Abstract Background: Everolimus (EVE) plus exemestane (EXE) doubled progression-free survival (PFS) while maintaining quality of life versus EXE alone in postmenopausal women with hormone receptor positive (HR+), HER2-negative metastatic breast cancer (mBC) (BOLERO-2 phase 3; NCT00863655). Pretreatment serum activin A was previously reported as a prognostic factor in first-line hormone therapy (letrozole vs tamoxifen) (Novartis P025) and anti-HER2 mBC (lapatinib vs trastzmab) (CCTG MA.31) trials. Here we investigate the prognostic and predictive ability of activin A in BOLERO-2. Methods: Activin A levels were determined on pretreatment serum samples using ELISA. Cox-proportional hazards model was used to assess the efficacy of EVE in the activin A low and high subgroups (median cut-point), and the prognostic effect of activin A on PFS and overall survival (OS). Results: Baseline activin A levels were determined in 513 patients (71% of 725 BOLERO-2 patients randomized 2:1 to EVE+EXE or EXE). Predictive and prognostic signals are shown in the table below Predictive and prognostic signalsPredictive/PrognosticEnd-pointAct-ivin ATreatmentNEventsMedian PFSHR (95% CI); p valuepredictivePFSHEXE93832.5 (1.5-2.8)-predictivePFSHEVE+EXE1631325.4 (4.1-6.8)0.46 (0.34 - 0.60); <0.0001predictivePFSLEXE89774.2 (2.0 -5.4)-predictivePFSLEVE+EXE1681059.9 (8.1-12.5)0.38 (0.28 - 0.51); <0.0001predictiveOSHEXE936820.1 (13.8-22.6)-predictiveOSHEVE+EXE16312917.7 (15.7-22.3)1.04 (0.78 - 1.40); 0.78predictiveOSLEXE8939NA (34.7-NA)-predictiveOSLEVE+EXE1687241.4 (36.4-NA)1.02 (0.69 - 1.50); 0.93prognosticPFSH 2562154.1 (2.9-4.2)-prognosticPFSL 2571826.9 (6.7-8.5)0.54 (0.45 - 0.66); <0.0001prognosticOSH 25619718.0 (16.5-21.1)-prognosticOSL 25711142.3 (38.5-NA)0.34 (0.27 - 0.42); <0.0001 . In multivariate analysis (including sensitivity to prior hormone therapy and visceral disease), activin A remained a significant independent prognostic factor for PFS and OS [HR 0.57 (0.46-0.69) and 0.34 (0.27-0.43), respectively]. Conclusions: Higher serum activin A was strongly associated with shorter PFS and OS in HR+/HER2- mBC patients. Everolimus was efficacious regardless of serum activin A level. These results are similar to our previous studies in phase 3 trials of letrozole-tamoxifen (Novartis P025), and HER2-targeted therapy, lapatinib vs trastuzmab (CCTG MA.31): pretreatment serum activin A was prognostic for outcome, but was not a predictive factor for treatment arm selection. Citation Format: Ali SM, Chen D, Ali A, Krecko L, Leitzel K, Vasekar M, Nagabhairu V, Marks E, Polimera H, Richardson A, May M, He W, Patel P, Lavin M, Hofsess S, Sweetman R, Hortobagyi G, Baselga J, Lipton A. Serum activin A and outcomes in HR+ /HER2- metastatic breast cancer patients treated with everolimus: Results from BOLERO-2 [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P1-07-09.

Background : The Y537S and D538G mutations in ESR1 are recurrent alterations in metastatic breast cancer (MBC) that promote ligand-independent receptor activation and resistance to estrogen deprivation therapy in laboratory models. The clinical prevalence of these mutations is not well established, but has been reported to be ∼10% by tumor sequencing. Cell free DNA (cfDNA) analysis typically reflects tumor-derived genetic alterations and can be used to characterize a population of patients at a common time point. We hypothesized that cfDNA could be used to detect ESR1 mutation at entry and its clinical impact in a large subset from BOLERO-2 that randomized patients with MBC to Exemestane (EXE) or EXE plus Everolimus (EVE). Methods : Patients with ER+/HER2– MBC with prior exposure to at least 1 non-steroidal aromatase inhibitor (NSAI) were enrolled in BOLERO-2. cfDNA was extracted from 560 baseline plasma samples using QIAamp Circulating DNA kit or QIAsymphony DSP Virus/Pathogen kit. Samples were analyzed by droplet digital PCR for Y537S and D538G. Cox-proportional hazards model was used to assess progression free survival (PFS) in patient subgroups defined by each ESR1 mutation, and the prognostic effect of each ESR1 mutation on overall survival (OS). Results : Of 541 evaluable patients (74.7% of study population), 156 (28.8%) had mutation in ESR1 in D538G (21.1%) and/or Y537S (13.3%) with 30 samples having both mutations. Sequencing of 302 archival tumor specimens (244 primary and 57 metastases) from this study only yielded 4 instances of D538G (1.3%) and 1 Y537S (0.3%). In the overall population, both mutations were poor prognostic factors associated with shorter OS (Table 1). PFS results were different for the two mutations. D538G but not Y537S mutation was associated with a shorter PFS with EXE compared to wild type (WT), (hazard ratios, D538G: 1.44 [95%CIs, 1.04-1.99] and Y537S: 0.92 [95%CIs, 0.44-1.93]). The D538G mutant group derived a similar benefit as WT from the addition of EVE to EXE, whereas the Y537S group did not (Table 2). Conclusions : cfDNA analysis identifies a high rate of the Y537S and D538G ESR1 mutations in ER+ NSAI-treated MBC. As the two mutations may only represent 50-60% of all activating ESR1 mutations, the clinical prevalence of ESR1 mutations in ER+ MBC may be much higher than previously reported. Both mutations appear to be associated with a more aggressive disease biology. Interestingly, these two activating mutations appear to have differential effects on EXE and EVE sensitivity, highlighting new areas for research in ER biology. Citation Format: Chandarlapaty S, Sung P, Chen D, He W, Samoila A, You D, Bhatt T, Patel P, Voi M, Gnant M, Hortobagyi G, Baselga J, Moynahan ME. cfDNA analysis from BOLERO-2 plasma samples identifies a high rate of ESR1 mutations: Exploratory analysis for prognostic and predictive correlation of mutations reveals different efficacy outcomes of endocrine therapy–based regimens. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr S2-07.

Abstract Background: The Y537S and D538G mutations in ESR1 are recurrent alterations in metastatic breast cancer (MBC) that promote ligand-independent receptor activation and resistance to estrogen deprivation therapy in laboratory models. The clinical prevalence of these mutations is not well established, but has been reported to be ∼10% by tumor sequencing. Cell free DNA (cfDNA) analysis typically reflects tumor-derived genetic alterations and can be used to characterize a population of patients at a common time point. We hypothesized that cfDNA could be used to detect ESR1 mutation at entry and its clinical impact in a large subset from BOLERO-2 that randomized patients with MBC to Exemestane (EXE) or EXE plus Everolimus (EVE). Methods: Patients with ER+/HER2– MBC with prior exposure to at least 1 non-steroidal aromatase inhibitor (NSAI) were enrolled in BOLERO-2. cfDNA was extracted from 560 baseline plasma samples using QIAamp Circulating DNA kit or QIAsymphony DSP Virus/Pathogen kit. Samples were analyzed by droplet digital PCR for Y537S and D538G. Cox-proportional hazards model was used to assess progression free survival (PFS) in patient subgroups defined by each ESR1 mutation, and the prognostic effect of each ESR1 mutation on overall survival (OS). Results: Of 541 evaluable patients (74.7% of study population), 156 (28.8%) had mutation in ESR1 in D538G (21.1%) and/or Y537S (13.3%) with 30 samples having both mutations. Sequencing of 302 archival tumor specimens (244 primary and 57 metastases) from this study only yielded 4 instances of D538G (1.3%) and 1 Y537S (0.3%). In the overall population, both mutations were poor prognostic factors associated with shorter OS (Table 1). Table 1AlterationNEventsMedian OS (95%CI)HR (95%CI)WT38521732.1 (28.1-36.4) D538G835726.0 (19.2-32.4)1.25 (1.02-1.54)Y537S423020.0 (13.0-29.3)2.31 (1.34-3.97)Double mt302415.2 (10.9-27.4)1.77 (1.31-2.39) PFS results were different for the two mutations. D538G but not Y537S mutation was associated with a shorter PFS with EXE compared to wild type (WT), (hazard ratios, D538G: 1.44 [95%CIs, 1.04-1.99] and Y537S: 0.92 [95%CIs, 0.44-1.93]). The D538G mutant group derived a similar benefit as WT from the addition of EVE to EXE, whereas the Y537S group did not (Table 2). Table 2AlterationGroupNEventsMedian PFS (95%CI)HR (95%CI)WTEXE1281163.9 (2.8-4.2)0.4 (0.31-0.51) EXE/EVE2571728.5 (6.9-9.9) D538GEXE24222.7 (1.4-2.8)0.34 (0.2-0.57) EXE/EVE59455.8 (4.2-8.4) Y537SEXE21164.1 (1.4-6.7)0.98 (0.49-1.94) EXE/EVE21194.2 (1.4-5.4) Conclusions: cfDNA analysis identifies a high rate of the Y537S and D538G ESR1 mutations in ER+ NSAI-treated MBC. As the two mutations may only represent 50-60% of all activating ESR1 mutations, the clinical prevalence of ESR1 mutations in ER+ MBC may be much higher than previously reported. Both mutations appear to be associated with a more aggressive disease biology. Interestingly, these two activating mutations appear to have differential effects on EXE and EVE sensitivity, highlighting new areas for research in ER biology. Citation Format: Chandarlapaty S, Sung P, Chen D, He W, Samoila A, You D, Bhatt T, Patel P, Voi M, Gnant M, Hortobagyi G, Baselga J, Moynahan ME. cfDNA analysis from BOLERO-2 plasma samples identifies a high rate of ESR1 mutations: Exploratory analysis for prognostic and predictive correlation of mutations reveals different efficacy outcomes of endocrine therapy–based regimens. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr S2-07.

Guy Jerusalem, Fabrice Andre, David Chen, Douglas Robinson, Mustafa Ozguroglu, Istvan Lang, Michelle White, Masakazu Toi, Tetiana Taran, Luca Gianni CHU Sart Tilman Liege and University of Liege, Liege, Belgium; Institut Gustav Roussy, Villejuif, France; Novartis Pharmaceuticals, East Hanover, New Jersey, USA; Cerrahpasa Medical Faculty, Istanbul University, Istanbul, Turkey; Orszagos Onkologiai Intezet, Budapest, Hungary; Monash Medical Center Moorabbin, Victoria, Australia; Kyoto University, Kyoto, Japan; Ospedale San Raffaele, Milan, Italy

ABSTRACT Background In RECORD-1, everolimus significantly increased median PFS (primary end point) over placebo in VEGFr-TKI-refractory patients with mRCC. We investigated angiogenesis pathway molecules, sVEGFR-2, VEGF-A, and bFGF, in plasma as potential biomarkers of everolimus efficacy in RECORD-1. Material and methods Patients received everolimus 10 mg daily (n = 277) or placebo (n = 139), both with BSC; placebo patients could cross over to everolimus at disease progression. Pre-dose blood samples were collected on day 1 of the first four 28-day treatment cycles; plasma levels of sVEGFR-2, VEGF-A, and bFGF were assessed using ELISA. A mixed effects model was used to assess treatment effect over time on each biomarker. Hazard ratios (HR) for prognostic effects were obtained using log baseline biomarker values as continuous variables in a stratified Cox proportional hazards model. Results Plasma values for sVEGFR-2, VEGF-A, and bFGF were available for 45/45/39% of everolimus patients and 50/50/45% of placebo patients. Baseline characteristics of patients with biomarker data were similar to the overall population. Mean log baseline values for sVEGFR-2, VEGF-A, and bFGF were similar for both arms; 9.1/5.1/1.6 for everolimus and 9.1/5.2/1.8 for placebo, respectively. Median PFS was significantly improved with everolimus vs placebo, regardless of baseline levels of any of the biomarkers analyzed (P Conclusions Everolimus provided significant clinical benefit over placebo, regardless of baseline biomarker levels. However, lower VEGF-A level was seen as a potential prognostic factor for longer PFS. Plasma levels of bFGF and sVEGFR-2 were significantly down-regulated from baseline by everolimus treatment. Disclosure S. Oudard: Stephane Oudard received honoraria from Bayer, Novartis, Pfizer, Roche, and Sanofi-Aventis. B. Escudier: Bernard Escudier has received honorarium from Novartis, Pfizer, GSK, Aveo, and Bayer. J. Thompson: John Thompson received clinical study support from Novartis. V. Grunwald: Viktor Grunwald served as consultant to Roche, Bayer, Novartis, Pfizer, GlaxoSmithKline, and Aveo/Astellas, received honoraria from GlaxoSmithKline, Novartis, and Pfizer, and received research funding from Pfizer and GlaxoSmithKline. S. Bracarda: Sergio Bracarda has served as advisor for Pfizer, Bayer-Schering, GlaxoSmithKline, Novartis, Aveo/Astellas, Boheringer-Ingelheim, Johnson & Johnson, and Sanofi-Aventis and has received speaker fees from Pfizer, Novartis, and Sanofi-Aventis. A. Panneerselvam: Ashok Panneerselvam is an employee of Novartis Pharmaceuticals Corporation. S. Gogov: Sven Gogov is an employee of Novartis Pharma AG. D. Chen: David Chen is an employee of Novartis Pharmaceuticals Corporation. R.J. Motzer: Robert Motzer has received research funding from Novartis. All other authors have declared no conflicts of interest.

ABSTRACT Background RADIANT-3 was a phase III study investigating the effect of the mammalian target of rapamycin inhibitor everolimus on progression-free survival (PFS) in patients with advanced pancreatic neuroendocrine tumors (pNET; Yao et al, NEJM, 2011). Everolimus significantly improved PFS compared with placebo (11 vs 4.6 months, P Methods Baseline plasma levels of VEGF-A, PlGF, sVEGFR1, and sVEGFR2 were determined by ELISA using multiplexed MSD platform. The optimal cutoffs for these markers were explored using the “survival tree analysis” method. Interaction of treatment and baseline marker status ( Results PFS was significantly improved to a similar extent in patients receiving everolimus compared with patients who received placebo, regardless of baseline levels of markers (P Conclusions These exploratory analyses demonstrated consistent everolimus efficacy in all patients with advanced pNET irrespective of their baseline VEGF pathway biomarker levels. However, levels of VEGF-A, PlGP, and sVEGFR1 are potential prognostic factors for pNET. Marker Cutoff (pg/mL) Median PFS Prognostic effect HR [95% CI]; P value Treatment effect P value VEGF-A 246.1 8.3 vs 5.5 1.50 [1.17-1.92]; PlGF 32.06 8.0 vs 4.2 1.52 [1.14-2.02]; .004 SVEGFR1 226.2 8.3 vs 5.5 1.62 [1.27-2.07]; SVEGFR2 24503.1 10.8 vs 5.7 1.30 [0.96-1.76]; .090 Disclosure J.C. Yao: Consultant/advisory role for Novartis, Ipsen, Pfizer, Endo; Honoraria from Novartis; Research funding from Novartis, Genentech. M. Shah: Honoraria for presentations and participation in Advisory Board Meetings: Novartis, Ipsen, Pfizer Research grant: Novartis. A. Panneerselvam: Novartis employee. S. Stergiopoulos: Novartis employee. D. Chen: Novartis employee. M. Pavel: Honoraria for presentations and participation in Advisory Board Meetings: Novartis, Ipsen, Pfizer Research grant: Novartis. All other authors have declared no conflicts of interest.