PET- and CT-Based Imaging Criteria for Response Assessment of Gastroenteropancreatic Neuroendocrine Tumors Under Radiopharmaceutical Therapy.

Despite well-documented limitations, current guidelines recommend the use of size-based RECIST 1.1 for response assessment of gastroenteropancreatic neuroendocrine tumors (GEP-NETs) under radiopharmaceutical therapy (RPT). We hypothesize that functional criteria are superior to RECIST 1.1 for response assessment and progression-free survival (PFS) prediction, and molecular scores can be used in prognosticating PFS. This single-center, retrospective study included 178 patients with GEP-NETs (G1 and G2) who were treated with at least 2 consecutive cycles of RPT with [Lu]Lu-DOTATATE and who underwent somatostatin receptor PET/CT at baseline and after 2 cycles of RPT (follow-up). PFS was defined as the time between baseline and clinical progression, as reported by a GEP-NET multidisciplinary tumor board (MDT) assessment or reported death. The differences in categorization and PFS between RECIST 1.1, Choi (functional criteria), and the MDT were evaluated, and 3-y PFS with MDT defined PFS as the reference. The predictive values of the different scores in somatostatin receptor standardized reporting and data system and Krenning (molecular scores) for PFS were analyzed. Choi criteria classified a higher number of patients as having progressive disease and partial response and a lower number of patients as having stable disease compared with RECIST 1.1 ( < 0.01). The PFS of patients with progressive disease according to RECIST 1.1 and Choi criteria was shorter than that of patients with stable disease and partial response ( < 0.05). Choi criteria showed a nonsignificantly higher concordance with the MDT than with RECIST 1.1. There was a shift in category from a Krenning score of 4 to a score of 3 between baseline and follow-up ( < 0.01). At baseline, a Krenning score of 3 was associated with a shorter median PFS compared with a score of 4 ( < 0.05). In addition to RECIST 1.1, further PET- and CT-based imaging criteria have the potential to assess response and predict PFS in patients with GEP-NETs undergoing RPT. Our data support the assumption to use Choi criteria for prediction of PFS and agreement in response assessment. At baseline, the Krenning score can be used to predict therapy response after 2 cycles of RPT.