No Prognostic Impact of Graft-to-Recipient Weight Ratio on Hepatocellular Carcinoma Recurrence Following Living Donor Liver Transplantation

Background

Hepatocellular carcinoma (HCC) is a well-established indication for liver transplantation (LT) and is a significant reason for living donor liver transplantation (LDLT) in Asian countries [1]. Nevertheless, HCC recurrence after LT leads to reduced patient survival, even with aggressive approaches to recurrent tumor management. Consequently, the selection of candidates for LT is conducted with meticulous consideration to minimize the likelihood of HCC recurrence.

Post-transplant HCC recurrence rates have been reported to be higher in patients undergoing LDLT than in deceased donor LT [2,3]. The higher recurrence rates in patients undergoing LDLT may be due to the expedited process of patient evaluation and selection for LDLT, resulting in transplantation of more aggressive tumors [3]; excessive manipulation of the native liver for preservation of the retrohepatic inferior vena cava [4]; and regeneration of the partial liver graft, enhancing HCC recurrence [5].

A single-center study from Korea reported that a graft-to-recipient weight ratio (GRWR) <0.8% was associated with higher tumor recurrence rates in patients with HCC beyond, but not within, the Milan criteria [5]. By contrast, other studies found that small-sized liver grafts do not raise the risk of HCC recurrence after LDLT [6,7]. Several animal studies, however, have indicated that regeneration of the partial liver graft enhances HCC recurrence [8–10].

To address the conflicting findings concerning the prognostic impact of GRWR, this retrospective, single-center, observational study aimed to investigate whether GRWR played a role in influencing the occurrence of HCC recurrence following LDLT.

Material and Methods

PATIENT SELECTION:

This retrospective observational single-center study evaluated the association between GRWR and post-transplant HCC recurrence. We conducted a search within the LT database of the Asan Medical Center to find adult patients who had undergone primary LDLT for HCC between January 2006 and December 2016, covering 11 years. Inclusion criteria were patients aged ≥18 years at the time of LDLT, with a pre-LT diagnosis of HCC. Exclusion criteria were patients incidentally diagnosed with HCC through liver examination at the time of transplantation, those who had undergone re-transplantation, individuals with combined HCC-cholangiocarcinoma, and those who had died within less than 3 months after LT. Additionally, patients who had undergone dual-donor LT were excluded to prevent potential bias associated with multiple donors. A total of 856 patients were included in this study.

PATIENT GROUPING AND FOLLOW-UP:

Small-sized liver grafts usually indicate those with GRWR <0.8%, but the range of GRWR ≥0.8% in the control group was too wide to compare objectively. Thus, these patients were divided into 4 GRWR groups, consisting of patients with GRWR <0.8%, ≥0.8% and <1.0%, ≥1.0%, and <1.2%, and ≥1.2% [6]. Patients were also divided into 2 groups based on the Milan criteria, consisting of those within and beyond the Milan criteria; and into those based on α-fetoprotein–des-γ-carboxy prothrombin–tumor volume (ADV) scores, with a cutoff of 5log [11–13]. The mammalian target of rapamycin inhibitors were not administered before HCC recurrence in the study patients. Patient follow-up was extended until April 2021 or until the time of patient death, and this was accomplished by comprehensive review of institutional medical records, along with support from the National Health Insurance Service in Korea.

ETHICS STATEMENT:

The study protocol received approval from the institutional review board of Asan Medical Center (No. 2021-1611), which waived the need for informed consent given the retrospective nature of this investigation. The study adhered to ethics guidelines outlined in the World Medical Association Declaration of Helsinki 2013.

STATISTICAL ANALYSIS:

Numerical data were expressed as either median with corresponding ranges or as mean with standard deviations (SD). Continuous variables underwent comparisons via the t test or analysis of variance (ANOVA), while categorical variables were assessed using chi-square tests. Survival curves were constructed using the Kaplan-Meier method and compared via log-rank tests. Statistical analyses were performed using SPSS version 22 (IBM, New York, NY, USA), with statistical significance denoted by P values <0.05.

Results

PATIENT DEMOGRAPHICS:

The 856 patients included in the present study were classified into 4 groups according to the GRWR values, with 54 (6.3%), 272 (31.8%), 274 (32.0%), and 256 (29.9%) patients having GRWRs <0.8%, 0.8–0.99%, 1.0–1.19%, and ≥1.2%, respectively. The clinicopathologic features of these 4 groups were similar, except for mean patient age and the proportions of patients who underwent ABO-incompatible LT (Table 1). Because these 2 factors are not considered significant risk factors for HCC recurrence, propensity score matching was not performed.

DISEASE-FREE AND OVERALL PATIENT SURVIVAL:

Out of the 856 patients included in the study, HCC recurrence was observed in 179 patients (20.9%), while 123 patients (14.4%) died due to various causes during a median follow-up duration of 79 months. Among the 123 patients who died, 98 died due to HCC recurrence-related reasons, while the remaining 25 died due to other causes. The 1-, 3-, 5-, and 10-year disease-free survival (DFS) rates were 88.0%, 81.7%, 79.1%, and 77.7%, respectively, whereas the 1-, 3-, 5-, and 10-year overall survival (OS) rates were 96.6%, 88.7%, 86.7%, and 84.2%, respectively (Figure 1).

DISEASE-FREE AND OVERALL PATIENT SURVIVAL ACCORDING TO GRWR:

Evaluation of the effects of GRWR on patient survival outcomes revealed that the 5-year DFS rates in patients with GRWRs <0.8%, 0.8–0.99%, 1.0–1.19%, and ≥1.2% did not differ significantly, being 83.5%, 76.7%, 79.6%, and 80.2%, respectively (P=0.545) (Figure 2). Similarly, there were no significant differences observed in the 5-year OS rates among these 4 groups, being 86.9%, 82.9%, 89.7%, and 87.2%, respectively (P=0.313) (Figure 2).

DISEASE-FREE AND OVERALL PATIENT SURVIVAL ACCORDING TO GRWR IN PATIENTS WITHIN AND BEYOND MILAN CRITERIA:

Of the 856 patients included in this study, 602 were within the Milan criteria and 254 were beyond the Milan criteria. Assessment of the 602 patients with HCC within the Milan criteria showed that the 5-year DFS rates of those with GRWRs <0.8%, 0.8–0.99%, 1.0–1.19%, and ≥1.2% were 92.0%, 86.8%, 85.6%, and 88.9%, respectively (P=0.398), and that their 5-year OS rates were 92.6%; 89.1%, 92.6%, and 91.0%, respectively (P=0.676) (Figure 3).

Assessment of the 254 patients with HCC beyond the Milan criteria showed that the 5-year DFS rates of those with GRWRs <0.8%, 0.8–0.99%, 1.0–1.19%, and ≥1.2% were 51.3%, 55.7%, 65.5%, and 56.9%, respectively, whereas their 5-year OS rates were 67.7%, 70.0%, 83.1%, and 77.0%, respectively (P=0.649) (Figure 4).

DISEASE-FREE AND OVERALL PATIENT SURVIVAL ACCORDING TO GRWR WITH ADV SCORE CUTOFF AT 5LOG:

Of the 856 patients included in this study, 661 had ADV scores <5log and 195 had ADV scores ≥5log. Assessment of the 661 patients with ADV scores <5log revealed that the 5-year DFS rates of those with GRWRs <0.8%, 0.8–0.99%, 1.0–1.19%, and ≥1.2% were 92.1%, 85.8%, 87.9%, and 86.6%, respectively (P=0.633); whereas their 5-year OS rates were 94.8%, 87.9%, 95.2%, and 91.6%, respectively (P=0.285) (Figure 5).

Assessment of the 195 patients with ADV scores ≥5log revealed that the 5-year DFS of those with GRWRs <0.8%, 0.8–0.99%, 1.0–1.19%, and ≥1.2% were 65.2%, 50.8%, 50.7%, and 51.2%, respectively (P=0.674), and that their 5-year OS rates were 71.1%, 68.6%, 70.9%, and 67.7%, respectively (P=0.906) (Figure 6).

Discussion

Forecasting the post-transplant prognosis in individuals who have received LT as a treatment for HCC poses a challenge, given the considerable variation in tumor burden, the heterogeneous nature of HCC tumor biology, and the potential impact of immunosuppressive therapy on outcomes. Previous studies have explored various factors to predict the likelihood of HCC recurrence after LT, including tumor size, number of tumors, vascular invasion, tumor differentiation, and levels of alpha-fetoprotein and des-γ-carboxy prothrombin [14–17]. Careful patient selection for LT has been suggested to mitigate the risk of post-transplant HCC recurrence [17–19].

Post-transplant HCC recurrence rates have been reported to be higher in patients who undergo LDLT than those who receive deceased donor LT [2,3], and one of the underlying causes of this difference was thought to be regeneration of the partial liver graft in LDLT, which enhances HCC recurrence [5]. A Korean single-center study that included 428 patients who underwent LDLT showed that GRWR <0.8% was associated with higher tumor recurrence rates in patients with HCC beyond, but not within, the Milan criteria [5]. Several animal studies have suggested that regeneration of the partial liver graft encourages HCC recurrence [8–10]. For instance, prior studies have indicated substantial activation of cellular signaling pathways associated with tumor invasion and migration in recipients of small-for-size liver grafts, potentially promoting increased tumor growth and metastasis following LT [8].

Other studies have reported that small-sized liver grafts do not raise the risk of HCC recurrence after LDLT [6,7]. A previous study from our center analyzed the influence of GRWR on HCC recurrence in 181 LDLT recipients with HCC within the University of California at San Francisco criteria. These patients were divided into 4 GRWR groups, consisting of those with low (<0.8%), middle (0.8–1.0%), and high (>1.0%) GRWR, and those who received whole liver grafts (GRWR >1.5%). A GRWR less than 0.8% was not a substantial risk factor for HCC recurrence, suggesting that a smaller living donor liver graft and the subsequent liver regeneration did not elevate the risk of HCC recurrence following LT, particularly when HCC satisfied the Milan or UCSF criteria [6]. An analysis of 295 patients in the China Liver Transplant Registry reported that small grafts, with GRWR ≤0.8%, can predict inferior survival outcomes but cannot predict earlier HCC recurrence [7].

The present study found that GRWR <0.8% was not prognostic for HCC recurrence or patient survival. Subgroup analyses also showed that GRWR <0.8% was not a significant risk factor for survival outcomes in patients within and beyond the Milan criteria, similar to findings in our previous study [6]. Subgroup analyses were also performed using an ADV score cutoff of 5log, which was shown to be more prognostic than the Milan criteria [12]. GRWR <0.8% was not found to be a significant risk factor for DFS or OS in patients with ADV scores <5log and ≥5log. Taken together, these findings indicate that GRWR <0.8% is not prognostic for post-transplant HCC recurrence. However, the proportion of patients with GRWR <0.8% was only 6.3% in the present study; thus, special attention should be paid to the interpretation of the results.

Although GRWR >0.8% is preferable, LDLT with GRWR between >0.6% and 0.8% has been performed if an alternative living donor is unavailable. This type of small-for-size graft has been associated with the development of small-for-size syndrome. Portal vein pressure was significantly higher in patients receiving small-sized liver grafts [20,21]. Although elevated portal venous pressure during the early post-transplant phase can trigger rapid liver regeneration, it has also been associated with graft dysfunction [9]. While multiple strategies have been employed to address early allograft dysfunction in LDLT, practices such as graft selection to ensure a GRWR exceeding 0.8%, facilitating complete venous drainage, and adjusting inflow have now become established standards in the field of LDLT [22,23]. Additional studies are required to evaluate the association between small-for-size syndrome and post-transplant HCC recurrence [24].

This study had several limitations, including its retrospective nature and the enrollment of patients from a single center in a region with a high prevalence of hepatitis B virus infection. Additionally, most of the HCC cases included in this study originated in livers infected with hepatitis B virus. Only 4 patients had very low GRWR (<0.7%) and the median GRWR in the group with GRWR <0.8% was 0.75%. Therefore, the prognostic effect of very low GRWR could not be investigated in this study.

Conclusions

The present high-volume study demonstrated that GRWR <0.8% did not have a significant prognostic impact on the oncological outcomes of patients undergoing LT for HCC. It is necessary to perform further high-volume multi-center studies to validate the findings of the present study.