Randomized Phase III Trial of Sorafenib Versus Placebo in Asian Patients with Advanced Hepatocellular Carcinoma

Reviewer: Eric Shinohara MD, MSCI
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 2, 2008

Presenter: Cheng A.
Presenter's Affiliation: National Taiwan University Hospital, Taipei, Taiwan
Type of Session: Scientific


  • Hepatocellular carcinoma (HCC) is the 8th most common cancer in the world, but yet it is the number three cause of cancer deaths worldwide. The areas with the highest incidence of HCC are East and Southeast Asia and Sub-Saharan Africa.
  • 5-year overall survival (OS) for patients with unresectable HCC is approximately 10%; in resectable patients, it varies from 10-73% depending on the stage.
  • Hepatitis B and C infections are known risk factors for HCC (approximately 100-fold increase in risk).  The prevalence of HCC is higher in Asian countries, partly due to a higher prevalence of hepatitis. In contrast, alcohol-related HCCs are more common in the West.
  • It is unclear how this disparity in the cause of HCC affects how patients in the East respond to therapy for HCC compared with patients in the West. 
    • There are several possible reasons for differences in response:
      • The different causal factors for HCC as stated above
      • Differences in surgical technique, use of transplantation, and other local treatment modalities differ between the East and West
      • Potential genetic differences
  • Sorafenib is a small molecular inhibitor of multiple kinases, including Raf, platelet-derived growth factor (PDGF), vascular endothelial growth factor (VEGF), and c-kit. It has anti-angiogenic and pro-apoptotic activities. T
  • he SHARP trial (Llovet J., et al., ASCO Annual Meeting, 2007) is a phase III trial which demonstrated that sorafenib treatment significantly improved overall survival in patients with advanced HCC compared with placebo, with a median overall survival (OS) of 10.7 versus 7.9 months. Sorafenib treatment was generally well tolerated in this study.
  • The present study investigates the safety and efficacy of sorafenib in patients with HCC from the Asia-Pacific region.

Materials and Methods

  • Eligibility and exclusion criteria:
    • Patients with advanced, measurable HCC were eligible for this study
    • Patients had to have an ECOG performance status of 0 to 2
    • Patients had to have a Child-Pugh status A
    • Patients could not have had prior systemic therapy for HCC
    • Patients had to have a life expectance of ≥12 weeks
  • Eligible patients were randomized 2:1 to either treatment with sorafenib (arm A) or with placebo (arm B). 
    • 400 mg of sorafenib was given twice a day
  • The endpoints for this study were
    • overall survival (OS)
    • time to progression (TTP)
    • time to symptomatic progression (evaluated using the FSHI8-TSP scoring system)
    • disease control rate (DCR) (defined as complete or partial response or stable disease for at least 28 days by Response Evaluation Criteria in Solid Tumors (RECIST) criteria)
    • safety 
  • There was no primary endpoint defined.


  • A total of 226 patients were randomized: 150 patients to the sorafenib arm and 76 to the placebo arm. 
  • Baseline characteristics were similar between the two arms. 
    • For the sorafenib arm, mean age was 51±13 years and 85% of patients were male. 
    • For the placebo arm, mean age was 52±12 years and 87% of patients were male.
  • Baseline tumor characteristics were similar between the two treatment arms. 
  • Results for the endpoints of this study were as follows:
Median (months) Sor vs Plac
Events: Sorafenib
Events: Placebo
Hazard Ratio (95% CI)
6.5 vs 4.2
102 (68%)
62 (82%)
0.68 (0.5-0.93)
2.8 vs 1.4
108 (72%)
58 (76%)
0.57 (0.42-0.79)
3.5 vs 3.4
126 (84%)
65 (86%)
0.89 (0.66-1.20)
2.8 vs 1.4
134 (89%)
73 (96%)
0.62 (0.46-0.83)
  • Response to treatment:

Partial Response
Stable Disease
Disease Progression
Disease Control Rate
35% (95% CI= 28-44%)
16% (95% CI= 8-26%)

  • Grade 3 or greater toxicity was as follows:

Hand foot skin reaction
Number of drug related adverse events
13 (9%)
1 (1%)

  • Overall toxicity was similar between both arms 48% versus 45%
  • The OS advantage was seen across most groups.
  • Patients in the present study were younger than those in the SHARP trial (51 yo versus 67 yo) which is thought to reflect the higher rate of hepatitis related HCC in Asia. It is thought Hepatitis related HCC occurs at a younger age. 81% of patients had hepatitis related HCC in this study versus 46% in the SHARP study.
  • OS and PFS were similar between the present study and the SHARP study (OS HR 0.68 (present study) versus 0.69 (SHARP study); PFS 0.62 (present study) versus 0.69 (SHARP study).

Author's Conclusions

  • Sorafenib significantly increases OS and TTP in Asian patients with HCC compared with placebo.
  • Sorafenib is well tolerated based on toxicity data.
  • There results were similar to those in previous studies (SHARP trial), despite Asian patients having more advanced disease (more lung metastasis, 50% versus 21%), a greater number of tumor sites (35% with 4 or more tumor sites versus 13%), and worse performance statuses compared with patients in the SHARP trial.
  • The results of the present study confirm the results of the SHARP trial, and demonstrate that sorafenib therapy has broad applicability in Asian patients.

Clinical/Scientific Implications

  • The present study confirms the results of the SHARP study, and demonstrates that sorafenib is effective in Asians with HCC. As the use of biological agents becomes broader, it is imperative that we develop molecular markers which can predict their efficacy. Because biological agents are targeted agents, defining the patient population that is likely to benefit is important. These agents do induce toxicity, as seen in the present study, and it is important to use these agents in patients who are likely to respond. There have already been studies which have demonstrated that k-ras mutation status is a predictor of response in patients with colorectal cancer treated with cetuximab. Furthermore, the benefit from biological agents has been modest thus far. By selecting patients who are likely to respond, the chances of finding a benefit to the biological agents is greatly increased, as seen with herceptin studies in the past. With sorafenib, this is a potential problem, as it targets multiple kinases, although the vascular endothelial growth factor and the MAPK pathways seem to be most likely for sorafenib’s beneficial effects in HCC.
  • There is a paucity of effective systemic agents in the treatment of HCC and the majority of treatment, such as surgery, liver transplant, and trans-arterial embolization (TAE) , are local treatments. Studies of doxorubicin and multi-agent chemotherapy have had mixed results and are largely disappointing.   Surafenib is a promising systemic agent which appears to be broadly effective but there are still a number of questions to be answered. The majority of these studies have been in patients with good liver function (Child Pugh A patients in the present study). It is unclear how toxic this therapy will be in patients with worse liver function or who undergo TAE (which can damage the liver). The majority of patients with HCC have poor hepatic function and small studies have suggested that standard dosing of sorafinib may not be appropriate in these patients.