Treatment of hepatocellular carcinoma before liver transplantation can curb local tumor progression and thereby prolong patients' transplantation eligibility.
Retrospective case-control pilot study. Twelve of 39 patients receiving liver transplantation for hepatocellular carcinoma had treatment before transplantation. Pretreatment included radiofrequency ablation (n = 8), percutaneous ethanol injection (n = 2), both modalities (n = 1), and tumor resection (n = 1). Twelve control subjects without pretreatment who were age-, sex-, and score-matched on the Model for End-stage Liver Disease and Child-Turcotte-Pugh classification were selected. The primary outcome measure was the waiting period for transplantation.
Patients with pretreatment waited on the transplant list significantly longer than their counterparts without pretreatment (median, 484 vs 253 days; P = .03).
Treatment before transplantation with tumor ablation or resection is associated with a longer waiting period on the transplant list. This enables patients who might otherwise be removed from the list because of tumor progression to receive transplantation.
Hepatocellular carcinoma (HCC) is the most common primary liver malignancy1 and the most frequent solid organ tumor.2 It is one of the world's 10 most common tumors,3 and its burden on society is not inconsequential, accounting for more than 1 million deaths per year.2 There has been a steep rise in incidence in the United States to 2.4 cases per 100 000 people,4 a rate that is expected to increase further in the coming years because of delayed consequences of the hepatitis C virus epidemic of the 1970s and 1980s.5
Patients who have small tumors may be eligible for multiple treatment alternatives, including partial hepatectomy (PH). Although PH provides a chance of cure, only a small percentage of patients with HCC undergo resection.6 Contraindications to PH include limited liver reserve and tumor size and location, and the resectability rate in patients with underlying cirrhosis is as low as 10%.7 For most patients, other locoregional therapeutic options exist, including transarterial chemoembolization, percutaneous ethanol injection (PEI), and radiofrequency ablation (RFA). All of these treatments, including resection, are associated with a high rate of tumor recurrence, because the entire liver is at risk for HCC.8- 10
Liver transplantation offers the dual advantages of eradicating the tumor and replacing the diseased liver. This modality has become the treatment of choice, especially for patients with end-stage liver disease.11 Favorable outcomes with 5-year tumor-free survival in the range of 70% have been reported in recent studies12- 15 for patients with stage I or II disease (ie, 1 nodule that is 2-5 cm or ≤3 nodules that are all <3 cm, without vascular invasion). The outcomes of orthotopic liver transplantation (OLT) in patients with stage III disease (ie, 1 nodule that is >5 cm or ≤3 nodules with at least 1 that is >3 cm) and higher have not been encouraging.15 Therefore, the United Network for Organ Sharing has adopted transplant listing criteria that exclude patients with stage III disease and higher.16
Orthotopic liver transplantation remains problematic because of the limited availability of organs.17 Patients with tumor progression beyond stage II while awaiting appropriate donor livers contribute to a high rate of being rejected from the transplant list. Dropout rates of 23% in 6 months18 and up to 40% in 12 months19,20 have been reported. When dropouts are included in the 2-year survival calculation, Llovet and colleagues14 have indicated that properly selected patients fare better with PH than OLT. Appropriate care for patients on the transplant list is controversial, and there is a lack of evidence-based recommendations.
Local tumor control while patients await OLT appears to hold promise to decrease the transplant list dropout rate in the era of an increasing waiting period. Recent suggestions of therapy while on the waiting list include PH, transarterial chemoembolization, and ablative techniques,21,22 but definitive studies are required.23 Although there are data that imply that RFA is safe in such a setting,22 no studies, to our knowledge, have compared the experience of patients who receive ablation as pretreatment with that of patients who do not. This study was undertaken to determine if such interventions intended as a bridge to OLT extend the waiting period for transplantation.
This was a retrospective case-control pilot study with a median follow-up of 18 months. Thirty-nine patients underwent OLT for end-stage liver disease associated with HCC between January 1999 and August 2003. All patients were placed on the transplant list and ultimately underwent transplantation at the same tertiary referral center.
Twelve patients were identified, each having had ablation or resection before transplantation. Methods of tumor ablation included RFA or PEI. Eight patients received RFA and 2 patients underwent PEI. One patient had both modalities of ablative therapy. Only 1 patient had resection of tumor before transplantation and was thus included in the pretreatment group. Of the 9 RFA procedures, 6 were performed percutaneously and 2 were done laparoscopically. The remaining case involved an open technique. Intraoperative ultrasonography was used for real-time monitoring during the ablation, and subsequent 4-phase computed tomography or contrast magnetic resonance imaging was used to document the degree of ablation every 3 months, on average.
Patients were matched with 12 control subjects without pretreatment before transplantation. Demographic matching criteria included age and sex. To facilitate comparisons of groups with similar levels of disease severity, patients and controls were also matched on Model for End-stage Liver Disease (MELD) and Child-Turcotte-Pugh scores. Tumor size was not a factor in control selection.
The primary outcome measure of interest was the waiting period for transplantation, defined as the date of placement on the transplant list to the date of first transplantation. Secondary outcome measures included pretreatment-associated morbidity, percentage of explanted viable tumor, and posttransplant recurrence and mortality.
The transplantation criteria changed in February 2002 to include the MELD score. We conducted subanalyses to compare the rates of pretreatment and the waiting period before and after February 2002.
One-tailed Wilcoxon rank sum tests were used to compare continuous data for patients and controls (eg, waiting time, which was highly skewed). Fisher exact tests were used for categorical data comparisons. All analyses were performed using R version 1.8.1 (Free Software Foundation, Boston, Mass).
This study was approved by the Human Subjects Review Committee of the University of Washington.
The results are summarized in Table 1. Patients with and without pretreatment had median ages of 52 and 54 years, respectively. Each group had 10 men and 2 women. The mean MELD and Child-Turcotte-Pugh scores were 10.5 and 7.0, respectively, in the group with pretreatment and 11.5 and 8.1 in the group without pretreatment. Although control subjects were not matched to patients based on tumor size, there was no significant difference in the size of the largest dimension of their tumor (median, 2.45 vs 2.50 cm; P = .53). The primary outcome measure was days on the transplant waiting list. Patients receiving pretreatment waited on the transplant list significantly longer than their counterparts who did not receive pretreatment (median, 484 vs 253 days; P = .03).
Eleven (55%) of 20 patients enrolled before February 2002 received pretreatment, and of the remaining 4 patients enrolled after that time, 1 received pretreatment. The median waiting period for transplantation before February 2002 was 517 and 307 days (P = .08) for the groups with and without pretreatment, respectively. The median waiting period after that date was 149 and 92 days (P = .25) for the 2 groups, respectively. The reduction in sample size in the subanalyses precluded achieving statistically significant results.
There were no significant complications associated with ablative therapy. Pathologic review of the explanted livers showed a mean ± SD of 25% ± 38.6% (range, 0%-95%) residual tumor at the sites of previous ablation. The distribution of residual tumor is presented in Figure 1. All but 1 patient had residual tumor on microscopic evaluation. One patient in the pretreatment group died 4 months after transplantation because of aplastic anemia. There were 2 HCC recurrences in the matched controls after transplantation and none in the pretreatment group (P = .48).
Distribution of residual tumor. All but 1 patient had residual tumor on microscopic evaluation.
Many treatments now exist for patients with HCC and well-compensated liver function; however, none, to our knowledge, have been evaluated in randomized controlled studies.24 Most clinicians agree that ultimately these patients need OLT; however, treatment of tumor progression in the interim period is variable, and dropout rates from the transplant list are as high as 40%.19,20 We evaluated whether pretreatment mostly with RFA affected the length of time patients remained on the transplant list compared with those without pretreatment.
Patients in this pilot study who received pretreatment before OLT remained on the transplant list significantly longer than their matched control subjects not receiving pretreatment (median, 484 vs 253 days; P = .03). These extra days are vital to increasing the probability for an OLT match in today's era of increasing need for organs in short supply.
After the implementation of the MELD score in February 2002, subanalyses of that data showed that the waiting period dropped to about one third of the previous waiting period. When each group, before and after February 2002, was further subdivided into those who received pretreatment and those who did not, we concluded that there was a similar distribution of patients and controls. In addition, a similar trend of waiting periods was observed regardless of whether the patient was listed before or after February 2002; the waiting period of those receiving pretreatment was 40% longer.
Ablation appears to be a safe, well-tolerated procedure, as no increase in posttransplant morbidity was observed among patients compared with controls. When RFA was used as the primary local ablative therapy, histologically proven residual disease was found in 91% of the explanted livers. Of these, viable tumor occupied on average 40% of the ablated sites. The reported local recurrence rate varies significantly in the literature and has been as high as 79% among patients whose explanted livers were examined microscopically.22 Although most studies of local recurrence are based on imaging results (ie, computed tomography, magnetic resonance imaging, and positron emission tomography), few histologic analyses have been conducted beyond a year from the time of ablation. Our findings are unique because of the long duration (ie, >1 year) between the time of ablation and the pathologic analysis; hence, our observed rate of disease recurrence may be more indicative of the long-term outcome following RFA.
Table 2 compares the local recurrence rate following RFA in contemporary series, including 2 studies that included pathologic data from explanted livers.8,22,25,26 Compared with conventional imaging techniques, microscopic analysis provides much greater sensitivity in detecting small-volume disease. Other factors contributing to postablation recurrence include RFA technique (ie, percutaneous, open, or laparoscopic), tumor size, and proximity of tumor to vessels.
Although it is accepted that patients with HCC need OLT, the timing of it is controversial. Should available organs be reserved for the sickest patients? Is there a role for pretransplant therapy? Graziadei et al27 documented that patients meeting United Network for Organ Sharing criteria and receiving pretreatment with transarterial chemoembolization to arrest tumor progression were able to remain on the transplant list, with excellent outcome after OLT. Belghiti and colleagues28 showed in a comparative study that PH extended the time on the transplant list a mean of 20 months and was an effective treatment, yielding no decrease in long-term survival after OLT. Had these patients been placed on the transplant list before the intervention, PH would have extended their ability to wait to 600 days. The Center for the Study of Liver Disease in Hong Kong described a series of 135 patients with 5-year survival of 70% when PH was the primary therapy followed by salvage OLT for HCC recurrence.29 Finally, another study18 of PH before OLT asserts an increase in the ultimate transplantation rate to 10.7% on a 24-month waiting list.
If pretreatment is determined to be appropriate, which modality, PH or ablation, is most effective? Ablative techniques as pretreatment are not well studied.18 Llovet et al18 demonstrated that PEI increased the rate of OLT by greater than 10% at 1 year. There are rare reports of the use of RFA during the waiting period for transplantation. Restaging and treating occult nodules with RFA has been shown to be beneficial in reducing the dropout rate.30 Also, in a case series by Fontana et al,22 RFA was an effective bridge to OLT therapy for 15 of 23 transplant candidates. As there is a paucity of comparative data on the subject, our pilot study documenting an extended waiting period with the use of ablative techniques encourages us to examine the questions further.
Alternative explanations for a comparatively longer waiting period in the pretreatment group were considered. It is possible that the matched controls received OLT faster because of deterioration of liver function and an associated advancement of priority status on the transplant list, suggesting that the key difference is not extension of the waiting period in the pretreatment group, but rather shortening of the waiting period in the control group. Although possible, we believe that this scenario is unlikely given that the ablative therapy in the pretreatment group would be prone to cause liver parenchymal damage and be associated with possible deterioration more so than nonintervention. Furthermore, the MELD scores of the 2 groups were equivalent at the time of enlistment for transplantation.
Patients in this pilot study were selected from the population of those successfully transplanted. An important question regarding ablative therapy is whether it enables patients who would otherwise be dropped from the list to remain on the transplant list after receiving RFA or PEI. We have shown that patients who receive pretreatment remain on the list longer than their counterparts who do not receive any pretreatment; however, we have not answered the question of whether ablation prevents dropout from the waiting list. An improved study design to address the role of ablation would include all patients who were originally listed for transplantation, comparing the rates of successful transplantation and time-to-event data between similar groups. An analysis of all pretreatments (PH, RFA, and PEI) as a follow-up to this pilot study is now under way. Conclusions of that study, in combination with this study's results showing that ablation is associated with a longer term on the waiting list, will help determine the feasibility of randomized controlled trials to definitively establish the role of pretreatment modalities in this patient population.
Treatment before transplantation with tumor ablative therapy or limited resection is associated with a longer waiting period on the liver transplant list. This may be one reason why patients who might otherwise be removed from the list because of tumor progression can be successfully transplanted. Pretreatment with RFA or PEI is safe and effective in bridging certain patients with HCC to transplantation, but when used as the primary therapeutic modality, RFA is associated with a high rate of histologically proven recurrence.
Robert R. Selby, MD, Los Angeles, Calif: For practical purposes, there is only one curative therapy for HCC in the cirrhotic liver, and that is liver transplantation. The converging problems of liver donor unavailability and cirrhotic liver unresectability create a dilemma that has left us struggling for interim therapies that would allow us to bridge the time to transplantation without allowing tumor progression. Add to that the development of a new form of intrahepatic therapy, RFA, whose utility and efficacy are not completely defined, and we have the ingredients of a very timely paper.
In that context, this paper is quite germane. It takes a retrospective look at cirrhotic patients with HCC who have been closely matched physiologically and who have comparable tumor sizes. Both the controls and the pretreated patients have gone on to transplantation. The primary end point is the time from candidate listing to liver transplantation, with the secondary end points of pretreatment morbidity and efficacy. The conclusions of the paper are that pretreatment results in patient transplantability that persists a long time following pretreatment and that this pretreatment is safe.
The MELD and CTP scores of patients at listing were quite low, which I assume means they were listed for HCC rather than liver failure. My first question is what were the MELD and CTP scores at transplantation? Were patients transplanted for liver failure or for time spent on the waiting list? Here, you have 2 groups of patients who were transplanted, and 1 group waited twice as long as the other. What is the explanation? Did patients with pretreatment demonstrate any radiologic progression while they were waiting, or did the tumors remain inert? Was there radiologic tumor progression in the controls? Rather than just knowing that there was 25% viable tumor in the treated sites of the explanted liver, I would like to know more pathologic information. What were the histologic and radiologic tumor stages of transplanted patients for both subsets? If they were similar at the outset, what were they at transplantation? In the beginning, you established a comparison between the 2 groups, but you have not completely developed that comparison. Although it is impressive that pretreated patients were still transplantable for cure at 484 days, how do we know that the same would not have been true in the control group?
Early in the paper, you say that tumor progression leads to a high dropout rate. But in your matched controls, there was no dropout since it was a retrospective study, and you defined controls that went on to transplantation. I think that in your study design you should pick a primary end point that reflects the ability of your therapy to control local tumor progression and spend more effort on the comparative pathologic data between the groups. Although the results in the study imply, they do not necessarily justify your conclusion, and although that data may be available, it has not been completely elucidated here.
Lastly, just a comment about the nature of RFA. In my mind, this is a spatially 2-dimensional therapy, even when employed intraoperatively. I believe that most failures result from inadequate application due to the difficulty to appreciate the 3-dimensional aspects of tumors on ultrasound imaging, either intraoperatively or in the radiology suite. Perhaps stereotactic localization and treatment, such as neoadjuvant therapy with external beam radiation (CyberKnife), may offer improved control with low morbidity, although I am not yet aware of any trials. Have you any knowledge of such practice?
Ryutan Hirose, MD, San Francisco, Calif: I have some issues that have been expressed by Dr Selby. You could still examine all of the patients that were listed, even in a retrospective fashion, and look at your dropout rate of patients that were either treated or not treated. I don't know that that is a favorable end point for anybody to wait longer for their transplant. I don't think that is actually something that is a good thing to happen to someone waiting for a liver transplant.
I also have some questions about the modalities of treatment. Several patients had ethanol injection, 1 person had resection, and the rest had RFA. How many patients were done by interventional radiologists in terms of the RFA vs surgeons, either laparoscopically or open? I believe that the laparoscopic approach often provides superior results than those performed by interventional radiologists.
Also, there have been data that suggest that percutaneous ethanol injections have a higher recurrence rate than RFA. In addition, there have been several studies that give us mixed messages about the role of resection in these patients with cirrhosis, whether it's a good thing or a bad thing. I want to know how the authors would decide which modality a patient should have. They also did not mention transarterial chemoembolization, which we think is also another modality by which these patients can be bridged towards transplantation. I also want to know what the authors think of the role of living donor transplantation in these patients. And finally, I would like to know whether patients ought to be downstaged. We realize that, with Mazzaferro's criteria if you use them, you get very good results. There is no argument with that. But there is a question of whether someone who is slightly outside those bounds can be downstaged either with RFA or resection and whether those patients might not qualify then for liver transplantation as well.
Paul D. Hansen, MD, Portland, Ore: I am intrigued by the high rate of residual disease. My experience is primarily with colorectal or other metastatic disease, less so with primary hepatocellular cancers. I have found over the last 7 years or so of doing RFA that, when you burn very aggressively, you get a better result. One of the concerns I have with patients who have hepatocellular cancer and cirrhosis is that, because they are often Child's B or C patients, I am somewhat less aggressive about how I burn, because I don't want to kill too much surrounding tissue. That may be one reason for the higher recurrence rate.
The second issue is that between 1998 and 2003 there was tremendous change in RFA devices. We started out with a 50-W 3-cm device, and we now use a 200-W 7-cm device with saline infusion. That should greatly affect our ability to successfully ablate the tumor. I am wondering if the author could comment on their change in devices used over time. Did they notice better results as time went on?
John P. Roberts, MD, San Francisco: Everybody is trying to figure out in the transplant community and probably in the nontransplant community whether to ablate and wait or ablate and then transplant when an organ becomes available. Ablate and wait means that you wait and see if there is a recurrence, and ablate and transplant is when you transplant them based on their stage. The priority of transplantation is unclear for those patients who have been ablated, primarily because we don't know if they have cancer still. Should they be prioritized for transplantation at the same risk as a patient who is not ablated? That's the area in which we lack clarity.
There was a study from Virginia, fairly similar, in which they found a 25% risk of having recurrence of tumor at the ablation site and another 25% risk of having tumor elsewhere in the liver. I was wondering if we could hear about tumor at the ablation site vs tumor elsewhere in the explanted liver, because I think this would help us understand the ablate and wait vs the ablate and transplant issue.
Susan L. Orloff, MD, Portland: I agree with many of the questions from the audience. They are some of the questions that I had, and I think this is a very interesting topic to discuss, especially for those of us involved in liver transplantation. What are the specific reasons for no pretreatment in the group of patients that were not pretreated? Most programs, I believe, treat the patients that have tumors (pretransplant), and why were these patients specifically not treated? That would be important to know.
Dr Yeung: I want to thank the discussants for many insightful comments regarding our study, which represents the first phase of a series of planned studies to examine the value of tumor ablation as a bridging modality in patients awaiting liver transplantation for HCC.
The aim of ablation treatment is to halt disease progression and thereby to extend patients' eligibility on the waiting list, as our data would support. Additionally, we would expect a decrease in the dropout rate for the treated group and ultimately be able to achieve equal or better long-term survival following transplantation. These latter hypotheses will be answered in the subsequent phase of the study.
Dr Selby asked about possible reasons for the observed differences in the waiting periods between the treated and untreated groups. As indicated in the presentation, there are a number of possibilities besides increasing the duration of eligibility secondary to a decreased dropout among those treated with ablation. For example, those in the untreated group may have more severe liver disease, although this is unlikely since both groups were matched for their MELD scores. Alternatively, there could be a delay to transplantation due to complications or recovery from ablation. Again, this is not supported by the observed median hospital stay of less than 24 hours following percutaneous or laparoscopic RFA and by the lack of significant postoperative morbidity. Lastly, there may be psychosocial issues in the treated group that could delay their time to transplant. However, this should not affect our calculation of the waiting period from the time of listing, when these issues have already been worked out. Of these possible explanations, we favor the interpretation that pretransplant therapy is effective in reducing disease progression and maintaining eligibility status for transplant.
Drs Hirose and Hansen accurately pointed out that local recurrence following RFA depends on a variety of factors, including the techniques employed. In our series, the ablations were equally divided between percutaneous, open, and laparoscopic approaches. The first of these was conducted by our radiologists and the latter 2 by our surgeons. The numbers in each group are too small for meaningful comparison. However, I think that our higher recurrence rate is more related to (1) the method of detection, microscopic vs gross imaging techniques; (2) length of the follow-up period, ours approaching 1½ years; and (3) the number of ablation sessions. Our patients received only one ablative procedure as long as their tumors remained within transplant criteria. Other factors such as size of lesions, evolving technology, and reporting bias would all contribute to the highly variable rates of recurrence reported in the literature. Several recent posttransplant series have consistently found over 50% local recurrence following RFA. Improvements in tumor targeting, heat distribution (especially for the larger lesions), and methods of real-time assessment of the ablation zone are much needed to achieve better outcomes in local control.
Finally, stemming from the paucity of evidence, it is highly desirable to conduct prospective analyses to determine the real impact of pretransplant tumor-directed therapies on the long-term survival of this population.
Correspondence: Raymond S. Yeung, MD, Department of Surgery, University of Washington School of Medicine, 1959 NE Pacific St, Seattle, WA 98195 (firstname.lastname@example.org).
Accepted for publication April 9, 2004.
This paper was presented at the 75th Annual Meeting of the Pacific Coast Surgical Association; February 15, 2004; Maui, Hawaii; and is published after peer review and revision. The discussions that follow this article are based on the originally submitted manuscript and not the revised manuscript.
We acknowledge the contribution of Lois Yamamoto for assistance in study coordination.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Surgery editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 18
Customize your page view by dragging & repositioning the boxes below.
Care at the Close of Life: Evidence and Experience
Integrating Palliative Care for Liver Transplant Candidates: "Too Well for Transplant, Too Sick for Life"
All results at
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.