Automatic Radiotherapy Planning for Glioblastoma Radiotherapy With Sparing of the Hippocampus and nTMS-Defined Motor Cortex

BackgroundNavigated transcranial magnetic stimulation (nTMS) of the motor cortex has been successfully implemented into radiotherapy planning by a number of studies. Furthermore, the hippocampus has been identified as a radiation-sensitive structure meriting particular sparing in radiotherapy. This study assesses the joint protection of these two eloquent brain regions for the treatment of glioblastoma (GBM), with particular emphasis on the use of automatic planning.Patients and MethodsPatients with motor-eloquent brain glioblastoma who underwent surgical resection after nTMS mapping of the motor cortex and adjuvant radiotherapy were retrospectively evaluated. The radiotherapy treatment plans were retrieved, and the nTMS-defined motor cortex and hippocampus contours were added. Four additional treatment plans were created for each patient: two manual plans aimed to reduce the dose to the motor cortex and hippocampus by manual inverse planning. The second pair of re-optimized plans was created by the Auto-Planning algorithm. The optimized plans were compared with the “Original” plan regarding plan quality, planning target volume (PTV) coverage, and sparing of organs at risk (OAR).ResultsA total of 50 plans were analyzed. All plans were clinically acceptable with no differences in the PTV coverage and plan quality metrics. The OARs were preserved in all plans; however, overall the sparing was significantly improved by Auto-Planning. Motor cortex protection was feasible and significant, amounting to a reduction in the mean dose by >6 Gy. The dose to the motor cortex outside the PTV was reduced by >12 Gy (mean dose) and >5 Gy (maximum dose). The hippocampi were significantly improved (reduction in mean dose: ipsilateral >6 Gy, contralateral >4.6 Gy; reduction in maximum dose: ipsilateral >5 Gy, contralateral >5 Gy). While the dose reduction using Auto-Planning was generally better than by manual optimization, the radiated total monitor units were significantly increased.ConclusionConsiderable dose sparing of the nTMS-motor cortex and hippocampus could be achieved with no disadvantages in plan quality. Auto-Planning could further contribute to better protection of OAR. Whether the improved dosimetric protection of functional areas can translate into improved quality of life and motor or cognitive performance of the patients can only be decided by future studies.

Clinical Validation and Treatment Plan Evaluation Based on Auto-Delineation of the Clinical Target Volume for Prostate Cancer Radiotherapy

Purpose Clinical target volumes (CTV) and organs at risk (OAR) could be auto-contoured to save workload. The goal of this study was to assess a convolutional neural network (CNN) for totally automatic and accurate CTV and OAR in prostate cancer, while also comparing anticipated treatment plans based on auto-contouring CTV to clinical plans. Methods From January 2013 to January 2019, 217 computed tomography (CT) scans of patients with locally advanced prostate cancer treated at our hospital were collected and analyzed. CTV and OAR were delineated with a deep learning based method, which named CUNet. The performance of this strategy was evaluated using the mean Dice similarity coefficient (DSC), 95th percentile Hausdorff distance (95HD), and subjective evaluation. Treatment plans were graded using predetermined evaluation criteria, and % errors for clinical doses to the planned target volume (PTV) and organs at risk(OARs) were calculated. Results The defined CTVs had mean DSC and 95HD values of 0.84 and 5.04 mm, respectively. For one patient's CT scans, the average delineation time was less than 15 seconds. When CTV outlines from CUNetwere blindly chosen and compared to GT, the overall positive rate in clinicians A and B was 53.15% vs 46.85%, and 54.05% vs 45.95%, respectively (P>0.05), demonstrating that our deep machine learning model performed as good as or better than human demarcation Furthermore, 8 testing patients were chosen at random to design the predicted plan based on the auto-courtoring CTV and OAR, demonstrating acceptable agreement with the clinical plan: average absolute dose differences of D2, D50, D98, Dmean for PTV are within 0.74%, and average absolute volume differences of V45, V50 for OARs are within 3.4%. Without statistical significance (p>0.05), the projected findings are comparable to clinical truth. Conclusion The experimental results show that the CTV and OARs defined by CUNet for prostate cancer were quite close to the ground reality.CUNet has the potential to cut radiation oncologists' contouring time in half. When compared to clinical plans, the differences between estimated doses to CTV and OAR based on auto-courtoring were small, with no statistical significance, indicating that treatment planning for prostate cancer based on auto-courtoring has potential.

An algorithm for thoracic re-irradiation using biologically effective dose: a common language on how to treat in a “no-treat zone”

Background Re-irradiation (re-RT) is a technically challenging task for which few standardized approaches exist. This is in part due to the lack of a common platform to assess dose tolerance in relation to toxicity in the re-RT setting. To better address this knowledge gap and provide new tools for studying and developing thresholds for re-RT, we developed a novel algorithm that allows for anatomically accurate three-dimensional mapping of composite biological effective dose (BED) distributions from nominal doses (Gy). Methods The algorithm was designed to automatically convert nominal dose from prior treatment plans to corresponding BED value maps (voxel size 2.5 mm3 and α/β of 3 for normal tissue, BED3). Following the conversion of each plan to a BED3 dose distribution, deformable registration was used to create a summed composite re-irradiation BED3 plan for each patient who received two treatments. A proof-of-principle analysis was performed on 38 re-irradiation cases of initial stereotactic ablative radiotherapy (SABR) followed by either re-SABR or chemoradiation for isolated locoregional recurrence of early-stage non-small cell lung cancer. Results Evaluation of the algorithm-generated maps revealed appropriate conversion of physical dose to BED at each voxel. Of 14 patients receiving repeat SABR, there was one case each of grade 3 chest wall pain (7%), pneumonitis (7%), and dyspnea (7%). Of 24 patients undergoing repeat fractionated radiotherapy, grade 3 events were limited to two cases each of pneumonitis and dyspnea (8%). Composite BED3 dosimetry for each patient who experienced grade 2–3 events is provided and may help guide development of precise cumulative dose thresholds for organs at risk in the re-RT setting. Conclusions This novel algorithm successfully created a voxel-by-voxel composite treatment plan using BED values. This approach may be used to more precisely examine dosimetric predictors of toxicities and to establish more accurate normal tissue constraints for re-irradiation.

A pilot study of machine-learning based automated planning for primary brain tumours

Purpose High-quality radiotherapy (RT) planning for children and young adults with primary brain tumours is essential to minimize the risk of late treatment effects. The feasibility of using automated machine-learning (ML) to aid RT planning in this population has not previously been studied. Methods and materials We developed a ML model that identifies learned relationships between image features and expected dose in a training set of 95 patients with a primary brain tumour treated with focal radiotherapy to a dose of 54 Gy in 30 fractions. This ML method was then used to create predicted dose distributions for 15 previously-treated brain tumour patients across two institutions, as a testing set. Dosimetry to target volumes and organs-at-risk (OARs) were compared between the clinically-delivered (human-generated) plans versus the ML plans. Results The ML method was able to create deliverable plans in all 15 patients in the testing set. All ML plans were generated within 30 min of initiating planning. Planning target volume coverage with 95% of the prescription dose was attained in all plans. OAR doses were similar across most structures evaluated; mean doses to brain and left temporal lobe were lower in ML plans than manual plans (mean difference to left temporal, – 2.3 Gy, p = 0.006; mean differences to brain, – 1.3 Gy, p = 0.017), whereas mean doses to right cochlea and lenses were higher in ML plans (+ 1.6–2.2 Gy, p < 0.05 for each). Conclusions Use of an automated ML method to aid RT planning for children and young adults with primary brain tumours is dosimetrically feasible and can be successfully used to create high-quality 54 Gy RT plans. Further evaluation after clinical implementation is planned.

Comparing Multi-Objective Local Search Algorithms for the Beam Angle Selection Problem

In intensity-modulated radiation therapy, treatment planners aim to irradiate the tumour according to a medical prescription while sparing surrounding organs at risk as much as possible. Although this problem is inherently a multi-objective optimisation (MO) problem, most of the models in the literature are single-objective ones. For this reason, a large number of single-objective algorithms have been proposed in the literature to solve such single-objective models rather than multi-objective ones. Further, a difficulty that one has to face when solving the MO version of the problem is that the algorithms take too long before converging to a set of (approximately) non-dominated points. In this paper, we propose and compare three different strategies, namely random PLS (rPLS), judgement-function-guided PLS (jPLS) and neighbour-first PLS (nPLS), to accelerate a previously proposed Pareto local search (PLS) algorithm to solve the beam angle selection problem in IMRT. A distinctive feature of these strategies when compared to the PLS algorithms in the literature is that they do not evaluate their entire neighbourhood before performing the dominance analysis. The rPLS algorithm randomly chooses the next non-dominated solution in the archive and it is used as a baseline for the other implemented algorithms. The jPLS algorithm first chooses the non-dominated solution in the archive that has the best objective function value. Finally, the nPLS algorithm first chooses the solutions that are within the neighbourhood of the current solution. All these strategies prevent us from evaluating a large set of BACs, without any major impairment in the obtained solutions’ quality. We apply our algorithms to a prostate case and compare the obtained results to those obtained by the PLS from the literature. The results show that algorithms proposed in this paper reach a similar performance than PLS and require fewer function evaluations.

Volumetric Modulated Arc Therapy Capabilities for Treating Lower-Extremity Skin Affected by Several Merkel Cell Carcinoma Nodules: When Technological Advances Effectively Achieve the Palliative Therapeutic Goal while Minimising the Risk of Potential Toxicities

The peculiar and rare clinical condition below clearly requires a customized care approach in the context of personalized medicine. An 80-year-old female patient who was subjected in 2018 to surgical removal of a cutaneous Merkel cell carcinoma (MCC) nodule located on the posterior surface of the left thigh and to three subsequent palliative radiotherapy treatments developed a fourth relapse in October 2020, with fifteen nodular metastases located in the left thigh and leg. Since the overall macroscopic disease was still exclusively regionally located and microscopic spread was likely extended also to clinically negative skin of the thigh and leg, we performed an irradiation of the whole left lower extremity. For this purpose the total target (65.5 cm) was divided into three sub-volumes. Dose prescription was 30 Gy in 15 daily fractions. A sequential boost of 10 Gy in 5 daily fractions was planned for macroscopic nodules. Plans were calculated by means of volumetric modulated arc therapy (VMAT) with the field overlap technique. Thanks to this, we obtained a homogeneous dose distribution in the field junction region; avoidance structures were delineated in the central part of the thigh and leg with the aim of achieving an optimal superficial dose painting and to reduce bone exposure to radiation. This case study demonstrates that VMAT allows for a good dose coverage for circumferential cutaneous targets while sparing deeper organs at risk. A reproducible image-guided set-up is fundamental for an accurate and safe dose delivery. However, local treatments such as radiotherapy for very advanced MCC of the lower extremities might have limited impact due to the high probability of systemic progression, as illustrated in this case. Radiation is confirmed as being effective in preventing MCC nodule progression toward skin wounding.

Optimized Conformal Total Body Irradiation Among Recipients of TCRαβ/CD19-Depleted Grafts in Pediatric Patients With Hematologic Malignancies: Single-Center Experience

Total body irradiation (TBI) in combination with chemotherapy is widely used as a conditioning regimen in pediatric and adult hematopoietic stem cell transplantation (HSCT). The combination of TBI with chemotherapy has demonstrated superior survival outcomes in patients with acute lymphoblastic and myeloid leukemia when compared with conditioning regimens based only on chemotherapy. The clinical application of intensity-modulated radiation therapy (IMRT)-based methods (volumetric modulated arc therapy (VMAT) and TomoTherapy) seems to be promising and has been actively used worldwide. The optimized conformal total body irradiation (OC-TBI) method described in this study provides selected dose reduction for organs at risk with respect to the most significant toxicity (lungs, kidneys, lenses). This study included 220 pediatric patients who received OC-TBI with subsequent chemotherapy and allogenic HSCT with TCRαβ/CD19 depletion. A group of 151 patients received OC-TBI using TomoTherapy, and 40 patients received OC-TBI using the Elekta Synergy™ linac with an Agility-MLC (Elekta, Crawley, UK) using volumetric modulated arc therapy (VMAT). Twenty-nine patients received OC-TBI with supplemental simultaneous boost to bone marrow—(SIB to BM) up to 15 Gy: 28 patients (pts)—TomoTherapy; one patient—VMAT. The follow-up duration ranged from 0.3 to 6.4 years (median follow-up, 2.8 years). Overall survival (OS) for all the patients was 63% (95% CI: 56–70), and event-free survival (EFS) was 58% (95% CI: 51–65). The cumulative incidence of transplant-related mortality (TRM) was 10.7% (95% CI: 2.2–16) for all patients. The incidence of early TRM (&lt;100 days) was 5.0% (95% CI: 1.5–8.9), and that of late TRM (&gt;100 days) was 5.7 (95% CI: 1.7–10.2). The main causes of death for all the patients were relapse and infection. The concept of OC-TBI using IMRT VMAT and helical treatment delivery on a TomoTherapy treatment unit provides maximum control of the dose distribution in extended targets with simultaneous dose reduction for organs at risk. This method demonstrated a low incidence of severe side effects after radiation therapy and predictable treatment effectiveness. Our initial experience demonstrates that OC-TBI appears to be a promising technique for the treatment of pediatric patients.

Artificial Intelligence: A New Tool in Oncologist's Armamentarium

Artificial intelligence (AI) has become an essential tool in human life because of its pivotal role in communications, transportation, media, and social networking. Inspired by the complex neuronal network and its functions in human beings, AI, using computer-based algorithms and training, had been explored since the 1950s. To tackle the enormous amount of patients' clinical data, imaging, histopathological data, and the increasing pace of research on new treatments and clinical trials, and ever-changing guidelines for treatment with the advent of novel drugs and evidence, AI is the need of the hour. There are numerous publications and active work on AI's role in the field of oncology. In this review, we discuss the fundamental terminology of AI, its applications in oncology on the whole, and its limitations. There is an inter-relationship between AI, machine learning and, deep learning. The virtual branch of AI deals with machine learning. While the physical branch of AI deals with the delivery of different forms of treatment—surgery, targeted drug delivery, and elderly care. The applications of AI in oncology include cancer screening, diagnosis (clinical, imaging, and histopathological), radiation therapy (image acquisition, tumor and organs at risk segmentation, image registration, planning, and delivery), prediction of treatment outcomes and toxicities, prediction of cancer cell sensitivity to therapeutics and clinical decision-making. A specific area of interest is in the development of effective drug combinations tailored to every patient and tumor with the help of AI. Radiomics, the new kid on the block, deals with the planning and administration of radiotherapy. As with any new invention, AI has its fallacies. The limitations include lack of external validation and proof of generalizability, difficulty in data access for rare diseases, ethical and legal issues, no precise logic behind the prediction, and last but not the least, lack of education and expertise among medical professionals. A collaboration between departments of clinical oncology, bioinformatics, and data sciences can help overcome these problems in the near future.

Accelerated hypofractionated radiotherapy for chest wall and nodal irradiation using hybrid techniques

Aim: This study compares three different hybrid plans, for left-sided chest wall (CW) and nodal stations irradiation using a hypofractionated dose regimen. Materials and methods: Planning target volumes (PTVs) of 25 breast cancer patients that included CW, supraclavicular (SCL) and internal mammary node (IMN) were planned with 3 different hybrid techniques: 3DCRT+IMRT, 3DCRT+VMAT and IMRT+VMAT. All hybrid plans were generated with a hypofractionated dose prescription of 40·5 Gy in 15 fractions. Seventy per cent of the dose was planned with the base-dose component and remaining 30% of the dose was planned with the hybrid component. All plans were evaluated based on the PTVs and organs at risk (OARs) dosimetric parameters. Results: The results for PTVs parameters have shown that the 3DCRT+IMRT and 3DCRT+VMAT plans were superior in uniformity index to the IMRT+VMAT plan. The OARs dose parameters were comparable between hybrid plans. The IMRT+VMAT plan provided a larger low dose volume spread to the heart and ipsilateral lung (p < 0·001). The 3DCRT+VMAT plan required less monitor units and treatment time (p = 0·005) than other plans. Conclusion: The 3DCRT+VMAT hybrid plan showed superior results with efficient treatment delivery and provide clinical benefit by reducing both low and high dose levels.