Relationship between metabolic toxicity and efficacy of everolimus in patients with neuroendocrine tumors: a pooled analysis from the randomized, phase 3 RADIANT-3 and RADIANT-4 trials
Fazio, N.; Carnaghi, C.; Buzzoni, R.; Valle, J.W.; Herbst, F.; Ridolfi, A.; Strosberg, J.; Kulke, M.H.; Pavel, M.E.; Yao, J.C.
Cancer 127(15): 2674-2682
Hyperglycemia and hypercholesterolemia are class effects of mammalian target of rapamycin inhibitors such as everolimus. This post hoc pooled analysis assessed the potential impact of these events on the efficacy of everolimus. Patients with advanced, low- or intermediate-grade pancreatic, gastrointestinal, or lung neuroendocrine tumors received either oral everolimus at 10 mg/d or a placebo in the RAD001 in Advanced Neuroendocrine Tumors 3 (RADIANT-3) and RAD001 in Advanced Neuroendocrine Tumors 4 (RADIANT-4) trials. A landmark progression-free survival (PFS) analysis by central review was performed for patients treated for at least 16 weeks (n = 308) and according to the occurrence of any-grade adverse events (AEs) within this treatment period. The overall PFS with everolimus from the pooled analysis was 11.4 months (95% confidence interval, 11.01-13.93 months), which was consistent with the findings of RADIANT-3 and RADIANT-4. Overall, 19.1% and 9.8% of patients in RADIANT-3 and 11.9% and 6.4% of patients in RADIANT-4 developed any-grade hyperglycemia and hypercholesterolemia, respectively (regardless of the study drug). The duration of everolimus exposure was longer in patients who developed these AEs versus patients without these AEs. Overall, 308 patients were exposed to treatment for at least 16 weeks (hyperglycemia, 39 of 269 patients; hypercholesterolemia, 20 of 288 patients). No association was observed between the development of these AEs and PFS (18.8 and 14.1 months with and without hyperglycemia, respectively, and 14.1 and 14.8 months with and without hypercholesterolemia, respectively). Although limitations apply because of the small number of AEs observed, there was no significant impact of these AEs on PFS; this suggests similar efficacy in the presence or absence of these events.