HEPATOBILIARY MALIGNANCIES
Ischemia Time Impacts Recurrence of Hepatocellular Carcinoma After Liver Transplantation Shunji Nagai,1,2 Atsushi Yoshida,2 Marcelo Facciuto,1 Dilip Moonka,3 Marwan S. Abouljoud,2 Myron E. Schwartz,1 and Sander S. Florman1 Although experimental evidence has indicated that ischemia-reperfusion (I/R) injury of the liver stimulates growth of micrometastases and adhesion of tumor cells, the clinical impact of I/R injury on recurrence of hepatocellular carcinoma (HCC) after liver transplantation (LT) has not been fully investigated. To study this issue, we conducted a retrospective review of the medical records of 391 patients from two transplant centers who underwent LT for HCC. Ischemia times along with other tumor/recipient variables were analyzed as risk factors for recurrence of HCC. Subgroup analysis focused on patients with HCC who had pathologically proven vascular invasion (VI) because of the associated increased risk of micrometastasis. Recurrence occurred in 60 patients (15.3%) with median time to recurrence of 0.9 years (range, 40 days-4.6 years). Cumulative recurrence curves according to cold ischemia time (CIT) at 2-hour intervals and warm ischemia time (WIT) at 10-minute intervals showed that CIT >10 hours and WIT >50 minutes were associated with significantly increased recurrence (P 5 0.015 and 0.036, respectively). Multivariate Cox’s regression analysis identified prolonged cold (>10 hours; P 5 0.03; hazard ratio [HR] 5 1.9) and warm (>50 minutes; P 5 0.003; HR 5 2.84) ischemia times as independent risk factors for HCC recurrence, along with tumor factors, including poor differentiation, micro- and macrovacular invasion, exceeding Milan criteria, and alphafetoprotein >200 ng/mL. Prolonged CIT (P 5 0.04; HR 5 2.24) and WIT (P 5 0.001; HR 5 5.1) were also significantly associated with early (within 1 year) recurrence. In the subgroup analysis, prolonged CIT (P 5 0.01; HR 5 2.6) and WIT (P 5 0.01; HR 5 3.23) were independent risk factors for recurrence in patients with VI, whereas there was no association between ischemia times and HCC recurrence in patients with no VI. Conclusion: Reducing ischemia time may be a useful strategy to decrease HCC recurrence after LT, especially in those with other risk factors. (HEPATOLOGY 2015;61:895-904)
O
ver the past few decades, recurrence of hepatocellular carcinoma (HCC) after liver transplantation (LT) and its associated risk factors has been extensively investigated.1-3 Tumor size and number are generally accepted as predictive of recurrence, and many selection criteria for transplantation have used these characteristics, including the Milan, University of California San Francisco, and Up-to-7
criteria.4-6 Vascular invasion (VI) and tumor differentiation have also been shown to be significant predictors of recurrence.7-10 Several experimental models have shown that ischemia-reperfusion (I/R) injury of the liver promotes cancer cell implantation and growth.11-13 We hypothesize that growth of HCC micrometastases may be accelerated in LT recipients who receive liver grafts
Abbreviations: AFP, alpha-fetoprotein; ALT, alanine aminotransaminase; AST, aspartate aminotransaminase; CIT, cold ischemia time; DCD, donation after cardiac death; HCC, hepatocellular carcinoma; HR, hazard ratio; HTK, histidine-tryptophan-ketoglutarate; IQR, interquartile range; I/R, ischemia-reperfusion; LT, liver transplantation; MELD, Model for End-Stage Liver Disease; PRBCs, packed red blood cells; OR, odds ratio; UW, University of Wisconsin; VI, vascular invasion; WIT, warm ischemia time. From the 1Recanati/Miller Transplantation Institute, Icahn School of Medicine at Mount Sinai, New York, NY; 2Division of Transplant and Hepatobiliary Surgery, Henry Ford Transplant Institute, Henry Ford Hospital, Detroit, MI; and 3Division of Gastroenterology, Henry Ford Hospital, Detroit, MI. Received January 2, 2014; accepted August 1, 2014. Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.27358/suppinfo. Preliminary results were presented in 2011 at the International Liver Transplant Society, Valencia, Spain, and in 2012 at the American Transplant Congress, Boston, MA. 895
896
NAGAI ET AL.
exposed to prolonged ischemia times, leading to early recurrence of HCC after LT. Our study examines the relationship between ischemia time of the liver graft and the incidence and timing of HCC recurrence after LT based on experience from two large transplant centers in the United States: Mount Sinai Medical Center in New York City and Henry Ford Hospital in Detroit. We then performed validation analysis using a separate data set from Indiana University Hospital in Indianapolis.
Patients and Methods Study Population. Medical records of 271 patients at Mount Sinai Medical Center and 178 patients at Henry Ford Hospital who underwent LT for HCC were retrospectively reviewed. Patients with pathologically proven HCC in their explants were eligible for this study; mixed cholangiocarcinoma/HCC was excluded. Patients who died within 30 days of LT were excluded, as were patients who received living donor organs or organs procured after cardiac death. This study was approved by the institutional review boards at both centers. LT Procedure. Several techniques were used for the LT surgery, specifically in regard to the hepatic outflow reconstruction. Bicaval and piggyback techniques (standard piggyback and side-to-side cavacavostomy) were used based upon surgical preference. After the reconstruction of hepatic veins and portal vein, the transplanted liver was reperfused, followed by the hepatic artery reconstruction. Donor livers were flushed with CUSTODIOL histidine-tryptophanketoglutarate (HTK) solution (Essential Pharmaceuticals, LLC, Newtown, PA) or University of Wisconsin (UW) solution (Bridge to Life Solutions LLC, Columbia, SC). Donor livers were generally flushed with 46 L of solution in situ, and portal flush was added as needed on the back table. Cold ischemia time (CIT) was defined as the time from donor cross-clamping to the removal of the liver from the cold preservative solution preceding implantation. Warm ischemia time (WIT) was defined as the time from when the liver is taken out of the cold preservative solution in preparation for implantation until reperfusion.
HEPATOLOGY, March 2015
Post-Transplant Management. Maintenance immunosuppresion was with tacrolimus, mycophenolate mofetil, and steroids. Sirolimus was often added when recurrence of HCC was diagnosed, but was not routinely used in an adjuvant fashion. Abdomen and chest computed tomography or magnetic resonance imaging and alpha-fetoprotein (AFP) levels were monitored until the fifth year after LT. Biopsy was used selectively to establish the diagnosis of recurrent HCC; pathological confirmation was not required if the lesions progressed and were clinically consistent with HCC. Recurrence Analysis. HCC recurrence analysis was performed on the eligible patients. Variables regarding patient characteristics included age, sex, underlying liver disease, previous history of surgery or tumordirected therapies, and Model for End-Stage Liver Disease (MELD) score. Tumor factors included pretransplant AFP level, size and number of tumors, presence and degree of VI, and tumor grade. Operative factors included surgical techniques, operative time, transfusion of packed red blood cells (PRBCs), CIT, and WIT. Subgroup analysis was performed focused on patients with VI because of the associated increased risk of micrometastasis. In order to clarify degree of I/ R injury, post-transplant peak levels of aspartate transaminase (AST) and alanine transaminase (ALT) within the first week after LT were compared according to CIT and WIT. Validation of the Results of the Present Study. In order to validate the results of this study, data collected from another transplant center, Indiana University Hospital, was used to assess the association between ischemia times and HCC recurrence. The prediction model for HCC recurrence was applied to their data set with the same inclusion and exclusion criteria. The same statistical analysis methods were applied. Medical records of 305 patients who underwent LT for HCC were independently analyzed by the surgeon (R.M.) in Indiana University Hospital. Retrospective analysis of the transplant database has been approved by the institutional review board at the Indiana University School of Medicine. Statistical Analysis. Continuous variables were analyzed using the Student t test; for discrete variables, chi-square analysis was performed. Data are shown as
Address reprint requests to: Sander S. Florman, M.D., F.A.C.S., The Mount Sinai Medical Center, One Gustave L. Levy Place, Box 1104, New York, NY 10029-6574. E-mail: [email protected]; fax: 1-212-348-2474. C 2014 by the American Association for the Study of Liver Diseases. Copyright V View this article online at wileyonlinelibrary.com. DOI 10.1002/hep.27358 Potential conflict of interest: Nothing to report.
HEPATOLOGY, Vol. 61, No. 3, 2015
NAGAI ET AL.
897
Fig. 1. Comparison of post-transplant liver enzyme levels according to ischemia times (median and IQR). (A) There was a significant trend for post-transplant AST peak levels among CIT groups (P 5 0.02). Patients with prolonged CIT (10 hours was significantly associated with higher HCC recurrence rate (P 5 0.015 for CIT >10 hours vs.
50 minutes. WIT >50 minutes were significantly associated with higher HCC recurrence rate (P 5 0.036 for WIT >50 minutes vs. 50 minutes; Gray’s test; Fig. 2B). Based on these results, the thresholds for CIT and WIT were set at 10 hours and 50 minutes, respectively. On univariate analysis, tumor size and number, pathologically exceeding Milan criteria, poor differentiation, and micro- and macrovascular invasion were associated with HCC recurrence (Table 1). CIT and WIT were associated with recurrence as dichotomized variables (P 5 0.005 and 0.046, respectively); CIT was also associated with recurrence as a continuous variable (P 5 0.01; hazard ratio [HR] 51.12 per hour). The amount of intraoperative PRBC transfusion was associated with HCC recurrence in a continuous manner (P 5 0.001; HR 5 1.03 per unit). HCC recurrence was not associated with the type of preservation solution (UW or HTK solution; P 5 0.98). Significant factors on univariate analysis and clinically relevant factors were included in multivariate analyses. As the first multivariate analysis, CIT and WIT were dichotomized at 10 hours and 50 minutes,
HEPATOLOGY, Vol. 61, No. 3, 2015
NAGAI ET AL.
899
Table 1. Risk Factor for Recurrence of HCC Univariate Variables
Recipient age (per year) Female recipient Hepatitis C MELD score >15 AFP >200 ng/mL Exceeding Milan criteria Tumor size (per cm) Tumor number Poor differentiation VI Microvascular Macrovascular Donor age (per year) Previous hepatic resection Previous RFA Previous TACE Operation time (per hour) IVC replacement Preservation solution UW (ref.) HTK PRBCs (per unit) 0 unit (ref.) 1-10 units >10 units CIT (per hour) >10 hours WIT (per min) >50 minutes
Multivariate
P Value*
HR (95% CI)
0.99 0.14 0.19 0.61 0.25
1.0 (0.97-1.03) 0.55 (0.25-1.21) 0.71 (0.43-1.18) 0.99 (0.96-1.03) 0.73 (0.42-1.26)
P Value*
HR (95% CI)
1.89 (1.00-3.57) 2.06 (1.14-3.71)
10 hours (8.5 vs. 4.6 units; P < 0.001) were
900
NAGAI ET AL.
HEPATOLOGY, March 2015
associated with larger amount of transfusion; there was no association with WIT. On multivariate analysis, MELD score (P 5 0.003; odds ratio [OR] 5 1.05) and CIT >10 hours (P < 0.001; OR 5 3.95) remained as independent predictors. Subgroup Analysis of Ischemia Time in Patients With/Without VI. Subgroup analysis was performed according to status of pathologically proven VI. Interaction between VI, and ischemia times were identified to be significant (P < 0.001 for CIT >10 hours and P < 0.001 for WIT >50 minutes), which validated
this subgroup analysis. Pathologically proven VI was identified in 122 patients, 35 (28.7%) of whom developed recurrence. On multivariate analysis, CIT >10 hours (P 5 0.01; HR 5 2.6) and WIT >50 minutes (P 5 0.01; HR 5 3.23) were independent predictors of recurrence. In a similar analysis among patients without VI, neither CIT (P 5 0.99) nor WIT (P 5 0.21) was associated with HCC recurrence (Table 3). Based on a discriminant analysis, the following formula was established to predict risk of recurrence in patients with VI:
D524:112:21ðAFP > 200 ng=mLÞ13:75ðexceeding Milan criteriaÞ12:94 ðpoor differentiationÞ 12:94 ðCIT > 10 hours Þ12:87 ðWIT > 50 minÞ: ðScore 0 for not meeting criterion; score 1 for meeting criterion; D > 0 predicts recurrence:Þ
Table 2. Risk Factor for Early Recurrence of HCC (Within 1 Year) Multivariate Variables
P Value*
HR
95% CI
AFP >200 ng/mL Exceeding Milan criteria† Poor differentiation VI Microvascular Macrovascular Previous TACE PRBCs (per unit) CIT >10 hours WIT >50 minutes
0.03 0.08 0) and standard (D < 0) risk groups was significant (P < 0.001) and the rates were similar between the standard risk group and negative VI group (P 5 0.77). One- and three-year HCC recurrence rates in the high-risk, standard-risk, and negative VI group were 47.9% and 59.8%, 3.2% and 10.5%, and 3.0% and 8.5%, respectively (Fig. 3). Recurrence was accurately predicted in 85 of 111 (76.6%) patients with VI. Validation Analysis With Indiana University Hospital’s Data Set. Of 305 patients who underwent LT for HCC at Indiana University Hospital, 268 met
Table 3. Risk Factor for Recurrence of HCC in Patients With HCC Accompanied With VI VI Group (n 5 122) Variables
AFP >200 ng/mL Exceeding Milan criteria† Poor differentiation Previous TACE PRBCs CIT >10 hours WIT >50 minutes
P Value*
0.08 0.02 0.005 0.28 0.12 0.01 0.01
Non-VI Group (n 5 269)
HR
95% CI
2.08 3.13 3.1 1.54 1.02 2.6 3.23
0.92-4.7 1.25-7.88 1.4-6.88 0.7-3.4 0.99-1.05 1.23-5.49 1.24-8.38
*Cox’s proportional regression analysis. Exceeding Milan Criteria was included in the multivariate analysis, instead of tumor size and number. Abbreviations: TACE, transarterial chemoembolization; CI, confidence interval. †
P Value*
0.89 0.8 0.006 0.18 0.44 0.99 0.21
HR
95% CI
1.08 1.13 4.16 1.77 0.96 1.01 1.92
0.34-3.44 0.45-2.85 1.49-11.58 0.76-4.12 0.87-1.06 0.33-3.13 0.69-5.35
HEPATOLOGY, Vol. 61, No. 3, 2015
NAGAI ET AL.
901
(P 5 0.02; HR 5 2.17), but not between CIT >10 hours and recurrence (P 5 0.37; HR 5 1.48). Our prediction model was applied to this population, and the risk of HCC recurrence was clearly stratified among the groups. Adjusting death before HCC recurrence as a competing risk event, the differences of HCC recurrence rate between the high- and standard-risk groups, as well as between the standardrisk and negative VI groups, were significant (P 5 0.025 and 0.012, respectively; Gray’s test). Oneand three-year recurrence rates in the high-risk, standard-risk, and negative VI groups were 27.8% and 53.7%, 3.6% and 19.3%, 3.8% and 9.3%, respectively (Fig. 4).
Discussion
Fig. 3. Comparison of cumulative recurrence rate of HCC according to the prediction model. The classification was decided based on the following formula.
D524:112:21 ðAFP > 200 ng=mLÞ 13:75 ðexceeding MilancriteriaÞ12:94 ðpoor differentiationÞ 12:94 ðCIT > 10 hours Þ12:87 ðWIT > 50 minÞ: ðScore of 0 for not meeting criteria; score of 1 for meeting criteriaÞ If a patient has no or one risk factor, they are standard risk (D < 0), whereas if they have more than one risk factor, they are high risk (D > 0). Cumulative incidence curves of HCC recurrence and death before HCC recurrence. With adjusting death before HCC recurrence as a competing risk event, the difference of HCC recurrence between the high- and standard-risk groups was significant (P < 0.001) and the rates were similar between the standard-risk group and negative VI group (P 5 0.77).
inclusion criteria. HCC recurrence developed in 51 (19%). On univariate analysis, WIT >50 minutes was significantly associated with higher HCC recurrence rate (P 5 0.045; HR 5 1.94). CIT >10 hours showed tendency for increased risk of recurrence, although not statistically significant (P 5 0.19; HR 5 1.76). Neither WIT nor CIT was associated with HCC recurrence in a continuous manner (P 5 0.34 and 0.52). After adjusting the risk with positive VI, tumor size, and number (exceeding Milan criteria), there was significant association between WIT >50 minutes and HCC recurrence
Currently, prioritization of patients with HCC for LT is based on tumor characteristics alone.4,6,10,16-20 However, other factors may significantly affect HCC recurrence and survival. This study demonstrates that prolonged cold and/or warm ischemia of the liver graft are independent predictors of both overall and early (within 1 year) HCC recurrence, supporting our hypothesis that I/R injury of the liver graft accelerates growth and implantation of micrometastases of HCC present at the time of LT. A study of patients with HCC in the Scientific Registry of Transplant Recipients database demonstrated decreased overall survival among recipients who received organs procured after cardiac death versus those who received conventionally procured organs, suggesting an impact of greater I/R injury (interpretation of this study is, however, limited by the lack of data on HCC recurrence).21 Though the 13 patients in our series who received liver graft procured after cardiac death were excluded from the study, 5 of 13 developed recurrence (P 5 0.04; HR 5 1.61; univariate Cox’s regression analysis, data not shown in the results). Because HCC patients are disadvantaged by organ allocation based on MELD score in some regions, livers from donation after cardiac death (DCD) donors might be used for advanced HCC patients, especially those who do not meet criteria for MELD exceptional points. However, considering potential influence of donor WIT on HCC recurrence, the indication of DCD LT for HCC patients needs to be cautiously applied. Experimental evidence has suggested a number of biological mechanisms for the association between cancer outcome and I/R injury to the liver.11,12 I/R induces mechanical injury to hepatic sinusoids, which leads
902
NAGAI ET AL.
Fig. 4. Comparison of recurrence rate of HCC according to the prediction model by using the data set from the other transplant center (Indiana University Hospital) as the validation of the prediction model. Cumulative incidence curves of HCC recurrence and death before HCC recurrence. With adjusting death before HCC recurrence as a competing factor, the difference of HCC recurrence between the high- and standard-risk groups was significant (P 5 0.025) and between the standard-risk group and negative VI group (P 5 0.012).
to hepatic microcirculatory barrier dysfunction and activates cell signals related to invasion and migration.11,22 Hypoxia induces genes and cytokines involved in angiogenesis, cellular proliferation, growth, and adhesion: for example, in the absence of oxygen, hypoxia-inducible transcription factor 1 binds to hypoxia-response elements, activating the expression of hypoxia-response genes, such as vascular endothelial growth factor.23-26 Most of the recurrences among our patients were in extrahepatic organs, suggesting that the oncological effects of prolonged ischemia time may be systemic. Man et al. reported that infiltrative growth pattern of tumor cells accompanied with venous invasion and cell signal related to tumor cell invasion were observed not only in liver tumor cells, but also in the lung metastasis nodules in rats undergoing I/R injury and major hepatectomy, compared to the control group.11 Activation of cancer cell migration and invasion pathways by I/R injury is one reported mechanism by which accelerated growth and implantation of extrahepatic micrometastases might occur.13 Significantly high AST and
HEPATOLOGY, March 2015
ALT levels in the prolonged ischemia time groups provide evidence of greater I/R injury in these patients. Nevertheless, the biological mechanisms accounting for acceleration of extrahepatic recurrence has not been clearly demonstrated yet, and further investigations are necessary. Similar to previously reported studies,27 we observed a significant association of PRBC transfusion with HCC outcome after LT on univariate analysis. Mechanisms underlying the oncological impact of blood loss and/or transfusion have been proposed for various tumors, including HCC, colon cancer, gastric cancer, and pancreatic cancer.28-31 Blood transfusion can impair immunity and enhance inflammation.32,33 Profuse hemorrhage during surgery may lead to more tumor manipulation that could result in dissemination of cancer cells.34 Systemic hypoperfusion resulting from massive hemorrhage impairs oxygen delivery to vital organs, leading to systemic inflammation and production of cytokines that may reduce antitumor immunity.35 In our study, large volume transfusion was associated with overall HCC recurrence, but did not remain as an independent risk factor on multivariate analysis. We observed a significant correlation between ischemia times and transfusion requirements. Poor early graft function associated with prolonged ischemia can lead to intraoperative coagulopathy and increased blood loss.36 This correlation could at least partly underlie the connection between ischemia time and HCC recurrence. Because the entire liver is removed at LT, post-LT HCC recurrence must be the result of undetectable micrometastases present at the time of LT; if ischemia accelerates tumor growth, its impact should be greater in patients more likely to harbor micrometastases (e.g., vascular invasion). Subgroup analysis in patients with vascular invasion showed that long ischemia time, both CIT and WIT, are independent predictors of HCC recurrence in addition to exceeding Milan criteria, poor tumor differentiation, and elevated AFP level. In the presence of any two of these risk factors, the 1year recurrence rate is 47.9%. Though not of use in case selection given that the required data are only available after LT, our formula enables risk stratification that may be useful in establishing the post-LT monitoring schedule, choosing the immunosuppression protocol, and selecting patients for clinical trials of adjuvant strategies.37 In order to validate our findings and determine the efficacy of our recurrence formula for high-risk HCC patients, we reviewed and applied our analysis to
HEPATOLOGY, Vol. 61, No. 3, 2015
another center’s database (Indiana University). The analysis for the validation of our findings and prediction model confirmed increased risk of HCC recurrence associated with prolonged ischemia times. WIT showed more-significant association, which was consistent with our findings. Neither WIT nor CIT correlated in a continuous manner in this data set, whereas WIT showed significant association in a continuous manner, but not CIT, in our series. The negative influence probably remarkably increased at some thresholds and did not seem to have linear correlation with time. The cut-off time for increased risk of HCC recurrence remains to be elucidated, because of the limited sample size. In terms of the risk stratification, we could emphasize the validity of our prediction model and the importance of modifications of post-transplant management for these populations at high risk, because the patients from the different data set categorized in the high-risk group also showed much earlier and higher rates of HCC recurrence, as our population demonstrated. Thus, these findings further supported our conclusion. In order to clarify the clinical implication of ischemia time on HCC recurrence in LT patients, a multicenter, clinical study should be conducted. In addition, further experimental models are necessary to unravel biological mechanisms accounting for potential oncological influence of I/R injury. Pending the results of those studies, it might be reasonable to make all efforts to reduce ischemia time before transplant in patients with HCC. Ischemia times could be shortened by improved selection of surgical procedures and systematic changes in the organ allocation process. Transporting the liver graft and timing of recipient surgery could be further optimized by efficient communication among transplant teams. Thus, we suggest that transplant surgeons, hepatologists, and coordinators should recognize the possible influence of ischemia time on HCC recurrence after LT. In conclusion, we have demonstrated that prolongation of ischemia time, both warm and cold, predicts early HCC recurrence after LT. Our results suggest that interventions to shorten ischemia times may improve outcomes for patients with HCC who undergo LT, especially those with other risk factors for recurrence. Acknowledgment: The authors thank Drs. Richard S. Mangus, M.D., M.S., and A. Joseph Tector, M.D., Ph.D., Transplant Division, Indiana University Hospital, for their participation in the validation analysis. The authors also thank Drs. Shozo Mori, M.D., Ph.D., Mizuki Ninomiya, M.D., Ph.D., and Lloyd Brown, M.D.,
NAGAI ET AL.
903
for data gathering and analysis of Mount Sinai Medical Center and Henry Ford Hospital.
References 1. Prasad KR, Young RS, Burra P, Zheng SS, Mazzaferro V, Moon DB, Freeman RB. Summary of candidate selection and expanded criteria for liver transplantation for hepatocellular carcinoma: a review and consensus statement. Liver Transpl 2011;17(Suppl 2):S81-S89. 2. Nagai S, Abouljoud MS, Kazimi M, Brown KA, Moonka D, Yoshida A. Peritransplant lymphopenia is a novel prognostic factor in recurrence of hepatocellular carcinoma after liver transplantation. Transplantation 2014;97:694-701. 3. Halazun KJ, Hardy MA, Rana AA, Woodland DC, Luyten EJ, Mahadev S, et al. Negative impact of neutrophil-lymphocyte ratio on outcome after liver transplantation for hepatocellular carcinoma. Ann Surg 2009;250:141-151. 4. D’Amico F, Schwartz M, Vitale A, Tabrizian P, Roayaie S, Thung S, et al. Predicting recurrence after liver transplantation in patients with hepatocellular carcinoma exceeding the up-to-seven criteria. Liver Transpl 2009;15:1278-1287. 5. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693-699. 6. Yao FY, Ferrell L, Bass NM, Watson JJ, Bacchetti P, Venook A, et al. Liver transplantation for hepatocellular carcinoma: expansion of the tumor size limits does not adversely impact survival. HEPATOLOGY 2001;33:1394-1403. 7. DuBay D, Sandroussi C, Sandhu L, Cleary S, Guba M, Cattral MS, et al. Liver transplantation for advanced hepatocellular carcinoma using poor tumor differentiation on biopsy as an exclusion criterion. Ann Surg 2011;253:166-172. 8. Gouw AS, Balabaud C, Kusano H, Todo S, Ichida T, Kojiro M. Markers for microvascular invasion in hepatocellular carcinoma: where do we stand? Liver Transpl 2011;17(Suppl 2):S72-S80. 9. Lim KC, Chow PK, Allen JC, Chia GS, Lim M, Cheow PC, et al. Microvascular invasion is a better predictor of tumor recurrence and overall survival following surgical resection for hepatocellular carcinoma compared to the Milan criteria. Ann Surg 2011; 254:108-113. 10. Mazzaferro V, Llovet JM, Miceli R, Bhoori S, Schiavo M, Mariani L, et al. Predicting survival after liver transplantation in patients with hepatocellular carcinoma beyond the Milan criteria: a retrospective, exploratory analysis. Lancet Oncol 2009;10:35-43. 11. Man K, Ng KT, Lo CM, Ho JW, Sun BS, Sun CK, et al. Ischemiareperfusion of small liver remnant promotes liver tumor growth and metastases—activation of cell invasion and migration pathways. Liver Transpl 2007;13:1669-1677. 12. van der Bilt JD, Kranenburg O, Nijkamp MW, Smakman N, Veenendaal LM, Te Velde EA, et al. Ischemia/reperfusion accelerates the outgrowth of hepatic micrometastases in a highly standardized murine model. HEPATOLOGY 2005;42:165-175. 13. Ku Y, Kusunoki N, Shiotani M, Maeda I, Iwasaki T, Tominaga M, et al. Stimulation of haematogenous liver metastases by ischaemiareperfusion in rats. Eur J Surg 1999;165:801-807. 14. Gray R. A class of K-sanple tests for comparing the cumulative incidence of a competing risk. Ann Stat 1988;16:1141-1154. 15. Bruix J, Sherman M; Practice Guidelines Committee, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma. HEPATOLOGY 2005;42:1208-1236. 16. Kaido T, Takada Y, Uemoto S. Usefulness of the Kyoto criteria as selection criteria for living donor liver transplantation for hepatocellular carcinoma. Liver Transpl 2010;16:538-540. 17. Piardi T, Gheza F, Ellero B, Woehl-Jaegle ML, Ntourakis D, Cantu M, et al. Number and tumor size are not sufficient criteria to select
904
18.
19.
20.
21.
22.
23.
24.
25.
26. 27.
28.
29.
NAGAI ET AL.
patients for liver transplantation for hepatocellular carcinoma. Ann Surg Oncol 2012;19:2020-2026. Llovet JM, Fuster J, Bruix J; Barcelona Clinic Liver Cancer Group. The Barcelona approach: diagnosis, staging, and treatment of hepatocellular carcinoma. Liver Transpl 2004;10(2 Suppl 1):S115-S120. Kaido T, Takada Y, Uemoto S. Selection criteria for patients with hepatocellular carcinoma in liver transplantation. Transplantation 2009;88: 442-443; author reply, 443. Makuuchi M, Kokudo N, Arii S, Futagawa S, Kaneko S, Kawasaki S, et al. Development of evidence-based clinical guidelines for the diagnosis and treatment of hepatocellular carcinoma in Japan. Hepatol Res 2008;38:37-51. Croome KP, Wall W, Chandok N, Beck G, Marotta P, HernandezAlejandro R. Inferior survival in liver transplant recipients with hepatocellular carcinoma receiving donation after cardiac death liver allografts. Liver Transpl 2013;19:1214-1223. Itoh K, Yoshioka K, Akedo H, Uehata M, Ishizaki T, Narumiya S. An essential part for Rho-associated kinase in the transcellular invasion of tumor cells. Nat Med 1999;5:221-225. Axelson H, Fredlund E, Ovenberger M, Landberg G, Pa˚hlman S. Hypoxia-induced dedifferentiation of tumor cells—a mechanism behind heterogeneity and aggressiveness of solid tumors. Semin Cell Dev Biol 2005;16:554-563. Brahimi-Horn C, Pouyssegur J. The role of the hypoxia-inducible factor in tumor metabolism growth and invasion. Bull Cancer 2006;93: E73-E80. Carmeliet P, Dor Y, Herbert JM, Fukumura D, Brusselmans K, Dewerchin M, et al. Role of HIF-1alpha in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature 1998;394:485-490. Harris AL. Hypoxia—a key regulatory factor in tumour growth. Nat Rev Cancer 2002;2:38-47. Kaido T, Takada Y, Egawa H, Uemoto S. The influence of intraoperative homologous blood transfusion on prognosis after liver transplantation for hepatocellular carcinoma. Hepatogastroenterology 2009;56:808-812. Nagai S, Fujii T, Kodera Y, Kanda M, Sahin TT, Kanzaki A, et al. Impact of operative blood loss on survival in invasive ductal adenocarcinoma of the pancreas. Pancreas 2011;40:3-9. Katz SC, Shia J, Liau KH, Gonen M, Ruo L, Jarnagin WR, et al. Operative blood loss independently predicts recurrence and survival after resection of hepatocellular carcinoma. Ann Surg 2009;249:617-623.
HEPATOLOGY, March 2015
30. Ojima T, Iwahashi M, Nakamori M, Nakamura M, Naka T, Katsuda M, et al. Association of allogeneic blood transfusions and long-term survival of patients with gastric cancer after curative gastrectomy. J Gastrointest Surg 2009;13:1821-1830. 31. Amato A, Pescatori M. Perioperative blood transfusions for the recurrence of colorectal cancer. Cochrane Database Syst Rev 2006;(1): CD005033. 32. Lieberman MD, Shou J, Sigal RK, Yu J, Goldfine J, Daly JM. Transfusion-induced immunosuppression results in diminished host survival in a murine neuroblastoma model. J Surg Res 1990;48:498-503. 33. Ydy LR, Slhessarenko N, de Aguilar-Nascimento JE. Effect of perioperative allogeneic red blood cell transfusion on the immune-inflammatory response after colorectal cancer resection. World J Surg 2007;31:20442051. 34. Yamanaka N, Okamoto E, Fujihara S, Kato T, Fujimoto J, Oriyama T, et al. Do the tumor cells of hepatocellular carcinomas dislodge into the portal venous stream during hepatic resection? Cancer 1992;70:22632267. 35. Jubert AV, Lee ET, Hersh EM, McBride CM. Effects of surgery, anesthesia and intraoperative blood loss on immunocompetence. J Surg Res 1973;15:399-403. 36. Totsuka E, Fung JJ, Hakamada K, Ohashi M, Takahashi K, Nakai M, et al. Synergistic effect of cold and warm ischemia time on postoperative graft function and outcome in human liver transplantation. Transplant Proc 2004;36:1955-1958. 37. Roayaie S, Frischer JS, Emre SH, Fishbein TM, Sheiner PA, Sung M, et al. Long-term results with multimodal adjuvant therapy and liver transplantation for the treatment of hepatocellular carcinomas larger than 5 centimeters. Ann Surg 2002;235:533-539.
Authors’ names in bold designate shared co-first authorship.
Supporting Information Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.27358/suppinfo.
Ischemia Time Impacts Recurrence of Hepatocellular Carcinoma After Liver Transplantation Shunji Nagai,1,2 Atsushi Yoshida,2 Marcelo Facciuto,1 Dilip Moonka,3 Marwan S. Abouljoud,2 Myron E. Schwartz,1 and Sander S. Florman1 Although experimental evidence has indicated that ischemia-reperfusion (I/R) injury of the liver stimulates growth of micrometastases and adhesion of tumor cells, the clinical impact of I/R injury on recurrence of hepatocellular carcinoma (HCC) after liver transplantation (LT) has not been fully investigated. To study this issue, we conducted a retrospective review of the medical records of 391 patients from two transplant centers who underwent LT for HCC. Ischemia times along with other tumor/recipient variables were analyzed as risk factors for recurrence of HCC. Subgroup analysis focused on patients with HCC who had pathologically proven vascular invasion (VI) because of the associated increased risk of micrometastasis. Recurrence occurred in 60 patients (15.3%) with median time to recurrence of 0.9 years (range, 40 days-4.6 years). Cumulative recurrence curves according to cold ischemia time (CIT) at 2-hour intervals and warm ischemia time (WIT) at 10-minute intervals showed that CIT >10 hours and WIT >50 minutes were associated with significantly increased recurrence (P 5 0.015 and 0.036, respectively). Multivariate Cox’s regression analysis identified prolonged cold (>10 hours; P 5 0.03; hazard ratio [HR] 5 1.9) and warm (>50 minutes; P 5 0.003; HR 5 2.84) ischemia times as independent risk factors for HCC recurrence, along with tumor factors, including poor differentiation, micro- and macrovacular invasion, exceeding Milan criteria, and alphafetoprotein >200 ng/mL. Prolonged CIT (P 5 0.04; HR 5 2.24) and WIT (P 5 0.001; HR 5 5.1) were also significantly associated with early (within 1 year) recurrence. In the subgroup analysis, prolonged CIT (P 5 0.01; HR 5 2.6) and WIT (P 5 0.01; HR 5 3.23) were independent risk factors for recurrence in patients with VI, whereas there was no association between ischemia times and HCC recurrence in patients with no VI. Conclusion: Reducing ischemia time may be a useful strategy to decrease HCC recurrence after LT, especially in those with other risk factors. (HEPATOLOGY 2015;61:895-904)
O
ver the past few decades, recurrence of hepatocellular carcinoma (HCC) after liver transplantation (LT) and its associated risk factors has been extensively investigated.1-3 Tumor size and number are generally accepted as predictive of recurrence, and many selection criteria for transplantation have used these characteristics, including the Milan, University of California San Francisco, and Up-to-7
criteria.4-6 Vascular invasion (VI) and tumor differentiation have also been shown to be significant predictors of recurrence.7-10 Several experimental models have shown that ischemia-reperfusion (I/R) injury of the liver promotes cancer cell implantation and growth.11-13 We hypothesize that growth of HCC micrometastases may be accelerated in LT recipients who receive liver grafts
Abbreviations: AFP, alpha-fetoprotein; ALT, alanine aminotransaminase; AST, aspartate aminotransaminase; CIT, cold ischemia time; DCD, donation after cardiac death; HCC, hepatocellular carcinoma; HR, hazard ratio; HTK, histidine-tryptophan-ketoglutarate; IQR, interquartile range; I/R, ischemia-reperfusion; LT, liver transplantation; MELD, Model for End-Stage Liver Disease; PRBCs, packed red blood cells; OR, odds ratio; UW, University of Wisconsin; VI, vascular invasion; WIT, warm ischemia time. From the 1Recanati/Miller Transplantation Institute, Icahn School of Medicine at Mount Sinai, New York, NY; 2Division of Transplant and Hepatobiliary Surgery, Henry Ford Transplant Institute, Henry Ford Hospital, Detroit, MI; and 3Division of Gastroenterology, Henry Ford Hospital, Detroit, MI. Received January 2, 2014; accepted August 1, 2014. Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.27358/suppinfo. Preliminary results were presented in 2011 at the International Liver Transplant Society, Valencia, Spain, and in 2012 at the American Transplant Congress, Boston, MA. 895
896
NAGAI ET AL.
exposed to prolonged ischemia times, leading to early recurrence of HCC after LT. Our study examines the relationship between ischemia time of the liver graft and the incidence and timing of HCC recurrence after LT based on experience from two large transplant centers in the United States: Mount Sinai Medical Center in New York City and Henry Ford Hospital in Detroit. We then performed validation analysis using a separate data set from Indiana University Hospital in Indianapolis.
Patients and Methods Study Population. Medical records of 271 patients at Mount Sinai Medical Center and 178 patients at Henry Ford Hospital who underwent LT for HCC were retrospectively reviewed. Patients with pathologically proven HCC in their explants were eligible for this study; mixed cholangiocarcinoma/HCC was excluded. Patients who died within 30 days of LT were excluded, as were patients who received living donor organs or organs procured after cardiac death. This study was approved by the institutional review boards at both centers. LT Procedure. Several techniques were used for the LT surgery, specifically in regard to the hepatic outflow reconstruction. Bicaval and piggyback techniques (standard piggyback and side-to-side cavacavostomy) were used based upon surgical preference. After the reconstruction of hepatic veins and portal vein, the transplanted liver was reperfused, followed by the hepatic artery reconstruction. Donor livers were flushed with CUSTODIOL histidine-tryptophanketoglutarate (HTK) solution (Essential Pharmaceuticals, LLC, Newtown, PA) or University of Wisconsin (UW) solution (Bridge to Life Solutions LLC, Columbia, SC). Donor livers were generally flushed with 46 L of solution in situ, and portal flush was added as needed on the back table. Cold ischemia time (CIT) was defined as the time from donor cross-clamping to the removal of the liver from the cold preservative solution preceding implantation. Warm ischemia time (WIT) was defined as the time from when the liver is taken out of the cold preservative solution in preparation for implantation until reperfusion.
HEPATOLOGY, March 2015
Post-Transplant Management. Maintenance immunosuppresion was with tacrolimus, mycophenolate mofetil, and steroids. Sirolimus was often added when recurrence of HCC was diagnosed, but was not routinely used in an adjuvant fashion. Abdomen and chest computed tomography or magnetic resonance imaging and alpha-fetoprotein (AFP) levels were monitored until the fifth year after LT. Biopsy was used selectively to establish the diagnosis of recurrent HCC; pathological confirmation was not required if the lesions progressed and were clinically consistent with HCC. Recurrence Analysis. HCC recurrence analysis was performed on the eligible patients. Variables regarding patient characteristics included age, sex, underlying liver disease, previous history of surgery or tumordirected therapies, and Model for End-Stage Liver Disease (MELD) score. Tumor factors included pretransplant AFP level, size and number of tumors, presence and degree of VI, and tumor grade. Operative factors included surgical techniques, operative time, transfusion of packed red blood cells (PRBCs), CIT, and WIT. Subgroup analysis was performed focused on patients with VI because of the associated increased risk of micrometastasis. In order to clarify degree of I/ R injury, post-transplant peak levels of aspartate transaminase (AST) and alanine transaminase (ALT) within the first week after LT were compared according to CIT and WIT. Validation of the Results of the Present Study. In order to validate the results of this study, data collected from another transplant center, Indiana University Hospital, was used to assess the association between ischemia times and HCC recurrence. The prediction model for HCC recurrence was applied to their data set with the same inclusion and exclusion criteria. The same statistical analysis methods were applied. Medical records of 305 patients who underwent LT for HCC were independently analyzed by the surgeon (R.M.) in Indiana University Hospital. Retrospective analysis of the transplant database has been approved by the institutional review board at the Indiana University School of Medicine. Statistical Analysis. Continuous variables were analyzed using the Student t test; for discrete variables, chi-square analysis was performed. Data are shown as
Address reprint requests to: Sander S. Florman, M.D., F.A.C.S., The Mount Sinai Medical Center, One Gustave L. Levy Place, Box 1104, New York, NY 10029-6574. E-mail: [email protected]; fax: 1-212-348-2474. C 2014 by the American Association for the Study of Liver Diseases. Copyright V View this article online at wileyonlinelibrary.com. DOI 10.1002/hep.27358 Potential conflict of interest: Nothing to report.
HEPATOLOGY, Vol. 61, No. 3, 2015
NAGAI ET AL.
897
Fig. 1. Comparison of post-transplant liver enzyme levels according to ischemia times (median and IQR). (A) There was a significant trend for post-transplant AST peak levels among CIT groups (P 5 0.02). Patients with prolonged CIT (10 hours was significantly associated with higher HCC recurrence rate (P 5 0.015 for CIT >10 hours vs.
50 minutes. WIT >50 minutes were significantly associated with higher HCC recurrence rate (P 5 0.036 for WIT >50 minutes vs. 50 minutes; Gray’s test; Fig. 2B). Based on these results, the thresholds for CIT and WIT were set at 10 hours and 50 minutes, respectively. On univariate analysis, tumor size and number, pathologically exceeding Milan criteria, poor differentiation, and micro- and macrovascular invasion were associated with HCC recurrence (Table 1). CIT and WIT were associated with recurrence as dichotomized variables (P 5 0.005 and 0.046, respectively); CIT was also associated with recurrence as a continuous variable (P 5 0.01; hazard ratio [HR] 51.12 per hour). The amount of intraoperative PRBC transfusion was associated with HCC recurrence in a continuous manner (P 5 0.001; HR 5 1.03 per unit). HCC recurrence was not associated with the type of preservation solution (UW or HTK solution; P 5 0.98). Significant factors on univariate analysis and clinically relevant factors were included in multivariate analyses. As the first multivariate analysis, CIT and WIT were dichotomized at 10 hours and 50 minutes,
HEPATOLOGY, Vol. 61, No. 3, 2015
NAGAI ET AL.
899
Table 1. Risk Factor for Recurrence of HCC Univariate Variables
Recipient age (per year) Female recipient Hepatitis C MELD score >15 AFP >200 ng/mL Exceeding Milan criteria Tumor size (per cm) Tumor number Poor differentiation VI Microvascular Macrovascular Donor age (per year) Previous hepatic resection Previous RFA Previous TACE Operation time (per hour) IVC replacement Preservation solution UW (ref.) HTK PRBCs (per unit) 0 unit (ref.) 1-10 units >10 units CIT (per hour) >10 hours WIT (per min) >50 minutes
Multivariate
P Value*
HR (95% CI)
0.99 0.14 0.19 0.61 0.25
1.0 (0.97-1.03) 0.55 (0.25-1.21) 0.71 (0.43-1.18) 0.99 (0.96-1.03) 0.73 (0.42-1.26)
P Value*
HR (95% CI)
1.89 (1.00-3.57) 2.06 (1.14-3.71)
10 hours (8.5 vs. 4.6 units; P < 0.001) were
900
NAGAI ET AL.
HEPATOLOGY, March 2015
associated with larger amount of transfusion; there was no association with WIT. On multivariate analysis, MELD score (P 5 0.003; odds ratio [OR] 5 1.05) and CIT >10 hours (P < 0.001; OR 5 3.95) remained as independent predictors. Subgroup Analysis of Ischemia Time in Patients With/Without VI. Subgroup analysis was performed according to status of pathologically proven VI. Interaction between VI, and ischemia times were identified to be significant (P < 0.001 for CIT >10 hours and P < 0.001 for WIT >50 minutes), which validated
this subgroup analysis. Pathologically proven VI was identified in 122 patients, 35 (28.7%) of whom developed recurrence. On multivariate analysis, CIT >10 hours (P 5 0.01; HR 5 2.6) and WIT >50 minutes (P 5 0.01; HR 5 3.23) were independent predictors of recurrence. In a similar analysis among patients without VI, neither CIT (P 5 0.99) nor WIT (P 5 0.21) was associated with HCC recurrence (Table 3). Based on a discriminant analysis, the following formula was established to predict risk of recurrence in patients with VI:
D524:112:21ðAFP > 200 ng=mLÞ13:75ðexceeding Milan criteriaÞ12:94 ðpoor differentiationÞ 12:94 ðCIT > 10 hours Þ12:87 ðWIT > 50 minÞ: ðScore 0 for not meeting criterion; score 1 for meeting criterion; D > 0 predicts recurrence:Þ
Table 2. Risk Factor for Early Recurrence of HCC (Within 1 Year) Multivariate Variables
P Value*
HR
95% CI
AFP >200 ng/mL Exceeding Milan criteria† Poor differentiation VI Microvascular Macrovascular Previous TACE PRBCs (per unit) CIT >10 hours WIT >50 minutes
0.03 0.08 0) and standard (D < 0) risk groups was significant (P < 0.001) and the rates were similar between the standard risk group and negative VI group (P 5 0.77). One- and three-year HCC recurrence rates in the high-risk, standard-risk, and negative VI group were 47.9% and 59.8%, 3.2% and 10.5%, and 3.0% and 8.5%, respectively (Fig. 3). Recurrence was accurately predicted in 85 of 111 (76.6%) patients with VI. Validation Analysis With Indiana University Hospital’s Data Set. Of 305 patients who underwent LT for HCC at Indiana University Hospital, 268 met
Table 3. Risk Factor for Recurrence of HCC in Patients With HCC Accompanied With VI VI Group (n 5 122) Variables
AFP >200 ng/mL Exceeding Milan criteria† Poor differentiation Previous TACE PRBCs CIT >10 hours WIT >50 minutes
P Value*
0.08 0.02 0.005 0.28 0.12 0.01 0.01
Non-VI Group (n 5 269)
HR
95% CI
2.08 3.13 3.1 1.54 1.02 2.6 3.23
0.92-4.7 1.25-7.88 1.4-6.88 0.7-3.4 0.99-1.05 1.23-5.49 1.24-8.38
*Cox’s proportional regression analysis. Exceeding Milan Criteria was included in the multivariate analysis, instead of tumor size and number. Abbreviations: TACE, transarterial chemoembolization; CI, confidence interval. †
P Value*
0.89 0.8 0.006 0.18 0.44 0.99 0.21
HR
95% CI
1.08 1.13 4.16 1.77 0.96 1.01 1.92
0.34-3.44 0.45-2.85 1.49-11.58 0.76-4.12 0.87-1.06 0.33-3.13 0.69-5.35
HEPATOLOGY, Vol. 61, No. 3, 2015
NAGAI ET AL.
901
(P 5 0.02; HR 5 2.17), but not between CIT >10 hours and recurrence (P 5 0.37; HR 5 1.48). Our prediction model was applied to this population, and the risk of HCC recurrence was clearly stratified among the groups. Adjusting death before HCC recurrence as a competing risk event, the differences of HCC recurrence rate between the high- and standard-risk groups, as well as between the standardrisk and negative VI groups, were significant (P 5 0.025 and 0.012, respectively; Gray’s test). Oneand three-year recurrence rates in the high-risk, standard-risk, and negative VI groups were 27.8% and 53.7%, 3.6% and 19.3%, 3.8% and 9.3%, respectively (Fig. 4).
Discussion
Fig. 3. Comparison of cumulative recurrence rate of HCC according to the prediction model. The classification was decided based on the following formula.
D524:112:21 ðAFP > 200 ng=mLÞ 13:75 ðexceeding MilancriteriaÞ12:94 ðpoor differentiationÞ 12:94 ðCIT > 10 hours Þ12:87 ðWIT > 50 minÞ: ðScore of 0 for not meeting criteria; score of 1 for meeting criteriaÞ If a patient has no or one risk factor, they are standard risk (D < 0), whereas if they have more than one risk factor, they are high risk (D > 0). Cumulative incidence curves of HCC recurrence and death before HCC recurrence. With adjusting death before HCC recurrence as a competing risk event, the difference of HCC recurrence between the high- and standard-risk groups was significant (P < 0.001) and the rates were similar between the standard-risk group and negative VI group (P 5 0.77).
inclusion criteria. HCC recurrence developed in 51 (19%). On univariate analysis, WIT >50 minutes was significantly associated with higher HCC recurrence rate (P 5 0.045; HR 5 1.94). CIT >10 hours showed tendency for increased risk of recurrence, although not statistically significant (P 5 0.19; HR 5 1.76). Neither WIT nor CIT was associated with HCC recurrence in a continuous manner (P 5 0.34 and 0.52). After adjusting the risk with positive VI, tumor size, and number (exceeding Milan criteria), there was significant association between WIT >50 minutes and HCC recurrence
Currently, prioritization of patients with HCC for LT is based on tumor characteristics alone.4,6,10,16-20 However, other factors may significantly affect HCC recurrence and survival. This study demonstrates that prolonged cold and/or warm ischemia of the liver graft are independent predictors of both overall and early (within 1 year) HCC recurrence, supporting our hypothesis that I/R injury of the liver graft accelerates growth and implantation of micrometastases of HCC present at the time of LT. A study of patients with HCC in the Scientific Registry of Transplant Recipients database demonstrated decreased overall survival among recipients who received organs procured after cardiac death versus those who received conventionally procured organs, suggesting an impact of greater I/R injury (interpretation of this study is, however, limited by the lack of data on HCC recurrence).21 Though the 13 patients in our series who received liver graft procured after cardiac death were excluded from the study, 5 of 13 developed recurrence (P 5 0.04; HR 5 1.61; univariate Cox’s regression analysis, data not shown in the results). Because HCC patients are disadvantaged by organ allocation based on MELD score in some regions, livers from donation after cardiac death (DCD) donors might be used for advanced HCC patients, especially those who do not meet criteria for MELD exceptional points. However, considering potential influence of donor WIT on HCC recurrence, the indication of DCD LT for HCC patients needs to be cautiously applied. Experimental evidence has suggested a number of biological mechanisms for the association between cancer outcome and I/R injury to the liver.11,12 I/R induces mechanical injury to hepatic sinusoids, which leads
902
NAGAI ET AL.
Fig. 4. Comparison of recurrence rate of HCC according to the prediction model by using the data set from the other transplant center (Indiana University Hospital) as the validation of the prediction model. Cumulative incidence curves of HCC recurrence and death before HCC recurrence. With adjusting death before HCC recurrence as a competing factor, the difference of HCC recurrence between the high- and standard-risk groups was significant (P 5 0.025) and between the standard-risk group and negative VI group (P 5 0.012).
to hepatic microcirculatory barrier dysfunction and activates cell signals related to invasion and migration.11,22 Hypoxia induces genes and cytokines involved in angiogenesis, cellular proliferation, growth, and adhesion: for example, in the absence of oxygen, hypoxia-inducible transcription factor 1 binds to hypoxia-response elements, activating the expression of hypoxia-response genes, such as vascular endothelial growth factor.23-26 Most of the recurrences among our patients were in extrahepatic organs, suggesting that the oncological effects of prolonged ischemia time may be systemic. Man et al. reported that infiltrative growth pattern of tumor cells accompanied with venous invasion and cell signal related to tumor cell invasion were observed not only in liver tumor cells, but also in the lung metastasis nodules in rats undergoing I/R injury and major hepatectomy, compared to the control group.11 Activation of cancer cell migration and invasion pathways by I/R injury is one reported mechanism by which accelerated growth and implantation of extrahepatic micrometastases might occur.13 Significantly high AST and
HEPATOLOGY, March 2015
ALT levels in the prolonged ischemia time groups provide evidence of greater I/R injury in these patients. Nevertheless, the biological mechanisms accounting for acceleration of extrahepatic recurrence has not been clearly demonstrated yet, and further investigations are necessary. Similar to previously reported studies,27 we observed a significant association of PRBC transfusion with HCC outcome after LT on univariate analysis. Mechanisms underlying the oncological impact of blood loss and/or transfusion have been proposed for various tumors, including HCC, colon cancer, gastric cancer, and pancreatic cancer.28-31 Blood transfusion can impair immunity and enhance inflammation.32,33 Profuse hemorrhage during surgery may lead to more tumor manipulation that could result in dissemination of cancer cells.34 Systemic hypoperfusion resulting from massive hemorrhage impairs oxygen delivery to vital organs, leading to systemic inflammation and production of cytokines that may reduce antitumor immunity.35 In our study, large volume transfusion was associated with overall HCC recurrence, but did not remain as an independent risk factor on multivariate analysis. We observed a significant correlation between ischemia times and transfusion requirements. Poor early graft function associated with prolonged ischemia can lead to intraoperative coagulopathy and increased blood loss.36 This correlation could at least partly underlie the connection between ischemia time and HCC recurrence. Because the entire liver is removed at LT, post-LT HCC recurrence must be the result of undetectable micrometastases present at the time of LT; if ischemia accelerates tumor growth, its impact should be greater in patients more likely to harbor micrometastases (e.g., vascular invasion). Subgroup analysis in patients with vascular invasion showed that long ischemia time, both CIT and WIT, are independent predictors of HCC recurrence in addition to exceeding Milan criteria, poor tumor differentiation, and elevated AFP level. In the presence of any two of these risk factors, the 1year recurrence rate is 47.9%. Though not of use in case selection given that the required data are only available after LT, our formula enables risk stratification that may be useful in establishing the post-LT monitoring schedule, choosing the immunosuppression protocol, and selecting patients for clinical trials of adjuvant strategies.37 In order to validate our findings and determine the efficacy of our recurrence formula for high-risk HCC patients, we reviewed and applied our analysis to
HEPATOLOGY, Vol. 61, No. 3, 2015
another center’s database (Indiana University). The analysis for the validation of our findings and prediction model confirmed increased risk of HCC recurrence associated with prolonged ischemia times. WIT showed more-significant association, which was consistent with our findings. Neither WIT nor CIT correlated in a continuous manner in this data set, whereas WIT showed significant association in a continuous manner, but not CIT, in our series. The negative influence probably remarkably increased at some thresholds and did not seem to have linear correlation with time. The cut-off time for increased risk of HCC recurrence remains to be elucidated, because of the limited sample size. In terms of the risk stratification, we could emphasize the validity of our prediction model and the importance of modifications of post-transplant management for these populations at high risk, because the patients from the different data set categorized in the high-risk group also showed much earlier and higher rates of HCC recurrence, as our population demonstrated. Thus, these findings further supported our conclusion. In order to clarify the clinical implication of ischemia time on HCC recurrence in LT patients, a multicenter, clinical study should be conducted. In addition, further experimental models are necessary to unravel biological mechanisms accounting for potential oncological influence of I/R injury. Pending the results of those studies, it might be reasonable to make all efforts to reduce ischemia time before transplant in patients with HCC. Ischemia times could be shortened by improved selection of surgical procedures and systematic changes in the organ allocation process. Transporting the liver graft and timing of recipient surgery could be further optimized by efficient communication among transplant teams. Thus, we suggest that transplant surgeons, hepatologists, and coordinators should recognize the possible influence of ischemia time on HCC recurrence after LT. In conclusion, we have demonstrated that prolongation of ischemia time, both warm and cold, predicts early HCC recurrence after LT. Our results suggest that interventions to shorten ischemia times may improve outcomes for patients with HCC who undergo LT, especially those with other risk factors for recurrence. Acknowledgment: The authors thank Drs. Richard S. Mangus, M.D., M.S., and A. Joseph Tector, M.D., Ph.D., Transplant Division, Indiana University Hospital, for their participation in the validation analysis. The authors also thank Drs. Shozo Mori, M.D., Ph.D., Mizuki Ninomiya, M.D., Ph.D., and Lloyd Brown, M.D.,
NAGAI ET AL.
903
for data gathering and analysis of Mount Sinai Medical Center and Henry Ford Hospital.
References 1. Prasad KR, Young RS, Burra P, Zheng SS, Mazzaferro V, Moon DB, Freeman RB. Summary of candidate selection and expanded criteria for liver transplantation for hepatocellular carcinoma: a review and consensus statement. Liver Transpl 2011;17(Suppl 2):S81-S89. 2. Nagai S, Abouljoud MS, Kazimi M, Brown KA, Moonka D, Yoshida A. Peritransplant lymphopenia is a novel prognostic factor in recurrence of hepatocellular carcinoma after liver transplantation. Transplantation 2014;97:694-701. 3. Halazun KJ, Hardy MA, Rana AA, Woodland DC, Luyten EJ, Mahadev S, et al. Negative impact of neutrophil-lymphocyte ratio on outcome after liver transplantation for hepatocellular carcinoma. Ann Surg 2009;250:141-151. 4. D’Amico F, Schwartz M, Vitale A, Tabrizian P, Roayaie S, Thung S, et al. Predicting recurrence after liver transplantation in patients with hepatocellular carcinoma exceeding the up-to-seven criteria. Liver Transpl 2009;15:1278-1287. 5. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693-699. 6. Yao FY, Ferrell L, Bass NM, Watson JJ, Bacchetti P, Venook A, et al. Liver transplantation for hepatocellular carcinoma: expansion of the tumor size limits does not adversely impact survival. HEPATOLOGY 2001;33:1394-1403. 7. DuBay D, Sandroussi C, Sandhu L, Cleary S, Guba M, Cattral MS, et al. Liver transplantation for advanced hepatocellular carcinoma using poor tumor differentiation on biopsy as an exclusion criterion. Ann Surg 2011;253:166-172. 8. Gouw AS, Balabaud C, Kusano H, Todo S, Ichida T, Kojiro M. Markers for microvascular invasion in hepatocellular carcinoma: where do we stand? Liver Transpl 2011;17(Suppl 2):S72-S80. 9. Lim KC, Chow PK, Allen JC, Chia GS, Lim M, Cheow PC, et al. Microvascular invasion is a better predictor of tumor recurrence and overall survival following surgical resection for hepatocellular carcinoma compared to the Milan criteria. Ann Surg 2011; 254:108-113. 10. Mazzaferro V, Llovet JM, Miceli R, Bhoori S, Schiavo M, Mariani L, et al. Predicting survival after liver transplantation in patients with hepatocellular carcinoma beyond the Milan criteria: a retrospective, exploratory analysis. Lancet Oncol 2009;10:35-43. 11. Man K, Ng KT, Lo CM, Ho JW, Sun BS, Sun CK, et al. Ischemiareperfusion of small liver remnant promotes liver tumor growth and metastases—activation of cell invasion and migration pathways. Liver Transpl 2007;13:1669-1677. 12. van der Bilt JD, Kranenburg O, Nijkamp MW, Smakman N, Veenendaal LM, Te Velde EA, et al. Ischemia/reperfusion accelerates the outgrowth of hepatic micrometastases in a highly standardized murine model. HEPATOLOGY 2005;42:165-175. 13. Ku Y, Kusunoki N, Shiotani M, Maeda I, Iwasaki T, Tominaga M, et al. Stimulation of haematogenous liver metastases by ischaemiareperfusion in rats. Eur J Surg 1999;165:801-807. 14. Gray R. A class of K-sanple tests for comparing the cumulative incidence of a competing risk. Ann Stat 1988;16:1141-1154. 15. Bruix J, Sherman M; Practice Guidelines Committee, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma. HEPATOLOGY 2005;42:1208-1236. 16. Kaido T, Takada Y, Uemoto S. Usefulness of the Kyoto criteria as selection criteria for living donor liver transplantation for hepatocellular carcinoma. Liver Transpl 2010;16:538-540. 17. Piardi T, Gheza F, Ellero B, Woehl-Jaegle ML, Ntourakis D, Cantu M, et al. Number and tumor size are not sufficient criteria to select
904
18.
19.
20.
21.
22.
23.
24.
25.
26. 27.
28.
29.
NAGAI ET AL.
patients for liver transplantation for hepatocellular carcinoma. Ann Surg Oncol 2012;19:2020-2026. Llovet JM, Fuster J, Bruix J; Barcelona Clinic Liver Cancer Group. The Barcelona approach: diagnosis, staging, and treatment of hepatocellular carcinoma. Liver Transpl 2004;10(2 Suppl 1):S115-S120. Kaido T, Takada Y, Uemoto S. Selection criteria for patients with hepatocellular carcinoma in liver transplantation. Transplantation 2009;88: 442-443; author reply, 443. Makuuchi M, Kokudo N, Arii S, Futagawa S, Kaneko S, Kawasaki S, et al. Development of evidence-based clinical guidelines for the diagnosis and treatment of hepatocellular carcinoma in Japan. Hepatol Res 2008;38:37-51. Croome KP, Wall W, Chandok N, Beck G, Marotta P, HernandezAlejandro R. Inferior survival in liver transplant recipients with hepatocellular carcinoma receiving donation after cardiac death liver allografts. Liver Transpl 2013;19:1214-1223. Itoh K, Yoshioka K, Akedo H, Uehata M, Ishizaki T, Narumiya S. An essential part for Rho-associated kinase in the transcellular invasion of tumor cells. Nat Med 1999;5:221-225. Axelson H, Fredlund E, Ovenberger M, Landberg G, Pa˚hlman S. Hypoxia-induced dedifferentiation of tumor cells—a mechanism behind heterogeneity and aggressiveness of solid tumors. Semin Cell Dev Biol 2005;16:554-563. Brahimi-Horn C, Pouyssegur J. The role of the hypoxia-inducible factor in tumor metabolism growth and invasion. Bull Cancer 2006;93: E73-E80. Carmeliet P, Dor Y, Herbert JM, Fukumura D, Brusselmans K, Dewerchin M, et al. Role of HIF-1alpha in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature 1998;394:485-490. Harris AL. Hypoxia—a key regulatory factor in tumour growth. Nat Rev Cancer 2002;2:38-47. Kaido T, Takada Y, Egawa H, Uemoto S. The influence of intraoperative homologous blood transfusion on prognosis after liver transplantation for hepatocellular carcinoma. Hepatogastroenterology 2009;56:808-812. Nagai S, Fujii T, Kodera Y, Kanda M, Sahin TT, Kanzaki A, et al. Impact of operative blood loss on survival in invasive ductal adenocarcinoma of the pancreas. Pancreas 2011;40:3-9. Katz SC, Shia J, Liau KH, Gonen M, Ruo L, Jarnagin WR, et al. Operative blood loss independently predicts recurrence and survival after resection of hepatocellular carcinoma. Ann Surg 2009;249:617-623.
HEPATOLOGY, March 2015
30. Ojima T, Iwahashi M, Nakamori M, Nakamura M, Naka T, Katsuda M, et al. Association of allogeneic blood transfusions and long-term survival of patients with gastric cancer after curative gastrectomy. J Gastrointest Surg 2009;13:1821-1830. 31. Amato A, Pescatori M. Perioperative blood transfusions for the recurrence of colorectal cancer. Cochrane Database Syst Rev 2006;(1): CD005033. 32. Lieberman MD, Shou J, Sigal RK, Yu J, Goldfine J, Daly JM. Transfusion-induced immunosuppression results in diminished host survival in a murine neuroblastoma model. J Surg Res 1990;48:498-503. 33. Ydy LR, Slhessarenko N, de Aguilar-Nascimento JE. Effect of perioperative allogeneic red blood cell transfusion on the immune-inflammatory response after colorectal cancer resection. World J Surg 2007;31:20442051. 34. Yamanaka N, Okamoto E, Fujihara S, Kato T, Fujimoto J, Oriyama T, et al. Do the tumor cells of hepatocellular carcinomas dislodge into the portal venous stream during hepatic resection? Cancer 1992;70:22632267. 35. Jubert AV, Lee ET, Hersh EM, McBride CM. Effects of surgery, anesthesia and intraoperative blood loss on immunocompetence. J Surg Res 1973;15:399-403. 36. Totsuka E, Fung JJ, Hakamada K, Ohashi M, Takahashi K, Nakai M, et al. Synergistic effect of cold and warm ischemia time on postoperative graft function and outcome in human liver transplantation. Transplant Proc 2004;36:1955-1958. 37. Roayaie S, Frischer JS, Emre SH, Fishbein TM, Sheiner PA, Sung M, et al. Long-term results with multimodal adjuvant therapy and liver transplantation for the treatment of hepatocellular carcinomas larger than 5 centimeters. Ann Surg 2002;235:533-539.
Authors’ names in bold designate shared co-first authorship.
Supporting Information Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.27358/suppinfo.
