Poster Presentation Clinical Oncology Society of Australia Annual Scientific Meeting 2019

Outcomes based on plasma biomarkers for the phase 3 CELESTIAL trial of cabozantinib (C) versus placebo (P) in advanced hepatocellular carcinoma (aHCC) (#340)

F Parnis 1 , L Rimassa 2 , R K Kelley 3 , T Meyer 4 , B-Y Ryoo 5 , P Merle 6 , J-W Park 7 , J F Blanc 8 , H-Y Lim 9 , A Tran 10 , A E Borgman-Hagey 11 , D O Clary 11 , E Wang 11 , A-L Cheng 12 , A B El-Khoueiry 13 , G K Abou-Alfa 14
  1. Adelaide Cancer Centre, Adelaide, South Australia, Australia
  2. Humanitas Cancer Center, Humanitas Clinical and Research Center, IRCCS, Rozzano (Milan), Italy
  3. UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA
  4. Royal Free Hospital, London, United Kingdom
  5. Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea
  6. Groupement Hospitalier Lyon Nord, Lyon, France
  7. National Cancer Center, Goyang, South Korea
  8. Hôpital Haut-Lévêque, CHU Bordeaux, France
  9. Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
  10. Hepatology, Groupe Hospitalier L’Archet, Nice, France
  11. Exelixis Inc, Alameda, CA
  12. National Taiwan University College of Medicine, Taipei, Taiwan
  13. USC Norris Comprehensive Cancer Center, Los Angeles, CA
  14. Memorial Sloan Kettering Cancer Center, New York, NY

Background: C inhibits tyrosine kinases implicated in HCC progression and resistance to antiangiogenic therapy, including VEGFR, MET, and AXL. In the phase 3 CELESTIAL trial (NCT01908426), C significantly improved overall survival (OS) and progression-free survival (PFS) vs P in previously treated aHCC (Abou-Alfa, NEJM 2018). Here, we evaluate outcomes based on plasma biomarkers in CELESTIAL.

Methods: 707 patients (pts) were randomized 2:1 to receive C (60 mg qd) or P. Pts were Child-Pugh A and ECOG PS ≤1, must have received prior sorafenib, and could have received up to 2 prior systemic regimens for HCC. Plasma samples were collected at baseline and on treatment for 674 pts and analyzed by Luminex assay (Assay Gate, Ijamsville, MD). OS and PFS were evaluated for low and high baseline biomarker levels dichotomized at the median. Prognostic factors were identified by comparing low vs high biomarker levels within each treatment arm, using the criterion that the HR 95% CI does not include 1.0. Biomarkers were considered predictive if p≤0.05 for the interaction between treatment and subgroup.

Results: For low and high levels of all 13 biomarkers analyzed, hazard ratios favored C over P for both OS and PFS (all C vs P HRs<0.55; not shown). Analyses comparing low vs high biomarker levels identified MET, AXL, HGF, GAS6, VEGF-A, ANG2, IL-8, IGF-1, and EPO as possible prognostic factors for OS in one or both treatment arms. No baseline biomarkers were found to be predictive of an OS benefit with C (p interaction >0.05).

Conclusion: C treatment was associated with improved OS and PFS vs P in previously treated aHCC irrespective of baseline biomarker levels. Low baseline levels of MET, HGF, GAS6, VEGF-A, ANG2, and IL-8 and high levels of IGF-1 were identified as potential prognostic biomarkers for longer OS in the P group.