O-016 - Exploratory analysis of tumor growth rate in patients with advanced gastrointestinal stromal tumors (GIST) treated with regorafenib in the GRID phas...

Date 04 July 2015
Event WorldGI 2015
Session Oral and LBA abstracts
Topics Anti-Cancer Agents & Biologic Therapy
Presenter C. Kappeler
Citation Annals of Oncology (2015) 26 (suppl_4): 108-116. 10.1093/annonc/mdv235
Authors C. Kappeler1, J. Blay2, A. Wagner1, P. Reichardt3, Y.-. Kang4, H. Joensuu5, K. Schaefer1, J. Chung6, P.G. Casali7, G. Demetri8
  • 1Bayer Pharma AG, Berlin/DE
  • 2Centre Léon Bérard, Lyon/FR
  • 3HELIOS Klinikum Berlin-Buch, Berlin/DE
  • 4Asan Medical Center, Seoul/KR
  • 5Helsinki University Hospital and University of Helsinki, Helsinki/FI
  • 6Bayer Healthcare Pharmaceuticals, Whippany/US
  • 7Adult Mesenchymal Tumour Medical Oncology Unit, Milan/IT
  • 8Boston/US



In the randomized double-blind (DB) phase 3 GRID trial, regorafenib significantly improved the primary endpoint of progression-free survival (PFS) vs placebo in patients with metastatic and/or unresectable GIST who progressed on imatinib and sunitinib (HR 0.27; one-sided p < 0.0001). After progression on DB placebo, randomized patients were allowed to crossover to open-label (OL) regorafenib, while patients initially randomized to regorafenib were allowed to continue as OL regorafenib. This exploratory analysis evaluated the growth rate of target lesions during DB and OL treatment.


Target lesions were assessed at baseline, q4 weeks for the first 3 months, q6 weeks for the next 3 months, and then q8 weeks until the end of treatment. Changes in target lesion diameters over time were approximated by a parabola-like 3-parametric model separately for the DB and OL periods. Tumor growth rate (TGR), defined as the percent change from baseline per month in the sum of target lesion diameters, are given by the slope at the earliest and latest time points of the model curve.


Of the 133 patients randomized to DB regorafenib, 41 continued OL regorafenib; 129 and 37 were evaluable for TGR analysis, respectively. Of 66 patients randomized to DB placebo, 56 (85%) crossed over to OL regorafenib and 65 and 53 were evaluable, respectively. During the DB period, TGR for regorafenib patients was typically close to zero early reflecting target lesion stabilization and then positive from nadir to progression reflecting tumor growth (Table). TGR remained low (5.2 and 4.6) in the regorafenib patients who continued OL regorafenib after progression. In contrast, patients treated with placebo during the DB period had early and late tumor growth. In the subgroup of placebo patients who crossed over to OL regorafenib at progression, TGR decreased from 13.6 with placebo to below zero (-2.6) after starting OL regorafenib, which was similar to early TGR in patients treated with regorafenib during the DB period.


Regorafenib reversed or stabilized tumor growth during the DB period with a similar benefit in DB-placebo patients who switched to regorafenib at progression. TGR for regorafenib-treated patients before and after progression was less than that of patients treated with placebo, suggesting that despite evidence of tumor growth or prior RECIST progression, regorafenib continued to slow tumor growth relative to that observed in the absence of regorafenib. The clinical relevance of these findings for treatment decisions needs to be further investigated. TGR may be an additional efficacy parameter to consider when monitoring treatment on regorafenib and should be explored with other tyrosine kinase inhibitors.

Table: O-016