Current clinical evidence for nucleos(t)ide analogues in patients with HBV-related hepatocellular carcinoma

ABSTRACT

Introduction: Hepatocellular carcinoma (HCC) is a leading cause of death globally and is frequently seen following Hepatitis B virus (HBV) or Hepatitis C virus infection. Areas with high HBV infection rates, such as Asia and sub-Saharan Africa, are therefore also high-risk areas for HCC.

Areas covered: This review identifies and discusses the current evidence from robust clinical trials which have investigated the benefits of Nucleos(t)ide analogue (NA) antiviral therapy in HBV-related HCC patients, including HCC patients that underwent liver transplantation and HCC patients with or without curative treatment. In addition, we assess how this evidence has influenced current clinical practice, with a particular focus on those areas of high HBV infection rates.

Expert commentary: A number of studies have assessed whether NA antiviral treatment can improve the prognosis of HBV-related HCC patients. In this review we evaluate the current evidence, including that from trials in Asia, for antiviral NA treatments in HBV-related HCC patients. We also focus on those NAs with a high genetic barrier to resistance (i.e. ETV or TDF), on different therapeutic approaches, and on the future evidence that is required in this field.

KEYWORDS:

• Hepatocellular carcinoma
• Hepatitis B virus
• antiviral
• nucleos(t)ide analogue
• clinical practice
• Asia

1. Introduction

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related death, with approximately 750,000 new cases worldwide reported each year [1]. Currently, surgical resection, liver transplantation (LT), chemotherapy, and local tumor ablation are the four main treatment options for HCC patients. An estimated 78% of HCC cases result from chronic hepatitis B virus (HBV) and hepatitis C virus (HCV) infection [2], with areas of high HBV prevalence showing significant geographical correlation with HCC incidence and mortality worldwide [3]. Approximately 50% of the total cases and almost all of childhood HCC is caused by chronic HBV (CHB) infection [3]. In fact, HBV infection is the dominant risk factor in most areas of high HCC incidence in Asia and Sub-Saharan Africa, with the exception of Japan, where the major risk factor for HCC is chronic HCV infection [3]. High HBV DNA levels have been found to be a critical risk factor for HCC development [4,5] as well as recurrence of HCC post-surgical resection [6]. Nucleos(t)ide analogue (NA) antiviral therapy is recognized as an important treatment option to control HBV DNA load in patients and consequent HCC development (Figure 1) [7–10]. In HBV-infected patients already diagnosed with HCC, antiviral NA therapy can also improve the disease prognosis by reducing the risk of HCC recurrence and improving survival rate. In this review, we summarize the current evidence and practice for NA antiviral therapy in HBV-related HCC patients [7,11].

Figure 1. Current steps in the management of HBV-related HCC.

2. Clinical benefits/advantages of NA antiviral therapy in HBV-related HCC following diagnosis

2.1. Current antiviral therapy for CHB

Current global treatment strategies using standard antiviral therapy for HBV involve either a finite course of pegylated interferon (PegIFN) or an indefinite duration of treatment with NA’s [12]. PegIFN therapy has limited clinical use due to an adverse safety profile and the risk of liver disease exacerbation. It is not recommended for patients who have hepatic decompensation, immunosuppression, or medical or psychiatric contraindications [13].

Five oral NAs have previously been approved for the treatment of CHB [14] in Asia, including lamivudine (LAM), adefovir dipivoxil (ADV), entecavir (ETV), telbivudine (LdT), and tenofovir disoproxil fumarate (TDF). In addition, a new NA (tenofovir alafenamide; TAF) has recently been approved in the USA. The low genetic barrier to resistance of agents such as LAM, ADV, and LdT has limited their administration in CHB patients. NAs such as ETV and TDF are potent inhibitors of HBV with a high genetic barrier to resistance. Superior virological response has been observed with ETV at 48 weeks, with 67% patients in the ETV group having undetectable serum HBV DNA levels compared to 36% in the LAM group (P < 0.001) [15]. Treatment with TDF has also been shown to result in a stable response, with viral suppression of nearly 100% after 4 years of therapy [16,17]. Therefore, both agents have been recommended as first-line treatments for CHB patients [12,13,18].

All available antiviral therapies for HBV have been proven to be associated with a reduced risk of HCC development [8–10,12]. A nationwide cohort study from Taiwan revealed that NA-treated patients had a lower 7-year incidence of HCC (7.32; 95% CI, 6.77–7.87%) than untreated controls (22.7; 95% CI, 22.1–23.3%; P < 0.001) [19]. In patients with HCC, antiviral NA therapy was found to enhance patients’ tolerance for HCC treatment and further improve prognosis [19]. The most recent Asia Pacific Association for the Study of the Liver (APASL) guidelines recommend that NA treatment be given to patients with HBV-related HCC (before, during and after chemotherapy, resection or LT), if they have detectable serum HBV DNA [13].

2.1.1. Anti-HBV treatment with NAs in patients awaiting liver transplantation

HCC patients with a single lesion ≥2 cm and <5 cm, or no more than three lesions (the largest of which is <3 cm in size), and no radiographic evidence of extrahepatic disease, are optimal candidates for LT [20]. Antiviral therapy should commence in patients with HCC secondary to CHB who are awaiting liver transplantation [21]. NAs are effective in suppressing HBV viral load pre-transplant and preventing HBV recurrence post-transplant, thus contributing to improved long-term prognosis [13].

2.1.2. Anti-HBV treatment with NAs in patients listed for LT

Pre-transplant therapy with NAs aims to reduce HBV DNA at the time of LT and maintain undetectable serum HBV DNA post-transplant in order to prevent HBV recurrence [21]. The use of NAs prior to LT in HCC patients is similar to that in patients who develop HBV-related hepatic decompensation [22]. A summary of key clinical studies of NA use in this patient population are summarized in Table 1.

Table 1. Studies of pre-transplant anti-HBV treatment.

Study	Study type	Patient population	Treatment	N	Follow-up duration	Clinical improvement measure	Clinical improvement results	Virologic breakthrough	Survival
Schiff [23]	Prospective open-label	LAM-resistant HBV waitlisted for LT	ADV monotherapy	N: 226 (pre-LT)	39 weeks (median follow-up duration)	Mean change in HBV DNA from baseline at 48 weeks Undetectable HBV DNA levels at 48 weeks	–3.5 (log10 copies/ml) in ADV pts 59% in those with detectable HBV DNA at baseline	Not applicable	82.3% of pre-LT pts survived at end of study
Hou [24]	Double-blind RCT	Compensated HBeAg +ve or –ve CHB pts	LdT vs LAM	N: 332 HBeAg ±ve LdT: 147 LAM:143 HBeAg – ve LdT: 20 LAM: 22	52 weeks	Mean change in HBV DNA from baseline at 52 weeks Undetectable HBV DNA levels at 52 weeks	HBeAg ±ve –6.3 LdT vs – 5.5 LAM (P < 0.001) HBeAg -ve –5.5 LdT vs – 4.8 LAM (P = NS) HBeAg ±ve 67% LdT vs 38% LAM (P < 0.001) HBeAg -ve 85% LdT vs 77% LAM (P = NS)	HBeAg ±ve 7.5% LdT vs 17.5% LAM (P = 0.009) HBeAg -ve 0% LdT vs 4.5% LAM (P = NS)	Not applicable
Lai [25]	Double-blind RCT	HBeAg +ve or –ve CHB pts	LdT vs LAM	N: 1367 LdT: 680 LAM: 687	52 weeks	Mean change in HBV DNA from baseline at 52 weeks Undetectable HBV DNA levels at 52 weeks	HBeAg ±ve –6.5 LdT vs – 5.5 LAM (P < 0.001) HBeAg -ve –5.2 LdT vs – 4.4 LAM (P < 0.001) HBeAg ±ve 60% LdT vs 40% LAM (P < 0.001) HBeAg -ve 88% LdT vs 71% LAM (P < 0.001)	HBeAg ±ve 5.9% LdT vs 15.3% LAM (P < 0.001) HBeAg -ve 2.3% LdT vs 12.5% LAM (P < 0.001)	Not applicable
Chan [26]	Double-blind RCT	Tx naïvedecompensated CHB pts	LdT vs LAM	N: 232 LdT: 116 LAM: 116	52 weeks	Composite of serum HBV DNA <10 000 copies/mL, normal serum ALT level and improvement in or stabilization of CTP	56.2% LdT vs 54.0% LAM (P = 0.74) at wk 52 39.1% LdT vs 36.4% LAM (P = 0.67) at wk 104	Not available	94% LdT vs 88% LAM at wk 52 87% LdT vs 79% LAM at wk 104
Chang [15]	Double-blind RCT	NA-naïveHBeAg +ve compensated CHB pts	ETV vs LAM	N: 709 ETV: 354 LAM: 355	48 weeks	2pt improvement in Knodell necroinflammatory score with no worsening in Knodell fibrosis score Mean change in HBV DNA from baseline at 48 weeks Undetectable HBV DNA levels at 48 weeks	72% ETV vs 62% LAM (P = 0.009) –6.9 ETV vs – 5.4 LAM (P < 0.001) 67% ETV vs 36% LAM (P < 0.001)	Not applicable	100% ETV vs 99.4% LAM (P = 0.5)
Lai [27]	Double-blind RCT	NA-naïve HBeAg -ve compensated CHB pts	ETV vs LAM	N: 638 ETV: 325 LAM: 313	48 weeks	2pt improvement in Knodell necroinflammatory score with no worsening in Knodell fibrosis score Mean change in HBV DNA from baseline at 48 weeks Undetectable HBV DNA levels at 48 weeks and ALT< 1.25xULN	70% ETV vs 61% LAM (P < 0.001) –5.1 ETV vs – 0.5 LAM (P < 0.001) 55% ETV vs 4% LAM (P < 0.001)	2% ETV vs 8% LAM	99.4% ETV vs 100% LAM (P = 0.5)
Sherman [28]	Double-blind RCT	LAM-refractory HBeAg +ve pts	ETV vs LAM	N: 286 ETV: 141 LAM: 145	48 weeks	2pt improvement in Knodell necroinflammatory score with no worsening in Knodell fibrosis score Mean change in HBV DNA from baseline at 48 weeks Undetectable HBV DNA levels at 48 weeks	68% ETV vs 32% LAM (P = 0.01) –5.0 ETV vs – 4.5 LAM (P < 0.001) 90% ETV vs 72% LAM (P < 0.001)	Not applicable	99.3% ETV vs 98.6% LAM

Studies have shown that LAM is generally safe and well tolerated in patients with decompensated cirrhosis, demonstrating clinical improvement compared with untreated controls [29–31]. Furthermore, a subset of patients with less advanced liver failure, as indicated by lower modified Child-Turcotte-Pugh (mCTP) scores, might also derive clinical benefit from LAM therapy, thus delaying the need for LT [31]. Unfortunately, long-term use of LAM is associated with increasing rates of antiviral resistance [13]. Virological breakthrough caused by antiviral resistance can induce hepatitis flares, worsen liver failure, and increase the risk of HBV recurrence post-LT, thus reducing survival rates [32].

ADV has activity against both wild-type HBV and LAM-resistant HBV [22,33]. In a prospective, open-label, compassionate-use study, ADV was used in wait-listed patients with LAM-resistant HBV infection [33]. After 48 weeks of treatment, serum HBV DNA was undetectable in 59% of patient with survival rates of 86%. Eleven patients (6%) in the wait-listed group experienced an increase of at least 0.5mg/dL from baseline in serum creatinine during ADV treatment [33]. In a cohort of 128 pre-transplant patients with evidence of LAM-resistant HBV, 48 weeks of ADV therapy resulted in significant improvements in virological, biochemical, and clinical parameters [33]. Although ADV is not recommended as a first-line therapy in patients with decompensated cirrhosis due to its slow viral suppression and risk of renal toxicity, it still presents a potential life-saving treatment option for pre-transplant patients with LAM-resistant HBV.

LdT treatment has shown greater antiviral and clinical efficacy than LAM in CHB patients with compensated cirrhosis [24,25] and also decompensated cirrhosis [26]. A randomized, double-blind trial comparing the efficacy of LdT and LAM in patients with decompensated cirrhosis found that the proportion of patients achieving undetectable HBV DNA (<300 copies/mL) and ALT normalization was significantly higher in LdT-treated patients than in LAM-treated patients after 76 weeks of treatment (56.3% vs. 32.9%, P = 0.018) [26]. However, up to 21.6% of HBeAg-positive and 8.6% of HBeAg-negative patients have been reported to show LdT resistance after 2 years of therapy [34].

Long-term ETV treatment leads to substantial improvement and even regression of fibrosis or cirrhosis [35]. Data from three large, randomized, multicenter, phase III studies involving patients with CHB confirmed that ETV provided a higher observed virologic efficacy than LAM in a subgroup of patients with advanced liver fibrosis/cirrhosis [15,27,28]. A retrospective analysis from Korea showed that ETV was also associated with a significantly lower risk of death or transplantation than LAM in a matched pair of patients with decompensated cirrhosis (HR, 0.43; 95% CI: 0.30–0.63; P < 0.001) [36]. In another multicenter, prospective cohort study of 707 patients with HBV-related decompensated cirrhosis [37], antiviral-treated patients had significantly higher transplant-free survival than untreated patients (5-year survival rates of 59.7% vs. 46.0%, respectively). Considering that maintaining potent suppression of HBV led to improved long-term outcomes, the results of this research support the current recommendations of ETV as a preferred first-line treatment for patients with HBV-related decompensated cirrhosis [12,13]. However, for LAM-resistant patients, ETV monotherapy was not an effective rescue therapy [38].

Tenofovir is a licensed drug for the treatment of CHB with potent activity against HBV [12]; it is also a preferred option in patients with compensated or decompensated cirrhosis [13]. A multinational, open-label, follow-up cohort study found that at least 5 year’s TDF treatment in CHB patients resulted in 87% of patients showing histological improvement and 51% demonstrating regression of fibrosis (P < 0.0001) [39]. TDF with or without ETV also exhibited progressive improvement in the Model for End-Stage Liver Disease (MELD) scores in patients with decompensated cirrhosis in real-life clinical practice [40]. TDF also produced a similar treatment response and clinical outcome to ETV in patients with severe acute HBV exacerbation [41].

Based on these data, the European Association for the Study of the Liver (EASL) clinical practice guidelines recommend pre-transplant therapy using a potent NA with a high genetic barrier to resistance for all HBsAg-positive patients undergoing LT for HBV-related HCC or end-stage liver disease to achieve the lowest possible level of HBV DNA before LT [12].

2.1.3. Post-transplant NA therapy to prevent HBV recurrence

HBV recurrence in post-transplant patients who have not received antiviral prophylaxis can be as high as 80%, and is associated with a severe course or failure of LT [42]. Effective and safe antiviral therapy can contribute to additional survival benefits. Current prophylaxis strategies include Hepatitis B immune globulin (HBIg) alone, NA monotherapy, or combination therapy of HBIg, and NA with or without HBIg withdrawal after a certain period (Figure 2) [21]. HBIg monotherapy requires indefinite treatment leading to prohibitively expensive costs post-transplant and was found to be less effective in candidates with detectable serum HBV DNA at the time of LT [13]. With the introduction of NAs, HBIg monotherapy has been abandoned as an appropriate prophylactic strategy. Hence, here we focus primarily on the efficacy of NAs (monotherapy or combined with HBIg) in preventing post-transplant HBV recurrence (Table 2).

Table 2. Studies of post-transplant anti-HBV treatment.

Study	Study type	Treatment	N	Follow-up duration	Virologic breakthrough	HBV recurrence	Survival
Ahn [43]	Retrospective chart review	NA + HBIg vs NA mono	N: 28 NA +HBIg16 NA mono: 12	16.8 months (median follow-up duration post-LT in group 1)	Not applicable	HBV recurrence in 14% of pts(all noncompliant, 3 in HBIG+NA, 1 in NA mono)	No differences in survival
Fung [44]	Prospectiveobservational	HBsAg +ve pts treated with ETV as primary prophylaxis for prevention of HBV recurrence	N: 80 (all HBsAg positive)	26 months (median follow-up duration post-transplant)	10 pts had reappearance of HBsAg; 0 pts had evidence of virologic breakthrough	94% had undetectable levels of HBV DNA (<35 copies/mL) at the last follow-up	No HBV-related mortality in the total cohort
Cholongitas [45]	Prospective observational	Withdrawal of HBIg in stable pts who had been under HBIg+NA (at least 12 months) therapy after LT	N: 47 LAM+ADV: 23 LAM+TDF: 5 TDF: 10 ETV: 9	24 months (median follow-up duration post-HBIg withdrawal)	All pts had undetectable HBV DNA levels	No reported HBV recurrence	No HBV-related mortality was reported
Perrillo [46]	Prospective Open-label trial	ETV combined with various HBIg regimens after CHB-related LT	N: 65 Evaluated: 61 (received at least 4 weeks of ETV treatment) Completed treatment: 55	Follow-up duration was 72 weeks	100% evaluated pts had undetectable HBV DNA 3% of pts had nonviremic HBsAg recurrence	0% reported virologic recurrence	1/61 pts died due to HCC recurrence with metastasis
Tanaka [47]	Retrospective chart review	TDF (± LAM) and one year of HBIg	N: 24 TDF + HBIg: 15 TDF+LAM+HBIg: 9	29.1 months (median post-LT follow-up duration)	Undetectable HBsAg and HBV DNA levels during follow-up period	No recurrent HBV infection reported in median follow-up period	Three deaths (12.5%) during follow-up period Over all patient survival post-LT: Year 1: 100%; Year 5: 84.1%
Ueda [48]	Retrospective chart review	Tx group: ETV + HBIg Control: LAM + HBIg	N: 89 Tx group: 26 Control: 63	Median follow-up: Tx group: 25.1 months Control: 70.6 months	Not applicable	No HBV recurrence with ETV during median follow-up HBV recurrence rates in control pts: Year 1: 2%; Year 3: 4%; Year 5: 6%	Survival rates of ETV pts were 73% at both 1 and 3 years Survival rates of control: Year 1: 81%; Year 3: 78%
Gao [49]	Retrospective chart review	Low-dose HBIg plus LAM or ETV	N: 253 LAM + HBIg: 169 ETV + HBIg: 84	Mean txduration: LAM group: 61.8 ± 18.3 months ETV group: 57.1 ± 15.9 months	Not available	No HBV recurrence reported in the ETV group HBV recurrence rate of 9.46% in the LAM group	There was no significant difference in the survival rate between the two groups
Kim [50]	Retrospective chart review	ETV + HBIg post-LT	N: 154	28.0 months (median follow-up duration)	Not available	3.2% Post-LT HBV recurrence post-LT pts 0% ETV resistant mutants Overall rates of HBV recurrence: Year 1: 0.6%; Year 2: 1.6%; Year 4: 6.2%	Not available
Lee [51]	Retrospective chart review	HBIg ± ETV	N: 667	51.0 months (median follow-up duration)	Median time to HBV recurrence was 22 months	11.4% HBV recurrence at 51 months	54.2% survival at 10 years in pts with HBV recurrence

Figure 2. Post-transplant treatment options for the prevention of HBV recurrence.

X: negative impact; ?: unclear impact; ✓: positive impact; HBV: hepatits B virus; HBIg: hepatitis B immune globulin; NA: nucleos(t)ide analogue; LAM: lamivudine; ETV: entecavir; TDF: tenofovir disoproxilfumarate; IM: intramuscular.

2.1.4. NA monotherapy

Malkan et al. [52] investigated LAM as a single-agent treatment in LT recipients for the prevention of post-transplant HBV recurrence. Patient survival was 81% over a median of 24 months with 19% having evidence of breakthrough infection with the tyrosine-methionine-aspartate-aspartate (YMDD) mutant of HBV. Other studies have also identified YMDD variants post-LT, leading to death from graft failure [53,54]. A cohort study of LAM monotherapy before and after LT demonstrated that the 1- and 2-year cumulative post-transplant survival for patients with YMDD mutants was 83% and 28%, respectively [55].

The recent approval of highly potent NAs with minimal risk of genetic resistance, such as ETV and TDF, has allowed physicians to re-visit NA (ETV and TDF) monotherapy as post-transplant prophylaxis. However, available data for ETV and/or TDF monotherapy as prophylaxis were limited and based on small retrospective studies. Ahn reported four patients receiving NA monotherapy for preventing post-LT HBV recurrence, among whom one patient received TDF monotherapy [43]. The patient had no evidence of post-LT HBV recurrence during the follow-up [43]. Fung et al. investigated the efficacy of ETV monotherapy as a treatment option in patients with CHB who received a liver transplant [44]. It is still unclear which subgroup of patients is suitable for this regimen; the efficacy of a ’rescue therapy’ in patients who failed ETV/TDF monotherapy as post-transplant prophylaxis also remains to be elucidated.

2.1.5. Combination therapy with NA and HBIg

Combining NAs with HBIg as prophylaxis for CHB recurrence has produced general recurrence rates of <10%, which is substantially lower than HBIg or NAs as monotherapy [56]. Therefore, the combination regimen is currently the best-practice approach recommended by the European Association for the Study of the Liver (EASL), Asian Pacific Association for the Study of the Liver (APASL) and American Association for the Study of Liver Disease (AASLD) guidelines [12,13,18]. The efficacy of NA and HBIg in combination has mainly been investigated in patients receiving LAM and HBIg. Data from a study of 165 LT recipients (94 of 165 diagnosed as HCC) suggest that LAM and individualized low-dose intramuscular HBIg provided effective prophylaxis against HBV recurrence [57]. A meta-analysis consisting of two prospective and four retrospective studies showed that the OR for risk reduction in HBV recurrence with HBIg and LAM versus HBIg alone was 0.08 (95% CI: 0.03–0.21) [58].

Indefinite therapy is usually required due to the persistence of HBV in the liver or extrahepatic sites, which may trigger HBV recurrence post-transplant [13]. Patients receiving long-term treatment with LAM are at high risk of developing LAM-resistant mutants. Therefore, a combination of HBIg and NA with a high genetic barrier to resistance (i.e. ETV or TDF) is increasingly in demand. A systematic review found that patients treated with HBIg and ETV/TDF experienced significantly lower rates of HBV recurrence than patients receiving HBIg and LAM (1.3% vs. 7.1%, P = 0.0005) [59]. Despite these promising results, the optimal protocol for post-LT HBV prophylaxis has not been established, as HBIg has several limitations including high cost, need for parenteral administration, and poor safety profile. HBIg discontinuation after prior combination therapy or followed by NAs is the most common strategy currently utilized. A number of studies have assessed and demonstrated the efficacy of HBIg discontinuation in patients receiving NAs with a high genetic barrier to resistance [45,59,60].

2.2. Patients eligible for surgical resection or local ablation therapy

Hepatic resection is the most efficient treatment available for HCC patients with tumors >5 cm, nodules >3 cm, or with tumor macroscopic vascular invasion, who are less likely to benefit from LT or local ablative therapy [61]. Hepatic resection remains a radical therapeutic option that may offer a chance for disease-free survival in these patients [62,63]. For cases not suitable for liver resection, local tumor ablation is an alternative option for curative or palliative treatment of HCC, with a number of tumor ablative procedures being available such as radiofrequency ablation (RFA), transarterial chemoembolization (TACE), microwave (MW), and percutaneous ethanol injection (PCEI) [64]. However, the prognosis for HCC patients remains discouraging because of the frequent incidence of intrahepatic metastasis and the high recurrence rate [65]. In patients undergoing hepatic resection for HBV-related HCC, accumulating evidence suggests that viral load at the time of resection is an independent risk factor for HCC recurrence [6,63]. Liver resection itself could reactivate HBV replication, especially in patients who have not received prophylactic antiviral therapy [65,66].

2.3. Recurrence of HCC

Intrahepatic recurrence of HCC after hepatic resection has two distinct etiologies: recurrence via metastasis and new primary tumors [67]. Administration of antiviral therapy after HCC surgery cannot prevent intrahepatic early recurrence, or recurrence outside of the liver, both of which account for at least 70 and 10% of HCC recurrences after resection, respectively [68].

2.3.1. NAs against HCC recurrence after surgical resection or local ablation therapy

Higher serum HBV DNA level (> 10⁶ copies/ml) has been shown to be associated with late tumor recurrence [69]. The impact of postoperative antiviral treatment on survival in patients with HBV or HCV infection-related primary HCC after curative therapy was therefore evaluated in a number of studies [6,69–77] (Table 3). A meta-analysis of these studies found that there was a trend of lower HCC recurrence in the treatment groups (33.3%) compared with controls (52.9%) (OR: 0.43, 95% CI: 0.17–1.07, P = 0.07) [78]. Chan et al. also found that commencing antiviral therapy after hepatectomy improves the prognosis of HCC in preoperatively antiviral-naïve patients with HBV infection (5-year survival 71.2% vs. 43.5%, P = 0.005) [72]. Similarly, a cohort study of patients receiving curative liver resection for HCC in Taiwan demonstrated a significant risk reduction of tumor recurrence in patients receiving NAs versus untreated patients (45.6% vs. 54.6%, P < 0.001) [73]. This benefit was also seen following RFA (2-year recurrence rate: 41.8% vs. 54.3%; P < 0.05) [74].

Table 3. Recurrence of HCC. Risk factors and survival rates.

Study	Patient population	Treatment	N	Identified risk factor(s) for HCC recurrence	Follow-up duration	HCC recurrence	Survival
Hung [6]	Liver resection for HBV-related HCC	Not applicable	N: 72 Tumor recurrence: 30 No tumor recurrence: 42	HBV viral load >2,000 IU/mL (4 log10 copies/mL) (P = 0.001) α-fetoprotein (AFP) level >1,000 ng/mL (P = 0.02) Tumor size >5 cm (P = 0.02) Age >60 yrs (P = 0.01)	18.9 months (median)	42% recurrence Median time to recurrence: 7.3 months	7 deaths due to HCC recurrence (23.2%) 100% of patients in the non-recurrence (n = 42) group survived (P = 0.001)
Kuzuya [70]	Resection or radiofrequency ablation for HBV-related HCC	Tx group: LAM therapy Control: No LAM therapy	N: 49 LAM group: 16 Control: 33	Not available	Not available	Cumulative HCC recurrence rates did not differ significantly between the two groups (P = 0.622)	Cumulative survival rates of LAM patients were higher than those in the control group (P = 0.063)
Someya [71]	Cirrhotic patients receiving ablation for HBV-related HCC		N: 80 Abnormal AST: 38 (6 received IFN) Normal AST: 42 (5 received IFN)	Not available	Not available	Tumor recurrence rates in treated and untreated patients with abnormal AST, respectively: Year 1: 16.7% and 37.9%; Year 2: 16.7% and 60.1%; Year 3: 16.7% and 83.4 (P = 0.0139)	Not available
Chan [72]	Hepatectomy for HBV-related HCC	Tx group: NA therapy Control: No NA therapy	N: 136 Tx group: 42 (LAM: 38; ETV: 4) Control: 94	Not available	1 month after hepatectomy to confirm complete tumor clearance and then every 3 months for surveillance	Tumor recurrence developed in 48% of tx patients and 65% of control patients (P = 0.28)	Overall survival rates in in the tx group: Year 1: 88.1%; Year 3: 79.1%; Year 5: 71.2% Overall survival rates in in the tx group: Year 1: 66.5%; Year 3: 51.4%; Year 5: 51.4%
Wu [73]	Curative liver resection for HBV-related HCC	Tx group: NA therapy Control: No NA therapy	N: 4569 Tx group: 518 (LAM, ETV, LdT) Control: 4051	Liver cirrhosis was an independent risk factor for HCC recurrence (P < 0.001)	Median follow-up durations: Control group: 1.57 years Tx group: 2.18 years	HCC recurrence occurred in 43.6% of untreated patients and 20.5% of treated patients (P < 0.001)	Six year overall mortalities for tx patients were 29.0% and for untreated 42.4% (P < 0.001)
Lee [74]	NA-naïve and NA-experienced receiving radiofrequency ablation for HBV-related HCC.	Tx group: NA therapy Control: No NA therapy	N: 399 NA-treated: 133 Control: 266	Not available	Not available	2-year HCC recurrence rate of NA-treated vs control group: 41.8% vs 54.3%; P < 0.05)	Not available
Hann [75]	Local tumor ablation forNA-naïve HBV-related HCC	Tx group: NA therapy Control: No NA therapy	N: 25 Tx group: 16 Control: 9	Not available	Up to 12 years	HCC recurrence occurred in 100% of untreated patients and 18.8% (including residual tumor activity) of treated patients	Median survival: 80 months in Tx group vs 16 months in Control group (P = 0.001)
Piao [76]	Local tumor ablation or surgery for HBV-related HCC	Tx group: LAM therapy Control: No NA therapy	N: 70 Tx group: 30 Control: 40	Not available	Median follow-up: Tx group: 24 months Control group: 31 months	HCC recurrence: At 12 months: Tx group: 25% Control group: 42% At 24 months: Tx group: 54% Control group: 59%	Mortality rate: At 12 months: Tx group: 0% Control group: 8% At 24 months: Tx group: 9% Control group: 13%
Wu [69]	Liver resection for HBV-related HCC	Not applicable	N: 193 Tumor recurrence: 134 No tumor recurrence: 59	Early recurrence (<2 years) AFP >20ng/mL Cut margin ≤1cm Macroscopic venous invasion Multinodularity Late recurrence (≥2 years) HBV DNA >10^6 ICG-15 > 10% Ishak activity >6 Multinodularity	58.2 months (median)	49% early recurrence (<2 years) 69% total recurrence Median time to recurrence: 10 months	Risk factors associated with poor survival: Age > 60 years Microscopic venous invasion Performance status ≥1 AFP >20ng/mL Creatinine >1.5mg/dL Multinodularity
Kim [77]	Curative liver resection for HBV-related HCC	Not applicable	N: 157	Persistent detectable serum HBV DNA (P = 0.041)	35 months (median)	5-year cumulative recurrence: Persistent HBV DNA group: 72.9% No persistent HBV DNA group: 54.7% (P = 0.043)	Not available

2.3.2. Patients receiving chemotherapy for HBV-related HCC

Reactivation of HBV is common in HBsAg-positive individuals who are undergoing chemotherapy, especially in patients with hematologic diseases [79,80]. For patients with HBV-related HCC, HBV reactivation and subsequent liver failure or early termination of chemotherapy has a negative effect on HCC survival [80]. It is believed that chemotherapy may enhance the replication of HBV in hepatocytes and suppress the cellular immune response at the same time [81,82]. All candidates for chemotherapy should be screened for HBsAg and anti-HBc prior to initiation of treatment [13]. The risk and severity of HBV reactivation appears to be proportionate to the degree of immunosuppression and the intensity of cancer chemotherapy.

Meta-analyses have shown that LAM can reduce the risk of HBV reactivation by 4–7 fold during chemotherapy in patients with a positive HBsAg [83]. In patients with solid tumors or lymphoma, prophylactic LAM could reduce the severity of HBV reactivation [84,85]. However, the optimal treatment duration of preventive LAM is still unclear, with studies ranging from 3 to 6 months [86,87]. A limitation of LAM is the development of resistance, ranging from 3 to 8% in published studies [80]. Both ADV and ETV have lower resistance rates than LAM and studies have shown that both drugs are effective in preventing HBV reactivation in patients receiving chemotherapy [88–90], with one retrospective study demonstrating that ETV had significantly lower rates of HBV reactivation (0 vs. 12.4%, P = 0.024) [90]. ETV also had significantly fewer hepatitis flares due to HBV reactivation (0.0% vs. 11.6%, P = 0.039) [91].

2.4. Practical management strategies

Several organizations have published consensus guidelines on chronic HBV infection, including the EASL, AASLD, APASL and National Institute for Health and Clinical Excellence (NICE). Peg-IFN-α-2a, IFN-α-2b, LAM, ADV, ETV, LdT, TDF are the most commonly used approved antiviral drugs for the management of chronic HBV infection. For HBV patients, regardless of HBeAg-positive or negative, all guidelines recommend ETV, TDF and PEG-IFN-α-2a as the first-line treatment [13,18,20].

However, antiviral therapy may not prove to be adequate in patients with very advanced liver disease who should be considered for liver transplantation simultaneously [12]. In HBV-related HCC patients receiving curative surgical resection, one retrospective cohort study reported that adjuvant therapy with NAs could reduce the HCC recurrence rate and mortality by 9% and 13.4%, respectively [73]. However, the study failed to compare the baseline tumor staging, surgical performances, and liver function or the development of antiviral resistance. Large randomized trials establishing the benefit of NA treatment in HBV-related HCC after curative resection are therefore necessary. As high HBV DNA levels are closely related to the recurrence of HBV-related HCC after surgical resection, studies have also compared the differential ability of NAs to suppress HBV viral load. ETV was reported to suppress HBV replication more rapidly and potently than LAM and ADV in compensated HBV patients [27,92,93]. Fung et al. have however summarized various case reports and small studies of LAM that suggest an increased risk of hepatic failure, which could lead to death, following biochemical flares after virologic breakthrough [94]; however, a meta-analysis of trials comparing ETV and LAM in CHB decompensated cirrhosis found no overall difference in mortality between the treatments [92]. It is therefore reasonable to suggest that both ETV and TDF would have a superior treatment efficacy in HBV patients than other antiviral drugs, although further robust evidence for the long-term effect of TDF in the HBV-related HCC patient population is required. Data on the long-term safety of TDF in the HIV patient population have demonstrated that TDF therapy could lead to renal insufficiency and hypophosphatemic osteomalacia, although this has not yet been demonstrated in HCC patients [95]. The EASL guidelines state that ETV is the most appropriate first-line agent in individuals with renal failure, whereas TDF is considered an appropriate alternative in LAM-resistant patients [12]. However, the impact of either agent on survival in patients with HBV-related HCC remains to be fully elucidated.

3. Conclusion

The goal of antiviral therapy for HBV-related HCC is to suppress viral replication, reduce the recurrence rate of HCC, and increase patient overall survival. To date, the use of antiviral therapy in HBV-related HCC has been recommended by several global organizations. Both pre- and post-operative antiviral therapy could reduce the recurrence rate of HCC in patients, but little data are currently available comparing the efficacy of different NAs. Studies have recommended ETV and TDF as first-line antiviral agents for treatment-naïve patients with decompensated cirrhosis both pre- and post-transplantation, and using NAs with a low genetic barrier to drug resistance should be avoided. The use of antiviral agents should be combined with HBIg to further reduce HBV recurrence rates, while monotherapy of potent NAs with a high genetic barrier can be an alternative to combination therapy for post-transplant patients. For patients receiving chemotherapy treatment, LAM can reduce the risk and the severity of HBV reactivation but poses a drug resistance threat. Potent NAs with lower resistance rates than LAM, such as TVF and ETV, showed effectiveness in preventing HBV reactivation in patients receiving chemotherapy.

In future, large, multicenter randomized trials comparing the efficacy of different antiviral therapies in HBV-related HCC patients should be conducted.

4. Expert commentary

HCC is a leading cause of cancer-related death, with an estimated 750,000 new cases of HCC worldwide every year. It is further estimated that up to 80% of these cases are as a result of HBV or HCV infection, with evidence showing that HCC mortality and incidence is correlated with prevalence of HBV. Surgical resection, lT, chemotherapy, and local tumor ablation are currently the main treatment options available, however in cases related to HBV infection there should also be an antiviral component of treatment to prevent recurrence. Considering the continued high rates of Hepatitis B in sub-Saharan Africa and East Asia, there remains a high need for clear and effective antiviral treatment pathways in these regions for HBV patients that subsequently develop HCC. This need exists in three distinct but related patient populations, namely:

1. HBV-related HCC patients who have undergone lT. In these patients, a combination of HBIg and NAs with a high genetic barrier to resistance (i.e. ETV or TDF) prove to effectively prevent HBV recurrence posttransplant. It is still unclear whether HBIg discontinuation after prior combination therapy or NA monotherapy could provide equivalent long-term benefits, and this should be investigated in the future from both an efficacy and cost-effectiveness perspective.

2. HBV-related HCC patients with curative treatment. Accumulating evidence shows that NA antiviral treatment could reduce HCC recurrence and improve overall survival rate. This supports the use of NA antiviral therapy in the tertiary prevention of HCC.

3. HBV-related HCC patients without curative treatment. NAs have been shown to be effective in preventing HBV reactivation in patients receiving chemotherapy. However, future considerations should evaluate whether NAs should be administered in all or only a subset of late-stage HCC patients. This question should be considered from both a disease status perspective and an expected lifespan status.

An important next step would therefore be to conduct a series of clinical trials. Of particular interest would be the initiation of large multicenter trials with longer treatment duration from across Asia. This would enable specific sub-analyses and meta-analyses of the data to confirm the safety and efficacy of existing treatments in these high-risk populations, and would be of benefit in the generation of robust local guidelines. By understanding this issue more thoroughly, it will not only be possible to assess the best treatment option for each patient but will also provide data on the total healthcare cost associated with the long-term outcomes of the patients. This in turn will allow for more thorough total cost assessments, and potentially lead to an improved economic strategy for treatment of HBV-related HCC.

Although treatments are available for HBV-related HCC, future focus should also be on the prevention of HBV, thus avoiding HBV-related HCC. Safe and effective vaccines for HBV are available which have enabled infection rates to drop to less than 1% in Western Europe and North America. With improved prevention and clearer guidance for treatment of HBV-related HCC, we hope that the healthcare burden of this disease can be reduced across the region.

5. Five-year view

The current standard of antiviral therapy involves a finite course of pegIFN and an indefinite duration of NAs. Although ‘functional cure’, defined as HBsAg loss or seroconversion is rarely achieved with NAs, long-term treatment with NAs does reduce the incidence of cirrhosis, liver failure, and HCC. Therefore, NAs with potent antiviral effect and high genetic barrier are still likely to be a mainstay of long-term treatment in patients with HBV-related HCC.

In addition, a number of potential therapies are under preclinical and clinical investigation, and these have the potential to further improve treatment if approved. These include compounds that induce cccDNA degradation, compounds that inhibit HBV entry, compounds that inhibit the expression of viral proteins, and compounds that use immunotherapeutic approaches [13,96].

In addition to upcoming developments in treatment options, it is likely that there will be improvements in the identification of HBV patients at high-risk of HCC. A number of risk factors for HCC have already been identified [97], and researchers have developed risk prediction models that can be used by doctors to stratify patients for their risk of HCC [98]. This will allow preventative antiviral treatment for those patients at higher risk of HCC, thus reducing the overall burden of HBV-related HCC in a cost-effective manner. Future risk models should also take into account somatic and inherited biomarkers [98].

Key issues

• A high prevalence of HBV infection is associated with an increased incidence of HCC

• Antiviral Nucleos(t)ide analogues (NA) treatment improves HBV-related HCC prognosis

• Drug resistance can be an issue with some currently available NAs. NAs with a high genetic barrier to resistance are recommended by international guidelines.

• More data are required that directly compare the safety and efficacy of current NAs

Declaration of interest

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Additional information

Funding

Editorial and writing support was provided by Roopsha Brahma of MediTech Media, funded by Bristol-Myers Squibb, Shanghai.

Notes on contributors

Yiqi Yu

Conceived and designed the manuscript: Wenhong Zhang Search the literature: Yiqi Yu and Jingwen Ai Wrote the manuscript: Yiqi Yu and Jingwen Ai Design the table and figure: Yiqi Yu and Jingwen Ai