Background: Chromogranin A (CgA), neuron-specific enolase (NSE), gastrin, and glucagon are important biomarkers in pNET. The goals of this analysis were to characterize serum CgA, NSE, gastrin, and glucagon concentration changes from baseline in response to treatment with oral everolimus or placebo in patients with advanced pNET in the phase III RADIANT-3 trial (ESMO 2010, Abstract #LBA9). Methods: Patients with progressive, advanced low- or intermediate-grade pNET were randomized to everolimus 10 mg/d orally (n=207) or placebo (n=203). Serum samples were collected and analyzed for CgA, NSE, gastrin, and glucagon at baseline and, if elevated (>ULN), were repeated on day 1 of each subsequent cycle. Changes from baseline over time were analyzed using a mixed-effects model including least squares estimates and P values of treatment effect. Results: Patients with elevated baseline values had biomarkers assessed for a median number of 10 cycles in the everolimus arm and 4 cycles in the placebo arm. Everolimus vs placebo resulted in greater reductions of CgA (P<0.0001), NSE (P<0.0001), gastrin (P<0.0001), and glucagon (P<0.0001) over time. During the first cycle, the ratio in fold change from baseline for everolimus vs placebo was 0.617/1.081=0.57 (95% CI, 0.45-0.73) for CgA, 0.478/0.951=0.50 (95% CI, 0.32-0.80) for NSE, 0.581/0.833=0.70 (95% CI, 0.52-0.94) for gastrin, and 0.679/1.034=0.66 (95% CI, 0.49-0.89) for glucagon. This positive treatment effect with everolimus was maintained over subsequent treatment cycles; the ratio in fold change from baseline for everolimus vs placebo groups ranged from 0.39-0.57 for CgA, 0.23-0.51 for NSE, 0.46-0.70 for gastrin, and 0.45-0.66 for glucagon. Conclusions: In addition to the improvement in PFS seen in RADIANT-3, these results demonstrate that treatment with everolimus resulted in early, sustained decreases in serum levels of CgA, NSE, gastrin, and glucagon in patients with pNET.