5896 - Outcomes by baseline alpha-fetoprotein (AFP) levels in the phase 3 CELESTIAL trial of cabozantinib (C) versus placebo (P) in previously treated advanced hepatocellular carcinoma (HCC)

Background

High baseline AFP levels are associated with poor prognosis in patients (pts) with HCC and are associated with a distinct molecular profile. C, an inhibitor of MET, VEGFR, and AXL, improved overall survival (OS) and progression-free survival (PFS) in the phase 3 CELESTIAL trial (NCT01908426). Median OS was 10.2 mo with C vs 8.0 mo with P (HR 0.76, 95% CI 0.63–0.92; p = 0.0049), and median PFS was 5.2 mo with C vs 1.9 mo with P (HR 0.44, 95% CI 0.36–0.52; p < 0.0001).

Methods

707 pts were randomized 2:1 to receive C (60 mg qd) or P. Eligible pts had pathologic diagnosis of HCC, Child-Pugh score A, and ECOG PS ≤ 1. Patients must have received prior sorafenib and 1–2 prior lines of systemic therapy for HCC and must have progressed following at least one. Stratification was by disease etiology, geographic region, and extent of disease. Outcomes were analyzed by baseline AFP levels using cutoffs of 20, 200, and 400 ng/mL.

Results

Overall, 69% of pts had baseline AFP ≥20 ng/mL, 49% had AFP ≥200 ng/mL, and 41% had AFP ≥400 ng/mL. For pts with baseline AFP <400 ng/mL, median OS was 13.9 mo with C vs 10.3 mo with P (HR 0.81), and median PFS was 5.5 mo with C vs 1.9 mo with P (HR 0.47). For pts with baseline AFP ≥400 ng/mL, median OS was 8.5 mo with C vs 5.2 mo with P (HR 0.71), and median PFS was 3.9 mo with C vs 1.9 mo with P (HR 0.42). C also improved OS vs P in pts with AFP ≥200 ng/mL or ≥ 20 ng/mL, and C improved PFS irrespective of AFP levels (Table). Pts with high baseline AFP experienced high-grade transaminitis more frequently in both treatment groups; grade 3/4 elevated aspartate aminotransferase with C vs P was 8% vs 4% for AFP <400 ng/mL and 17% vs 11% for AFP ≥400 ng/mL.Table: 702P

Conclusions

C improved OS and PFS vs P in patients with previously treated advanced HCC across a range of baseline AFP levels. High AFP levels were associated with a larger treatment benefit with C for both OS and PFS.

Clinical trial identification

NCT01908426.

Legal entity responsible for the study

Exelixis, Inc.

Funding

Exelixis, Inc.

Editorial Acknowledgement

Disclosure

R.K. Kelley: Consulting or advisory role: Genentech for IDMC (self), Bayer, BMS, Astra Zeneca, DebioPharm, Agios (institution); Research funding: Adaptimmune, Agios, Astra Zeneca, Bayer, Bristol Myers Squibb, Celgene, Exelixis, Eli Lilly, Medimmune, Merck, Novartis, Regeneron, Sanofi, Taiho, Target Pharma Solutions, Tekmira. A.B. El-Khoueiry: Honoraria: BMS, Bayer, Exelixis, EMD Serono, Eisai, Roche, Cytomx, Merck; Consulting or advisory role: BMS, Bayer; Research funding: Astrazeneca, Astex. T. Meyer: Consulting or advisory role: Bayer, Eisai, BMS, BTG, Ipsen; Research funding: Bayer. L. Rimassa: Consulting or advisory role: Lilly, Bayer, Sirtex Medical, Italfarmaco, Sanofi, Arqule, Baxter, Ipsen, Exelixis, Amgen; Travel accomodations expenses: Arqule, Ipsen. S.L. Chan: Consulting or advisory role: AstraZeneca, MSD; Research funding: MSD, Pfizer, Sirtex. A. Tran: Consulting or advisory role: Gilead, Abbvie, MSD; Speakers’ bureau: Gilead, Abbvie, MSD; Travel, accomodations, expenses: Gilead, Abbvie, MSD. V.C. Tam: Honoraria: Celgene, Apobiologix; Travel accomodations expenses: Amgen. D.W. Markby, D.O. Clary: Employee, Stock ownership: Exelixis. A-L. Cheng: Consulting or Advisory role: Bayer, Schering Pharma, Bristol-Myers Squibb, Eisai, Merck Serono, Novartis, Ono Pharmaceutical, ONXEO; Speakers\' bureau: Novartis; Research funding: Sanofi. G.K. Abou-Alfa; Self consulting: Agios, Amgen, Antengene, Aptus, Aslan, Astellas, Astra Zeneca, Bayer, BMS, Boston Scientifc, Carsgen, Celgene, Casi, Daiichi, Debio, Delcath, Eisai, Exelixis, Halozyme, Inovio, Ipsen, Merck, Onxeo, PCI Biotech, Roche, Sanofi, Servier, Sillajen, Sirtex, Yakult Immediate; Family member consulting: Celgene, CytomX, Gilead, Halozyme, Sanofi, Silenseed; Institutional research: Agios, Array, Astra Zeneca, Bayer, BMS, Casi, Celgene, Exelixis, Genentech, Incyte, Lilly, Mabvax, MedImmune, Momenta, Novartis, OncoMed Pharmaceuticals, Roche. All other authors have declared no conflicts of interest.