1163P - Response evaluation using RECIST and Choi criteria in patients with well-differentiated pancreatic neuroendocrine tumors (PNET) treated with sunitin...

Date 29 September 2012
Event ESMO Congress 2012
Session Poster presentation I
Topics Anticancer agents
Neuroendocrine Tumours
Biological Therapy
Presenter Chantal Dreyer
Authors C. Dreyer1, O. Hentic2, M. Zappa3, P. Hammel2, M. Bouattour4, C. Mateescu5, S. Faivre6, P. Ruszniewski2, E. Raymond7
  • 1Oncologie, Beaujon University Hospital, 92110 - Clichy/FR
  • 2Gastroenterology, beaujon, Clichy/FR
  • 3Radiology, Beaujon, Clichy/FR
  • 4Oncology Department, Beaujon University Hospital, Clichy/FR
  • 5Oncology Department, Beaujon University Hospital, 92110 - Clichy/FR
  • 6Cancérologie (oncology), Beaujon University Hospital, Clichy/FR
  • 7Beaujon University Hospital, 92110 - Clichy/FR



Despite low response rate by RECIST, sunitinib and everolimus improved the progression-free survival of patients (pts) with well-differentiated pNET in two large phase III controlled trials. RECIST being a poor surrogate endpoint for PFS, this study aimed evaluating the value of CHOI criteria in PNET pts.


Data for this analysis were collected from pts with well-differentiated PNET treated consecutively with sunitinib or everolimus in our center between 10-2006 and 11-2010. All pts had tumor progression within 12 months (m) prior to treatment. Pts were considered evaluable if CT-scans were performed within 6 weeks prior treatment and <3 months following treatment initiation. A radiologist blinded for clinical data evaluated responses according to RECIST and CHOI criteria. Best responses were correlated with Kaplan Meier estimates of time-to-progression (TTP) and overall survival (OS).


A total of 25 pts were treated with either sunitinib or everolimus. Among them, three pts were evaluated using MRI and thus were not evaluable in this study. Twelve pts received sunitinib and ten pts were treated with everolimus (Male/female: 8/14, median age: 57, range 38-76). All pts had tumor progression within 12 m prior to sunitinib or everolimus. Pts were either treated in 1st line (n = 6), 2nd line (n = 9), 3rd line (n = 7) or 4th line (n = 1) after cytotoxic chemotherapy. At the first CT-scan evaluation, 2 pts presented a partial response (PR), 18 pts had a stable disease (SD), and 2 pts had a progressive disease (PD) by RECIST. Using CHOI criteria, 11 PR, 8 SD, and 3 PD were observed. Nine of 18 pts (50%) with SD according to RECIST were reallocated to the responders (PR) using CHOI criteria. Median TTP and OS for all pts were 14.0 m and 35.8 m, respectively. According to RECIST, TTP was higher in pts with PR + SD compared to PD (14.2m vs 3.3m, p < 0.0001). According to CHOI criteria, TTP was 26.1m in pts with PR, 8.7m in pts with SD, and 3.5m in pts with PD (p = 0.038).


Response by CHOI criteria is a prognostic factor for TTP and may identify among SD pts by RECIST those benefiting the most of targeted therapy.


O. Hentic: Novartis.

P. Hammel: honoraria from novartis and pfizer.

S. Faivre: honoraria from pfizer and novartis.

P. Ruszniewski: honoraria from novartis and pfizer.

E. Raymond: honoraria from novartis and pfizer.

All other authors have declared no conflicts of interest.