Hepatocellular Carcinoma: A Roadmap to Reduce Incidence and Future Burden

The incidence of hepatocellular carcinoma (HCC) has sharply risen over the past 20 years in the United States and is now a leading cause of cancer-related death (1). More than 90% of HCC cases in the United States occur in patients with cirrhosis, the end-result from any chronic liver disease including hepatitis C (HCV), hepatitis B (HBV), alcohol-related liver disease, and nonalcoholic fatty liver disease (NAFLD). However, HCC does not occur uniformly across populations and disproportionately affects racial/ethnic minorities, with blacks, Hispanics, and Asians experiencing higher age-specific incidence rates than non-Hispanic whites.

In two studies recently published in the Journal, Han et al. and Pham et al. (2,3) used data from the Surveillance, Epidemiology, and End Results program of cancer registries to characterize racial/ethnic differences in incidence and projected future HCC burden. Incidence rates among most racial/ethnic groups have steadily increased, and both studies project blacks and Hispanics to have the highest rates by 2030. The largest increase in incidence will occur among adults older than age 65 years, particularly among Hispanics. In contrast, incidence among Asians, who had the highest rate in 2000, has declined since 2010. Among subgroups of Asians, Pham et al. (3) observed differences in incidence, with declining incidence rates in Chinese men but increasing rates in others, specifically Japanese and Filipino men and Vietnamese and Laotian women.

Declines in HCC incidence among Asians are likely related to improvements in global HBV vaccination and treatment programs. Some Asian countries, such as Taiwan and China, have implemented neonatal HBV vaccination programs and antiviral treatment of HBV-infected individuals, with statistically and clinically significant reductions in HBV-related HCC (4). Differences among Asian subgroups may also be related to nativity (ie, US vs foreign-born) and prevalence of HBV infection (5). For example, a higher proportion of Southeast Asians are foreign-born and from HBV-endemic regions; by contrast, a higher proportion of Japanese are born in the United States and fewer have HBV infection.

Beyond differences within Asian subgroups, increasing HCC incidence among other racial/ethnic groups is primarily related to an aging population with chronic HCV infection and a growing cohort of patients with NAFLD. In the United States, HCV infection is more prevalent among the “baby boomer” cohort, those born between 1945 and 1965, consistent with the cohort effects observed by Han et al. (2). Individuals in this cohort were likely exposed to HCV infection in the 1970s and 1980s though injection drug use and contaminated blood products prior to routine HCV screening. The introduction of highly effective direct acting antiviral agents for HCV infection, in parallel with US preventive services task force recommendations for one-time HCV screening for all baby boomers, has facilitated the treatment and cure of many HCV-infected individuals during the past 5 years. HCV treatment significantly reduces (although does not eliminate) risk of HCC, and most patients in academic centers and large integrated health systems with known HCV infection have now received treatment (6). Therefore, it is possible Han and colleagues’ projections of HCC burden using data from 2000 to 2013 may not account for recent HCV screening and treatment efforts, thereby overestimating future HCC burden. Recent studies suggest some countries in Europe, where incidence rates had previously increased, are now experiencing plateaus and even declines in HCC incidence, underscoring the importance of incorporating more recent data in projections of HCC burden (7).

Additional studies are also needed to truly understand racial/ethnic disparities in HCC incidence and burden. Just as Pham et al. (3) highlight the danger of aggregating trends among a heterogeneous group of Asians, similar analyses are needed for Hispanics. For cancers other than HCC, prior studies have demonstrated differences in incidence and prognosis between Hispanic subgroups such as Mexicans, Puerto Ricans, Cubans, and Dominicans (8). It may also be important to consider nativity when examining trends in HCC incidence because others have suggested incidence rates may differ between US-born and foreign-born Hispanics (9). Second, several studies have highlighted racial/ethnic disparities in HCC incidence, but all have ignored the interaction between race/ethnicity and socioeconomic status, despite data demonstrating high correlations between the two. Ignoring intersectionality may falsely attribute socioeconomic disparities to race/ethnicity and vice versa, and larger studies including both are needed (10). Finally, we do not yet know how disparities in HCC incidence translate to HCC-related mortality. A recent study demonstrated racial/ethnic disparities in survival among HCC patients, with blacks and Hispanics having significantly worse survival than non-Hispanic whites (11). However, after adjusting for factors such as type of medical insurance, Child Pugh class, Barcelona Clinic Liver Cancer stage, and receipt of HCC treatment, only blacks had higher mortality, and Hispanics had lower mortality, compared with whites. Further studies are needed to validate these findings and identify interventions to improve HCC prognosis among racial/ethnic minorities.

Moving forward, incidence trends reported by Pham and Han (2,3) will inform interventions to reduce HCC burden in the United States. Continued efforts to increase HBV vaccination and treatment will help further reduce HBV-related HCC and reinforce declines in HCC incidence observed in several Asian ethnic groups. Only one-half of all infants and less than 10% of actively infected individuals globally receive timely birth-dose vaccination or antiviral treatment, respectively (12). Similarly, efforts to increase HCV screening and treatment, particularly among baby boomers, will prevent HCV-related HCC, which may particularly benefit blacks and non-Hispanic whites, in whom HCV infection is the most common risk factor for HCC. Nearly one-half of all HCV-infected individuals have not been screened and remain unaware of their infection (13). Other patient subgroups, such as incarcerated individuals, those not engaged in routine medical care, active drug users, and Medicaid-eligible individuals also have a higher prevalence of untreated HCV infection. Unlike viral hepatitis, no pharmacologic treatment currently exists to treat NAFLD and subsequently reduce HCC risk, making this a particular area of need as NAFLD becomes the most common cause of cirrhosis and HCC in the United States. Developing a chemopreventive agent for patients with NAFLD can decrease NAFLD-related HCC incidence. This may be especially beneficial for Hispanics, who experience a greater burden of NAFLD-related HCC than other racial/ethnic groups due to a combination of genetic predisposition and environmental risk factors (eg, metabolic syndrome) (14). While awaiting pharmacological therapy, continued campaigns for lifestyle modifications including healthy diets, increased exercise, and decreased alcohol abuse may help. Although the projected burden of HCC in the United States is alarming, this future is not inevitable. The incidence trends observed by Han et al. and Pham et al. (2,3) paint a roadmap to reduce the increasing incidence and disparities in HCC in the United States.

Funding

This work was conducted with support from NIH RO1 MD12565 (Singal and Murphy) and CPRIT PP170121 (Singal).

Notes

Affiliations of authors: Department of Internal Medicine (AGS), Department of Clinical Sciences (AGS, CCM), and Harold C. Simmons Cancer Center (AGS, CCM), UT Southwestern Medical Center, Dallas, TX.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Conflicts of interest: Dr Singal was on the speakers bureau for Gilead and has received grant support from Abbvie. Dr Murphy has no relevant conflicts of interest.