Relevance of baseline carcinoembryonic antigen for first-line treatment against metastatic colorectal cancer with FOLFIRI plus cetuximab or bevacizumab (FIRE-3 trial). (January 2019)

• Record Type: Journal Article
• Authors: Holch, J.W.; Ricard, I.; Stintzing, S.; Fischer von Weikersthal, L.; Decker, T.; Kiani, A.; Vehling-Kaiser, U.; Heintges, T.; Kahl, C.; Kullmann, F.; Scheithauer, W.; Moehler, M.; Jelas, I.; Modest, D.P.; Westphalen, C.B.; von Einem, J.C.; Michl, M.; Heinemann, V.
• Is Part Of: European journal of cancer. Volume 106(2019)
• Page Start: 115
• Page End: 125
• Publication Date: 2019-01
• Subjects: Metastatic colorectal cancer -- Carcinoembryonic antigen -- Bevacizumab -- Cetuximab -- Prognostic biomarker -- Predictive biomarker; Cancer -- Periodicals; Neoplasms -- Periodicals; Cancer -- Périodiques; Cancer; Tumors; Electronic journals; Periodicals; Electronic journals; 616.994
• DOI: 10.1016/j.ejca.2018.10.001
• Languages: English
• ISSNs: 0959-8049

Abstract

Abstract: Purpose: Increased baseline carcinoembryonic antigen (CEA) serum level is associated with inferior overall survival (OS) in metastatic colorectal cancer (mCRC). However, limited data exist on its predictive relevance for targeted therapies. Therefore, we analysed its relevance in FIRE-3, a randomised phase III study. Experimental design: FIRE-3 evaluated first-line FOLFIRI plus cetuximab (FOLFIRI/Cet) versus FOLFIRI plus bevacizumab (FOLFIRI/Bev) in mCRC patients with RAS -WT tumour (i.e. wild-type in KRAS and NRAS exons 2–4). Herein, the impact of CEA on patient outcome was investigated. Results: Of 400 patients, 356 (89.0%) were evaluable for CEA. High CEA (>10 ng/ml; N = 237) compared to low CEA (≤10 ng/ml; N = 119) was associated with shorter OS in the FOLFIRI/Bev arm (hazard ratio [HR] = 1.50; P = 0.036), while no significant OS difference was observed in the FOLFIRI/Cet arm (HR = 1.07; P = 0.74). In patients with high CEA, FOLFIRI/Cet compared to FOLFIRI/Bev showed a greater OS benefit (HR = 0.56; P < 0.001) than in patients with low CEA (HR = 0.78; P = 0.30). Furthermore, FOLFIRI/Cet exhibited significantly superior objective response rate in patients with high CEA (odds ratio = 2.21; P = 0.006) in contrast to patients with low CEA (odds ratio = 0.90; P = 0.85). Conclusion: In patients with RAS -WT mCRC receiving first-line chemotherapy with FOLFIRI/Cet versus FOLFIRI/Bev, elevated CEA was associated with inferior survival in the bevacizumab arm, while this was not the case when cetuximab was applied. Comparison of OS and objective response rate according to treatment arms indicated that cetuximab was greatly superior to bevacizumab in patients with elevated CEA, while this effect was markedly lower and lost statistical significance in patients with low CEA. Highlights: High baseline CEA impacts survival in patients with metastatic colorectal cancer. In FIRE-3, high baseline CEA led to shorter survival in FOLFIRI+bevacizumab arm. Contrary, high baseline CEA did not show shorter survival in FOLFIRI+cetuximab arm. Survival benefit of FOLFIRI+cetuximab over FOLFIRI+bevacizumab depended on CEA. High baseline CEA may predict benefit from FOLFIRI+cetuximab over FOLFIRI+bevacizumab.