“Liver Transplantation for Hepatocellular Carcinoma using Grafts from Uncontrolled Circulatory Death Donation Versus Brain Death Donation”


 Controversy exists regarding whether the rate of hepatocellular carcinoma (HCC) recurrence after orthotopic liver transplantation (OLT) differs when using livers from donation after controlled circulatory death (DCD) versus livers from donation after brain death (DBD). The aim of this cohort study was to analyze rates of HCC recurrence, patient survival, and graft survival after OLT for HCC, comparing recipients of DBD livers (n=103) with recipients of uncontrolled DCD livers (uDCD; n=41). No significant differences in tumor size, tumor number, serum alpha-fetoprotein, proportion of patients within Milan criteria, or pre-OLT bridging therapies were identified between groups, although the waitlist period was significantly shorter in the uDCD group (p=0.04). HCC recurrence was similar between groups. Patient survival was similar between groups, but graft survival was lower in the uDCD group. Multivariate analysis identified recipient age (p=0.03), pre-OLT bridging therapy (p=0.02), and HCC recurrence (p=0.04) as independent risk factors for patient survival and pre-OLT transarterial chemoembolization (p=0.04) as the single risk factor for HCC recurrence. In conclusion, similar patient survival and lower graft survival were observed in the uDCD group. However, the use of uDCD livers appears to be justified due to shorter waitlist time and similar HCC recurrence in both groups.


Introduction
Hepatocellular carcinoma (HCC) represents the sixth most common malignancy and the third leading cause of cancer-related deaths around the world 1,2 . It is frequently diagnosed incidentally or during screening programs, as people who develop HCC typically do not manifest symptoms until the tumor has reached a late stage.
Orthotopic liver transplantation (OLT) is considered the treatment of choice for patients with early-stage HCC (Milan criteria). In Spain, 1,227 patients underwent OLT during 2019, with 282 (23%) receiving transplants for HCC 3 . However, the number of available liver grafts remains insu cient to treat all patients who require an OLT for malignant or benign disease. To increase the liver graft pool and thereby decrease waitlist mortality, new strategies have been developed, such as the use of livers from marginal donors 4 , including livers donated after controlled circulatory death (cDCD) 5,6 or uncontrolled circulatory death (uDCD) [7][8][9][10][11] .
In Type 2 uDCD donation, the donor sustained a witnessed out-of-hospital cardiac arrest and underwent unsuccessful cardiopulmonary resuscitation, whereas in type 3 cDCD donation, organs are recovered after death con rmation from patients with irreversible brain injury or respiratory failure, in whom lifesustaining treatment has been withdrawn 12 . Liver grafts from both uDCD and cDCD donation are subjected to longer warm ischemia periods, compared to grafts donated after brain death (DBD), which results in a higher likelihood of ischemia/reperfusion injury (IRI). Use of uDCD livers has been associated with higher rates of primary nonfunction (PNF) and biliary complications (BCs) than OLT with cDCD and DBD donors. This has been attributed to the longer period of ischemia with uDCD, which is the sum of the donor circulatory arrest time, duration of cardiopulmonary resuscitation, duration of normothermic regional perfusion (NRP), and recipient warm ischemia time (WIT). 7,11 However, use of uDCD livers may be justi ed given the length of time OLT candidates remain on the waitlist (mean, 124 days) 3 .
As patients with HCC often exhibit compensated, clinically stable disease with a relatively low Model for End-stage Liver Disease (MELD) score when placed on the waitlist, these individuals are usually ideal candidates for cDCD OLT, as they are able to tolerate potential complications related to the use of marginal livers. Nevertheless, several reports have demonstrated increased HCC recurrence after OLT using cDCD, as well as reduced patient and graft survival. [13][14][15] These results were explained in a mouse model, which showed that IRI is a strong stimulus for recurrent intrahepatic tumor growth in transplanted livers 13,15,16 .
To our knowledge, no previous study has evaluated the use of uDCD livers in patients with HCC. The aim of this study was to analyze use of liver grafts from uDCD donation in patients with HCC, comparing rates of patient and graft survival, HCC recurrence, and recurrence-free survival with a control group of patients with HCC who received livers from DBD donation.

Study population and study design
Between April 1986 and December 2016, 1,876 OLTs were performed in adults and children at our institution. From January 2006 to December 2016, 75 of these OLTs were performed using livers from uDCD donation. This retrospective cohort study compared 103 adults with HCC who underwent OLT using liver grafts from DBD donation (DBD Group) with 41 adults with HCC who underwent OLT with livers from uDCD donation (uDCD Group). There was a chronological correlation between cases and controls. This study was performed in accordance with the ethical guidelines of the declaration of Helsinki, and was approved by our Institutional Review Board. The study was closed on October 31, 2018, after a minimum follow-up of 22 months after OLT.
Informed consent was obtained from all recipients for OLT, including extended criteria donors. In the case of donors, informed consent for donation was also obtained from the next of kin prior to organ retrieval.
Criteria for acceptance of uDCD and DBD liver grafts All liver donors in this study were maintained with NRP for a maximum of 300 min before initiation of organ perfusion, in accordance with our previously described protocol for uDCD donors 11 . The criteria for acceptance of uDCD livers were as follows: donor age between 14 and 55 years; maximum transaminase levels <4 times the upper limit of normal; and absence of alcoholic disease, drug addiction, history of cancer, hepatitis B and/or C infection, HIV infection, violent death, or abdominal trauma. Additional criteria, which were assessed at the time of organ procurement, included good appearance, consistency, and vascularization of the liver graft, and no evidence of ischemia of the gallbladder, common bile duct, or intestines.
Graft donor warm ischemia time (DWIT), also called pre-NRP WIT, was de ned as the sum of circulatory arrest time and duration of pre-NRP cardiopulmonary resuscitation, whereas recipient WIT (RWIT) was de ned as the interval from removal from cold preservation solution to completion of portal vein anastomosis. We routinely performed liver graft biopsy before cold perfusion and discarded grafts with brosis or >30% macrosteatosis. We also discarded grafts when DWIT or RWIT exceeded the times established in the protocol.

Inclusion and exclusion criteria, candidate information, and transplant technique
The inclusion criteria for OLT recipients were age >18 years, HCC as the main indication for transplant, and within the Milan 17 or University of California, San Francisco (UCSF) criteria for transplantation 18 . We excluded patients who received partial or combined transplants, underwent retransplantation, were positive for HIV, underwent OLT for fulminant hepatitis, or received grafts from donors >70 years of age (for DBD recipients). Use of uDCD livers was avoided in patients with previous abdominal surgery or a MELD score >30. No exception points were added to the MELD score for patients with tumors >2 cm in diameter. To prevent tumor progression, transarterial chemoembolization (TACE) was usually performed in patients with several tumors or a single tumor and ascites, whereas radiofrequency ablation was generally used in patients with a single tumor. Patients who signed the informed consent for uDCD livers were included on both waitlists (uDCD and DBD).
Recipient hepatectomy was performed using the vena cava sparing technique (piggy-back). In most cases, biliary reconstruction was performed as a cholechocholedochostomy without T-tube.

Comparisons of donor and pre-OLT recipient variables
Donor variables common to both groups were compared between groups. These included age, sex, body mass index (BMI), cause of death, vasopressor use, transfusion, intensive care unit (ICU) stay, cardiac arrest, steatosis, cold ischemia time (CIT), and RWIT. The following pre-OLT recipient characteristics were also compared between groups: age, sex, BMI, OLT indication, MELD score, comorbidities, waitlist duration, and laboratory values. Likewise, a number of perioperative variables were compared between groups, including biliary reconstruction, transfusion of blood products, post-OLT liver function, immunosuppression, retransplantation, post-OLT complications, patient and graft survival, and recurrence-free survival. Pre-and post-OLT tumor characteristics and HCC recurrence were also compared between groups.
PNF was de ned as severe clinical deterioration requiring retransplantation or progressing to death, which was the consequence of irreversible liver graft failure within 10 days after OLT, in the absence of vascular complications. Among BCs, non-anastomotic biliary strictures (NABS) were de ned as any stricture, dilation, or irregularity of the intrahepatic or extrahepatic bile ducts of the liver (hilum), whereas anastomotic biliary strictures (ABS) were de ned as lesions localized to the anastomosis site. 19

Immunosuppression
The immunosuppressive regimen consisted of tacrolimus and prednisone. Corticosteroids were usually discontinued between 3 and 6 months after OLT. Tacrolimus trough levels were maintained between 10-15 ng/mL during the 1 st months after transplantation, between 8 and 12 ng/mL until the 6 th month, and between 5 and 8 ng/mL thereafter. Mild acute rejection was treated with increasing the dose of tacrolimus, whereas moderate or severe episodes was treated with 1 g methylprednisolone intravenously for 3 days.
Currently, we use a tacrolimus-based regimen with lower tacrolimus doses, which includes mycophenolate mofetil (MMF) or a mammalian target of rapamycin inhibitor (mTORi) in the presence of renal dysfunction, hypertension, diabetes, de novo tumor, or HCC as OLT indication. Conversion from tacrolimus to MMF or mTORi monotherapy is performed on long-term follow-up in recipients who undergo OLT for HCC or in patients with severe adverse tacrolimus effects but stable liver function.

Statistical analysis
Quantitative variables were expressed as mean and standard deviation or median and interquartile (25%-75%) range. Qualitative variables were expressed as percentages. Differences between qualitative variables were assessed by chi-square test or Fisher's exact test, as appropriate. Quantitative variables were compared using t-test or Mann-Whitney U test, depending on whether the data were normally distributed. Graft and patient survival rates were estimated using the Kaplan-Meier method, and survival curves were compared using the log-rank test. Donor and recipient variables with p values <0.10 in univariate analysis were subsequently investigated in multivariate analysis using Cox's regression model to evaluate associations between baseline variables and patient or graft survival. Results were expressed as hazard ratios (HRs) and 95% con dence intervals (CIs). P values <0.05 were considered statistically signi cant. All analyses were performed with SPSS Statistics Version 24.

Results
Donor and recipient characteristics uDCD donors were signi cantly younger than DBD donors (41.0±10.0 y vs. 48.0±13.0 y; p = 0.001). Mean BMI, as well as rates of cardiac arrest and vasopressor use, was signi cantly higher in the uDCD group. There were no statistically signi cant differences between groups for CIT, RWIT, and rates of donor microsteatosis or macrosteatosis.
Mean age was signi cantly higher in uDCD recipients than in DBD recipients (57.0±8.1 y vs. 61.0±6.6 y; p = 0.007). The prevalence of other OLT indications associated with HCC was similar in both groups, as was the prevalence of other morbidities. The median MELD score was signi cantly higher in recipients of uDCD donors (p = 0.008). Among laboratory variables, recipients of uDCD livers had signi cantly lower platelet counts and prothrombin rates (Table 1).

Perioperative characteristics and post-OLT complications
Types of biliary reconstruction techniques were similarly distributed in the two groups, whereas the volume of all intraoperative blood products was signi cantly higher in uDCD recipients. Liver function parameters were similar between groups on the 7th post-OLT day, but median gamma-glutamyl transpeptidase (GGT) and bilirubin were signi cantly lower in uDCD recipients on the 30th post-OLT day. Immunosuppressor use, hospital stay, or ICU stay did not differ signi cantly between groups.
The rate of retransplantation was signi cantly higher in recipients of uDCD livers (5 patients; 12.2%) than in recipients of DBD livers (1 patient; 1.0%; p = 0.01). Reasons for retransplantation were PNF and NABS (4 patients and 1 patient, respectively) in the uDCD group. For the 1 patient who underwent retransplantation in the DBD group, PNF was the indication (Table 2).   The time from OLT waitlist inclusion to OLT was also signi cantly longer in the DBD group than in the uDCD group (7.9 ± 5.4 mo vs. 5.3 ± 3.0 mo; p = 0.040). Similar proportions of patients in both groups (48.5% in DBD and 48.7% in uDCD; p = 0.413) received pre-OLT HCC bridging therapies; rates for pre-OLT TACE or radiofrequency ablation did not differ between groups. After histological examination of the explant liver, we observed no statistically signi cant differences between groups for median tumor size, number of tumors, and vascular or perineural invasion. The proportion of patients within Milan criteria decreased from pre-OLT to post-OLT (based on liver explant histological ndings), although there were no differences between groups either pre-or post-OLT. There were corresponding increases in the percentage of patients within UCSF from pre-OLT to post-OLT, with no signi cant between-group differences at either time. The proportion of patients exceeding UCSF criteria after OLT were similar in the DBD group (19 patients; 18.4%) and uDCD group (7 patients; 17.1%; p = 0.528). ± 44 mo in the uDCD group; p = 0.550), the rate of tumor recurrence was similar in both groups (7.8% in the DBD group vs. 7.3% in the uDCD group; p = 0.597). The median time from OLT to tumor recurrence diagnosis was longer in the DBD group than in the uDCD group, although the difference did not reach statistical signi cance (33 mo vs. 12 mo; p = 0.091). Locations of tumor recurrence were the liver, bones, or lungs, which were not signi cantly different between groups. Four (3.8%) patients died of tumor recurrence in the DBD group, whereas 1 (2.4%) patient died of recurrence in the uDCD group (p = 0.545).
Preliminary studies from the Scienti c Registry of Transplant Recipients demonstrated that use of cDCD livers increased the risk of death in OLT recipients with HCC 30 and was associated with inferior patient and graft survival 13 . More recently, other OLT series comparing outcomes between cDCD and DBD livers in patients with HCC showed no differences in HCC recurrence 5,21,31 , patient survival, or tumor-free survival 5,31 . Furthermore, Khorsandi et al. 31 found no difference in survival between DCD and DBD transplant recipients but noted that increased HCC size and number, microvascular invasion, and poor tumor differentiation were associated with increased cancer-speci c mortality, thereby highlighting the importance of tumor biology on survival and recurrence post-OLT in patients with HCC.
As with cDCD livers, uDCD livers have often been used in patients with HCC [7][8][9][10][11]34 , reaching up to 85-90% of patients in two recent series 9,10 (one of which involved 46% of patients exceeding the Milan criteria) 9 . These series also underline the importance of using uDCD livers in recipients with HCC who meet recommended indications and avoiding their use when post-OLT graft function is unpredictable 9 . In agreement with previous authors 9 , our results con rmed that use of uDCD livers confers the advantage of reducing waitlist time, compared with DBD livers. Similar to donor data of previously reported studies 7-10 , the mean age of our uDCD donors was signi cantly lower than DBD donors, yet the mean age of our uDCD recipients was signi cantly higher.
As in other cDCD donation 35,36 and uDCD donation experiences 8-11, 34,37 the main disadvantages of uDCD livers in our series, when compared with DBD livers, were the increased need for blood products and higher rates of PNF, ischemic BCs, and retransplantation. However, on the 30th post-OLT day, liver function of our uDCD recipients was equivalent to that of DBD recipients, re ecting the good recovery of uDCD livers.
Overall, our results support the use of cDCD 6 uDCD group). However, the overall HCC recurrence rate after OLT was similar in both groups (7.8% in DBD vs. 7.3% in uDCD), with parallel distributions of metastasis locations. Other authors have reported similar HCC recurrence rates between DBD and cDCD liver recipients (12.1% vs. 12.3%) 5 .
Previous studies reported no differences in 1-year patient survival between recipients of uDCD and DBD livers but lower 1-year graft survival in uDCD liver recipients 8,9,11,37 . Our study con rmed similar overall patient survival, disease-free survival, and tumor recurrence rates at 5 years between recipients receiving DBD and uDCD livers but lower 5-year graft survival in the uDCD group, which was mainly attributed to a higher incidence of ischemic BCs and PNF. The increased risk of these complications with cDCD 27,29 or uDCD livers 7,9,10,34,39 is well known, and when we excluded patients with PNF (the number of which has decreased substantially during the past 4 years) from our analysis, graft survival was not signi cantly different between groups.
Our multivariate analysis revealed recipient age, use of bridging therapies, and HCC recurrence as independent risk factors for patient survival, whereas pre-OLT bridging therapy with TACE was the only risk factor for HCC recurrence. The authors of a recent study found no association between locoregional therapy and worse patient survival, and they considered tumor biology (size, number, differentiation, and microvascular invasion) to be more relevant for patient survival 31 . Nevertheless, the need for bridging therapies may re ect unfavorable HCC biology and, therefore, a higher risk of recurrence and lower survival. Thus, bridging locoregional therapies for patients within Milan criteria do not improve post-OLT survival or recurrences in the majority of patients who fail to achieve complete pathologic response 40 . It has been suggested that delaying OLT for 3 months after inclusion on the waitlist can reduce early HCC recurrence by excluding patients with poor tumor biology 41 .
There are some limitations to this study. Data were collected retrospectively and were therefore subject to the typical biases expected with this study design. Although this study represents the largest single-center experience using uDCD livers in recipients with HCC, the sample size was limited for analyzing risk factors by multivariate analysis.

Conclusion
In conclusion, when comparing uDCD and DBD livers in HCC recipients, similar patient survival but lower graft survival was observed in the uDCD group, which was attributed to the group's higher incidence of PNF and BCs. Furthermore, HCC recurrence rates after OLT were similar with uDCD or DBD liver grafts, and TACE as a bridging therapy was identi ed as a risk factor for post-OLT recurrence.

Declarations
Data availability