Assessment of Incidence of and Surveillance Burden for Hepatocellular Carcinoma Among Patients With Hepatitis C in the Era of Direct-Acting Antiviral Agents

“[Introduction]
Hepatocellular carcinoma (HCC) is the fastest rising cause of cancer-related mortality in the US. The incidence of HCC has increased over the last 2 decades owing to the hepatitis C virus (HCV) epidemic. Most HCC cases are diagnosed in advanced stages, with a median survival less than 1 year. Regular surveillance for HCC may help improve early cancer detection rates when curative treatment can be applied and is recommended in patients with HCV-associated cirrhosis.

The availability of direct-acting antiviral agents (DAAs) for HCV treatment has substantially altered the landscape of HCV. Though new DAA regimens can result in a virological cure (ie, sustained virological response [SVR]) in more than 90% of patients with HCV, many patients remain at risk of developing HCC after virological cure. Current clinical guidelines recommend that patients with virologically cured HCV with cirrhosis (or advanced fibrosis) to undergo routine HCC surveillance every 6 months.

[Results]
This decision analytical model study was conducted from January 2019 to February 2020. In 2012, the annual incidence of HCC among patients with HCV (with viremia and virologically cured HCV) was 18 000 (95% UI, 11 000-32 000). The annual incidence of HCV-associated HCC is projected to increase to 24 000 (95% UI, 18 000-31 000) cases in 2021 and decrease to 13 000 (95% UI, 11 000-16 000) cases by 2040. In 2012, 1000 (95% UI, 500-2100) new HCC cases (5.3% of all new cases) developed in patients with virologically cured HCV; this number is projected to peak at 7000 (95% UI, 5000-9600) new cases (35.0% of all new cases) in 2031 and then decrease to 6000 (95% UI, 4300-8300) new cases (45.8% of all new cases) by 2040. Most of the HCC cases would develop in patients with cirrhosis. The number of new HCC cases among patients with cirrhosis who achieved virological cure increased substantially from 700 (95% UI, 300-1700) in 2012 to 4500 (95% UI, 3000-6700) in 2040.

[..] Our model projected that in 2012, a total of 1.24 million (95% UI, 0.96 million-1.71 million) patients with HCV (with viremia and virologically cured HCV) were at risk of developing HCC and would be candidates for routine HCC surveillance in the US. The number of candidates eligible for HCC surveillance is estimated to increase to 1.49 million (95% UI, 1.23 million to 1.85 million) in 2020 and then gradually decrease to 0.83 million (95% UI, 0.66 million to 1.04 million) by the end of 2040. In 2012, 106 000 (95% UI, 70 000-178 000) surveillance candidates were those who had achieved virological cure; this number is projected to peak at 649 000 (95% UI, 512 000-824 000) in 2030 and then decrease to 539 000 (95% UI, 421 000-687 000) individuals by 2040. However, the proportion of all HCC surveillance candidates who are virologically cured is estimated to increase from 8.5% in 2012 to 64.6% in 2040. We also projected that the proportion of surveillance candidates with cirrhosis (compensated and decompensated) is estimated to increase from 42.8% in 2012 to 49.6% in 2040, and that the proportion of patients with cirrhosis among patients with virologically cured HCV is estimated to increase from 34.5% to 43.5% in the same time period.

[..] The number of HCC surveillance candidates increased in the privately insured pool to 376 000 (95% UI, 277000-509000) patients (25.9%) by 2016 and decreased in the uninsured pool to 127 000 (95% UI, 96 000-168 000) patients (8.8%) because of the implementation of the Affordable Care Act in 2014. The number of HCC surveillance candidates covered by Medicare increased further because of the aging population of patients with HCV (with viremia and virologically cured HCV) and is expected to surpass the number of candidates who are privately insured from 2021 to 2040. Among all payers, the state-administered programs (Medicaid and state prisons) is estimated to have the highest burden of HCC surveillance: the number of HCC surveillance candidates incarcerated in state prisons or covered by Medicaid is projected to peak at 658 000 (95% UI, 557 000-786 000) in 2020 and decrease to 349 000 (95% UI, 280 000-427 000) by the end of 2040.

[Discussion]
[..] With the availability and wider use of new DAAs for HCV treatment, the burden of HCC and routine HCC surveillance attributable to HCV could change in the era of DAAs. In this modeling study, we found that the incidence of HCC is estimated to continue to increase until 2021. Between 2012 and 2040, 583 000 patients with HCV (with viremia and virologically cured HCV) were estimated to develop HCC, and 27% of those cases would be among patients with virologically cured HCV. The number of candidates eligible for HCC surveillance is estimated to increase from 1.24 million in 2012 to 1.49 million in 2020, while the burden of surveillance is expected to shift from patients with viremia to individuals who achieved virological cure. In 2012, 9% of all candidates for HCV surveillance were virologically cured, which is projected to increase to 65% by 2040. The average age of HCC incidence and surveillance candidates is also projected to shift from 55 years in 2012 to more than 70 years in 2040.

[..] Most of the patients with virologically cured HCV will transition from receiving care from liver specialists to receiving care from primary care physicians. However, most people do not receive regular surveillance in primary care settings and the knowledge of HCC surveillance in primary care settings remains low. Therefore, it is vital to emphasize that biannual HCC surveillance is warranted for these patients in primary care settings.

The optimal clinical management of patients with virologically cured HCV remains unclear. While the HCC management guidelines by the European Association for the Study of the Liver recommends routine HCC surveillance in patients with virologically cured HCV having advanced fibrosis and cirrhosis, the guideline by the American Association for the Study of Liver Diseases recommends routine HCC surveillance only in individuals with cirrhosis without specifying the status of virological cure.

The optimal surveillance strategy in patients with virologically cured HCV will rely on more evidence and studies in the following 3 aspects. First, long-term data are needed for a better understanding of the risk of HCC after DAA-induced virological cure in patients with HCV. It is unclear how much of the current knowledge of the HCC risk among patients with virologically cured HCV from an interferon-based regimen could be used to inform surveillance policies in the era of DAAs. Some studies suggest that patients with HCV cured by DAA regimens could have higher HCC risks than those with HCV cured by interferon-based regimens, but the increase was not found to be significant after adjusting for confounders, and the difference was not found to be conclusive. In addition, compared with the available data in the interferon-based treatment era, the observations in DAA era are subject to shorter follow-up time and present an older at-risk population.

Second, health economic data are needed to identify cost-effective HCC surveillance policies for patients with virologically cured HCV. The primary reason for not recommending routine surveillance in patients with virologically cured HCV with advanced fibrosis may be that biannual surveillance using ultrasonography and α1-fetoprotein testing for the lifetime was not found to be cost-effective in this cohort. Instead of a general surveillance, further research is warranted to identify risk-based surveillance in these patients that could provide a good use of limited resources. In addition, the optimal age for stopping surveillance is unknown. It is plausible that patients with virologically cured HCV may not need HCC surveillance throughout their remaining lifetime.

Third, tailoring surveillance policies to individual-level factors may be a useful approach, especially as emerging molecular biomarkers show their value in the prediction of HCC risks in addition to current clinical risk factors. Comprehensive risk prediction models can be developed by integrating multilevel risk factors to further refine the risk predictions. A tailored surveillance strategy could differentiate the patients at high risk who can benefit more from surveillance compared with the others, and thus further improve the cost-effectiveness and efficiency of medical resource use.”

Full article, Chen Q, Ayer T, Adee MG et al. JAMA Network Open, 2020.11.18