Long-term therapy with clevudine for chronic hepatitis B can be associated with myopathy characterized by depletion of mitochondrial DNA


  • Potential conflict of interest: Nothing to report.


Clevudine (Revovir), a pyrimidine nucleoside analogue, is a recently introduced antiviral drug. Clinical trials have demonstrated potent, sustained antiviral activity against hepatitis B virus without specific adverse events. The lack of cytotoxicity and absence of an effect on mitochondrial function have been considered the reasons for the fewer adverse events. However, it came to our attention that several hepatitis B patients developed myopathy during clevudine therapy. Our study was aimed to analyze the clinical and pathological features of patients with clevudine-induced myopathy with some consideration of its pathogenetic mechanism. Seven hepatitis B patients who developed severe skeletal myopathy during clevudine therapy were examined in this study. The demographic data, clinical features, pathological findings, and molecular studies of these patients were analyzed with speculation about the underlying pathogenic mechanisms. All seven patients were treated with clevudine for more than 8 months (8-13 months). In all, the main symptom was slowly progressive proximal muscular weakness over several months. A markedly elevated creatine kinase level and myopathic patterns on electromyography were found. Muscle biopsies revealed severe myonecrosis associated with numerous ragged red fibers, cytochrome c oxidase–negative fibers, and predominant type II fiber atrophy. Molecular studies using quantitative polymerase chain reaction showed a depletion of the mitochondrial DNA in the patients' skeletal muscle. Conclusion: To the best of our knowledge, this is the first report of myopathy associated with clevudine therapy. This study has clearly shown that long-term clevudine therapy can induce the depletion of mitochondrial DNA and lead to mitochondrial myopathy associated with myonecrosis. Careful clinical and laboratory attention should be paid to patients on long-term clevudine therapy for this skeletal muscle dysfunction. (HEPATOLOGY 2009.)

Hepatitis B virus (HBV) is the leading cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma worldwide. As carriers of HBV serve as reservoirs for vertical and horizontal transmissions, the prevalence of carriers and the number of acute and chronic liver diseases place HBV infection among the most frequent and important transmissible diseases of the hepatobiliary system. Appropriate therapeutic agents are required in order to prevent its irreversible sequelae.

Currently, the use of oral antiviral drugs is considered a milestone of chronic hepatitis B therapy. The majority of these are nucleoside and nucleotide analogues that interfere primarily with viral replication by inhibition of the viral DNA polymerase. However, their therapeutic effects are limited because of the development of drug resistance,1, 2 relapse following cessation of treatment,3, 4 and possible toxicity.5 Mitochondrial cytotoxicity is a well-known side effect of nucleoside analogues and is most frequently associated with therapy using nucleoside reverse transcriptase inhibitors in human immunodeficiency virus patients. Recently, clevudine (Revovir) was introduced as a potent antiviral drug for HBV, and no serious adverse events were reported during a 24-week clinical trial.6

However, we have recently seen several patients who developed muscle weakness during long-term clevudine therapy. The aim of this study was to describe the clinical and pathological features of these patients while considering the underlying pathomechanism of clevudine-associated myopathy.


ALT, alanine aminotransferase; AST, aspartate aminotransferase; ATPase, adenosine triphosphatase; CK, creatine kinase; COX, cytochrome c oxidase; COXII, cytochrome c oxidase subunit II; EM, electron microscopy; H&E, hematoxylin and eosin; HBeAg, hepatitis B e antigen; HBV, hepatitis B virus; LDH, lactate dehydrogenase; MGT, modified Gomori trichrome; mtDNA, mitochondrial DNA; NE, not evaluated; PCR, polymerase chain reaction; Q-PCR, quantitative polymerase chain reaction; SDH, succinate dehydrogenase.

Patients and Methods


Seven consecutive patients who had developed muscle weakness during long-term clevudine therapy for hepatitis B were enrolled in this study. All patients were previously diagnosed with chronic hepatitis B on the basis of serology showing the persistence of the hepatitis B surface antigen for more than 6 months. The patients were enrolled from four different university hospitals in Gyeongsang Province, South Korea. Five patients had been previously treated at clinics and transferred to university hospitals after the development of myopathy.


Clinical Evaluation.

We collected information on age, gender, clinical features, duration of clevudine therapy before symptom onset, interval between symptom onset and evaluation, daily dose of clevudine, and recovery time. Recovery time was defined as the duration between discontinuation of clevudine and improvement to normal muscle power. Muscle power was graded with the Medical Research Council scale and was measured in the proximal and distal part of upper and lower extremities and neck flexors. According to the Medical Research Council scale, muscle power is graded as follows: 0, no movement; 1, palpable contraction but no visible movement; 2, movement only with gravity eliminated; 3, movement against gravity but not against resistance; 4−, movement against minimal resistance; 4, movement against some resistance; 4+, movement against moderate to maximal resistance; and 5, normal power.7 The bulbar symptoms (dysarthria and dysphagia) were graded as follows: −, no speech or swallowing difficulties; +, slight speech dysfunction with nasal voice or mild swallowing difficulty with occasional choking; and ++, slurred and unintelligible speech or consistent swallowing difficulties with aspiration pneumonia.8 Laboratory tests, including complete blood count, liver and renal function tests, thyroid function tests, serum electrolytes, and creatine kinase (CK), chest X-rays, and electrocardiograms were performed for all patients. To evaluate potential mitochondrial damage, we collected information about serum levels of lactate and pyruvate, associated symptoms such as nausea and asthenia, and the presence of hepatic steatosis, optic neuritis, and neuropathy. The diagnosis of the presence of hepatic steatosis was based on ultrasound findings compared with previous ultrasound results from the start of clevudine therapy.

The laboratory data about HBV status, including hepatitis B e antigen, HBV DNA, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) levels at the start of clevudine therapy, were also obtained by a review of medical records. Virological response was defined as undetectable HBV DNA (<2,000 copies/mL).

The nerve conduction studies were performed with standard techniques,9 and needle electromyography was performed on at least two proximal and distal muscles from the upper and lower extremities.

Pathological Study.

Muscle biopsies were performed in six patients. All muscle samples were obtained from an open biopsy of the clinically affected muscles. In five patients, the muscle specimens were processed for frozen sections with standard methods10 and stained with hematoxylin and eosin, modified Gomori trichrome, reduced nicotinamide adenine dinucleotide tetrazolium reductase, cytochrome c oxidase (COX), succinate dehydrogenase, and adenosine triphosphatase (at pHs 10.8 and 4.3). In one patient, the muscle specimen was embedded in paraffin and was stained with hematoxylin and eosin, and immunohistochemistry for lymphocyte antigen was performed. For electron microscopy, the samples were fixed with 2% glutaraldehyde and embedded in epoxy resin according to the standard procedure. Semithin, 1-μm-thick resin sections were stained with toluidine blue for light microscopy. Ultrathin sections (60 nm thick) were cut with diamond knives on an ultramicrotome (Reichert Supernova, Leica, Germany), double-stained with uranyl acetate and lead citrate, and examined under an electron microscope (GEM1200EX-2, JEOL, Japan)

Quantification of Mitochondrial DNA (mtDNA) with Quantitative Polymerase Chain Reaction (Q-PCR).

The Q-PCR analysis was performed in three patients (patients 1, 3, and 4) and a normal control in order to quantify the mtDNA content in their skeletal muscle. All subjects gave written, informed consents for their DNA analysis, and our study was approved by the institutional review board of Pusan National University Hospital. For the quantification of mitochondrial and nuclear DNA, the cytochrome c oxidase subunit II (COXII) gene (mitochondrial) and β-actin gene (nuclear) were selected as target and reference genes, respectively. The Q-PCR analysis was performed with DNA from the skeletal muscles of the control and patients. Genomic DNA was isolated from frozen skeletal muscles as described.11 Primers for target and reference amplifications were as follows: for COXII, sense primer 5′-TGC CCG CCA TCA TCC TA-3′ and antisense primer 5′-CGT CTG TTA TGT AAA GGA TGC GT-3′,12 and for the β-actin gene, sense primer 5′-ATC ATG TTT GAG ACC TTC AA-3′ and antisense primer 5′-AGA TGG GCA CAG TGT GGG T-3′.13 Duplicate 20-μL polymerase chain reactions (PCRs) consisting of 50 ng of genomic DNA, 10 pM of each primer, and 2 μL of a mixture of LightCycler FastStart enzyme and LightCycler FastStart SYBR Green I reaction mix (Roche) were run on a LightCycler system for real-time PCR (Roche). The thermocycler conditions consisted of an initial denaturation of 95°C for 5 minutes and 50 cycles of denaturation (95°C for 10 seconds) and annealing (55°C for 5 seconds). After the 50 cycles, a melting curve analysis (55°C-95°C) was performed to confirm product specificity. After the equal amplification efficiencies of both genes were confirmed, the comparative cycle threshold method was used to quantify mtDNA and nuclear DNA by a comparison of the control and patients.


Initial hepatological information, including hepatitis B e antigen, HBV DNA, and ALT, was available for all patients except patient 6 (Table 1). Virological response was achieved in three patients (patients 2, 4, and 7) 6 to 10 months after clevudine therapy and was not achieved in patient 5. Follow-up viral level testing was not performed in patients 1 and 3, both of whom had been treated at local clinics.

Table 1. Clinical and Laboratory Data
 Patient 1Patient 2Patient 3Patient 4Patient 5Patient 6Patient 7
  1. The bulbar weakness was graded as follows: −, no speech or swallowing difficulties; +, slight speech dysfunction with nasal voice or mild swallowing difficulty with occasional choking; and ++, slurred and unintelligible speech or consistent swallowing difficulties with aspiration pneumonia. For muscle power, the Medical Research Council scale was used: 0, no movement; 1, palpable contraction but no visible movement; 2, movement only with gravity eliminated; 3, movement against gravity but not against resistance; 4−, movement against minimal resistance; 4, movement against some resistance; 4+, movement against moderate to maximal resistance; 5, normal power.

  2. Abbreviations: AST/ALT, aspartate aminotransferase/alanine aminotransferase; CK, creatine kinase; HBeAg, hepatitis B e antigen; HBV, hepatitis B virus; LDH, lactate dehydrogenase.

Age, years41323928444447
Initial HBV status at the start of clevudine therapy       
 HBV DNA, 103 copies/mL1,45938,64496248,959208,061Unknown24,222
 AST/ALT, U/L (reference range, 0-40)223/35179/1289/8664/9639/17Unknown67/82
Daily clevudine dose, mg30303030303030
Duration of clevudine therapy before symptom onset, months12881281313
Interval between symptom onset and evaluation, months4454744
Distribution and grade of weakness       
 Proximal/distal arm4+/54/54+/54+/55/54/54+/5
 Proximal/distal leg4/53/54/54+/54+/54/54−/5
Recovery time, weeks12121610121212
Laboratory tests (at the time of evaluation of myopathy)       
 CK value, U/L (reference range, 30-180)4108082479513334991171555
 AST/ALT, U/L35/12475/60279/76112/6283/8523/23140/68
 LDH, U/L (reference range, 218-472)86229191910863.6623572648

Patients' clinical and laboratory data during the evaluation of myopathy are summarized in Table 1. Patient 1 stopped taking clevudine by herself 20 days before evaluation. In others, clevudine was discontinued after the evaluations. All patients presented slowly progressive muscular weakness over several months. No patients were taking concomitant medications, except for patient 5, who had been taking colchicine for 1 year for gout. There was no definite muscle atrophy in any patient, and three of them complained of muscle pain. Motor examination showed symmetric proximal limb weakness in all patients, and weakness of the bulbar and neck muscles was observed in patients 3 and 7. The duration of clevudine therapy before the onset of symptoms was greater than 8 months (8-13 months), and the interval between symptom onset and evaluation was more than 4 months (4-7 months). All patients had regained normal muscle strength within 16 weeks after discontinuation of clevudine.

Blood tests showed elevated serum CK, AST, ALT, and lactate dehydrogenase levels. The serum CK decreased gradually and returned to normal after discontinuation of clevudine therapy in all patients (Fig. 1). A thyroid function test showed slightly decreased function in patients 2 and 3. Although these two patients were initially treated with thyroid hormone replacement therapy, the muscular weakness and CK level did not improve.

Figure 1.

Serial creatine kinase (CK) levels after discontinuation of clevudine therapy. A black arrow indicates the day on which clevudine was stopped, and a white arrow indicates the admission day for the evaluation of myopathy. The time of normalized muscle power is indicated as a black arrowhead.

Serum lactate and pyruvate levels were checked in three patients at the time of evaluation of myopathy (patients 3, 5, and 6). Lactate levels were 8.2, 2.0, and 3.1 mmol/L (reference range, 0.7-2.5 mmol/L) and pyruvate levels were 1.90, 1.33, and 0.60 mg/dL (reference range, 0.30-0.90 mg/dL) for patients 3, 5, and 6, respectively. Hepatic ultrasound was evaluated in four patients (patients 2, 3, 5, and 7), and attenuated hepatic steatosis was observed in one patient (patient 5). In the other three patients, no significant hepatic steatosis was observed in comparison with hepatic ultrasound prior to clevudine treatment. Two patients complained of asthenia, and nausea was not observed in any of the patients. No patient had clinical evidence of optic neuritis or peripheral neuropathy.

Nerve conduction studies were normal, and electromyography showed spontaneous activity, such as a positive sharp wave and fibrillation potential and a small-amplitude, short-duration motor unit potential compatible with active myopathic changes.

Muscle biopsy findings are summarized in Table 2. Muscle fiber necrosis, COX-deficient ragged red fibers, and type II fiber atrophy were remarkable and consistent features (Fig. 2A -D). In one patient (patient 3), muscle biopsy was performed twice with a 3-month interval. The initial muscle biopsy showed some ragged red fibers and mild type 2 fiber atrophy (Fig. 3A,B). Clevudine therapy was continued because it was not considered to be the cause of myopathy at that point in time. After 3 months of clinical worsening, the second muscle biopsy showed definite worsening of pathological features: numerous COX-deficient ragged red fibers, marked selective type 2 fiber atrophy, and marked myonecrosis (Fig. 3C,D). In terms of pathological findings, patient 1 showed the most severe features with massive muscle fiber necrosis. Electron microscopy revealed mitochondrial proliferation and enlarged mitochondria that contained rectangular inclusions (Fig. 2E,F).

Table 2. Pathological Findings
 Patient 1Patient 2Patient 3Patient 4Patient 5Patient 6
  1. Abbreviations: ATPase, adenosine triphosphatase; COX, cytochrome c oxidase; EM, electron microscopy; H&E, hematoxylin and eosin; MGT, modified Gomori trichrome; NE, not evaluated; SDH, succinate dehydrogenase.

Myonecrosis (H&E)+++++±
Mitochondrial proliferation (MGT)+NE++++
Type II fiber atrophy (ATPase)+NE+±±+
COX-negative fibers (COX)+NE++++
Ragged red fibers (SDH)NENE++++
Abnormal mitochondria (EM)+NE++++
Figure 2.

(A) Patient 1. Marked muscle fiber necrosis with degenerating and regenerating muscle fibers (hematoxylin and eosin; original magnification, ×200). (B) Patient 3. Numerous cytochrome c oxidase–negative fibers (cytochrome c oxidase; original magnification, ×200). (C) Patient 6. Mitochondrial proliferation (modified Gomori trichrome; original magnification, ×200). (D) Patient 4. Ragged red fibers (succinate dehydrogenase; original magnification, ×100). (E) Patient 1. Abnormal proliferation of mitochondria. (F) Patient 4. Enlarged mitochondria with rectangular inclusions. Bar: (A-D) 100 μm, (E) 2 nm, and (F) 0.5 nm.

Figure 3.

Serial pathological features of patient 3. (A,B) Degeneration of muscle fibers (hematoxylin and eosin; original magnification, ×200) and mild type 2 fiber atrophy (adenosine triphosphatase at pH 10.8; original magnification, ×100) from initial biopsy. (C,D) Marked myonecrosis (hematoxylin and eosin; original magnification, ×200) and marked selective type 2 fiber atrophy (adenosine triphosphatase at pH 10.8; original magnification, ×100) after 3 months of clinical worsening.

On Q-PCR analysis, the calculated mtDNA/nuclear DNA ratio, a measure of mtDNA content, was reduced to less than half in comparison with the control in all three patients (48.4%, 41.4%, and 47.8%, respectively; Fig. 4).

Figure 4.

Mitochondrial DNA contents of the patients and control measured as the cytochrome c oxidase subunit II (COXII)/β-actin ratio.


To the best of our knowledge, this is the first report of clinical myopathy associated with clevudine therapy. The link between the development of myopathy and clevudine use in these patients was clear and convincing. First, all patients developed their symptoms during their course of long-term clevudine therapy without an alternative cause of myopathy. In addition, all of our patients showed progressive improvement of their muscle weakness and reduction of serum CK levels after the discontinuation of clevudine without specific management.

Clevudine [1-(2-deoxy-2-fluoro-β-L-arabinofuranosyl)-5-methyluracil] is a pyrimidine nucleoside analogue that has been shown to have potent activity against HBV, and it was recently approved for use in Korea. The mechanism of action of clevudine involves the inhibition of the HBV polymerase,14 and in vivo efficacy studies performed in duck and woodchuck models showed marked, rapid inhibition of virus replication without significant toxicity.15 A 24-week clinical trial showed no meaningful difference between clevudine and placebo in the incidence of serious adverse events.6 However, our study clearly indicates that the long-term use of clevudine may cause myopathy. It is remarkable that the muscle complaints developed after clevudine treatment lasting longer than 32 weeks in all of our patients. This may explain why no adverse muscle events were reported during the 24-week clinical trial.

The most consistent pathological feature in our patients was the mitochondrial myopathy characterized by numerous ragged red fibers and COX-negative fibers. Variable degrees of muscle fiber necrosis were also associated with high serum CK levels. In addition, the Q-PCR analysis has confirmed that such pathological alterations are associated with the depletion of mtDNA. Mitochondrial dysfunction is a well-known side effect of the nucleoside analogues. The most famous example is zidovudine, which has been mainly used to treat human immunodeficiency virus infection.16 Zidovudine-induced myopathy is associated with functional and morphological alterations of mitochondria resembling those observed in our study. In zidovudine-induced myopathy, a molecular analysis of muscle biopsies from patients showed depletion of mtDNA,17 which is caused by the inhibition of mtDNA polymerase γ by the drug. Polymerase γ is unique among the cellular replicative DNA polymerases as it is highly sensitive to inhibition by nucleoside analogues used in the treatment of human immunodeficiency virus and chronic hepatitis B and C infections.18, 19 Likewise, evidence of mitochondrial dysfunctions has also been reported with other drugs used to treat hepatitis B, such as lamivudine, telbivudine, and fialuridine5, 20–25 Thus, it is not surprising that clevudine could also produce mitochondrial toxicity.

Previously, an in vitro study showed that clevudine had no effect on the mtDNA content, suggesting that clevudine does not interfere with the mtDNA synthesis via DNA polymerase.26 However, our study presents evidence of mitochondrial toxicity: numerous ragged red fibers and COX-negative fibers in muscle pathology and depletion of mtDNA in Q-PCR analysis. Although mtDNA depletion due to the inhibition of DNA polymerase is a well-known mechanism of myopathy, alternative mechanisms of mitochondrial toxicity are also possible.27, 28 Further investigation and appropriate testing are needed to monitor any other adverse effects on mitochondrial integrity and the functions of nucleoside analogues.

On the other hand, the myopathy may have been mediated by immune mechanisms, perhaps triggered by the HBV infection itself and caused by sensitization to viral antigens expressed in muscle cells.29 However, muscle biopsies showed scant evidence of inflammation, and the muscle weakness and CK elevations developed only after long-term clevudine therapy. Furthermore, improvement by antiviral therapy has been reported in HBV-associated myopathy.29 These findings suggest that HBV infection itself is not the cause of the myopathy.

One of our patients (patient 5) was also taking colchicine for the treatment of gout. However, his muscle pathology was clearly different from that expected in colchicine-induced myopathy, which is characterized as vacuolar changes due to disruption of the microtubular cytoskeleton.30 The findings from pathology reduce the possibility of colchicine as the cause of myopathy.

It is also noteworthy that two of our patients (patients 2 and 3) showed laboratory findings suggestive of mild hypothyroidism. Although hypothyroidism also can cause a myopathy, the serum CK level in these patients was unusually high for hypothyroid myopathy, and the CK level did not normalize after thyroid hormone replacement; both strongly suggest that hypothyroidism was unlikely the cause of the high CK and muscle weakness in these patients.

The number of patients developing clinical myopathy during long-term clevudine therapy is currently unclear. Approximately 3,000 patients with HBV have been treated with clevudine for more than 24 weeks in Gyeongsang Province, and at least seven patients have developed myopathy so far. To determine the incidence of clevudine-induced myopathy, further studies in a large population and cohorts are necessary.

In conclusion, our study indicates that long-term therapy with clevudine can cause depletion of mtDNA and lead to a myopathy characterized by mitochondrial dysfunction and myonecrosis. Therefore, it is advised that careful clinical observation with respect to the muscle-related symptoms should be made and regular measurements of serum CK and lactate levels should be performed in all chronic hepatitis B patients who are taking clevudine for more than 32 weeks.