The impact of HCV eradication on hepatocellular carcinogenesis

Abstract

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related death in the world. Infection with the hepatitis C virus (HCV) represents one of the most common risk factors for HCC development and cases of HCV-related complications have been rising over the last two decades. Though the standard of HCV therapy has been interferon (IFN)-based for many years, the therapeutic revolution spurred by the development of direct-acting antivirals (DAAs) promises to usher in a new era in which chronic hepatitis C becomes a rare disease. Based on long term follow-up in persons experiencing IFN-based sustained virologic responses (SVR), we can expect that rates of HCV-associated HCC will decrease significantly after the widespread adoption of DAAs, but there remains a persistent risk for HCC even among some patients with advanced fibrosis who have achieved SVR. As such, individuals treated for HCV with advanced fibrosis should continue to be screened regularly for HCC post-SVR. Furthermore, as the population of patients with SVR grows, it will become imperative to accurately identify those individuals at high-risk for developing HCC, appropriately allocate resources for screening, and consider cost-effective chemopreventive strategies. Risk factors include pre-existing advanced fibrosis/cirrhosis, older age, diabetes mellitus, and ethanol use. Additionally, laboratory biomarkers and genetic signatures are currently being identified that not only predict the likelihood of HCC development in SVR patients but may also serve as dynamic indicators of therapeutic response.

I. Introduction

Primary liver cancer, of which hepatocellular carcinoma (HCC) represents the most common histologic type, is one of the most common malignancies worldwide, comprising 5.6% of all cancers¹. While it is the fifth most common cancer globally, it is also the second most common cause of cancer death¹. Though survival rates for HCC have been improving over the last 30 years, they remain quite poor despite improving surveillance programs and therapeutic options. 5-year survival rates currently stand at ~15% in the United States (US)², but only ~5% in resource-poor countries³. In the US, HCC was the fastest-growing cause of cancer-related death in men and women between 1992 and 2010², though recent studies suggest that the rate of rise in incidence may be slowing⁴. Still, an estimated 33,000 new cases of HCC and 23,000 HCC-related deaths occurred in 2014 in the US alone⁵.

One of the most significant risk factors for the development of HCC is chronic infection with hepatitis C virus (HCV). In the US, 1.3% of the population, or 3.2 million people are thought to have chronic hepatitis C, many of whom are not aware that they are infected⁶. HCV is an RNA virus with a purely cytoplasmic life cycle, unlike the hepatitis B virus, which directly integrates into the genetic material of the cell. HCV-associated hepatocarcinogenesis is thought to be multifactorial, arising from direct viral oncogenic effects as well as from mutagenic insults to the hepatocyte genome accumulated from rounds of inflammation and fibrosis associated with chronic viral infection⁷. HCV infection itself is associated with a 15–20 fold increased risk of HCC development; observational data has suggested an annual incidence of HCC in individuals with HCV-associated cirrhosis of 1–5%⁸. Forecasting models predict that without treatment, 14.4% of all patients with HCV would develop HCC⁹. Thus, without further advances in screening or therapeutics, the incidence of HCC will continue to climb in the coming decades.

The goal of HCV therapy has been achievement of permanent clearance of the virus (reflected by undetectable HCV RNA using highly sensitive assays), also known as sustained virologic response (SVR). SVR has been shown to be associated with dramatic reductions in all-cause mortality and progression of liver disease¹⁰. The therapeutic landscape of HCV has undergone a sea change with the recent introduction of direct acting anti-viral agents (DAAs) and approval of all-oral interferon-free DAA regimens that now promise superb (>90%) SVR rates¹¹. As such, the rates of HCV-associated complications, including HCC, are expected to decrease in the near future. In this review, we will discuss the latest advances in HCV therapeutics, the impact of established interferon-based strategies on HCC incidence, risk factors for developing HCC after SVR, and how the arrival of DAAs and interferon-free regimens will impact the surveillance and management of patients at risk and with HCC in the near future.

II. The rise of direct acting antiviral agents (DAAs) in HCV treatment

Since its introduction in 1986, interferon (IFN)-based therapy has been the backbone of HCV therapy. However, despite advancement and refinement of IFN-based regimens over two decades, SVR rates of only 55% were observed¹². Furthermore, SVR rates varied with genotype and viral load and the regimen itself was poorly tolerated.

Significant advances in our understanding of the molecular virology, life cycle, and pathogenesis of HCV led directly to the development of the first DAAs. These agents were the first-generation NS3-4A protease inhibitors, telaprevir and boceprevir, which achieved SVR rates of 65–75% when used together with pegylated-IFN (PEG-IFN) and ribavirin^13,14. As such, they were approved by the FDA for use in “triple therapy” for HCV genotype 1 in 2011. Their approval was followed by a flurry of activity, notable for the development of several compounds targeting other stages of the HCV life cycle. Simeprevir, a once daily NS3-4A protease inhibitor, was approved in 2013 to be used in combination with PEG-IFN and ribavirin for treatment of HCV genotype 1, achieving comparable SVR rates as its predecessors, with better tolerability¹⁵. A major advance was the development of an NS5B polymerase inhibitor, sofosbuvir. Sofosbuvir is a member of a family of nucleotide analogues that work by causing early chain termination after being incorporated into newly synthesized viral RNA¹⁶. Given its mechanism of action and the conservation of the NS5B RNA polymerase active site, it is active against all HCV genotypes and has a high barrier to resistance, selecting only for viral mutants with exceedingly low replication fitness. As such, in a landmark trial enrolling individuals with predominantly HCV genotype 1 or 4, sofosbuvir-anchored triple therapy was found to achieve SVR rates of 90% after 12 weeks of therapy (SVR12)¹⁷. Moreover, SVR12 rates of 95% and 82% were attained with sofosbuvir and ribavirin alone in treatment-naïve and treatment-experienced persons, respectively with HCV genotype 2 or 3^17,18. Accordingly, in 2013, the FDA approved sofosbuvir for use as part of triple therapy for HCV genotypes 1 and 4 and with ribavirin alone for genotypes 2 and 3.

Of particular interest has been the development of all-oral IFN-free regimens utilizing two or more classes of DAAs to achieve the dual goal of rapid viral suppression and prevention of selection of resistant variants. This concept has been realized with the approval of sofosbuvir and ledipasvir (an NS5A inhibitor) by the FDA in October 2014 as a once daily co-formulation for the treatment of HCV genotype 1. This was done on the basis of three pivotal trials that studied this combination with and without ribavirin in both treatment-naïve and treatment-experienced patients. These clinical studies found that irrespective of ribavirin use, individuals treated with this combination achieved SVR12 rates of 94–99%^19–21. In addition, sofosbuvir and simeprevir for the treatment of HCV genotype 1 was approved by the FDA in November 2014, based on results from the COSMOS trial, which demonstrated >90% SVR12 rates and good safety and tolerability profiles²². Finally, the combination regimen of ombitasvir (NS5A inhibitor), ritonavir-boosted paritaprevir (NS3-4A protease inhibitor), dasabuvir (non-nucleoside NS5B inhibitor) and ribavirin was approved in December 2014 on the basis of several trials showing SVR12 rates >90%^23,24. Several other drugs have been recently approved as part of new regimens (see Table 1), and still others are currently in late phase clinical trials.

Table 1.

AASLD/IDSA recommended drug regimens for initial HCV treatment as of mid-2015.

Genotype	Treatment Regimen	Duration (weeks)	SVR	Estimated Cost	Level of evidence
1	Sofosbuvir/Ledipasvir	8 (NC, low VL)	94%21	$63,000	Class I, Level A
		12 (NC, high VL)	98%19	$94,500	Class I, Level A
		24 (LC)	99%20	$189,000	Class I, Level A
	rParitaprevir/Ombitasvir/Dasabuvir/Ribavirin	12 (NC)	96%24	$83,319	Class I, Level A
		24 (LC)	96%23	$166,638	Class I, Level A
	Sofosbuvir/Simeprevir ± Ribavirin	12 (NC)	92%22	$150,360	Class IIa, Level B
		24 (LC)	93%22	$300,720	Class IIa, Level B
2	Sofosbuvir/Ribavirin	12 (NC)	95%17	$86,500	Class I, Level A
		16 (LC)	83%17	$115,333	Class IIb, Level C
3	Sofosbuvir/Ribavirin	24	93%59	$173,000	Class I, Level B
4	rParitaprevir/Ombitasvir/Ribavirin	12	100%60	~$55,000	Class I, Level B
	Sofosbuvir/Ledipasvir	12	95%*	$94,500	Class IIb, Level B
	Sofosbuvir/Ribavirin	24	96%17	$173,000	Class IIa, Level B
5	Sofosbuvir/Ribavirin + PEG-IFN	12	100%17	$94,000	Class IIa, Level B
6	Sofosbuvir/Ledipasvir	12	100%17	$94,500	Class IIa, Level B

^*Kapoor et al, Hepatology 2014; 60:321A

VL = viral load; rParitaprevir = ritonavir-boosted paritaprevir; NC = non-cirrhotic; LC = liver cirrhosis

Published data regarding real-world experience with the new DAA regimens are scarce at this time. However, preliminary data from the HCV-TARGET group, a multicenter consortium prospectively characterizing the outcomes of practices in North America and Europe, paints an encouraging picture. Individuals with HCV genotype 1 treated with sofosbuvir and simeprevir ± ribavirin experienced similar rates of SVR compared to those seen in clinical trials (~90% at SVR4) as well as very low rates of viral breakthrough, discontinuation, and serious adverse effects (Jensen et al., Hepatology 2014; 60:219A). In the near future, we should expect to have comparable real-world data regarding the efficacy and safety profiles of other DAA regimens.

Current areas of active investigation include expanding the reach of DAA-based treatment to other HCV genotypes, HCV/HIV co-infected patients, persons with decompensated cirrhosis, and those experiencing HCV infection after liver transplantation. Promising data for all oral DAA combinations in each of these groups has been published or presented. In the coming years, we can anticipate that the landscape of HCV therapeutics will continue to rapidly evolve.

III. Risk reduction of HCC in individuals achieving SVR with IFN-based therapies

The achievement of SVR is a critical point in the progression of disease for individuals with chronic HCV infection. Recent studies investigating long-term clinical endpoints have found that for patients with all stages of fibrosis, SVR is associated with marked reductions in all-cause mortality²⁵, progression of fibrosis²⁶, need for liver transplantation¹⁰, extrahepatic complications²⁷, and hepatocarcinogenesis.

The link between chronic HCV infection and HCC was observed in the mid-1980s. Shortly after the identification of HCV, researchers in Europe and Asia later reported prevalence of anti-HCV antibodies in 58–75% of patients with HCC, whose liver disease was previously associated with non-A, non-B hepatitis²⁸. However, direct evidence of the benefits of antiviral therapy with regard to hepatocarcinogenesis were not apparent until Nishiguri et al²⁹ directly addressed this question with a randomized control trial. In this study, 90 Japanese patients with chronic HCV infection and compensated cirrhosis were randomized 1:1 to IFN monotherapy or symptomatic therapy and followed prospectively. After a mean observation period of 4.4 years for treated patients and 5.5 years for controls, only 2 treated patients developed HCC compared to 17 controls (RR 0.067, P = 0.01). Long-term follow-up of the same cohort revealed continued reduction in HCC development among treated patients (RR 0.244, P < 0.001), suggesting a durable amelioration of hepatocarcinogenic risk after antiviral therapy³⁰. Since then, several long-term observational and retrospective studies have corroborated these initial findings, though most of these studies have been performed in Japan where the prevalence of HCV-associated HCC is higher³¹.

In a more contemporary multicenter long-term follow-up study performed in Europe, 530 patients with biopsy-proven advanced fibrosis treated with IFN monotherapy, IFN and ribavirin, or PEG-IFN and ribavirin were followed for a median duration of 8.4 years. In addition to a significant improvement in all-cause mortality rate in patients with SVR (HR 0.26, P < 0.001), significantly fewer individuals with SVR developed HCC compared to non-responders (HR 0.19, P < 0.001)¹⁰. Similarly, a recent systematic meta-analysis synthesized the results from 30 observational studies performed in Asia, Europe, Canada, and the United States assessing the risk of HCC among HCV-infected persons treated with interferon-based therapy. Evaluating 25,906 participants, with follow-up ranging from 3.0–8.2 years, the meta-analysis found that among those who attained SVR, 1.5% developed HCC compared to 6.2% of individuals who did not attain SVR (adjusted HR 0.24). In a sub-analysis examining patients with advanced liver disease, 4.2% of responders developed HCC compared to 17.8% of non-responders³². The authors argue that this would suggest that though treating HCV at earlier stages of disease is more efficient in HCC prevention, more cases of HCC are prevented when individuals with advanced liver disease achieve SVR. In both studies, HCC was defined by histopathologic confirmation or by characteristic imaging findings on CT, MRI, or contrast-enhanced ultrasound (i.e. arterial-phase enhancement) in the setting of markedly elevated and/or increasing α-fetoprotein (AFP).

The suppressive effect of antiviral therapy has also been evaluated in the setting of curative resection of HCV-associated HCC. Though it has long been known that IFN therapy does not represent a viable option as a therapy for HCC, emerging data suggests that IFN therapy may have benefit in preventing HCC recurrence after viral cure. A recent single-center study evaluating 178 patients who underwent PEG-IFN/ribavirin combination therapy after curative treatment for HCV-related HCC found significant reductions in HCC recurrence, particularly in those who achieved SVR³³. A meta-analysis of 14 studies encompassing 1385 patients addressed the same question by examining the utility of IFN-based adjuvant therapy following curative resection or ablation of HCV-associated HCC³⁴. The authors found that in HCV-associated HCC, IFN treatment reduced mortality (RR 0.44, P < 0.00001) and decreased HCC recurrence by ~30%. There is also preliminary evidence that IFN-based therapy prior to surgical resection helps to decrease HCC recurrence rates³⁵. The mechanism by which interferon is able to provide additional benefit, particularly post-procedure, is unclear. It is likely a combination of both the anti-proliferative effects as well as the direct antiviral effects of IFN. Furthermore, IFN has direct anti-tumor effects including induction of pro-apoptotic genes via mechanisms not limited to upegulation of tumor suppressors, inhibition of angiogenesis, as well as upregulation of antitumor immune responses³⁶. IFN-α, the principal class of IFN used in HCV treatment, has been shown to upregulate natural killer (NK) cell activity and increase circulating levels of T helper lymphocytes, thereby augmenting the anti-tumor response by both innate and adaptive immune mechanisms³⁷. Interestingly, IFN-γ, whose receptors are primarily expressed on hepatocytes, has been shown to have multiple anti-tumor effects including upregulation of CD8 T cell and NK cell expression and function in mouse models³⁸. As such, IFN-γ has recently become a subject of interest as a novel tool for the treatment of HCV and HCC. Whether similar pleiomorphic benefits will be observed with DAA-based treatment, which have antiviral but no clear direct anti-tumor or anti-inflammatory effects, remains an open question. Furthermore, it is unclear whether or not such therapy will benefit patients who already have late-stage or metastatic HCC given their ineligibility for curative interventions.

In general, there are limitations to the data supporting long-term benefits of SVR related to HCC development. These include the use of predominantly observational and retrospective methods and the inherent biases that attend these approaches. Further weaknesses include the use of small cohorts, single center studies, and insufficient follow-up time given the gradual and asymptomatic pathogenesis of HCV-related liver disease, which is often measured in decades. Many of these studies have also been from Japan, where HCC incidence rates are significantly higher than in the United States/Europe³⁹.

Nevertheless, the data over the past two decades clearly demonstrate that the achievement of SVR is associated with a significant decrease in the risk of HCC development. Moreover, antiviral therapy can act as tertiary prophylaxis, reducing the risk of HCC recurrence after curative resection or transplant. With the high SVR rates achieved with DAA-based regimens, the overall incidence of HCC can thus be expected to continue to decrease. However, there still remains a non-zero risk of hepatocarcinogenesis that has been consistently observed in many studies. Cumulative HCC incidence in patients with SVR at 5 and 10 years post-SVR have ranged in recent studies from 2.3–8.8% and 3.1–11.1%, respectively³¹. In a prospective multicenter study that evaluated the long-term effect of IFN-based therapy, the authors found that 50% of those with SVR who developed HCC had HCC discovered 7 or more years after being clear of virus⁴⁰. Isolated reports of HCC developing as far as 13 years after achievement of SVR have been published⁴¹. This strongly suggests that for selected individuals, despite the resolution of the initial infection, there remains an elevated risk of hepatocarcinogenesis over that of healthy individuals.

IV. Risk factors for HCC in patients with SVR

Given the continued risk of hepatocarcinogenesis even after SVR, it is critical to be able to identify those individuals at highest risk in order to intervene with more aggressive screening and follow-up. This is especially true given the increasing number of patients who will achieve SVR with DAA-based regimens.

There have been many studies identifying predictors for HCC development in patients with chronic HCV infection but fewer have sought to identify predictors of HCC development in patients after SVR achievement. The most well established risk factor for developing HCC after SVR in HCV patients is advanced fibrosis or cirrhosis⁴². HCC rarely develops in HCV infection in the absence of cirrhosis. Consistent with this, a subgroup analysis of trials enrolling only patients with advanced fibrosis and cirrhosis found that 4.2% of these patients achieving SVR developed HCC compared to 1.5% when analyzing trials that enrolled individuals at all stages of fibrosis³². There is also emerging data that patients with underlying diabetes mellitus are also at higher risk of HCC after SVR⁴³. This may reflect the independent risk for HCC development incurred in patients with diabetes, though the mechanism by which this occurs is unclear and likely complex. Several other risk factors have also been consistently associated with HCC development in SVR patients including older age, male sex, and ethanol use³¹.

In an effort to guide prediction of high-risk SVR patients, several laboratory biomarkers have been identified to predict the likelihood of hepatocarcinogenesis in these patients. For example, in one study of 140 SVR patients with advanced fibrosis/cirrhosis from the HALT-C trial, the authors found that pre-IFN treatment platelet count, alkaline phosphatase, and older age were significantly associated with HCC development⁴⁴. Researchers have also identified post-treatment alanine aminotransferase (ALT) and AFP as possibly important biomarkers. In one study, a post-treatment ALT <40 IU/L and an AFP <6.0 ng/mL were identified to have high negative predictive values for HCC development. These markers identified those at high-risk for developing HCC in SVR patients and in non-responders⁴⁵. In another study, older age and a post-treatment AFP >5 ng/mL measured 24 weeks post-interferon treatment was significantly predictive of development of HCC in both SVR and non-responders⁴⁶. A recent retrospective study of non-cirrhotic patients reported that an elevated FIB-4 index⁴⁷, a predictor of liver fibrosis using a composite value of ALT/AST/age/platelet count, at SVR24 is also predictive of HCC development in SVR patients⁴⁸. Table 2 summarizes risk factors that have been identified for HCC incidence in patients with SVR. Moving forward, further studies identifying risk factors and suitable biomarkers to identify SVR patients at higher risk of HCC development will be needed, particularly in Western populations as the vast majority of investigations to date have been performed in Asia.

Table 2.

Risk factors for HCC development in treated HCV patients with SVR

Publication	SVR Patients	Observation period (yrs)	Risk factor	HR	95% CI	p value
Makiyama et al. 200442	1197	5.9	Age ≥50	7.38	1.74–31.33	0.007
			Male	5.50	1.29–23.44	0.02
			F3/F4 fibrosis	2.34	1.06–5.16	0.03
Ikeda et al. 200561	1056	4.7	Age ≥60	3.13	1.32–7.42	0.01
			AST > 100	3.10	1.31–7.31	0.01
			Plt <150	2.78	1.07–7.20	0.002
Chang et al. 201262	871	3.4	F3/F4 fibrosis	3.95	1.46–10.70	0.007
			Age ≥60	3.82	1.74–8.37	0.001
			Post-IFN AFP ≥20	3.15	1.60–6.19	0.001
			Plt <150	2.81	1.22–6.44	0.015
Arase et al. 201363	1751	8.1	Diabetes	4.76	1.60–14.10	0.005
			Male	3.42	1.01–11.63	0.049
			Alcohol intake	2.68	1.14–6.34	0.049
			Age, every 10 years	2.60	1.48–4.58	0.001
Oze et al. 201446	1425	3.3	Post-IFN AFP ≥5	8.10	2.74–23.94	<0.001
			Age ≥65	5.81	1.12–30.07	0.036
Yamashita et al. 201464	562	4.8	F2/F3/F4 fibrosis	10.7	2.2–192.1	<0.001
			Age ≥50	4.1	1.4–17.4	<0.01
			Ethanol ≥30 g/day	3.9	1.7–9.0	<0.01
			Pre-IFN AFP ≥8	2.6	1.2–6.1	<0.05
Huang et al. 201465	642	4.4	GGT ≥75 U/L	5.78	1.94–17.38	0.002
			Age ≥65	4.14	1.43–11.97	0.009
			F2/F3 fibrosis	3.74	1.04–13.52	0.04
Toyoda et al. 201548	522	7.2	Diabetes	2.08	1.02–4.01	0.05
			FIB-4 index	1.73	1.09–2.86	0.02
Chang et al. 201566	801	5.0	Age ≥60	3.75	1.82–7.73	<0.001
			Post-IFN AFP ≥20	2.80	1.55–5.08	0.001
			Plt <150	2.67	1.36–5.24	0.004
			F3/F4 fibrosis	2.24	1.12–4.50	0.023

Units: AFP (ng/mL), Platelets (10⁹/L), GGT (U/L)

In recent years, the availability of high-throughput sequencing and gene expression profiling methodologies has fueled an emerging effort to identify genetic biomarkers predictive of clinical outcomes in patients at high likelihood for developing complications of cirrhosis. Gene expression profiling has been performed on fixed tissues from livers of patients who underwent surgical resection of primary HCC. This analysis identified a reproducible 186-gene signature that correlated with survival⁴⁹. Interestingly, molecular analysis also suggested that the signature might reflect the “field defect” in the cirrhotic liver, in which recurrent HCC arises from a favorable microenvironment. This genetic signature was also prospectively studied in individuals with HCV-related Child-Pugh class A cirrhosis. Liver biopsies were obtained from patients who were subsequently followed for a median of 10 years. A genetic signature predicted to have “poor prognosis” was found to be significantly associated not only with mortality (HR 2.77, P = 0.004), but also with progression to advanced cirrhosis (HR 3.79, P < 0.001) and development of HCC (HR 2.65, P = 0.009)⁵⁰. This data suggests that the 186-gene signature is a sensitive predictor of clinical deterioration even in early stage cirrhosis and may identify those high-risk patients who would be most likely to benefit from therapeutic intervention. Repeat biopsy and re-evaluation of the genetic signature after therapy could also serve as a method to identify those at persistent risk for HCC. Single nucleotide polymorphisms (SNPs) in several genes including EGF, PNPLA3, and IL28B have also been identified that are associated with clinical deterioration, including HCC development, in HCV-associated cirrhosis⁵¹. It remains to be seen whether similar SNPs or other genetic markers will be able to identify patients at risk of further clinical deterioration after SVR.

V. Screening, Chemoprevention, and Conclusions

Current AASLD guidelines for HCC surveillance recommend that patients with chronic hepatitis C and cirrhosis should be screened using ultrasonography every 6 months⁵². Notably, AFP is not recommended as a screening tool given its poor sensitivity and positive predictive value. However, it is unclear whether or not to continue screening individuals who have achieved SVR. Given that the risk of disease progression, including HCC development, is reduced but not eliminated after viral clearance, we recommend continuing regular screening with ultrasonography even after successful treatment. Whether post-treatment AFP will be incorporated into post-SVR screening will depend on the reproducibility of some of the preliminary studies described above that suggest a predictive benefit of this value for the development of post-SVR HCC. No direct prospective comparison of regular post-SVR screening versus usual care has been performed. However, in a retrospective cohort study of 562 patients with SVR, the 5-year survival rate of those that received regular HCC screening (ultrasonography at least every 6 months) was 93% compared to 60% of those that did not.

However, surveillance is thought to only be cost-effective in patients with cirrhosis if annual HCC incidence is ≥1.5% per year, a rate higher than most estimates for patients with SVR³². This is further complicated by the fact that the population of SVR patients will rapidly expand in the coming decades with the advent of DAA-based therapeutic strategies. The medical community has already had difficulty implementing successful screening for routinely warranted cases.. Only 12% of new HCV-related HCC cases are diagnosed through screening in the US⁵³ and less than 20% of patients with cirrhosis who develop HCC have undergone regular surveillance, highlighting the challenges that clinicians already face in the management of these patients⁵⁴. For all these reasons, it will be crucial to identify those who would benefit most from continued screening to allocate resources in a cost-effective manner. These will likely include patients with pre-treatment cirrhosis, older patients, diabetics and others identified by laboratory and genetic biomarkers.

Chemopreventive strategies will also be a significant consideration in the management of patients at risk for developing HCC. For example, maintenance low-dose IFN³⁹ after a course of typical IFN-based treatment has been evaluated in large-scale randomized trials, in an effort to prevent further prevent disease progression and HCC development. However, the relative inefficacy and poor tolerability of long-term IFN-based regimens have made it a non-viable chemopreventive candidate. Other chemoprevention strategies have been investigated to prevent development of HCC in patients after curative treatment with antiviral therapy, including glycyrrhizin and ursodeoxycholic acid. However, these have produced limited results in small studies⁵⁵.

Ultimately, the decision to implement screening or chemopreventive strategies in patients with chronic HCV and in those with SVR will require an integrated evaluation of each patient’s risk factors. Ideally, the decision can be aided by a prognostic index that evaluates patient characteristics including laboratory biomarkers and their molecular (186-gene) signature, Moreover, a cost-effectiveness analysis suggested that the 186-gene signature enables personalized HCC surveillance with reduced aggregate medical care cost and extended patient life expectancy⁵⁰. One such prognostic index was recently validated that accurately predicts the risk of clinical deterioration in patients with compensated HCV cirrhosis including liver-related mortality, hepatic decompensation, and HCC⁵⁸. It is hoped that the signature can soon yield plausible circulating markers that can be interrogated from patient sera.

The ascendant era of DAA-centered HCV therapeutics will likely stem the rising tide of chronic HCV infection in the near future and with it, lower the incidence of HCV-related complications, including HCC. Unfortunately, economic barriers to achieving this goal remain at this point given the prohibitively high upfront cost of these therapies (Table 2). However, recent economic analyses suggest that the current price of DAA-based treatment for genotype 1, and for treatment-experienced patients with genotypes 2/3 may be cost-effective when considering quality of life and the overall saved cost of future hospitalizations for HCV-related complications^56,57. Supporting this notion, a population-based study from Canada reported significant health-care related costs associated with HCC with an estimated mean 5-year net cost of care of $77,509. Given the estimated 33,000 new cases in the US in 2014, HCC-associated care represents a $2.5 billion strain on the medical system. Thus, chemopreventive approaches for HCC including treatment of HCV are likely cost effective provided that they are reasonably priced and safe since the economic cost of HCC is so high with regard to survival cost and the need for transplant or locoregional therapies. The challenge moving forward will be to enrich our ability to select patients who continue to be at the highest risk of HCC for continued follow-up, screening, and treatment to maximize prognosis. Prospective trials will be needed to address questions of the durability of SVR after DAA treatment, the incidence of HCC after achievement of DAA-mediated SVR, and the benefit of continued screening in SVR patients.