533P - BATON-CRC: A phase 2 randomized trial comparing tivozanib (tivo) + mFOLFOX6 with bevacizumab (bev) + mFOLFOX6 in stage IV metastatic colorectal can...

Date 29 September 2014
Event ESMO 2014
Session Poster Display session
Topics Anti-Cancer Agents & Biologic Therapy
Colon Cancer
Rectal Cancer
Presenter Al Benson Iii.
Citation Annals of Oncology (2014) 25 (suppl_4): iv167-iv209. 10.1093/annonc/mdu333
Authors A. Benson Iii.1, J.A. Bridgewater2, I. Kiss3, F. Eskens4, J. Chen5, C. Sasse5, S. Vossen5, C. van Sant5, H. Ball5, A. Keating5, A. Krivoshik5
  • 1Northwestern University, Robert H. Lurie Comprehensive Cancer Center, 60611 - Chicago/US
  • 2Paul O'gorman Building, UCL Cancer Institute, UK WC1E 6DD - London/GB
  • 3Care Department Of Complex Oncology Care, Masaryk Memorial Cancer Institute, Brno/CZ
  • 4Dept. Of Medical Oncology, Erasmus EC Cancer Institute, NL-3015 CE - Rotterdam/NL
  • 5Global Development, Astellas Pharmaceuticals, Northbrook/US



Tivo is a selective oral VEGF TKI with a long half-life and activity against all 3 VEGFRs. Preclinical data showed antitumor activity of tivo when added to chemotherapy (chemo). In a phase 1b study of tivo + mFOLFOX6 in mCRC and other GI cancers, partial responses were shown and the combination was well tolerated. Based on these results, a randomized, open-label, phase 2 trial of tivo + mFOLFOX6 (arm A) vs bev + mFOLFOX6 (arm B) in previously untreated mCRC was initiated.


Eligible patients had no prior systemic chemo, no fluorouracil-containing adjuvant therapy in the previous 6 mo, and an ECOG PS ≤1. No prior VEGF therapy, including bev was permitted, nor a history of significant thromboembolic or vascular disorders within 6 mo before study entry. The primary end point was PFS by investigator radiologic assessment. Secondary end points included PFS by IRR, OS, ORR, DOR, TTF, and biomarker subgroup analysis of LDH; VEGF A, C, D; CD68; myeloid-derived gene signature and serum soluble cytokines. Subjects were randomized 2:1 and stratified by LDH, origin of cancer, and number of metastatic sites, and received mFOLFOX6 q2w on days 1 and 15 of each 28-day cycle with either tivo 1.5 mg qd for 21 days followed by 7 days off treatment or bev 5 mg/kg q2w.


Between 12/20/11 and 4/28/13, 265 subjects were randomized: 177 to arm A and 88 to arm B. A prespecified interim analysis included 95 PFS events and met prespecified futility criteria. mPFS (arm A vs arm B) was 9.4 mo vs 10.7 mo (P = .706); mORR: 45.2% vs 43.2% (P = .718). The overall safety profile was comparable between arms, but there were more drug-related ≥grade 3 treatment-emergent AEs in arm A (57.1% vs 35.6%), higher ALT or AST elevations >3 × ULN (16.6% vs 5.7%), and increase in grade for platelets (65.9% vs 43.7%). The most common toxicities included diarrhea, nausea, fatigue, neutropenia, and hypertension.


The addition of tivo to mFOLFOX6 vs bev + mFOLFOX6 met prespecified futility criteria and resulted in comparable PFS, ORR, TTF, and DOR with an acceptable safety profile. At the interim analysis, there were no significant associations between serum/tumor biomarkers and outcomes.


J.A. Bridgewater: • Advisory board: 2013, 1; I. Kiss: • Ad bds: Roche, Merck, Bayer • Speakers’ bureau: Roche, Merck, Amgen, Bayer, Sanofi; J. Chen, C. Sasse, S. Vossen, C. van Sant, H. Ball, A. Keating, A. Krisvoshik: Astellas employee • Stock • Salary; All other authors have declared no conflicts of interest.