The value of 18F-fluorodeoxyglucose positron-emission-tomography-computed tomography (FDG-PET/CT) for staging patients with (very) high-risk non-muscle invasive bladder cancer (NMIBC) is unknown.... Show moreThe value of 18F-fluorodeoxyglucose positron-emission-tomography-computed tomography (FDG-PET/CT) for staging patients with (very) high-risk non-muscle invasive bladder cancer (NMIBC) is unknown. In this study among NMIBC patients referred for RC, FDG-PET/CT detected metastases that were not detected by CT, leading to treatment changes in 10% of patients. However, the use of FDG-PET/CT should be weighed against its disad-vantages, including false-positive lesions. Introduction and Objectives: 18F-fluorodeoxyglucose positron-emission tomography-computed tomography (FDG-PET/CT) is increasingly used in the preoperative staging of patients with muscle-invasive bladder cancer. The clinical added value of FDG-PET/CT in high-risk non-muscle invasive bladder cancer (NMIBC) is unknown. In this study, the value of FDG-PET/CT in addition to contrast enhanced (CE)-CT was evaluated in high-risk NMIBC before radical cystec-tomy (RC). Materials and Methods: This is a retrospective analysis of consecutive patients with high risk and very-high risk urothelial NMIBC scheduled for RC in a tertiary referral center between 2011 and 2020. Patients underwent staging with CE-CT (chest and abdomen/pelvis) and FDG-PET/CT. We assessed the clinical disease stage before and after FDG-PET/CT and the treatment recommendation based on the stage before and after FDG-PET/CT. The accuracy of CT and FDG-PET/CT for identifying metastatic disease was defined by the receiver-operating curve using a reference -standard including histopathology/cytology (if available), imaging and follow-up. Results: A total of 92 patients were identified (median age: 71 years). In 14/92 (15%) patients, FDG-PET/CT detected metastasis (12 suspicious lymph nodes and 4 distant metastases). The disease stage changed in 11/92 (12%) patients based on additional FDG-PET/CT findings. FDG-PET/CT led to a different treatment in 9/92 (10%) patients. According to the reference standard, 25/92 (27%) patients had metastases. The sensitivit y, specificit y and accuracy of FDG-PET/CT was 36%, 93% and 77% respectively, versus 12%, 97% and 74% of CE-CT only. The area under the ROC curve was 0.643 for FDG-PET/CT and 0.545 for CT, P = .036. Conclusion: The addition of FDG-PET/CT to CE-CT imaging changed the treatment in 10% of patients and proved to be a valuable diagnostic tool in a selected subgroup of NMIBC patients scheduled for RC. Show less
BackgroundCheckpoint inhibitors provided sustained clinical benefit to metastatic lung cancer patients. Nonetheless, prognostic markers in metastatic settings are still under research. Imaging... Show moreBackgroundCheckpoint inhibitors provided sustained clinical benefit to metastatic lung cancer patients. Nonetheless, prognostic markers in metastatic settings are still under research. Imaging offers distinctive advantages, providing whole-body information non-invasively, while routinely available in most clinics. We hypothesized that more prognostic information can be extracted by employing artificial intelligence (AI) for treatment monitoring, superior to 2D tumor growth criteria.MethodsA cohort of 152 stage-IV non-small-cell lung cancer patients (NSCLC) (73 discovery, 79 test, 903CTs), who received nivolumab were retrospectively collected. We trained a neural network to identify morphological changes on chest CT acquired during patients' follow-ups. A classifier was employed to link imaging features learned by the network with overall survival.ResultsOur results showed significant performance in the independent test set to predict 1-year overall survival from the date of image acquisition, with an average area under the curve (AUC) of 0.69 (p < 0.01), up to AUC 0.75 (p < 0.01) in the first 3 to 5 months of treatment, and 0.67 AUC (p = 0.01) for durable clinical benefit (6 months progression-free survival). We found the AI-derived survival score to be independent of clinical, radiological, PDL1, and histopathological factors. Visual analysis of AI-generated prognostic heatmaps revealed relative prognostic importance of morphological nodal changes in the mediastinum, supraclavicular, and hilar regions, lung and bone metastases, as well as pleural effusions, atelectasis, and consolidations.ConclusionsOur results demonstrate that deep learning can quantify tumor- and non-tumor-related morphological changes important for prognostication on serial imaging. Further investigation should focus on the implementation of this technique beyond thoracic imaging. Show less
