169P - Serum CA19-9 response is an early predictive marker for the efficacy of regorafenib in refractory metastatic colorectal cancer

Date 19 December 2015
Event ESMO Asia 2015 Congress
Session Poster presentation 1
Topics Cytotoxic agents
Colon and Rectal Cancer
Biological therapy
Presenter Azusa Komori
Citation Annals of Oncology (2015) 26 (suppl_9): 42-70. 10.1093/annonc/mdv523
Authors A. Komori1, H. Taniguchi1, Y. Kito2, S. Hamauchi2, T. Masuishi3, H. Hasegawa1, S. Mitani1, Y. Narita1, S. Kadowaki1, T. Ura1, M. Ando1, K. Mori4, H. Yasui2, K. Muro1, K. Yamazaki5
  • 1Department Of Clinical Oncology, Aichi Cancer Center Hospital, 464-8681 - Nagoya/JP
  • 2Division Of Gastrointestinal Oncology, Shizuoka Cancer Center, 411-8777 - Shizuoka/JP
  • 3Department Of Clinical Oncology, Aichi Cancer Center Hospital, Nagoya/JP
  • 4Clinical Trial Coordination Office, Shizuoka Cancer Center, Shizuoka/JP
  • 5Division Of Gastrointestinal Oncology, Shizuoka Cancer Center, Shizuoka/JP



The oral multikinase inhibitor regorafenib improves overall survival in patients with chemo-refractory metastatic colorectal cancer (mCRC). However regorafenib induces various adverse events (AEs) which often impair patients' quality of life. Therefore the identification of early predictive markers for the efficacy of regorafenib is warranted.


We retrospectively examined 146 consecutive mCRC patients who received regorafenib in two institutions from May 2013 to March 2015. Clinical parameters including patients' backgrounds, AEs and changes in the value of biochemical parameters (LDH, CEA, and CA 19-9) from the initiation of regorafenib until day 28 were evaluated to identify predictors for progression-free survival (PFS).


Patient characteristics: median age (range), 64 (37-87) years; ECOG performance status 0/1/2, 28/63/9%; KRAS status wild/mutant/unknown, 55/44/1%; prior chemotherapy fluoroprymidine/oxaliplatin/irinotecan/bevacizumab/anti-EGFR antibodies (only for KRAS wild type)/TAS-102, 100/100/99/95/94/6%. Response rate and disease control rate were 0.7%, 31.9%, respectively. Median PFS was 2.1 months, and median overall survival was 6.6 months with the median follow-up time of 5.4 months. Major AEs in all courses (all garde/ ≥ grade3) were hand foot skin reaction 71/21%, hypertension 47/9% and proteinuria 50/6%. The results of univariate and multivariate analyses of PFS are shown in the table. The patients whose serum CA19-9 decreased after regorafenib had a significantly better PFS (median 15.7 vs. 8.7 weeks, p= 0.004) compared to patients whose CA19-9 level did not decrease. The early CA19-9 decrease remained an independent predictive factor (HR 0.408, 95%CI: 0.199-0.836, p = 0.014) from the results of multivariate analysis.

Univariate analysis Multivariate analysis
HR (95% CI) p-value HR (95% CI) p-value
PS(≥2) 1.154 (0.506-2.634) 0.733 - -
Age (≥70years) 1.199 (0.776-1.852) 0.413 - -
KRAS (mutant) 0.901 (0.623-1.303) 0.579 - -
Primary site(+) 1.761 (1.093-2.835) 0.020 1.141 (0.657-1.981) 0.639
No of metastatic sites(≥2) 1.742 (1.069-2.839) 0.026 1.879 (0.910-3.877) 0.088
Time from the first line (≥18months) 0.637 (0.408-0.995) 0.048 0.297 (0.151-0.586) 0.001
Hand foot skin reaction (≥Grade2) 0.743 (0.516-1.073) 0.114 - -
Hypertension (≥Grade2) 0.903 (0.617-1.322) 0.600 - -
CEA decreased (≥10%) 0.538 (0.344-0.841) 0.007 0.800 (0.426-1.503) 0.488
CA19-9 decreased (≥10%) 0.462 (0.269-0.792) 0.005 0.408 (0.199-0.836) 0.014


Early response of CA19-9 may predict the efficacy of regorafenib. Further studies are needed for external validation.

Clinical trial identification


T. Ura, K. Muro, K. Yamazaki: lecture fee: Bayer HealthCare Pharmaceuticals. All other authors have declared no conflicts of interest.