Lamivudine has limited impact on HBV for hepatitis B / HIV-coinfected in South African trial

Keith Alcorn
05 July 2011

The inclusion of lamivudine (3TC) in antiretroviral therapy had few additional benefits for South African HIV-positive patients co-infected with hepatitis B, South African investigators report in the online edition of AIDS.

Hepatitis B virus (HBV) outcomes were not significantly better in patients treated with lamivudine than in patients taking an alternative antiretroviral drug.

“Overall the efficacy of 3TC on markers of HBV disease in this HIV/HBV co-infected population was poor,” comment the investigators. They believe their results support the World Health Organization (WHO) recommendation that co-infected patients should receive antiretroviral therapy that includes tenofovir (Viread).



A protein made in the liver, needed to maintain a balance of the fluids in the body. In a blood test, lower than normal levels of albumin and total protein may indicate liver damage or disease. If there is not enough albumin, fluid may accumulate in the abdomen (ascites).

Rates of HIV and hepatitis B co-infection are high in many resource-limited settings, and liver disease caused by hepatitis B is a major cause of serious illness and death in co-infected individuals.

A number of anti-HIV drugs are also active against hepatitis B. These include 3TC, FTC (emtricitabine), and tenofovir. In resource-rich countries, antiretroviral therapy for co-infected patients is recommended to include tenofovir and 3TC or FTC in order to provide two antiretroviral drugs that are also active against hepatitis B.

In poorer countries, however, 3TC is the only one of these drugs which is widely available.

Investigators from the South African military wanted to see if the inclusion of 3TC, rather than alternative anti-HIV drugs, in first-line antiretroviral regimens was associated with improved HIV and hepatitis B treatment outcomes in co-infected patients.

They therefore retrospectively analysed results from the PHIDISA II study. Its participants included 1771 patients, 6% of whom had chronic hepatitis B infection.

The patients were starting HIV therapy because of a fall in their CD4 cell count to below 200 cells/mm3 or because they had developed an AIDS-defining illness.

HIV therapy was based on NRTI pairs of either 3TC combined with d4T (stavudine) or ddI (didadosine) with AZT (zidovudine).

The investigators compared four outcomes in the co-infected patients according to the use of 3TC or non-3TC regimens. These were:

·        Changes in HIV viral load and CD4 cell count.

·        Rates of hepatitis B flares.

·        Hepatitis B suppression.

·        Mortality.

Compared to patients with HIV-mono-infection, those with co-infection were significantly more likely to be male (86% vs. 64%, p < 0.0001).

At baseline, a number of laboratory markers associated with poorer long-term outcomes were significantly more unfavourable in co-infected patients than in individuals only infected with HIV. These included higher liver enzyme levels (ALTs) (p < 0.0001), lower albumin (p < 0.0001), and lower platelet counts (p = 0.0002).

After starting HIV therapy, changes in CD4 cell count and HIV viral load were similar between mono- and co-infected patients. Moreover, for co-infected patients, these HIV-related outcomes were comparable regardless of the use of 3TC.

Overall, only 2% in patients experienced a hepatitis flare after initiating therapy. But rates were much higher in the co-infected patients than among patients only infected with HIV (9% vs. 0.02%, p < 0.0001).

When analysis was restricted to the co-infected patients, rates of hepatitis flares were comparable between the 3TC-treated patients and individuals who were not taking this drug.

After a year of therapy, 33% of co-infected patients who were taking 3TC had an undetectable hepatitis B viral load (below 55 iu/m). This compared to 13% of co-infected patients who were taking an alternative anti-HIV drug – a non-significant difference.

Mortality rates were significantly higher among co-infected patients than mono-infected individuals (17% vs. 11%, p = 0.04). The investigators urge “urgent” action to improve the management and care of co-infected patients in resource-limited settings.

However, the risk of mortality was comparable for co-infected patients regardless of whether they were treated with 3TC. Baseline albumin (p = 0.001) and platelet count (p = 0.001) were the only significant risk factors for mortality.

“There is little benefit from the use of 3TC,” write the authors.

They conclude that their findings support the WHO recommendation for tenofovir therapy for all HIV/hepatitis B-co-infected patients.

Matthews GV et al. Impact of 3TC on HIV and HBV related outcomes in HIV/HBV individuals in a randomized controlled trial of antiretroviral therapy in South Africa. AIDS 25, online edition, doi: 10:1097/QAD.0b013e328349bbf5, 2011.