This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Non-alcoholic steatohepatitis (NASH)-related hepatocellular carcinoma (HCC) has become the second leading cause of HCC-related liver transplantation in the United States. This study investigated post-transplant recurrence and survival for patients transplanted for NASH-related HCC compared to non-NASH HCC etiologies. Retrospective review of the United Network for Organ Sharing (UNOS) Organ Procurement and Transplantation Network (OPTN) database identified 7,461 patients with HCC—1,405 with underlying NASH and 6,086 with non-NASH underlying diseases. After propensity score matching (PSM) to account for patient- and tumor-related confounders 1,175 remained in each group. Primary outcomes assessed were recurrence rate and recurrence-free survival. Recurrent malignancy at 5 years post-transplant was lower in NASH compared to non-NASH patients (5.80 vs. 9.41%,
Hepatocellular carcinoma (HCC) accounts for the fourth most cancer-related deaths in the United States (US) (
HCC develops through progressive hepatocellular inflammation, leading to fibrosis, cell death, and aberrant regeneration which results in tumor formation (
Owing to the underlying metabolic syndrome often associated with NASH, these patients carry higher rates of concomitant cardiovascular and endocrine comorbidities than non-NASH ESLD population (
We performed a retrospective review of the Organ Procurement and Transplantation (OPTN) database for all adult (≥18-year-old) deceased donor liver transplant recipients in the United States diagnosed with HCC in the setting of known underlying liver disease. Our study population included transplants from 4 November 2012 to 6 December 2020, with the initiation date coinciding to the date OPTN began tracking tumor characteristics on transplant hepatectomy specimens. Recipients were first classified by diagnosis of NASH (NASH: 1,405, non-NASH: 6,086;
Study design. NASH transplant recipients with HCC were first compared to non-NASH recipients with HCC. These patients were then propensity matched and further compared. Additional analysis was performed on the unmatched populations to compare with post-transplant HCC recurrence and post-recurrence survival between NASH and non-NASH populations.
We first set out to assess post-transplant HCC recurrence rate in NASH vs non-NASH patients. We defined recurrence rate as a post-transplant HCC-related death or a diagnosis of HCC recurrence, derived from a validation study showing reliability of HCC recurrence data in the UNOS OPTN database (
As most cases of recurrent HCC occur within 5 years (
We then assessed survival patterns in NASH and non-NASH patients who developed post-transplant recurrence. Here, patients with recurrent HCC after transplant were again divided by underlying diagnosis (NASH: 52, non-NASH: 365). Baseline recipient, donor, and transplant characteristics were compared, as were tumor characteristics on transplant hepatectomy. The primary outcome assessed was survival after recurrence.
To evaluate differences between NASH and non-NASH patient cohorts’ overall survival after transplant with and without recurrence, and to verify any trends seen only in the recurrence population, overall survival was reported in all four of those subgroups.
Continuous variables were evaluated for normality using the Shapiro Wilk test. Non-normally distributed variables were compared with a Wilcoxon rank-sum test and were represented as median interquartile range (IQR). Categorical variables were compared using a chi-square or Fisher’s exact test and were represented as numbers (percentage of population).
PSM of non-NASH to NASH patients was completed using 1:1 nearest-neighbor matching with a caliper width of 0.2. Covariates matched in propensity score models were identified
Post-transplant HCC recurrence rates were assessed using a competing risk-regression model with non-cancer-related death used as a competing outcome. Cumulative incidence of HCC recurrence was evaluated using Fine-Gray proportional sub distribution hazard ratio (SHR) models in NASH and non-NASH recipients. Post-transplant survival and survival after diagnosis of recurrence, as defined above, were reported via Kaplan-Meier curves with statistical significance assessed using Log-rank tests. Recurrence rates were compared using Cox Proportional Hazard regression modeling. These data were remained unadjusted as attempts at adjusted analyses yielded underpowered results. `For all comparisons two-sided statistical significance was set
Prior to propensity matching, 1,405 patients had NASH-related HCC compared to 6,086 with non-NASH diagnoses (
PSM resulted in 1,175 matched pairs with largely similar profiles (
Propensity score matched baseline characteristics between NASH and non-NASH recipients with HCC.
NASH | Non-NASH |
|
|
---|---|---|---|
Number | 1,175 | 1,175 | |
Median followup (days) | 1,070 (382–1,809) | 1,243 (688–1,903) | |
Recipient characteristics | |||
Age | 64 (60–68) | 64 (60–67) | 0.55 |
Female sex | 378 (32.17%) | 384 (32.68%) | 0.83 |
Ethnicity | 0.67 | ||
White | 882 (75.06%) | 865 (73.62%) | |
Black | 11 (0.94%) | 14 (1.19%) | |
Other | 282 (24.00%) | 296 (25.19%) | |
BMI | 31.79 (28.20–35.53) | 27.75 (24.64–31.66) | <0.01 |
Pre-exception MELD | 12 (9–16) | 12 (9–16) | 0.48 |
AFP | 0.98 | ||
<100 ng/ml | 1,097 (93.36%) | 1,094 (93.11%) | |
100–399 ng/ml | 62 (5.28%) | 65 (5.53%) | |
≥400 ng/ml | 16 (1.36%) | 16 (1.36%) | |
Locoregional therapy | |||
TACE | 752 (64.00%) | 759 (64.60%) | 0.79 |
TARE | 132 (11.23%) | 140 (11.91%) | 0.65 |
Ablation | 384 (32.68%) | 365 (31.06%) | 0.43 |
Other | 11 (0.94%) | 13 (1.11%) | 0.84 |
Number of locoregional treatments | 0.69 | ||
0 | 126 (10.72%) | 116 (9.87%) | |
1 | 727 (61.87%) | 747 (63.57%) | |
2 | 254 (21.62%) | 238 (20.26%) | |
≥3 | 68 (5.79%) | 74 (6.30%) | |
Disabled functional status | 165 (14.04%) | 184 (15.66%) | 0.29 |
Diabetes mellitus | 818 (71.57%) | 328 (28.20%) | <0.01 |
Portal vein thrombosis | 189 (16.11%) | 203 (17.32%) | 0.44 |
Hemodialysis | 10 (0.85%) | 19 (1.62%) | 0.13 |
Previous abdominal surgery | 626 (53.28%) | 610 (51.91%) | 0.53 |
Multiorgan | 20 (1.70%) | 23 (1.96%) | 0.76 |
Primary diagnosis | — | ||
NASH | 1,175 (100.00%) | 0 (0.0%) | |
HCV | 0 (0.0%) | 63 (5.41%) | |
HBV | 0 (0.0%) | 759 (65.15%) | |
EtOH | 0 (0.0%) | 251 (21.55%) | |
Other |
0 (0.0%) | 92 (7.90%) | |
Donor characteristics | |||
Age | 46 (30–58) | 45 (31–59) | 0.80 |
Female sex | 492 (41.87%) | 499 (42.47%) | 0.80 |
BMI | 27.46 (23.74–32.34) | 27.65 (23.56–31.96) | 0.76 |
Diabetes mellitus | 159 (13.53%) | 165 (14.04%) | 0.76 |
Macrosteatosis (%) | 5 (0–10) | 5 (0–10) | 0.08 |
Inotrope support | 566 (48.17%) | 556 (47.32%) | 0.71 |
LDRI | 1.58 (1.28–1.92) | 1.60 (1.28–1.94) | 0.22 |
Cause of death | 0.36 | ||
Anoxia | 420 (35.74%) | 459 (39.06%) | |
CVA | 391 (33.28%) | 391 (33.28%) | |
Head trauma | 337 (28.68%) | 302 (25.70%) | |
CNS tumor | 8 (0.68%) | 5 (0.43%) | |
Other | 19 (1.62%) | 18 (1.53%) | |
DCD | 84 (7.15%) | 83 (7.06%) | 0.99 |
Transplant details | |||
CIT (hours) | 5.90 (4.60–7.25) | 5.93 (4.50–7.55) | 0.43 |
Values are listed as number (percentage) or median ± interquartile range unless otherwise stated. BMI, body mass index; NASH, non-alcoholic steatohepatitis; AFP, alpha fetoprotein; TACE, transarterial chemoembolization; TARE, transarterial radioembolization; HCV, Hepatitis C Virus; EtOH, alcohol; CVA, cerebrovascular accident; LDRI, Liver Donor Risk Index; CNS, central nervous system; DCD, donation after cardiac death; CIT, cold ischemia time.
Includes metabolic, autoimmune and cholestatic diseases.
Propensity score matched tumor characteristics in transplant hepatectomy specimens.
NASH | Non-NASH |
|
|
---|---|---|---|
Number | 1,175 | 1,175 | |
No tumor on explant | 71 (6.04%) | 76 (6.47%) | 0.73 |
Number of tumors | 0.83 | ||
1 | 548 (46.64%) | 524 (44.60%) | |
2 | 268 (22.81%) | 269 (22.89%) | |
3 | 128 (10.89%) | 130 (11.06%) | |
≥4 | 160 (13.62%) | 176 (14.98%) | |
Largest tumor size (cm) | 2.5 (1.5–3.5) | 2.4 (1.5–3.5) | 0.59 |
Tumor differentiation |
0.75 | ||
Complete necrosis | 296 (25.19%) | 276 (23.49%) | |
Well | 274 (23.32%) | 270 (22.98%) | |
Moderate | 532 (45.28%) | 555 (47.23%) | |
Poor | 73 (6.21%) | 74 (6.30%) | |
Vascular invasion | 0.86 | ||
Microvascular | 125 (10.64%) | 134 (11.40%) | |
Macrovascular | 21 (1.79%) | 21 (1.79%) | |
Satellite lesions | 59 (5.02%) | 61 (5.19%) | 0.93 |
Values are listed as number (percentage) or median ± interquartile range unless otherwise stated.
Differentiation of worst tumor.
Comparing NASH to non-NASH transplant recipients, we observed reduced post-transplant HCC recurrence rate in NASH patients. After PSM, recurrence rates at 5 years were 5.80% in the NASH group and 9.41% in non-NASH patients (SHR: 0.61, 95% CI: 0.42–0.89,
Propensity score matched transplant outcomes by diagnosis of NASH.
NASH | Non-NASH | HR/SHR | 95% CI |
|
|
---|---|---|---|---|---|
Number | 1,175 | 1,175 | |||
Acute Rejection within 1 year | 77 (8.85%) | 62 (7.17%) | — | — | 0.78 |
Recurrent Malignancy | (SHR) | ||||
5-year | 5.80% | 9.41% | 0.61 | 0.42–0.89 | 0.01 |
Median time to recurrence |
426 (213–752) | 400 (195–796) | — | — | 0.59 |
Post-transplant survival | (HR) | ||||
Overall | — | — | 0.87 | 0.71–1.07 | 0.20 |
1-year | 92.98% | 94.06% | — | — | 0.32 |
3-year | 86.35% | 84.34% | — | — | 0.38 |
5-year | 80.71% | 78.40% | — | — | 0.30 |
Values are listed as percent, number (percentage) or median ± interquartile range unless otherwise stated.
For patients with recurrent HCC only.
Cumulative incidence of post-transplant HCC recurrence
We next assessed only patients with recurrent HCC after transplant. In this cohort, median follow-up for NASH patients was 2,059 days (IQR: 1,003–2,157) and 2,132 days (IQR: 1,445–2,409) for non-NASH patients. As shown in
Baseline characteristics in NASH and non-NASH recipients with HCC recurrence after transplant.
NASH | Non-NASH |
|
|
---|---|---|---|
Patients with recurrent HCC | 52 | 365 | |
Median followup (days) | 2,058 (1,002–2,156) | 2,133 (1,444–2,503) | |
Recipient characteristics | |||
Age | 65 (62–67) | 61 (57–65) | <0.01 |
Female sex | 19 (36.54%) | 64 (17.53%) | <0.01 |
Ethnicity | 0.01 | ||
White | 37 (71.15%) | 232 (63.56%) | |
Black | 0 (0.0%) | 50 (13.70%) | |
Other | 15 (28.85%) | 83 (22.74%) | |
BMI | 32.39 (29.21–35.39) | 27.40 (24.27–31.32) | <0.01 |
Pre-exception MELD | 12 (9–16) | 11 (8–15) | 0.66 |
AFP | 0.65 | ||
<100 ng/ml | 41 (80.39%) | 271 (75.70%) | |
100-399 ng/ml | 6 (11.76%) | 61 (17.04%) | |
≥400 ng/ml | 4 (7.84%) | 26 (7.26%) | |
Locoregional therapy | |||
TACE | 38 (73.08%) | 264 (72.33%) | 0.99 |
TARE | 6 (11.54%) | 24 (6.58%) | 0.24 |
Ablation | 17 (32.69%) | 98 (26.85%) | 0.41 |
Other | 0 (0.00%) | 4 (1.10%) | 0.99 |
Number of locoregional treatments | 0.99 | ||
0 | 6 (11.54%) | 41 (11.24%) | |
1 | 27 (51.92%) | 194 (53.15%) | |
2 | 14 (26.92%) | 94 (25.75%) | |
≥3 | 5 (9.62%) | 36 (9.86%) | |
Disabled functional status | 6 (11.54%) | 60 (16.44%) | 0.42 |
Diabetes mellitus | 31 (62.00%) | 96 (26.52%) | <0.01 |
Portal vein thrombus | 13 (25.00%) | 45 (12.36%) | 0.02 |
Hemodialysis | 0 (0.00%) | 6 (1.64%) | 0.99 |
Previous abdominal surgery | 22 (42.31%) | 154 (42.19%) | 0.99 |
Multiorgan recipient | 0 (0.00%) | 7 (1.92%) | 0.60 |
Primary diagnosis | — | ||
NASH | 52 (100.00%) | 0 (0.0%) | |
HCV | 0 (0.0%) | 245 (67.68%) | |
HBV | 0 (0.0%) | 18 (4.97%) | |
EtOH | 0 (0.0%) | 83 (22.93%) | |
Other |
0 (0.0%) | 16 (4.42%) | |
Donor characteristics | |||
Age | 42 (26−56) | 44 (30−56) | 0.73 |
Female sex | 22 (42.31%) | 151 (41.37%) | 0.99 |
BMI | 27.23 (23.99–31.65) | 27.27 (23.13–31.44) | 0.71 |
Diabetes mellitus | 5 (9.62%) | 47 (12.88%) | 0.66 |
Macrosteatosis | 5 (5–18) | 5 (0–10) | 0.06 |
Inotrope support | 26 (50.00%) | 182 (49.86%) | 0.99 |
LDRI | 1.53 (1.23–1.87) | 1.54 (1.27–1.87) | 0.83 |
Cause of death | 0.98 | ||
Anoxia | 20 (38.46%) | 132 (36.16%) | |
CVA | 18 (34.62%) | 127 (34.79%) | |
Head trauma | 14 (26.92%) | 100 (27.40%) | |
CNS tumor | 0 (0.00%) | 2 (0.55%) | |
Other | 0 (0.001%) | 4 (1.10%) | |
DCD | 5 (9.62%) | 25 (6.85%) | 0.40 |
Transplant details | |||
CIT (hours) | 6.05 (4.25–8.26) | 5.95 (4.66–7.58) | 0.58 |
Values are listed as number (percentage) or median ± interquartile range unless otherwise stated.
BMI, body mass index; NASH, non-alcoholic steatohepatitis; AFP, alpha fetoprotein; TACE, transarterial chemoembolization; TARE, transarterial radioembolization; HCV, Hepatitis C Virus; EtOH, alcohol; CVA, cerebrovascular accident; LDRI, liver donor risk index; CNS, central nervous system; DCD, donation after cardiac death; CIT, cold ischemia time.
Includes metabolic, autoimmune and cholestatic diseases.
Tumor characteristics in transplant hepatectomy specimens in patients with recurrent HCC after transplant.
NASH | Non-NASH |
|
|
---|---|---|---|
Patients with recurrent HCC | 52 | 365 | |
No tumor on explant | 0 (0.00%) | 9 (2.47%) | 0.61 |
Number of tumors | 0.13 | ||
1 | 21 (40.38%) | 138 (37.81%) | |
2 | 8 (15.38%) | 80 (21.92%) | |
3 | 11 (21.15%) | 35 (9.59%) | |
≥4 | 12 (23.08%) | 103 (28.22%) | |
Largest tumor size (cm) | 3.2 (2.1–4.6) | 2.8 (1.7–4.3) | 0.09 |
Tumor differentiation |
0.50 | ||
Complete necrosis | 6 (11.54%) | 39 (10.68%) | |
Well | 4 (7.69%) | 48 (13.15%) | |
Moderate | 29 (53.85%) | 207 (56.71%) | |
Poor | 14 (26.92%) | 71 (19.45%) | |
Vascular invasion | 0.11 | ||
Microvascular | 12 (23.08%) | 113 (30.96%) | |
Macrovascular | 6 (11.54%) | 18 (4.93%) | |
Satellite lesions | 5 (9.62%) | 38 (10.41%) | 0.99 |
Values are listed as number (percentage) or median ± interquartile range unless otherwise stated.
Differentiation of worst tumor.
We then compared outcomes in patients with recurrent malignancy. Here, we found no statistically significant differences in survival from time of recurrence in NASH compared to non-NASH patients (
Kaplan-Meier curves comparing survival following recurrence in NASH vs. non-NASH patients.
Outcomes in patients with recurrent HCC after transplant by diagnosis of NASH.
NASH | Non-NASH | HR | 95% CI |
|
|
---|---|---|---|---|---|
Patients with recurrent HCC | 52 | 365 | |||
Median time to death after recurrence (days) |
150 (73–375) | 227 (97–484) | — | — | 0.05 |
Survival after recurrence | |||||
Overall | — | — | 1.06 | 0.73–1.53 | 0.75 |
6 months | 53.99% | 67.02% | — | — | 0.10 |
1 year | 45.95% | 46.71% | — | — | 0.63 |
18 months | 29.03% | 34.43% | — | — | 0.45 |
Values are listed as percent, number (percentage) or median ± interquartile range unless otherwise stated.
For mortalities only.
In this study we compared NASH-related and non-NASH HCC transplant populations, specifically looking at recurrence rates as well as survival post recurrence. NASH patients were found to have a lower HCC recurrence rate at 5 years while post-transplant survival remained similar between the two groups.
Previous studies comparing NASH to non-NASH populations have provided conflicting results to date with regards to HCC outcomes. Billeter et al. utilized propensity-score matching to compare NASH-related and non-NASH HCC patients in 34 NASH patients receiving liver resection in a single institution and found no differences in 1-, 3-, or 5-year recurrence-free survival (
Understanding the biology of HCC in NASH-related and non-NASH patients is critical to understanding tumor behavior as well as response to transplantation and adjuvant treatment modalities. Unlike HCC secondary to non-NASH diseases, NASH-related HCC pathogenesis is uniquely affected by a cascade of insulin resistance which causes oxidative stress, inflammation, and fibrosis-stimulating cytokines (
Another important difference between NASH-related and non-NASH patients are tumor characteristics at time of surgical treatment. Utilizing the UNOS OPTN database, Lewin et. al. found that NASH patients receiving liver transplantation for HCC were less likely to have tumors with vascular invasion and/or poor differentiation upon explant and were less likely to have evidence of metastasis compared to other HCC etiologies (
While we observed lower recurrence rates in NASH HCC patients, those who did recur had shorter median survival than non-NASH patients. Some emerging data may help explain that by highlighting differences in NASH-related HCC response to adjuvant therapies. Locoregional therapy, namely TACE, has been shown to have lower complete response, more progression of disease, higher rates of residual disease, and more recurrence in 1-2-month follow-up imaging in the obese population (
Our study suffers several limitations which include but are not limited to the retrospective nature of a large, federally maintained database. It should be noted that HCC outcomes in this database lack granular details regarding some tumor and treatment characteristics. A recent study, however, showed that the UNOS OPTN observed HCC recurrence rate was not significantly lower than the expected rate, validating the use of the OPTN database in evaluating outcomes related to transplantation for HCC (
Currently, increased early detection of HCC and surgical treatment offers the best therapeutic opportunity for HCC patients with any etiology (
The original contributions presented in the study are included in the article/
The studies involving human participants were reviewed and approved by the Thomas Jefferson University Hospital Institutional Review Board. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.
RL, PA, and AB participated in research design, writing of the paper, performance of the research, and data analysis. KP, OS, AA, JC, WM, AF, CR, JG, AS, and HD participated in writing of the paper.
PA was supported by National Institutes of Health institutional training grant T32GM008562. HD was supported by the American Liver Foundation. This work was supported in part by Health Resources and Services Administration contract 234-2005-370011C.
The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at:
The propensity score match.
AFP, alpha feto-protein; BMI, body mass index; CIT, cold ischemia time; CNS, central nervous system; CVA, cerebrovascular accident; DAA, direct-acting antiviral; DCD, donation after cardiac death; ESLD, end-stage liver disease; EtOH, alcohol; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; ICU, intensive care unit; LDRI, liver donor risk index; MELD, model for end stage liver disease; NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis; OPTN, Organ Procurement and Transplantation Network; PSM, propensity score matched; PVT, portal vein thrombosis; TACE, trans-arterial chemoembolization; TARE, trans-arterial radioembolization; UNOS, United Network for Organ Sharing.