Search

Purpose: Pancreatic stereotactic body radiation therapy (SBRT) is limited to gross tumor without elective coverage for subclinical disease. Given a better understanding of recurrence patterns, we hypothesized that the addition of elective nodal irradiation (ENI) to pancreatic SBRT would be tolerable and would decrease locoregional progression.Methods and Materials: We conducted a retrospective 1:2 propensity-matched cohort study to compare toxicity and locoregional progression among patients treated with pancreatic SBRT with or without ENI. In the SBRT + ENI cohort, an elective target volume was delineated per Radiation Therapy Oncology Group guidelines and treated to 25 Gy in 5 fractions alongside 40 Gy in 5 fractions to gross disease. The primary outcome was the cumulative incidence of locoregional progression, with death as a competing risk.Results: Among 135 candidate controls treated with SBRT alone, 100 were propensity-matched to 50 patients treated with SBRT + ENI. All patients completed SBRT. Median potential radiographic follow-up was 28 months. The incidence of late and serious acute toxicity was similar between matched cohorts. However, SBRT + ENI was associated with a statistically significant increase in acute grade 1 to 2 nausea (60% vs 20%, P < .001). The 24-month cumulative incidences of locoregional progression with and without ENI were 22.6% (95% confidence interval [CI], 10.0%-35.1%) versus 44.6% (95% CI, 34.8%-54.4%; multivariable-adjusted hazard ratio, 0.39; 95% CI, 0.18-0.87; P = .021). This was stable in sensitivity analyses of uniform prescription dose, multiagent chemotherapy, and resectability. There were fewer peripancreatic (0% vs 7%), porta hepatis (2% vs 7%), and peri-aortic/aortocaval (5% vs 12%) recurrences after SBRT + ENI, but no difference in survival.Conclusions: Pancreatic SBRT + ENI was tolerable and did not increase late or serious acute toxicity relative to a matched cohort undergoing SBRT alone, but did increase acute grade 1 to 2 nausea. The addition of ENI to SBRT was associated with decreased locoregional progression but not improved survival. Further studies are warranted to determine whether ENI offers meaningful benefit. [ABSTRACT FROM AUTHOR]

• SABR doses ≥40 Gy appear to be superior than lower doses. • Modified FOLFIRINOX use appears to be superior to gemcitabine-based regimens. • Combination of SABR ≥40 Gy and modified FOLFIRINOX demonstrated the best outcomes. • Future prospective studies should test this combination for unresectable pancreatic cancer. We aimed to report the long-term impact of modern chemotherapy and SABR dose regimens on oncologic outcomes of unresectable pancreatic adenocarcinoma (PA). We reviewed the treatment characteristics and outcomes of all patients who received multi-fraction SABR for unresectable PA between February 2007 and August 2018 at our institution. Time-to-events were calculated from date of diagnosis treating death as a competing risk. A total of 149 patients were identified. Median follow-up was 15 months (range: 5–47). Median SABR dose was 33 Gy (range: 20–45) delivered in 5 fractions in 143 patients, and 3 or 6 fractions in 6 patients. 107 patients (72%) received gemcitabine-based chemotherapy while 31 (21%) received modified FOLFIRINOX (mFFX). Median OS was 16 months (95% CI, 14–17), with a 1-year cumulative incidence of LF of 14%. The combination of SABR doses ≥40 Gy and mFFX (n = 21) showed a superior PFS and OS to the use of GEM-based chemotherapy with <40 Gy SABR doses (median PFS: 14 vs. 10 months, HR: 0.46, 95% CI: 0.29–0.71, P = 0.003; median OS: 24 vs. 14 months, HR: 0.36, 95% CI: 0.22–0.59, P = 0.002), with 1-year PFS and OS of 67% and 90% compared to 35% and 59% for those who received GEM-based chemotherapy with <40 Gy SABR doses, respectively. The use of mFFX and a SABR dose ≥40 Gy in 5 fractions may be superior compared to regimens that utilize gemcitabine-based chemotherapy or SABR doses <40 Gy. [ABSTRACT FROM AUTHOR]

Purpose: Radiation therapy (RT) is prescribed for curative and palliative treatment for around 50% of patients with solid tumors. Radiation‐induced toxicities of healthy organs accompany many RTs and represent one of the main limiting factors during dose delivery. The existing RT planning solutions generally discard spatial dose distribution information and lose the ability to recognize radiosensitive regions of healthy organs potentially linked to toxicity manifestation. This study proposes a universal deep learning‐based algorithm for recognitions of consistent dose patterns and generation of toxicity risk maps for the abdominal area. Methods: We investigated whether convolutional neural networks (CNNs) can automatically associate abdominal computed tomography (CT) images and RT dose plans with post‐RT toxicities without being provided segmentation of abdominal organs. The CNNs were also applied to study RT plans, where doses at specific anatomical regions were reduced/increased, with the aim to pinpoint critical regions sparing of which significantly reduces toxicity risks. The obtained risk maps were computed for individual anatomical regions inside the liver and statistically compared to the existing clinical studies. Results: A database of 122 liver stereotactic body RT (SBRT) executed at Stanford Hospital from July 2004 and November 2015 was assembled. All patients treated for primary liver cancer, mainly hepatocellular carcinoma and cholangiocarcinoma, with complete follow‐ups were extracted from the database. The SBRT treatment doses ranged from 26 to 50 Gy delivered in 1–5 fractions for primary liver cancer. The patients were followed up for 1–68 months depending on the survival time. The CNNs were trained to recognize acute and late grade 3+ biliary stricture/obstruction, hepatic failure or decompensation, hepatobiliary infection, liver function test (LFT) elevation or/and portal vein thrombosis, named for convenience hepatobiliary (HB) toxicities. The toxicity prediction accuracy was of 0.73 measured in terms of the area under the receiving operator characteristic curve. Significantly higher risk scores (P < 0.05) of HB toxicity manifestation were associated with irradiation for the hepatobiliary tract in comparison to the risk scores for liver segments I–VIII and portal vein. This observation is in strong agreement with anatomical and clinical expectations. Conclusion: In this work, we proposed and validated a universal deep learning‐based solution for the identification of radiosensitive anatomical regions. Without any prior anatomical knowledge, CNNs automatically recognized the importance of hepatobiliary tract sparing during liver SBRT. [ABSTRACT FROM AUTHOR]

To develop a framework for automated prediction of hepatobiliary (HB) toxicity after liver stereotactic body radiation therapy (SBRT). A newly recognized toxicity type, named central or HB liver toxicity, had been reported, manifestation of which strongly correlates with the dose delivered to portal vein (PV) during SBRT. We propose a novel framework for automated HB toxicity prediction by combining deep learning-based auto-segmentation, PV anatomy analysis and the previously reported HB toxicity model. For validation of the framework, an IBR approved representative database of 72 patients treated with SBRT from primary (37) and metastatic (35) liver cancer was assembled. Each case included a pre-treatment CT, manual segmentations of tumor and PV, approved treatment plan, and the record of acute and late post-treatment toxicities. Performance of the developed framework was evaluated by quantitative comparison against manual predictions of HB toxicity, as well as post-treatment toxicity follow-ups. The manual and automated predictions of HB toxicity were in agreement for 94% cases using either V BED10 30 ≥ 45 cc or V BED10 40 ≥ 37 cc dosimetric predictors. When compared to post-treatment follow-ups for primary liver cancer, the proposed automated framework made 86% and 83% correct predictions in comparison to 83% and 80% correct manual predictions using V BED10 30 ≥ 45 cc or V BED10 40 ≥ 37 cc, respectively. The proposed framework automates the HB toxicity prediction with the accuracy similar to manual analysis-based HB toxicity prediction. The strategy is quite general and extendable to the automated prediction of toxicities of other organs. [ABSTRACT FROM AUTHOR]

Our aim was to report physician- and patient-reported outcomes of patients with localized breast cancer treated with moderate versus ultrahypofractionated whole breast irradiation (WBI) after breast-conserving surgery (BCS). Between February 2018 and February 2020, patients with localized breast cancer (pT0-3 pN0-1 M0) were offered participation in a phase 3 randomized clinical trial assessing adjuvant moderate hypofractionation (MHF) to 40 Gy in 15 fractions versus ultrahypofractionation (UHF) to 25 Gy in 5 fractions after BCS, with an optional simultaneously integrated boost. Toxicities, cosmesis, and quality of life were assessed at baseline, end of treatment (EOT), and 3 months, 1 year, 2 years, and 3 years from irradiation using validated metric tools. One hundred seven patients were randomized to MHF (n = 54) or UHF (n = 53) adjuvant WBI. The median follow-up was 42.8 months. Grade 2 radiation dermatitis was experienced by 4 patients (7.4%) in the MHF arm and 2 patients (3.7%) in the UHF arm at EOT (P =.726). No grade 3 or higher toxicities were observed. Deterioration of cosmesis by physician assessment was observed in 2 (6.7%) patients treated in the UHF arm and 1 (1.9%) patient treated in the MHF arm at EOT (P =.534), whereas at 3 months, only 1 (1.8%) patient treated in the MHF arm demonstrated deterioration of cosmesis (P =.315). At EOT, 91% and 94% of patients reported excellent/good cosmesis among those treated with MHF and UHF regimens, respectively (P =.550). At 3 months, more patients within the MHF arm reported excellent/good cosmesis compared with those in the UHF arm (100% vs 91%; P =.030). However, the difference in patient-reported cosmesis disappeared at the 1-, 2-, and 3-year time points. UHF WBI showed similar treatment-related late toxicities and similar provider-scored cosmesis compared with MHF radiation in patients treated adjuvantly after BCS. [ABSTRACT FROM AUTHOR]

Background: Accurate prediction of radiation toxicity of healthy organs‐at‐risks (OARs) critically determines the radiation therapy (RT) success. The existing dose–volume histogram‐based metric may grossly under/overestimate the therapeutic toxicity after 27% in liver RT and 50% in head‐and‐neck RT. We propose the novel paradigm for toxicity prediction by leveraging the enormous potential of deep learning and go beyond the existing dose/volume histograms. Experimental Design: We employed a database of 125 liver stereotactic body RT (SBRT) cases with follow‐up data to train deep learning‐based toxicity predictor. Convolutional neural networks (CNNs) were applied to discover the consistent patterns in 3D dose plans associated with toxicities. To enhance the predicting power, we first pretrain the CNNs with transfer learning from 3D CT images of 2644 human organs. CNNs were then trained on liver SBRT cases. Furthermore, nondosimetric pretreatment features, such as patients' demographics, underlying liver diseases, liver‐directed therapies, were inputted into the fully connected neural network for more comprehensive prediction. The saliency maps of CNNs were used to estimate the toxicity risks associated with irradiation of anatomical regions of specific OARs. In addition, we applied machine learning solutions to map numerical pretreatment features with hepatobiliary toxicity manifestation. Results: Among 125 liver SBRT patients, 58 were treated for liver metastases, 36 for hepatocellular carcinoma, 27 for cholangiocarcinoma, and 4 for other histologies. We observed that CNN we able to achieve accurate hepatobiliary toxicity prediction with the AUC of 0.79, whereas combining CNN for 3D dose plan analysis and fully connected neural networks for numerical feature analysis resulted in AUC of 0.85. Deep learning produces almost two times fewer false‐positive toxicity predictions in comparison to DVH‐based predictions, when the number of false negatives, i.e., missed toxicities, was minimized. The CNN saliency maps automatically estimated the toxicity risks for portal vein (PV) regions. We discovered that irradiation of the proximal portal vein is associated with two times higher toxicity risks (risk score: 0.66) that irradiation of the left portal vein (risk score: 0.31). Conclusions: The framework offers clinically accurate tools for hepatobiliary toxicity prediction and automatic identification of anatomical regions that are critical to spare during SBRT. [ABSTRACT FROM AUTHOR]

With the rapid development of imaging modalities and surgical techniques, the clinical entity representing tumors that are intermediate between resectable and unresectable pancreatic adenocarcinoma has been identified has been termed "borderline resectable" (BR). These tumors are generally amenable for resection but portend an increased risk for positive margins after surgery and commonly necessitate vascular resection and reconstruction. Although there is a lack of consensus regarding the appropriate definition of what constitutes a BR pancreatic tumor, it has been demonstrated that this intermediate category carries a particular prognosis that is in between resectable and unresectable disease. In order to downstage the tumor and increase the probability of clear surgical margins, neoadjuvant therapy is being increasingly utilized and studied. There is a lack of high-level evidence to establish the optimal treatment regimen for BR tumors. When resection with negative margins is achieved after neoadjuvant therapy, the prognosis for BR tumors approaches and even exceeds that for resectable disease. This review presents the current definitions, different treatment approaches, and the clinical outcomes of BR pancreatic cancer. [ABSTRACT FROM AUTHOR]

Purpose To further explore the correlation of central biliary tract (cHBT) radiation doses with hepatobiliary toxicity (HBT) after stereotactic body radiation therapy (SBRT) in a larger patient dataset. Methods We reviewed the treatment and outcomes of all patients who received SBRT for primary liver cancer (PLC) and metastatic liver tumors between July 2004 and November 2015 at our institution. The cHBT was defined as isotropic expansions (5, 10, 15, 20 and 25 mm) from the portal vein (PV). Doses were converted to biologically effective doses by using the standard linear quadratic model with α / β of 10 (BED 10 ). HBT was graded according to the Common Terminology Criteria for Adverse Events v4.03. Results Median follow-up was 13 months. Out of the 130 patients with complete follow-up records analyzed, 60 (46.1%) had liver metastases, 40 (30.8%) had hepatocellular carcinoma (HCC), 26 (20%) had cholangiocarcinoma (CCA) and 4 (3.1%) patients other PLC histologies. Thirty-three (25.4%) grade 2+ and 28 (21.5%) grade 3+ HBT were observed. Grade 3+ HBT was seen in 13 patients (50%) with CCA, 7 patients (17.5%) with HCC and 7 (11.7%) patients with liver metastases. SBRT doses to the cHBT were highly associated with HBT, but only for PLC patients when analyzed by histological subtype. The 15 mm expansion from the PV (cHBT 15 ) proved to be an appropriate surrogate for the cHBT. The strongest cHBT 15 dose predictors for G3+ HBT for PLC were the V BED10 40 ⩾37 cc ( p < 0.0001) and the V BED10 30 ⩾ 45 cc ( p < 0.0001). Conclusion SBRT doses to the cHBT are associated with occurrence of HBT only in PLC patients. Limiting the dose to the cHBT to V BED10 40 < 37 cc and V BED10 30 < 45 cc when treating PLC patients with SBRT may reduce the risk of HBT. [ABSTRACT FROM AUTHOR]

Purpose: Deep learning is an emerging technique that allows us to capture imaging information beyond the visually recognizable level of a human being. Because of the anatomic characteristics and location, on-board target verification for radiation delivery to pancreatic tumors is a challenging task. Our goal was to use a deep neural network to localize the pancreatic tumor target on kV x-ray images acquired using an on-board imager for image guided radiation therapy.Methods and Materials: The network is set up in such a way that the input is either a digitally reconstructed radiograph image or a monoscopic x-ray projection image acquired by the on-board imager from a given direction, and the output is the location of the planning target volume in the projection image. To produce a sufficient number of training x-ray images reflecting the vast number of possible clinical scenarios of anatomy distribution, a series of changes were introduced to the planning computed tomography images, including deformation, rotation, and translation, to simulate inter- and intrafractional variations. After model training, the accuracy of the model was evaluated by retrospectively studying patients who underwent pancreatic cancer radiation therapy. Statistical analysis using mean absolute differences (MADs) and Lin's concordance correlation coefficient were used to assess the accuracy of the predicted target positions.Results: MADs between the model-predicted and the actual positions were found to be less than 2.60 mm in anteroposterior, lateral, and oblique directions for both axes in the detector plane. For comparison studies with and without fiducials, MADs are less than 2.49 mm. For all cases, Lin's concordance correlation coefficients between the predicted and actual positions were found to be better than 93%, demonstrating the success of the proposed deep learning for image guided radiation therapy.Conclusions: We demonstrated that markerless pancreatic tumor target localization is achievable with high accuracy by using a deep learning technique approach. [ABSTRACT FROM AUTHOR]

Background: Deep learning auto‐segmentation (DLAS) models have been adopted in the clinic; however, they suffer from performance deterioration owing to the clinical practice variability. Some commercial DLAS software provide an incremental retraining function that enables users to train a custom model using their institutional data to account for clinical practice variability. Purpose: This study was performed to evaluate and implement the commercial DLAS software with the incremental retraining function for definitive treatment of patients with prostate cancer in a multi‐user environment. Methods: CT‐based target organs and organs‐at‐risk (OAR) delineation of 215 prostate cancer patients were utilized. The performance of three commercial DLAS software built‐in models was validated with 20 patients. A retrained custom model was developed using 100 patients and evaluated on the remaining data (n = 115). Dice similarity coefficient (DSC), Hausdorff distance (HD), mean surface distance (MSD), and surface DSC (SDSC) were utilized for quantitative evaluation. A multi‐rater qualitative evaluation was blindly performed with a five‐level scale. Visual inspection was performed in consensus and non‐consensus unacceptable cases to identify the failure modes. Results: Three commercial DLAS vendor built‐in models achieved sub‐optimal performance in 20 patients. The retrained custom model had a mean DSC of 0.82 for prostate, 0.48 for seminal vesicles (SV), and 0.92 for rectum, respectively. This represents a significant improvement over the built‐in model with DSC of 0.73, 0.37, and 0.81 for the corresponding structures. Compared to the acceptance rate of 96.5% and consensus unacceptable rate (i.e., both reviewers rated as unacceptable) of 3.5% achieved by manual contours, the custom model achieved a 91.3% acceptance rate and 8.7% consensus unacceptable rate. The failure modes of retrained custom model were attributed to the following: cystogram (n = 2), hip prosthesis (n = 2), low dose rate brachytherapy seeds (n = 2), air in endorectal balloon(n = 1), non‐iodinated spacer (n = 2), and giant bladder(n = 1). Conclusion: The commercial DLAS software with the incremental retraining function was validated and clinically adopted for prostate patients in a multi‐user environment. AI‐based auto‐delineation of the prostate and OARs is shown to achieve improved physician acceptance, overall clinical utility, and accuracy. [ABSTRACT FROM AUTHOR]