Abstract
Background & Aims
Racial and ethnic minorities are reported to have higher mortality related to hepatocellular carcinoma (HCC) than non-Hispanic whites. However, it is not clear whether differences in tumor characteristics or liver dysfunction among racial or ethnic groups affect characterization of causes for this disparity. We aimed to characterize racial and ethnic differences in HCC presentation, treatment, and survival.
Methods
We performed a retrospective study of patients diagnosed with HCC from January 2008 through July 2017 at 2 large health systems in the United States. We used multivariable logistic regression and Cox proportional hazard models to identify factors associated with receipt of curative therapy and overall survival.
Results
Among 1117 patients with HCC (35.9% white, 34.3% black, 29.7% Hispanic), 463 (41.5%) were diagnosed with early-stage HCC (Barcelona Clinic Liver Cancer stage 0/A) and 322 (28.8%) underwent curative treatment. Hispanic (odds ratio [OR], 0.75; 95% CI 0.55–1.00) and black patients (OR, 0.74; 95% CI, 0.56–0.98) were less likely to be diagnosed with early-stage HCC than white patients. Among patients with early-stage HCC, Hispanics were less likely to undergo curative treatment than whites (OR, 0.58; 95% CI, 0.36–0.91). Black patients with early-stage HCC were also less likely to undergo curative treatment than white patients, but this difference was not statistically significant (OR, 0.66, 95% CI, 0.43–1.03). Black and Hispanic patients had shorter median survival times than white patients (10.6 months and 14.4 months vs 16.3 months). After adjusting for type of medical insurance, Child Pugh class, Barcelona Clinic Liver Cancer stage, and receipt of HCC treatment, black patients had significantly higher mortality (HR 1.12, 95%CI 1.10 – 1.14) and Hispanic patients had lower mortality (HR 0.83, 95%CI 0.74 – 0.94) than white patients.
Conclusion
In a retrospective study of patients diagnosed with HCC, we found racial/ethnic differences in outcomes of HCC to be associated with differences in detection of tumors at early stages and receipt of curative treatment. These factors are intervention targets for improving patient outcomes and reducing disparities.
Keywords: Liver cancer, disparities, hispanic, black, early detection
INTRODUCTION
Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related death worldwide and a leading cause of morbidity and mortality in patients with cirrhosis.1 Due to the current epidemic of advanced hepatitis C virus infection and rapidly increasing incidence of non-alcoholic steatohepatitis (NASH), HCC incidence in the U.S. has doubled over the past two decades and is predicted to continue to increase over the next 20 years.2
HCC disproportionally affects racial/ethnic minority populations in the United States, with higher age-specific rates among non-Hispanic Black (Black), Hispanic, and Asian persons compared to non-Hispanic whites (Whites). Age-adjusted incidence rates are higher for both Hispanics and Blacks (6.3 and 5.0 per 100,000, respectively), compared to Whites (2.4 per 100,000)3. Hispanics have experienced the largest increase in HCC incidence among all racial/ethnic groups, with an estimated 4.7% increase in incidence each year since 2000 (vs. 4.3% in Blacks and Whites and 0.6% in Asians).3, 4
Racial and ethnic minority patients in the U.S. are also at risk for poor cancer-related outcomes. Racial and ethnic disparities in breast, colorectal, and cervical cancer process and outcome measures persist across the cancer care continuum, from screening underuse, to diagnostic delays, inadequate receipt or delays in treatment, and worse overall survival.5–8 Socioeconomic disadvantage, lack of health insurance, and language barriers limit access to cancer screening and treatment.9–11 However, some believe differences in prognosis extend beyond screening process disparities and could be related to biologic differences in tumor behavior.
In HCC, prior studies have suggested racial and ethnic-specific disparities in incidence, tumor characteristics, treatment receipt, and outcomes, with Blacks and Hispanics having poorer survival than White and Asians.12–17 Further, prior studies have found minority patients are less likely to receive HCC-directed therapy than non-minorities, including surgical resection and liver transplantation.12, 13, 15, 18 While lower receipt of curative therapy among minority populations is a common finding across studies, there are conflicting data on survival differences between racial and ethnic groups.19, 20 Previous studies are limited by the use of population-based administrative datasets which lack clinical granularity on liver disease etiology, underlying liver function, tumor burden, and patient performance status, making it difficult to determine exactly where and why disparities occur or the magnitude of their impact on patient outcomes.
Understanding and quantifying potential differences is the first step to identify intervention targets and inform strategies to reduce existing disparities. Therefore, the aim of our study was to compare HCC presentation, treatment, and survival among Hispanic, Black, and non-Hispanic White patients followed in two large urban health systems in the United States.
METHODS
Study Population
Using prospectively maintained HCC databases at Parkland Memorial Health and Hospital System (Parkland) and UT Southwestern Medical Center, we identified all patients diagnosed with HCC between January 2008 and July 2017. As the safety-net health system for Dallas County, Parkland serves a large population of socioeconomically disadvantaged patients with cirrhosis and HCC. Parkland patients receive HCC screening, diagnostic evaluation, and HCC-directed therapy for no or reduced cost through a financial assistance program funded by Dallas County taxes. Although liver transplantation is not available at Parkland, eligible patients without social and financial barriers can be referred to other health systems, including UT Southwestern, for transplant evaluation. As the only NCI-designated cancer center in North Texas, UT Southwestern acts as a tertiary care referral center for patients with HCC across Texas and surrounding states. Patients at Parkland and UT Southwestern are cared for by the same set of dedicated HCC providers and cases are reviewed in a shared multidisciplinary conference. Both Parkland and UT Southwestern have clinic availability to see patients newly diagnosed with HCC within 2 weeks of referral.
We confirmed all HCC diagnoses using the American Association for the Study of Liver Disease (AASLD) criteria. 21 We excluded: 1) patients without imaging studies and for whom we could not determine tumor characteristics; 2) those with a liver mass without characteristic imaging (arterial enhancement with delayed washout) or histological confirmation; and 3) those who received treatment at an outside institution prior to presentation at one of the study sites. The Institutional Review Board of the UT Southwestern Medical Center approved this study.
Data Collection
We reviewed each patient’s electronic medical record (EMR) to obtain patient demographics, clinical history, laboratory data, and imaging results at time of HCC diagnosis. Race/ethnicity was obtained by self-report from all patients at time of clinic visits and was categorized as non-Hispanic white (White), non-Hispanic black (Black), Hispanic, and other. Insurance status at HCC diagnosis was categorized as Medicare, Medicaid, private insurance, or no insurance. Clinical history of interest included comorbid conditions such as metabolic syndrome, alcohol and smoking history, cirrhosis etiology, liver function (Child Pugh score), healthcare utilization patterns, HCC-directed treatment, and overall survival (See Supplemental Material for details). Tumor characteristics of interest included the number of lesions, maximum tumor diameter, portal vein invasion, hepatic vein invasion, presence of distant metastases, and stage of the tumor at diagnosis. The Barcelona Clinic Liver Cancer (BCLC) stage was used for tumor staging. Treatment for HCC was considered curative if it included local ablative techniques, surgical resection or liver transplantation. If patients received multiple treatments, they were classified as having received the most curative therapy. For example, if a patient received TACE prior to liver transplantation, they were classified as having liver transplantation.
Statistical Analysis
Fisher exact and Chi-square tests were used to compare characteristics of the study population by race/ethnicity. Multivariable logistic regression models were used to evaluate the association between race/ethnicity and a) early tumor detection, defined as BCLC stage 0 or A, and b) curative treatment receipt. We estimated median overall survival, stratified by race/ethnicity, using Kaplan-Meier analysis. Overall survival was calculated from HCC diagnosis to the last known date of follow-up, liver transplantation, death, or end of study period (July 1, 2017). Log rank tests were used to compare survival distributions by race/ethnicity. We used univariate and multivariable Cox proportional hazard regression models to identify factors associated with overall survival, adjusting for clustering by health system. We included BCLC stage and receipt of curative treatment as a priori correlates of overall survival. Other covariates were included in the multivariable model if they were significantly associated with overall survival in univariate analysis. We report adjusted associations of covariates and overall survival as hazard ratios (HR) and 95% confidence intervals (CI). All tests were two-sided and performed at the 5% significance level. Statistical analysis was performed using Stata 14.0 (College Station, TX).
RESULTS
Patient Characteristics
We identified 1117 eligible HCC patients, of whom 401 (35.9%) were White, 384 (34.4%) Black, and 332 (29.7%) Hispanic. Eighty-three patients of other races were excluded from analyses. Baseline patient characteristics are detailed in Table 1. Blacks were significantly younger than Whites (p=0.003) but age was not significantly different between Hispanics and Whites (p=0.67). Compared to Whites, Blacks were more likely to have viral hepatitis (p<0.001) whereas Hispanics were more likely to have alcohol- or NASH-related cirrhosis (p<0.001). Similarly, obesity (BMI ≥30) was more common in Hispanics (p=0.02) but not Blacks (p=0.25) compared to Whites; however both Blacks (p<0.001) and Hispanics (p=0.001) were more likely to have any component of the metabolic syndrome compared to Whites. Degree of liver dysfunction significantly differed by race/ethnicity, with Child A cirrhosis less likely in Hispanics (p=0.003) and more likely in Blacks (p=0.03) than Whites. Specifically, Blacks had less ascites and hepatic encephalopathy than Whites. Performance status did not significantly differ between Blacks or Hispanics compared to Whites.
Table 1.
Patient clinical characteristics at time of HCC diagnosis, stratified by race/ethnicity
| White (n=401) | Black (n=384) | Hispanic (n=332) | p-value | |
|---|---|---|---|---|
|
| ||||
| Age (years) | 61.0 ± 9.4 | 59.0 ± 7.7 | 60.7 ± 10.0 | 0.006 |
|
| ||||
| Gender (% male) | 332 (82.8%) | 292 (76.2%) | 237 (71.4%) | 0.001 |
|
| ||||
| Hospital | < 0.001 | |||
| Parkland Health and Hospital | 187 (46.6) | 319 (83.1) | 267 (80.4) | |
| UT Southwestern Medical Center | 214 (53.4) | 65 (16.9) | 65 (19.6) | |
|
| ||||
| Medical insurance | < 0.001 | |||
| Commercial insurance | 97 (24.2) | 32 (8.3) | 37 (11.1) | |
| Medicare | 132 (32.9) | 104 (27.1) | 77 (23.2) | |
| Medicaid | 59 (14.7) | 106 (27.6) | 66 (19.9) | |
| Parkland financial assistance* | 113 (28.2) | 142 (37.0) | 152 (45.8) | |
|
| ||||
| Body mass index | < 0.001 | |||
| Normal (< 25) | 139 (35.7) | 156 (41.2) | 73 (22.1) | |
| Overweight (25 – < 30) | 136 (35.0) | 126 (33.3) | 133 (40.3) | |
| Obese (30 – < 35) | 68 (17.5) | 62 (16.4) | 81 (24.6) | |
| Morbidly obese (≥35) | 46 (11.8) | 35 (9.2) | 43 (13.0) | |
|
| ||||
| Marital status (% married) | 150 (37.9) | 86 (22.8) | 139 (42.9) | < 0.001 |
|
| ||||
| Education (% some college) | 101 (48.3) | 34 (21.3) | 20 (13.6) | < 0.001 |
|
| ||||
| Employment (% active employment) | 86 (27.7) | 43 (18.0) | 37 (17.1) | 0.003 |
|
| ||||
| Alcohol history (% active heavy use) | 60 (15.2) | 82 (21.9) | 56 (17.0) | 0.04 |
|
| ||||
| Metabolic syndrome, any component | 287 (71.6) | 323 (84.1) | 272 (81.9) | < 0.001 |
|
| ||||
| Performance status | 0.22 | |||
| ECOG 0–1 | 335 (84.2) | 310 (80.7) | 267 (80.9) | |
| ECOG 2 | 52 (13.0) | 50 (13.0) | 46 (13.9) | |
| ECOG 3 | 11 (2.8) | 24 (6.3) | 17 (5.2) | |
|
| ||||
| Hepatology subspecialty care during year prior to HCC diagnosis | 131 (32.8) | 84 (21.9) | 96 (29.0) | 0.003 |
|
| ||||
| Liver disease etiology | < 0.001 | |||
| Hepatitis C | 269 (67.4) | 315 (82.9) | 147 (45.0) | |
| Hepatitis B | 15 (3.8) | 30 (7.9) | 6 (1.8) | |
| Alcohol-related | 54 (13.5) | 16 (4.2) | 88 (26.9) | |
| NASH/cryptogenic | 46 (11.5) | 13 (3.4) | 77 (23.6) | |
|
| ||||
| Platelet count | 135 ± 88 | 174 ± 102 | 124 ± 83 | < 0.001 |
|
| ||||
| Bilirubin (mg/dL) | 2.1 ± 3.7 | 1.7 ± 2.5 | 2.3 ± 3.2 | 0.04 |
|
| ||||
| Presence of hepatic encephalopathy | 93 (23.3) | 33 (8.6) | 92 (27.8) | < 0.001 |
|
| ||||
| Presence of ascites | 195 (48.6) | 129 (33.6) | 177 (53.3) | < 0.001 |
|
| ||||
| Child Pugh class | < 0.001 | |||
| Child Pugh A | 194 (48.4) | 216 (56.4) | 124 (37.4) | |
| Child Pugh B | 153 (38.1) | 131 (34.2) | 147 (44.3) | |
| Child Pugh C | 54 (13.5) | 36 (9.4) | 61 (18.4) | |
Continuous data expressed as mean ± standard deviation
ECOG – Eastern Cooperative Oncology Group; NASH – nonalcoholic steatohepatitis
Parkland offers a sliding fee scale program, which provides access to subspecialty medical care, including hepatocellular carcinoma screening and treatment, at low cost for uninsured Dallas County residents
Percentages are reported among patients with non-missing data
Clinical Presentation and Tumor Stage
One-third (30.7%) of patients received care at the tertiary care health system, with Hispanics and Blacks more likely than Whites to be seen at the safety-net system. HCC was detected via surveillance in 40.7% of patients, incidentally in 36.5%, and symptomatically in 22.8% of patients. Despite a higher proportion of Whites receiving Hepatology care in the year prior to HCC diagnosis compared to non-Whites (p=0.003), the proportion of HCC detected by surveillance was similar among groups.
Differences in tumor characteristics, by race/ethnicity, are detailed in Table 2. There was no difference in number of nodules between groups, including similar proportions of unifocal HCC; however, Blacks were noted to have larger overall tumor burden than Whites, including higher proportions of infiltrative HCC, vascular invasion and distant metastases. Overall, only 44.0% of patients were diagnosed with HCC within Milan Criteria. Compared to Whites, Blacks were significantly less likely be detected within Milan Criteria (p=0.005). Hispanics were also less likely than Whites to be found within Milan Criteria but this difference did not reach statistical significance (p=0.11).
Table 2.
Tumor characteristics and treatment, stratified by race/ethnicity
| White (n=401) | Black (n=384) | Hispanic (n=332) | p-value | |
|---|---|---|---|---|
|
| ||||
| Surveillance during year prior to HCC diagnosis | 172 (43.2) | 147 (38.9) | 136 (41.5) | 0.47 |
|
| ||||
| Number of HCC nodules | 0.06 | |||
| One | 212 (52.9) | 186 (48.6) | 177 (53.5) | |
| Two | 70 (17.5) | 49 (12.8) | 48 (14.5) | |
| Three or more | 119 (29.6) | 148 (38.6) | 106 (32.0) | |
|
| ||||
| Maximum tumor diameter (cm) | 5.1 ± 4.1 | 5.8 ± 4.4 | 5.3 ± 4.2 | 0.06 |
|
| ||||
| Distant metastases | 60 (15.0) | 90 (23.4) | 59 (17.8) | 0.008 |
|
| ||||
| Portal vein invasion | 91 (22.9) | 116 (31.0) | 69 (21.2) | 0.005 |
|
| ||||
| HCC within Milan Criteria | 199 (49.6) | 152 (39.6) | 145 (43.7) | 0.02 |
|
| ||||
| BCLC Stage | 0.001 | |||
| Stage 0/A | 185 (46.1) | 149 (38.8) | 129 (39.0) | |
| Stage B | 76 (19.0) | 60 (15.6) | 55 (16.6) | |
| Stage C | 79 (19.7) | 119 (31.0) | 75 (22.7) | |
| Stage D | 61 (15.2) | 56 (14.6) | 72 (21.8) | |
|
| ||||
| Treatment during follow-up | < 0.001 | |||
| Liver Transplantation | 43 (10.7) | 7 (1.8) | 23 (6.9) | |
| Resection | 51 (12.7) | 63 (16.4) | 18 (5.4) | |
| Local ablation | 49 (12.2) | 30 (7.8) | 38 (11.5) | |
| TACE/TARE/SBRT | 115 (28.7) | 83 (21.6) | 102 (30.8) | |
| Systemic therapy | 26 (6.5) | 58 (17.5) | 36 (10.8) | |
| Best supportive care | 117 (29.2) | 134 (34.9) | 115 (34.6) | |
|
| ||||
| Median time to first treatment, days (IQR) | 49 (31 – 79) | 49 (30 – 78) | 47 (29–71.5) | |
BCLC – Barcelona Clinic Liver Cancer; HCC – hepatocellular carcinoma; SBRT – stereotactic body radiation therapy; TACE – transarterial chemoembolization; TARE – transarterial radioembolization
Numbers in parentheses represent percentages, except where otherwise specified
There was variation in BCLC tumor stage, with 5.5% of patients diagnosed with BCLC stage 0 tumors, 35.8% BCLC A, 16.9% BCLC B, 25.0% BCLC C, and 16.8% BCLC stage D HCC. BCLC D HCC diagnosis was related to Child C cirrhosis in 72% of cases, poor ECOG performance status in 21%, and both in 7% of cases. In univariate analysis, Hispanics (OR 0.75, 95%CI 0.55 – 1.00) and Blacks (OR 0.74, 95%CI 0.56 – 0.98) were less likely to found at BCLC stage 0/A than Whites. After adjusting for receipt of HCC surveillance, Hispanics were still less likely than Whites to be detected at BCLC stage 0/A (OR 0.70, 95%CI 0.51 – 0.98); however Black-White differences were no longer statistically significant (OR 0.76, 95%CI 0.55 – 1.04).
Curative Treatment Receipt
In total, 28.8% of patients received curative treatment during median follow-up of 8.5 (IQR 3.1 – 21.1) months (Table 2). Among 463 patients with early stage HCC, less than two-thirds (58.1%) underwent curative treatment, with a higher proportion in Whites than Hispanics and Blacks (64.9% vs. 51.5% and 55.0%, respectively; p=0.04). In univariate analysis, curative treatment receipt was less likely in Hispanics (OR 0.51, 95%CI 0.37 – 0.70) and Blacks (OR 0.60, 95%CI 0.44 – 0.81) than Whites. Among those with BCLC stage 0/A HCC, Hispanics remained less likely to undergo curative treatment than Whites (OR 0.58, 95%CI 0.36 – 0.91); however, the difference in curative treatment between Blacks and Whites was not statistically significant (OR 0.66, 95%CI 0.43 – 1.03). Similar results were seen in multivariable analyses after adjusting for BCLC tumor stage, age, and insurance status. Hispanics were significantly less likely to undergo curative treatment than Whites (OR 0.57, 95%CI 0.39 – 0.85); but the difference between Blacks and Whites was not statistically significant (OR 0.79, 95%CI 0.54 – 1.15).
Overall Survival
Median overall survival for all patients was 13.8 months, with 1-year and 3-year survival rates of 53.0% and 29.5%. Overall survival, stratified by race/ethnicity, is illustrated in Figure 1. For Whites, median overall survival was 16.3 (IQR 5.2 – 61.2) months, compared to 14.4 (IQR 3.9 – 52.7) months for Hispanics and 10.6 (IQR 3.3 – 34.1) months for Blacks. In univariate analysis, Blacks had worse survival than Whites (HR 1.34, 95%CI 1.11 – 1.61) but survival was similar between Hispanics and Whites (HR 1.07, 95%CI 0.78 – 1.50). In exploratory subgroup analyses (Table 3), Black-White disparities in survival were most notable for BCLC stages A and D but less pronounced for BCLC stages B and C. Survival was similar between Hispanics and Whites across all BCLC stages.
Figure 1.
Overall survival, stratified by race/ethnicity
Table 3.
Median overall survival in months, stratified by race/ethnicity
| White (n=401) | Black (n=384) | Hispanic (n=332) | Black vs. White (HR, 95%CI) | Hispanic vs. White (HR, 95%CI) | |
|---|---|---|---|---|---|
| Overall | 16.3 (5.2 – 61.2) | 10.6 (3.3 – 34.1) | 14.4 (3.9 – 52.7) | 1.34 (1.12 – 1.60) | 1.08 (0.89 – 1.30) |
| BCLC stage A | 45 (17.0 – 94.6) | 28.4 (13.2 – 55.7) | 42.6 (17.0 – 82.6) | 1.4 (0.97 – 1.91) | 0.93 (0.64 – 1.36) |
| BCLC stage B | 14.8 (6.0 – 29.8) | 16.8 (8.4 –27.8) | 16.0 (7.9 – 22.0) | 0.92 (0.59 – 1.42) | 1.11 (0.72 – 1.72) |
| BCLC stage C | 5.8 (1.8 – 10.3) | 4.5 (2.0 – 7.9) | 4.0 (2.2 – 7.9) | 1.13 (0.83 – 1.55) | 1.07 (0.75 – 1.52) |
| BCLC stage D | 2.3 (1.2 – 13.6) | 1.5 (0.7 – 3.1) | 3.5 (1.1 – 8.5) | 1.77 (1.19 – 2.62) | 0.97 (0.66 – 1.43) |
BCLC – Barcelona Clinic Liver Cancer
Interquartile ranges (IQR) reported for median survival time
The association between Black race and survival was mitigated but still statistically significant when adjusting for insurance status, Child Pugh class, BCLC tumor stage, and type of HCC-directed treatment (HR 1.12, 95%CI 1.10 – 1.14) (Table 4). When adjusting for insurance, status, Child Pugh class, BCLC tumor stage, and type of HCC-directed treatment, there was a significant difference in overall survival between Hispanics and Whites, with reduced mortality among Hispanics (HR 0.83, 95%CI 0.74 – 0.94).
Table 4.
Factors associated with overall survival
| Univariate Analysis | Multivariable Analysis | |||
|---|---|---|---|---|
|
| ||||
| Hazard Ratio | 95% CI | Hazard Ratio | 95% CI | |
|
| ||||
| Race/Ethnicity | ||||
| White | Reference | Reference | Reference | Reference |
| Hispanic | 1.07 | 0.78 – 1.50 | 0.83 | 0.74 – 0.94 |
| Black | 1.34 | 1.11 – 1.61 | 1.12 | 1.10 – 1.14 |
|
| ||||
| Type of medical coverage | ||||
| Commercial | Reference | Reference | Reference | Reference |
| Medicare | 0.99 | 0.51 – 1.94 | 0.87 | 0.51 – 1.49 |
| Medicaid | 1.51 | 1.24 – 1.84 | 0.79 | 0.73 – 0.85 |
| Financial Assistance | 1.99 | 1.51 – 2.62 | 0.96 | 0.86 – 1.08 |
| Other | 1.20 | 0.87 – 1.65 | 0.83 | 0.80 – 0.87 |
|
| ||||
| Child Pugh Class | ||||
| Child Pugh A | Reference | Reference | Reference | Reference |
| Child Pugh B | 1.99 | 1.70 – 2.34 | 1.52 | 1.36 – 1.70 |
| Child Pugh B | 3.39 | 1.50 – 7.64 | 0.89 | 0.64 – 1.24 |
|
| ||||
| BCLC stage | ||||
| Stage 0/A | Reference | Reference | Reference | Reference |
| Stage B | 2.39 | 2.26 – 2.54 | 1.79 | 1.79 – 1.80 |
| Stage C | 6.19 | 4.38 – 8.74 | 3.22 | 2.06 – 5.04 |
| Stage D | 6.39 | 2.80 – 14.57 | 4.34 | 3.45 – 5.45 |
|
| ||||
| Treatment | ||||
| Liver transplantation | Reference | Reference | Reference | Reference |
| Resection | 3.30 | 2.47 – 4.41 | 5.40 | 3.87 – 7.53 |
| Local ablation | 5.56 | 5.45 – 5.68 | 7.98 | 5.65 – 11.3 |
| TACE/TARE/SBRT | 10.8 | 7.03 – 16.6 | 11.10 | 5.82 – 21.2 |
| Systemic therapy | 31.2 | 16.2 – 60.3 | 16.9 | 12.0 – 23.9 |
| Best supportive care | 47.8 | 22.8 – 100.3 | 31.3 | 17.9 – 54.7 |
BCLC – Barcelona Clinic Liver Cancer; SBRT – stereotactic body radiation therapy; TACE – transarterial chemoembolization; TARE – transarterial radioembolization
Health system was associated with overall survival on univariate analysis, with the safety-net health system associated with increased mortality (HR 1.48, 95%CI 1.25 – 1.75). See Supplemental Material for subgroup analyses by BCLC stage and health system.
DISCUSSION
In this study, we used clinically granular data from a comprehensive EMR to compare HCC presentation and outcomes by race/ethnicity at two large US hospital systems. We found Hispanic and Blacks were less likely to be diagnosed with early stage HCC compared to Whites. Both Blacks and Hispanics were also less likely than Whites to receive curative treatment for HCC, including among patients with early stage HCC. Although Hispanics had shorter absolute survival than Whites, they had improved survival and stage-adjusted survival after adjusting for insurance coverage, Child Pugh class, BCLC tumor stage, and receipt of HCC treatment. In contrast, blacks had significantly worse survival than Whites, which persisted after adjusting for these factors.
As with most malignancies, HCC tumor burden at diagnosis impacts both prognosis and treatment decisions. HCC most commonly occurs in the background of cirrhosis; therefore, prognosis and treatment options depend on the degree of underlying liver dysfunction at presentation. A higher proportion of Hispanics and Blacks were diagnosed at a later BCLC stage, however this appears to be driven by different factors. Compared to Whites, Hispanics had decreased tumor burden but more advanced liver dysfunction, whereas Blacks had better liver function but were more often detected symptomatically with larger tumor stage. This difference in initial presentation suggests that potential targets for interventions to improve prognosis in these two groups may differ. Disparities in tumor stage at presentation may be in part due to racial/ethnic differences in liver disease etiology and recognition of cirrhosis, HCC surveillance utilization, and access to primary and subspecialty healthcare.22–25
Differences in survival likely involve a combination of medical, financial, and social factors. Some studies have hypothesized this difference may be related to biologic differences in tumor behavior, while others have highlighted socio-demographic inequalities across the HCC cancer care continuum.24, 26, 27 Provider-level factors such as poor physician-patient communication and inherent provider biases, in addition to system-level factors such as care coordination, available technology and treatment delays, may also play a role.28,29 While a prior VA study found race/ethnicity was not associated with all-cause mortality in HCC, this may be attributable to system-level differences between the VA and the health systems evaluated in our study20.
In our study, we found differences in survival were mitigated after adjusting for BCLC tumor stage and receipt of HCC treatment, suggesting these factors may explain differences in prognosis and be appropriate intervention targets for studies to reduce racial disparities in HCC prognosis. Underuse of HCC surveillance is one of the main reasons for late stage tumor detection in the United States, with fewer than 20% of at-risk patients receiving consistent semi-annual HCC surveillance.24 Although prior studies suggest lower HCC surveillance in racial/ethnic minorities and persons of low SES,23 less than half of patients in each racial/ethnic group in our study received HCC surveillance in the year prior to HCC diagnosis. Similarly, less than one-third of patients in our study received curative treatment, including less than two-thirds of patients with BCLC stage 0/A HCC, with lower rates in Blacks and Hispanics than Whites. Other studies have demonstrated the underuse of HCC treatments, with less than 25% of U.S. patients receiving curative treatment in clinical practice.27 Further, it has been demonstrated that patients at safety-net hospitals are less likely to receive HCC treatment even when diagnosed at an early stage30. As observed in other studies, Blacks and Hispanics in our study had less access to liver transplantation than Whites (1.8% and 6.9% vs 10.7%, respectively), likely related in part to a lower proportion of minority patients having insurance coverage.31 Further studies are needed to explore reasons for late stage tumor detection and underuse of curative treatment, particularly among racial/ethnic minorities, to inform future intervention strategies.32–34
Existing disparities research in HCC has historically centered upon the use of large, administrative databases including Medicare and SEER (Surveillance, Epidemiology and End Results) registries. However, these large datasets have several important limitations, of particular importance in HCC compared to other cancers. They have limited clinical data regarding liver dysfunction and performance status, non-specific tumor staging categorized as localized, regional, or distant, only capture the first HCC-directed treatment, are prone to ascertainment bias as HCC is diagnosed based on imaging criteria and not histologically in most cases, and poor capture of race/ethnicity. Although our study addresses many of these issues, it has some notable limitations. Our study’s retrospective nature results in potential for missing data, unmeasured cofounders, and measurement bias (e.g. ECOG performance status). Second, our findings may not be generalizable to other care settings, although safety-net health systems care for a disproportionate share of racial/ethnic minority patients in the U.S.30 For example, both health systems included a multi-disciplinary tumor clinic, which has been shown to improve HCC-related outcomes but may not be universally available in all healthcare systems.32 Further, underinsured or uninsured patients at Parkland are eligible for HCC surveillance and treatment at low- or no-cost using a patient assistance program that is not widely available in the US. Additionally, while some patients at Parkland are referred for liver transplantation at outside institutions, the referrals are not consistently recorded in the EMR and therefore not included in our analysis. Third, our conclusions must be interpreted in light of the inherent complexity of racial and ethnic health disparities. Race/ethnicity in both health systems was self-reported and may not be collected reliably or account for multi-ethnic individuals who are often forced into a single category. Self-reported race/ethnicity may differ from genetic determination, with some studies showing equal heterogeneity within races as between races.35 Race and socioeconomic status are also often highly correlated, complicating interpretation of any observed disparities.36 Finally, Hispanic patients represent a heterogeneous group of patients; in our study, Hispanic patients were primarily of Mexican descent and thus our findings may not generalize to other Hispanic populations. Overall, we believe our study’s limitations are outweighed by its strengths including its large, well-characterized cohort and its racially and socio-economically diverse population.
In summary, we found differences in HCC presentation and outcomes between racial/ethnic groups. Both Blacks and Hispanics were less likely to be diagnosed with early stage HCC and undergo curative treatment than Whites, even among those with early stage HCC. Blacks and Hispanics both had worse absolute survival than Whites, but these racial and ethnic disparities in survival were reduced if not eliminated after adjusting for tumor stage and receipt of HCC treatment. Our data suggest interventions to improve early tumor detection and curative treatment receipt would likely improve HCC outcomes and reduce disparities. Our findings have important implications for health policy and highlight the need for further study on racial-ethnic disparities in HCC including identification of additional actionable intervention targets to reduce disparities in prognosis.
Supplementary Material
Background
Racial and ethnic minority patients with hepatocellular carcinoma may receive less curative treatment and have higher mortality than non-Hispanic whites, however prior data are conflicting on the magnitude and factors underlying these disparities.
Findings
Black patients had significantly higher mortality and Hispanic patients had lower mortality compared to white patients after adjusting for tumor stage, liver function, receipt of HCC treatment, and insurance status.
Implications for patient care
Understanding racial/ethnic differences in hepatocellular carcinoma surveillance, treatment, and mortality is the necessary first step to allow for identification of potential intervention targets to improve patient outcomes and reduce existing disparities.
Acknowledgments
Financial support: This work was conducted with support from NCI RO1 CA222900, NIH RO1 MD012565, and NIH UL1-TR001105. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Abbreviations
- AASLD
American Association for the Study of Liver Disease
- BCLC
Barcelona Clinic Liver Cancer
- BMI
body mass index
- ECOG
Eastern Cooperative Oncology Group
- EMR
electronic medical record
- HCC
hepatocellular carcinoma
- HCV
Hepatitis C virus
- NAFLD
nonalcoholic fatty liver disease
- NASH
nonalcoholic steatohepatitis
- PNPLA3
patatin-like phospholipase domain-containing protein 3
- SBRT
stereotactic body radiation therapy
- SEER
Surveillance, Epidemiology, and End Results
- TACE
transarterial chemoembolization
- TARE
transarterial radioembolization
Footnotes
Conflicts of Interests: None of the authors have any conflicts of interest
Author Contributions:
Nicole E. Rich was involved in study concept and design, acquisition of data, interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content.
Caitlin Hester was involved in interpretation of data and critical revision of the manuscript for important intellectual content.
Mobolaji Odewole was involved in acquisition of data and critical revision of the manuscript for important intellectual content.
Caitlin C. Murphy was involved in interpretation of data and critical revision of the manuscript for important intellectual content.
Neehar D. Parikh was involved in interpretation of data and critical revision of the manuscript for important intellectual content.
Jorge A. Marrero was involved in interpretation of data and critical revision of the manuscript for important intellectual content.
Adam C. Yopp was involved in study concept and design, acquisition of data, analysis and interpretation of data, critical revision of the manuscript for important intellectual content, and study supervision.
Amit G. Singal was involved in study concept and design, acquisition of data, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content, and study supervision.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA: A Cancer Journal for Clinicians. 2015;65:87–108. doi: 10.3322/caac.21262. [DOI] [PubMed] [Google Scholar]
- 2.White DL, Thrift AP, Kanwal F, et al. Incidence of Hepatocellular Carcinoma in All 50 United States, From 2000 Through 2012. Gastroenterology. 2017;152:812–820. doi: 10.1053/j.gastro.2016.11.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.El-Serag HB, Lau M, Eschbach K, et al. Epidemiology of hepatocellular carcinoma in Hispanics in the United States. Arch Intern Med. 2007;167:1983–9. doi: 10.1001/archinte.167.18.1983. [DOI] [PubMed] [Google Scholar]
- 4.El-Serag HB. Epidemiology of viral hepatitis and hepatocellular carcinoma. Gastroenterology. 2012;142:1264–1273 e1. doi: 10.1053/j.gastro.2011.12.061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Patel N, Yopp AC, Singal AG. Diagnostic Delays are Common Among Patients wtih Hepatocellular Carcinoma. J Natl Compr Canc Netw. 2014 doi: 10.6004/jnccn.2015.0074. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Singal AG, Waljee AK, Patel N, et al. Therapeutic delays lead to worse survival among patients with hepatocellular carcinoma. J Natl Compr Canc Netw. 2013;11:1101–8. doi: 10.6004/jnccn.2013.0131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Shavers VL, Brown ML. Racial and ethnic disparities in the receipt of cancer treatment. J Natl Cancer Inst. 2002;94:334–57. doi: 10.1093/jnci/94.5.334. [DOI] [PubMed] [Google Scholar]
- 8.Bach PB, Schrag D, Brawley OW, et al. Survival of blacks and whites after a cancer diagnosis. JAMA. 2002;287:2106–2113. doi: 10.1001/jama.287.16.2106. [DOI] [PubMed] [Google Scholar]
- 9.Artinyan A, Mailey B, Sanchez-Luege N, et al. Race, ethnicity, and socioeconomic status influence the survival of patients with hepatocellular carcinoma in the United States. Cancer. 2010;116:1367–77. doi: 10.1002/cncr.24817. [DOI] [PubMed] [Google Scholar]
- 10.Busch EL, Martin C, DeWalt DA, et al. Functional health literacy, chemotherapy decisions, and outcomes among a colorectal cancer cohort. Cancer Control. 2015;22:95–101. doi: 10.1177/107327481502200112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kroenke CH, Kwan ML, Neugut AI, et al. Social networks, social support mechanisms, and quality of life after breast cancer diagnosis. Breast Cancer Res Treat. 2013;139:515–27. doi: 10.1007/s10549-013-2477-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mathur AK, Osborne NH, Lynch RJ, et al. Racial/ethnic disparities in access to care and survival for patients with early-stage hepatocellular carcinoma. Archives of Surgery. 2010;145:1158–1163. doi: 10.1001/archsurg.2010.272. [DOI] [PubMed] [Google Scholar]
- 13.Ha J, Yan M, Aguilar M, et al. 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. doi: 10.1002/cncr.30103. [DOI] [PubMed] [Google Scholar]
- 14.Stewart SL, Kwong SL, Bowlus CL, et al. Racial/ethnic disparities in hepatocellular carcinoma treatment and survival in California, 1988–2012. World Journal of Gastroenterology. 2016;22:8584–8595. doi: 10.3748/wjg.v22.i38.8584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Xu L, Kim Y, Spolverato G, et al. Racial disparities in treatment and survival of patients with hepatocellular carcinoma in the United States. Hepatobiliary Surgery and Nutrition. 2016;5:43–52. doi: 10.3978/j.issn.2304-3881.2015.08.05. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Berg SA, Modayil M, Nair TD, et al. Racial differences characterizing Hispanics with hepatocellular carcinoma. Journal of Clinical Oncology. 2014;32:e17564–e17564. [Google Scholar]
- 17.Harrison LE, Reichman T, Koneru B, et al. Racial discrepancies in the outcome of patients with hepatocellular carcinoma. Archives of Surgery. 2004;139:992–996. doi: 10.1001/archsurg.139.9.992. [DOI] [PubMed] [Google Scholar]
- 18.Siegel AB, McBride RB, El-Serag HB, et al. Racial disparities in utilization of liver transplantation for hepatocellular carcinoma in the United States, 1998–2002. Am J Gastroenterol. 2008;103:120–7. doi: 10.1111/j.1572-0241.2007.01634.x. [DOI] [PubMed] [Google Scholar]
- 19.Aparo S, Goel S, Lin D, et al. Survival analysis of Hispanics in a cohort of patients with hepatocellular carcinoma. Cancer. 2014;120:3683–3690. doi: 10.1002/cncr.28867. [DOI] [PubMed] [Google Scholar]
- 20.Serper M, Taddei TH, Mehta R, et al. Association of Provider Specialty and Multidisciplinary Care With Hepatocellular Carcinoma Treatment and Mortality. Gastroenterology. 2017;152:1954–1964. doi: 10.1053/j.gastro.2017.02.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Heimbach J, Kulik LM, Finn R, et al. Aasld guidelines for the treatment of hepatocellular carcinoma. Hepatology. 2017 doi: 10.1002/hep.29086. [DOI] [PubMed] [Google Scholar]
- 22.Farvardin S, Patel J, Khambaty M, et al. Patient-reported barriers are associated with lower hepatocellular carcinoma surveillance rates in patients with cirrhosis. Hepatology. 2017;65:875–884. doi: 10.1002/hep.28770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Singal AG, XL, Tiro J, et al. Racial, Social, and Clinical Determinants of Hepatocellular Carcinoma Surveillance. Am J Med. 2014 doi: 10.1016/j.amjmed.2014.07.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Singal AG, Yopp A, CSS, et al. Utilization of hepatocellular carcinoma surveillance among American patients: a systematic review. J Gen Intern Med. 2012;27:861–7. doi: 10.1007/s11606-011-1952-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Singal AG, Yopp AC, Gupta S, et al. Failure Rates in the Hepatocellular Carcinoma Surveillance Process. Cancer Prev Res (Phila) 2012;5:1124–1130. doi: 10.1158/1940-6207.CAPR-12-0046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sloane D, Chen H, Howell C. Racial disparity in primary hepatocellular carcinoma: tumor stage at presentation, surgical treatment and survival. J Natl Med Assoc. 2006;98:1934–9. [PMC free article] [PubMed] [Google Scholar]
- 27.Tan D, Yopp A, Beg MS, et al. Meta-analysis: underutilisation and disparities of treatment among patients with hepatocellular carcinoma in the United States. Aliment Pharmacol Ther. 2013;38:703–12. doi: 10.1111/apt.12450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Dalton-Fitzgerald E, Tiro J, Kandunoori P, et al. Practice Patterns and Attitudes of Primary Care Providers and Barriers to Surveillance of Hepatocellular Carcinoma in Patients with Cirrhosis. Clin Gastroenterol Hepatol. 2014 doi: 10.1016/j.cgh.2014.06.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Penner LA, Dovidio JF, Gonzalez R, et al. The Effects of Oncologist Implicit Racial Bias in Racially Discordant Oncology Interactions. J Clin Oncol. 2016;34:2874–80. doi: 10.1200/JCO.2015.66.3658. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mokdad AA, Murphy CC, Pruitt SL, et al. Effect of hospital safety net designation on treatment use and survival in hepatocellular carcinoma. Cancer. 2017 doi: 10.1002/cncr.31066. [DOI] [PubMed] [Google Scholar]
- 31.Wong RJ, Devaki P, Nguyen L, et al. 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;20:528–35. doi: 10.1002/lt.23820. [DOI] [PubMed] [Google Scholar]
- 32.Yopp AC, Mansour JC, Beg MS, et al. Establishment of a Multidisciplinary Hepatocellular Carcinoma Clinic is Associated with Improved Clinical Outcome. Ann Surg Oncol. 2014;21:1287–95. doi: 10.1245/s10434-013-3413-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Singal AG, Tiro JA, Marrero JA, et al. Mailed Outreach Program Increases Ultrasound Screening of Patients With Cirrhosis for Hepatocellular Carcinoma. Gastroenterology. 2017;152:608–615 e4. doi: 10.1053/j.gastro.2016.10.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Singal AG, Mittal S, Yerokun OA, et al. Hepatocellular Carcinoma Screening Associated with Early Tumor Detection and Improved Survival Among Patients with Cirrhosis in the US. Am J Med. 2017;130:1099–1106 e1. doi: 10.1016/j.amjmed.2017.01.021. [DOI] [PubMed] [Google Scholar]
- 35.Foster MW, Sharp RR. Race, ethnicity, and genomics: social classifications as proxies of biological heterogeneity. Genome Res. 2002;12:844–50. doi: 10.1101/gr.99202. [DOI] [PubMed] [Google Scholar]
- 36.Peters NA, Javed AA, He J, et al. Association of socioeconomics, surgical therapy, and survival of early stage hepatocellular carcinoma. J Surg Res. 2017;210:253–260. doi: 10.1016/j.jss.2016.11.042. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.