Biologic disease-modifying antirheumatic drugs (DMARDs) have emerged as essential and effective therapies for a wide spectrum of conditions including treating cancer, preventing transplant rejection or graft-versus-host disease, and suppressing autoimmune diseases [1]. Comorbidities in biologic recipients can be problematic; increased replication or even reactivation of latent viral infection is one of the major concerns [2]. Hepatitis B viral (HBV) reactivation has been a possibility in cancer chemotherapy and tumor necrosis factor alpha (TNF-α) inhibitor treatment [3-5]. The clinical spectrum of viral reactivation can be diverse, ranging from simple transaminitis to characteristic hepatitis flare, decompensated hepatic failure, or even death [1]. Recent treatment guidelines recommend prophylactic or preemptive anti-viral therapy for patients with chronic hepatitis B [6]. In contrast, prophylaxis for patients with chronic hepatitis C during biologic DMARD use is not emphasized. Hepatitis C virus (HCV) infection is one of the leading causes of liver cirrhosis and hepatocellular carcinoma, and it often requires liver transplantation [7]. Immunosuppression is also known to induce viral replication of HCV [1,8-10], and clinical outcomes of HCV reactivation owing to biologic DMARDs appear to be no less severe than the outcomes of reactivated HBV [1,8,9]. Thus, identifying patients at risk or early detection of HCV reactivation is pivotal for decreasing the incidence of related morbidities.

Rheumatoid arthritis (RA) is one of the most common chronic inflammatory diseases for which biologic DMARDs are used. The prevalence of HCV infection in RA patients ranges from 0.02% to 0.65% [11,12]. Although we now encounter an increasing number of RA patients with concomitant chronic HCV infection who have an inadequate response to conventional DMARDs, prescribing biologic DMARDs to this group of patients is often reserved owing to concerns about increased viral replication or overt viral hepatitis [13].

Only a few studies have followed the outcomes of virologic breakthrough in RA patients with chronic HCV infection treated with biologic DMARDs. The aim of this study was to investigate the changes in transaminase and viral load in HCV-infected Korean RA patients after they began taking biologic agents.

Patient enrollment

This was a multicenter retrospective study for which we enrolled RA patients with chronic HCV infection who were identified from 12 University Hospitals in Korea between November 2014 and November 2015. All patients met the 2010 American College of Rheumatology and European League Against Rheumatism classification criteria of RA [14]. We defined biologic DMARD users as patients who received at least one biologic DMARD during the course of RA treatment. HCV infection had been identified by the presence of anti-HCV antibodies, and the diagnosis was verified by the attending physician. The protocol was approved by the ethics committee in each institute (no. 1402-080-558).

Collection of study data

Baseline and longitudinal clinical data had been obtained including body mass index, history of cigarette smoking, date of RA diagnosis, duration of disease, comorbidities, rheumatoid factor (RF) and anti-cyclic citrullinated peptide antibody (anti-CCP) titers, erythrocyte sedimentation rate and C-reactive protein levels at diagnosis, prior and concomitant use of conventional DMARDs, and number of biologic agents (etanercept, adalimumab, infliximab, golimumab, tocilizumab, abatacept, and rituximab) prescribed during the study period. Information related to chronic HCV infection included the HCV genotype, viral RNA titer, list of anti-viral treatments, and extrahepatic manifestations.

Definitions of acute exacerbation and reactivation of hepatitis C

We evaluated safety profiles including changes in transaminase (i.e., serum aspartate aminotransferase [AST], alanine aminotransferase [ALT]), and viral HCV RNA titer during biologic therapy. Data on serum AST and ALT had been collected every three months from initiation of the biologic agent. We were also able to obtain available serum HCV RNA levels before and after biologic therapy. We defined HCV reactivation as an increase in HCV RNA viral load >1 log₁₀ IU/mL (virologic breakthrough) plus a ≥threefold increase in serum ALT or AST that could be explained by no other cause [2,15,16].

Characteristics of RA patients

We identified 78 RA patients with concomitant HCV infection (60 women and 18 men); demographics and clinical characteristics are summarized in Table 1. A total of 17 patients received biologic DMARD(s) during RA treatment. Mean age (SD) at RA diagnosis was 52.8 (13.1) years and mean disease duration was 10.0 (6.1) years. Six of the 17 patients (35.3%) had been diagnosed with hepatitis C prior to RA. One patient was co-infected with HBV and HCV. Two patients had extrahepatic manifestations of HCV infection: hypothyroidism (n=1), and mixed cryoglobulinemia (n=1). The most common cDMARD prescribed was hydroxychloroquine (83.6%), followed by methotrexate (50.8%) and sulfasalazine (36.1%); see Figure 1 for details. Regarding anti-viral treatment, one patient previously received interferon (IFN) and ribavirin (RBV) as acute hepatitis C treatment before initiating DMARD therapy (approximately 46 months ago).

Table 1 . Characteristics of patients with chronic hepatitis C at the time of rheumatoid arthritis diagnosis.

Variable	Biologic DMARD ever-users(n=17)	ConventionalDMARD only-users (n=61)	Total patients (n=78)
Age (yr)	52.8±13.1	57.1±10.6	56.2±11.2
Female gender	14 (82.4)	46 (75.4)	60 (76.9)
Disease duration of RA (yr)	10.0±6.1	7.1±4.6	7.8±5.0
BMI (n=69)	20.2±6.6	23.7±0.5	23.1±3.3
Alcohol consumer (n=69)	1 (5.9)	7 (11.5)	8 (11.6)
Smoker (n=71)	0	10 (16.4)	10 (14.1)
Previous	0	7 (11.5)	7 (9.0)
Current	0	3 (4.9)	3 (3.8)
RF positive (n=78)	15 (88.2)	46 (75.4)	61 (78.2)
Titer (IU/mL) (n=69)*	246.7±278.5	108.6±170.3	140.7±206.8
Anti-CCP positive (n=55)	9 (81.8)	32 (72.7)	41 (74.5)
Titer (IU/mL) (n=38)†	112.3±131.7	46.2±36.7	103.6±125.1
ESR (mm/h) (n=76)‡	50.2±21.6	39.4±25.0	41.8±24.5
CRP (mg/dL) (n=71)§	2.1±2.2	1.9±3.4	2.0±3.2
Comorbidities (n=77)
Fatty liver	2 (11.8)	6 (9.8)	8 (10.4)
Diabetes mellitus	2 (11.8)	11 (18.0)	13 (16.9)
Hypertension	3 (17.6)	25 (41.0)	28 (36.4)
Duration of HCV infection, years	7.6±1.9	6.6±3.3	8.0±5.9
HCV genotype (n=24)
Genotype 1a	0	0	0
Genotype 1b	1 (5.9)	6 (9.8)	7 (9.0)
Genotype 2a	2 (11.8)	14 (23.0)	16 (20.5)
Genotype 2b	0	1 (1.6)	1 (1.3)
Anti-viral treatment	1 (5.9)	14 (23.0)	15 (19.2)
Peg-interferon	0	4 (6.6)	4 (5.2)
Ribavirin	0	1 (1.6)	1 (1.3)
Peg-interferon+ribavirin	2 (11.8)	8 (13.1)	10 (12.8)
Sofosbuvir+ribavirin	0	0	0
HBV co-infection	1 (5.9)	2 (3.3)	3 (3.9)

Values are presented as mean±standard deviation or number (%). DMARD: disease-modifying antirheumatic drug, RA: rheumatoid arthritis, BMI: body mass index, RF: rheumatoid factor, Anti-CCP: anti-cyclic citrullinated protein antibody, ESR: erythrocyte sedimentation rate, CRP: C-reactive protein, HCV: hepatitis C virus, HBV: hepatitis B virus. p-values were generated by using Mann-Whitney U test; *p=0.024, ^†p=0.300, ^‡p=0.080, ^§p=0.402..

Figure 1. Conventional disease-modifying antirheumatic drug (DMARD) use in rheumatoid arthritis patients with chronic hepatitis C. HCQ: hydroxychloroquine, MTX: methotrexate, SSLZ: sulfasalazine, LEF: leflunomide, TAC: tacrolimus, CsA: cyclosporine A.

Clinical traits of biologic DMARD users

Seventeen patients were prescribed with biologic DMARD(s) during RA treatment: etanercept (n=8), adalimumab (n=8), infliximab (n=2), tocilizumab (n=2), abatacept (n=1), and golimumab (n=1). The mean treatment durations were 35.6 months for etanercept, 31.2 months for adalimumab, 10.4 months for infliximab, 22.9 months for tocilizumab, 14.1 months for abatacept, and 1.9 months for golimumab (Table 2). Etanercept was the agent with the longest exposure (23.7 patientㆍyears) followed by adalimumab (20.8 patientㆍyears). Detailed information on each patient is provided in Table 3.

Table 2 . Biologic agents used in 17 rheumatoid arthritis patients with chronic hepatitis C.

Agents	Mean treatment duration (months)	Duration of exposure (patientㆍyears)
Etanercept (n=8)	35.6 (4.1∼87.4)	23.7
Adalimumab (n=8)	31.2 (8.6∼73.3)	20.8
Infliximab (n=2)	10.4 (6.3∼14.4)	1.7
Tocilizumab (n=2)	22.9 (18.1∼27.8)	3.8
Abatacept (n=1)	14.1	1.2
Golimumab (n=1)	1.9	0.2

Values are presented as mean (minimum to maximum) or mean only..

Table 3 . Clinical characteristics of biologic users in rheumatoid arthritis patients with chronic hepatitis C.

No	Gen-der	Age (yr)	Disease duration of RA (yr)	RF titer (IU/mL)	Anti-CCP titer (IU/mL)	Duration of HCV infection (yr)	HCV RNA titer before biologic use (IU/mL)	Prophylactic anti-viral therapy	Biologic agent(s)	Treatment duration (mo)	Concomit-ant DMARD use	Baseline AST/ALT (U/L)	Follow-up HCV RNA titer (IU/mL) (mo)	Peak AST/ALT (U/L) within the 2nd year (mo)	Subsequent anti-viral therapies
1	F	71	7.1	NC	1.5	16.2	17,600	No	ETX	87.4	None	190/218	NC	105/39 (8)	-
2	F	69	10.5	634	35.5	4.3	2,760,000	Previous (IFN+RBV)	ETX	5.3	MTX	41/28	NC	33/22 (9)	-
3	F	37	15.1	152	>100	4.8	Negative	No	ETX	67.5	None	21/16	Negative	40/22 (24)	-
4	F	76	18.8	68.1	28.3	4.7	NC	No	ETX	18.5	None	22/9	NC	30/10 (9)	-
5	F	58	8.5	224	NC	0.3	NC	No	ETX→ ABT	4.1→3.8	MTX →MTX	15/9	NC	21/18 (15)	-
6	F	73	17.3	175	NC	17.3	Negative	No	ETX→ ADA	19.6→19.1	MTX →None	17/12	Negative	23/15 (8)	-
7	F	78	7.8	36	NC	Un-known	3,811,947	No	ETX → GOL → TCZ	9.8→1.9 →27.7	None	68/34	NC	165/140 (21.3) →53/13 (22.6)	-
8	F	59	3.5	218.2	NC	18.5	NC	No	ADA	29.4	MTX, HCQ	85/83	25,590 (3) →52,512 (27)	238/162 (16)	-
9	F	69	21.5	43	69	1.3	NC	No	ADA	40.6	MTX	37/20	2,330,000 (24.7) →17,500 (27.5)	134/75 (24.7) →44/23(27.5)	IFN+RBV
10	F	39	8.8	12.3	NC	8.8	Negative	No	ADA	73.3	MTX	12/8	Negative	24/28 (6)	-
11	M	62	7.8	26.1	96.1	7.6	NC	No	ADA	8.6	MTX	18/25	NC	22/29 (6)	-
12	F	42	3.4	67.5	>200	3.5	Negative	No	ADA	29.7	MTX	26/13	NC	23/11 (9)	-
13	F	68	1.6	405.9	NC	2.9	NC	No	ADA	11.1	MTX	18/15	<15 (34.5) →15 (43.1)	20/13 (24)	-
14	F	68	18.1	960.8	>100	6.8	Negative	No	ADA→ETX	Unknown	Unknown	Unknown	NC	Unknown	-
15	M	74	7	251	>600	7	NC	No	IFX	6.3	SSLZ	18/13	NC	19/14 (4)	-
16	F	53	3.9	26.5	<7	22.3	NC	No	IFX	14.4	None	69/18	205,000 (21.1)	48/26 (9)	-
17	M	73	9.4	647	>100	25.5	839,227	No	TCZ	18.1	SSLZ	37/30	2,110,902 (4.5)→ 3,958,308 (14.4)	366/438 (18.1)	Sofosbuvir+RBV

RA: rheumatoid arthritis, RF: rheumatoid factor, Anti-CCP: anti-cyclic citrullinated protein antibody, HCV: hepatitis C virus, DMARDs: disease-modifying antirheumatic drugs, AST: aspartate aminotransferase, ALT: alanine transaminase, NC: not checked, IFN: interferon, RBV: ribavirin, ETX: etanercept, ABT: abatacept, ADA: adalimumab, GOL: golimumab, TCZ: tocilizumab, MTX: methotrexate, HCQ: hydroxychloroquine, SSLZ: sulfasalazine..

Biologic DMARD users had higher levels of inflammatory markers and RF and/or anti-CCP titers at RA diagnosis than did conventional DMARD (-only) users. Methotrexate and sulfasalazine were more commonly prescribed during biologic DMARD use (Figure 1). One patient received IFN and RBV as acute hepatitis C treatment approximately 46 months before starting etanercept.

Outcomes of HCV replication and transaminitis in biologic DMARD users

There was no case of HCV over-reactivation in RA patients with chronic HCV infection who were exposed to biologic DMARD(s). Methotrexate was the most common anchoring drug at initiating biologic therapy (8/16, 50%). In the majority of patients, transaminase level was stable during biologic therapy. Liver function in eight RA patients who received etanercept were stable throughout the course of etanercept (patients 1∼7 and 14). Three patients who received etanercept as the first biologic DMARD switched to abatacept (patient 5), adalimumab (patient 6), or golimumab followed by tocilizumab (patient 7). Patient 7 developed a transient increase in ALT at 21.3 months (tocilizumab), but it normalized within two months. Two of eight patients (patients 8 and 9) who used adalimumab developed transaminitis; patient 8 discontinued adalimumab at 29 months because of repeatedly elevated ALT with increased HCV RNA, and ALT normalized after discontinuing adalimumab. Patient 9 experienced elevated ALT plus a substantially high viral RNA titer at 24.7 months and was preemptively treated with IFN and RBV (Table 3). The viral RNA titer rapidly decreased within three months.

Among patients treated with tocilizumab, one (patient 17) with a history of compensated liver cirrhosis and hepatocellular carcinoma developed an increase of 4.7 times in viral RNA titer plus a 10-fold increase in ALT. He was immediately treated with sofosbuvir and RBV, and his viral RNA titer decreased markedly within a month. Changes in transaminase and viral RNA were unremarkable in other patients who received infliximab (patients 15 and 16) and abatacept (patient 5).

Treatment guidelines for RA have evolved during the past decade driven by the incorporation of biologic DMARDs, enabling patients to better achieve clinical remission or low disease activity. In patients infected with chronic viral hepatitis, the decision to treat with biologic therapy is often reserved due to concerns about viral reactivation and/or deterioration of hepatic function. We reviewed 78 Korean RA patients with chronic HCV infection; 17 patients had been treated with biologic DMARDs, mainly TNF-α inhibitors, and most had stable liver function during biologic therapy. Two patients experienced transaminitis plus notably increased viral RNA titers compared with baseline, but preemptive anti-viral therapy readily solved the events.

Adverse hepatic outcomes of TNF-α inhibitors in HCV-infected RA patients have been reported in the literature [13,17-27]. In one study, the safety profiles of TNF-α inhibitors in HCV-infected patients appeared to be acceptable, irrespective of agent or underlying chronic inflammatory condition [13]. One study showed that etanercept was most frequently used in RA patients with chronic HCV infection [12]; patients maintained their hepatic function and viral load, with only one of 61 patients experiencing HCV reactivation [13,28]. Safety signals of etanercept plus methotrexate in a prospective, multicenter study concluded that TNF-α inhibitors can be used in daily clinical practice without increased risk of hepatotoxicity or viral replication in HCV carriers [20]. In our study, etanercept users also showed no significant hepatic dysfunction or virologic breakthrough during follow-up (23.7 patientㆍyears).

Long-term use of TNF-α inhibitors in HBV and HCV co-infectedpatients may augment viral replication, requiring anti-viral therapy [18]. In our study, one out of three HBV and HCV co-infected patients received adalimumab followed by etanercept with no alarming safety signals. In fact, HCV RNA was negative and HBV DNA titer was low when the patients began the biologics. Unfortunately, whether the patient received prophylactic anti-viral therapy for HBV infection was unknown in this case. In contrast, two of eight patients treated with adalimumab experienced transaminitis plus increased viral RNA titers. This result suggests that there may be a higher risk of hepatic events in users of monoclonal TNF-α inhibitors.

Two case reports suggested that short-term tocilizumab treatment may not affect hepatitis C viral load or serum transaminase [29,30]. The authors concluded that tocilizumab can be utilized in RA patients with chronic HCV infection, although monitoring liver function and viral RNA levels is required. On the contrary, there is a report of HCV reactivation in an asymptomatic carrier with RA within the first 12 months of tocilizumab treatment [31]. In our study, one patient with compensated liver cirrhosis experienced acute deterioration of hepatic function with an HCV RNA titer that was 4.7 times higher within 24 months treatment of tocilizumab. A study showed that interleukin-6 contributes to liver regeneration and protection against liver injury [32], therefore tocilizumab needs to be used with caution in patients with liver cirrhosis, especially with high copies of HCV RNA.

Previous studies reported 8.8% to 15% in HCV infected RA patients who received TNF inhibitors experienced transaminitis [13,19]. In our study, 23.5% patients developed transaminitis during biologic therapies, which was relatively higher than previous studies.

Eshbaugh and Zito [33] described four cases of abatacept-treated RA patients with hepatitis C; the authors observed no adverse event of hepatic function deterioration or increased viral load and suggested that abatacept could be used in patients with HCV infection. As for rituximab, its use is recommended with caution in patients with HBV infection [34]. Furthermore, Chen et al. [35] reported that B-cell target therapy may influence HCV reactivation to a greater extent than TNF-α inhibitors. No patient had been treated with rituximab in our study.

This is the first Korean multicenter study to investigate virologic or biochemical changes in RA patients who were receiving biologic therapy and who were also infected with HCV. There are some limitations that need to be discussed. First, this was a retrospective study based on a relatively limited number of HCV-infected patients, especially biologic DMARD users, with heterogeneous clinical characteristics, different biologics, and different follow-up periods; etanercept and adalimumab were the agents that were most prescribed, with the latter being associated with more events of interest. Second, there were missing data on viral RNA titers at certain time points, which made it difficult to assess the presence of virologic breakthrough in the presence of transaminitis. Third, biologic DMARD prescription as well as its selection or combination with a conventional DMARD was determined by the attending physician, and selected patients were likely prone to have better prognoses in terms of safety. Thus, our patients may not have fully featured the untoward outcomes in HCV-infected RA patients.

Although we could not highlight the biologic DMARD of choice, etanercept is one option for RA patients with chronic HCV infection. Importantly, HCV-infected RA patients are not free from developing hepatitis or increased viral loads during biologic DMARD use. Therefore, physicians should be vigilant in monitoring liver function and viral RNA titer before and during biologic therapy.

This work was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute, funded by the Ministry of Health & Welfare, Republic of Korea (grant number HI14C1277), a grant from the Korea Healthcare Technology R&D Project, Ministry of Healthcare & Welfare, Republic of Korea (grant number. HI10C2020), and a grant from the Ministry of Science, ICT and Future Planning (grant number. NRF-2015M3A9B6052011).

No potential conflict of interest relevant to this article was reported.