Cabozantinib prolonged survival of patients with previously treated advanced hepatocellular carcinoma (HCC), even among patients with high serum alpha-fetoprotein (AFP levels), according to an exploratory analysis published in Clinical Cancer Research.1

High baseline AFP has been associated with poor prognosis with HCC and systemic treatment is associated with a decrease in AFP from baseline and improved survival. AFP has not been characterized in the setting of cabozantinib treatment, which is approved by the US Food and Drug Administration for advanced HCC after progression with sorafenib. The aim of this exploratory analysis is to evaluate the role of serum AFP at baseline and changes in response to cabozantinib treatment.

The phase 3 CELESTIAL trial randomly assigned 707 patients with previously treated advanced HCC to receive cabozantinib or placebo. Cabozantinib was demonstrated to prolong overall survival (OS) and progression-free survival (PFS) compared with placebo.

Continue Reading

This analysis stratified patients from the CELESTIAL trial according to baseline AFP levels, using 400 ng/mL as a cutoff. Changes in serum AFP were assessed at 8 weeks, which coincided with the first tumor assessment in the trial. AFP response was defined as a ≥20% decrease from baseline.

Baseline AFP levels were similar between the cabozantinib and placebo arms, with medians of 154.7 ng/mL (interquartile range [IQR], 14.0-2998.9) and 202.5 ng/mL (IQR, 10.2-5174.9), respectively.

Cabozantinib treatment resulted in prolonged OS compared with placebo in both baseline AFP groups, but was significant only among patients with a higher baseline AFP level. Among patients with lower baseline AFP levels, the median OS was 13.9 months with cabozantinib compared with 10.3 months with placebo (hazard ratio [HR], 0.81; 95% CI, 0.62-1.04). In the group with higher AFP levels, cabozantinib resulted in a median OS of 8.5 months compared with 5.2 months with placebo (HR, 0.71; 95% CI, 0.54-0.94).

PFS was also longer with cabozantinib in both baseline AFP groups. In the group with low baseline AFP levels, the median PFS with cabozantinib was 5.5 months compared with 1.9 months with placebo (HR, 0.47; 95% CI, 0.37-0.60). Among patients with higher baseline AFP levels, the median PFS was 3.9 and 1.9 months with cabozantinib or placebo, respectively (HR, 0.42; 95% CI, 0.32-0.55).

AFP response occurred more frequently with cabozantinib, with a rate of 50% compared with 13% in the placebo group. AFP response was associated with longer OS and PFS, regardless of treatment. Among the cabozantinib-treated patients, the median OS with an AFP response was 16.1 months compared with 9.1 months for patients who did not experience an AFP response (HR, 0.61; 95% CI, 0.41-0.74). Among patients who received placebo, the median OS with an AFP response was 11.3 months compared with 7.2 months with no response (HR, 0.79; 95% CI, 0.41-1.55).

In the cabozantinib arm, the median PFS was 7.3 months with an AFP response compared with 4.0 months with no AFP response (HR, 0.55; 95% CI, 0.41-0.74). In the placebo arm, the median PFS was 3.8 or 1.9 months with or without an AFP response, respectively (HR, 0.51; 95% CI, 0.27-0.96).

“Our analysis shows improved outcomes with cabozantinib relative to placebo in patients with previously treated HCC across a range of baseline AFP levels,” the authors wrote. They noted that given the number of emerging therapies for HCC, “further investigation of AFP kinetics in patients treated with newly available therapies is warranted.”


Kelley RK, Meyer T, Rimassa L, et al. Serum alpha-fetoprotein levels and clinical outcomes in the phase III CELESTIAL study of cabozantinib versus placebo in patients with advanced hepatocellular carcinoma. Clin Cancer Res. Published July 29, 2020. doi: 10.1158/1078-0432.CCR-19-3884