Acta Radiologica | 2019
Response evaluation for immunotherapy through semi-automatic software based on RECIST 1.1, irRC, and iRECIST criteria: comparison with subjective assessment
Abstract
Background Pseudoprogression is difficult to diagnose in patients undergoing immunotherapy. Subjective response assessment is still common in clinical practice. Purpose To evaluate the differences between response evaluation criteria in solid tumors version 1.1 (RECIST 1.1), immune-related response criteria (irRC), and modified RECIST 1.1 for immunotherapy (iRECIST) through semi-automatic software, and to compare iRECIST-based response evaluation with subjective assessment. Material and Methods The best overall response of each patient based on RECIST 1.1, irRC, and iRECIST was determined on CT scans through semi-automatic software and the differences between the criteria were evaluated. Criteria-based response evaluation through semi-automatic software was compared with subjective assessment on radiology report by correlating the best overall response to overall survival. Results A total of 21 patients were included (five patients with melanoma, 12 patients with non-small-cell lung cancer, and four patients with hepatocellular carcinoma). Two patients with progressive disease by RECIST 1.1 but non-progressive disease by irRC and iRECIST eventually experienced tumor response and had favorable outcomes, indicating pseudoprogression. The survival difference between patients with non-progressive disease and progressive disease was better stratified through iRECIST-based response evaluation (P\u2009=\u20090.078) than that through subjective assessment (P\u2009=\u20090.501). Conclusion Pseudoprogression in immunotherapy may be captured through semi-automatic software utilizing irRC or iRECIST criteria. iRECIST-based response evaluation may provide a better survival stratification compared with subjective assessment.