Background: The mTORC1 inhibitor everolimus has been approved in combination with the aromatase inhibitor exemestane for the treatment of hormone receptor-positive (HR+) human epidermal growth factor receptor 2-negative (HER2-) metastatic breast cancer (HR+ mBC) progressing on prior therapy with a non-steroidal aromatase inhibitor. To date, no predictive biomarkers of tumor sensitivity/resistance for everolimus-based treatments have been identified. We hypothesized that precocious changes in the Standardized Uptake Volume (∆SUV%), as assessed by 18F-Fluorodeoxyglucosepositron-emission tomography (18F-FDG PET/CT), may be a marker of everolimus efficacy. Methods: This was a retrospective study including 31 HR+ HER2- patients treated with everolimus and exemestane in two Italian centers between 2013 and 2018. The objective of the study was to investigate ∆SUV% as a predictive marker of everolimus antitumor efficacy. 18F-FDG PET/CT scans were performed at baseline and after three months of treatment. Patients were defined as long responders (LRs) if disease progression occurred at least 10 months after treatment initiation and long survivors (LSs) if death occurred later than 36 months after starting therapy. ROC analysis was used to determine the optimal cut-off values of ∆SUV% to distinguish LRs from non-LRs and LSs from non-LSs. Progression-free survival (PFS) and overall survival (OS) were estimated by Kaplan-Meier method. Results: The SUVmax values decreased significantly from baseline to 3 months after therapy (p = 0.003). Dynamic changes of SUVmax (Delta SUV) had a higher accuracy in discriminating long-responders from non-long-responders (AUC = 0.67, Delta SUV cut-off = 28.8%) respects to its ability to identify long survivors from no-long survivors (AUC = 0.60, Delta SUV cut-off = 53.8%). Patients were divided into groups according to the Delta SUV cut-offs and survival outcomes were evaluated: patients with a decrease of ∆SUV% ≥ 28.8% had significantly better PFS (10 months-PFS: 63.2%, 95% CI: 37.9-80.4% and 16.7%, 95% CI: 2.7-41.3% respectively, p = 0.005). As regard as OS, patients with ∆SUV% ≥ 53.8% had longer OS when compared to patients with ∆SUV% < 53.8% (36 month-OS: 82.5% vs. 45.9% vs. p = 0.048). Conclusion: We found two precocious ∆SUV% thresholds capable of identifying HR+ HER2-mBC patients, which would achieve long-term benefit or long-term survival during everolimus-exemestane therapy. These results warrant further validation in prospective studies and should be integrated with molecular biomarkers related to tumor metabolism and mTORC1 signaling.
Keywords: 18F-FDG PET/CT; everolimus; metastatic breast cancer; predictive biomarker; ∆SUV%.