Photon vs proton hypofractionation in prostate cancer: A systematic review and meta-analysis.

BACKGROUND: High-level evidence on hypofractionated proton therapy (PT) for localized and locally advanced prostate cancer (PCa) patients is currently missing. The aim of this study is to provide a systematic literature review to compare the toxicity and effectiveness of curative radiotherapy with photon therapy (XRT) or PT in PCa. METHODS: PubMed, Embase, and the Cochrane Library databases were systematically searched up to April 2022. Men with a diagnosis of PCa who underwent curative hypofractionated RT treatment (PT or XRT) were included. Risk of grade (G) ≥ 2 acute and late genitourinary (GU) OR gastrointestinal (GI) toxicity were the primary outcomes of interest. Secondary outcomes were five-year biochemical relapse-free survival (b-RFS), clinical relapse-free, distant metastasis-free, and prostate cancer-specific survival. Heterogeneity between study-specific estimates was assessed using Chi-square statistics and measured with the I2 index (heterogeneity measure across studies). RESULTS: A total of 230 studies matched inclusion criteria and, due to overlapped populations, 160 were included in the present analysis. Significant lower rates of G ≥ 2 acute GI incidence (2 % vs 7 %) and improved 5-year biochemical relapse-free survival (95 % vs 91 %) were observed in the PT arm compared to XRT. PT benefits in 5-year biochemical relapse-free survival were maintained for the moderate hypofractionated arm (p-value 0.0122) and among patients in intermediate and low-risk classes (p-values < 0.0001 and 0.0368, respectively). No statistically relevant differences were found for the other considered outcomes. CONCLUSION: The present study supports that PT is safe and effective for localized PCa treatment, however, more data from RCTs are needed to draw solid evidence in this setting and further effort must be made to identify the patient subgroups that could benefit the most from PT.

Can we predict pathology without surgery? Weighing the added value of multiparametric MRI and whole prostate radiomics in integrative machine learning models.

OBJECTIVE: To test the ability of high-performance machine learning (ML) models employing clinical, radiological, and radiomic variables to improve non-invasive prediction of the pathological status of prostate cancer (PCa) in a large, single-institution cohort. METHODS: Patients who underwent multiparametric MRI and prostatectomy in our institution in 2015-2018 were considered; a total of 949 patients were included. Gradient-boosted decision tree models were separately trained using clinical features alone and in combination with radiological reporting and/or prostate radiomic features to predict pathological T, pathological N, ISUP score, and their change from preclinical assessment. Model behavior was analyzed in terms of performance, feature importance, Shapley additive explanation (SHAP) values, and mean absolute error (MAE). The best model was compared against a naïve model mimicking clinical workflow. RESULTS: The model including all variables was the best performing (AUC values ranging from 0.73 to 0.96 for the six endpoints). Radiomic features brought a small yet measurable boost in performance, with the SHAP values indicating that their contribution can be critical to successful prediction of endpoints for individual patients. MAEs were lower for low-risk patients, suggesting that the models find them easier to classify. The best model outperformed (p ≤ 0.0001) clinical baseline, resulting in significantly fewer false negative predictions and overall was less prone to under-staging. CONCLUSIONS: Our results highlight the potential benefit of integrative ML models for pathological status prediction in PCa. Additional studies regarding clinical integration of such models can provide valuable information for personalizing therapy offering a tool to improve non-invasive prediction of pathological status. CLINICAL RELEVANCE STATEMENT: The best machine learning model was less prone to under-staging of the disease. The improved accuracy of our pathological prediction models could constitute an asset to the clinical workflow by providing clinicians with accurate pathological predictions prior to treatment. KEY POINTS: • Currently, the most common strategies for pre-surgical stratification of prostate cancer (PCa) patients have shown to have suboptimal performances. • The addition of radiological features to the clinical features gave a considerable boost in model performance. Our best model outperforms the naïve model, avoiding under-staging and resulting in a critical advantage in the clinic. •Machine learning models incorporating clinical, radiological, and radiomics features significantly improved accuracy of pathological prediction in prostate cancer, possibly constituting an asset to the clinical workflow.

More than Five Decades of Proton Therapy: A Bibliometric Overview of the Scientific Literature.

BACKGROUND: The therapeutic potential of proton therapy (PT) was first recognized in 1946 by Robert Wilson, and nowadays, over 100 proton centers are in operation worldwide, and more than 60 are under construction or planned. Bibliometric data can be used to perform a structured analysis of large amounts of scientific data to provide new insights, e.g., to assess the growth and development of the field and to identify research trends and hot topics. The aim of this study is to provide a comprehensive bibliometric analysis of the current status and trends in scientific literature in the PT field. METHODS: The literature on PT until the 31st December 2022 in the Scopus database was searched, including the following keywords: proton AND radiotherapy AND cancer/tumor in title, abstract, and/or keywords. The open-source R Studio's Bibliometrix package and Biblioshiny software (version 2.0) were used to perform the analysis. RESULTS: A total of 7335 documents, mainly articles (n = 4794, 65%) and reviews (n = 1527, 21%), were collected from 1946 to 2022 from 1054 sources and 21,696 authors. Of these, roughly 84% (n = 6167) were produced in the last 15 years (2008-2022), in which the mean annual growth rate was 13%. Considering the corresponding author's country, 79 countries contributed to the literature; the USA was the top contributor, with 2765 (38%) documents, of whom 84% were single-country publications (SCP), followed by Germany and Japan, with 535 and 531 documents of whom 66% and 93% were SCP. Considering the themes subanalysis (2002-2022), a total of 7192 documents were analyzed; among all keywords used by authors, the top three were radiotherapy (n = 1394, 21% of documents), intensity-modulated radiotherapy (n = 301, 5%), and prostate cancer (n = 301, 5%). Among disease types, prostate cancer is followed by chordoma, head and neck, and breast cancer. The change in trend themes demonstrated the fast evolution of hotspots in PT; among the most recent trends, the appearance of flash, radiomics, relative biological effectiveness (RBE), and linear energy transfer (LET) deserve to be highlighted. CONCLUSIONS: The results of the present bibliometric analysis showed that PT is an active and rapidly increasing field of research. Themes of the published works encompass the main aspects of its application in clinical practice, such as the comparison with the actual photon-based standard of care technique and the continuing technological advances. This analysis gives an overview of past scientific production and, most importantly, provides a useful point of view on the future directions of the research activities.

Oral Surgery and Osteoradionecrosis in Patients Undergoing Head and Neck Radiation Therapy: An Update of the Current Literature.

Osteoradionecrosis (ORN) is a serious long-term complication of head and neck radiotherapy (RT), which is often triggered by dental extractions. It results from avascular aseptic necrosis due to irradiated bone damage. ORN is challenging to treat and can lead to severe complications. Furthermore, ORN causes pain and distress, significantly reducing the patient's quality of life. There is currently no established preventive strategy. This narrative review aims to provide an update for the clinicians on the risk of ORN associated with oral surgery in head and neck RT patients, with a focus on the timing suitable for the oral surgery and possible ORN preventive treatments. An electronic search of articles was performed by consulting the PubMed database. Intervention and observational studies were included. A multidisciplinary approach to the patient is highly recommended to mitigate the risk of RT complications. A dental visit before commencing RT is highly advised to minimize the need for future dental extractions after irradiation, and thus the risk of ORN. Post-RT preventive strategies, in case of dento-alveolar surgery, have been proposed and include antibiotics, hyperbaric oxygen (HBO), and the combined use of pentoxifylline and tocopherol ("PENTO protocol"), but currently there is a lack of established standards of care. Some limitations in the use of HBO involve the low availability of HBO facilities, its high costs, and specific clinical contraindications; the PENTO protocol, on the other hand, although promising, lacks clinical trials to support its efficacy. Due to the enduring risk of ORN, removable prostheses are preferable to dental implants in these patients, as there is no consensus on the appropriate timing for their safe placement. Overall, established standards of care and high-quality evidence are lacking concerning both preventive strategies for ORN as well as the timing of the dental surgery. There is an urgent need to improve research for more efficacious clinical decision making.

Impact of inter-observer variability on first axillary level dosimetry in breast cancer radiotherapy: An AIRO multi-institutional study.

This study quantified the incidental dose to the first axillary level (L1) in locoregional treatment plan for breast cancer. Eighteen radiotherapy centres contoured L1-L4 on three different patients (P1,2,3), created the L2-L4 planning target volume (single centre planning target volume, SC-PTV) and elaborated a locoregional treatment plan. The L2-L4 gold standard clinical target volume (CTV) along with the gold standard L1 contour (GS-L1) were created by an expert consensus. The SC-PTV was then replaced by the GS-PTV and the incidental dose to GS-L1 was measured. Dosimetric data were analysed with Kruskal-Wallis test. Plans were intensity modulated radiotherapy (IMRT)-based. P3 with 90° arm setup had statistically significant higher L1 dose across the board than P1 and P2, with the mean dose (Dmean) reaching clinical significance. Dmean of P1 and P2 was consistent with the literature (77.4% and 74.7%, respectively). The incidental dose depended mostly on L1 proportion included in the breast fields, underlining the importance of the setup, even in case of IMRT.

Hypofractionated proton therapy for benign tumors of the central nervous system: A systematic review of the literature.

AIMS: Aim of the present analysis was to report results of a systematic review of the literature in the setting of patients treated with hypoF PT for benign lesions of the central nervous system (CNS). METHODS: The methodology complied with the PRISMA recommendations. PubMed, EMBASE and Scopus databases were interrogated in September 2022. RESULTS: Twelve papers have been selected including patients treated for base of the skull meningiomas (6 papers), vestibular schwannoma (3 papers) and pituitary adenomas (3 papers). Clinical outcomes were evaluated with both radiologic images and clinical parameters. Long-term toxicity was reported in all but one series with an incidence ranging from 2 % to 7 % in patients treated for base of skull meningioma and 1-9 % for schwannoma. CONCLUSIONS: HypoF PT is a safe and effective treatment in selected benign tumors of the CNS. Further dosimetric and clinical comparisons are required to better refine the patients' selection criteria.

Cyberknife Radiosurgery for Prostate Cancer after Abdominoperineal Resection (CYRANO): The Combined Computer Tomography and Electromagnetic Navigation Guided Transperineal Fiducial Markers Implantation Technique.

In this technical development report, we present the strategic placement of fiducial markers within the prostate under the guidance of computed tomography (CT) and electromagnetic navigation (EMN) for the delivery of ultra-hypofractionated cyberknife (CK) therapy in a patient with localized prostate cancer (PCa) who had previously undergone chemo-radiotherapy for rectal cancer and subsequent abdominoperineal resection due to local recurrence. The patient was positioned in a prone position with a pillow under the pelvis to facilitate access, and an electromagnetic fiducial marker was placed on the patient's skin to establish a stable position. CT scans were performed to plan the procedure, mark virtual points, and simulate the needle trajectory using the navigation system. Local anesthesia was administered, and a 21G needle was used to place the fiducial markers according to the navigation system information. A confirmatory CT scan was obtained to ensure proper positioning. The implantation procedure was safe, without any acute side effects such as pain, hematuria, dysuria, or hematospermia. Our report highlights the ability to use EMN systems to virtually navigate within a pre-acquired imaging dataset in the interventional room, allowing for non-conventional approaches and potentially revolutionizing fiducial marker positioning, offering new perspectives for PCa treatment in selected cases.

Comparison of automated segmentation techniques for magnetic resonance images of the prostate.

BACKGROUND: Contouring of anatomical regions is a crucial step in the medical workflow and is both time-consuming and prone to intra- and inter-observer variability. This study compares different strategies for automatic segmentation of the prostate in T2-weighted MRIs. METHODS: This study included 100 patients diagnosed with prostate adenocarcinoma who had undergone multi-parametric MRI and prostatectomy. From the T2-weighted MR images, ground truth segmentation masks were established by consensus from two expert radiologists. The prostate was then automatically contoured with six different methods: (1) a multi-atlas algorithm, (2) a proprietary algorithm in the Syngo.Via medical imaging software, and four deep learning models: (3) a V-net trained from scratch, (4) a pre-trained 2D U-net, (5) a GAN extension of the 2D U-net, and (6) a segmentation-adapted EfficientDet architecture. The resulting segmentations were compared and scored against the ground truth masks with one 70/30 and one 50/50 train/test data split. We also analyzed the association between segmentation performance and clinical variables. RESULTS: The best performing method was the adapted EfficientDet (model 6), achieving a mean Dice coefficient of 0.914, a mean absolute volume difference of 5.9%, a mean surface distance (MSD) of 1.93 pixels, and a mean 95th percentile Hausdorff distance of 3.77 pixels. The deep learning models were less prone to serious errors (0.854 minimum Dice and 4.02 maximum MSD), and no significant relationship was found between segmentation performance and clinical variables. CONCLUSIONS: Deep learning-based segmentation techniques can consistently achieve Dice coefficients of 0.9 or above with as few as 50 training patients, regardless of architectural archetype. The atlas-based and Syngo.via methods found in commercial clinical software performed significantly worse (0.855[Formula: see text]0.887 Dice).

Quality assurance for automatically generated contours with additional deep learning.

OBJECTIVE: Deploying an automatic segmentation model in practice should require rigorous quality assurance (QA) and continuous monitoring of the model's use and performance, particularly in high-stakes scenarios such as healthcare. Currently, however, tools to assist with QA for such models are not available to AI researchers. In this work, we build a deep learning model that estimates the quality of automatically generated contours. METHODS: The model was trained to predict the segmentation quality by outputting an estimate of the Dice similarity coefficient given an image contour pair as input. Our dataset contained 60 axial T2-weighted MRI images of prostates with ground truth segmentations along with 80 automatically generated segmentation masks. The model we used was a 3D version of the EfficientDet architecture with a custom regression head. For validation, we used a fivefold cross-validation. To counteract the limitation of the small dataset, we used an extensive data augmentation scheme capable of producing virtually infinite training samples from a single ground truth label mask. In addition, we compared the results against a baseline model that only uses clinical variables for its predictions. RESULTS: Our model achieved a mean absolute error of 0.020 ± 0.026 (2.2% mean percentage error) in estimating the Dice score, with a rank correlation of 0.42. Furthermore, the model managed to correctly identify incorrect segmentations (defined in terms of acceptable/unacceptable) 99.6% of the time. CONCLUSION: We believe that the trained model can be used alongside automatic segmentation tools to ensure quality and thus allow intervention to prevent undesired segmentation behavior.