Erectile function preservation after salvage radiation therapy for biochemically recurrent prostate cancer after prostatectomy: Five-year results of the SAKK 09/10 randomized phase 3 trial.

Objectives: To evaluate effects of dose intensified salvage radiotherapy (sRT) on erectile function in biochemically recurrent prostate cancer (PC) after radical prostatectomy (RP). Materials and methods: Eligible patients had evidence of biochemical failure after RP and a PSA at randomization of ≤ 2 ng/ml. Erectile dysfunction (ED) was investigated as secondary endpoint within the multicentre randomized trial (February 2011 to April 2014) in patients receiving either 64 Gy or 70 Gy sRT. ED and quality of life (QoL) were assessed using CTCAE v4.0 and the EORTC QoL questionnaires C30 and PR25 at baseline and up to 5 years after sRT. Results: 344 patients were evaluable. After RP 197 (57.3 %) patients had G0-2 ED while G3 ED was recorded in 147 (42.7 %) patients. Subsequently, sexual activity and functioning was impaired. 5 years after sRT, 101 (29.4 %) patients noted G0-2 ED. During follow-up, 44.2 % of patients with baseline G3 ED showed any improvement and 61.4 % of patients with baseline G0-2 ED showed worsening. Shorter time interval between RP and start of sRT (p = 0.007) and older age at randomization (p = 0.005) were significant predictors to more baseline ED and low sexual activity in the long-term. Age (p = 0.010) and RT technique (p = 0.031) had a significant impact on occurrence of long-term ED grade 3 and worse sexual functioning. During follow-up, no differences were found in erectile function, sexual activity, and sexual functioning between the 64 Gy and 70 Gy arm. Conclusion: ED after RP is a known long-term side effect with significant impact on patients' QoL. ED was further affected by sRT, but dose intensification of sRT showed no significant impact on erectile function recovery or prevalence of de novo ED after sRT. Age, tumor stage, prostatectomy and RT-techniques, nerve-sparing and observation time were associated with long-term erectile function outcome.ClinicalTrials.gov. Identifier: NCT01272050.

Real-world treatment patterns and medical costs of prostate cancer patients in Switzerland - A claims data analysis.

BACKGROUND: Prostate cancer (PC) is the most prevalent cancer in men in Switzerland. However, evidence on the real-world health care use of PC patients is scarce. The aim of this study is to describe health care utilization, treatment patterns, and medical costs in PC patients over a period of five years (2014-2018). METHOD: We used routinely collected longitudinal individual-level claims data from a major provider of mandatory health insurance in Switzerland. Due to the lack of diagnostic coding in the claims data, we identified treated PC patients based on the treatments received. We described health care utilization and treatment pathways for patients with localized and metastatic PC. Costs were calculated from a health care system perspective. RESULTS: A total of 5591 PC patients met the inclusion criteria. Between 2014 and 2018, 1741 patients had outpatient radiotherapy for localized or metastatic PC and 1579 patients underwent radical prostatectomy. 3502 patients had an androgen deprivation therapy (ADT). 9.5% of these patients had a combination therapy with docetaxel, and 11.0% had a combination with abiraterone acetate. Docetaxel was the most commonly used chemotherapy (first-line; n = 413, 78.4% of all patients in chemotherapy). Total medical costs of PC in Switzerland were estimated at CHF 347 m (95% CI 323-372) in 2018. CONCLUSION: Most PC patients in this study were identified based on the use of ADT. Medical costs of PC in Switzerland amounted to 0.45% of total health care spending in 2018. Treatment of metastatic PC accounted for about two thirds of spending.

Converting between the International Prostate Symptom Score (IPSS) and the Expanded Prostate Cancer Index Composite (EPIC) urinary subscales: modeling and external validation.

BACKGROUND: Prostate-related quality of life can be assessed with a variety of different questionnaires. The 50-item Expanded Prostate Cancer Index Composite (EPIC) and the International Prostate Symptom Score (IPSS) are two widely used options. The goal of this study was, therefore, to develop and validate a model that is able to convert between the EPIC and the IPSS to enable comparisons across different studies. METHODS: Three hundred forty-seven consecutive patients who had previously received radiotherapy and surgery for prostate cancer at two institutions in Switzerland and Germany were contacted via mail and instructed to complete both questionnaires. The Swiss cohort was used to train and internally validate different machine learning models using fourfold cross-validation. The German cohort was used for external validation. RESULTS: Converting between the EPIC Urinary Irritative/Obstructive subscale and the IPSS using linear regressions resulted in mean absolute errors (MAEs) of 3.88 and 6.12, which is below the respective previously published minimal important differences (MIDs) of 5.2 and 10 points. Converting between the EPIC Urinary Summary and the IPSS was less accurate with MAEs of 5.13 and 10.45, similar to the MIDs. More complex model architectures did not result in improved performance in this study. The study was limited to the German versions of the respective questionnaires. CONCLUSIONS: Linear regressions can be used to convert between the IPSS and the EPIC Urinary subscales. While the equations obtained in this study can be used to compare results across clinical trials, they should not be used to inform clinical decision-making in individual patients. TRIAL REGISTRATION: This study was retrospectively registered on clinicaltrials.gov on January 14th, 2022, under the registration number NCT05192876.

Impact of Imaging-Guided Localization on Performance of Tailored Axillary Surgery in Patients with Clinically Node-Positive Breast Cancer: Prospective Cohort Study Within TAXIS (OPBC-03, SAKK 23/16, IBCSG 57-18, ABCSG-53, GBG 101).

BACKGROUND: Tailored axillary surgery (TAS) is a novel surgical concept for clinical node-positive breast cancer. It consists of the removal of the sentinel lymph nodes (LNs), as well as palpably suspicious nodes. The TAS technique can be utilized in both the upfront and neoadjuvant chemotherapy (NACT) setting. This study assessed whether/how imaging-guided localization (IGL) influenced TAS. PATIENTS AND METHODS: This was a prospective observational cohort study preplanned in the randomized phase-III OPBC-03/TAXIS trial. IGL was performed at the surgeon's discretion for targeted removal of LNs during TAS. Immediate back-up axillary lymph node dissection (ALND) followed TAS according to TAXIS randomization. RESULTS: Five-hundred patients were included from 44 breast centers in six countries, 151 (30.2%) of whom underwent NACT. IGL was performed in 84.4% of all patients, with significant variation by country (77.6-100%, p < 0.001). No difference in the median number of removed (5 vs. 4, p = 0.3) and positive (2 vs. 2, p = 0.6) LNs by use of IGL was noted. The number of LNs removed during TAS with IGL remained stable over time (p = 0.8), but decreased significantly without IGL, from six (IQR 4-6) in 2019 to four (IQR 3-4) in 2022 (p = 0.015). An ALND was performed in 249 patients, removing another 12 (IQR 9-17) LNs, in which a median number of 1 (IQR 0-4) was positive. There was no significant difference in residual nodal disease after TAS with or without IGL (68.0% vs. 57.6%, p = 0.2). CONCLUSIONS: IGL did not significantly change either the performance of TAS or the volume of residual nodal tumor burden. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03513614.

An Innovative Non-Linear Prediction Model for Clinical Benefit in Women with Newly Diagnosed Breast Cancer Using Baseline FDG-PET/CT and Clinical Data.

Objectives: We aimed to develop a novel non-linear statistical model integrating primary tumor features on baseline [18F]-fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT), molecular subtype, and clinical data for treatment benefit prediction in women with newly diagnosed breast cancer using innovative statistical techniques, as opposed to conventional methodological approaches. Methods: In this single-center retrospective study, we conducted a comprehensive assessment of women newly diagnosed with breast cancer who had undergone a FDG-PET/CT scan for staging prior to treatment. Primary tumor (PT) volume, maximum and mean standardized uptake value (SUVmax and SUVmean), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were measured on PET/CT. Clinical data including clinical staging (TNM) but also PT anatomical site, histology, receptor status, proliferation index, and molecular subtype were obtained from the medical records. Overall survival (OS), progression-free survival (PFS), and clinical benefit (CB) were assessed as endpoints. A logistic generalized additive model was chosen as the statistical approach to assess the impact of all listed variables on CB. Results: 70 women with newly diagnosed breast cancer (mean age 63.3 ± 15.4 years) were included. The most common location of breast cancer was the upper outer quadrant (40.0%) in the left breast (52.9%). An invasive ductal adenocarcinoma (88.6%) with a high tumor proliferation index (mean ki-67 expression 35.1 ± 24.5%) and molecular subtype B (51.4%) was by far the most detected breast tumor. Most PTs displayed on hybrid imaging a greater volume (12.8 ± 30.4 cm3) with hypermetabolism (mean ± SD of PT maximum SUVmax, SUVmean, MTV, and TLG, respectively: 8.1 ± 7.2, 4.9 ± 4.4, 12.7 ± 30.4, and 47.4 ± 80.2). Higher PT volume (p < 0.01), SUVmax (p = 0.04), SUVmean (p = 0.03), and MTV (<0.01) significantly compromised CB. A considerable majority of patients survived throughout this period (92.8%), while five women died (7.2%). In fact, the OS was 31.7 ± 14.2 months and PFS was 30.2 ± 14.1 months. A multivariate prediction model for CB with excellent accuracy could be developed using age, body mass index (BMI), T, M, PT TLG, and PT volume as predictive parameters. PT volume and PT TLG demonstrated a significant influence on CB in lower ranges; however, beyond a specific cutoff value (respectively, 29.52 cm3 for PT volume and 161.95 cm3 for PT TLG), their impact on CB only reached negligible levels. Ultimately, the absence of distant metastasis M displayed a strong positive impact on CB far ahead of the tumor size T (standardized average estimate 0.88 vs. 0.4). Conclusions: Our results emphasized the pivotal role played by FDG-PET/CT prior to treatment in forecasting treatment outcomes in women newly diagnosed with breast cancer. Nevertheless, careful consideration is required when selecting the methodological approach, as our innovative statistical techniques unveiled non-linear influences of predictive biomarkers on treatment benefit, highlighting also the importance of early breast cancer diagnosis.

Re-irradiation for head and neck cancer: outcome and toxicity analysis using a prospective single institution database.

Purpose: Re-irradiation (re-RT) in head and neck cancer is challenging. This study prospectively explored the feasibility of re-RT in patients with loco-regionally recurrent or second primary head and neck cancer (LRR/SP HNC). Methods: From 2004 to 2021, 61 LRR/SP HNC patients were treated with re-RT, defined as having a second course of RT with curative intent resulting in a cumulative dose of ≥100 Gy in an overlapping volume. Postoperative or definitive dynamic intensity-modulated and/or volumetric modulated re-RT was administered using twice daily hyperfractionation to 60 Gy combined with cisplatin or carboplatin/5-fluorouracil. Overall survival (OS), progression-free survival (PFS), locoregional control (LRC) and distant metastasis control (DMC) were analyzed and prognostic factors evaluated. Toxicity was prospectively recorded and graded. Results: The median follow-up was 9.8 months. In 41 patients (67.1%), complete administration of the intended treatment was not feasible. In 9 patients (15%) re-RT was interrupted prematurely and in other 9, the complete re-RT dose was lower than 60 Gy, and 37 patients (61%) could not receive or complete chemotherapy. Two-year OS, PFS and LRC rates were 19%, 18% and 30%, respectively. 20 patients (33%) received the complete intended treatment, and 1- and 2-year OS rates were 70% and 47%, respectively. Charlson comorbidity index was an important predictor for treatment completion. Multivariate analysis revealed recurrent N stage 0-1, age, chemotherapy administration and re-RT dose of 60 Gy as prognostic factors for clinical outcomes. No grade 5 re-RT-related toxicity was observed. The most common new grade ≥3 acute toxicities were dysphagia (52%) and mucositis (46%). Late toxicity included grade ≥3 dysphagia in 5% and osteoradionecrosis in 10% of evaluable patients, respectively. 6 patients (10%) were alive after 9 years without progression and no late toxicity grade ≥3, except for 2 patients presenting with osteoradionecrosis. Conclusion: Hyperfractionated re-RT with 60 Gy combined with platinum-based chemotherapy was a curative treatment option with acceptable toxicity in LRR/SP patients. Patients with higher comorbidity had a higher probability of failing to receive and complete the intended therapy. Consequently, they derived unsatisfactory benefits from re-RT, highlighting the importance of patient selection.

Dose-Volume Histogram Parameters and Quality of Life in Patients with Prostate Cancer Treated with Surgery and High-Dose Volumetric-Intensity-Modulated Arc Therapy to the Prostate Bed.

INTRODUCTION: Prostate bed radiotherapy (RT) is a major affecter of patients' long-term quality of life (QoL). To ensure the best possible outcome of these patients, dose constraints are key for optimal RT planning and delivery. However, establishing refined dose constraints requires access to patient-level data. Therefore, we aimed to provide such data on the relationship between OAR and gastrointestinal (GI) as well as genitourinary (GU) QoL outcomes of a homogenous patient cohort who received dose-intensified post-operative RT to the prostate bed. Furthermore, we aimed to conduct an exploratory analysis of the resulting data. METHODS: Patients who were treated with prostate bed RT between 2010 and 2020 were inquired about their QoL based on the Expanded Prostate Cancer Index Composite (EPIC). Those (n = 99) who received volumetric arc therapy (VMAT) of at least 70 Gy to the prostate bed were included. Dose-volume histogram (DVH) parameters were gathered and correlated with the EPIC scores. RESULTS: The median age at the time of prostate bed RT was 68.9 years, and patients were inquired about their QoL in the median 2.3 years after RT. The median pre-RT prostate-specific antigen (PSA) serum level was 0.35 ng/mL. The median duration between surgery and RT was 1.5 years. The median prescribed dose to the prostate bed was 72 Gy. A total of 61.6% received prostate bed RT only. For the bladder, the highest level of statistical correlation (p < 0.01) was seen for V10-20Gy, Dmean and Dmedian with urinary QoL. For bladder wall, the highest level of statistically significant correlation (p < 0.01) was seen for V5-25Gy, Dmean and Dmedian with urinary QoL. Penile bulb V70Gy was statistically significantly correlated with sexual QoL (p < 0.05). A larger rectal volume was significantly correlated with improved bowel QoL (p < 0.05). Sigmoid and urethral DVH parameters as well as the surgical approach were not statistically significantly correlated with QoL. CONCLUSION: Specific dose constraints for bladder volumes receiving low doses seem desirable for the further optimization of prostate bed RT. This may be particularly relevant in the context of the aspiration of establishing focal RT of prostate cancer and its local recurrences. Our comprehensive dataset may aid future researchers in achieving these goals.

Association of Axillary Dissection With Systemic Therapy in Patients With Clinically Node-Positive Breast Cancer.

Importance: The role of axillary lymph node dissection (ALND) to determine nodal burden to inform systemic therapy recommendations in patients with clinically node (cN)-positive breast cancer (BC) is currently unknown. Objective: To address the association of ALND with systemic therapy in cN-positive BC in the upfront surgery setting and after neoadjuvant chemotherapy (NACT). Design, Setting, and Participants: This was a prospective, observational, cohort study conducted from August 2018 to June 2022. This was a preplanned study within the phase 3 randomized clinical OPBC-03/TAXIS trial. Included were patients with confirmed cN-positive BC from 44 private, public, and academic breast centers in 6 European countries. After NACT, residual nodal disease was mandatory, and a minimum follow-up of 2 months was required. Exposures: All patients underwent tailored axillary surgery (TAS) followed by ALND or axillary radiotherapy (ART) according to TAXIS randomization. TAS removed suspicious palpable and sentinel nodes, whereas imaging-guidance was optional. Systemic therapy recommendations were at the discretion of the local investigators. Results: A total of 500 patients (median [IQR] age, 57 [48-69] years; 487 female [97.4%]) were included in the study. In the upfront surgery setting, 296 of 335 patients (88.4%) had hormone receptor (HR)-positive and Erb-B2 receptor tyrosine kinase 2 (ERBB2; formerly HER2 or HER2/neu)-negative disease: 145 (49.0%) underwent ART, and 151 (51.0%) underwent ALND. The median (IQR) number of removed positive lymph nodes without ALND was 3 (1-4) nodes compared with 4 (2-9) nodes with ALND. There was no association of ALND with the proportion of patients undergoing adjuvant chemotherapy (81 of 145 [55.9%] vs 91 of 151 [60.3%]; adjusted odds ratio [aOR], 0.72; 95% CI, 0.19-2.67) and type of systemic therapy. Of 151 patients with NACT, 74 (51.0%) underwent ART, and 77 (49.0%) underwent ALND. The ratio of removed to positive nodes was a median (IQR) of 4 (3-7) nodes to 2 (1-3) nodes and 15 (12-19) nodes to 2 (1-5) nodes in the ART and ALND groups, respectively. There was no observed association of ALND with the proportion of patients undergoing postneoadjuvant systemic therapy (57 of 74 [77.0%] vs 55 of 77 [71.4%]; aOR, 0.86; 95% CI, 0.43-1.70), type of postneoadjuvant chemotherapy (eg, capecitabine: 10 of 74 [13.5%] vs 10 of 77 [13.0%]; trastuzumab emtansine-DM1: 9 of 74 [12.2%] vs 11 of 77 [14.3%]), or endocrine therapy (eg, aromatase inhibitors: 41 of 74 [55.4%] vs 36 of 77 [46.8%]; tamoxifen: 8 of 74 [10.8%] vs 6 of 77 [7.8%]). Conclusion: Results of this cohort study suggest that patients without ALND were significantly understaged. However, ALND did not inform systemic therapy recommendations.

Trends in use of neoadjuvant systemic therapy in patients with clinically node-positive breast cancer in Europe: prospective TAXIS study (OPBC-03, SAKK 23/16, IBCSG 57-18, ABCSG-53, GBG 101).

PURPOSE: The aim of this study was to evaluate clinical practice heterogeneity in use of neoadjuvant systemic therapy (NST) for patients with clinically node-positive breast cancer in Europe. METHODS: The study was preplanned in the international multicenter phase-III OPBC-03/TAXIS trial (ClinicalTrials.gov Identifier: NCT03513614) to include the first 500 randomized patients with confirmed nodal disease at the time of surgery. The TAXIS study's pragmatic design allowed both the neoadjuvant and adjuvant setting according to the preferences of the local investigators who were encouraged to register eligible patients consecutively. RESULTS: A total of 500 patients were included at 44 breast centers in six European countries from August 2018 to June 2022, 165 (33%) of whom underwent NST. Median age was 57 years (interquartile range [IQR], 48-69). Most patients were postmenopausal (68.4%) with grade 2 and 3 hormonal receptor-positive and human epidermal growth factor receptor 2-negative breast cancer with a median tumor size of 28 mm (IQR 20-40). The use of NST varied significantly across the countries (p < 0.001). Austria (55.2%) and Switzerland (35.8%) had the highest percentage of patients undergoing NST and Hungary (18.2%) the lowest. The administration of NST increased significantly over the years (OR 1.42; p < 0.001) and more than doubled from 20 to 46.7% between 2018 and 2022. CONCLUSION: Substantial heterogeneity in the use of NST with HR+/HER2-breast cancer exists in Europe. While stringent guidelines are available for its use in triple-negative and HER2+ breast cancer, there is a need for the development of and adherence to well-defined recommendations for HR+/HER2-breast cancer.

Radiomics and machine learning for predicting the consistency of benign tumors of the central nervous system: A systematic review.

PURPOSE: Predicting the consistency of benign central nervous system (CNS) tumors prior to surgery helps to improve surgical outcomes. This review summarizes and analyzes the literature on using radiomics and/or machine learning (ML) for consistency prediction. METHOD: The Medical Literature Analysis and Retrieval System Online (MEDLINE) database was screened for studies published in English from January 1st 2000. Data was extracted according to the PRISMA guidelines and quality of the studies was assessed in compliance with the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). RESULTS: Eight publications were included focusing on pituitary macroadenomas (n = 5), pituitary adenomas (n = 1), and meningiomas (n = 2) using a retrospective (n = 6), prospective (n = 1), and unknown (n = 1) study design with a total of 763 patients for the consistency prediction. The studies reported an area under the curve (AUC) of 0.71-0.99 for their respective best performing model regarding the consistency prediction. Of all studies, four articles validated their models internally whereas none validated their models externally. Two articles stated making data available on request with the remaining publications lacking information with regard to data availability. CONCLUSIONS: The research on consistency prediction of CNS tumors is still at an early stage regarding the use of radiomics and different ML techniques. Best-practice procedures regarding radiomics and ML need to be followed more rigorously to facilitate the comparison between publications and, accordingly, the possible implementation into clinical practice in the future.

The Apple Watch spO2 sensor and outliers in healthy users.

It is unclear how frequently the Apple Watch produces spO2 measurements outside of the normal range in healthy individuals at rest. We conducted a head-to-head comparison in 38 healthy individuals between two watches and two medical-grade pulse oximeters. Fourteen percent of watch measurements yielded spO2 values below 95%, with no values below 92%. Results suggest that outliers measured by the watch should not be a cause for concern in otherwise healthy individuals.

Applications of Machine Learning in Palliative Care: A Systematic Review.

Objective: To summarize the available literature on using machine learning (ML) for palliative care practice as well as research and to assess the adherence of the published studies to the most important ML best practices. Methods: The MEDLINE database was searched for the use of ML in palliative care practice or research, and the records were screened according to PRISMA guidelines. Results: In total, 22 publications using machine learning for mortality prediction (n = 15), data annotation (n = 5), predicting morbidity under palliative therapy (n = 1), and predicting response to palliative therapy (n = 1) were included. Publications used a variety of supervised or unsupervised models, but mostly tree-based classifiers and neural networks. Two publications had code uploaded to a public repository, and one publication uploaded the dataset. Conclusions: Machine learning in palliative care is mainly used to predict mortality. Similarly to other applications of ML, external test sets and prospective validations are the exception.

Accuracy of the Apple Watch Oxygen Saturation Measurement in Adults: A Systematic Review.

The purpose of this review is to summarize the research on the accuracy of oxygen saturation (spO2) measurements using the Apple Watch (Apple Inc., Cupertino, California). The Medline and Google Scholar databases were searched for papers evaluating the spO2 measurements of the Apple Watch vs. any kind of ground truth and records were analyzed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The five publications with 973 total patients that met the inclusion criteria all used the Apple Watch Series 6 and described 95% limits of agreement of +/- 2.7 to 5.9% spO2. However, outliers of up to 15% spO2 were reported. Only one study had patient-level data uploaded to a public repository. The Apple Watch Series 6 does not show a strong systematic bias compared to conventional, medical-grade pulse oximeters. However, outliers do occur and should not cause concern in otherwise healthy individuals. The impact of race on measurement accuracy should be investigated.

Machine Learning for the Detection and Segmentation of Benign Tumors of the Central Nervous System: A Systematic Review.

Objectives: To summarize the available literature on using machine learning (ML) for the detection and segmentation of benign tumors of the central nervous system (CNS) and to assess the adherence of published ML/diagnostic accuracy studies to best practice. Methods: The MEDLINE database was searched for the use of ML in patients with any benign tumor of the CNS, and the records were screened according to PRISMA guidelines. Results: Eleven retrospective studies focusing on meningioma (n = 4), vestibular schwannoma (n = 4), pituitary adenoma (n = 2) and spinal schwannoma (n = 1) were included. The majority of studies attempted segmentation. Links to repositories containing code were provided in two manuscripts, and no manuscripts shared imaging data. Only one study used an external test set, which raises the question as to whether some of the good performances that have been reported were caused by overfitting and may not generalize to data from other institutions. Conclusions: Using ML for detecting and segmenting benign brain tumors is still in its infancy. Stronger adherence to ML best practices could facilitate easier comparisons between studies and contribute to the development of models that are more likely to one day be used in clinical practice.

Cranial Radiotherapy for Prostate Cancer Patients With Brain Metastases Inaccessible to Stereotactic Radiotherapy.

BACKGROUND/AIM: To evaluate patients and treatment characteristics as well as clinical outcome in patients with intracranial metastases from prostate cancer (PCA) treated with palliative radiotherapy. PATIENTS AND METHODS: Fifteen patients treated for intracranial metastases of PCA were identified. The median age of patients was 69 years. 80% of patients received whole brain radiotherapy and 20% received partial brain radiotherapy. Clinical outcome was assessed. Univariate analysis was performed to analyze the impact of patient specific parameters on survival. RESULTS: There was no >G2 acute or any late toxicity. Median time from the first diagnosis of PCA to first diagnosis of intracranial metastases was 62 months (range=15-160 months). Median survival from first diagnosis of intracranial metastases was 14 weeks (range=0-126 weeks) and 6 weeks (range=0-47 weeks) from the start of radiotherapy. In univariate analysis, survival was significantly better for patients with an Eastern Cooperative Oncology Group (ECOG) performance status 1 compared to ECOG 2-3 [18 weeks (range=5-47 weeks) vs. 3 weeks (range=0-21 weeks), p=0.030] and Recursive Partitioning Analysis (RPA) class 2 compared to RPA class 3 [18 weeks (range=5-47 weeks) vs. 6 weeks (range=0-21 weeks), p=0.045]. CONCLUSION: Overall survival of the patients with wide-spread intracranial metastases from PCA was poor. The decision for a radiotherapy should be done on individual patient basis.

Convolutional Neural Networks to Detect Vestibular Schwannomas on Single MRI Slices: A Feasibility Study.

In this study. we aimed to detect vestibular schwannomas (VSs) in individual magnetic resonance imaging (MRI) slices by using a 2D-CNN. A pretrained CNN (ResNet-34) was retrained and internally validated using contrast-enhanced T1-weighted (T1c) MRI slices from one institution. In a second step, the model was externally validated using T1c- and T1-weighted (T1) slices from a different institution. As a substitute, bisected slices were used with and without tumors originating from whole transversal slices that contained part of the unilateral VS. The model predictions were assessed based on the categorical accuracy and confusion matrices. A total of 539, 94, and 74 patients were included for training, internal validation, and external T1c validation, respectively. This resulted in an accuracy of 0.949 (95% CI 0.935-0.963) for the internal validation and 0.912 (95% CI 0.866-0.958) for the external T1c validation. We suggest that 2D-CNNs might be a promising alternative to 2.5-/3D-CNNs for certain tasks thanks to the decreased demand for computational power and the fact that there is no need for segmentations. However, further research is needed on the difference between 2D-CNNs and more complex architectures.

Classifying the Acquisition Sequence for Brain MRIs Using Neural Networks on Single Slices.

Background Neural networks for analyzing MRIs are oftentimes trained on particular combinations of perspectives and acquisition sequences. Since real-world data are less structured and do not follow a standard denomination of acquisition sequences, this impedes the transition from deep learning research to clinical application. The purpose of this study is therefore to assess the feasibility of classifying the acquisition sequence from a single MRI slice using convolutional neural networks. Methods A total of 113 MRI slices from 52 patients were used in a transfer learning approach to train three convolutional neural networks of different complexities to predict the acquisition sequence, while 27 slices were used for internal validation. The model then underwent external validation on 600 slices from 273 patients belonging to one of four classes (T1-weighted without contrast enhancement, T1-weighted with contrast enhancement, T2-weighted, and diffusion-weighted). Categorical accuracy was noted, and the results of the predictions for the validation set are provided with confusion matrices. Results The neural networks achieved a categorical accuracy of 0.79, 0.81, and 0.84 on the external validation data. The implementation of Grad-CAM showed no clear pattern of focus except for T2-weighted slices, where the network focused on areas containing cerebrospinal fluid. Conclusion Automatically classifying the acquisition sequence using neural networks seems feasible and could be used to facilitate the automatic labelling of MRI data.

Adherence to Contouring and Treatment Planning Requirements Within a Multicentric Trial: Results of the Quality Assurance of the SAKK 09/10 trial.

PURPOSE: To evaluate the results of the radiation therapy (RT) quality assurance (QA) program of the phase 3 randomized SAKK 09/10 trial in patients with biochemically recurrent prostate cancer after prostatectomy. METHODS AND MATERIALS: Within the Schweizerische Arbeitsgemeinschaft für Klinische Krebsforschung (SAKK) 09/10 trial testing 64-Gy versus 70-Gy salvage RT, a central collection of treatment plans was performed and thoroughly reviewed by a dedicated medical physicist and radiation oncologist. Adherence to the treatment protocol and specifically to the European Organization for the Research and Treatment of Cancer (EORTC) guidelines for target volume definition (classified as deviation observed yes vs no) and its potential correlation with acute and late toxicity (Common Terminology Criteria for Adverse Events version 4.0) and freedom from biochemical progression (FFBP) were investigated. RESULTS: The treatment plans for 344 patients treated between February 2011 and April 2014 depicted important deviations from the EORTC guidelines and the recommendations per trial protocol. For example, in up to half of the cases, the delineated structures deviated from the protocol (eg, prostate bed in 48.8%, rectal wall [RW] in 41%). In addition, variations in clinical target volume (CTV) and planning target volume (PTV) occurred frequently (eg, CTV and PTV deviations in up to 42.4% and 25.9%, respectively). The detected deviations showed a significant association with a lower risk of grade ≥2 gastrointestinal acute toxicity when the CTV did not overlap the RW versus when the CTV overlapped the RW (odds ratio [OR], 0.43; 95% confidence interval [CI], 0.22-0.85; P = .014), and a higher rate of grade ≥2 late genitourinary (GU) toxicity when the CTV overlapped the RW (OR, 2.58; 95% CI, 1.17-5.72; P = .019). A marginally significant lower risk of grade ≥2 late GU toxicity was observed when the prostate bed did not overlap versus did overlap the RW (OR, 0.51; 95% CI, 0.25-1.03; P = .06). In addition, a marginally significant decrease in FFBP was observed in patients with PTV not including surgical clips as potential markers of the limits of the prostate bed (hazard ratio, 1.44; 95% CI, 0.96-2.17; P = .07). CONCLUSIONS: Despite a thorough QA program, the central review of a phase 3 trial showed limited adherence to treatment protocol recommendations, which was associated with a higher risk of toxicity by means of acute or late gastrointestinal or GU toxicity and showed a trend toward worse FFBP. Data from this QA review might help to refine future QA programs and prostate bed delineation guidelines.

Convolutional Neural Networks for Classifying Laterality of Vestibular Schwannomas on Single MRI Slices-A Feasibility Study.

Introduction: Many proposed algorithms for tumor detection rely on 2.5/3D convolutional neural networks (CNNs) and the input of segmentations for training. The purpose of this study is therefore to assess the performance of tumor detection on single MRI slices containing vestibular schwannomas (VS) as a computationally inexpensive alternative that does not require the creation of segmentations. Methods: A total of 2992 T1-weighted contrast-enhanced axial slices containing VS from the MRIs of 633 patients were labeled according to tumor location, of which 2538 slices from 539 patients were used for training a CNN (ResNet-34) to classify them according to the side of the tumor as a surrogate for detection and 454 slices from 94 patients were used for internal validation. The model was then externally validated on contrast-enhanced and non-contrast-enhanced slices from a different institution. Categorical accuracy was noted, and the results of the predictions for the validation set are provided with confusion matrices. Results: The model achieved an accuracy of 0.928 (95% CI: 0.869-0.987) on contrast-enhanced slices and 0.795 (95% CI: 0.702-0.888) on non-contrast-enhanced slices from the external validation cohorts. The implementation of Gradient-weighted Class Activation Mapping (Grad-CAM) revealed that the focus of the model was not limited to the contrast-enhancing tumor but to a larger area of the cerebellum and the cerebellopontine angle. Conclusions: Single-slice predictions might constitute a computationally inexpensive alternative to training 2.5/3D-CNNs for certain detection tasks in medical imaging even without the use of segmentations. Head-to-head comparisons between 2D and more sophisticated architectures could help to determine the difference in accuracy, especially for more difficult tasks.