The main objective of the study is to assess changes in fat distribution over 48 weeks of treatment in patients who currently successfully use zidovudine (AZT) and lamivudine (3TC) as part of their regimen and who will either continue these antiretrovirals or who will switch these antiretrovirals to tenofovir disoproxil fumarate (TDF) and emtricitabine (FTC). Each of these medications is commonly used for the treatment of HIV-1 infection.

Combinations of two nucleoside analogue RT inhibitors (NRTI) are by far the most commonly utilized backbone of combination antiretroviral regimens upon which additional third or fourth agents, a non-nucleoside RT inhibitor (NNRTI) or a (boosted) protease inhibitor (PI), confer sufficient potency for long-term efficacy. The choice of the specific two NRTI is made on the basis of potency and durability, short-and long-term toxicities, drug-drug interactions, the propensity to select for resistance mutations, and dosing convenience. Currently, the most frequently used NRTI combination is zidovudine plus lamivudine, often co-formulated as Combivir®.

The thymidine analogues NRTI have been implicated in the pathogenesis of the lipodystrophy syndrome and of peripheral lipoatrophy in particular. Exposure to zidovudine is associated with a slower and possibly lesser degree of limb fat loss than exposure to stavudine. The mechanisms by which thymidine analogue NRTI contribute to lipoatrophy remain uncertain, although there is evidence that drug-induced mitochondrial toxicity at the level of subcutaneous adipose tissue is involved2 3. NRTI are known inhibitors of mitochondrial DNA polymerase gamma and greater degrees of mtDNA depletion in subcutaneous adipose tissue are associated with greater loss of limb fat as measured by DEXA. In addition, NRTI may detrimentally affect mitochondrial function by mechanisms other than mtDNA depletion. It has also been postulated that the normal regional distribution of body fat may be under hypothalamic autonomic neuronal control, and that effects of NRTI on the central and/or peripheral autonomic pathways may be underlying the selective loss of fat in subcutaneous but not in central regions of the body4.

Withdrawal of thymidine analogues from the antiretroviral regimens of patients with clinical signs of lipodystrophy (and replacing it by abacavir or in more recent studies by tenofovir) has been demonstrated to be able to partially revert limb fat loss in a number of studies, including randomized placebo-controlled trials 5-10. Of note, most patients in these studies had stavudine replaced. Carr11 and Martin6 reported an increase in limb fat of 0.39 kg and 1.26 kg 24 and 104 weeks, respectively, after replacing stavudine or zidovudine by abacavir in the Mitox study (the majority of patients was on a stavudine-based regimen at entry, the minority on a ZDV-based regimen). Moyle8 reported an increase of 1.08 kg of limb fat 48 weeks after withdrawal of thymidine analogues. The reported increases in limb fat have been greatest in patients that were pretreated with stavudine compared to patients pretreated with zidovudine. The rate of peripheral fat loss is slower in patients using zidovudine compared to patients using stavudine, suggesting that patients using stavudine may have usually lost more peripheral fat to begin with than patients using zidovudine. Therefore the potential increase in peripheral fat after withdrawal of the thymidine analogue is expected to be greatest in patients using stavudine. The current study, in contrast to earlier studies, will be limited to patients who with respect to thymidine analogue exposure have solely been exposed to zidovudine and never to stavudine. In order to assure that participants in our study have already had the opportunity to lose a reasonable amount of peripheral fat, we will only include patients who have been continuously treated with zidovudine and lamivudine for at least 2 years. Nolan12 reported that treatment-naïve patients receiving zidovudine-based therapy, after 3 years had lost an average of 31% of limb fat compared to baseline12. Patients who continue to use zidovudine appear to continue to progressively loose limb fat at a rate of several percentage-points per year. By limiting our sample to patients who have been treated with zidovudine and lamivudine for over 2 years we can be confident that the average loss of limb fat is large enough to allow for an objectively measurable and statistically significant regain in limb fat to occur after discontinuation of zidovudine and lamivudine.

Continued treatment with zidovudine plus lamivudine is expected to result in a (further) decrease in the mass of peripheral adipose tissue. Switching to emtricitabine plus tenofovir disoproxil fumarate is expected to result in an increase or at least the absence of a further loss of peripheral adipose tissue. The combination of emtricitabine plus tenofovir disoproxil fumarate is an attractive backbone of antiretroviral combination therapy. This combination has potent antiretroviral activity, has a favorable resistance profile, ease of once-daily dosing, can be conveniently co-formulated in a single tablet, has no special food requirements, is well tolerated, and thus far has not been particularly associated with the occurrence of mitochondrial toxicity or lipoatrophy. With respect to the latter, the recent Gilead 903 trial has been very informative in the sense that no limb fat wasting (measured by DEXA) was observed between 96 and 144 weeks of treatment as opposed to the progressive limb fat loss which was seen in patients randomized to the stavudine-containing control arm13.

In this study we therefore aim to demonstrate a small but statistically significant mean increase in peripheral body fat in the patients who switch to emtricitabine and tenofovir disoproxil fumarate as opposed to a further decrease in peripheral fat in the patients who continue the use of zidovudine and lamivudine. Such increases in peripheral fat are not expected to be observable for the treating physicians and the patients themselves. However, the demonstration of even a small sub-clinical effect will provide important information for future patient management with respect to strategies on how to best prevent progressive lipoatrophy. The Mitox study demonstrated that substituting abacavir for stavudine resulted in an increase in limb fat of 390 grams after 24 weeks11. This increase was likewise not noticeable to clinicians and patients, and could only be detected with the use of DEXA scanning. Nevertheless, the results of Mitox had a huge impact on the management of patients: clinicians started to pro-actively substitute other compounds for stavudine, and to exclude stavudine from first-line regimens in order to prevent (progression of) lipoatrophy. If our study would demonstrate a similar increase in peripheral fat in patients substituting emtricitabine plus tenofovir disoproxil fumarate for zidovudine and lamivudine, this could potentially likewise impact on the continued long-term use of ZDV-based regimens. Although the cleanest comparison would be to replace only zidovudine by tenofovir disoproxil fumarate, we also replace lamivudine by emtricitabine. Given that most experts agree that the difference between lamivudine and emtricitabine is probably not very important and emtricitabine has the additional advantage of allowing the possible use of the fixed dose once daily combination of emtricitabine and tenofovir disoproxil fumarate (Truvada®), we have chosen to replace lamivudine by emtricitabine as well.

Purpose: Antiretroviral regimens containing emtricitabine plus tenofovir disoproxil fumarate might be associated with a lower incidence and severity of lipoatrophy. In this study we compare the effect of proactively switching zidovudine and lamivudine to emtricitabine plus tenofovir disoproxil fumarate on peripheral fat loss with continued lamivudine and zidovudine-based treatment in HIV-1 infected adults also receiving a non-nucleoside reverse transcriptase inhibitor or (boosted) protease inhibitor. In the patients who discontinue the use of zidovudine and switch to emtricitabine and tenofovir disoproxil fumarate we expect to demonstrate a regaining or at least a lack of further loss of peripheral adipose tissue.

Inclusion Criteria:

• Providing written informed consent
• HIV-1 infected patients.
• At least 18 years of age.
• Males or non-pregnant, non-lactating females. Women of childbearing age must have a negative urine pregnancy test at screening. All female participants must be encouraged to utilise adequate contraception for the month preceding entry and for the duration of the study.
• Treatment for at least two years with a first-line regimen of zidovudine plus lamivudine (as either a fixed dose combination or dosed separately) plus either an NNRTI or a (boosted) PI. Patients may have previously used multiple drugs from both the NNRTI or the PI classes.
• Plasma HIV-1 RNA levels < 50 copies/mL for at least 6 months at screening. Isolated measurements of plasma HIV-1 RNA levels above 50 but below 200 copies/ml (socalled blips) are allowed.

Exclusion Criteria:

• Prior treatment with an NRTI other than zidovudine or lamivudine.
• Use of a triple NRTI antiretroviral regimen or a regimen including unboosted saquinavir, fusion inhibitors or hydroxyurea
• Prior virological treatment failure, defined as having had to switch antiretroviral therapy because of virologic failure in the opinion of the physician.
• HIV-2 co-infection.
• Renal impairment and/or use of nephrotoxic agents which in the opinion of the investigator are a contraindication for the use of tenofovir disoproxil fumarate.
• Clinically relevant laboratory abnormalities: anemia, thrombocytopenia, leucopenia, elevated liver transaminases, elevated bilirubin, elevated amylase, elevated lipase.
• Use of co-medication, other than antiretroviral drugs, with a known pharmacological interaction with one or more of the study drugs
• Active alcohol or drug use, sufficient in the investigator's opinion to prevent compliance with the dosing schedule and evaluations (methadone and buprenorphine use is allowed, although the dose of methadone might need to be adjusted).
• Anticipated non-compliance with the protocol.
• Presence of a newly (within 30 days prior to the time of enrolment) diagnosed HIV-related opportunistic infection or condition which may interfere with the ability to comply with the study.
• Chronic active viral hepatitis or other chronic liver disease, which in the opinion of the investigator is a contraindication for the use of any of the study drugs.
• Women who have the intention to become pregnant during the study period.
• Patients who have received within 4 weeks prior to entry, or who have an anticipated need for treatment with radiation therapy or cytotoxic chemotherapeutic agents during the protocol study period.
• Patients who have taken any investigational drug 30 days prior to the start of the study
• Patients with malabsorption syndrome or other gastrointestinal dysfunction which may interfere with drug absorption or prevent the patient from taking oral medication.