Tenofovir safe and effective for long-term hepatitis B treatment with little bone loss

Tenofovir (Viread) continues to be safe and effective for treating chronic hepatitis B through eight years of follow-up, researchers reported at The Liver Meeting 2012, the 63^rd Annual Meeting of the American Association for the Study of Liver Diseases (AASLD) last week in Boston. Another study showed minimal bone loss among tenofovir-treated patients using the FRAX method.

Several nucleoside/nucleotide analogues are active against hepatitis B virus (HBV), but with long-term use HBV develops resistance to older drugs such as lamivudine (Epivir-HBV), thereby compromising sustained effectiveness.

Tenofovir – which has demonstrated potent activity against both HBV and HIV – has a higher barrier to resistance. Gilead Sciences' Study 102 and Study 103showed that tenofovir maintains HBV viral load suppression, is well tolerated and shows no evidence of resistance through six years for previously untreated chronic hepatitis B patients.

Scott Fung, fromToronto General Hospital, and colleagues conducted a phase 3b study of tenofovir for treatment-experienced hepatitis B patients. Participants were receiving lamivudine at study entry but had HBV viral load of 1000 IU/ml or more and documented lamivudine resistance (e.g., M204I/V or L180M mutations); some had also used adefovir (Hepsera) for 48 weeks or less.

This analysis included 280 participants, about half of whom were hepatitis B 'e' antigen (HBeAg) positive. Approximately two-thirds were enrolled in Europe and one-third in North America. They had been infected with HBV for ten years on average and had been taking lamivudine for about four years.

About three-quarters of participants were men, about 60% were white, about 35% were Asian and the average age was 47 years. HBV genotype A was most common (22%), followed by B (14%), C (19%) and D (45%). At baseline, the mean HBV DNA level was about 6.5 log₁₀ copies/mL.

About 60% of participants had elevated alanine aminotransferase (ALT) at study entry. People with pre-existing kidney function impairment (creatinine clearance < 50 ml/min) were excluded, as were individuals with HIV or hepatitis C co-infection. At baseline about 30% of participants had evidence of low bone density (osteopenia or osteoporosis) on spine DEXA scans and about 20% had bone loss shown on hip scans.

Participants were randomly assigned to take tenofovir alone or co-formulated tenofovir/emtricitabine (Truvada, approved for HIV treatment). Safety and efficacy were evaluated over 96 weeks. About 90% of participants in both study arms completed all 96 weeks.

Assessments included kidney biomarkers (serum creatinine, creatinine clearance, serum phosphorus) because tenofovir can cause kidney toxicity in susceptible individuals, as well as DEXA bone density scans because the drug has been linked to bone loss in people with HIV.

An intent-to-treat analysis showed good – and statistically equivalent – efficacy in the tenofovir monotherapy and tenofovir/emtricitabine arms: 89 and 86%, respectively, had undetectable HBV DNA at 96 weeks (< 400 copies/ml or < 69 IU/ml). ALT normalisation, indicating reduced liver inflammation, occurred in 62 vs 63%, respectively.

Other measures of response to hepatitis B treatment were less impressive. HBeAg loss occurred in 15% of tenofovir monotherapy and 13% of tenofovir/emtricitabine recipients. HBeAg seroconversion occurred in 11 and 10%, respectively. Only one person in the combination arm experienced hepatitis B surface antigen (HBsAg) loss, considered the best indication of a cure for hepatitis B.

Amongst 18 participants who underwent drug resistance testing, no tenofovir resistance was detected in either treatment arm through 96 weeks.

Tenofovir monotherapy and tenofovir/emtricitabine were both generally safe and well tolerated. Only one participant experienced a serious adverse event, and three people (1%) discontinued treatment due to adverse events.

Looking specifically at kidney-related side-effects, no participants saw a confirmed serum creatinine increase of 0.5 mg/dL or more from baseline. Just 1% had reduced serum phosphorus below 2 mg/dL, whilst 3% had creatinine clearance below 50 ml/min (most of whom had sub-normal kidney function at baseline).

The researchers judged that no study participants developed clinically relevant bone loss according to spine and hip bone mineral density T-scores and Z-scores, which compare bone density to population norms. Although five people sustained trauma-associated bone fractures, none had non-traumatic fractures due to bone fragility in the absence of injury.

"A high rate of HBV DNA suppression with no detectable [tenofovir] resistance was achieved with [tenofovir] in patients with documented lamivudine resistance through 96 weeks," the researchers concluded. "[Tenofovir] was safe and well tolerated, with a low rate of renal events and no evidence of clinically relevant bone loss."

Given the similar safety and effectiveness of the tenofovir monotherapy and combination therapy arms, Dr Fung said that tenofovir alone should be considered adequate.

Bone Loss

In a related study, Upkar Gill from the Institute of Cell and Molecular Science in London and colleagues looked more extensively at bone density changes among chronic hepatitis B patients taking tenofovir.

The study was designed to show whether the World Health Organization's FRAX score is consistent with DEXA scans and bone-related biomarkers such as serum alkaline phosphatase and calcium levels for determining bone loss in this population. The WHO algorithm is less expensive than the other methods and can be widely used in resource-limited settings.

FRAX is a web-based tool that calculates ten-year fracture risk using several variables including patient age, sex, body mass index, individual and family history of fractures, smoking and use of alcohol or steroids. The FRAX score may help determine which individuals could benefit from DEXA scans, ongoing monitoring or lifestyle modification to reduce their risk, the researchers noted.

The analysis included 122 chronic hepatitis B patients who had used tenofovir for at least 48 weeks, as well as 48 individuals not exposed to the drug. About 70% were men, two-thirds were Asian and the rest were about evenly split between black and white participants. The median age was 45 years in the tenofovir group and 36 in the control group.

The tenofovir group had similar proportions of people with mild, moderate and severe liver fibrosis, but in the control group nearly two-thirds had mild liver damage. About 10% of participants were smokers and about 16% said they drank alcohol.

The investigators found that tenofovir recipients had a significantly lower average hip T-score than unexposed study participants. T-scores for the lumbar spine and femoral neck (hip joint), however, did not differ significantly.

Small changes in bone density were seen soon after starting tenofovir, but levels soon reached a plateau with no further bone loss progression. There was no correlation between duration of tenofovir use and bone loss at any site. A similar pattern has been seen in studies of bone loss among people with HIV taking tenofovir.

There was no overall correlation between bone biochemistry markers and changes in bone mineral density. Tenofovir recipients had a tendency towards increased serum alkaline phosphatase, but the difference did not reach statistical significance.

In a univariate analysis, older age, body mass index, smoking and tenofovir use were all significant risk factors for bone loss. However, in a multivariate analysis controlling for other factors, the effect of tenofovir was no longer significant.

Overall, 75% of tenofovir recipients were determined to have a low risk of fractures according to pre-DEXA FRAX scores, and none required treatment for bone loss. FRAX predicted major hip fractures as well as DEXA scans and did not miss any high-risk patients. Based on these findings, the researchers concluded that FRAX has 100% sensitivity, 84% specificity, a negative predictive value of 100% and a positive predictive value of 40% for projecting bone breaks in this population.

"We demonstrate the utility of FRAX as an alternative tool to assess fracture risk over time", they concluded. "We recommend the use of FRAX to consolidate treatment decisions in chronic hepatitis B", reducing the need for costly DEXA scans.

Using this method, Gill explained, people with low risk for fractures can be reassured, medium-risk patients can be monitored and high-risk individuals can receive interventions to strengthen their bones and reduce their chances of breaks.