TC

BA

t

0d

2

COMPLICATIONS

umor Recurrence After Liver Transplantation for Hepatocellulararcinoma: Recurrence Pathway and Prognostic Factors

. Pérez-Saborido, S. Jiménez de los Galanes, J.C. Menéu-Dı́az, C. Jiménez Romero,

. Moreno Elola-Olaso, Y. Fundora Suárez, V. Barra Valencia, and E. Moreno-González

ABSTRACT

Introduction. Liver transplantation (OLT) has been advocated as a good managementoption for patients with carcinoma hepatocellular (HCC). More recurrences are extrahepaticdue to many pathological factors.Patients and Methods. From April 1986 to December 2003, we performed 95. OLTs forHCC including 73% men of mean age of 54.7 years and 25.3% not filling Mazzaferro’s criteria.Results. The recurrence incidence was 15.8% (n � 15), including only extrahepatic lesionsin 11 (mainly lung recurrence, seven) and hepatic plus extrahepatic in four. Main late mortalitywas due to tumor recurrence (n � 12, 33.3%). No differences were observed among sex,preoperative chemoembolization, age, Child, Okuda, etiology, or satellite nodules. A greaterincidence of tumor recurrence was observed with a preoperative biopsy (45.5% vs 5.9%, P �.0001); and alpha fetoprotein (AFP) � 200 ng/mL (37.5% vs 13.3%, P � .08); known HCC(25.5% vs 3.1%, P � .008); vascular invasion (42.1% vs 10.3%, P � .001); � 5 cm single nodule(50% vs 13%, P � .004); more than three nodules (50% vs 13.9%, P � .01); moderately topoorly differentiated tumors (37.5% vs 12.7%, P � .01); pTNM IV (50% vs 8.7%, P � .0001);and not meeting Milan criteria (40.9% vs 9.2%, P � .001). These are the same factors forextrahepatic recurrence. For hepatic recurrence the prognostic factors were: vascular invasion(15.8% vs 1.5%, P � .008), more than three nodules (25% vs 2.5%, P � .004), moderately topoorly differentiated tumors (18.8% vs 1.4%, P � .003), pTNM IV (16.7% vs 1.4%, P � .006),and not meeting Milan criteria (13.6% vs 1.5%, P � .01).Conclusions. Recurrence incidence with Milan criteria was less than 10%, mainlyextrahepatic (lung). Prognostic factors for tumor recurrence were pathological features,namely vascular invasion, more than three nodules, size larger than 5 cm, moderately to

From the General, Digestive and Abdominal Organs Transplanta-ion Surgical Department, “12 de Octubre” Hospital, Madrid, Spain.

Digestive and Abdominal Organs Transplantation Surgical De-partment, “12 de Octubre” Hospital, Avda de Córdoba Km

Address reprint requests to B. Pérez Saborido, General, 5.400, 28041 Madrid, Spain. E-mail: perezsb@hotmail.com

041-1345/07/$–see front matter © 2007 by Elsevier Inc. All rights reserved.oi:10.1016/j.transproceed.2007.06.059 360 Park Avenue South, New York, NY 10010-1710

304 Transplantation Proceedings, 39, 2304–2307 (2007)

LcOpinciner

ioama

PS

FSp9o

F

MPpph

T

A2dSpdtnpc

S

Wtdwp

TUMOR RECURRENCE AFTER LIVER TRANSPLANTATION 2305

poorly differentiated tumors, pTNM IV stage. The use of preoperative chemoembolizationdid not decrease the recurrence rate. A preoperative biopsy increased the incidence of

extrahepatic recurrence.

uc

RL

A(61sa

tp(

R

TplhrprO

C

Taioaabwenaar(

P

Tvgwt(

IVER TRANSPLANTATION (OLT) has been advo-cated as a good management option for patients with

hronic liver damage and hepatocellular carcinoma (HCC).LT offers the advantage of radical tumor removal even for

atients with multifocal disease or severe cirrhosis, remov-ng the possibility of metachronous lesions and restoringormal liver function.1 Currently OLT is the treatment ofhoice for early HCC, namely, solitary HCC of 5 cm or lessn diameter or no more than three tumors 3 cm or lessodules in diameter, in the absence of vascular invasion orxtrahepatic disease. Actuarial survival is good as areecurrence-free survival and recurrence rates.2,3

Among the prognostic factors, pathological variables aremportant features determining long-term outcome—numberf nodules, tumor size, vascular invasion, satellite nodules,nd differentiation grade.1–7 Our objective was to deter-ine the incidence of recurrence, the recurrence pathway

nd the prognostic factors in our experience.

ATIENTS AND METHODStudy Time and Patients

rom June 1989 to December 2003, with follow-up finished ineptember 2004 (mean follow up of 44.2 � 41.7 months), weerformed 961 OLTs including 9.8% in patients with HCC (n �5). The male/female ratio was 74 (72.9%) to 21 (22.1%) with anverall mean age of 54.7 � 8.7 years (range 15 to 70).

unctional Status and Preoperative Variables

ost patients were Child B (45.1 %, n � 41) or C (36.3%, n � 33).reoperative transarterial chemoembolization was performed in 31atients (33.7%) while on the waiting list. The mean alpha feto-rotein (AFP) value was 309 � 1421 ng/mL with nine patientsaving a value greater than 200 ng/mL (12%).

umor Status

mong the pathological variables, we observed vascular invasion in1.1% (n � 20). About the histological grade, 17.9% of patientsisplayed moderately to poorly differentiated tumors (n � 17).atellite nodules were present in nine patients (9.5%) and 11atients (11.7%) had bilateral involvement; 54.7% of patientsisplayed a tumor size smaller than 3 cm, and only 12 (12.6%) hadumors larger than 5 cm. Eight patients had more than threeodules (8.4%), as did 20 patients (21.1%) when presented withTNM stage IV. Of the patients, 25.3% did not fill Mazzaferro’sriteria.

tatistical Analysis

e performed a retrospective analysis on a prospective database ofransplanted patients, using SPPS v10 for Windows. Quantitativeata are expressed as the mean values and standard deviations, asell as qualitative data, by frequency and rate, which were com-

ared with the chi-squared test. Survival curves were calculated P

sing the Kaplan-Meier method with log-rank tests to compareurves. A statistical level of P � .05 was taken as significant.

ESULTSong-Term Survival

t the end of the study 56 of the 87 patients were alive64.4%) with 1-, 3-, and 5-year actuarial survivals of 85%,8.7%, and 60.2%, respectively, and a mean survival time of22.1 � 13.6 months. The 1-, 3-, and 5-year disease-freeurvivals (DFS) were 81%, 65%, and 57%, respectively withmean DFS time of 114 � 14.2 months.During the follow-up the main cause of mortality was

umor recurrence (35.5%) and medical causes (29%). Sixatients (19.4%) died due to liver disease recurrence and 516.1%) due to new tumors.

ecurrence Incidence and Recurrence Pathway

he recurrence rate was 15.8% (15 patients): only extrahe-atic recurrences occurred in 11 patients (66.7%), mainly

ung recurrences (seven patients, 7.4%) although othersad disease in several organs (n � 3), adrenal (n � 2), bone,etroperitoneal and peritoneal (n � 1 for each) and hepaticlus extrahepatic recurrence in four patients (33.3%). Mostecurrences (75%) developed during the first year afterLT.

linical Prognostic Factors

able 1 shows some of the prognostic factors in univariatend multivariate analyses in relation to recurrence. Nonfluence was observed with gender, Child, or Okuda statusr etiology (viral or not). We can see greater rates of overallnd extrahepatic recurrences among patients with preoper-tive AFP levels greater than 200 ng/mL (37.5% vs 13.3%)ut without statistical significance (P � .08); no differencesere observed in hepatic recurrence. No significant differ-nces were observed in recurrence incidence (extrahepaticor hepatic) between the patients with or without preoper-tive chemoembolization. Finally, patients who underwentpreoperative biopsy displayed a significant greater recur-

ence rate due to extrahepatic disease: 45.5% versus 5.9%P � .001) but no hepatic recurrence.

athological Prognostic Factors of Tumor Recurrence

able 2 shows a univariate analysis of all pathologicalariables. A tumor size larger than 5 cm was related to areater rate of extrahepatic recurrence (50% vs 13%)ithout affecting hepatic recurrence incidence. More than

hree nodules was related to both greater extrahepatic50% vs 13%) and hepatic recurrences (25% vs 2.5%).

atients who fulfilled Mazzaferro’s criteria displayed lower

ivhteers1dt(gtww

D

Ocacslc2o2ti

rap(sec5pwer(TvSbmlsad

G

P

C

E

P

P

t

I

M

T

N

D

S

V

E

p

2306 SABORIDO, DE LOS GALANES, Díaz ET AL

ncidences of extrahepatic and hepatic recurrences: 9.2%ersus 40.9% for extrahepatic and 1.5% versus 13.6% forepatic recurrences. Among the two important factors washe degree of histological differentiation: patient with mod-rately or poorly differentiated tumors showed greaterxtrahepatic (37.5% vs 12.7%) and hepatic (18.8% vs 1.4%)ecurrence: The same observation applied to vascular inva-ion: 42.1% versus 10.3% in extrahepatic and 15.8% versus.5% in hepatic recurrence. No differences were observedue to the presence of satellite nodules or of encapsulatedumors. Finally, patients with advanced tumor stagespTNM IV) displayed worse prognosis with significantlyreater extrahepatic and hepatic recurrences. When theumor was known previous to the transplant, the prognosisas worse: 25.5% versus 3.1% for extrahepatic recurrencesithout differences in hepatic recurrence.

ISCUSSION

LT is the management of choice for HCC and liverirrhosis with good long-term outcomes longer survivalsnd lower recurrence incidences than liver resection espe-ially with the application of Milan criteria: namely, aolitary tumor below 5 cm or no more than three nodules noarger than 3 cm3. With the application of these selectionriteria, the recurrence incidence has decreased to 10% to0%.1–4 In this series, we observed a recurrence incidencef 15.8%, but it was an historical series (from 1986) and5.3% of patients did not fullfill the Milan criteria. Amonghe patients who met the Milan criteria, the recurrence

Table 1. Clinical Factors and Overall Extrahepatic and HepaticRecurrence Incidence

Overall and ExtrahepaticRecurrence Hepatic Recurrence

Factor Rate (%) P* OR Rate (%) P* OR

enderMale 19.1 NS 5.9 NSFemale 10.5 0

reopAFP� 200 ng/mL 13.3 .08 3.9 3.3 NS� 200 ng/mL 37.5 12.5hildA 21.4 NS 0 NSB–C 15.9 5.8

tiologyVirus 20.6 NS 4.8No 8.7 4.3 NS

reoperative FNA-bYes 45.5 .0001 13.3 4.5 NSNo 5.9 3.9

reoperative TACEYes 17 NS 3.2 NSNo 19.4 5.7

Abbreviations: OR, odds ratio; FNA-b, fine-needle aspiration biopsy; TACE,ransarterial chemoembolization; NS, not significant; AFP, alpha fetoprotein.

*Chi-square test.

ncidence was less than 10%.P

However, not only tumor size and number of nodules areelated to recurrence; other clinical and pathological vari-bles are important prognostic factors,1–6 especially theresence of vascular invasion and the differentiation grademoderately to poorly differentiated tumors).7–11 In oureries, more than three nodules was clearly related toxtrahepatic (50%) and hepatic recurrences (25%). Inontrast, tumor size only related to extrahepatic recurrence:0% among subjects with tumors larger than 5 cm. Asreviously published, two important factors in our seriesere, vascular invasion (42.1% survival among subjects withxtrahepatic and 15.8% among those with hepatic recur-ences) and moderately to poorly differentiated tumors37.5% of extrahepatic and 18.8% of hepatic recurrence).he problem was the preoperative diagnosis of the presence ofascular invasion or the histological differentiation grade.ome authors have suggested the utility of a preoperative liveriopsy to obtain the diagnosis,7 but the risk is high, not only ofajor complications, but also of abdominal implantation and

ong-term tumor recurrence, as we have observed.12 Ourtudy, showed an extrahepatic recurrence rate of 45.5%mong patients with a preoperative biopsy while there were noifferences in hepatic recurrence. A relation has been previ-

Table 2. Pathological Factors and Overall Extrahepatic andHepatic Recurrence Incidence

Overall and ExtrahepaticRecurrence Hepatic Recurrence

Factor Rate (%) P* OR Rate (%) P* OR

ncidentalYes 3.1 .008 0.09 0 NSNo 25.5 7.3azzaferroYes 9.2 .001 6.8 1.5 .01 10.1No 40.9 13.6

umor size�5 cm 13 .04 6.7 3.9 NS�5 cm 50 10odules number�3 13 .01 6.1 2.5 .004 12.8�3 50 25

ifferentiationWD 12.7 .018 4.1 1.4 .003 16.1MD-PD 37.5 18.8

attelitosisNo 15.2 NS 3.8 NSYes 37.5 12.5

ascular invasionNo 10.3 .001 6.3 1.5 .008 12.5Yes 42.1 15.8

ncapsulatedNo 20 NS 7.5 NSYes 14.9 2.1

TNMI–III 8.7 .0001 10.5 1.4 .006 13.6IV 50 16.7

Abbreviations: WD, well differentiated; MD, moderately differentiated;

D, poorly differentiated; OR, odds ratio.*Chi-square test.

oMn

alwh

llpop

tdtrtraf

R

t

t

fw

hP

co

ct

sp3

dhc

fl

ct

gp

prc

lc

zt

ht5

r

TUMOR RECURRENCE AFTER LIVER TRANSPLANTATION 2307

usly observed between tumor size and vascular invasion orD-PD tumors1,8,10; therefore, we do not believe that it is

ecessary to perform a liver biopsy.In a Spanish multicenter study, an elevated AFP level was

prognostic factor for long-term outcome.13 We observed aarger incidence of extrahepatic recurrence among patientsith an AFP level above 200 ng/mL (37.5% vs 13.3%),owever, no relation was observed to hepatic recurrence.Some authors have observed a benefit in survival and a

ower recurrence rate employing preoperative chemoembo-ization.14,15 In our experience as has been previouslyublished,16 we have not observed a significant decrease inverall, extrahepatic, or hepatic recurrence rates amongatients with preoperative chemoembolization.In conclusion, a tumor size larger than 5 cm, more than

hree nodules, vascular invasion, moderately to poorlyifferentiated tumors, and advanced TNM stage are impor-ant prognostic factors for tumor recurrence. Mainly, theecurrences are extrahepatic occurring before 1 year afterhe liver transplantation. Preoperative liver biopsies wereelated to larger extrahepatic recurrences, so they must bevoided. Furthermore, preoperative chemoembolizationailed to confer a benefit.

EFERENCES

1. Hemming AW, Cattral MS, Reed AI, et al: Liver transplan-ation for hepatocellular carcinoma. Ann Surg 233:652, 2001

2. Molementi EP, Klintmalm GB: Hepatocellular cancer in liverransplantation. J Hepatobiliary Pancreat Surg 8:427, 2001

3. Mazzaferro V, Regalia E, Doci R, et al: Liver transplantationor the treatment of small hepatocellular carcinoma in patientsith cirrosis. N Engl J Med 334:693, 19964. Pérez B, Loinaz C, Gimeno A, et al: Liver transplantation for

epatocellular carcinoma: our experience from 1986. Transplant

roce 35:1285, 2003 h

5. Klintmalm GB: Liver transplantation for hepatocellular car-inoma: a registry repor of the impact of tumor characteristics onut come. Ann Surg 228:479, 19986. Molmenti EP, Klintmalm GB: Liver transplantation in asso-

iation with hepatocellular carcinoma: an update of the Interna-ional Tumor Registry. Liver Transpl 8:736, 2002

7. Jonas S, Bechstein WO, Steinmullet T, et al: Vascular inva-ión and hitopahologic gradin determine outcome alter liver trans-lantation for hepatocellular carcinoma in cirrhosis. Hepatology3:1080, 20018. Ariizumi S, Takasaki K, Yamamoto M, et al: Histopathologic

ifferentiation of the main nodule determines outcome afterepatic resection for synchronous multicentric hepatocellular car-inomas. Hepatogastroenterology 51:500, 2004

9. Cillo U, Vitale A, Bassanello M, et al: Liver transplantationor the treatment of moderately or well-differentiated hepatocellu-ar carcinoma. Ann Surg 239:150, 2004

10. Ker CG, Chen HY, Chen KS, et al: Clinical significance ofell differentiation in hepatocellular carcinoma. Hepatogastroen-erology 50:475, 2003

11. Tamura S, Kato T, Berho M, et al: Impact of histologicalrade of hepatocellular carcinoma on the outcome of liver trans-lantation. Arch Surg 136:25, 2001 (discussion 31)12. Saborido BP, Diaz JC, de los Galanes SJ, et al: Does

reoperative fine needle aspiration-biopsy produce tumor recur-ence in patients following liver transplantation for hepatocellulararcinoma? Transplant Proc 37:3874, 2005

13. Figueras J, Ibáñez L, Ramos E, et al: Selection criteria foriver transplantation in early-stage hepatocellular carcinoma andirrosis: results of a multicenter study. Liver Transpl 7:877, 2001

14. Spreafico C, Marchiano A, Regalia E, et al: Chemoemboli-ation of hepatocellular carcinoma in patients who undergo liverransplantation. Radiology 192:687, 1994

15. Harnois DM, Steers J, Andrews JC, et al: Preoperativeepatic artery chemoembolization followed by orthotopic liverransplantation for hepatocellular carcinoma. Liver Transpl Surg:192, 199916. Pérez Saborido B, Menéu JC, Moreno E, et al: Is transarte-

ial chemoembolization necessry before liver transplantation for

epatocellular carcinoma. Am J Surg 190:383, 2005