Introduction

Hepatitis B virus (HBV) and hepatitis C virus (HCV) are the leading causes of chronic liver disease that can progress to cirrhosis and increase the risk of hepatocellular carcinoma. Both infections remain a global public health issue, with an estimated 58 million individuals living with HCV and 296 million with HBV.1,2 Nearly 2 billion people globally have been exposed to an HBV infection, as indicated by the presence of antibodies to the virus core antigen.2 Although HBV and HCV are phylogenetically distant pathogens, they share a similar transmission route, which increases the odds of coinfection. The rate at which this phenomenon occurs can vary in different geographic regions from 1% to 15%.3,4

Although initial in vitro observations implied that HBV and HCV could replicate in the same hepatocytes without interfering with each other, clinical data do not support this claim. Firstly, higher liver disease severity and increased rate of cirrhosis have been observed in coinfected individuals than in HBV or HCV monoinfections.5-7 Secondly, studies in coinfected patients indicate that HCV is more likely to suppress HBV replication.4,8-11 All in all, further research is needed to understand the effect of coinfection on various disease markers, particularly in patients with 2 different HBV serologic patterns: 1) negative for hepatitis B virus surface antigen (HBsAg) but positive for antibodies against hepatitis B core protein (anti-HBc), and 2) positive for HBsAg.

While a chronic HBV infection is not entirely curable at present, despite available treatment suppressing viral replication and lowering the risk of cirrhosis and hepatocellular carcinoma, HCV infections can be successfully and safely treated with direct-acting antiviral agents (DAAs).12-14 Therefore, it is pivotal to establish whether HCV treatment with DAA can lead to HBV reactivation in coinfected individuals. Initial studies with short-term follow-up suggest that the risk of such a phenomenon may be low, even though some authors indicate that HBV reactivation during DAA can be underreported.3,15-18 However, long-term observations are also needed to understand whether some distant HBV reactivation events may occur in HCV-treated patients. Moreover, studies with long-term follow-up of patients who experienced HBV reactivation during DAA therapy are needed to fill the knowledge gap. It is also crucial to understand the effectiveness of DAA treatment in patients coinfected with HBV. Studies assessing HBsAg and anti-HBc in all patients treated for HCV were carried out only in small populations. In large studies, such as the only Polish research project on this subject by Jaroszewicz et al,3 which is also unique in terms of its population size, only some patients had the anti-HBc presence assessed.3 This clearly highlights the need for additional research explicitly designed to characterize HBV-infected patients receiving anti-HCV treatment, and to assess the risk of HBV reactivation during anti-HCV treatment and effectiveness of DAA therapy. Such observations are essential for future guidelines focusing specifically on managing HBV/HCV coinfection. The only recommendations we currently have are remarks in various guidelines on managing HCV or HBV infection or expert opinions.9,19-23

This study aimed to assess the prevalence of HBV serologic markers in patients with chronic HCV infection, to compare patients and liver disease characteristics with reference to the HBV status, to evaluate the risk of HBV reactivation during DAA therapy, to monitor patients with reactivation on a long-term basis, and to assess the safety and effectiveness of the antiviral treatment.

Patients and methods

Study population

The analyzed population consisted of all consecutive patients with chronic HCV infection receiving antiviral treatment with interferon (IFN)-free regimens from July 1, 2015 to December 31, 2022 at a single hepatology center. The patients were treated under a reimbursable therapeutic program of the National Health Fund. The choice of the DAA regimen was made by an attending physician, guided by the product characteristics, requirements of the drug program, and recommendations of the Polish Group of Experts for HCV.21,24-27 The patients provided their informed consent for treatment and processing of their personal data, as required by the drug program and national regulations. Ethical review and approval were waived due to a retrospective design of the study.

Data were obtained retrospectively from medical records. Baseline data comprised demographic and clinical parameters, including comorbidities, medications used, information on the severity of liver disease, history of previous antiviral therapy, current DAA regimen, and laboratory test results, such as aminotransferase activity, international normalized ratio, complete blood count, creatinine, bilirubin, and albumin levels. All patients were assessed for the presence of HIV antibodies. Data were also collected on a diagnosis of alcoholic liver disease and autoimmune hepatitis as other factors contributing to damage to this organ.28

Severity of liver disease

Liver disease advancement was assessed by noninvasive measurement of liver stiffness (elastography) using an Aixplorer device (Supersonic, Aix-en-Provence, France). Values obtained in kilopascals (kPa) were presented in the form of corresponding degrees of fibrosis (F) on the METAVIR scale according to the guidelines of the European Association for the Study of the Liver,21 and a cutoff value of 13 kPa was used to identify individuals with F4 fibrosis, who were considered cirrhotic. These patients were assessed for the presence of esophageal varices, past or present liver failure in the form of ascites and encephalopathy, and were scored on the Child–Pugh (CP) scale; grades B and C corresponded to decompensated cirrhosis. Data on liver transplantation and a diagnosis of hepatocellular carcinoma were also reported.

Hepatitis C virus infection and direct-acting antiviral treatment

Data on HCV genotype, baseline viral load measured by polymerase chain reaction (PCR) with a lower limit of detection 10 IU/ml, extrahepatic manifestations, history of previous antiviral therapy, and current DAA regimen were collected, including genotype-specific and pangenotypic options. A measure of the DAA therapy effectiveness was sustained virologic response, defined as undetectable HCV RNA at least 12 weeks after the end of the therapy. Patients with detectable viral load at this time point were considered virologic failures, while those without HCV RNA assessment 12 weeks after the end of the therapy due to loss to follow-up were considered nonvirologic failures. Data on the course of the therapy, adverse events, including severe ones, and deaths during treatment and up to 12 weeks after its completion were also gathered.

Hepatitis B virus coinfection

Each patient was tested for coinfection with HBV. The tests for HBV infection included determining HBsAg and anti-HBc antibodies. All patients with positive HBsAg were assessed for the presence of hepatitis B envelope antigen (HBeAg) and anti-HBe antibodies. HBsAg- and anti-HBc-positive patients were also tested for HBV DNA by real-time PCR with a lower detection limit of 10 IU/ml. Information on therapy for HBV infection was also collected, including the type of drug and the time of its use before DAA therapy. HBsAg-positive and anti-HBc-positive individuals were monitored for HBV reactivation during DAA treatment and up to 24 weeks after its completion. The patients who developed reactivation were monitored for an extended period of time until the disorder resolved or during anti-HBV therapy, if included. HBV reactivation was defined as a sudden increase of at least 100-fold in the HBV DNA concentration in individuals with previously detectable HBV DNA or HBsAg or HBV DNA detection in individuals with previously undetectable anti-HBc antibodies.29

Statistical analysis

Continuous data were reported as medians and interquartile ranges (IQRs). Categorical data were summarized by frequencies and percentages. For nominal variables, the significance of difference was calculated by the χ2 test or the Fisher exact test (as appropriate in the case of a small group size). For continuous, non-normally distributed variables, the group comparisons were performed using the nonparametric Mann–Whitney test (Gaussian distribution of continuous variables was checked with the Shapiro–Wilk test). Sustained virologic response (SVR) was calculated as intent-to-treat analysis, and after exclusion of lost to follow-up patients as per protocol (PP). The Bonferroni corrections (a raw P value multiplied by the number of comparisons) were applied to any multiple comparisons to account for α inflation and limit the probability of type 1 error. A P value below 0.05 was considered significant. Statistical analyses were performed using Statistica v. 13 packkage (StatSoft, Tulsa, Oklahoma, United States) and GraphPad Prism 5.1 package (GraphPad Software, Inc., La Jolla, California, United States).

Results

Characteristics of the study population

The study population consisted of 1118 patients with chronic HCV infection. They were divided into 3 subgroups according to their HBV status: HBsAg- and anti-HBc-negative (n = 950; 85%; group A), HBsAg-negative and anti-HBc-positive (n = 160; 14.3%; group B), and 8 individuals positive for HBsAg (0.7%; group C).

The study group comprised more women than men (597/1118; 53%). This was also true for patients in the groups A and B, while insignificantly more men were among the HBsAg-positive patients (Table 1). Median (IQR) age of the patients was the highest in the group B (58 [44–67] years) and the lowest in the group C (36 [30.5–43] years), with significant differences between the groups. As compared with the individuals without HBV coinfection, the patients in the groups B and C had more comorbidities (77.7% vs 81.3% and 87.5%, respectively), although the differences were not significant. Alcoholic liver disease was diagnosed in 42 patients, 35 in the group A (3.7%) and 7 in the group B (4.4%). No patients were diagnosed with autoimmune hepatitis. Biochemical markers were comparable between the groups, except for lower platelet count and albumin level noted in the group B (Table 1).

Table 1. Baseline characteristics of patients with reference to their hepatitis B virus status

Parameter

Group A: HBsAg(–) / anti-HBc(–) (n = 950)

Group B: HBsAg(–) / anti-HBc(+) (n = 160)

Group C: HBsAg(+) (n = 8)

P value

A vs B

A vs C

B vs C

Sex

Women

509 (53.6)

86 (53.8)

2 (25)

>0.99

0.48

0.45

Men

441 (46.4)

74 (46.3)

6 (75)

Age, y

Overall

49 (36–63)

58 (44–67)

36 (30.5–43)

<⁠0.001

0.06

0.006

Women

54 (36–65)

59.5 (47–69)

37 (35–39)

0.009

0.78

0.3

Men

45 (36–61)

56.5 (42–64)

34 (30–47)

0.02

0.18

0.06

BMI, kg/m2

25.4 (22.6–28.4)

25.3 (22.4–28.9)

22.9 (19.8–27.6)

>0.99

0.45

0.54

Comorbidities

Any comorbidity

738 (77.7)

130 (81.3)

7 (87.5)

0.93

>0.99

>0.99

Hypertension

334 (35.2)

66 (41.3)

1 (12.5)

0.42

0.81

0.45

Diabetes

119 (12.5)

27 (16.9)

0

0.39

>0.99

>0.99

Renal disease

72 (7.6)

21 (13.1)

2 (25)

0.06

0.36

0.9

Autoimmune diseases

69 (7.3)

10 (6.3)

0

>0.99

>0.99

>0.99

Non-HCC tumors

45 (4.7)

15 (9.4)

1 (12.5)

0.06

0.99

>0.99

Alcoholic liver disease

35 (3.7)

7 (4.4)

0

>0.99

>0.99

>0.99

Concomitant medications

611 (64.3)

115 (71.9)

7 (87.5)

>0.99

0.51

>0.99

Extrahepatic manifestations of HCV

Any

497 (52.3)

86 (53.8)

4 (50)

>0.99

>0.99

>0.99

Cryoglobulinemia

410 (43.2)

85 (53.1)

4 (50)

>0.99

>0.99

>0.99

Thyroid abnormalities with the presence of antithyroid antibodies

87 (9.2)

12 (7.5)

0

>0.99

>0.99

>0.99

Thrombocytopenia in noncirrhotics

40 (4.2)

5 (3.1)

0

>0.99

>0.99

>0.99

Laboratory parameters

ALT, IU/l

56 (36–90)

55 (36–98)

100.5 (49–141.5)

>0.99

0.42

0.57

Bilirubin, mg/dl

0.8 (0.6–1)

0.8 (0.6–1.1)

0.8 (0.7–0.9)

0.99

>0.99

>0.99

Albumin, g/dl

4.1 (3.8–4.3)

4.0 (3.6–4.2)

4.2 (4–4.6)

0.003

0.6

0.14

INR

1 (1–1.1)

1 (1–1.1)

1.1 (1–1.2)

>0.99

0.39

0.54

Creatinine, mg/dl

0.9 (0.8–1)

0.9 (0.8–1)

1 (0.8–1.1)

>0.99

0.72

0.9

Hemoglobin, g/dl

14.3 (13.4–15.4)

14 (13.1–15)

14.5 (13.3–16.5)

0.14

>0.99

0.9

Platelets, × 103/μl

189 (140–231)

172.5 (118.5–207.5)

149 (108.4–167)

0.02

0.09

0.36

HCV RNA, × 106 IU/ml

0.9 (0.3–2.6)

1 (0.3–2.8)

0.2 (<⁠0.1–1.7)

>0.99

0.21

0.21

Data are presented as number (percentage) or median (interquartile range).

SI conversion factors: to convert ALT to μkat/l, multiply by 0.0167; bilirubin to μmol/l, by 17.104; albumin to g/l, by 10; creatinine to μmol/l, by 88.4; hemoglobin to g/l, by 10.

Abbreviations: ALT, alanine transaminase; anti-HBc, antibodies against hepatitis B core protein; BMI, body mass index; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; INR, international normalized ratio

Severity of liver disease

The group C had the highest percentage of patients with F4 fibrosis (37.5%), followed by the group B (29.4%), as compared with 21.1% in the group A. However, these differences were not significant following the Bonferroni correction due to a small number of patients, particularly in the HBsAg-positive population (Table 2). The patients with cirrhosis in the group B were more likely to be rated as grade B or C on the CP scale than those without HBV coinfection. All patients with cirrhosis in the HBsAg-positive group were scored as A on the CP scale. Among the HBsAg-negative / anti-HBc-positive cirrhotic individuals, 5 presented moderate ascites and 6 had encephalopathy. Twelve patients in the group A and 3 in the group B had a history of hepatocellular carcinoma, and 3 and 1, respectively, were liver transplant recipients; these events were not reported among the HBsAg-positive patients.

Table 2. Characteristics of liver disease with reference to hepatitis B virus status

Parameter

Group A: HBsAg(–) / anti-HBc(–) (n = 950)

Group B: HBsAg(–) / anti-HBc(+) (n = 160)

Group C: HBsAg(+) (n = 8)

P value

A vs B

A vs C

B vs C

F4 liver fibrosis

200 (21.1)

47 (29.4)

3 (37.5)

0.14

>0.99

>0.99

Child–Pugh scale

B

33 (3.5)

12 (7.5)

0

0.06

0.14

>0.99

C

5 (0.5)

3 (1.9)

0

0.06

0.51

>0.99

History of hepatic decompensation

Ascites

31 (3.3)

9 (5.6)

1 (12.5)

0.51

0.51

>0.99

Encephalopathy

11 (1.2)

6 (3.8)

0

0.03

0.13

>0.99

Esophageal varices

90 (9.5)

18 (11.3)

2 (25)

>0.99

0.84

0.73

Hepatic decompensation at baseline

Moderate ascites – responded to diuretics

21 (2.2)

8 (5)

1 (12.5)

0.16

0.09

>0.99

Tense ascites – not responded to diuretics

3 (0.3)

2 (1.3)

0

0.3

0.78

>0.99

Encephalopathy

7 (0.7)

6 (3.8)

0

0.003

0.03

>0.99

HCC history

12 (1.3)

3 (1.9)

0

>0.99

>0.99

>0.99

OLTx history

3 (0.3)

1 (0.6)

0

>0.99

>0.99

>0.99

HIV coinfection

2 (0.2)

0

0

>0.99

>0.99

>0.99

Abbreviations: OLTx, orthotopic liver transplantation; others, see Table 1

Hepatitis C virus infection and direct-acting antiviral treatment

Among the HBsAg-positive patients, genotype 3 type infections were significantly more common (37.5%) (Table 3). The percentage of treatment-naive patients was the highest in the group A (82%), and the lowest in the group C (62.5%). A comparable proportion of patients in all analyzed groups received genotype-specific and pangenotypic regimens.

Table 3. Hepatitis C virus infection and antiviral treatment characteristics with reference to hepatitis B virus status

Parameter

Group A: HBsAg(–) / anti-HBc(–) (n = 950)

Group B: HBsAg(–) / anti-HBc(+) (n = 160)

Group C: HBsAg(+) (n = 8)

P value

A vs B

A vs C

B vs C

GT

1

50 (5.3)

7 (4.4)

0

0.6

0.6

0.24

1a

12 (1.3)

1 (0.6)

0

1b

762 (80.2)

134 (83.8)

5 (62.5)

3

100 (10.5)

11 (6.9)

3 (37.5)

4

24 (2.5)

7 (4.4)

0

6

2 (0.2)

0

0

Mixed GT

2 (0.2)

2 (1.3)

0

GT3 / non-GT3

100 (10.5) / 850 (89.5)

11 (6.9) / 149 (93.1)

3 (37.5) / 5 (62.5)

0.45

0.03

0.06

GT1b / non-GT1b

762 (80.2)/188 (19.8)

134 (83.8)/26 (16.2)

5 (62.5)/3 (37.5)

0.87

0.6

0.42

History of previous therapy

Treatment-naive

779 (82)

124 (77.5)

5 (62.5)

0.15

0.03

0.06

Null-responder

61 (6.4)

20 (12.5)

0

Relapser

68 (7.2)

11 (6.9)

3 (37.5)

Discontinuation due to safety reasons

42 (4.4)

5 (3.1)

0

Current treatment regimen

GT-specific treatment regimens

Overall

493 (51.9)

96 (60)

4 (50)

0.18

>0.99

>0.99

ASV + DCV

18 (1.9)

1 (0.6)

0

LDV/SOF ± RBV

145 (15.3)

30 (18.7)

3 (37.5)

OBV/PTV/r ± DSV ± RBV

193 (20.3)

39 (24.4)

1 (12.5)

GZR/(EBR ± RBV)

133 (14)

26 (16.3)

0

SOF + SMV ± RBV

4 (0.4)

0

0

Pangenotypic regimens

Overall

457 (48.1)

64 (40)

4 (50)

0.18

>0.99

>0.99

SOF + DCV ± RBV

1 (0.1)

1 (0.6)

0

SOF + RBV

29 (3.1)

4 (2.5)

1 (12.5)

GLE/PIB

253 (26.6)

32 (20)

2 (25)

SOF/VEL ± RBV

173 (18.2)

27 (16.9)

1 (12.5)

SOF/VEL/VOX

1 (0.1)

0

0

Current RBV-containing therapies

118 (12.4)

31 (19.4)

5 (62.5)

0.06

0.003

0.01

Abbreviations: ASV, asunaprevir; DCV, daclatasvir; DSV, dasabuvir; EBR, elbasvir; GLE, glecaprevir; GT, genotype; GZR, grazoprevir; LDV, ledipasvir; OBV, ombitasvir; PIB, pibrentasvir; PTV, paritaprevir; RBV, ribavirin; SMV, simeprevir; SOF, sofosbuvir; VEL, velpatasvir; VOX, voxilaprevir; others, see Table 1

The response to DAA therapy did not differ significantly between the studied groups, with the SVR calculated for the PP population at 98.4%, 98.7%, and 100% in the groups A, B, and C, respectively (Figure 1). The number of patients lost to follow-up in these groups was 15/950 (1.5%), 3/160 (1.8%), and 0/8, respectively.

Figure 1. Direct-acting antiviral treatment effectiveness with reference to hepatitis B virus status

Abbreviations: ITT, intent-to-treat; PP, per protocol; others, see Table 1

Most patients completed the therapy according to schedule, regardless of their HBV status (Table 4). Thirty-one patients in the group A and 6 in the group B experienced severe adverse events. A total of 13 deaths were reported, 10 in the group A and 3 in the group B, 12 of which involved patients with cirrhosis, 8 of them decompensated.

Table 4. Safety of direct-acting antiviral therapy with reference to hepatitis B virus status

Parameter

HBsAg(–) / anti-HBc(–) (n = 950)

HBsAg(–) / anti-HBc(+) (n = 160)

HBsAg(+) (n = 8)

Treatment course

According to schedule

922 (97)

152 (95)

8 (100)

Therapy modification (RBV dose)

15 (1.6)

5 (3.1)

0

Therapy discontinuation

13 (1.4)

3 (1.9)

0

Patients with at least 1 AE

136 (14.3)

31 (19.4)

0

Serious AEs

31 (3.3)

6 (3.8)

0

AEs leading to treatment discontinuation

5 (0.5)

2 (1.3)

0

AEs of particular interest (in cirrhotics)

Ascites

9 (0.9)

2 (1.3)

0

Hepatic encephalopathy

7 (0.7)

5 (3.1)

0

Gastrointestinal bleeding

1 (0.1)

0

0

Death

10 (1.1)a

3 (1.9)b

0

Data are presented as number (percentage).

a 2 × COVID-19, 1 × cerebral stroke, 1 × liver impairment with gastrointestinal bleeding, 1 × renal failure, 2 × cardiac arrest, 1 × chronic lymphocytic leukemia, 2 × HCC

b Liver and renal impairment, HCC, pelvic bone fracture

Abbreviations: AE, adverse event; others, see Tables 1 and 3

Hepatitis B virus coinfection

The 8 HBsAg-positive patients accounted for 0.7%, and the 160 anti-HBc-positive individuals for 14.3% of the total analyzed population. Of the HBsAg-positive patients, 6 were men, and 5 were treated with antiviral therapy for more than a year, 4 with entecavir and 1 with lamivudine. Of those treated, 1 patient had a low HBV viral load at the start of DAA therapy and became negative at week 4. In another patient, entecavir therapy was introduced before DAA treatment as part of prophylaxis against HBV reactivation; the patient started an IFN-free regimen with a viral load of 124 IU/ml of HBV DNA, but at week 4 he was negative. All patients were negative for HBeAg and had anti-HBe antibodies (Table 5).

Table 5. Baseline characteristics of hepatitis C virus-infected patients positive for hepatitis B virus surface antigen

Sex

Age

Genotype

Comorbidities

Comedications

Fibrosis; CP grade

HBe / anti-HBe

HBV DNA

M

51

3

Osteopenia

Entecavira, testosterone

1

Negative / positive

124 IU/ml

M

47

3

Hemophilia A

Lamivudine, coagulation factor VIII

2

Negative / positive

Negative

F

39

1b

Fibrodysplasia ossificans progressiva

Entecavir

2

Negative / positive

62 IU/ml

M

37

3

Acute lymphoblastic leukemia, epilepsy, cholelithiasis, depression

Entecavir, lamotrigine

4; CP-A

Negative / positive

Negative

F

35

1b

Neuroblastoma, kidney failure, hypertension, endometriosis

Losartan

4; CP-A

Negative / positive

Negative

M

31

1b

Renal failure

Entecavir

1

Negative / positive

Negative

M

30

1b

Acute lymphoblastic leukemia, skin cancer, epilepsy, Barrett esophagus

Entecavir, levetiracetam, topiramate, spironolactone

4; CP-A

Negative / positive

Negative

M

30

1b

3

Negative / positive

Negative

a Entecavir included before direct-acting antivirals as prophylaxis of HBV reactivation

Abbreviations: CP, Child–Pugh scale; F, female; HBe, hepatitis B envelope protein; M, male; others, see Tables 1 and 3

Three cases meeting the definition of HBV reactivation were reported during DAA therapy and follow-up, 1 in a patient positive for HBsAg (12.5%) and 2 in those with anti-HBc antibodies present (1.25%). All were clinically silent and diagnosed based on laboratory findings, and all patients with HBV reactivation achieved SVR after DAA treatment.

Reactivation in the HBsAg-positive genotype 1b-infected man with F3 fibrosis occurred at the end of 12-week ombitasvir / paritaprevir /ritonavir + dasabuvir+ ribavirin therapy. Alanine aminotransferase (ALT) activity was within the normal range, but HBV DNA determination showed a viral load of 19 IU/ml. A follow-up test performed 12 weeks after the end of DAA therapy was negative (Table 6). The patient was monitored periodically due to the presence of HBsAg and continued to be positive for 4 years. However, with negative HBV DNA and the absence of progression of liver fibrosis, the patient did not meet the criteria for inclusion of antiviral therapy. A recent evaluation carried out 5 years after DAA therapy showed negative HBsAg with positive anti-HBs antibodies at 130 mIU/ml, negative HBV DNA, and liver stiffness of F2.

Table 6. Characteristics of hepatitis C virus-infected patients with hepatitis B virus reactivation

Parameter

Patient 1

Patient 2

Patient 3

Sex; age, y

M; 30

F; 53

F; 35

Genotype

1b

4

1b

Fibrosis

3

1

1

Comorbidities

None

After kidney transplant, renal failure, hypertension

Rheumatoid arthritis

Comedications

None

Mofetil mycophenolate, tacrolimus, prednisone, losartan, clonidine, lacidipine, bisoprolol

Methylprednisolone

Anti-HBV therapy

None

Lamivudine

None

Baseline HBsAg

Positive

Negative

Negative

Baseline HBeAg/anti-HBe

Negative / positive

Not done

Not done

Baseline HBV DNA

Negative

Negative

435 IU/ml

HBV at reactivation

12 week FU

24 week FU

12 week FU

ALT at reactivation

Normal

Normal

16 × ULN

HBV DNA at reactivation

19 IU/ml

107 IU/ml

105 IU/ml

HBsAg/HBeAg at reactivation

Positive / negative

Positive / negative

Positive / negative

Anti-HBV therapy at reactivation

None

Tenofovir

Tenofovir

Outcome

5-year FU: ALT normal, HBsAg negative, HBV DNA negative, F2

4-year FU: ALT normal, HBsAg positive, HBV DNA negative, F1

4-year FU: ALT normal, HBsAg negative, HBV DNA negative, F1

DAA regimen

OBV/PTV/r+DSV +RBV, 12 weeks

GZR/EBV, 16 weeks

GZR/EBV, 12 weeks

SVR

Yes

Yes

Yes

Abbreviations: FU, follow-up; SVR, sustained virologic response; r, ritonavir; ULN, upper limit of normal; others, see Tables 1, 3, and 5

Both cases of reactivation in the group with positive anti-HBc antibodies involved women on immunosuppressive therapy (Table 6). One of them, a 53-year-old individual with genotype 4 infection and F1 fibrosis, was on lamivudine for 4 years due to immunosuppressive therapy after kidney transplant. Twenty-four weeks after the end of a 16-week grazoprevir / elbasvir (GZR/EBR) therapy, the appearance of HBsAg with normal ALT activity, negative HBeAg, and HBV DNA of 107 IU/ml were reported. She was switched to tenofovir at a dose adjusted to impaired renal function (estimated glomerular filtration rate 30–60 ml/min/1.73 m2), and over 2.5 years of DDA therapy the HBV DNA level was reduced and then undetectable. After 4 years of follow-up, at the moment of this paper writing, the patient is still positive for HBsAg and remains on tenofovir. The second patient, a 35-year-old woman with genotype 1b infection and F1 fibrosis, diagnosed with rheumatoid arthritis, on a low dose of methylprednisolone (4 mg daily), had HBV DNA of 435 IU/ml at the start of a 12-week GZR/EBR therapy, with negative HBsAg and anti-HBs antibodies at 14 mIU/ml. She did not receive anti-HBV therapy, and was HBsAg-negative at the end of DAA therapy; HBV DNA was not assessed. Twelve weeks after the end of DAA treatment, an increase in ALT activity up to 16 times above the normal upper limit (reference range, 5–40 U/l) was observed, accompanied by detection of HBsAg without HBeAg and an increase in HBV viral load to 105 IU/ml. She started treatment with tenofovir, and after the next 12 weeks, ALT activity was within the normal range, while HBsAg and HBV DNA were negative. As the patient was still receiving steroid therapy, she remained on tenofovir, which was stopped 18 months after discontinuation of methylprednisolone. Now she is monitored, and 6 months after tenofovir discontinuation, ALT activity is normal, and HBsAg and HBV DNA are negative. In both patients, follow-up elastography examinations, the most recently performed 4 years after DAA therapy completion, showed liver stiffness consistent with F1 fibrosis.

Discussion

Testing patients with chronic hepatitis C for HBV markers is recommended, considering the increased risk of infection due to the common transmission route of both viruses.27 Our study provides important insights into the clinical characteristics of coinfected patients, the risk of HBV reactivation during DAA treatment of HCV infection, and the effectiveness of antiviral therapy. It also encourages further studies conducted in other populations, ultimately enabling to shape future recommendations and guidelines on HBV/HCV coinfection management.

Most studies evaluating HBV coinfection in patients with chronic HCV infection are limited to assessing the presence of HBsAg, and data on the prevalence of anti-HBc antibodies in this patient population are scarce.3,29 In our analysis, which included a real-world single-center cohort of more than 1000 patients with chronic HCV infection, the prevalence of HBsAg was 0.7%, and over 14% showed the presence of anti-HBc antibodies. The same percentage of HBsAg-positive patients was documented in a retrospective study from Egypt30 involving 3300 patients, and all 25 individuals with dual HBV-HCV infection were young men at a mean age of 35 years, which was consistent with our findings. Another retrospective Egyptian multicenter study31 in a very large population of almost 300 000 patients with chronic HCV infection showed a rate of 0.8% of patients positive for HBsAg, which is comparable to our results. The lowest prevalence of HBV coinfection was documented in a study from Southern Ethiopia,32 in which it was only 0.3%, but it should be noted that the study analyzed only 19 patients with chronic hepatitis C. Importantly though, contrary to our research, anti-HBc antibodies were not evaluated in any of the above-cited papers. In 10 studies from the United States, the median (IQR) prevalence of HBsAg among HCV-infected patients was 1.2% (0.2%–5.8%), and the markers of past HBV infection were found in a median of 43% (4.7%–62.6%) of participants.33-38 One of possible explanations for different outcomes of our and the American studies may be different characteristics of the analyzed populations in terms of additional risk factors, for example, intravenous drug use, dialysis, HIV coinfection, or sexual behavior.

The period covered by the analysis is also significant. A study by Tyson et al38 evaluating HBV markers in the population of HCV-infected American veterans between 1997 and 2005, found that 1.4% of them were positive for HBsAg and 34.7% for anti-HBc antibodies, with a downward trend in subsequent years of evaluation. Independent risk factors for the presence of HBsAg were younger age and male sex, which is consistent with our findings. The risk was significantly higher in drug-addicted individuals and those living with HIV, which we were unable to prove, because there were only 2 HIV-infected patients in our study population. The only Polish study to date evaluating these parameters, conducted by Jaroszewicz et al,3 showed the percentage of HBsAg to be 1.1% in a cohort of more than 10 000 patients with chronic HCV infection. The prevalence of anti-HBc was estimated at 20.2%, although it should be emphasized that not all patients were tested in this research. This indicates the advantage of our analysis, and may be a possible explanation for the differences in the obtained results. In the study by Jaroszewicz et al,3 HBsAg-positive HCV-infected patients were also younger, with a higher percentage of men, HIV infections, and advanced fibrosis. Moreover, in our analysis, HBsAg-positive patients more often presented with F4 fibrosis than those with HCV monoinfection, supporting the thesis of more advanced liver disease when infected with both viruses reported in the literature.39,40 In our analysis, not only HBsAg-positive patients but also those with a past HBV infection had F4 fibrosis more often than the patients without HBV markers, which indicates that a prior HBV infection also contributes to the advancement of liver disease, supporting the findings of other authors.41 However, due to the small size of the studied groups, these differences were not significant, advocating further research in this field.

We found out that the presence of HBV serologic markers (HBsAg or anti-HBc) negatively affected the effectiveness and safety of DAA antiviral therapy. Irrespective of their HBV status, the patients achieved a very high SVR, exceeding 98%, following treatment with both genotype-specific and pangenotypic regimens, thus supporting conclusions from other real-world studies.3,42

Despite the lack of a negative impact on effectiveness of DAA therapy, our knowledge on the presence of HBV infection markers in patients with chronic hepatitis C has gained even more importance with the introduction of DAAs due to the risk of reactivation.18 Since the first report indicating the possibility of HBV infection reactivation in patients treated with DAAs was submitted to the Food and Drug Administration in 2016, many papers on this topic have been published.3,16,17,43

According to the available literature data, the incidence of HBV reactivation in HBsAg-positive patients during DAA treatment and follow-up after treatment varies widely, depending on the HBeAg status, anti-HBV therapy, DAA regimen, sensitivity of the diagnostic methods used, frequency of assessments carried out, length of observation, and size of the analyzed populations.9 Studies in large cohorts have confirmed the risk of such events at 0%–9.7%, most of which occurred during the treatment course and were observed with multiple different DAA regimens.3,44-46 Studies involving smaller cohorts of HBsAg-positive patients, ranging from 2 to 36 individuals, have documented a 2%–75% risk.9

The spectrum of clinical manifestations associated with HBV reactivation is also wide, ranging from laboratory abnormalities to life-threatening hepatitis, the latter affecting HBV endemic areas and HBeAg-positive patients.18,47

Based on these data, according to the current recommendations, all patients with diagnosed HCV infection planned for DAA treatment should be screened for HBsAg.20,46,47 Coinfected individuals fulfilling the standard criteria for HBV treatment should receive nucleoside / nucleotide analogs (NA), while HBsAg-positive patients not meeting the criteria for anti-HBV treatment should receive NA prophylaxis prior to DAA, and be monitored for reactivation for at least 12 weeks after anti-HCV therapy.

Our analysis documented HBV reactivation in 1 of 8 patients with baseline HBsAg-positive and HBeAg-negative results, corresponding to a 12.5% risk. The reactivation was clinically silent, not accompanied by an increase in ALT activity above the normal levels, and was recognized only by routinely performed HBV DNA tests, which showed a low viral load at the end of DAA therapy. Long-term follow-up showed a reduction in liver stiffness on elastography and HBsAg seroconversion, which occurred in the 5th year after the end of DAA therapy. Due to the lack of publications on long-term follow-up of patients after HBV reactivation during DAA treatment, it is difficult to relate this to other reports.

Based on data from the literature, the risk of reactivation among HCV-infected patients with anti-HBc antibodies is much lower.9 In the current study, HBV reactivation occurred in 2 out of 160 patients with these characteristics, corresponding to 1.25%, which was in line with other reports claiming the risk in the range from 0% to 8%.

Importantly, both women remained on immunosuppressive therapy, and in 1 of them reactivation occurred even despite several years of lamivudine treatment with baseline suppression of HBV viral load. Our results are consistent with observations by other authors indicating the risk of HBV reactivation in immunocompromised patients.16,48 Thus, they support national guidelines recommending anti-HBc testing before DAA therapy in immunocompromised patients, while the European and American guidelines recommend such testing in all individuals qualified for DAA treatment.20,49,50 All guidelines mentioned above recommend monitoring ALT levels in HBsAg-negative patients with anti-HBc antibodies present and evaluating for HBsAg and HBV DNA if ALT activity increases. If HBsAg and / or HBV DNA are detected, NA therapy should be initiated. Tenofovir was included in both of our patients with desirable effect, and no progression of liver fibrosis was observed during long-term follow-up. The long-term observation of patients with HBV reactivation is one of the study’s strengths that should be emphasized. None of the analyses published so far have provided such a prolonged follow-up of patients with reactivation of HBV infection during DAA therapy. The second strength is testing for the presence of both HBsAg and anti-HBc antibodies in all patients qualified for DAA therapy. In addition, it should be noted that all patients with a positive result in any of the abovementioned tests were tested for HBV DNA, which together with a large size of the analyzed population, ensured reliability of our findings. Since the analysis was performed in a single center, all patient procedures were repeatable and consistent. However, we are aware of the limitations of the analysis. As a retrospective study, it was subject to the risk of entry errors, bias, and underestimated adverse events. Due to the single-center study, the results may not represent the entire HCV-infected population. The small share of HIV-coinfected patients did not allow for the assessment of the impact of this infection on the described phenomena.

Conclusions

This real-world population study of HCV-infected patients found that 0.7% were positive for HBsAg and 14.3% had a past HBV infection. HBsAg-positive patients were younger and had more advanced liver fibrosis. During DAA treatment, the risk of HBV reactivation in the HBsAg-positive patients was 12.5%, while in the population of anti-HBc-positive individuals, it was much lower and concerned immunosuppressed patients. Neiter the presence of HBV markers nor the occurrence of reactivation reduced the effectiveness of DAA therapy.