Gastrointestinal Cancer Unit, October 12 University Hospital, Madrid, Spain
Gastrointestinal Cancer Clinical Research Unit, Clinical Research Program, Spanish National Cancer Research Center, Fuenlabrada University Hospital, Camino del Molino 2, Fuenlabrada, Madrid, Spain 28942
Hepatocellular carcinoma (HCC) is the fifth most common solid tumor, and its incidence has been progressively increasing in the Western world, in which it causes more than 600,000 deaths per year.1 At present, the only potentially curative treatments that are available are liver transplantation and liver resection; the former is indicated for patients with cirrhosis who have a limited functional reserve for partial hepatectomy. Even when the restrictive Milan criteria are followed, the relapse rate after transplantation is approximately 10% to 30%, and according to a recent review,2 there is currently no consensus on the most suitable therapeutic approach to this complication.
One interesting area of research in this field is the impact of immunosuppression on carcinogenesis and the development of HCC. Mammalian target of rapamycin (mTOR) inhibitors have been shown to have a direct antitumorigenic effect and to inhibit cell growth.3-6 In experimental models of HCC, the mTOR pathway has been aberrantly activated in up to 50% of cases.7 Although the currently available data are preliminary and come from retrospective studies,8, 9 it may be possible to switch to mTOR-based immunosuppressive therapy after recurrence10, 11 or even to use an mTOR inhibitor as a first-line treatment whenever the histopathological characteristics of the tumor (eg, microvascular involvement, an undifferentiated tumor, or noncompliance with the Milan criteria) indicate a high risk of relapse after transplantation.12, 13
Sorafenib, an inhibitor of multiple tyrosine kinases (including b-Raf), has recently been approved as a first-line treatment for advanced HCC.14 The activation of the Ras/mitogen-activated protein kinase pathway is a common finding in neoplastic processes (including HCC) and is a determinant for promoting cell proliferation and the survival of tumor cells. This makes sorafenib an interesting drug when its use as an adjuvant treatment after recurrence or as a treatment for nonsurgical relapse is being investigated. Although there are no studies assessing the safety and efficacy of the concomitant administration of sorafenib and mTOR inhibitors and the literature on this combination is scarce and mainly preclinical,11, 15 their combined use could have a synergistic effect.16, 17 To the best of our knowledge, this study presents the first reported series of patients with HCC recurrence after liver transplantation who were switched to mTOR-based immunosuppression in combination with an antineoplastic systemic treatment with sorafenib.
PATIENTS AND METHODS
This was a descriptive, open, multicenter, retrospective, uncontrolled cohort study designed to assess the combined use of an mTOR inhibitor (everolimus or sirolimus) and sorafenib in the treatment of patients with HCC recurrence after liver transplantation. The study was performed at 5 Spanish liver transplant units in accordance with the principles of the Declaration of Helsinki, the Good Clinical Practice guidelines, and current Spanish legislation on the protection of personal data. The study was also approved by the local ethics committee (October 12 University Hospital, Madrid, Spain), and all centers followed the same study protocol. All the patients included in the study were informed about the therapeutic protocol, accepted the treatment, and gave their written informed consent.
Between march 2008 and July 2010, this study enrolled all patients with a history of recurrent HCC after liver transplantation who were being treated with mTOR inhibitors and sorafenib for a tumor relapse that was not susceptible to locoregional therapy. In all cases, radiological confirmation was obtained with conventional imaging techniques (computed axial tomography and/or magnetic resonance imaging) before the treatment initiation. Two populations were identified for later analysis. The first was termed the safety population and included 31 patients who received sorafenib and an mTOR inhibitor either as an adjuvant or palliative treatment. The second population was designated the efficacy population and included the 26 patients who received this combination as a systemic treatment for tumor relapse. Five cases were excluded from the efficacy population: 4 patients who received treatment with mTOR and sorafenib as an adjuvant therapy after a radical treatment of their relapse and 1 patient who had received an mTOR inhibitor as immunosuppressive therapy before the tumor relapse. There were no differences in terms of the performance status, drug tolerance, Child-Pugh liver function status, or any other clinical parameters between these 2 groups.
The patients included in the study were switched to mTOR inhibitor (everolimus or sirolimus)–based immunosuppression after the confirmation of recurrence. Everolimus or sirolimus was employed at the investigator's discretion. Two to 4 weeks after the switch (when stable graft function was ensured), sorafenib (Nexavar, Bayer AG, Leverkusen, Germany) was added. In the case of everolimus (Certican, Novartis, Basel, Switzerland), the mean dose was 2.23 mg/day (range = 1.3-4.0 mg/day). For patients who received sirolimus (Rapamune, Pfizer, New York, NY), the mean dose was 2.63 mg/day (range = 1.0-4.0 mg/day). The mean plasma levels of everolimus were within the treatment levels accepted for maintenance immunosuppression. Sorafenib was used at doses approved for the treatment of advanced HCC (400 mg twice daily), and dose adjustments were made according to each patient's tolerance. Because of the absence of previous information on the potential toxicity or immunological consequences of this combination, the initial dosage of sorafenib was 400 mg/day in 16 cases, and this was progressively increased up to the maximum tolerated dose (800 mg/day). This treatment was maintained until the appearance of unacceptable toxicity, death, hepatic decompensation, or tumor progression justifying the suspension of the treatment. In 3 cases, tumor progression was treated with local therapies without the suspension of the combined treatment with sorafenib and the mTOR inhibitor.
Treatment Discontinuation and Dose Reductions
The temporary discontinuation of treatment was indicated for patients who presented with toxicity secondary to the therapeutic regimen according to the National Cancer Institute (NCI) Common Toxicity Criteria (CTC; version 3.0)18 or at the investigator's discretion. This withdrawal was maintained until the values returned to the threshold safety level, and subsequent dose reductions were made whenever they were necessary.
Assessment of the Efficacy and Toxicity
The baseline assessment included a detailed history and physical examination, a complete blood count with a differential leukocyte count, biochemistry and liver function tests, a coagulation study, and a tumor assessment with abdominal 3-phase contrast-enhanced computed axial tomography or magnetic resonance imaging. The adverse effects were continually assessed during all follow-up visits. If any adverse event was suspected, an assessment was completed with appropriate additional tests for each case.
All patients who received treatment with an mTOR inhibitor and sorafenib (the safety population) were considered for toxicity assessments.
A radiological evaluation of the treatment response was performed every 12 weeks during the follow-up. The objective efficacy parameters were considered to be the tumor response rate, the tumor progression-free interval, and the overall survival. The response to treatment was determined by the assessment of the target tumor lesions present at the beginning of the study and the appearance of new lesions. This response was classified according to the Response Evaluation Criteria in Solid Tumors (RECIST)19 through the establishment of the proportions of complete or partial responses, disease stabilization, and tumor progression. The objective response rate was calculated as the proportion of patients in whom a complete or partial response was observed versus the total number of assessable patients. The clinical benefit rate was defined as the proportion of patients achieving a partial response or disease stabilization according to RECIST that lasted more than 3 months. Progression-free survival was calculated from the beginning of the sorafenib treatment until progression or death from any cause. Overall survival for sorafenib was calculated from the beginning of the sorafenib treatment until death from any cause or the date of the last follow-up visit. The overall postrecurrence survival rate was calculated from the date of the first recurrence until death from any cause or the date of the last follow-up visit.
The severity of the reported toxicity was evaluated at each follow-up visit with the NCI CTC (version 3.0).18
The statistical analysis was performed with the SPSS statistics package for Windows (version 15.0). The toxicity assessment included all patients who had received at least 1 dose of the investigational therapeutic combination (31 cases). The efficacy parameters were determined for the 26 patients who received the treatment. The objective response rates and the proportions of adverse effects were calculated with regular estimates and 95% confidence intervals (CIs). The overall survival rate with sorafenib, the overall postrecurrence survival rate, and the progression-free survival rate were estimated with the Kaplan-Meier method.
CI, confidence interval; CTC, Common Toxicity Criteria; HCC, hepatocellular carcinoma; mTOR, mammalian target of rapamycin; NCI, National Cancer Institute; RECIST, Response Evaluation Criteria in Solid Tumors.
Thirty-one patients were enrolled between March 2008 and July 2010. Their main characteristics are shown in Table 1. In 29 patients (93.5%), HCC with concurrent cirrhosis was the indication for transplantation (48.4% had hepatitis C virus cirrhosis, and 45.2% had alcoholic cirrhosis); 2 cases (6.5%) presented with incidental HCC. Most patients underwent deceased donor liver transplantation; living donor liver transplantation was performed in only 2 cases (6.5%). All cases except those receiving a living donor liver transplant were selected according to the Milan criteria for liver transplantation. The median time from transplantation to recurrence was 22.6 months (range = 2.2-103.1 months). The immunosuppressive protocol used until recurrence is shown in Table 1. Recurrence was solely hepatic in 5 patients (16.1%), extrahepatic in 20 patients (64.5%), and both hepatic and extrahepatic in 6 patients (19.4%). In 9 cases (29.0%), there was synchronous recurrence at multiple locations.
Table 1. Baseline Characteristics of the Patients and Their Disease (n = 31)
ECOG Performance Status [n(%)].
This patient was excluded from efficacy population analyses.
After the diagnosis of posttransplant HCC recurrence, an mTOR inhibitor (everolimus or sirolimus) was introduced to replace calcineurin inhibitor immunosuppressants in 30 of the 31 patients: 22 patients received everolimus, and 8 received sirolimus (1 patient was already receiving everolimus). The types, doses, and serum levels of the new immunosuppressive therapy are shown in Table 2. In the patient already receiving everolimus, sorafenib was added to the treatment. The median time since treatment initiation with mTOR inhibitors to the start of the sorafenib treatment was 1.1 months (range = 0-21.5) in the case of everolimus and 1.4 months (range = 0-21.2) in the case of sirolimus.
Table 2. Maintenance Immunosuppression After the Diagnosis of Posttransplant HCC Recurrence: Pharmacokinetic and Dosing Data for mTOR Inhibitors
Plasma Level (ng/mL)
Mean ± standard deviation
2.6 ± 0.3
8.9 ± 1.1
Range (minimum to maximum)
Mean ± standard deviation
2.2 ± 0.2
5.5 ± 0.4
Range (minimum to maximum)
The administration of sorafenib was begun at the full dosage (800 mg/day) for 10 patients and at 400 mg/day for the other 16 patients. Four of these cases did not reach the full dosage because of side effects (2 patients were maintained at 400 mg/day, and the other 2 patients reached 600 mg/day). In 8 cases (30.8%), the sorafenib dosage was reduced because of toxicity. In 6 of these cases, the dosage was reduced from 800 to 600 mg, and in 2 cases, it was reduced from 800 and 600 to 400 mg (the total daily dosage).
Two of the 26 cases (7.7%) in the efficacy analysis could not be evaluated because of the short treatment period with sorafenib (<4 weeks) and the lack of imaging tests for tumor reassessment. There was 1 partial response, and there were 13 cases with disease stabilization as the best response. Table 3 shows the best tumor responses observed during the study and the overall objective response rate.
Table 3. Best Tumor Responses and Global Clinical Benefit Rate: Efficacy Population (n = 26)
NOTE: The data are presented as numbers and percentages.
The responses were not assessable because of the short treatment period with sorafenib (<4 weeks) and the lack of imaging tests for tumor reassessment.
The median overall survival since the start of the treatment with sorafenib was 19.3 months (95% CI = 13.4-25.1 months; Fig. 1). The median time to disease progression after the initiation of the sorafenib treatment was 6.77 months (95% CI = 2.3-11.1 months; Fig. 2). The median overall survival after relapse was 40.1 months (95% CI = 10.1-70.1 months; Fig. 3).
The most common adverse events probably related to the treatment are summarized in Table 4. The toxicity of the immunosuppressant treatment or its combination with sorafenib led to dose reductions in 2 of the 23 patients treated with everolimus and in 4 of the 8 patients treated with sirolimus. In 1 of the everolimus-treated patients, the toxicity required treatment discontinuation, and in another case, the treatment was changed to sirolimus. There were 4 recorded episodes of graft dysfunction; in 2 cases, the profile was considered mild with a rise in aminotransferase levels between 1.5 and 2 times the upper limit of normal. Treatment suspension and dose adjustments were not required. In the other 2 cases, hypertransaminasemia was found (levels > 10 times the upper limit of normal), and the temporary suspension of sorafenib and its delayed reintroduction at a reduced dose were required. There were no recorded or suspected episodes of acute or chronic graft rejection related to mTOR inhibitors and sorafenib.
Table 4. Hematological and Nonhematological Toxicity
Grade ≥ 3/4
NOTE: The data are presented as numbers and percentages and were obtained from patients who had received a combination of an mTOR inhibitor and sorafenib (n = 31). All data were graded according to the NCI CTC (version 3.0).
Severe toxicity episodes were infrequent and included 2 cases of upper digestive hemorrhaging, 1 case of central nervous system hemorrhaging, and 1 case of severe biventricular heart failure. Neither of the 2 episodes of severe upper digestive hemorrhaging among the 31 patients enrolled in this study (6.5%) seemed to be related to portal hypertension or peptic disease. In both cases, severe diffuse bleeding was observed via endoscopy. In one of these cases, death occurred soon after the bleeding episode, and in the other case, everolimus was stopped, and the patient was maintained on sorafenib. The central nervous system hemorrhage occurred 18 days after the initiation of the combination treatment with sorafenib and everolimus. Although the treatment was immediately withdrawn, the patient died 4 weeks after the bleeding episode. In this case, there was the unconfirmed suspicion of leptomeningeal metastases as the underlying cause of the bleeding. Finally, the patient with serious heart failure had a history of arteriosclerosis and peripheral artery disease, diabetes mellitus, and chronic kidney failure. An ischemic origin unrelated to sorafenib could not be ruled out.
At present, there is no consensus on the management of posttransplant HCC recurrence. In the different published series,10-12, 20-22 the mean time to recurrence has ranged from 8 to 23 months, and the majority of the cases have presented with dissemination with extrahepatic involvement, mainly in the bones and lungs. The current series of patients confirms previously reported data: a mean time to relapse of 22.6 months; multiple metastatic involvement in 29.0% of the cases involving primarily the lungs (41.9%), bones (19.3%), or lymph nodes (19.3%); and only 5 cases (16.1%) with the liver as the sole site.
The overall 5-year survival rates after recurrence in the first reported series were 22% to 23%,20, 21 although the most recently published series have reported overall 5-year survival rates higher than 40%,10-12 probably because of earlier diagnoses and more patients amenable to surgery or locoregional ablative therapies.
In all series, patients undergoing surgery with a curative intent (liver resection with or without metastasectomy) had a significantly better prognosis (80% 5-year survival)22 than those who were not candidates for radical treatments (0% 5-year survival). Locoregional therapies such as radiofrequency ablation and chemoembolization have not been extensively studied in this setting, although they have been occasionally used. Liver retransplantation, although performed anecdotally,23 is not an accepted indication, although it can be considered in select cases with de novo HCC in a cirrhotic transplanted liver.24 Because many patients with posttransplant recurrence have no underlying cirrhosis, liver resection is an attractive therapeutic option. However, most patients develop extrahepatic recurrences, and when it is possible, surgical treatment is rarely curative because the rate of re-recurrence is higher than 50%11, 20, 21; there is no established adjuvant treatment. The patients included in this study represent a special population with a poor prognosis because they experienced HCC relapse that was not susceptible to local therapies with a radical intent. There is growing evidence indicating that calcineurin inhibitors (cyclosporine and tacrolimus) could favor tumor growth and recurrence.25-27 On the other hand, mTOR inhibitors have a direct antitumorigenic effect, inhibit cell growth,3-6 and have been approved at an appropriate and specific antineoplastic dose for other solid tumors such as renal cell carcinoma.28 Data from retrospective studies8, 9 suggest that even at immunosuppressive doses (which are lower than those used in the oncological setting), mTOR inhibitors could be useful in patients with posttransplant HCC recurrence.
Sorafenib, a multiple tyrosine kinase inhibitor, has been recently approved as the first systemic treatment for advanced HCC in patients with cirrhosis,14 but to date, data on its use in the transplant setting are scarce.29 Preclinical models suggest that the combination of an mTOR inhibitor and sorafenib could be beneficial because of their synergistic effect. According to a recent publication,15 a patient treated with everolimus and sorafenib showed encouraging results: 3 months after the combined treatment was started, the patient presented with a partial response according to RECIST, and 8 months later, the patient was still alive with a significant reduction in the tumor size and few side effects (except for a grade II digestive hemorrhage that required the temporary withdrawal of the mTOR inhibitor). In addition to this publication, a recent letter to the editor mentioned 3 more cases treated with this combination, but the results were somewhat less positive (1 case of disease stabilization).30 Besides these publications, we have found no other documented series of patients treated with this combination.
Our report reinforces the described case and sets the overall survival time after the initiation of the combined treatment with an mTOR inhibitor and sorafenib at 19.3 months. In our study, most side effects of this combination were grade 1/2 and were consistent with those described in a nontransplant setting. It is of paramount importance for transplant patients that no cases of rejection were described in this study, and only 2 patients developed significant liver dysfunction (hypertransaminasemia). There does not seem to be any temporal relationship between the initiation of treatment and the early appearance of vascular events or the risk of digestive hemorrhaging. In the case with severe biventricular heart failure, the myocardial damage was probably due to the antiangiogenic effect of the treatment on a chronically hypoxemic myocardium. This conclusion is based on the baseline characteristics of the patient in whom this effect occurred (diabetes, chronic kidney failure, episodes of intermittent claudication, and overt signs of peripheral artery disease due to digital necrosis); the heart failure was not due to a direct toxic effect of the drug on myocytes. According to our experience, the combination of sorafenib and an mTOR inhibitor has a significant but manageable toxicity profile, and a careful assessment of potential vascular or hemorrhagic complications should be performed for all patients. This combination should not be considered an option for patients with a history of bleeding diathesis or coagulopathy; for patients currently using anticoagulants such as vitamin K antagonists, unfractionated heparin, or low-molecular-weight heparin; or for patients with significant cardiovascular impairment, unstable angina, or myocardial infarction or with any history of cerebral vascular accidents, transient ischemic attacks, or grade 2 or higher peripheral vascular disease.
The main limitations of this study are inherent to its design, and this is due in large part to the lack of previous data on the use of this combination. Because this was an uncontrolled study, the absence of a control group of patients (for comparisons of the results) and the retrospective nature of the study reduced the evidence level of the conclusions. Nevertheless, the main study variables were objective and were easily assessable, and this reduced the placebo and Hawthorne effects as much as possible.
In conclusion, the results of this study suggest that a combined treatment with mTOR inhibitors and sorafenib for posttransplant HCC recurrence not amenable to curative treatments should be considered an experimental therapeutic option for this poor prognosis population. The lack of relevant immunological complications and the potential efficacy have to be balanced against the significant toxicity with this new therapeutic approach. Therefore, we think that this experimental combination warrants further evaluation in prospective, controlled clinical studies.
The authors thank Traducciones Lexic S.L. for its contribution to the drafting of this article.