International Hepatology

Sorafenib use while waiting for liver transplant: We still need to wait

Richard S. Finn

Department of Medicine, Division of Hematology/Oncology, Geffen School of Medicine, University of California-Los Angeles, 10833 Le Conte Ave.,11-934 Factor Bldg., Los Angeles, CA 90095, USA

COMMENTARY ON:Use of sorafenib in patients with hepatocellular carcinomabefore liver transplantation: a cost–benefit analysis whileawaiting data on sorafenib safety. Vitale A, Volk ML, PastorelliD, Lonardi S, Farinati F, Burra P, Angeli P, Cillo U. Hepatology2010 Jan;51(1):165–73.

http://www.ncbi.nlm.nih.gov/pubmed/19877181

� 2011 European Association for the Study of the Liver. Publishedby Elsevier B.V. All rights reserved.

With its approval in 2007 for advanced hepatocellular carci-noma (HCC), sorafenib became the first systemic agent with pro-ven activity against the disease. To date, two large randomizedstudies in advanced stage HCC (BCLC C and a subset of BCLC B)have demonstrated its anti-cancer activity and ability to extendsurvival in patients with Childs A cirrhosis [1,2]. Since then, sev-eral studies have been launched to evaluate sorafenib’s potentialrole in less advanced HCC. Uniquely, sorafenib’s ability toimprove survival is not mediated by inducing significant tumorshrinkage, but by its ability to slow progression and prolongthe time to tumor progression (TTP). While some data has beenpresented [3], several ongoing studies including the SPACE(NCT00855218) and STORM (NCT0692770) studies are evaluatingsorafenib as an adjuvant (after definitive therapy) to transarterialchemoembolization (TACE) and curative resection or radiofre-quency ablation (RFA), respectively. The current study by Vitaleand colleagues uses a cost–benefit analysis to determine thepotential utility of using sorafenib in the neo-adjuvant setting,prior to liver transplant [4]. The clinical need to ask this questionis real, as the authors highlight, drop-out from HCC progressionbeyond Milan criteria is one of the main reasons why patientsawaiting transplant for HCC do not receive an organ.

The Milan criteria was initially described by Mazzaferro andcolleagues over a decade ago and has served as the benchmarkfor prioritizing patients with HCC and otherwise lower MELDscores [5]. However, even with priority, there is still a relativeshortage of livers available and HCC patients are competing withthose with decompensated liver disease for organs. Though thereare regional differences for wait times, the optimal management

Journal of Hepatology 20

Received 17 May 2011; received in revised form 11 August 2011; accepted 11 August2011

E-mail address: Rfinn@mednet.ucla.edu

of patients with tumors within Milan has not been determinedfrom prospective studies and our natural inclination is ‘‘to dosomething’’ when a patient has a known malignancy. Retrospec-tive analyses do suggest that locoregional therapies such as TACEand RFA may be effective ‘‘bridge therapies’’ to transplant [6,7];the basic notion being that existing lesions can be kept withinsize criteria with these approaches. A recent consensus confer-ence on transplant and HCC endorsed this concept, especiallywhen wait times are greater than 6 months [8]. However, thesemodalities fail as they do not prevent the development of metas-tases or the development of new, de novo HCC in a cirrhotic liver.

To determine the potential utility of sorafenib while awaitingtransplant, the authors of this manuscript develop a Markovmodel based on the hazard ratio for sorafenib to decrease the riskof progression in the two studies in advanced HCC [1,2]. Themodel compared the use of sorafenib as a bridging therapy forpatients with well-compensated liver disease and HCC to nobridging therapy unless wait times were greater than 6 months,when loco-regional therapies were taken into account. The studyassessed sorafenib’s neo-adjuvant use on survival as measured byquality-adjusted life days, transplant probability, costs, willing-ness to pay, and net health benefit. Using a HR of 0.47 for inter-mediate stage HCC (BCLC B) derived in a subset analysis of theSHARP study, the authors found that with a monthly drop outprobability of 5% and median time to transplant of 3 months,the gain in liver transplant probability due to sorafenib was 5%and increased with length of waiting time and decreasing hazardratio. These data are shown in Fig. 1 adapted from the manu-script. Additionally, the models demonstrated a median survivalbenefit of 94 quality adjusted life days (QALDs) for sorafenib,and the net health benefit was 37 QALDs. Net health benefitwas sensitive to not only waiting time (>6 months), but alsothe effectiveness of other (locoregional) therapies. This is animportant challenge to the conclusions of the study as in manycenters wait times are a year or longer and in that case, patientsare surely being treated with a locoregional therapy – the benefitof which (e.g. HR), if any, is unknown. Further, while subset anal-yses from randomized studies in advanced disease found that theHR for sorafenib in BCLC B patients was less than for BCLC Cpatients (i.e. those with earlier stage derived a greater benefit),we do not know that this is the case for patients with tumorstaged within Milan criteria where the true HR is unknown.Given the challenge of facing a patient with potentially curableHCC and a clear wait time of >6 months, treating physicians offer

12 vol. 56 j 723–725

Sorafenib HR0.230.470.710.93

Sorafenib HR0.230.470.710.93

Gai

n in

LT

prob

abilit

y (%

)

02468

10

30 90 180 365

12141618

Median time to LT (d)

B

AQ

ALD

s

050

100150200

1 86 431 102 267 334 400

250

350300

Median time to LT (d)

Fig. 1. Adapted from [4].

International Hepatology

patients locoregional therapy while waiting, so why not offerthem sorafenib in addition or even instead?

While the work by Vitale is of interest and highlights an areaof unmet medical need, besides the true efficacy (HR) of sorafenibin this setting, the main issues to prevent its routine use is thelack of available safety data in this population. Drugs with anti-angiogenic activity have been associated with surgical complica-tions such as bleeding and delayed wound healing when used inthe perioperative setting [9]. In the renal carcinoma literature,neoadjuvant sorafenib specifically has not been associated withsurgical complications when stopped at least one day before sur-gery [10,11]. However, there are concerns unique to liver trans-plantation such as vascular anastomoses, liver regeneration,and graft rejection. A recent laboratory study evaluated theeffects of sorafenib on liver regeneration in a mouse model[12]. Mice were treated in 3 groups; group 1 received sorafenibfor 14 days until the day prior to hepatectomy, group 2 receivedsorafenib as group one and continued after surgery, and group 3received it only after surgery. The study demonstrated that whensorafenib was stopped one day before surgery there were noeffects on liver regeneration, however, there was a decrease inliver regeneration in the other two groups. While these dataand the urology literatures are reassuring, at the current timethere is a lack of safety data to recommend the use of sorafenibin the pre-transplant setting. Importantly, unlike in the case ofelective surgery where the last dose of sorafenib can be con-trolled, the nature of liver transplant makes this timing difficultand convincing safety data needs to be generated with sorafenibin the pre-transplant setting, specifically in regards to whensorafenib should be stopped. Sorafenib is an oral agent generallygiven twice daily. In a Phase I study of patients with preservedliver function, the half-life ranged from 24 to 38 h [13] and doesnot appear to be altered in cases of liver dysfunction [14]. Con-ceivably, if a patient’s tumor is controlled, sorafenib exposurein the pre-transplant period could be minimized by discontinuingdrug at a pre-specified MELD score when the chance of beingcalled for transplant becomes more likely. Based on the available

724 Journal of Hepatology 201

data, a minimum of 24 h would be desired, but longer would bepreferred. In addition, pre-clinical models have demonstrated anacceleration in the development of metastasis in mouse modelsafter short-term exposure to VEGF receptor tyrosine kinase inhib-itors [15]. This theoretical possibility of actually promoting recur-rence is concerning and would need to be assessed prospectively.

In reality, wait times for most patients with HCC are>6 months and many are treated with locoregional therapydespite the lack of prospective randomized data supporting itsuse in this setting. There is currently a double-blind multi-centerPhase III study comparing TACE and placebo versus TACE plussorafenib in patients with HCC before liver transplant (HeiLivCa)[16]. Results from this study should help clinicians better under-stand the risk and benefits of using sorafenib in this setting. How-ever, for patients that are not candidates for loco-regionaltherapy, there should not be a tendency to use sorafenib aloneas a bridge therapy at this time.

Ultimately, we hope to move beyond anatomic stage for theselection of patients with HCC for transplant. More molecularstudies are needed to identify and validate prognostic markersof favorable clinical behavior. In addition, the identification ofpredictive markers for those patients that are most likely to ben-efit from a management strategy that includes sorafenib isneeded.

Conflict of interest

R.F. acted as consultant for Bayer, Onyx, and Bristol MyersSquibb.

References

[1] Llovet JM et al. Sorafenib in advanced hepatocellular carcinoma. N Engl JMed 2008;359:378–390.

[2] Cheng AL et al. Efficacy and safety of sorafenib in patients in the Asia-Pacificregion with advanced hepatocellular carcinoma: a phase III randomised,doubleblind, placebo-controlled trial. Lancet Oncol 2009;10:25–34.

[3] Okita K, Imanaka K, Chida N, Tak WY, Nakachi K, Takayama T, et al. Phase IIIstudy of sorafenib in patients in Japan and Korea with advanced hepatocel-lular carcinoma (HCC) treated after transarterial chemoembolization (TACE).Abstract 128, 2010.

[4] Vitale A et al. Use of sorafenib in patients with hepatocellular carcinomabefore liver transplantation: a cost–benefit analysis while awaiting data onsorafenib safety. Hepatology 2010;51:165–173.

[5] Mazzaferro V et al. Liver transplantation for the treatment of smallhepatocellular carcinomas in patients with cirrhosis. N Engl J Med1996;334:693–699.

[6] Graziadei IW et al. Chemoembolization followed by liver transplantation forhepatocellular carcinoma impedes tumor progression while on the waitinglist and leads to excellent outcome. Liver Transpl 2003;9:557–563.

[7] Lu DS et al. Percutaneous radiofrequency ablation of hepatocellular carci-noma as a bridge to liver transplantation. Hepatology 2005;41:1130–1137.

[8] Clavien PA, Lesurtel M, Bossuyt PM, Gores GJ, Langer B, Perrier A, et al.Recommendations for liver transplantation for hepatocellular carcinoma: aninternational consensus conference report. Lancet Oncol 2011. [Epub aheadof print].

[9] Bose D et al. Vascular endothelial growth factor targeted therapy in theperioperative setting: implications for patient care. Lancet Oncol 2010;11:373–382.

[10] Cowey CL et al. Neoadjuvant clinical trial with sorafenib for patients withstage II or higher renal cell carcinoma. J Clin Oncol 2010;28:1502–1507.

[11] Margulis V et al. Surgical morbidity associated with administration oftargeted molecular therapies before cytoreductive nephrectomy or resectionof locally recurrent renal cell carcinoma. J Urol 2008;180:94–98.

[12] Hora C, Romanque P, Dufour JF. Effect of sorafenib on murine liverregeneration. Hepatology 2011;53:577–586.

2 vol. 56 j 723–725

JOURNAL OF HEPATOLOGY

[13] Strumberg D et al. Phase I clinical and pharmacokinetic study of the Novel

Raf kinase and vascular endothelial growth factor receptor inhibitor BAY 43–9006 in patients with advanced refractory solid tumors. J Clin Oncol2005;23:965–972.

[14] Miller AA et al. Phase I and pharmacokinetic study of sorafenib in patientswith hepatic or renal dysfunction: CALGB 60301. J Clin Oncol 2009;27:1800–1805.

Journal of Hepatology 201

[15] Ebos JM et al. Accelerated metastasis after short-term treatment with apotent inhibitor of tumor angiogenesis. Cancer Cell 2009;15:232–239.

[16] Hoffmann K et al. Prospective, randomized, double-blind, multi-center,Phase III clinical study on transarterial chemoembolization (TACE) combinedwith Sorafenib versus TACE plus placebo in patients with hepatocellularcancer before liver transplantation – HeiLivCa [ISRCTN24081794]. BMCCancer 2008;8:349.

2 vol. 56 j 723–725 725