Treatment with tenofovir associated with more bone loss than abacavir, but long-term consequences unclear

Treatment with the antiretroviral drug tenofovir is associated with more bone loss than therapy with abacavir, European investigators report in the October 15^th edition of Clinical Infectious Diseases.

However, the authors of an editorial that accompanies the study cast doubt on the clinical significance of the study’s findings.

Moreover, both the study investigators and the editorial’s authors write that further research is needed into the causes and long-term consequences of bone loss in patients taking antiretroviral therapy.

Bone loss is an increasing concern for patients with HIV. Recently published European HIV treatment guidelines recommend monitoring of the bone metabolism of patients taking antiretroviral therapy.

The causes of bone loss are uncertain. It is known that HIV itself can causes loss of bone. But antiretroviral therapy, especially treatment with tenofovir, has been associated with decreased bone density.

All anti-HIV drugs can cause side-effects, and the main alternative to tenofovir therapy -  abacavir - has been associated with a risk of hypersensitivity reaction and a possible increase in the risk of heart attack.

Aware of the possible side-effects of both drugs, investigators from the European ASSERT study analysed 48 weeks of data to see if treatment with tenofovir was indeed associated with greater bone loss and greater bone turnover than treatment with abacavir.

The study involved 385 patients starting HIV therapy for the first time. They were randomised on an equal basis to take treatment that included tenofovir-FTC (Truvada) or abacavir-3TC (Kivexa). All the patients took efavirenz (Sustiva). This drug was selected because it was thought to cause no bone side-effects. However, the authors of the editorial note that subsequent research has suggested that therapy with this drug may indeed lead to bone loss.

Bone mineral density in the hip and lumbar spine was measured at baseline and then at regular intervals over 48 weeks. Several markers of bone turnover were also analysed using blood tests.

At baseline, there were no significant differences clinical or demographic differences between the patients in the two study arms. The median age was 37 years, 81% were male, 78% were white, 9% were co-infected with hepatitis C virus, 37% smoked, and 67% had a body mass index (BMI) below 25.

On entry to the study, the patients had a median CD4 cell count of 240 cells/mm³, and median viral load was 5 log₁₀ copies/ml.

Large numbers of patients withdrew before the end of the study (tenofovir-FTC = 28%; abacavir-3TC = 33%).

By week 48, reductions in bone mineral density in the hip were seen in both study arms. However, loss of bone was significantly greater for those taking tenofovir than those treated with abacavir (-3.6 vs. -1.7%, p < 0.001).

Loss of bone was at a constant rate through the 48 weeks of the study for those taking abacavir-3TC. However, in patients treated with tenofovir-FTC, bone mineral density declined sharply during the first six months of therapy, but then loss stabilised at a rate similar to that observed in the other study arm.

Lower baseline BMI, non-black race, and use of illicit drugs were all significantly associated with greater bone loss on the hip (all p < 0.03).

Patients taking tenofovir-FTC also had greater bone loss in the lumbar spine (-2.4% vs -1.6%, p = 0.036). In both arms of the study, there was a steep decline during the first 24 weeks of therapy, but this was then followed by some recovery.

Factors associated with an increased risk of loss of bone in the lumbar spine included lower baseline BMI, non-black race and older age (p < 0.03).

The investigators also noted that by the end of the study a higher proportion of those taking tenofovir-FTC than those taking abacavir-3TC had lost 6% or more of bone density at both the hip (13% vs. 3%) and lumbar spine (10% vs. 5%).

In addition, blood markers showed that those taking tenofovir had higher rates of bone turn-over than those treated with abacavir.

However, the authors of the accompanying editorial note that it is unclear if the statistically significant associations found between tenofovir therapy and bone loss translates into clinical significance. “The clinical value of the ASSERT data is uncertain”, they write. They therefore call for larger and longer trials to assess the relationship between HIV therapy and bone loss and its clinical significance.

In addition, they raise some questions about the study’s methodology, and note the high drop-out rate and gender imbalance of the study population.They also note that no information was provided about two important factors associated with bone loss: sex hormone and vitamin D levels. In addition, the relationship between bone density and fracture risk was not explored.