10 February 2016 

Does tenofovir (TDF) cause liver injury?

Tenofovir is an important antiviral drug used to treat HIV and hepatitis B virus and is made in two formulations:

  • Tenofovir DF (tenofovir disoproxil fumarate; TDF)
  • TAF (tenofovir alafenamide)

The original formulation of tenofovir is TDF (Viread). This drug is in the following fixed-dose combinations:

  • Truvada (tenofovir + FTC)
  • Atripla (tenofovir + FTC + efavirenz)
  • Complera (tenofovir + FTC + rilpivirine)
  • Stribild (tenofovir + FTC + cobicistat + elvitegravir)

The new formulation of tenofovir, TAF, will gradually become available, initially in several of the fixed-dose formulations mentioned above. The first TAF-containing medicine approved in Canada (and other high-income countries) is Genvoya. This medicine is similar to Stribild except instead of having TDF, Genvoya contains TAF.

Advantages of TAF

TDF is not as well absorbed as TAF, so a larger dose must be taken. As a result, a relatively large concentration of tenofovir accumulates in the blood. Some researchers think that these high blood levels of tenofovir may be responsible for reports of tissue injury (in the bones and kidneys) of some TDF users.

In contrast, when TAF is taken, it is quickly absorbed and concentrates inside cells of the immune system. This is useful in two ways:

  • TAF can protect the cells that HIV targets
  • there is a reduced risk of tenofovir harming bones and kidneys

A signal of possible toxicity

Researchers with a large database called DAD have been collecting health-related information on HIV-positive people from clinics in Australia, Europe and the U.S. From time to time this data is analysed as researchers seek any links between the use of anti-HIV drugs and possible side effects.

The DAD dataset has accumulated information on about 49,000 HIV-positive people. Beginning in 2004, DAD researchers began regularly seeking information about serious liver injury. In the latest analysis from DAD, researchers found a statistically increased risk for serious liver injury and/or liver cancer among a very small proportion of people (less than 1%) who had been exposed to TDF as well as some older anti-HIV drugs. Due to the study’s built-in limitations—it cannot prove that TDF was the cause of liver injury and/or liver cancer—its findings must be treated cautiously. At a minimum, DAD’s findings need to be confirmed in other large datasets and scientific investigation is needed to understand how TDF might injure the liver. In the meantime it is important to bear in mind that the DAD researchers did not advise doctors to cease prescribing TDF.

Study details

The DAD researchers specifically sought information on what doctors called end-stage liver disease (ESLD). In cases of ESLD, the liver has become so severely injured that it can no longer function properly. Such a severe degree of liver injury can arise from years of untreated viral hepatitis (caused by hepatitis B or C viruses) or chronic exposure to harmful substances such as excessive levels of alcohol.


For the present study, DAD researchers focused on 45,544 participants for whom relevant data were available for analysis about liver issues. They found that a total of 209 cases of ESLD and 110 cases of liver cancer occurred between the years 2004 and 2014, for a total of 319 events. These participants were in DAD for an average of eight years before these events occurred.

The most common symptoms related to ESLD were those that affected the brain (43%), including difficulty thinking clearly, problems with memory, and excessive sleeping. Less common symptoms were internal bleeding (27%) and kidney dysfunction (15%).

The average profile of DAD participants with ESLD and liver cancer was as follows:

  • 47 years
  • 80% men, 20% women
  • CD4+ count – 266 cells/mm3
  • 83% had hepatitis B or C virus or both co-infections

The seriousness of ESLD or liver cancer

Once participants were diagnosed with ESLD or liver cancer, DAD was only able to capture further data on them for about three months. It is likely that this short duration was related to the relatively short lifespan of participants after such a serious diagnosis was made. For instance, after their diagnosis, half of the 319 participants were dead within about seven months.

Links to treatment

Statistical analysis found that the use of the following drugs was linked to an increased risk for ESLD or liver cancer:

  • d4T (stavudine, Zerit)
  • ddI (didanosine, Videx)
  • TDF

Other nukes, such as the following, were not linked to an increased risk for ESLD or liver cancer:

  • 3TC (lamivudine) and in Kivexa, Triumeq, Combivir and Trizivir
  • FTC (emtricitabine, Emtriva and in several co-formulated medicines mentioned at the beginning of this report)

All of the above drugs belong to a class of anti-HIV medicines commonly called nukes (nucleoside or nucleotide analogues).

The other anti-HIV drug that was associated with an increased risk for ESLD or liver cancer was the protease inhibitor amprenavir (Agenerase) and its subsequent formulation fosamprenavir (Telzir, Lexiva).

Overall, the proportion of DAD participants who developed ESLD or liver cancer was low—319 out of 45,544 people. This is less than 1% of all participants in DAD. Thus, ESLD or liver cancer linked to the use of d-drugs (d4T, ddI) and/or TDF was very rare in this study.

Advice from DAD

The DAD researchers recommend that the use of d-drugs “should be avoided.” This is consistent with the advice from leading HIV treatment guidelines in high-income countries, which discourage the use of d-drugs because they have been linked to painful injury to nerves (peripheral neuropathy) and, in the case of d4T, changes in body shape.

The DAD team also suggests that doctors monitoring patients who have previously used d-drugs “might consider” intensifying their monitoring of the liver with blood tests. If such test results are confirmed to be abnormal, the DAD researchers say that additional procedures can be used to assess the health of the liver. Such procedures could include specialized ultrasound scans of the liver (Fibroscan) or liver biopsy. These may be helpful in identifying which patients are at increased risk for ESLD or liver cancer.

Focus on TDF

According to the DAD researchers, the association that they found between the use of TDF and ESLD or liver cancer was unexpected. This association was found even in participants who did not have viral hepatitis infection.

As this association with TDF is new, researchers do not know of any biological mechanism that could possibly explain this drug’s connection to serious liver injury. However, they suspect that TDF might injure the energy-producing parts (mitochondria) of liver cells.

A separate analysis of data (in press) from 21,000 HIV-positive people by DAD has also found that a persistent increase in liver enzymes in the blood is associated with use of tenofovir. None of these participants were co-infected with hepatitis-causing viruses. This finding suggests the possibility that tenofovir is associated with liver inflammation in some people who use it.

Also, a placebo-controlled trial of pre-exposure prophylaxis (PrEP) using Truvada (TDF + FTC) in healthy HIV-negative women has found an increased risk of moderately elevated liver enzyme levels in the blood among some TDF users. However, in this study TDF was not used for very long periods and adherence was not as good as in other studies.

All of these data suggest that further investigation of TDF’s potential to cause liver injury is a good idea.

Focus on protease inhibitors

Fosamprenavir (the subsequent formulation that replaced the initial formulation of amprenavir) is almost never prescribed in high-income countries today. Instead, other protease inhibitors—such as darunavir (Prezista and in Prezcobix) + ritonavir (Norvir), atazanavir (Reyataz) + ritonavir and, in some cases, lopinavir + ritonavir (in Kaletra)—are used. DAD found that, in general, these other protease inhibitors were not associated with any increased risk of ESLD or liver cancer. However, the research team stated that there was insufficient information on darunavir in its dataset.

Key points

Though serious, ESLD and liver cancer were very rare events in DAD, affecting less than 1% of participants who used nukes or other drugs that could have increased their risk for these complications.

Hepatitis-causing viruses (hepatitis B and C) also play a role in inciting and inflaming liver injury. It is likely that in the future, as potent all-oral anti-HCV treatments become more widely available in Canada and other high-income countries, more people will be cured of hepatitis C (HCV) and there will be fewer cases of HCV-associated complications such as ESLD and liver cancer. However, note that nearly 20% of participants who developed these complications did not have HCV or HBV co-infection.

So-called d-drugs are mentioned in this study. The use of these drugs is discouraged in treatment guidelines in high-income countries because they are associated with painful injury to nerves (peripheral neuropathy) and, in the case of d4T, changes in body shape (the HIV lipodystrophy syndrome).

The DAD researchers found that the risk of developing ESLD or liver cancer did not begin to decrease for participants until they had stopped taking d-drugs for six years. They stated that this suggests that exposure to d4T and ddI may have caused “irreversible tissue damage.” This finding of persistent injury or delayed recovery is consistent with another much smaller study that implicated one of the d-drugs (ddI) in liver injury. Insufficient numbers of people quit taking TDF (it is generally well tolerated), so little information is available from DAD on recovery from TDF-associated liver injury.

Caution needed

  1. DAD is observational in design. Such studies, due to built-in limitations, can never prove cause and effect; that is, DAD can never prove that TDF (or any other drug) causes serious liver injury. Such a link can only be proven by a very large study of a more robust statistical design. However, before embarking on such an expensive and lengthy venture, other researchers who operate databases that also collect data from HIV-positive people should consider performing similar analyses to confirm or refute DAD’s findings.
  2. DAD did not collect extensive information on alcohol consumption. Serious liver injury can occur due to excessive alcohol intake and this may be one possible factor that could have inadvertently biased researchers’ interpretation of the data. However, a previous study in the Netherlands investigated data from over 18,000 HIV-positive people for liver injury. These researchers assessed exposure to anti-HIV drugs and had data on alcohol intake. The Dutch study was able to rule out excess alcohol intake (by excluding such people from its analysis) and found that exposure to d-drugs was significantly linked to an increased risk of liver injury. In that study, the Dutch researchers focused on one aspect of liver injury: elevated blood pressure within the blood vessels of the liver (portal hypertension). Like DAD, the Dutch researchers found that cases of severe d-drug-related liver injury were very rare—occurring in less than 1% of people who used d-drugs. In some cases Dutch researchers also found that liver injury also occurred in people who used a d-drug in combination with TDF. However, a major issue with the Dutch study is that it was observational in design.

The research that is required

TDF is generally well tolerated and is an important part of effective anti-HIV regimens. It has been used safely by tens of thousands of HIV-positive people for many years. The DAD researchers did not advise doctors to cease prescribing TDF. Further analyses in other large databases are needed to confirm or refute DAD’s findings. If confirmed, additional research will be required to better understand how TDF might injure the liver and which patients who use this drug could be susceptible to such injury. The good news from DAD is that such injury is likely very rare.


CATIE's hepatitis C information

Understanding Cirrhosis of the Liver: First steps for the newly diagnosed – Canadian Association of Hepatology Nurses (CAHN), CATIE

Genvoya approved in Canada – What you need to knowTreatmentUpdate 212

TAF is comingTreatmentUpdate 211

—Sean R. Hosein


  1. Ryom L, Lundgren JD, De Wit S, et al. Use of antiretroviral therapy and risk of end-stage liver disease and hepatocellular carcinoma in HIV-positive persons. AIDS. 2016; in press.
  2. Kovari H, Sabin CA, Ledergerber B, et al. Antiretroviral drugs and risk of chronic ALT elevation in HIV-monoinfected persons: The D:A:D Study. Open Forum Infectious Diseases. 2016; in press.
  3. Mandala J, Nanda K, Wang M, et al. Liver and renal safety of tenofovir disoproxil fumarate in combination with emtricitabine among African women in a pre-exposure prophylaxis trial. BMC Pharmacology & Toxicology. 2014 Dec 24;15:77.
  4. Scourfield A, Jackson A, Waters L, et al. The value of screening HIV-infected individuals for didanosine-related liver disease? Antiviral Therapy. 2011;16(6):941-
  5. Eisenberg EJ, He GX, Lee WA.  Metabolism of GS-7340, a novel phenyl monophosphoramidate intracellular prodrug of PMPA, in blood. Nucleosides, Nucleotides & Nucleic acids. 2001 Apr-Jul;20(4-7):1091-8.
  6. Schouten JN, Van der Ende ME, Koëter T, et al. Risk factors and outcome of HIV-associated idiopathic noncirrhotic portal hypertension. Alimentary Pharmacology & Therapeutics.  2012 Nov;36(9):875-85.