Abstract
Background
Socio-economic inequalities among different racial/ethnic groups have increased in many high-income countries. It is unclear, however, whether increasing socio-economic inequalities are associated with increasing differences in survival in liver transplant (LT) recipients.
Methods
Adults undergoing first time LT for hepatocellular carcinoma (HCC) between 2002 and 2017 recorded in the Scientific Registry of Transplant Recipients (SRTR) were included and grouped into three cohorts. Patient survival and graft survival stratified by race/ethnicity were compared among the cohorts using unadjusted and adjusted analyses.
Results
White/Caucasians comprised the largest group (n=9,006, 64.9%), followed by Hispanic/Latinos (n=2,018, 14.5%), Black/African Americans (n=1,379, 9.9%), Asians (n=1,265, 9.1%) and other ethnic/racial groups (n=188, 1.3%). Compared to Cohort I (2002-2007), the 5-year survival of Cohort III (2012-2017) increased by 18% for Black/African Americans, by 13% for Whites/Caucasians, by 10% for Hispanic/Latinos, by 9% for patients of other racial/ethnic groups and by 8% for Asians (All P values<0.05). Despite Black/African Americans experienced the highest survival improvement, their overall outcomes remained significantly lower than other ethnic∕racial groups (adjusted HR for death=1.20; 95%CI 1.05-1.36; P=0.005; adjusted HR for graft loss=1.21; 95%CI 1.08-1.37; P=0.002).
Conclusion
The survival gap between Black/African Americans and other ethnic/racial groups undergoing LT for HCC has significantly decreased over time. However, Black/African Americans continue to have the lowest survival among all racial/ethnic groups.
Abbreviations:
LT (Liver transplantation), HCC (Hepatocellular carcinoma), U.S. (United States), HR (Hazard Ratio), HCV (Hepatitis C Virus), SRTR (Scientific Registry of Transplant Recipients), HRSA (Health Resources and Services Administration), OPTN (Organ Procurement and Transplantation Network), STROBE (Strengthening of the Reporting of Observational Studies in Epidemiology), MELD (Model for End Stage Liver Disease), AFP (Alpha Feto Protein), BMI (Body Mass Index)Introduction
In a previous study,
1
we reported that among 6048 patients who underwent liver transplant (LT) for hepatocellular carcinoma (HCC) in the United States (U.S.) between 2002 and 2013, Black/African Americans had the lowest survival rate compared to patients belonging to other ethnic/racial groups. The five-year survival was 68% for White/Caucasians and 56% for Black/African Americans.- Molinari M.,A.S.
- Tsung A.
- Bou Samra P.
- Jonaissaint N.
Adult African Americans undergoing cadaveric liver transplantation for hepatocellular carcinoma within the Milan criteria have the lowest 5-year survival among all the ethnic groups in the United States: analysis of USA national data between January 2002 and June 2013.
Hepatoma Res. 2018; 4
1
After accounting for recipient and donor characteristics, the adjusted hazard ratio (aHR) for mortality was significantly higher for Black/African Americans in comparison to White/Caucasian patients (aHR = 1.52 (95% CI 1.28–1.80; P < 0.001).- Molinari M.,A.S.
- Tsung A.
- Bou Samra P.
- Jonaissaint N.
Adult African Americans undergoing cadaveric liver transplantation for hepatocellular carcinoma within the Milan criteria have the lowest 5-year survival among all the ethnic groups in the United States: analysis of USA national data between January 2002 and June 2013.
Hepatoma Res. 2018; 4
1
- Molinari M.,A.S.
- Tsung A.
- Bou Samra P.
- Jonaissaint N.
Adult African Americans undergoing cadaveric liver transplantation for hepatocellular carcinoma within the Milan criteria have the lowest 5-year survival among all the ethnic groups in the United States: analysis of USA national data between January 2002 and June 2013.
Hepatoma Res. 2018; 4
Other studies have described similar findings.
2
, 3
, 4
, 5
, 6
, 7
, 8
, 9
, 10
, - Molinari M.,A.S.
- Tsung A.
- Jorgensen D.
- Bou Samra P.
- Jonaissaint N.
Adult African Americans undergoing cadaveric liver transplantation for hepatocellular carcinoma within the Milan criteria have the lowest 5-year survival among all the ethnic groups in the United States: analysis of USA national data between January 2002 and June 2013.
Hepatoma Res. 2018; 4
11
These observations suggest that factors not associated with the burden of the disease have important effects on patient outcomes.2
,7
,11
, 12
, 13
, 14
, 15
, 16
, 17
, 18
, 19
, 20
, 21
, 22
During the last few decades, the interest in understanding these factors has grown at a rapid pace. Many studies have indicated that the level of education, the household income, the degree of support provided by family or friends and access to healthcare services are very important social determinants of health (SDH) that impact the quality of life and the longevity of patients, especially those affected by malignancies.2
,7
,11
, 12
, 13
, 14
, 15
, 16
, 17
, 18
, 19
, 20
, 21
, 22
, 23
, 24
In the U.S., unfavorable SDHs are more frequently found in ethnic/racial minorities. For example, Black/African Americans patients in need of solid organ transplantation are less likely referred for surgery.
7
,25
, 26
, 27
The lower-than-expected transplant rate in Black/African American patients might be due to the lower rate of healthcare insurance among members of this community.25
,26
Because adequate access to healthcare is a requirement for all patients approved for transplant surgery, irrespective of their ethnicity/race,28
,29
it is unclear why postoperative survival of Black/African American patients is consistently lower than patients from other ethnic/racial groups.2
,6
,27
,30
,31
Patients who undergo LT for HCC intersect the areas of solid organ transplantation and oncology. Due to the complexity of their care and the need for very close monitoring of their immunosuppression, LT patients have more frequent encounters with healthcare providers than patients treated with other modalities such as hepatic resections. Another important difference is that LT patients undergo a more stringent selection process that include proof of compliance to therapeutic recommendations, abstinence from alcohol or illicit drugs, adequate social support, and evidence of financial stability necessary to maintain immunosuppression therapies and cover the expenses of ongoing monitoring of drug levels and radiological tests. Since patients undergoing LT represents a very selected groups of patients, monitoring disparities in their outcomes provides a unique opportunity that is unmatched by other groups of patients who might face more barriers to access healthcare services.
Since the evidence that socio-economic inequalities are growing in many countries around the world,
32
, 33
, 34
we postulated that the survival difference described in previous studies between Black/African Americans and patients of other ethnic/racial groups after LT for HCC might have also increased. To test this hypothesis, we examined patient and graft survival after LT for HCC during three different periods. Since the introduction of direct antiviral agents (DAAs) for the treatment of viral hepatitis C (HCV) occurred during the study period, we performed sensitivity analysis to assess whether changes in postoperative survival were more pronounced in patients affected by HCV.Patients and methods
This study used data from the Scientific Registry of Transplant Recipients (SRTR). The SRTR data system includes data on all donors, wait-listed candidates, and transplant recipients in the U.S., submitted by the members of the Organ Procurement and Transplantation Network (OPTN). The Health Resources and Services Administration (HRSA), U.S. Department of Health and Human Services provides oversight to the activities of the OPTN and SRTR contractors. Due to the anonymity of the variables, the need for recipient individual consents was waived by the institutional review board that approved this study with protocol number 13060220. All the procedures and methods used to conduct this study were in compliance with the declaration of Helsinki on ethical principles for medical research involving human subjects.
35
This study is reported following the Strengthening of the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.36
Study population
Adults (age ≥ 18 years old) recipients of LT for HCC between March 1, 2002, and July 31, 2017, were candidates for this study. The start date of the data collection coincided with the introduction of the MELD score for the allocation of liver grafts in the U.S.
37
Patients who underwent redo LT, recipients of live donor or incompatible ABO blood group grafts, multi-visceral transplants or with the diagnosis of mixed HCC-cholangiocarcinoma or other non-HCC malignancies were excluded.38
Stratification of the study population
The study population was divided into five groups based on recipients' ethnicity/race as reported in the SRTR. Group I was represented by Whites/Caucasian recipients (n = 9006; 64.9%), Group II was represented by Black/African Americans (n = 1379; 9.9%), Group III was represented by Hispanic/Latinos (n = 2018; 14.5%), Group IV was represented by Asian/Pacific Islanders (n = 1265; 9.1%), and Group V was represented by recipients of other races/ethnicities (n = 188; 1.3%).
Cohorts
Based on the date of LT, participants were grouped into three cohorts with comparable duration of follow-up. Cohort I included patients transplanted between March 1, 2002 and February 28, 2007 (n = 2140; 15.4%). Cohort II included patients transplanted between March 1, 2007, and February 29, 2012 (n = 4692; 33.8%), and Cohort III included recipients operated between March 1, 2012, and July 31, 2017 (n = 7024; 50.6%). The date of the last follow-up was December 31, 2019.
Data collection and definitions
Parameters related to recipients' demographic characteristics, oncological factors, clinical characteristics, and cold ischemia time were analyzed. Recipients' characteristics were age, sex, body mass index (BMI), ethnicity/race, time on the waitlist, predisposing factor for HCC, the calculated MELD score at the time of transplant surgery.
39
Social determinants of heath were the highest degree of education, type of healthcare insurance and employment status. Oncological parameters were the maximum tumor size, the highest serum level of alpha-fetoprotein (AFP) before LT. Donors' characteristics included age, sex, ethnicity/racial group, and donor risk index.40
The MELD score was calculated using the formula MELD = 3.78 × ln[serum bilirubin (mg/dL)] + 11.2 × ln[INR] + 9.57 × ln[serum creatinine (mg/dL)] + 6.43
41
without exception points for HCC. BMI was calculated using the World Health Organization formula BMI = body weight (Kg)/height2 (m).42
Time-dependent variables were cold ischemia time, defined as the interval between the donor aorta cross-clamp and the time when the liver graft was removed from the cold storage during the index operation.World Health Organization. Obesity and overweight. [Accessed 2015 Jan]. Available from: URL: http//www.who.int/mediacentre/factsheets/fs311/en. [cited 2021 August 5].
Outcomes
The primary outcome was patient death. The secondary outcome was graft loss. Overall survival was calculated as the time between the date of LT and the date of death, the date of the last recorded follow-up, or the date of completion of the study. Graft survival was calculated as the time interval between the date of LT and the date of graft loss, or the date of relisting, or the date of re-transplantation or the date of death, whichever came first.
Statistical analysis
Continuous variables with normal distribution are reported using the mean and standard deviation (SD). Median and interquartile ranges (IQR) are used for variables with non-normal distribution. Frequency and percentages are used for categorical variables. Analysis of variance, chi-square, and Kruskal–Wallis tests were used for summary statistics, and all estimates were adjusted using Bonferroni correction for multiple comparisons.
43
The Kaplan–Meier method was used to estimate the proportion of patients who were alive after transplantation.
44
,45
The Log-rank test46
was used to compare survivals among cohorts or ethnic groups. Cox regression analysis was performed to test differences in patient and graft survivals among ethnic/racial groups after adjusting for recipient age, gender, MELD score, cold ischemia time, wait time, AFP, the maximum diameter of the largest tumor, era of transplantation, type of health insurance, highest level of education and employment status.Censoring was used when patients were alive at the last follow-up or the study's completion date. Censoring for graft survival was used when the graft was still functioning at the last follow-up or the date of completion of the study.
Group I (Whites/Caucasian patients) and Cohort 1 (March 1, 2002 to February 28, 2007) represented the references for all respective comparisons. All statistical analyses were completed using R 3.6.0 (R Foundation for Statistical Computing, Vienna, Austria), and SPSS (Statistics for Windows, version 26, SPSS Inc., Chicago, Ill, USA). All statistical analyses were two-tailed, and P-values <0.05 were considered significant.
Results
Characteristics of the study population
A total of 17,366 subjects were screened and 13,856 were included (Fig. 1). The demographic and the clinical characteristics of the entire population and of each ethnic/racial group are reported in Table 1. The mean age at transplantation was 59.3 years old (SD = 7.4), male sex represented 77.9% of the entire cohort, and the mean calculated MELD score at transplantation was 13.5 (SD = 7.3). The mean size of the largest tumor was 2.5 cm (SD = 1.3), and the median serum level of AFP was 13.0 ng/mL (IQR 6.0–52.0). A single HCC nodule was diagnosed in 57.3% of patients, two lesions were diagnosed in 19.6% of patients and three or more tumors were diagnosed in 8.2% of patients.
Figure 1Flow chart representing the number of patients screened, the number of patients excluded and the reasons for their exclusion. Among 17,366 potential candidates for this study, 3510 did not satisfy the inclusion criteria, leaving a total number of 13,856 recipients who represented the study population
Table 1Summary of the demographic and clinical characteristics of the study population, and of each racial/ethnic group
| Characteristics | All recipients | White/Caucasian | Black/African American | Hispanic/Latino | Asian | Other |
|---|---|---|---|---|---|---|
| n. 13,856 | n. 9006 (64.9%) | n. 1379 (9.9%) | n. 2018 (14.5%) | n. 1265 (9.1%) | n. 188 (1.3%) | |
| Recipient | ||||||
| Age (years), mean ± std | 59.34 ± 7.47 | 59.6 ± 7.1 | 58.9 ± 7.6 | 58.8 ± 7.7 | 58.9 ± 9.0 | 57.2 ± 8.1 |
| Male, n. (%) | 10,788 (77.9) | 7202 (80.0) | 1012 (73.4) | 1493 (74.0) | 946 (74.8) | 135 (71.8) |
| BMI | 28.37 ± 5.23 | 28.7 ± 5.2 | 28.4 ± 5.5 | 29.0 ± 5.1 | 24.7 ± 3.8 | 29.4 ± 5.5 |
| History of diabetes, n. (%) | 4134 (29.8) | 2568 (28.5) | 396 (28.7) | 756 (37.5) | 356 (28.1) | 58 (30.9) |
| Pre-transplant dialysis, n. (%) | 269 (1.9) | 175 (1.9) | 21 (1.5) | 56 (2.8) | 14 (1.1) | 3 (1.6) |
| Median time on waitlist (days) [IQR] | 141.00 [46.00, 290.50] | 129.0 [42.0, 267.0] | 127.0 [47.0, 278.0] | 189.5 [61.0, 362.0] | 192.0 [75.0, 373.0] | 130.5 [40.0, 277.0] |
| MELD score, mean ± std | 13.55 ± 7.30 | 13.6 ± 7.1 | 13.4 ± 7.0 | 14.8 ± 8.5 | 11.4 ± 6.3 | 14.8 ± 8.3 |
| Highest degree of education | ||||||
| Grade 1–8 | 800 (5.7) | 245 (2.7) | 32 (2.3) | 363 (17.9) | 155 (12.2) | 5 (2.7) |
| Grade 9–12 | 5625 (40.5) | 3609 (40.1) | 640 (46.4) | 929 (46.4) | 352 (27.8) | 95 (50.5) |
| Associate/bachelor degree | 2042 (14.7) | 1435 (15.9) | 160 (11.6) | 185 (9.1) | 245 (19.3) | 17 (9.0) |
| College/technical school | 3118 (22.5) | 2198 (24.4) | 345 (25.0) | 319 (15.8) | 209 (16.5) | 47 (25.0) |
| Post college/graduate degree | 894 (6.4) | 625 (6.9) | 55 (4.0) | 52 (2.5) | 156 (12.3) | 6 (3.2) |
| Unknown | 1377 (9.3) | 894 (9.9) | 147 (10.7) | 170 (8.4) | 148 (11.6) | 18 (9.6) |
| Primary healthcare source of payment | ||||||
| Private insurance | 7643 (55.2) | 5199 (57.7) | 699 (50.6) | 939 (46.6) | 709 (56.0) | 97 (51.5) |
| Public Insurance | 5687 (41.0) | 3456 (38.3) | 625 (45.3) | 1010 (50.1) | 516 (40.7) | 73 (38.8) |
| Self paid | 67 (0.5) | 31 (0.3) | 1 (0.0) | 5 (0.2) | 30 (2.3) | 0 (0.0) |
| Foreign government | 57 (0.4) | 54 (0.5) | 0 (0.0) | 2 (0.1) | 0 (0.0) | 1 (0.5) |
| Donation/free care | 17 (0.1) | 8 (0.0) | 1 (0.0) | 4 (0.2) | 3 (0.2) | 1 (0.5) |
| Department of veteran affairs | 385 (2.8) | 257 (2.8) | 52 (3.7) | 54 (2.7) | 6 (0.4) | 16 (8.5) |
| Other | 0 | 1 (0.0) | 1 (0.0) | 4 (0.2) | 1 (0.0) | 0 (0.0) |
| Employment status | ||||||
| On disability | 3026 (21.8) | 1889 (20.9) | 353 (25.5) | 509 (25.2) | 226 (17.8) | 49 (26.0) |
| Student | 7 (0.0) | 5 (0.0) | 0 (0.0) | 1 (0.0) | 1 (0.0) | 0 (0.0) |
| Homemaker | 125 (0.0) | 67 (0.0) | 3 (0.0) | 23 (1.1) | 31 (2.4) | 1 (0.5) |
| Working with income | 3347 (24.1) | 2267 (25.1) | 316 (22.9) | 362 (17.9) | 367 (29.0) | 35 (18.6) |
| Retired | 1617 (11.6) | 1097 (12.1) | 118 (8.5) | 204 (10.1) | 176 (13.9) | 22 (11.7) |
| Unknown | 5734 (41.3) | 3681 (40.8) | 589 (42.7) | 919 (45.5) | 464 (36.6) | 81 (43.0) |
| Viral hepatitis C | ||||||
| Positive | 8580 (61.9) | 5721 (63.5) | 1093 (79.3) | 1261 (62.5) | 383 (30.3) | 122 (64.9) |
| Negative | 4677 (33.8) | 2879 (32.0) | 238 (17.3) | 681 (33.7) | 819 (64.7) | 60 (31.9) |
| Undetermined | 299 (2.2) | 189 (2.1) | 22 (1.6) | 42 (2.1) | 41 (3.2) | 5 (2.7) |
| Unknown | 282 (2.0) | 206 (2.3) | 25 (1.8) | 32 (1.6) | 18 (1.4) | 1 (0.5) |
| Tumor | ||||||
| Diameter of the largest tumor (cm), mean ± std | 2.53 ± 1.37 | 2.5 ± 1.4 | 2.6 ± 1.3 | 2.6 ± 1.4 | 2.5 ± 1.3 | 2.5 ± 1.4 |
| Maximum number of tumors, n. (%) | ||||||
| 0 | 549 (3.9) | 334 (3.7) | 45 (3.3) | 87 (4.3) | 74 (5.9) | 9 (4.8) |
| 1 | 7936 (57.3) | 5207 (57.8) | 794 (57.6) | 1103 (54.7) | 722 (57.1) | 110 (58.5) |
| 2 | 2716 (19.6) | 1756 (19.5) | 278 (20.2) | 385 (19.1) | 265 (21.0) | 32 (17.0) |
| >2 | 1135 (8.2) | 739 (8.2) | 109 (7.9) | 181 (9.0) | 98 (7.7) | 8 (4.3) |
| Missing | 1520 (11.0) | 970 (10.8) | 153 (11.1) | 262 (13.0) | 106 (8.4) | 29 (15.4) |
| Serum alpha fetoprotein (ng/mL) median [IQR] | 13.00 [6.00, 52.00] | 12.0 [5.0, 42.0] | 28.0 [10.0, 102.0] | 15.0 [6.0, 58.0] | 13.0 [5.0, 86.0] | 18.0 [7.0, 66.0] |
| Donor | ||||||
| Age (years), mean ± std | 43.26 ± 16.64 | 43.3 ± 16.4 | 42.5 ± 16.4 | 43.5 ± 16.9 | 44.2 ± 18.2 | 40.3 ± 17.4 |
| Male, n. (%) | 8232 (59.4) | 5526 (61.4) | 813 (59.0) | 1146 (56.8) | 652 (51.5) | 95 (50.5) |
| Race, n. (%) | ||||||
| White | 9066 (65.4) | 6210 (68.9) | 847 (61.4) | 1191 (59.0) | 685 (54.2) | 133 (70.7) |
| African American | 2385 (17.2) | 1535 (17.0) | 340 (24.7) | 283 (14.0) | 208 (16.4) | 19 (10.1) |
| Hispanic | 1849 (13.3) | 973 (10.8) | 150 (10.9) | 464 (23.0) | 239 (18.9) | 23 (12.2) |
| Asian | 416 (3.0) | 203 (2.3) | 29 (2.1) | 64 (3.2) | 115 (9.1) | 5 (2.7) |
| Other | 140 (1.0) | 85 (0.9) | 13 (0.9) | 16 (0.8) | 18 (1.42) | 8 (4.3) |
| Donor risk index, mean ± std | 1.6 ± 0.35 | 1.5 ± 0.3 | 1.8 ± 0.4 | 1.5 ± 0.3 | 1.6 ± 0.4 | 1.7 ± 0.4 |
| Operative | ||||||
| CIT (h), mean ± std | 6.6 ± 2.5 | 6.6 ± 2.5 | 6.4 ± 2.4 | 6.6 ± 2.5 | 6.8 ± 2.5 | 6.8 ± 2.6 |
| CIT (h), median [IQR] | 6.2 [4.9, 8.0] | 6.1 [4.9, 8.0] | 6.0 [4.7, 7.8] | 6.3 [4.8, 8.0] | 6.5 [5.0, 8.0] | 6.4 [5.0, 8.0] |
Bold: parameters with P value < 0.0001 when compared to the respective value of Caucasian/White liver transplant recipients (reference group). Abbreviations: BMI, body mass index; MELD, model for end-stage liver disease; IQR, interquartile range; CIT, cold ischemia time.
Compared to other ethnic/racial groups, Black/African American patients had the highest serum level of alpha feto-protein (28 ng/mL vs. 12.0–18.0), the highest prevalence of hepatitis C viral (HCV) infections (79.3% vs. 30.3–64.9%) and received grafts with the highest donor risk index (1.8 vs. 1.5–1.7).
Cohorts
The trends of the demographic and clinical characteristics of the study population during the three Cohorts are reported in Table 2. Patients in Cohort III were older (60.8 vs. 56.5 years), had higher BMIs (28.7 vs. 27.6), were more likely diabetic (32.8% vs. 23.7%), more likely on dialysis (2.6% vs. 0.6%), and had a longer waiting time (211 vs. 49 days) in comparison to patients in Cohort I. Similarly, the mean size of the tumors (3.0 vs. 2.3 cm), the serum level of AFP (20 vs. 11 ng/dL), and the proportion of patients with more than two lesions were significantly higher (9.3% vs. 6.7%) in Cohort III than in Cohort I.
Table 2Recipient, tumor, and donor characteristics of the study population by time of liver transplantation
| Characteristics | Cohort I | Cohort II | Cohort III |
|---|---|---|---|
| March 2002–Feb 2007 | March 2007–Feb 2012 | March 2012–July, 2017 | |
| n = 2140 (15.4%) | n = 4692 (33.9%) | N = 7024 (50.7%) | |
| Recipient | |||
| Age (years), mean ± std | 56.5 ± 8.4 | 58.4 ± 7.1 | 60.8 ± 7.0 |
| Male, n. (%) | 1704 (79.6) | 3675 (78.3) | 5409 (77.0) |
| BMI (kg/m2), mean ± std | 27.6 ± 4.9 | 28.2 ± 5.3 | 28.7 ± 5.3 |
| History of diabetes, n. (%) | 507 (23.7) | 1326 (28.3) | 2301 (32.8) |
| Pre-transplant dialysis, n. (%) | 12 (0.6) | 76 (1.6) | 181 (2.6) |
| Median time on waitlist (days) [IQR] | 49.5 [18.00, 138.0] | 104.0 [37.0, 225.0] | 211.0 [93.0, 379.0] |
| MELD score, mean ± std | 13.0 ± 6.2 | 13.6 ± 7.1 | 13.7 ± 7.7 |
| Ethnicity/race, n. (%) | |||
| White | 1354 (63.3) | 3008 (64.1) | 4644 (66.1) |
| Black/African American | 182 (8.5) | 450 (9.6) | 747 (10.6) |
| Hispanic/Latino | 303 (14.2) | 668 (14.2) | 1047 (14.9) |
| Asian | 263 (12.3) | 498 (10.6) | 504 (7.2) |
| Other | 38 (1.8) | 68 (1.5) | 82 (1.2) |
| Highest degree of education | |||
| Grade 1–8 | 110 (5.1) | 272 (5.7) | 418 (5.9) |
| Grade 9–12 | 771 (36.0) | 1807 (38.5) | 3407 (43.3) |
| Associate/bachelor degree | 273 (12.7) | 655 (13.9) | 1114 (15.8) |
| College/technical school | 436 (20.3) | 1036 (22.0) | 1646 (23.4) |
| Post college/graduate degree | 143 (6.6) | 270 (5.7) | 481 (6.8) |
| Unknown | 407 (19.0) | 652 (13.8) | 318 (4.5) |
| Primary healthcare source of payment | |||
| Private insurance | 1343 (62.7) | 2741 (58.4) | 3559 (50.6) |
| Public Insurance | 689 (32.1) | 1758 (37.5) | 3233 (46.0) |
| Self paid | 37 (1.7) | 14 (0.3) | 16 (0.2) |
| Foreign government | 7 (0.3) | 20 (0.4) | 30 (0.4) |
| Donation/free care | 5 (0.2) | 11 (0.2) | 1 (0.0) |
| Department of veteran affairs | 52 (2.4) | 148 (3.2) | 185 (2.6) |
| Other | 7 (0.3) | 0 | 0 |
| Employment status | |||
| On disability | 461 (21.5) | 1411 (30.0) | 1154 (16.4) |
| Student | 1 (0.0) | 3 (0.0) | 3 (0.0) |
| Homemaker | 19 (0.8) | 47 (1.0) | 59 (0.8) |
| Working with income | 395 (18.4) | 1214 (25.8) | 1738 (24.7) |
| Retired | 211 (9.8) | 713 (15.1) | 693 (9.8) |
| Unknown | 1053 (49.2) | 1304 (27.7) | 3377 (48.0) |
| Viral hepatitis C | |||
| Positive | 1230 (57.5) | 2994 (63.8) | 4356 (62.0) |
| Negative | 657 (30.7) | 1475 (31.4) | 2545 (36.2) |
| Undetermined | 168 (7.9) | 75 (1.6) | 56 (0.8) |
| Unknown | 67 (3.1) | 148 (3.2) | 67 (1.0) |
| Tumor | |||
| Diameter of the largest tumor (mm), mean ± std | 3.0 ± 1.2 | 2.7 ± 1.3 | 2.3 ± 1.4 |
| Max number of tumors, n. (%) | |||
| 0 | 10 (0.5) | 283 (6.0) | 256 (3.6) |
| 1 | 1164 (54.4) | 2436 (51.9) | 4336 (61.7) |
| 2 | 438 (20.5) | 989 (21.1) | 1289 (18.4) |
| >2 | 199 (9.3) | 464 (9.89) | 171 (6.7) |
| Missing | 329 (15.4) | 519 (11.1) | 672 (9.6) |
| Alpha feto-protein, median [IQR] | 20.0 [7.0, 105.0] | 15.0 [6.0, 63.0] | 11.0 [5.0, 40.0] |
| Donor | |||
| Age (years), mean ± std | 42.9 ± 17.2 | 43.1 ± 16.7 | 43.5 ± 16.4 |
| Male, n. (%) | 1283 (60.0) | 2771 (59.1) | 4178 (59.5) |
| Race, n. (%) | |||
| White | 1354 (63.3) | 3088 (64.1) | 4644 (66.1) |
| Black/African American | 182 (8.5) | 450 (9.6) | 747 (10.6) |
| Hispanic/Latino | 303 (14.2) | 668 (14.2) | 1047 (14.9) |
| Asian | 263 (12.3) | 498 (10.6) | 504 (7.2) |
| Other | 38 (1.8) | 68 (1.4) | 82 (1.2) |
| Donor risk index | 1.6 ± 0.4 | 1.6 ± 0.4 | 1.5 ± 0.3 |
| Operative | |||
| CIT (h), mean ± std | 7.3 ± 2.7 | 6.7 ± 2.5 | 6.2 ± 2.3 |
| CIT (h), median [IQR] | 7.0 [5.5, 9.0] | 6.4 [5.0, 8.0] | 6.0 [4.6, 7.5] |
Bold: parameters with P value < 0.0001 when compared to the respective value of Caucasian/White liver transplant recipients (reference group). Abbreviations: BMI, body mass index; MELD, model for end-stage liver disease; IQR, interquartile range; CIT, cold ischemia time.
Patient and graft survival by ethnicity
The median follow-up of this study was 12.1 years (95% CI 11.4–12.9). The five-year patient survival was 84% for multi-racial or patients from other ethnic groups, 82% for Asians, 79% for Hispanic/Latinos, 76% for White/Caucasians and 71% for Black/African Americans (log-rank P < 0.001). Five-year graft survival was 81% for multi-racial or patients from other ethnic groups, 80% for Asians, 75% for Hispanic/Latinos, 73% for White/Caucasians and 67% for Black/African Americans (log-rank P < 0.001) (Fig. 2). After adjusting for age, gender, serum AFP, tumor size, number of tumors, waiting time, BMI, history of diabetes, need for preoperative dialysis, MELD score, donor risk index and year of transplantation (cohort effect), employment status, highest degree of education, and type of health insurance, both patient and graft survival of African American/Black patients was significantly lower than White/Caucasian patients. The adjusted hazard ratio for mortality of Black/African American patients was 1.15 (95% CI 1.03–1.29) (P = 0.017), and the adjusted hazard ration for graft survival was 1.18 (95% CI 1.06–1.31) (P = 0.003).
Figure 2Kaplan–Meier functions of patient and graft survival after liver transplantation for hepatocellular carcinoma in the United States between March 1, 2002, and June 30, 2017, stratified by ethnic/racial group
Patient and graft survival by cohorts
Patient and graft survival improved during the study period irrespective of ethnic/racial groups. The five-year patient survival was 71% for Cohort I, 73% for Cohort II, and 88% for Cohort III (log-rank P < 0.001). The five-year graft survival was 66% for Cohort I, 70% for Cohort II and 80% for Cohort III (Log-rank P < 0.001).
The Kaplan–Meier curves and the unadjusted hazard ratios for patient and graft survivals stratified by ethnic/racial group and cohort are reported in Figure 3, Figure 4 respectively.
Figure 3Kaplan–Meier functions of the survival of patients undergoing liver transplantation for hepatocellular carcinoma stratified by ethnic/racial group and by cohort (Panel a). The reference group for the unadjusted hazard ratios were White/Caucasian patients transplanted during the period between March 1, 2002 and February 28, 2007 (Cohort I) (Panel b)
Figure 4Kaplan–Meier functions of graft survival after liver transplantation for hepatocellular carcinoma stratified by ethnic/racial group and by cohort (Panel a). The reference group for the unadjusted hazard ratios was grafts transplanted into White/Caucasian patients during the period between March 1, 2002 and February 28, 2007 (Cohort I) (Panel b)
Among all the racial/ethnic groups, Black/African Americans experienced the most significant improvement in both patient and grafts survival. Compared to Cohort I, the survival of Black/African Americans improved by 18% (60% → 78%) while the survival of White/Caucasians improved by 13% (69% → 82%). During the same period, the survival of Hispanic/Latinos improved by 10% (73% → 83%), of Asians by 8% (78% → 86%), and of patients of other ethnic/racial groups by 8% (79% → 87%). The graft survival of African American/Black patients improved by 19% (56% → 75%) while the graft survival of White/Caucasians improved by 15% (64% → 79%), of Hispanic/Latinos by 12% (70% → 82%), of Asians by 10% (76% → 86%), and of patients of other ethnic/racial groups by 16% (70% → 86%) (Fig. 5).
Figure 5Changes in 5-year patient and graft survival rates after liver transplantation for hepatocellular carcinoma stratified by time-cohorts and ethnicity/racial groups
Sensitivity analysis for hepatitis C virus status
In 2011, direct antiviral agents (DAAs) were approved by the US Food and Drug Administration (FDA) for the treatment of patients with HCV. As the prevalence of HCV infection was different among ethnic/racial groups, we performed a sensitivity analysis to assess whether the longitudinal improvements observed in patient and graft survival was dependent on HCV status. Between Cohorts I and III, patient five-year survival improved by 8% (76% → 84%) for patients without HCV and by 12% (69% → 81%) for patients with HCV (Log-rank P < 0.001). During the same period, five-year graft survival improved by 15% (64% → 79%) for patients with HCV and by 10% (71% → 81%) for patients without HCV (Log-rank P < 0.001). Changes in patient and graft survival stratified by HCV status and race/ethnicity confirmed that HCV positive Black/African American patients experienced the most pronounced survival increment among all the ethnic/racial groups (Fig. 6). There was a statistically significant positive interaction between ethnicity/racial group and HCV status for both patient survival (P = 0.03) and graft survival (P = 0.02) suggesting that the effects on survival was different among racial/ethnic groups based on HCV status.
Figure 6Changes in 5-year patient and graft survival rates after liver transplantation for hepatocellular carcinoma stratified by viral hepatitis C status (HCV), time-cohorts, and ethnicity/racial groups
Discussion
It is well known that in the U.S., the outcomes of transplant patients are not homogeneous among different socio-economic and ethnic/racial groups.
2
,4
,48
In a previous study, we found that Black/African Americans patients had the lowest survival rate among all the racial/ethnic groups after LT for HCC.10
However, whether the survival difference between Black/African Americans and White/Caucasians had increased over the years has remained an unanswered question that was addressed in the current study.- Molinari M.,A.S.
- Tsung A.
- Jorgensen D.
- Bou Samra P.
- Jonaissaint N.
Adult African Americans undergoing cadaveric liver transplantation for hepatocellular carcinoma within the Milan criteria have the lowest 5-year survival among all the ethnic groups in the United States: analysis of USA national data between January 2002 and June 2013.
Hepatoma Res. 2018; 4
As socio-economic inequalities result in health-related disadvantages for some groups of patients, particularly for individuals belonging to ethnic/racial minorities,
49
,50
we hypothesized that increasing socio-economic disparities observed in our society might have contributed to the increasing of the survival gap between Black/African American patients and White/Caucasian LT recipients.32
, 33
, 34
To test this hypothesis, we analyzed the outcomes of adult patients after first time LT for HCC in the U.S. stratified by ethnicity/race over three different periods. The main findings of our analyses were that after adjusting for important clinical and socio-demographic confounders, Black/African Americans had a significantly lower survival rate than other ethnic/racial groups.
The survival gap between Black/African Americans and White/Caucasian patients, however, had progressively decreased over time. While the five-year survival gap was 9% (60% vs. 69%) between Black/African Americans and White/Caucasian patients transplanted between 2002 and 2007 (Cohort I), the survival gap decreased to 4% (78% vs. 82%) in patients transplanted between 2012 and 2017 (Cohort III). More importantly, we found that Black/African Americans had the most significant survival increase among all the racial groups during the study period.
To better understand the possible reasons for these positive trends, we tested whether during the study period the selection of Black/African American recipients had become more stringent. If that were the case, we would have expected a drop in the proportion of Black/African American being transplanted for HCC. The results of our analysis however did not confirm this hypothesis. In fact, the percentage of Black/African American patients transplanted for HCC in the USA during the study period had increased. In addition, we did not find any significant differences in demographic, oncological, and clinical characteristics between Black/African American recipients transplanted in Cohort I versus patients transplanted in Cohorts II or III.
A more plausible explanation for the significant survival improvement observed among Black/African American patients was that during the study period, DAAs for the treatment of HCV were introduced in the United States. Since HCV is more prevalent among Black/African Americans, we suspect that the availability of DAAs benefitted this group more significantly than others.
51
The sensitivity analyses performed to test this assumption supported our hypothesis. In fact, we found that HCV-positive patients had a more pronounced survival increment during the study period, and that Black/African American patients with HCV had the highest survival increment among all the subgroups. However, further studies will be necessary to test the strength and reproducibility of these findings.Besides the introduction of DAAs, many other factors might have influenced the results of our study. Although our aims did not include the identification of possible reasons for the survival improvement observed in our population, the finding of this study gave us the opportunity to reflect on some possibilities. These included the implementation of better perioperative management and more aggressive oncological locoregional therapies before LT, improved surgical techniques and advancements in the long-term care of patients with hyperlipidemia, renal insufficiency, diabetes, obesity, and other metabolic conditions that have been linked to the chronic use of immunosuppression medications.
Although innovative, this study has several limitations. Because of its retrospective design, the cohort was subject to selection bias. Furthermore, studies based on national registries such as the SRTR are at risk of errors in data collection. Another limitation is that adjustment for estimated household income using zip coding or further analysis to test differences between recipients living in rural versus urban areas were not performed. Lastly, it is important to keep in mind that the self-designation of race/ethnicity as recorded in SRTR does not reflect the complexity and the variety of the social, cultural, and biological attributes of the US population.
Despite the above limitations, this study has several strengths. The large number of patients included in our analysis provides the opportunity of assessing trends at a national level since we used data extracted from the SRTR. Another strength is the fact that this study addressed a very relevant healthcare question. Inequalities among different racial and socio-economic groups are among of the most critical public health problems in the U.S. and many other countries in the world.
18
Therefore, it is important that healthcare disparities are closely monitored as specific policies might become necessary to correct inequalities in the future.52
, 53
, - Pritchard R.S.
- Fisher E.S.
- Teno J.M.
- Sharp S.M.
- Reding D.J.
- Knaus W.A.
- et al.
Influence of patient preferences and local health system characteristics on the place of death. SUPPORT Investigators. Study to understand prognoses and preferences for risks and outcomes of treatment.
J Am Geriatr Soc. 1998; 46: 1242-1250
54
, 55
In conclusion, there has been a significant improvement in the 5-year survival rate of patients undergoing LT for HCC in the U.S. Between 2002 and 2017. Black/African American patients have experienced the most significant survival increment among all the ethnic/racial groups, and this trend was particularly pronounced for HCV positive recipients. We hypothesize that the introduction of new oral DAAs for the treatment of HCV was the main reason for these findings. Despite these positive results, the adjusted 5-year survival of Black/African American patients undergoing LT for HCC in the United States remains the lowest among all the racial/ethnic groups. Further studies are necessary to assess whether similar trends occurred patients undergoing LT for other indications.
Disclaimer
The data reported here have been supplied by the Hennepin Healthcare Research Institute (HHRI) as the contractor for the Scientific Registry of Transplant Recipients (SRTR). The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy of or interpretation by the SRTR or the U.S. Government.
Funding
JB has received research grant support from Gilead , Endra Life Sciences , and Pfizer , and has institutional research contracts with Intercept, Pfizer, Galectin, Intercept, Genentech, and Inventiva.
Contributions
Hao Liu: Study design, data analysis, writing and revision of the manuscript.
Christof Kaltenmeier: Study design, data analysis, and revision of the manuscript.
Naudia Jonassaint: Study design and revision of the manuscript.
Zhang Xingyu: Data analysis and revision of the manuscript.
Jaideep Behari: Study design and revision of the manuscript.
Ramon Bataller: Study design and revision of the manuscript.
Andres Duarte-Rojo: Study design and revision of the manuscript.
Shahid Malik: Study design and revision of the manuscript.
Dempsey L Hughes: Study design and revision of the manuscript.
Dheera Reddy: Study design and revision of the manuscript.
Ann Thompson: Study design and revision of the manuscript.
Hasan Al Harakeh: Study design and revision of the manuscript.
Roy Hill: Study design and revision of the manuscript.
Emilia Diego: Study design and revision of the manuscript.
Andrea di Martini: Study design and revision of the manuscript.
Michele Molinari: Study design, data analysis, writing and revision of the manuscript.
Conflict of interest
None declared.
References
- Adult African Americans undergoing cadaveric liver transplantation for hepatocellular carcinoma within the Milan criteria have the lowest 5-year survival among all the ethnic groups in the United States: analysis of USA national data between January 2002 and June 2013.Hepatoma Res. 2018; 4
- Combination of racial/ethnic and etiology/disease-specific factors is associated with lower survival following liver transplantation in African Americans: an analysis from UNOS/OPTN database.Clin Transplant. 2014; 28: 755-761
- Association of insurance with cancer care utilization and outcomes.CA Cancer J Clin. 2008; 58: 9-31
- Racial/ethnic disparities in hepatocellular carcinoma treatment and survival in California, 1988-2012.World J Gastroenterol. 2016; 22: 8584-8595
- Independent prognostic factors for posttransplant survival in hepatocellular carcinoma patients undergoing liver transplantation.Cancer Med. 2017; 6: 26-35
- Racial and insurance disparities in the receipt of transplant among patients with hepatocellular carcinoma.Cancer. 2010; 116: 1801-1809
- Ethnic disparities and liver transplantation rates in hepatocellular carcinoma patients in the recent era: results from the surveillance, epidemiology, and end results registry.Liver Transpl. 2014; 20: 528-535
- Racial disparities in liver transplantation for hepatocellular carcinoma are not explained by differences in comorbidities, liver disease severity, or tumor burden.Hepatol Commun. 2019; 3: 52-62
- Racial disparities in treatment and survival of patients with hepatocellular carcinoma in the United States.Hepatobiliary Surg Nutr. 2016; 5: 43-52
- Adult African Americans undergoing cadaveric liver transplantation for hepatocellular carcinoma within the Milan criteria have the lowest 5-year survival among all the ethnic groups in the United States: analysis of USA national data between January 2002 and June 2013.Hepatoma Res. 2018; 4
- Race, ethnicity, and socioeconomic status influence the survival of patients with hepatocellular carcinoma in the United States.Cancer. 2010; 116: 1367-1377
- Disparities in COVID-19 Outcomes by race, ethnicity, and socioeconomic status: a systematic-review and meta-analysis.JAMA Netw Open. 2021; 4: e2134147
- Race/ethnicity-specific disparities in cancer incidence, burden of disease, and overall survival among patients with hepatocellular carcinoma in the United States.Cancer. 2016; 122: 2512-2523
- The impact of race on survival after hepatocellular carcinoma in a diverse American population.Dig Dis Sci. 2018; 63: 515-528
- Effects of cancer stage and treatment differences on racial disparities in survival from colon cancer: a United States population-based study.Gastroenterology. 2016; 150: 1135-1146
- Cancer facts & figures for African Americans 2019-2021.([cited 2021 10-02-2021]; Available from:)
- Race and ethnicity in access to and outcomes of liver transplantation: a critical literature review.Am J Transplant. 2009; 9: 2662-2668
- Unequal treatment: confronting racial and ethnic disparities in health care.J Natl Med Assoc. 2002; 94: 666-668
- Development of a support for Black Lives Matter measure among racially-ethnically diverse college students.Am J Community Psychol. 2021; 68: 100-113
- Implicit racism, colour blindness, and narrow definitions of discrimination: why some White people prefer ‘All Lives Matter’ to ‘Black Lives Matter’.Br J Soc Psychol. 2021; 60: 1136-1153
- A 10-year social media analysis exploring hospital online support of Black lives Matter and the Black community.JAMA Netw Open. 2021; 4: e2126714
- High social vulnerability and “textbook outcomes” after cancer operation.J Am Coll Surg. 2021; 232: 351-359
- Race, ethnicity, culture, and disparities in health care.J Gen Intern Med. 2006; 21: 667-669
- Social determinants of health: how environmental factors affect health.Sr Care Pharm. 2020; 35: 56-67
- Black patients with cirrhosis have higher mortality and lower transplant rates: results from a metropolitan cohort study.Hepatology. 2021; 74: 926-936
- Sociodemographic differences in early access to liver transplantation services.Am J Transplant. 2009; 9: 2092-2101
- Persistent disparities in liver transplantation for patients with hepatocellular carcinoma in the United States, 1998 through 2007.Cancer. 2011; 117: 4531-4539
- Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis.N Engl J Med. 1996; 334: 693-699
- Liver transplantation for hepatocellular carcinoma. Working group report from the ILTS transplant oncology consensus conference.Transplantation. 2020; 104: 1136-1142
- Ethnic differences in hepatocellular carcinoma: implications for liver transplantation.Dig Dis Sci. 2008; 53: 551-555
- Effect of race on outcome of orthotopic liver transplantation: a cohort study.Lancet. 2002; 359: 287-293
- The globarl increase in the socioeconomic achievement gap, 1964 to 2015.Am Sociol Rev. 2019; 84: 517-544
- Most Americans say there is too much economic inequality in the U.S., but fewer than half call it a top priority.Pew Res Cent. 2020; : 1-45
- N.K., socioeconomic disparities in health: pathways and policies.Health Aff. 2002; 21: 60-76
- The Helsinki Declaration of the world medical association.Gac Med Mex. 2001; 137: 387-390
- The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.Lancet. 2007; 370: 1453-1457
- Scientific registry of transplant recipients: collecting, analyzing, and reporting data on transplantation in the United States.Transplant Rev. 2013; 27: 50-56
- Impact of body mass index on outcomes of 48281 patients undergoing first time cadaveric liver transplantation.World J Transplant. 2016; 6: 356-369
- A model to predict survival in patients with end-stage liver disease.Hepatology. 2001; 33: 464-470
- Characteristics associated with liver graft failure: the concept of a donor risk index.Am J Transplant. 2006; 6: 783-790
- A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts.Hepatology. 2000; 31: 864-871
World Health Organization. Obesity and overweight. [Accessed 2015 Jan]. Available from: URL: http//www.who.int/mediacentre/factsheets/fs311/en. [cited 2021 August 5].
- Teoria statistica delle class e calcolo delle probabilita′.Pubbl Reg Ist Super Sci Econ Commer Firenze. 1936; 8: 3-62
- Survival analysis: properties and Kaplan-Meier method.Dtsch Med Wochenschr. 2007; 132: e36-e38
- Nonparametric estimation from incomplete observations.J Am Stat Assoc. 1958; 53: 457-481
- Asymptomatically efficient rank invariant test procedures.J R Stat Soc. 1972; 135: 185-207
- Regression models and life tables.J R Stat Soc Ser B. 1972; 34: 187-202
- Racial disparity in primary hepatocellular carcinoma: tumor stage at presentation, surgical treatment and survival.J Natl Med Assoc. 2006; 98: 1934-1939
- A developmental science perspective on social inequality.Curr Dir Psychol Sci. 2020; 29: 610-616
- The misperception of racial economic inequality.Perspect Psychol Sci. 2019; 14: 899-921
- Recent advancement of direct-acting antiviral agents (DAAs) in hepatitis C therapy.Mini Rev Med Chem. 2018; 18: 584-596
- Understanding the treatment preferences of seriously ill patients.N Engl J Med. 2002; 346: 1061-1066
- Influence of patient preferences and local health system characteristics on the place of death. SUPPORT Investigators. Study to understand prognoses and preferences for risks and outcomes of treatment.J Am Geriatr Soc. 1998; 46: 1242-1250
- Covid-19-implications for the health care system.N Engl J Med. 2020; 383: 1483-1488
- Socioeconomic status in health research: one size does not fit all.JAMA. 2005; 294: 2879-2888
Article Info
Publication History
Published online: July 18, 2022
Accepted:
July 13,
2022
Received:
May 16,
2022
Identification
Copyright
© 2022 The Author(s). Published by Elsevier Ltd on behalf of International Hepato-Pancreato-Biliary Association Inc.
User License
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
01534-9&id=gr1.jpg){kind=link}
01534-9&id=gr2.jpg){kind=link}
01534-9&id=gr3.jpg){kind=link}
01534-9&id=gr4.jpg){kind=link}
01534-9&id=gr5.jpg){kind=link}
01534-9&id=gr6.jpg){kind=link}