The role of interim 18F-FDG PET/CT in prediction of response to ipilimumab treatment in metastatic melanoma.

Abstract

The aim of the present study was to assess the value of interim 18F-FDG PET/CT performed after the first two cycles of ipilimumab treatment in the prediction of the final clinical response to this type of immunotherapy. The study group comprised 41 patients with unresectable metastatic melanoma scheduled for ipilimumab therapy. Whole-body 18F-FDG PET/CT was performed before the start of ipilimumab treatment (baseline PET/CT) and after the initial two cycles of ipilimumab treatment (interim PET/CT). Evaluation of patient response to treatment was based on the European Organization for Research and Treatment of Cancer (EORTC) 1999 criteria for PET as well as the recently proposed PET Response Evaluation Criteria for Immunotherapy (PERCIMT). The patients' best clinical response, assessed at a median of 21.4months (range 6.3-41.9months) was used as reference. According to their best clinical response, the patients were divided into two groups: those showing clinical benefit (CB) including stable disease, partial response and complete response (31 patients), and those showing no clinical benefit (no-CB including progressive disease (10 patients). According to the EORTC criteria, interim PET/CT demonstrated progressive metabolic disease (PMD) in 20 patients, stable metabolic disease (SMD) in 11 patients, partial metabolic response (PMR) in 8 patients, and complete metabolic response (CMR) in 2 patients. According to the PERCIMT, interim PET/CT demonstrated PMD in 9 patients, SMD in 24 patients, PMR in 6 patients and CMR in 2 patients. On the basis of the interim PET, the patients were divided in a similar manner to the division according to clinical response into those showing metabolic benefit (MB) including SMD, PMR and CMR, and those showing no metabolic benefit (no-MB) including PMD. According to this dichotomization, the EORTC criteria showed a sensitivity (correctly predicting CB) of 64.5%, a specificity (correctly predicting no-CB) of 90.0%, a positive predictive value (PPV) of 95.2%, a negative predictive value (NPV) of 45.0% and an accuracy of 70.7% in predicting best clinical response. The PERCIMT showed a sensitivity of 93.6%, a specificity of 70.0%, a PPV of 90.6%, a NPV of 77.8% and an accuracy of 87.8%. The McNemar test showed that the PERCIMT had a significantly higher sensitivity than EORTC criteria (p = 0.004), while there was no significant difference in specificity (p = 0.5). The agreement between the two sets of criteria was poor (McNemar test p = 0.001, and accordingly kappa = 0.46). The application of the recently proposed PERCIMT to interim 18F-FDG PET/CT provides a more sensitive predictor of final clinical response to immunotherapy than the application of the EORTC criteria in patients with metastatic melanoma.