Imaging-based Toxicity and Response Pattern Assessment Following CAR T-Cell Therapy

• Published in: Radiology Vol. 302; no. 2; pp. 438 - 445
• Main Authors: Smith, Daniel A, Kikano, Elias, Tirumani, Sree Harasha, de Lima, Marcos, Caimi, Paolo, Ramaiya, Nikhil H
• Format: Journal Article
• Language: English
• Published: United States 01.02.2022
• Subjects: Cell- and Tissue-Based Therapy - adverse effects; Cytokine Release Syndrome - diagnostic imaging; Cytokine Release Syndrome - etiology; Cytokine Release Syndrome - immunology; Female; Humans; Immunotherapy, Adoptive - adverse effects; Lymphoma, Large B-Cell, Diffuse - therapy; Male; Middle Aged; Neurotoxicity Syndromes - diagnostic imaging; Neurotoxicity Syndromes - etiology; Neurotoxicity Syndromes - immunology; Receptors, Chimeric Antigen - immunology; Retrospective Studies; Tomography, X-Ray Computed
• ISSN: 0033-8419; 1527-1315
• DOI: 10.1148/radiol.2021210760

Background Chimeric antigen receptor (CAR) T-cell immunotherapy is increasingly used for refractory lymphoma but may lead to cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS). Imaging may assist in clinical management. Associations between CRS or ICANS grade and imaging findings remain not fully established. Purpose To determine associations between imaging findings and clinical grade of CRS or ICANS, evaluate response patterns, and assess imaging use following CAR T-cell treatment. Materials and Methods Patients with refractory B-cell lymphoma who received CAR T-cell infusion between 2018 and 2020 at a single center were analyzed retrospectively. Clinical CRS or ICANS toxicity grade was assessed using American Society for Transplantation and Cellular Therapy, or ASTCT, consensus grading. Thoracic and head images (radiographs, CT scans, MRI scans) were evaluated. Associations between imaging findings and clinical CRS or ICANS grade were analyzed. Wilcoxon signed-rank and χ tests were used to assess associations between thoracic imaging findings, clinical CRS toxicity grade, and imaging-based response. Response to therapy was evaluated according to Deauville five-point scale criteria. Results A total of 38 patients (mean age ± standard deviation, 59 years ± 10; 23 men) who received CAR T-cell infusion were included. Of these, 24 (63% [95% CI: 48, 79]) and 11 (29% [95% CI: 14, 44]) experienced clinical grade 1 or higher CRS and ICANS, respectively. Patients with grade 2 or higher CRS were more likely to have thoracic images with abnormal findings (10 of 14 patients [71%; 95% CI: 47, 96] vs five of 24 patients [21%; 95% CI: 4, 37]; = .002) and more likely to have imaging evidence of pleural effusions (five of 14 [36%; 95% CI: 10, 62] vs two of 24 [8.3%; 95% CI: 0, 20]; = .04) and atelectasis (eight of 14 [57%; 95% CI: 30, 84] vs six of 24 [25%; 95% CI: 7, 43]; = .048). Positive imaging findings were identified in three of seven patients (43%) with grade 2 or higher ICANS who underwent neuroimaging. The best treatment response included 20 of 36 patients (56% [95% CI: 39, 72]) with complete response, seven of 36 (19% [95% CI: 6, 33]) with partial response, one of 36 (2.8% [95% CI: 0, 8]) with stable disease, and eight of 36 (22% [95% CI: 8, 36]) with progressive disease. Conclusion Thoracic imaging findings, including pleural effusions and atelectasis, correlated with cytokine release syndrome grade following chimeric antigen receptor (CAR) T-cell infusion. CAR T-cell therapy yielded high response rates. © RSNA, 2021 See also the editorial by Langer in this issue.