Concern about the long-term safety of oral anti-HBV drug therapy Yimin Mao, Department of Gastroenterology, Shanghai Institute of Digestive Diseases, Renji Hospital, School of Medicine, Shanghai Jiaotong University, China Yimin Mao, Department of Gastroenterology, Shanghai Renji Hospital, China A large body of evidence-based medical evidence has shown that oral nucleoside (acid) analogues are effective in suppressing viral replication in patients with chronic viral hepatitis B (CHB), but improvements in alternative endpoints such as histology, virology, serology and biochemistry The impact on long-term clinical outcomes of the disease has yet to be further confirmed1 and therefore, the appropriate course of treatment for CHB with the available nucleoside (acid) analogues is not known with certainty. Long-term therapy may be the overall strategy to achieve long-term clinical benefit in terms of delaying and halting liver disease progression, reducing the incidence of cirrhosis/HCC, improving quality of life, and prolonging survival.2 However, the ensuing questions of safety, resistance profile, risk/effects, and pharmacoeconomics of long-term therapy are serious issues that we should and must face. In this paper, we discuss the possible safety issues. I. Mitochondrial toxicity Mitochondrial toxicity is a common safety concern for orally administered nucleoside (acid) analogs. The main function of mitochondria is to oxidize fatty acids and pyruvate into adenosine triphosphate (ATP), and once its function is impaired, it leads to impaired cellular energy production, which in turn causes cellular damage. The mitochondrial toxicity of nucleoside (acid) analogs is related to their pharmacological mechanism, which inhibits HBVDNA polymerase activity with varying degrees of inhibition of human mitochondrial DNA (mtDNA) polymerase γ, which depletes intracellular mtDNA.3, 4 The classic case is the now-defunct drug Fialuridine (FIAU), which causes lactic acidosis when used for more than 8 to 10 weeks. FIAU irreversibly binds to human mtDNA, leading to mitochondrial failure.3 Mitochondrial toxicity has been found to varying degrees in seven in vitro and in vivo studies of antiretroviral agents approved for the treatment of HIV infection.4 Five nucleoside analogs approved for the treatment of CHB were found to inhibit mitochondrial DNA polymerase gamma to a lesser extent than antiretrovirals in in vitro studies, and notably entecavir showed little mitochondrial toxicity in in vitro studies.5 Mitochondrial toxicity can produce systemic systemic lesions, and clinical manifestations can include myopathy, neuropathy, fatty liver, pancreatitis, and megaloblasts. pancreatitis, macrocytosis, hyperlactatemia, lactic acidosis, and nephrotoxicity. Therefore, all currently marketed nucleoside analogs have “black box” warnings in their instructions and product labels to alert people to the potential mitochondrial toxicity of these drugs. In addition to the direct toxic effects of the drug, other host factors such as age, gender, genetic background, comorbidities, nutritional status, and individual differences may also influence the clinical manifestations and degree of mitochondrial toxicity. At present, few adverse reactions related to mitochondrial toxicity have been reported in clinical trials of nucleoside analogs for the treatment of chronic hepatitis B. However, in reports of monitoring of the expanded population application of drugs after marketing, some adverse reactions related to mitochondrial toxicity occurred, although low, but were serious, such as myopathy, neuropathy, pancreatitis, and reversible renal damage, which should be given sufficient clinical attention and closely monitored. Combination therapy may be one of the treatment strategies for chronic hepatitis B. However, it is not known whether different drug combinations produce synergistic mitochondrial toxicity, and basic and clinical studies should strengthen the attention to this aspect. In animal and human studies, both adefovir and tenofovir have been found to cause dose-dependent nephrotoxicity, the mechanism of which is not fully understood, but which may involve alterations in renal tubular transport proteins, apoptosis, and mitochondrial toxicity. Nephrotoxicity is mainly tubular damage and is usually reversible after cessation of treatment. The clinical manifestation of most nephrotoxicity is a slight increase in blood creatinine and a decrease in blood phosphorus after 4 to 12 months of dosing. Patients with proximal tubular dysfunction may develop renal tubular acidosis, characterized by metabolic acidosis, low blood phosphorus and glycosuria. Other rare types of nephrotoxicity are nephrogenic uremia and acute renal failure, with a higher probability of occurrence in patients with pre-existing renal insufficiency or on other nephrotoxic drugs.3 Adefovir nephrotoxicity was first identified. High doses of 60-120 mg daily of adefovir for the treatment of HIV infection were associated with varying degrees of nephrotoxicity in 22%-50% of patients; mild nephrotoxicity was seen in approximately one-third of patients treated with CHB beyond 6 months at a dose of 30 mg/day.6 Therefore, the currently approved dose for the treatment of CHB is 10 mg daily. Tenofovir is similar in molecular structure to adefovir, but compared to adefovir Adefovir, the frequency of nephrotoxicity is lower compared to Adefovir. In randomized controlled clinical studies of CHB, tenofovir nephrotoxicity was even barely observed. However, instances of significant nephrotoxicity have been found in studies of long-term treatment of HIV-infected patients.7 The incidence of nephrotoxicity, particularly subclinical renal insufficiency, has not been adequately evaluated with long-term tenofovir treatment. To minimize the risk of nephrotoxicity, it is recommended that HBV-infected patients with renal insufficiency first undergo dose or dosing interval adjustment (see Table 1); second, serum creatinine, blood phosphorus, and urine samples should be monitored regularly (every 2-3 months if mildly impaired, or every month if moderately or severely impaired), and especially HBV-infected patients with renal insufficiency who have undergone dose adjustment should be closely monitored infected patients who have undergone dose adjustment. Third, avoid other nephrotoxic drugs if possible, and maintain proper hydration. Finally, serum creatinine and phosphate levels should be monitored regularly in patients on long-term use of adefovir or tenofovir. Dose or dosing interval adjustments are required if creatinine levels exceed the baseline value by more than 0.5 mg/dL or if blood phosphorus levels are below 2.0 mg/dL.3 Table 1: Recommended dose or dosing interval for patients with renal insufficiency HBV infection Lamivudine Adefovir Entecavir Tepivudine Tenofovir Dose (mg/day) GFR > 50 mL/min 100 mg/day† 10 mg /day 0.5 mg/day*† 600 mg/day 300 mg/day 30-49 mL/min 50 mg/day 10 mg/day 0.25 mg/day or 0.50 mg/two days 600 mg/two days q 48 hours 10-29 mL/min 15-25 mg/day 10 mg/two days 0.15 mg/day or 0.50 mg/three days 600 mg/three days q 72-96 hrs Dialysis 10 mg/day 10 mg/week 0.05 mg/day or 0.50 mg/week 600 mg/week Weekly III. Pregnancy and Reproductive Toxicity Clinical studies to reduce vertical transmission in pregnant women with chronic HBV infection are very limited. Retrospective studies have shown that lamivudine treatment reduces the risk of vertical transmission in pregnant women in late pregnancy.8 Another randomized controlled trial showed that the probability of HBV transmission at one year was 18% and 39% in HBV high-carrier mothers, 56 treated with lamivudine versus 52 untreated, respectively.3 However, it is difficult to exclude the effect of HBIg and vaccination on trial outcomes in both trials. Therefore, because of the potential uncertain risk to the fetus associated with the possible uncertain benefit of these drugs, and considering the general clinical benefit of standard postpartum immunoprophylaxis, antiviral drugs are not currently recommended for pregnant women in late pregnancy. All approved anti-HBV drugs have potential reproductive toxicity, with telbivudine and tenofovir in category B (no known teratogenicity or embryotoxicity in animal studies, but no human data available) and the other three in category C (teratogenic or embryotoxic in animal studies, no human data available). Therefore, there are prominent warnings in the instructions and product labels to alert people to the potential adverse fetal risks associated with the use of these drugs before and during pregnancy. US data show that in the pregnant HIV population, the incidence of birth defects in infants of pregnant women taking lamivudine and tenofovir in the first trimester was 3.1% (85/2784) and 2.2% (11/491), respectively, an incidence similar to that of the general US population.9 However, data are lacking in pregnant women with chronic HBV infection. Similarly, there are insufficient data on the risk of congenital defects in the fetus due to entecavir, adefovir and telbivudine. Therefore, it is recommended that this class of drugs should be avoided during the first trimester of pregnancy, and if they must be used they should be used as much as possible in class B.3 Despite the lack of evidence, most experts recommend that mothers who continue to use oral nucleoside analogues avoid breastfeeding for the first year. IV. Safety of Approved Drugs (i) Lamivudine Lamivudine is the first approved and by far the most widely used oral nucleoside analogue for the treatment of CHB. Early controlled clinical trials over a one-year period showed that lamivudine was well tolerated with similar frequency, type and severity of adverse events as placebo. However, resistance to long-term use was identified in subsequent clinical studies and post-marketing surveillance, with varying degrees of HBV breakthrough due to lamivudine resistance after more than 4 to 5 years of long-term use becoming the main adverse event reported.10 The probability of resistance at 5 years of treatment has now been confirmed to be 70%. Lamivudine should be used with caution in primary care patients, and although initial treatment may be well tolerated, the risk of resistance with long-term use is a clinical selection concern. Optimal treatment regimens have been introduced in China to mitigate this risk. Although, reversible myopathy, pancreatitis, and other manifestations of mitochondrial damage are rarely seen in co-infected patients with HBV and HIV treated with lamivudine, muscle wasting and asymptomatic macrocytosis have been reported in HIV-infected patients receiving daily cocktail therapy containing high doses of lamivudine.11 The role of lamivudine in triggering these adverse effects is not known, but should be of clinical attention. (ii) Adefovir Adefovir was approved for the treatment of CHB in 2002. Early dose exploratory clinical trials in development showed that the incidence of nephrotoxicity was positively correlated with dose, making the 10 mg/day dose the appropriate dose for its approval. In a 1-year registration-controlled clinical trial, the 10 mg/day adefovir arm demonstrated comparable rates of Grade I nephrotoxicity (i.e., a rise in blood creatinine above 0.5 mg/dL above baseline) compared to the placebo arm, demonstrating the renal safety of the 10 mg/day dose for 1 year of treatment. Adefovir nephrotoxicity mostly manifested as a mild reversible increase in blood creatinine levels unrelated to hypophosphatemia, which may improve with extended dosing intervals or continued dosing. There is insufficient evidence that prolonged treatment at the 10 mg/day dose increases the incidence of nephrotoxicity. However, after all, reversible increases in blood creatinine above baseline values of 0.5 mg/dL have been reported in 3% and 8% of patients in e antigen-negative and positive adefovir clinical studies, respectively, and a small number of patients may develop hypophosphatemia; therefore, the potential nephrotoxicity of adefovir should be closely monitored when applied clinically at regular doses. Because adefovir effectively inhibits lamivudine-resistant strains, studies of combination adefovir therapy in lamivudine-resistant patients are increasing, and the nephrotoxicity of adefovir in combination therapy has received widespread attention.3 In a study including 145 lamivudine-resistant patients, mild nephrotoxicity was present in 8% of patients, but when the investigators extended the adefovir dosing interval, all patients were able to continue the combination regimen for treatment. In another study comparing lamivudine with a combination of lamivudine and adefovir in a randomized controlled trial, no nephrotoxicity was observed in 115 patients with primary HBeAg-positive chronic hepatitis B on combination therapy. The overall resistance rate seen with long-term treatment with adefovir is clinically acceptable relative to other marketed nucleoside analogues, with a 5-year probability of occurrence of adefovir resistance of roughly 28%.12 Once resistance occurs, similarly, varying degrees of HBV breakthrough can occur, which is a potential adverse effect of its long-term treatment. (iii) Entecavir Entecavir was approved for hepatitis B treatment in 2005. The dose was 0.5 mg/day for primary patients and 1.0 mg/day for lamivudine-resistant patients. In a 1-year registered controlled clinical trial, the frequency and severity of adverse events with entecavir in the clinic and laboratory were similar to those of the control lamivudine. The risk of mitochondrial toxicity from long-term treatment with entecavir is relatively small compared to other nucleoside analogs. Preclinical studies have shown that no mitochondrial toxicity was observed in cell cultures of entecavir in combination with lamivudine, adefovir or tenofovir, and no increased production of mitochondrial toxicity of other nucleoside analogues was observed in combination intervention studies. Furthermore, no corresponding manifestations of mitochondrial toxicity or other adverse events were observed in clinical studies over a period of up to 5 years. Due to the trial design, the sample size of patients after 5 years was significantly reduced from 663 in the first year to 108 in the fifth year, and the reported incidence of long-term treatment resistance of 1.2% in primary care patients at 5 years is non-ITT data and, therefore, somewhat unconvincing, but entecavir has the lowest reported incidence of resistance among the currently available nucleoside analogs and is a good choice for primary care patients. In lamivudine-resistant patients, entecavir is of limited value, with resistance occurring in 37% of patients treated for 4 years.13 Because it was found in preclinical development that animals were more likely to develop solid tumors with long-term use of high-dose entecavir compared with placebo, clinical studies with large samples were conducted globally after its launch to further observe the effects in humans. There is no direct evidence that entecavir increases the incidence of malignancy in humans.3 (iv) Tebivudine Tebivudine is the fourth oral nucleoside analogue approved for the treatment of chronic hepatitis B at a dose of 600 mg/day. In a 2-year multinational participatory international multicenter clinical trial, a higher proportion of significant elevations in serum creatine phosphokinase (CPK) with grade 3 to 4 toxicity were seen with telbivudine than with the control lamivudine (12.9% versus 4.1%), while other adverse events were similar.14 Both patients in the study who developed symptomatic myopathy as a result of telbivudine treatment eventually discontinued the drug before the myopathy resolved. Therefore, regular monitoring of CPK and corresponding musculoskeletal symptoms before and during treatment with telbivudine is essential. To date, there have been no published reports of lactic acidosis associated with tibivudine treatment alone. However, there are preliminary reports of moderate peripheral neuropathy in 17% of patients treated with the combination of telbivudine and pegylated interferon alpha-2a.3 The combination of telbivudine with lamivudine has not demonstrated an advantage over treatment alone. The safety and efficacy of the combination of tenofovir with adefovir and tenofovir are also currently being evaluated. (v) Tenofovir Tenofovir is an oral nucleoside analogue approved last year for the treatment of CHB at a dose of 300 mg/day. In randomized controlled clinical trials, tenofovir was associated with similar adverse events as the control adefovir. Although no severe or symptomatic nephrotoxicity has been identified in patients with chronic hepatitis B, nephrotoxicity was found in 4% of patients in clinical studies for HIV infection, and individual patients developed nephrogenic uremia and acute renal failure.3 Therefore, concerns about its nephrotoxicity in the long-term treatment of chronic hepatitis B are not superfluous, and regular monitoring of serum creatinine and blood phosphorus may provide early detection of potential nephrotoxicity. Based on current studies, tenofovir-induced nephrotoxicity is mostly reversible after early discontinuation of the drug. Recent studies have found that HIV-infected patients on long-term tenofovir therapy can develop decreased bone mineral density and bone tenderness.15 These signs should also be taken into clinical consideration, and it is recommended that long-term treatment recipients receive regular bone densitometry and that calcium and vitamin D supplementation may help prevent or treat decreased bone mineral density and bone tenderness. Similarly, the pros and cons of bone metabolism in pediatric patients are debatable and the risks/effects need to be carefully weighed. Tenofovir has shown efficacy against lamivudine-resistant strains in both in vivo and in vitro studies, but the phenotype and genotype of tenofovir resistance have not been clearly defined.3 V. Research directions Most randomized controlled clinical trials, especially registration trials, have small sample sizes and short duration, making it difficult to observe a comprehensive and sufficient number of events that fully reflect the truth about the study drug, and therefore, by randomized controlled clinical trials only Some rare and important adverse reactions and certain consequences can only be identified in longer-term, larger sample size exposures.1 Moreover, the representation of study populations in clinical trials is often incomplete, and their efficacy and safety results cannot be extrapolated to all clinical patients. The understanding of the safety of long-term oral nucleoside analogs through existing clinical trials is clearly still only the tip of the iceberg, and therefore the response is even less well founded, and a large number of clinical studies are still needed to answer our concerns. Studies should focus on the occurrence and regression of known common safety concerns of mitochondrial toxicity/nephrotoxicity, the impact of different dose adjustment strategies on clinical outcomes of mitochondrial toxicity/nephrotoxicity, the risk/effects/pharmacoeconomics of combination therapy, the occurrence of drug resistance and coping strategies for single or combination long-term therapy, the observation of safety of specific drugs (e.g., tumorigenesis of entecavir, bone mineral density and bone metabolic parameters of tenofovir and bone metabolism parameters, etc.), efficacy and safety in special populations (children, elderly, cirrhotic decompensation, renal insufficiency, pregnant women, oncology patients, etc.) and risk/effects of various antiviral drugs in pregnant women infected with HBV, etc. It is crucial to conduct large-scale, long-term prospective clinical studies and strengthen monitoring after marketing, which can provide us with information on the safety of general and special populations with chronic HBV infection and provide evidence for scientific assessment of their risks/effects and rational clinical application. References: 1. Mao Yimin, Zeng Minde. A call for better clinical trials in hepatology. Liver 2009, Vol. 14, No. 2: 91-92 2. Lok ASF, McMahon BJ. AASLD practice guideline: chronic hepatitis B. HEPATOLOGY 2007;45:507-539. 3. Robert Side Effects of Long-Term Oral Antiviral Therapy for Hepatitis B. HEPATOLOGY 2009, Vol. 49, No. 5, Suppl.,S185-S195 4. Martin JL, Brown CE, Matthews-Davis N, Reardon JE. Effects of antiviral nucleoside on human DNA polymerases and mitochondrial DNA synthesis. Antimicrob Agents Chemother. 1994;38:2743-2749. 5. Mazzucco CE, Hamatake RK, Colonno RJ, Tenney DJ. Entecavir for treatment of hepatitis B virus displays no in vitro mitochondrial Antimicrob Agents Chemother 2008;52:598-605. 6. Izzedine H, Hulot JS, Launay-Vacher V, Marcellini P, Hadziyannis SJ, Currie G, et al. Adefovir Dipivoxil International 437 Study Group, renal safety of adefovir dipivoxil in patients with chronic hepatitis B: two doubleblind, randomized, placebo controlled studies. Kidney Int 2004;66:1153-1158. 7. Izzedine H, Isnard-bagnis C, Hulot JS, Vittecoq D, Cheng A, Jais CK, et al. Renal safety of tenofovir in HIV treatment-experienced patients. AIDS 2004;18:1074-1076. 8, Xu WM, Cui YT, Wang L, Yang H, Liang ZQ, Li XM, et al. Lamivudine in late pregnancy to prevent perinatal transmission of hepatitis B virus infection: a multicentre, randomized, double-blind, placebo-controlled study. J Viral Hepat 2009;16:94-103. 9. Dybul M, Fauci AS, Bartlett JG, Kaplan JE, Pau AK. Guidelines for using Recommendations on the Panel on Clinical Practices for the Treatment of HIV. MMWR Recomm Rep 2002;51:1-55. 10. Lok AS, Lai CL, Leung N, Yao GB, Cui ZY, Schiff ER, et al. Long-term safety of lamivudine treatment in patients with chronic hepatitis B. Gastroenterology 2003;125:1714-1722. 11, Khawcharoenporn T, Shikuma CM, Williams AE, Chow DC. Lamivudine associated macrocytosis in Int J STD AIDS 2007;18:39-40. 12, Hadziyannis SJ, Tassopoulos NC, Heathcote EJ, Chang TT, Kitis G, Rizzetto M, et al. Long-term therapy with adefovir dipivoxil for HBeAg negative chronic hepatitis B for up to 5 years. Gastroenterology 2006;131:1743-1751. 13, Han SH, Chang TT, Chao YC, Yoon SK, Gish RG, Cheinquer H, et al. Five years of continuous entecavir for nucleoside-naı¨ve HBeAg (_) chronic hepatitis B: results from study ETV 901 [Abstract]. HEPATOLOGY2008;48(4 Suppl 1):705A. 14, Liaw YF, Gane E, Leung N, Zeuzem S, Wang Y, Lai CL et al. 2-year GLOBE Trial results: telbivudine is superior to lamivudine in patients with chronic hepatitis B. Gastroenterology 2009;136:486-495. 15, Cassetti LJ, Madruga JV, Suleiman JM, Etzel A, Zhong L, Cheng AK, et al. The safety and efficacy of tenofovir DF in combination with lamivudine and efavirenz through 6 years in antiretroviral-naı¨ve, HIV-1 infected HIV Clin Trials 2007;8:164-172.