Hepatocellular carcinoma (HCC) is the most common primary malignancy of the liver and its incidence is rising.1 Both primary liver transplantation (LT) and locoregional therapy, done with curative intent can cure early-stage disease, which is usually defined as HCC within Milan criteria. Locoregional therapy done with curative intent (referred to as “curative locoregional therapy [CLRT]” in this article) usually refers to resection. However, resection is not possible in many patients because of location or size of the tumor, or the severity of the cirrhosis and portal hypertension. Thermoablation and cryoablation are forms of locoregional therapy that have been used with curative intent, in patients who are not resection candidates. Salvage LT (SLT) may be done if HCC recurs after CLRT. Providers generally prefer primary LT rather than CLRT (with or without SLT) to treat HCC in patients who have cirrhosis and portal hypertension, because it treats both the HCC and the cirrhosis. Liver transplantation gives excellent survival in patients with HCC within Milan or University of California at San Francisco criteria, but there are insufficient organs available.2 The overall survival (OS) and disease-free survival (DFS) for CLRT alone, and CLRT-SLT combined, as compared with primary LT for HCC remain unclear. We thus performed a systematic review and meta-analysis of the literature of all studies that evaluated OS and DFS of patients with early-stage HCC that were treated with either CLRT or primary LT. We included all forms of CLRT where the stated intent was to cure the patient, including resection, resection with SLT and thermal ablation.
MATERIALS AND METHODS
Selection of Studies
We performed a literature search of MEDLINE (1946 to April 2015), PUBMED, Scopus databases, using keywords and/or medical subject headings for HCC, LT, hepatectomy, catheter or radiofrequency ablation (RFA), salvage therapy, survival rate or treatment outcome. For the Scopus database, only free text searches with truncations were carried out. We restricted the search to human studies and studies published in English language. We also performed a search of clinical trials.gov using the keywords HCC, transplantation, surgical resection (SR), RFA, and/or locoregional therapy to identify unpublished trials that may be included in our study. Two reviewers (A.R.M., and S.P.) independently screened the database search for titles and abstracts and retrieved relevant studies. The bibliographies of the included articles were searched manually for any additional references.
We included all studies that compared the survival of patients with HCC after CLRT (including SLT), with primary LT. We included studies of any method of CLRT such as SR, RFA, and/or microwave ablation. We only included studies that reported the 3- and/or 5-year OS and DFS. All studies included in the meta-analysis are nonrandomized, observational cohort studies, and thus subject to potential confounders.
We analyzed 3- and 5-year OS, and DFS, of patients with HCC who were treated with (i) CLRT versus primary LT; (ii) Intention-to-treat analysis of CLRT-SLT versus primary LT. In addition, we performed subgroup analysis of CLRT versus primary LT in patients within (i) Child A cirrhosis or no cirrhosis and (ii) single HCC lesions. We chose the subgroups for their relevance in clinical practice before we performed the analysis (a priori).
We excluded abstracts, case reports, review articles, clinical guidelines, and consensus documents. We excluded studies that did not provide information on 3- or 5-year survival, or reported survival of 1 treatment modality only. We also excluded studies that compared different modalities of CLRT only. We only included reports where patients in both arms of the study were otherwise potential candidates for LT (eg, we excluded studies of patients with metastasis, coexisting serious illness). We also excluded studies of palliative locoregional therapy such as transcatheter arterial chemoembolization, bland embolization or radioembolization. Finally, we excluded studies that included subsets from previously published articles by the same authors.
Two reviewers (A.R.M., and S.P.) independently extracted data from each study using a standardized data extraction sheet. We extracted data on the last name of the first author, year of publication, mean age and sex of the study participants, presence or absence of cirrhosis, Child-Pugh class, mean Model for End-Stage Liver Disease (MELD) score, tumor characteristics, type of CLRT used, the mean follow-up period and percentage of recurrence of HCC. We also extracted data on the OS and DFS at 3-, 5-, and 10-year posttreatment, when available. The reviewers then cross checked each other’s results and resolved disparities in extracted data by discussion. We assessed the quality of the studies using the modified Newcastle Ottawa score for cohort studies3 and assessed reporting bias using funnel plots.
We used Review Manager 5.2 software (The Cochrane Collaboration, Oxford, UK) for this meta-analysis, and the Mantel-Haenszel method to calculate the pooled odds ratios (ORs) and 95% confidence intervals (CIs) of OS and DFS of patients at 3 years and 5 years, for all strata.4 We assessed heterogeneity by means of a χ2 test (Cochran Q statistic) and quantified this with the I2 statistic. If I2 > 50%, there was significant heterogeneity and we used a random-effects model. Generally, I2 was used to evaluate the level of heterogeneity, assigning the categories low, moderate, and high to I2 values of 25%, 50%, and 75%, respectively. We performed a meta-regression analysis of subgroups according to Child-Pugh class of cirrhosis and single HCC lesions. Funnel plots were obtained to assess risk of bias. We used the two 1-sided test to determine noninferiority of curative locoregional therapy followed by salvage liver transplant compared to primary LT.5 We set the noninferiority margin at 20%, meaning that the CLRT-SLT should not be more than 20% inferior to primary LT. We chose 20% as the margin of noninferiority based on our clinical judgment and that a 20% loss of efficacy was an acceptable trade-off for the benefit of saving the life of a transplant candidate with advanced cirrhosis but no HCC, who would otherwise have died if the liver had been used to transplant a patient with HCC. We established that noninferiority was present if the lower limit of the 95% CI of the pooled OR for the difference between CLRT-LRT and primary LT was above 0.8 (we set the significance value α = 0.025 and interval = (1 - 2α) × 100% of CI).
Selection of Studies
The search of databases yielded 506 references. Among the 506 references in the databases, there were 480 studies after removal of repeat articles. We excluded 414 articles on title and abstract including 137 review articles, 18 case series or case reports, 7 letters to the editor, comments or updates, 8 articles on basic sciences, 34 studies about bridging or down-staging prior to transplantation, 36 studies about palliative locoregional therapies such as transcatheter arterial chemoembolization/radioembolization, 103 studies that did not involve treatment with either CLRT or LT, 12 that were consensus statements or expert opinions, 8 that described surgical techniques of HCC, 15 that were about predictors of outcome with HCC and 36 that were not in English. We retrieved 66 articles to assess for eligibility and excluded 18 of these as they did not have a comparison group. The remaining 48 retrospective studies comparing the survival of patients with HCC with CLRT versus LT were included in our meta-analysis. The search of clinicaltrials.gov yielded 516 results; however, none of the trials yielded data that could be included in our meta-analysis.
We included 48 studies with 9835 patients (5736 patients with CLRT and 4119 patients with primary LT) in the meta-analysis.6-50 The mean age was 58.6 and 54.3 years in the CLRT and LT groups; and 77.3% and 79.3% were men, respectively. The patients had a mean MELD score of 10.6 and 15; and were Child-Pugh class A in 76.8% and 45.4% in the CLRT and primary LT groups, respectively.
SR was used in 40 studies, RFA in 3 studies; microwave ablation in 1 study, SLT in 12 studies and a combination of LRT in 8 studies. In all studies, the CLRT was compared to primary LT which was with deceased donor liver transplant in 37 studies and live donor liver transplant in 11 studies. Patients were all within Milan criteria in 20 studies, all outside Milan criteria in 5 studies, some within and some outside of Milan criteria in 23 studies. In the CLRT group, 10 studies had patients with Child A cirrhosis only, whereas 38 studies had patients with Child A, Child B, or Child C cirrhosis in the CLRT group. The liver transplant group included Child’s A, B, or C patients in all studies.
(1) CLRT versus Primary LT: This overall analysis included all 48 studies,6-50 in which 5736 patients received CLRT, mostly with SR, and 4119 patients received Primary LT. CLRT was significantly inferior to primary liver transplant for 3-year OS (OR, 0.69; CI, 0.56-0.85; P < 0.01), 5-year OS (OR, 0.59; 95% CI, 0.48-0.71; P < 0.01), and 5-year DFS (OR, 0.30; 95% CI, 0.22-0.41; P < 0.01) in patients with HCC (Table 1; Figures 1 and 2). When stratified by the type of CLRT, both SR alone and thermal ablation alone were significantly inferior to primary LT (P < 0.01), for both OS and DFS in patients with HCC.
(A) CLRT versus Primary LT in patients with Child’s A cirrhosis: In this subgroup analysis, we included 10 studies where CLRT was done in patients with Child’s A cirrhosis; these included 723 patients with CLRT and 628 patients with Primary LT.6-14,48 Resection was the most common form of CLRT. There was no significant difference in the 3-year OS (OR, 0.79; 95% CI, 0.5-1.25; P = 0.31), and 5-year OS (OR, 0.72; 95% CI, 0.31-1.34; P = 0.30) for CLRT and primary LT. We could not show noninferiority, as the lower limit of the 95% CI was below 0.8. The 5-year DFS for CLRT was inferior to primary LT (OR, 0.3; 95% CI, 0.13-0.64; P < 0.01) (Table 1, Figures 3-4).
(B) CLRT versus Primary LT in patients with single HCC lesions: In this subgroup analysis, we included 5 studies where CLRT was done in patients with single HCC lesion; these incorporated 573 patients with CLRT and 408 patients with Primary LT.8,12,22,25,26 There was no significant difference in the 5-year OS (OR, 0.79; 95% CI, 0.61-1.01; P = 0.11) for CLRT and primary LT. We could not show noninferiority, because the lower limit of the 95% CI was below 0.8. The 5 year DFS for CLRT was inferior to primary LT (OR, 0.34; 95% CI, 0.2-0.6; P < 0.01) (Table 1).
(2) CLRT-SLT versus primary LT—an intention-to-treat analysis: In this analysis, we included 5 studies where patients initially had resection and in those with recurrent HCC after CLRT, some but not all, went on to receive salvage liver transplant.15,21-24 The outcomes of the group overall, including those receiving or not receiving salvage LT, were compared with the outcomes of all patients listed for primary liver transplant to treat HCC. Each of these 5 studies15,21-24 had reported outcomes using an intention-to-treat analysis. The characteristics of the studies and their methodological quality are provided in Table 2.
The studies include 457 patients who received SR, and 999 patients who received primary LT. Patients mostly had Child-Pugh class A cirrhosis, and had 1 to 2 nodules of HCC ranging from 5 to 6.5 cm in diameter. The patients who underwent SR were all transplantable at the time of inclusion in the studies (patients with HCC beyond Milan criteria were potentially transplantable with exception points after down-staging), and were followed up for a mean of 52.2 months after resection.
HCC recurred after SR in 68.1% (range, 49%-80%) of the patients and salvage LT was done in 32.5% (range, 22-49%) of these patients. In those with recurrence, 67.5% (range, 51%-78%) were not eligible for salvage LT because of HCC recurrence outside Milan criteria (51%), advanced age (41%) and acquired comorbidities or patient refusal (8%). A mean of 31.9% (range, 20-37%) had no recurrence of HCC and did well after surgery.
In the intention-to-treat analysis, the 5-year OS for CLRT-SLT was not significantly different from primary LT, (OR, 1.0; 95% CI, 0.6-1.7; P = 1.0). We could not show noninferiority, as the lower limit of the 95% CI was below 0.8. The DFS was worse for CLRT-SLT (OR, 0.31; 95% CI, 0.15-0.61; P < 0.01) (Table 1; Figures 5 and 6).
Most providers consider LT to be the better treatment modality than SR or other forms of locoregional therapy done with curative intent to treat, early-stage primary liver cancer, even though many studies have shown that SR provides good OS in these patients. Our meta-analysis of 48 studies showed that 3- and 5-year OS and DFS is better after primary liver transplant compared with all types of curative locoregional therapy (resection, thermal ablation, CLRT-SLT) when these are analyzed together as group. However, these studies are potentially confounded. To address these confounding factors, we did subanalyses of studies that compared similar patients.
Subanalysis of studies that only included patients with Child A or no cirrhosis showed that the OS with CLRT was not significantly different to primary LT. However, we could not show noninferiority due to wide CIs. In addition, the DFS was worse for CLRT than primary LT. DFS was worse despite the fact that the patients in the CLRT group has lower MELD scores and had fewer HCC lesions than patients in the primary LT group. However, even though the DFS after CLRT in subjects with Child A cirrhosis is worse than primary LT, CLRT should still be considered an important option, in the setting of current organ shortage. This allows a donor liver to be allocated to a sick cirrhotic patient with high MELD score instead of a Child’s A cirrhosis patient with a low MELD score who can potentially be cured of cancer by locoregional therapy. The absence of noninferiority is in part due to the available data being quite limited, and hence lacking sufficient power to allow conclusive inference and further supports the need for adequately powered prospective studies. About 30% of patients survive long term after CLRT and never need a transplant.51 Thus, the United Network for Organ Sharing recommends that physicians perform resection or do other forms of locoregional therapy for patients who have HCC that is within Milan criteria, where feasible, to save donor organs for the recipient pool.
Salvage transplant for tumor recurrence after CLRT can be done to rescue the patient if there is recurrence. To ensure that SLT is not detrimental to the patient, short- and long-term DFS and OS should not be worse than that in primary liver transplant. Studies have shown that the OS and DFS are similar for SLT and primary LT.7,8,14-20,48 However, these studies included patients who actually received a salvage liver transplant if they had a recurrence of HCC. This introduces a number of potential confounding factors, particularly survivor bias. We therefore analyzed all studies that used an “intention-to-treat” design where patients treated with CLRT had the option of SLT if the tumor recurred, and compared them with patients who were placed on the waiting list for liver transplant. Surprisingly, only 32.5% (range, 22-49%) of patients with tumor recurrence after initial CLRT in effect went on to get a transplant. Although our meta-analysis showed that the outcomes for CLRT-SLT were not significantly different to primary LT with respect to OS at 3 and 5 years, we could not show noninferiority. Also, we found that the DFS was worse for CLRT-SLT at 5 years. This contradicts outcomes based on Markov-decision analysis.52 The difference is likely due to the 40-80% rate of salvage transplantation for tumor recurrence that the authors built into the Markov model, which was much higher than the 32.5% (20%-49%) rate in real life. On the other hand, we only had retrospective cohort studies available to analyze. Selection bias in these studies could account for the low rate of salvage transplantation. Good quality randomized controlled studies are needed to answer this question.
To our knowledge, our study is the first meta-analysis in the literature that compares CLRT-SLT versus primary liver transplant in an intention-to-treat basis. The meta-analysis highlights an unexpected finding in all the “intention-to-treat” studies, namely, that the majority of patients with recurrence after CLRT are unsalvageable with SLT, because their tumors are outside of Milan criteria. The 5 studies making up the “intention-to-treat” subanalysis included patients with 1 to 2 nodules of HCC ranging from 5 to 6.5 cm in diameter. Many of these patients would have been excluded from transplant in the United States, because tumor was beyond Milan criteria. It is possible that the patients chosen for CLRT started with a higher disease burden which could account for the higher tumor recurrence in the CLRT group. Also, 41% did not receive a SLT because they were considered to be too old, suggesting they may have been chosen for CLRT-SLT because they were older to start with.
Tumor recurrence in the CLRT-SLT studies was outside Milan Criteria when first detected, in 51% of patients. This suggests that survival may be improved if researchers could identify this subgroup at the outset, and possibly choose primary LT as initial treatment. Sala et al53 showed that microvascular invasion and presence of satellite nodules predicted early, advanced recurrence. Similarly, Fuks et al15 identified microscopic vascular invasion, satellite nodules, tumor size > 3 cm, poorly differentiated tumor and liver cirrhosis as predictors of recurrence outside Milan criteria. Guerrini et al24 found that the number of HCC nodules and alpha-fetoprotein level predicted recurrence outside Milan criteria. The authors of the “intention-to-treat” studies may have chosen resection over transplant for patients with biologically aggressive HCC. It is likely that these patients would have had early recurrence after primary LT too. Thus, although 5-year DFS was worse in the salvage liver transplant group in the intention-to-treat analysis, confounding factors, such as age, tumor biology, and tumor size, among others, may have caused the worse DFS. Randomized controlled studies are needed to establish whether long-term DFS is truly worse after CLRT-SLT than after primary LT, though this may not be feasible.
Although the meta-analysis did not include high-quality randomized controlled studies, it did include reasonable quality cohort studies, based on the Modified Newcastle Ottawa quality assessment scale of cohort studies. The absence of randomized controlled trial data is a major limitation of this meta-analysis. Another limitation of our study is the presence of significant heterogeneity among the studies, as shown by the asymmetry of the funnel plots. The subgroup analysis eliminated some, but not all of the heterogeneity. We believe that the heterogeneity among the studies was due to the clinical diversity among the studies. We however anticipated the clinical heterogeneity among the studies and used the random effects model to perform the meta-analysis. Finally, because we included diverse group of therapies such as SR or RFA under CLRT, our results do not generalize equally to each type of CLRT.
This meta-analysis clearly shows that primary liver transplant is the better option for the patient from an individual’s perspective but we do not address the broader issue of whether this is the best use of a scarce resource for the population of listed patients. Patients listed with exception points for HCC have a substantial survival advantage over non-HCC patients.51 Because there are not enough organs available, more non-HCC patients die on the waiting list as more HCC patients are transplanted. This also drives geographic discrepancy in liver organ allocation.54 In the United States, the organ procurement and transplant network sets policy to mitigate the overwhelming number of transplants going to HCC patients, to allow more transplants to go to the sick liver failure population, who die at much higher rates than HCC patients.54
Primary LT has better OS than curative locoregional therapy in patients with HCC and Child-Pugh class B/C cirrhosis. In patients with Child-Pugh class A cirrhosis and a single HCC lesion, primary LT was not superior to CLRT on primary efficacy testing. However, we could not show noninferiority for CLRT as compared to primary LT on noninferiority testing. There was no recurrence of HCC in about 30% of patients at 5 years after CLRT and these patients did not require SLT. A substantial number of lives would thus be saved if organs from this subgroup went instead to candidates without HCC who would otherwise not have received an organ. Unfortunately, about 70% of patients had recurrence of HCC after CLRT, of which only 30% were eligible for SLT. The possibility that a large proportion of patients with HCC would have a recurrence after CLRT and not be eligible for SLT would make this an unpalatable option for many patients.
Prospective controlled studies are greatly needed to compare CLRT and primary LT. Further research is also needed to reduce the recurrence of HCC beyond Milan criteria in patients undergoing resection, so as to make SLT more viable. Information from this meta-analysis should help decision making with the patient and inform the policy debate about allocation of deceased donor livers.
The authors thank Bruce and Joyce Miller for their fellowship research memorial award.
1. Altekruse SF, McGlynn KA, Reichman ME. Hepatocellular carcinoma incidence, mortality, and survival trends in the United States from 1975 to 2005. J Clin Oncol
2. Massie AB, Caffo B, Gentry SE, et al. MELD exceptions and rates of waiting list outcomes. Am J Transplant
3. Wells G, Shea B, O’connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of non-randomized studies in meta-analyses. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp
. Published July 2000.
4. Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst
5. Schuirmann DJ. A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability. J Pharmacokinet Biopharm
6. Bellavance EC, Lumpkins KM, Mentha G, et al. Surgical management of early-stage hepatocellular carcinoma: resection or transplantation? J Gastrointest Surg
7. Belghiti J, Cortes A, Abdalla EK, et al. Resection prior to liver transplantation for hepatocellular carcinoma. Ann Surg
8. Margarit C, Escartín A, Castells L, et al. Resection for hepatocellular carcinoma is a good option in Child-Turcotte-Pugh class A patients with cirrhosis who are eligible for liver transplantation. Liver Transpl
9. Bigourdan JM, Jaeck D, Meyer N, et al. Small hepatocellular carcinoma in Child A cirrhotic patients: hepatic resection versus transplantation. Liver Transpl
10. Choi SH, Choi GH, Kim SU, et al. Role of surgical resection for multiple hepatocellular carcinomas. World J Gastroenterol
11. Foltys D, Zimmermann T, Kaths M, et al. Hepatocellular carcinoma in Child's A cirrhosis: a retrospective analysis of matched pairs following liver transplantation vs. liver resection according to the intention-to-treat principle. Clin Transplant
12. Llovet JM, Fuster J, Bruix J. Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology
13. Poon RT, Fan ST, Lo CM, et al. Difference in tumor invasiveness in cirrhotic patients with hepatocellular carcinoma fulfilling the Milan criteria treated by resection and transplantation: impact on long-term survival. Ann Surg
14. Scatton O, Zalinski S, Terris B, et al. Hepatocellular carcinoma developed on compensated cirrhosis: resection as a selection tool for liver transplantation. Liver Transpl
15. Fuks D, Dokmak S, Paradis V, et al. Benefit of initial resection of hepatocellular carcinoma followed by transplantation in case of recurrence: an intention-to-treat analysis. Hepatology
16. Hwang S, Lee SG, Moon DB, et al. Salvage living donor liver transplantation after prior liver resection for hepatocellular carcinoma. Liver Transpl
17. Liu F, Wei Y, Wang W, et al. Salvage liver transplantation for recurrent hepatocellular carcinoma within UCSF criteria after liver resection. PLoS One
18. Moon JI, Kwon CH, Joh JW, et al. Primary versus salvage living donor liver transplantation for patients with hepatocellular carcinoma: impact of microvascular invasion on survival. Transplant Proc
19. Shao Z, Lopez R, Shen B, et al. Orthotopic liver transplantation as a rescue operation for recurrent hepatocellular carcinoma after partial hepatectomy. World J Gastroenterol
20. Wu L, Hu A, Tam N, et al. Salvage liver transplantation for patients with recurrent hepatocellular carcinoma after curative resection. PLoS One
21. Adam R, Azoulay D, Castaing D, et al. Liver resection as a bridge to transplantation for hepatocellular carcinoma on cirrhosis: a reasonable strategy? Ann Surg
. 2003;238:508–518; discussion 518–9.
22. Sapisochin G, Castells L, Dopazo C, et al. Single HCC in cirrhotic patients: liver resection or liver transplantation? Long-term outcome according to an intention-to-treat basis. Ann Surg Oncol
23. Del Gaudio M, Ercolani G, Ravaioli M, et al. Liver transplantation for recurrent hepatocellular carcinoma on cirrhosis after liver resection: University of Bologna experience. Am J Transplant
24. Guerrini GP, Gerunda GE, Montalti R, et al. Results of salvage liver transplantation. Liver Int
25. Bismuth H, Chiche L, Adam R, et al. Liver resection versus transplantation for hepatocellular carcinoma in cirrhotic patients. Ann Surg
26. Bronowicki JP, Boudjema K, Chone L, et al. Comparison of resection, liver transplantation and transcatheter oily chemoembolization in the treatment of hepatocellular carcinoma. J Hepatol
27. Castells A, Bruix J, Bru C, et al. Treatment of small hepatocellular carcinoma in cirrhotic patients: a cohort study comparing surgical resection and percutaneous ethanol injection. Hepatology
28. Cillo U, Vitale A, Brolese A, et al. Partial hepatectomy as first line treatment for patients with hepatocellular carcinoma. J Surg Oncol
29. Colella G, Bottelli R, De Carlis L, et al. Hepatocellular carcinoma: comparison between liver transplantation, resective surgery, ethanol injection, and chemoembolization. Transpl Int
. 1998;11(Suppl 1):S193–S196.
30. De Carlis L, Giacomoni A, Pirotta V, et al. Surgical treatment of hepatocellular cancer in the era of hepatic transplantation. J Am Coll Surg
31. Facciuto ME, Koneru B, Rocca JP, et al. Surgical treatment of hepatocellular carcinoma beyond Milan criteria. Results of liver resection, salvage transplantation, and primary liver transplantation. Ann Surg Oncol
32. Fan ST, Poon RT, Yeung C, et al. Outcome after partial hepatectomy for hepatocellular cancer within the Milan criteria. Br J Surg
33. Fan HL, Chen TW, Hsieh CB, et al. Liver transplantation is an alternative treatment of hepatocellular carcinoma beyond the Milan criteria. Am J Surg
34. Figueras J, Jaurrieta E, Valls C, et al. Resection or transplantation for hepatocellular carcinoma in cirrhotic patients: outcomes based on indicated treatment strategy. J Am Coll Surg
35. Harada N, Shirabe K, Ikeda Y, et al. Surgical management of hepatocellular carcinoma in Child-Pugh class B cirrhotic patients: hepatic resection and/or microwave coagulation therapy versus living donor liver transplantation. Ann Transplant
36. Ho CM, Lee PH, Chen CL, et al. Long-term outcomes after resection versus transplantation for hepatocellular carcinoma within UCSF criteria. Ann Surg Oncol
37. Kitisin K, Packiam V, Steel J, et al. Presentation and outcomes of hepatocellular carcinoma patients at a western centre. HPB (Oxford)
38. Koniaris LG, Levi DM, Pedroso FE, et al. Is surgical resection superior to transplantation in the treatment of hepatocellular carcinoma? Ann Surg
39. Lee KK, Kim DG, Moon IS, et al. Liver transplantation versus liver resection for the treatment of hepatocellular carcinoma. J Surg Oncol
40. Michel J, Suc B, Montpeyroux F, et al. Liver resection or transplantation for hepatocellular carcinoma? Retrospective analysis of 215 patients with cirrhosis. J Hepatol
41. Otto G, Heuschen U, Hofmann WJ, et al. Survival and recurrence after liver transplantation versus liver resection for hepatocellular carcinoma: a retrospective analysis. Ann Surg
42. Perry JF, Charlton B, Koorey DJ, et al. Outcome of patients with hepatocellular carcinoma referred to a tertiary centre with availability of multiple treatment options including cadaveric liver transplantation. Liver Int
43. Philosophe B, Greig PD, Hemming AW, et al. Surgical management of hepatocellular carcinoma: resection or transplantation? J Gastrointest Surg
44. Pierie JP, Muzikansky A, Tanabe KK, et al. The outcome of surgical resection versus assignment to the liver transplant waiting list for hepatocellular carcinoma. Ann Surg Oncol
45. Shah SA, Cleary SP, Tan JC, et al. An analysis of resection vs transplantation for early hepatocellular carcinoma: defining the optimal therapy at a single institution. Ann Surg Oncol
46. Sotiropoulos GC, Drühe N, Sgourakis G, et al. Liver transplantation, liver resection, and transarterial chemoembolization for hepatocellular carcinoma in cirrhosis: which is the best oncological approach? Dig Dis Sci
47. Tong MJ, Chavalitdhamrong D, Lu DS, et al. Survival in Asian Americans after treatments for hepatocellular carcinoma: a seven-year experience at UCLA. J Clin Gastroenterol
48. Vennarecci G, Ettorre GM, Antonini M, et al. First-line liver resection and salvage liver transplantation are increasing therapeutic strategies for patients with hepatocellular carcinoma and child a cirrhosis. Transplant Proc
49. Yamamoto J, Iwatsuki S, Kosuge T, et al. Should hepatomas be treated with hepatic resection or transplantation? Cancer
50. Zhou J, Wang Z, Qiu SJ, et al. Surgical treatment for early hepatocellular carcinoma: comparison of resection and liver transplantation. J Cancer Res Clin Oncol
51. Washburn K, Edwards E, Harper A, et al. Hepatocellular carcinoma patients are advantaged in the current liver transplant allocation system. Am J Transplant
52. Majno PE, Sarasin FP, Mentha G, et al. Primary liver resection and salvage transplantation or primary liver transplantation in patients with single, small hepatocellular carcinoma and preserved liver function: an outcome-oriented decision analysis. Hepatology
53. Sala M, Fuster J, Llovet JM, et al. High pathological risk of recurrence after surgical resection for hepatocellular carcinoma: an indication for salvage liver transplantation. Liver Transpl
54. Northup PG, Intagliata NM, Shah NL, et al. Excess mortality on the liver transplant waiting list: unintended policy consequences and model for end-stage liver disease (MELD) inflation. Hepatology