heterogeneous irradiation: a new discussion of an old problem.

PURPOSE: To investigate the capacity of two phenomenological expressions to describe the population tumor response in case of a heterogeneous irradiation of the tumor. The generalization of the individual tumor control probability (TCP) models to include the case of a heterogeneous irradiation is a trivial problem. However, an analytical solution that results in a closed form population TCP formula for the heterogeneous case is, unfortunately, a very complex mathematical problem. Therefore we applied a numerical approach to the problem. METHOD: Pseudo-experimental data sets are constructed through the generation of dose distributions and population TCP data obtained by a numerical solution of a multi-dimensional integral over an individual TCP model. The capacity of the following two phenomenological - Poisson and equivalent uniform dose (EUD) based - TCP expressions: [Figure: see text] to describe the population tumor response in case of heterogeneous irradiation is investigated through their fitting to the psuedo-experimental data sets. RESULTS AND CONCLUSIONS. While both expressions produce statistically acceptable fits to the pseudo-experimental data within 2% TCP error band, the use of the second expression is preferable since it produces considerably better fits to the data sets.

Twenty-year review of radiotherapy for vaginal cancer: an institutional experience.

OBJECTIVE: To evaluate clinical outcome, prognostic factors and chronic morbidity with radiotherapy for vaginal cancer treatment. MATERIALS AND METHODS: 68 patients with vaginal cancer treated by radical or adjuvant radiotherapy (RT) were selected. Five with rare subtypes of histopathology and 8 with adenocarcinoma were excluded from this study. 76.4% of the remainder had early-stage diseases (stage I: 14, II: 28, III: 9, and IV: 4). The patients in the years from which they were treated were almost evenly distributed (1st 5 years: 13, 2nd: 14, 3rd: 16, and 4th: 12). There were four treatment groups: external beam radiotherapy (EBRT) alone (n=18), brachytherapy (BT) alone (n=4), EBRT and BT (n=30), and surgery plus RT (n=3). RESULTS: Median follow-up was 50.3 months ranging from 3 to 213 months. 5-year overall survival (OS) was 55.6%, disease-specific survival (DSS) was 77.3%, disease-free survival was 74.2%, and local control was 87.7%. Independent prognostic factors for DSS and OS were tumor stage, site and size (p<0.05). Late radiation toxicity was minimal in the bladder (4.6%) and bowel (4.6%). Vaginal morbidity was observed in 35 patients (63.6%). It was lowest in the BT alone (0%), and highest in the EBRT and BT group (82.1%), especially for those received more than 70 Gy (p=0.05, Odds ratio=4.64, 95% confidence interval: 1.01-21.65). CONCLUSION: This retrospective review suggested that tumor stage, site, and size were important prognostic factors in patients with vaginal cancer. Higher radiation dose was associated with more frequent vaginal toxicity.