Abstract
77
Objectives: Peptide receptor radionuclide therapy (PRRT) is an effective treatment for metastatic somatostatin expressing NETs. PRRT response assessment is challenging due to heterogeneity of the disease, limitations of morphological response criteria in slowly growing tumors, necrotic changes after therapy, lack of standardized parameters and timing of functional imaging, and suboptimal sensitivity and specificity of clinical biomarkers. This study evaluates the prognostic value of PET parameters associated with response and objective assessment to PRRT.
Methods: In this retrospective study, we reviewed 30 patients (mean age at diagnosis 58 ± 10.3 years, range 33-74 years) with metastatic NETs who received a mean cumulative dose of 25.98 GBq (range 7.42-29.62 GBq, over 1-4 cycles) 177Lu-DOTATATE between April 2017 through October 2020. Ga-68 DOTATATE parameters SUVmax, SUVmax-avg, (SUVmax average of multiple lesions), and whole-body somatostatin receptor-expressing tumor volume (SRETVwb) were determined at baseline and follow-up. Baseline parameters and any changes after PRRT were correlated with PFS and OS. Response was defined by RECIST 1.1 during a median follow-up time of 12 months (range 1-32) until disease progression. Logistic regression models determined the relationship of PET parameters and best overall response to PRRT by RECIST 1.1; optimal cut-off of baseline PET parameters predictive of response was determined by the area under the receiver operator curve (AUC-ROC). Cox regression models were performed to identify PET parameters associated with PFS or OS.
Results: 4/30 (13.3%) patients showed a partial response, while stable disease was observed in 16/30 (53.3%) patients. 10/30(33.3%) patients had progression of disease. Baseline SUVmax-avg and changes in SRETVwb predicted the best response to PRRT. The optimal cut-off for response was SUVmax-avg 18.402 with 10 SUVmax-avg increments being associated with a relative OR of 2.53 (95% CI: 0.96-6.64; p = 0.06; AUC 0.74). The optimal cut-off for changes in SRETVwb for response was 15.49 mL with each 100 mL decrease being associated with a relative OR of 0.45 (95% CI: 0.26-0.91; p = 0.02; AUC 0.84). In terms of PFS, an increment in 100 mL of SRETVwb was associated with lower PFS, HR 1.52 (95% CI: 1.21-1.9; p <0.001). Increases of 100 mL of SRETVwb were associated with lower OS, HR 1.71 (95% CI: 1.25-2.34; p <0.001).
Conclusions: Functional images provide predictive biomarkers in patients with NETs that can improve patients' selection for PRRT, assess response, and may have a prognostic value of PFS in metastatic NET patients. Additional PET parameters are also being reviewed and validation studies are needed.
