Predicting successful clinical candidates for fiducial-free lung tumor tracking with a deep learning binary classification model.

OBJECTIVES: The CyberKnife system is a robotic radiosurgery platform that allows the delivery of lung SBRT treatments using fiducial-free soft-tissue tracking. However, not all lung cancer patients are eligible for lung tumor tracking. Tumor size, density, and location impact the ability to successfully detect and track a lung lesion in 2D orthogonal X-ray images. The standard workflow to identify successful candidates for lung tumor tracking is called Lung Optimized Treatment (LOT) simulation, and involves multiple steps from CT acquisition to the execution of the simulation plan on CyberKnife. The aim of the study is to develop a deep learning classification model to predict which patients can be successfully treated with lung tumor tracking, thus circumventing the LOT simulation process. METHODS: Target tracking is achieved by matching orthogonal X-ray images with a library of digital radiographs reconstructed from the simulation CT scan (DRRs). We developed a deep learning model to create a binary classification of lung lesions as being trackable or untrackable based on tumor template DRR extracted from the CyberKnife system, and tested five different network architectures. The study included a total of 271 images (230 trackable, 41 untrackable) from 129 patients with one or multiple lung lesions. Eighty percent of the images were used for training, 10% for validation, and the remaining 10% for testing. RESULTS: For all five convolutional neural networks, the binary classification accuracy reached 100% after training, both in the validation and the test set, without any false classifications. CONCLUSIONS: A deep learning model can distinguish features of trackable and untrackable lesions in DRR images, and can predict successful candidates for fiducial-free lung tumor tracking.

Resection with intraoperative cesium-131 brachytherapy as salvage therapy for recurrent brain tumors.

OBJECTIVE: The authors' objective was to examine the safety and efficacy of salvage intracranial cesium-131 brachytherapy in combination with resection of recurrent brain tumors. METHODS: The authors conducted a retrospective chart review of consecutive patients treated with intraoperative intracranial cesium-131 brachytherapy at a single institution. Permanent suture-stranded cesium-131 seeds were implanted in the resection cavity after maximal safe tumor resection. The primary outcomes of interest were local, locoregional (within 1 cm), and intracranial control, as well as rates of overall survival (OS), neurological death, symptomatic adverse radiation effects (AREs), and surgical complication rate graded according to Common Terminology Criteria for Adverse Events version 5.0. RESULTS: Between 2016 and 2020, 36 patients received 40 consecutive cesium-131 implants for 42 recurrent brain tumors and received imaging follow-up for a median (interquartile range [IQR]) of 17.0 (12.7-25.9) months. Twenty patients (55.6%) with 22 implants were treated for recurrent brain metastasis, 12 patients (33.3%) with 16 implants were treated for recurrent atypical (n = 7) or anaplastic (n = 5) meningioma, and 4 patients (11.1%) were treated for other recurrent primary brain neoplasms. All except 1 tumor (97.6%) had received prior radiotherapy, including 20 (47.6%) that underwent 2 or more prior radiotherapy treatments and 23 (54.8%) that underwent prior resection. The median (IQR) tumor size was 3.0 (2.3-3.7) cm, and 17 lesions (40.5%) had radiographic evidence of ARE prior to salvage therapy. Actuarial 1-year local/locoregional/intracranial control rates for the whole cohort and patients with metastases and meningiomas were 91.6%/83.4%/47.9%, 88.8%/84.4%/45.4%, and 100%/83.9%/46.4%, respectively. No cases of local recurrence of any histology (0 of 27) occurred after gross-total resection (p = 0.012, log-rank test). The 1-year OS rates for the whole cohort and patients with metastases and meningiomas were 82.7%, 79.1%, and 91.7%, respectively, and the median (IQR) survival of all patients was 26.7 (15.6-36.4) months. Seven patients (19.4%) experienced neurological death from progressive intracranial disease (7 of 14 total deaths [50%]), 5 (13.9%) of whom died of leptomeningeal disease. Symptomatic AREs were observed in 9.5% of resection cavities (n = 4), of which 1 (2.4%) was grade 3 in severity. The surgical complication rate was 16.7% (n = 7); 4 (9.5%) of these patients had grade 3 or higher complications, including 1 patient (2.4%) who died perioperatively. CONCLUSIONS: Cesium-131 brachytherapy resulted in good local control and acceptable rates of symptomatic AREs and surgical complications in this heavily pretreated cohort, and it may be a reasonable salvage adjuvant treatment for this patient population.