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
BACKGROUND: Accurate diagnosis and staging are crucial to ensure uniform allocation to the optimal treatment methods for non-small cell lung cancer (NSCLC) patients, but may differ among... Show moreBACKGROUND: Accurate diagnosis and staging are crucial to ensure uniform allocation to the optimal treatment methods for non-small cell lung cancer (NSCLC) patients, but may differ among multidisciplinary tumor boards (MDTs). Discordance between clinical and pathologic TNM stage is particularly important for patients with locally advanced NSCLC (stage IIIA) because it may influence their chance of allocation to curative-intent treatment. We therefore aimed to study agreement on staging and treatment to gain insight into MDT decision-making.RESEARCH QUESTION: What is the level of agreement on clinical staging and treatment recommendations among MDTs in stage IIIA NSCLC patients?STUDY DESIGN AND METHODS: Eleven MDTs each evaluated the same 10 pathologic stage IIIA NSCLC patients in their weekly meeting (n = 110). Patients were selected purposively for their challenging nature. All MDTs received exactly the same clinical information and images per patient. We tested agreement in cT stage, cN stage, cM stage (TNM 8th edition), and treatment proposal among MDTs using Randolph's free-marginal multirater kappa.RESULTS: Considerable variation among the MDTs was seen in T staging (K, 0.55 [95% CI, 0.34-0.75]), N staging (K, 0.59 [95% CI, 0.35-0.83]), overall TNM staging (K, 0.53 [95% CI, 0.35-0.72]), and treatment recommendations (K, 0.44 [95% CI, 0.32-0.56]). Most variation in T stage was seen in patients with suspicion of invasion of surrounding structures, which influenced such treatment recommendations as induction therapy and type. For N stage, distinction between Ni and N2 disease was an important source of discordance among MDTs. Variation occurred between 2 patients even regarding M stage. A wide range of additional diagnostics was proposed by the MDTs.INTERPRETATION: This study demonstrated high variation in staging and treatment of patients with stage IIIA NSCLC among MDTs in different hospitals. Although some variation may be unavoidable in these challenging patients, we should strive for more uniformity. Show less
