Guided Radiation Therapy Research Articles

AUDIT TO EVALUATE CURRENT RADIOTHERAPY (RT) PLANNING PATHWAY: COULD CURRENT PRACTICE BE IMPROVED BY THE TIMING OF MAGNETIC RESONANCE IMAGING (MRI)?

Abstract AIMS To evaluate the current guidance for glioblastoma (GBM) and primary brain radiotherapy (RT) planning. Current practice in this centre utilises pre and post-surgical MRI for RT target delineation. This study aims to explore the practicality and impact of scheduling an RT planning-specific MRI no more than 2 weeks before RT treatment start date (TSD). Time from surgery to RT TSD will also be explored. METHOD Fifty GBM patients, treated between August 2022 and October 2023, were audited retrospectively. Included patients were treated with radical intent and must have had surgery or biopsy. Age, gender or performance status (PS) were excluded from audit criteria. Age of MRI used for RT planning and surgery to TSD was noted. RESULTS Early results showed that the average age of MRI used for RT planning was 7 weeks old and Surgery to TSD was 7 weeks. A new pathway was introduced after 37 patients as these results were considered unacceptable. The new pathway included 1 reserved MRI appointment and one reserved RT Mask/CT appointment per week. 13 patients’ pathways were audited after introduction of the new pathway. The results showed that the average age of MRI used for RT planning was 2 weeks old and surgery to TSD was 6 weeks. CONCLUSION Limitations of the audit included disparity between number of patients before and after introduction of the new pathway. This was because of the time limitation for completion of the study. Improvement was seen in timing of MRI to TSD. However, the small improvement in timing of surgery to TSD was impacted by human factors rather than organisational influence, for example, surgical deconditioning, delays for wound healing and patient choice. This study has shown that the ideal timing of MRI can be achieved.

Assessing tumor microstructure with time-dependent diffusion imaging: Considerations and feasibility on clinical MRI and MRI-Linac.

Quantitative imaging biomarkers (QIBs) can characterize tumor heterogeneity and provide information for biological guidance in radiotherapy (RT). Time-dependent diffusion MRI (TDD-MRI) derived parameters are promising QIBs, as they describe tissue microstructure with more specificity than traditional diffusion-weighted MRI (DW-MRI). Specifically, TDD-MRI can provide information about both restricted diffusion and diffusional exchange, which are the two time-dependent effects affecting diffusion in tissue, and relevant in tumors. However, exhaustive modeling of both effects can require long acquisitions and complex model fitting. Furthermore, several introduced TDD-MRI measurements can require high gradient strengths and/or complex gradient waveforms that are possibly not available in RT settings. In this study, we investigated the feasibility of a simple analysis framework for the detection of restricted diffusion and diffusional exchange effects in the TDD-MRI signal. To promote the clinical applicability, we use standard gradient waveforms on a conventional 1.5 T MRI system with moderate gradient strength (Gmax=45 mT/m), and on a hybrid 1.5 T MRI-Linac system with low gradient strength (Gmax=15 mT/m). Restricted diffusion and diffusional exchange were simulated in geometries mimicking tumor microstructure to investigate the DW-MRI signal behavior and to determine optimal experimental parameters. TDD-MRI was implemented using pulsed field gradient spin echo with the optimized parameters on a conventional MRI system and a MRI-Linac. Experiments in green asparagus and 10 patients with brain lesions were performed to evaluate the time-dependent diffusion (TDD) contrast in the source DW-images. Simulations demonstrated how the TDD contrast was able to differentiate only dominating diffusional exchange in smaller cells from dominating restricted diffusion in larger cells. The maximal TDD contrast in simulations with typical cancer cell sizes and in asparagus measurements exceeded 5% on the conventional MRI but remained below 5% on the MRI-Linac. In particular, the simulated TDD contrast in typical cancer cell sizes (r=5-10µm) remained below or around 2% with the MRI-Linac gradient strength. In patients measured with the conventional MRI, we found sub-regions reflecting either dominating restricted diffusion or dominating diffusional exchange in and around brain lesions compared to the noisy appearing white matter. On the conventional MRI system, the TDD contrast maps showed consistent tumor sub-regions indicating different dominating TDD effects, potentially providing information on the spatial tumor heterogeneity. On the MRI-Linac, the available TDD contrast measured in asparagus showed the same trends as with the conventional MRI but remained close to typical measurement noise levels when simulated in common cancer cell sizes. On conventional MRI systems with moderate gradient strengths, the TDD contrast could potentially be used as a tool to identify which time-dependent effects to include when choosing a biophysical model for more specific tumor characterization.

Assessing Surface Guided Radiation Therapy Benefits for Paediatric Cancer Patients: Dosimetric Implications of Intrafractional Motion - An Institutional Review

<i>Introduction/Background</i>: SGRT, a real-time imaging technique, offers continuous monitoring and motion control during <i>treatment</i>. The investigation aims to assess potential dosimetric alterations in target coverage due to intrafractional motion, considering its impact on patient safety and treatment efficiency. <i>Materials and Methods</i>: A retrospective chart review was conducted to assess intrafractional shifts in 18 paediatric cancer patients. Patient setup employed SGRT using AlignRT (Vision RT Ltd., UK), and the PTV was aligned with CBCT. The study introduced induced shifts of 3 mm, 5 mm, and 7 mm during treatment delivery, assessing their impact on portal dosimetry results for both treatment fields. The gamma index criteria (3%, 3 mm) were employed to evaluate dosimetric accuracy. <i>Results</i>: A total of 18 patients were included, and induced shifts were analyzed for their impact on the planned gamma index values. Significant differences were observed between the Planned Gamma Index and induced shifts of 3 mm, 5 mm, and 7 mm for both treatment fields, highlighting the dosimetric implications of intrafractional motion in paediatric cases. <i>Conclusion</i>: Surface Guided Radiation Therapy (SGRT) is concluded to offer a comprehensive array of benefits for paediatric cases. The dosimetric implications of induced shifts underscore the importance of SGRT in ensuring accurate and safe treatment for paediatric cancer patients.

High-speed bioimpedance-based gating system for radiotherapy: Prototype and proof of principle.

To investigate a novel bioimpedance-based respiratory gating system (BRGS) designed for external beam radiotherapy and to evaluate its technical characteristics in comparison with existing similar systems. The BRGS was tested on three healthy volunteers in free breathing and breath-hold patterns under laboratory conditions. Its parameters, including the time delay (TD) between the actual impedance change and the gating signal, temperature drift, root mean square (RMS) noise, and signal-to-noise ratio (SNR), were measured and analyzed. The gate-on TD and the gate-off TD were found to be 9.0±2.0ms [mean±standard deviation (M±SD)] and 7.2±1.3ms, respectively. The temperature drift of the BRGS output signal was 0.02 Ω after 30 min of operation. RMS noise averaged 0.14±0.05 Ω (M±SD) for all subjects and varied from 0.08 to 0.20 Ω with repeated measurements. A significant difference in SNR (p<0.001) was observed between subjects, ranging from 4 to 15. The evaluated bioimpedance-based gating system showed a high performance in real-time respiratory monitoring and may potentially be used as an external surrogate guidance for respiratory-gated external beam radiotherapy. Direct comparison with commercially available systems, 4D correlation studies, and expansion of the patient sample are goals for future preclinical studies.

Clinical and Dosimetric Comparison Between Non-image Guided Radiation Therapy and Fiducial-Based Image Guided Radiation Therapy With or Without Reduced Margin in Intensity Modulated Radiation Therapy for Prostate Cancer

PurposeThis study aimed to compare the outcomes and toxicities between patients treated with image-guided radiation therapy (IGRT) using fiducial markers and non-IGRT in intensity-modulated radiation therapy (IMRT) for prostate cancer. Methods and MaterialsIn total, 518 patients with intermediate- and high-risk prostate cancer received IMRT with 78 Gy in 39 fractions after neoadjuvant androgen deprivation therapy for at least 3 months. Of these patients, 371 were in the non-IGRT group and 147 in the IGRT group, including the IGRT-A group using the same margins as the non-IGRT group and the IGRT-B group using reduced margins. The median follow-up periods for the non-IGRT, IGRT-A, and IGRT-B groups were 99 months, 88 months, and 63 months, respectively. Results: The 5-year biochemical recurrence-free survival rates in the non-IGRT, IGRT-A, and IGRT-B groups were 88%, 95%, and 98% (non-IGRT vs. IGRT-A: p = 0.396, IGRT-A vs. IGRT-B: p = 0.426), respectively. Those for intermediate- and high-risk patients were 94%, 93%, and 96% (non-IGRT vs. IGRT-A: p = 0.916, IGRT-A vs. IGRT-B: p = 0.646) and 87%, 96%, and 100% (non-IGRT vs. IGRT-A: p = 0.500, IGRT-A vs. IGRT-B: p = 0.483), respectively. For the non-IGRT and IGRT-A groups, the rates of acute grade ≥ 2 gastrointestinal (GI) toxicities and late grade ≥ 2 genitourinary (GU) toxicities were 17% and 7% (p = 0.019); and 28% and 16% (p = 0.028), respectively. In the IGRT-A and IGRT-B groups, the rates of acute grade ≥ 2 GU toxicities were 45% and 21% (p = 0.003). All V60Gy and V70Gy values of the bladder and rectal walls in the IGRT-B group were smaller than those in the IGRT-A group. Conclusions: IGRT with fiducial markers results in lower acute and late toxicities compared with non-IGRT in IMRT for intermediate- and high-risk prostate cancer. Moreover, the toxicities are further decreased by reducing the margins in the treatment planning under IGRT. These processes do not decrease the biochemical recurrence-free survival rates.

Optimal Correction Strategy of Image Guided Radiation Therapy Including the Paraortic Lymph Node Region in Patients With Cervical Cancers

The clinically accepted planning target volume margin for radiation therapy to the paraortic nodal region in cervical cancer patients is 5 mm. However, the comprehensive alignment and variability from the pelvic bone to all lumbar vertebrae are undetermined. This study aims to quantify the residual setup errors between the pelvic bone and lumbar vertebrae and determine the optimal correction strategy for patients with cervical cancer. Fifteen patients underwent pretreatment mega-voltage computed tomography scans (375 total fractions). Residual setup errors and required margins for each lumbar vertebra were calculated based on registrations accounting for pelvic rotation and translation. The systematic residual errors (1 SD) at L1, L2, L3, L4, and L5 using pelvic bone registration were 6.5, 4.9, 3.1, 1.5, and 0.6 mm in the anterior-posterior (AP) direction, 3.1, 2.3, 1.4, 0.6, and 0.3 mm in the right-left direction, and 2.7, 2.2, 1.7, 1.0, and 0.5 mm in the superior-inferior direction, respectively. The residual setup errors were the largest in the AP direction. Registration based on the pelvic bone required margins in the AP direction of 16.0, 12.1, 7.7, 3.6, and 1.3 mm for L1, L2, L3, L4, and L5, respectively, whereas registration based on L3 required margins of 8.8, 4.8, 4.4, 7.1, and 7.7 mm for L1, L2, L4, L5, and pelvic bone, respectively. Considerable local setup variability was found in patients with cervical cancer. After reviewing the corrective strategies, we determined that L3-based registration effectively minimized the required margins.

Magnetic Resonance Imaging Guided Radiation Therapy for Splenomegaly: Clinical Experiences and Technical Tips

IntroductionSplenomegaly is a common manifestation in chronic lymphoid and myeloid malignancies. While splenectomy is the preferred treatment for symptomatic splenomegaly, it carries significant risks. Radiotherapy (RT) has traditionally been considered a palliative option. This study explores the use of Magnetic Resonance-Guided Radiotherapy (MRgRT) for splenic irradiation (SI) in patients with myelofibrosis (MFI) and myelodysplastic/myeloproliferative neoplasms (MDS/MPN). MethodsThis single-center retrospective analysis includes patients with MFI and MDS/MPN who underwent MRgRT SI between 2018 and 2022. Ten 1 Gy fractions were delivered to the PTV (spleen+3/5mm margin). An adaptive online/offline strategy has been used to reduce the dose to healthy organs. Dosimetric data and clinical outcomes, including pain relief, gastrointestinal symptoms, and hematological values, were assessed. ResultsTwelve patients completed SI without interruption, with supportive transfusions as needed for cytopenias. Pain and gastrointestinal symptoms relief was observed in most cases. The mean percentage reduction in spleen volume was 53.61%, with an average cranio-caudal extension reduction of 77.78%. Twenty-nine (24.2%) out of 120 fractions were online adapted and 14 (11.7%) were replanned offline. Non-hematological toxicities were not reported. At a median follow-up of 12.9 months, 6 patients died, while 9 patients underwent hematopoietic cell transplantation, with 6 of them surviving. ConclusionThis study demonstrates MRgRT SI feasibility in MFI and MDS/MPN patients, offering symptom relief and significant spleen volume reduction. Real-time setup verification and adaptive planning allowed for tailored treatment with reduced margins, minimizing healthy tissue exposure. Larger prospective studies with longer follow-up are needed to further validate its efficacy and safety.

Development and first implementation of a novel multi-modality cardiac motion and dosimetry phantom for radiotherapy applications.

Cardiac applications in radiation therapy are rapidly expanding including magnetic resonance guided radiation therapy (MRgRT) for real-time gating for targeting and avoidance near the heart or treating ventricular tachycardia (VT). This work describes the development and implementation of a novel multi-modality and magnetic resonance (MR)-compatible cardiac phantom. The patient-informed 3D model was derived from manual contouring of a contrast-enhanced Coronary Computed Tomography Angiography scan, exported as a Stereolithography model, then post-processed to simulate female heart with an average volume. The model was 3D-printed using Elastic50A to provide MR contrast to water background. Two rigid acrylic modules containing cardiac structures were designed and assembled, retrofitting to an MR-safe programmable motor to supply cardiac and respiratory motion in superior-inferior directions. One module contained a cavity for an ion chamber (IC), and the other was equipped with multiple interchangeable cavities for plastic scintillation detectors (PSDs). Images were acquired on a 0.35 T MR-linac for validation of phantom geometry, motion, and simulated online treatment planning and delivery. Three motion profiles were prescribed: patient-derived cardiac (sine waveform, 4.3mm peak-to-peak, 60 beats/min), respiratory (cos4 waveform, 30mm peak-to-peak, 12 breaths/min), and a superposition of cardiac (sine waveform, 4mm peak-to-peak, 70 beats/min) and respiratory (cos4 waveform, 24mm peak-to-peak, 12 breaths/min). The amplitude of the motion profiles was evaluated from sagittal cine images at eight frames/s with a resolution of 2.4mm × 2.4mm. Gated dosimetry experiments were performed using the two module configurations for calculating dose relative to stationary. A CT-based VT treatment plan was delivered twice under cone-beam CT guidance and cumulative stationary doses to multi-point PSDs were evaluated. No artifacts were observed on any images acquired during phantom operation. Phantom excursions measured 49.3±25.8%/66.9±14.0%, 97.0±2.2%/96.4±1.7%, and 90.4±4.8%/89.3±3.5% of prescription for cardiac, respiratory, and cardio-respiratory motion profiles for the 2-chamber (PSD) and 12-substructure (IC) phantom modules respectively. In the gated experiments, the cumulative dose was<2% from expected using the IC module. Real-time dose measured for the PSDs at 10Hz acquisition rate demonstrated the ability to detect the dosimetric consequences of cardiac, respiratory, and cardio-respiratory motion when sampling of different locations during a single delivery, and the stability of our phantom dosimetric results over repeated cycles for the high dose and high gradient regions. For the VT delivery, high dose PSD was<1% from expected (5-6 cGy deviation of 5.9Gy/fraction) and high gradient/low dose regions had deviations<3.6% (6.3 cGy less than expected 1.73Gy/fraction). A novel multi-modality modular heart phantom was designed, constructed, and used for gated radiotherapy experiments on a 0.35 T MR-linac. Our phantom was capable of mimicking cardiac, cardio-respiratory, and respiratory motion while performing dosimetric evaluations of gated procedures using IC and PSD configurations. Time-resolved PSDs with small sensitive volumes appear promising for low-amplitude/high-frequency motion and multi-point data acquisition for advanced dosimetric capabilities. Illustrating VT planning and delivery further expands our phantom to address the unmet needs of cardiac applications in radiotherapy.

Quantifying societal burden of radiation-induced small bowel toxicity in patients with rectal cancer.

Advancements in rectal cancer (RC) treatment not only led to an increase in lives saved but also improved quality of life (QoL). Notwithstanding these benefits, RC treatment comes at the price of gastrointestinal morbidity in many patients. Health economic modelling poses an opportunity to explore the societal burden of such side-effects. This study aims to quantify radiation-induced late small bowel (SB) toxicity in survivors of RC for Three-Dimensional Conformal Radiation Therapy (3D-CRT), Intensity Modulated Radiation Therapy (IMRT) and Intensity Modulated Radiation Therapy - Image Guided Radiation Therapy (IMRT/IGRT). Materials and A model-based health economic evaluation was performed. The theoretical cohort consists of a case-mix of survivors of RC aged 25-99 years according to Belgian age-specific incidence rates. A societal perspective was adopted. The base case analysis was complemented with one-way deterministic analyses, deterministic scenario analyses and probabilistic sensitivity analysis (1,000 iterations). Results were presented as mean lifetime incremental cost (€) and utility (QALYs) per patient. The analyses showed that the use of innovative radiotherapy (RT) improves lifetime QoL in survivors of RC by 0.11 QALYs and 0.05 QALYs by preferring IMRT/IGRT and IMRT over 3D-CRT, respectively. The use of IMRT/IGRT and IMRT results in an incremental cost-saving of €3,820 and €1,863 per patient, solely by radiation-induced SB toxicity, compared to 3D-CRT. It is important to consider late toxicity effects in decisions regarding investments and reimbursement as our analysis highlighted the potential long-term cost-savings and improved QoL of novel RT techniques in patients with rectal cancer.

Sensitivity Response Analysis of Optical Surface Monitoring Systems Using the Fitzpatrick Scale: A Phantom Study

PurposeOptical Surface Monitoring systems (OSMS) have gained substantial attention in modern radiotherapy, specifically in the context of Surface Guided Radiation Therapy (SGRT) which offers real-time patient surface monitoring, ensuring accurate and effective radiotherapy treatments. The aim of this paper is to evaluate the OSMS camera sensitivity towards different skin tones, categorized according to the Fitzpatrick scale, a universal classification of human skin tones, using a phantom. Methods and MaterialThe study utilized Catalyst and SentinelTM OSMS systems (C-Rad, Uppsala, Sweden). The Alderson RANDO Female Pelvis phantom, located at the isocenter in CT simulation and treatment rooms, served as an experimental subject. Eighteen skin tone matching cotton cloths, selected based on Von Luschan's chromatic and Fitzpatrick scales, were wrapped around the phantom for sensitivity evaluation. Camera sensitivity was optimized by adjusting threshold/gain (100-600%) and integration time during individual scans in both rooms. Temporal response analysis spanned two months, with sixteen measurements for each OSMS taken in varying light conditions. ResultsThe OSMS systems successfully detected the surface of cloth-covered phantoms with varying mean (standard deviation) integration times: 550 (34) to 950 (43) μs for the Sentinel system and 2300 (71) to12000 (400) μs for the Catalyst system. The sensitivity parameters differed for each skin tone, with lighter skin requiring shorter integration times and gain/threshold values. Darker skin tones necessitated higher parameters for optimal surface images. The reliability of the systems declined with excessive parameters, leading to noise and compromised accuracy in patient positioning. ConclusionsOptimized sensitivity parameters tailored to individual skin tones are crucial for effective real-time patient surface monitoring in radiation therapy, as variations in skin colour can affect the accuracy of measurements. The precision of skin colour measurements in OSMS relies on carefully adjusting camera sensitivity parameters. However, careful consideration is essential, as larger values are required for darker skin tones, compromising reliability. This suggests the need for exploring alternative image guidance methods for patients with darker skin tones.

Improving motion management in radiation therapy: findings from a workshop and survey in Australia and New Zealand

Motion management has become an integral part of radiation therapy. Multiple approaches to motion management have been reported in the literature. To allow the sharing of experiences on current practice and emerging technology, the University of Sydney and the New South Wales/Australian Capital Territory branch of the Australasian College of Physical Scientists and Engineers in Medicine (ACPSEM) held a two-day motion management workshop. To inform the workshop program, participants were invited to complete a survey prior to the workshop on current use of motion management techniques and their opinion on the effectiveness of each approach. A post-workshop survey was also conducted, designed to capture changes in opinion as a result of workshop participation. The online workshop was the most well attended ever hosted by the ACPSEM, with over 300 participants and a response to the pre-workshop survey was received from at least 60% of the radiation therapy centres in Australia and New Zealand. Motion management is extensively used in the region with use of deep inspiration breath-hold (DIBH) reported by 98% of centres for left-sided breast treatments and 91% for at least some right-sided breast treatments. Surface guided radiation therapy (SGRT) was the most popular session at the workshop and survey results showed that the use of SGRT is likely to increase. The workshop provided an excellent opportunity for the exchange of knowledge and experience, with most survey respondents indicating that their participation would lead to improvements in the quality of delivery of treatments at their centres.

Evaluation of the offset couch parameter between kilovoltage on-board imaging and cone-beam computed tomography in patients with prostate cancer

Background: External Beam Radiation Therapy (EBRT) is a curative therapy technique for prostate cancer. Since the prostate is unstable and surrounded by the bladder and rectum, precision of the target location is critical. Image Guided Radiation Therapy (IGRT) can improve treatment precision. The bladder and rectum may alter volume during IGRT, shifting the prostate’s position and resulting in missed target volume doses and extra organs at risk (OARs) doses. Objective: To assess setup error and residual error during patient positioning, as well as the current IGRT protocol efficiency, in prostate cancer patients while recommending a planning target volume (PTV) margin. Materials and methods: The offset couch parameter of on-board imaging (OBI) and cone-beam computed tomography (CBCT) was computed to determine the error distribution, magnitude, and error difference between treatment phases. The systematic and random errors were calculated using the van Herk equation to determine the planning target volume (PTV) margin. Results: The setup error was -0.86 to 0.25 mm, and the residual error was -0.15 to 0.32 mm. The couch displacement percentage for OBI was 29.44% to 58.89%, and for CBCT was 8.10% to 34.12%. The systematic error was 1.65 to 3.21. The random error was 1.78 to 3.29. The setup error was greatest in the longitudinal (Lng) direction, residual error was greatest in the vertical (Vrt) direction, and systematic and random error were greatest in the Vrt and lateral (Lat) direction, respectively. The PTV margin was greatest in the Vrt direction, while the Lng direction was the narrowest margin for every treatment phase. Conclusion: The highest setup error occurs in the Lng direction for all treatment phases. For the 46 Gy and 60 Gy phases, the highest residual error is in the Vrt direction. However, in the 78 Gy phase, the error is relatively close to 0.01mm in every direction. The current IGRT protocol is effective in detecting setup and residual errors. The 78 Gy phase has the greatest PTV margin, whereas the 46 Gy phase shows the narrowest margins in all directions.

On Patient Experience and Anxiety During Treatment With Magnetic Resonance–Guided Radiation Therapy

Purpose: To assess patient experience and anxiety during MR-guided radiotherapy (MRgRT) on a 1.5 Tesla (T) MR-Linac when offered calming video content. Methods and Materials: A single center study was conducted within the Multi-OutcoMe EvaluatioN of radiation Therapy Using the MR-Linac (MOMENTUM) cohort. Patients were offered to watch calming video content on a video monitor during treatment. Questionnaires were used to assess patient experience (MRL-PE) and anxiety (State Trait Anxiety Inventory, STAI) at first treatment fraction (M1) and at third, fourth or fifth treatment fraction (M2). Paired T-tests were used to test for significant differences, and effect sizes (ES) to determine the importance of the difference. Results: Between November 2021 and November 2022, 66 patients were included. The majority was male (n=59, 89%). MRgRT was most frequently delivered to prostate cancer (n=45, 68%) followed by a lesion in the pancreas (n=8, 12%). At M1 and M2, 24/59 patients (41%) preferred to watch calming video content. One patient was not able to look at the video-monitor comfortably at first fraction. Patient experience was generally favorable or neutral; tingling sensations were reported by 17% of patients. Anxiety levels were high (16%), moderate (18%) or low to none (67%) prior to M1. STAI-scores were 33 (SD 9) prior to M1 and 29 (SD 7) after M1 (ES 0.7, p<0.001). STAI-scores were 32 (SD 9) prior to M2 and 31 (SD 8) (ES 0.4, p=0.009). Conclusions: Patients were able to comfortably view the video monitor during MRgRT. Consequently, this set up could be used for future applications, such as biofeedback. A sizeable minority of patients preferred to watch calming video's, which distracted them during treatment. Although the patient's experience was overall excellent, anxiety was reported. Anxiety levels were highest prior to treatment, which decreased after treatment.

Abstract PO4-07-01: Deep learning guided radiation therapy machine selection via a multi-headed dose prediction model

Abstract Background: Accelerated Partial Breast Irradiation (APBI) is gaining popularity for the treatment of appropriately selected breast cancer patients in lieu of whole breast irradiation regimens due to its comparable clinical outcomes and improved patient convenience and quality-of-life. Currently, several machines for delivery of APBI are available such as CyberKnife, Elekta Unity, Varian Ethos and Varian TrueBeam. Each treatment machine yields a unique distribution of radiation dose, so a given modality may afford particular advantages or liabilities based on a patient’s anatomy. Existing methods for identification of an ideal modality that best suits a patient’s unique anatomical features are generally lacking, manual, and resource intensive. To overcome this limitation and personalize modality selection, we propose a deep learning (DL) based multi-modality dose prediction model. The model takes a patient’s computed tomography (CT) scan, segmented planning target volumes (PTVs) and segmented organs at risk (OARs) as input, and outputs predicted dose distributions for all trained modalities. Differences may then be evaluated between each radiation treatment modality’s corresponding dose distribution and dose volumetric histograms (DVH) to select the optimal modality for each patient. Methods: Our dataset contains 16 partial breast patients, each with 4 unique treatment plans generated for 4 distinct modalities – CyberKnife, Elekta Unity, Varian Ethos and Varian Truebeam. 10, 2, and 4 patients were dedicated to the train, validation and test datasets respectively. Our model’s architecture is a U-net with a shared encoder and 4 independent dose decoders, in which each decoder predicts dose for one modality. Each training patient’s CT scan, PTVs, OARs, and a PTV distance map are input to the model during training, and the shared encoder shares the input data’s latent representations with each of the 4 decoders. The 4 distinct dose distributions from each training patient are used as labels for the 4 corresponding decoders to supervise model training. The mean squared error, mean squared log error, and DVH losses are minimized during model training. The model trained until validation loss converged. The 4 dose distributions were predicted in ~ 1 minute per test patient, yielding 16 dose distributions in the test dataset. Model performance across the test dataset was evaluated by calculation of the mean absolute percent error (MAPE) of dose differences within the patient’s body between the model’s predicted doses and label doses, relative to patient prescription dose. Results: The performance of the multi-modal dose prediction model was evaluated in terms of dose differences throughout each patient’s body across the 16 test dose distributions as compared to the actual planned doses. The average MAPE of the multi-modal dose predictions was 1.27 ± 0.38% of a patient’s prescription dose as compared to the actual planned dose across the test dataset. For a prescription dose of 3000 cGy, that corresponds to an average MAPE of 38.1 ± 11.4 cGy. Conclusion: Our multi-modality dose prediction model’s novelty lies in its rapid provision and comparison of accurate and detailed dosimetric data for each treatment modality. This model can empower treating radiation oncologists to make informed decisions in radiation therapy machine selection on a patient-by-patient basis, potentially optimizing radiation treatment plans by reduction of delivered dose to organs at risk. Table. Citation Format: Austen Maniscalco, Ezek Mathew, David Parsons, Asal Rahimi, Mona Arbab, Prasanna Alluri, Xingzhe Li, Mu-Han Lin, Steve Jiang, Dan Nguyen. Deep learning guided radiation therapy machine selection via a multi-headed dose prediction model [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO4-07-01.

3448: Overview of guidance and recommendations for skin radiotherapy

3448: Overview of guidance and recommendations for skin radiotherapy

2193: Impact of Surface Guided Radiation Therapy (SGRT) on treatment delivery in Pediatric patients.

2193: Impact of Surface Guided Radiation Therapy (SGRT) on treatment delivery in Pediatric patients.

Immediate lymphatic reconstruction: Lessons learned over eight years

Immediate lymphatic reconstruction (ILR) is recognized as a surgical approach used to reduce the risk of developing secondary lymphedema, and evidence demonstrating the efficacy of ILR is favorable. Our Lymphatic Center has become a centralized location offering ILR for the risk-reduction in breast cancer-related lymphedema (BCRL) in New England. Over the course of our experience, we made several modifications and adapted our approach to enhance the operative success of this procedure. These include advancements in our use of indocyanine green (ICG) imaging to identify baseline lymphatic anatomical variation, utilization of fluorescein isothiocyanate for lymphatic vessel visualization, application of the lymphosome concept to guide arm injection sites, verification of anastomotic patency (using ICG), localization of reconstruction to guide radiation therapy, incorporation of intraoperative tools to facilitate better anatomic visualization of the axilla, and addition of a lower extremity vein graft to mitigate venous-related complications. Collecting information from each surgery in a standardized manner, including intraoperative lymphatic channel measurements, and deploying clips for possible future radiation exposure, enables future studies on ILR patient outcomes. In this contribution, we aimed to share our institutional modifications with the surgical community to facilitate further adoption, conversation, and advancement of ILR for the risk-reduction in BCRL.

The Utility of Spectroscopic MRI in Stereotactic Biopsy and Radiotherapy Guidance in Newly Diagnosed Glioblastoma.

Current diagnostic and therapeutic approaches for gliomas have limitations hindering survival outcomes. We propose spectroscopic magnetic resonance imaging as an adjunct to standard MRI to bridge these gaps. Spectroscopic MRI is a volumetric MRI technique capable of identifying tumor infiltration based on its elevated choline (Cho) and decreased N-acetylaspartate (NAA). We present the clinical translatability of spectroscopic imaging with a Cho/NAA ≥ 5x threshold for delineating a biopsy target in a patient diagnosed with non-enhancing glioma. Then, we describe the relationship between the undertreated tumor detected with metabolite imaging and overall survival (OS) from a pilot study of newly diagnosed GBM patients treated with belinostat and chemoradiation. Each cohort (control and belinostat) were split into subgroups using the median difference between pre-radiotherapy Cho/NAA ≥ 2x and the treated T1-weighted contrast-enhanced (T1w-CE) volume. We used the Kaplan-Meier estimator to calculate median OS for each subgroup. The median OS was 14.4 months when the difference between Cho/NAA ≥ 2x and T1w-CE volumes was higher than the median compared with 34.3 months when this difference was lower than the median. The T1w-CE volumes were similar in both subgroups. We find that patients who had lower volumes of undertreated tumors detected via spectroscopy had better survival outcomes.

Characterization of the IDENTIFYTM surface scanning system for radiation therapy setup on a closed-bore linac.

In radiation therapy, surface guidance can be used for patient setup and intra-fraction motion monitoring. The surface guided radiation therapy (SGRT) system from Varian Medical systems, IDENTIFYTM, consists of three pods, including cameras and a random pattern projector, mounted on the ceiling. The information captured by the cameras is used to make a reconstruction of the surface. The aim of the study was to assess the technical performance of this SGRT system on a closed-bore linac. Phantom measurements were performed to assess the accuracy, precision, reproducibility and temporal stability of the system, both in align and in load position. Translations of the phantoms in lateral, longitudinal, and vertical direction, and rotations around three axes (pitch, roll and yaw) were performed with an accurate, in-house built, positioning stage. Different phantom geometries and different surface colors were used, and various ambient light intensities were tested. The accuracy of the IDENTIFYTM system at the closed-bore linac was 0.07mm and 0.07 degrees at load position, and 0.06mm and 0.01 degrees at align position for the white head phantom. The precision was 0.02mm and 0.02 degrees in load position, and 0.01mm and 0.02 degrees in align position. The accuracy for the Penta-Guide phantom was comparable to the white head phantom, with 0.06mm and 0.01 degrees in align position. The system was slightly less accurate for translations of the CIRS lung phantom in align position (0.20mm, 0.05 degrees). Reproducibility measurements showed a variation of 0.02mm in load position. Regarding the temporal stability, the maximum drift over 30min was 0.33mm and 0.02 degrees in load position. No effect of ambient light level on the accuracy of the IDENTIFYTM system was observed. Regarding different surface colors, the accuracy of the system for a black phantom was slightly worse compared to a white surface, but not clinical relevant. The IDENTIFYTM system can adequately be used for motion monitoring on a closed-bore linac with submillimeter accuracy. The results of the performed measurements meet the clinical requirements described in the guidelines of the AAPM and the ESTRO.

A kV–MV approach to CBCT metal artifact reduction using multi-layer MV-CBCT

Objective. To demonstrate that complete cone beam CT (CBCT) scans from both MV-energy and kV-energy LINAC sources can reduce metal artifacts in radiotherapy guidance, while maintaining standard-of-care x-ray doses levels. Approach. MV-CBCT and kV-CBCT scans are acquired at half normal dose. The impact of lowered dose on MV-CBCT data quality is mitigated by the use of a 4-layer MV-imager prototype and reduced LINAC energy settings (2.5 MV) to improve photon capture. Additionally, the MV-CBCT is used to determine the 3D position and pose of metal implants, which in turn is used to guide model-based poly-energetic correction and interleaving of the kV-CBCT and MV-CBCT data. Certain edge-preserving regularization steps incorporated into the model-based correction algorithm further reduce MV data noise. Main results. The method was tested in digital phantoms and a real pelvis phantom with large 2.5″ spherical inserts, emulating hip replacements of different materials. The proposed method demonstrated an appealing compromise between the high contrast of kV-CBCT and low artifact content of MV-CBCT. Contrast-to-noise improved 3-fold compared to MV-CBCT with a clinical 1-layer architecture at matched dose (37 mGy) and edge blur levels. Visual delineation of the bladder and prostate improved noteably over kV- or MV-CBCT alone. Significance. The proposed method demonstrates that a full MV-CBCT scan can be combined with kV-CBCT to reduce metal artifacts without resorting to complicated beam collimation strategies to limit the MV-CBCT dose contribution. Additionally, significant improvements in CNR can be achieved as compared to metal artifact reduction through current clinical MV-CBCT practices.