Uridine supplementation for the treatment of antiretroviral therapy-associated lipoatrophy: a randomized, double-blind, placebo-controlled trial

Transcription

1 Antiviral Therapy 12: Uridine supplementation for the treatment of antiretroviral therapy-associated lipoatrophy: a randomized, double-blind, placebo-controlled trial Jussi Sutinen 1,2, Ulrich A Walker 3, Ksenia Sevastianova 1,4 *, Hartwig Klinker 5, Anna-Maija Häkkinen 6, Matti Ristola 2 and Hannele Yki-Järvinen 1 1 Division of Diabetes, Department of Medicine, Helsinki University Central Hospital, Helsinki, Finland 2 Division of Infectious Diseases, Department of Medicine, Helsinki University Central Hospital, Helsinki, Finland 3 Department of Rheumatology and Clinical Immunology, Medizinische Universitätsklinik, Freiburg, Germany 4 Minerva Institute for Medical Research, Helsinki, Finland 5 Department of Medicine, Division of Infectious Diseases, Medizinische Poliklinik, University of Wuerzburg, Germany 6 Department of Oncology, Helsinki University Central Hospital, Helsinki, Finland *Corresponding author: Tel: ; Fax: ; ksenia.sevastianova@helsinki.fi These data were presented at the 7th International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV. Dublin, Ireland, November 2005, Abstract L7, and the 10th European AIDS Conference, Dublin, Ireland, November 2005, Absract PS5/6. Background: Highly active antiretroviral therapy (HAART) is associated with loss of subcutaneous fat (lipoatrophy) presumably due to mitochondrial toxicity of nucleoside reverse transcriptase inhibitors. In vitro, uridine abrogates thymidine analogue-induced toxicity in adipocytes. Methods: A total of 20 patients with HAART-associated lipoatrophy were randomized to receive either a dietary uridine supplement (36 g three times a day for 10 consecutive days/month) or placebo, for 3 months. Body composition was measured using dual energy X-ray absorptiometry, magnetic resonance imaging and proton spectroscopy. Data are mean ± standard error of mean. Results: The mean increases in limb fat (880 ±140 versus 230 ±270 g; P<0.05), intra-abdominal fat (210 ±80 versus 80 ±70 cm 3 ; P<0.05) and total body fat (1,920 ±240 versus 240 ±520 g; P<0.01) were significantly greater in the uridine than in the placebo group. Within the uridine group, the changes from baseline to 3 months were statistically significant in total limb fat (P<0.001), intra-abdominal fat (P<0.05) and total body fat (P<0.001). The proportion of limb fat to total fat increased from 18% to 25% (P<0.05) in the uridine group. Liver fat content and lean body mass remained unchanged in both groups. High-density lipoproteincholesterol concentrations decreased in the uridine and increased in the placebo group, whereas fasting serum insulin concentrations did not change. Uridine supplementation was well tolerated and the virological effect of HAART was not affected. Conclusion: Uridine supplementation significantly and predominantly increased subcutaneous fat mass in lipoatrophic HIV-infected patients during unchanged HAART. Introduction Highly active antiretroviral treatment (HAART) has improved the morbidity and mortality of HIV-1- infected subjects [1,2]. Unfortunately, HAART is also associated with severe metabolic adverse effects [3]. A major and perhaps the most stigmatizing side effect of HAART is lipoatrophy, an abnormal loss of subcutaneous fat that may even result in discontinuation of therapy [4]. The pathophysiological basis for this acquired form of lipodystrophy is likely to be multifactorial, but several lines of evidence point towards an important role of antiretrovirals, in general, and nucleoside reverse transcriptase inhibitors (NRTIs) in particular [5]. Clinical trials have demonstrated that the absence or presence of the thymidine analogues stavudine (d4t) and zidovudine (AZT) in the NRTI backbone of antiretroviral regimens influences an individual s risk of developing lipoatrophy [6]. In vitro data suggest that clinically relevant concentrations of some NRTIs impair polymerase-γ and therefore interfere with the replication of mitochondrial DNA (mtdna) [7]. Suggested pathways eventually leading to NRTIinduced mitochondrial toxicity include effects not only on mtdna but also on mitochondrial RNA, nucleotide phosphorylation and the mitochondrial respiratory chain [8] International Medical Press

2 J Sutinen et al. De novo synthesis of pyrimidine nucleotides is dependent on an intact mitochondrial respiratory chain, a dysfunction of which may ultimately lead to deficiency of intracellular pyrimidines [9]. If cellular pyrimidine deficiency is crucial in the pathophysiology of lipoatrophy, this consequence of mitochondrial toxicity should be overcome by providing exogenous pyrimidines, even if the primary cause for mitochondrial toxicity, that is, NRTI drugs, remains unchanged. Recent in vitro investigations have indeed demonstrated that uridine, which is a precursor for all pyrimidines, is able to prevent and revert NRTI-related mitochondrial toxicity [10 13]. Although theoretically possible, in vitro studies have not shown uridine to compromise the antiretroviral efficacy of NRTIs [11,14 16]. Uridine supplements have previously been used to prevent 5-fluorouracil (5-FU)-induced cytotoxicity in the treatment of solid malignancies [17 19] and as supplementation for hereditary orotic aciduria [20,21]. Uridine treatment is considered safe, with diarrhoea being the dose-limiting side effect. Intravenous administration has also been associated with febrile reactions and phlebitis [17 19]. Uridine can be detected in small quantities in everyday food and drink, for example, in beer [22], but its low concentrations make effective supplementation by dietary means alone unfeasible. In the present study we examined the ability of oral uridine supplementation to revert lipoatrophy without changing current HAART. Patients and methods Study design This investigator-initiated study was performed in a randomized and double-blinded fashion. Eligible patients were randomized to treatment with either NucleomaxX (Pharma Trade Healthcare, Spånga, Sweden), a dietary supplement with a high bioavailability of uridine [23], 36 g (one sachet) three times daily for 10 consecutive days each month, or a similar looking, smelling and tasting placebo using the same dosing schedule for 3 months. Both the uridine supplement and the placebo sachets contained an identical amount of calories (1,660 kj/100 g). The optimal dosing of uridine supplementation is not known. The dosing rationale for the current study (10 days on, 20 days off) was based on data demonstrating (i) a dose-dependent beneficial effect of uridine in vitro on NRTI-induced toxicity [12]; (ii) a Phase I pharmacokinetic study demonstrating that uridine serum levels that are beneficial in vitro can be also achieved in vivo with oral administration [23]; (iii) the fact that exogenous uridine rapidly enters tissue cells from plasma where uridine pools turn over with half-lives of h [24]; and (iv) a relatively rapid reversal of mitochondrial toxicity in the presence of uridine, compared with a more prolonged onset in its absence [13,14,19]. Hence, the dosing of the current study was chosen to achieve high peak uridine concentrations on one hand and to benefit from the residual effect after 10-day supplementation on the other. Eligible patients were assigned to one of the two groups using minimization of differences (calculated for the variables listed below) between the treatment groups as the method of randomization [25,26]. This method of randomization is preferable to stratified randomization as it ensures more accurate matching of baseline variables in small studies [25]. The following variables (relative weight of each variable is given in parentheses) were considered: use of d4t (2 ), age (0.5 ), gender (0.5 ), body mass index (BMI) (1 ), fasting plasma glucose (1 ), fasting serum cholesterol (1 ) and triglyceride concentrations (1 ), use of hypolipaemic (1 ) and oral antidiabetic (1 ) agents. The relative weights were determined empirically to reflect the importance of a given matching variable at baseline. The person responsible for randomization of patients to the treatment groups worked independently from the investigators responsible for enrollment. The primary aim of the study was to determine if uridine supplementation increases the amount of peripheral subcutaneous fat (= total limb fat). Secondary aims included evaluation of effects of uridine on markers of insulin resistance and its safety in HIV-1-infected patients receiving concurrent HAART. Measurements of body composition and laboratory markers of insulin resistance were performed at baseline and after 3 months of treatment. In order to assess the safety of the intervention, adverse events were recorded, and total blood count, alanine aminotransferase (ALT), creatinine, potassium and sodium concentrations were measured from a fasting blood sample at baseline, and after the first, second and third 10-day course of treatment (on days 11, 41 and 71) and at 3 months. HIV-1 viral loads were assessed at baseline and on days 41, 71 and at 3 months. Serum uridine concentrations were measured at baseline and after the third course of uridine/placebo treatment (day 71). Patients were seen by the study physician at baseline, after the second course of treatment (on day 41) and at 3 months. Patients Patients were recruited from the HIV outpatient clinic of Helsinki University Central Hospital. The inclusion criteria included (i) clinically stable HIV-1 infection, (ii) age over 18 years, (iii) treatment with HAART for at least 18 months with no changes in the regimen International Medical Press

3 Uridine in HAART-associated lipoatrophy during 12 weeks prior to randomization, (iv) current use of either d4t or AZT, and (v) lipoatrophy defined as self-reported and investigator-confirmed symptoms of loss of subcutaneous fat with or without increased abdominal girth, breast size or development of a buffalo hump. Exclusion criteria included allergy to milk proteins (NucleomaxX, the uridine supplement used, contains milk protein), current use of didanosine (ddi), pregnancy and lactation. Current use of ddi was an exclusion criterion since in vitro models have not demonstrated any effect of uridine on adipocytes in the presence of ddi, which is a purine rather than a pyrimidine analogue [12,13]. The purpose, nature and potential risks of the study were explained to the study subjects before their written informed consent was obtained. The protocol was approved by the ethics review committee of the Helsinki University Central Hospital. The study is compliant with the revised CONSORT statement guidelines ( Measures of body composition Limb fat, truncal fat, total body fat and lean body mass were measured using dual-energy X-ray absorptiometry (DEXA) (Lunar Prodigy, Madison, WI, USA). Intra-abdominal and abdominal subcutaneous fat were quantified by analysing a total of 16 T1-weighted trans-axial magnetic resonance imaging (MRI) scans as previously described [27]. Liver fat content was measured using proton spectroscopy as previously described [28]. Patients self-assessment of the severity of lipoatrophy was evaluated using a questionnaire in which patients were asked to rate lipoatrophy using a score from 1 (no lipoatrophy) to 5 (very severe lipoatrophy) in six different anatomical sites (face, arms, legs, abdomen, hip and buttocks). Analytical procedures Plasma glucose, serum total and high-density lipoprotein (HDL) cholesterol as well as triglyceride concentrations, glycosylated haemoglobin A1c (Hb A1c ) percentage and plasma ALT activity were measured, and venous blood gas analyses performed as previously described [27]. Serum free insulin concentrations were determined by time-resolved fluoroimmunoassay using an Insulin Kit (AUTOdelfia; Wallac, Turku, Finland). The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated from the formula: fasting glucose (mmol/l) fasting insulin (mu/l)/22.5 [29]. Serum C-reactive protein (CRP) was analysed using a commercial kit (Ultrasensitive CRP Kit; Orion Diagnostica, Espoo, Finland). Plasma lactate concentrations were measured using an enzymatic kit from Roche Diagnostics using an autoanalyzer (Cobas Integra 400/800; Roche Instrument Center, Rotkreutz, Switzerland). Anion gap was calculated from the electrolyte values using the formula ([Na + ] + [K + ]) ( [Cl ] + [HCO 3 ]), all units expressed in mmol/l. Serum uridine concentrations were measured using high performance liquid chromatography, as described in [30]. CD4 + T-cell count was determined using flowcytometry (FACSort/FACSCalibur; Becton Dickinson, San José, CA, USA). HIV-1 viral load was measured using HPS Cobas TaqMan 48 HIV-1 Test (Roche Diagnostics, Branchburg, NJ, USA) with a detection limit of 50 copies/ml. Statistical analysis The hypothesis was to test the effect of uridine supplementation in thymidine analogue-associated lipoatrophy, the null hypothesis being that uridine has no effect. Paired t-tests were used for comparison of changes between baseline and 3 months within the uridine and placebo groups. Unpaired t-tests were used to compare changes between the uridine and placebo group. Logarithmic transformation was performed on data that were not normally distributed. Categorical variables were compared using Fisher s exact test. Safety laboratory follow-up data was analysed using repeated measures of analysis of variance (ANOVA). Analyses were intent-to-treat, that is, all randomized patients were included. Data are expressed as means ±SEM. P-values <0.05 were considered statistically significant. Sample size was calculated based on a study showing an increase of 570 g in arm fat after switching d4t to abacavir [31]. In the present study, a sample size of 10 patients in each group was based on a significance level of 0.05 in a two-group t-test with 80% power to detect a difference in means of total limb fat of 700 g, assuming that the common standard deviation is 500 g, and allowing for a patient drop-out-rate of 10%. Results A total of 39 patients were screened for the study: 15 did not meet the inclusion criteria and four refused to participate. Of the remaining patients, 10 were randomized to receive uridine supplementation and 10 to receive placebo. One patient in the uridine group discontinued the study after 2 days due to the taste of the product. One patient in the placebo group died of myocardial infarction. The remaining 18 patients completed the study. Antiretroviral and all other medication remained unchanged in all participants throughout the study. At baseline, serum uridine concentrations were not different between the groups (5.2 ±0.5 versus 4.4 ±0.5 µmol/l in the uridine versus placebo group; NS) and similar to those previously described by the same laboratory for HIV-1-negative Antiviral Therapy 12:1 99

4 J Sutinen et al. subjects (5.6 ±1.1 µmol/l) [23]. On day 71, serum uridine concentration increased to 16.9 ±1.9 µmol/l (P<0.001) in the uridine group, but remained stable in the placebo group (5.0 ±0.4 µmol/l; NS). The difference between the groups was also statistically highly significant (P<0.001). At baseline, the uridine and placebo groups were similar with respect to age, gender, body composition, metabolic parameters and HIV-1-related characteristics (Tables 1 3). The mean increases in total limb fat, truncal fat, total body fat mass and intra-abdominal fat were significantly greater in the uridine versus the placebo group (Figure 1, Table 2). After 3 months of treatment with uridine, total limb fat mass increased by 29%, truncal fat mass by 10% and total body fat mass by 14% as measured by DEXA (Table 2). Also the percentage fat in the limbs, in the truncal area and total body increased significantly in the uridine but not the placebo group (Table 2). In the uridine group, limb fat accounted for 18 ±3% and truncal fat for 78 ±3% of total fat mass at baseline. After 3 months of treatment, the proportion of limb fat to total fat mass increased to 25 ±2% (P<0.05), whereas that of truncal fat to total fat mass decreased to 72 ±2% (P=0.050). The increase in total limb fat was similar in both AZT-treated (800 ±240 g) and d4t-treated (990 ±110 g) patients in the uridine group. Intra-abdominal fat mass measured by MRI and body weight increased in the uridine group (Table 2). Abdominal subcutaneous fat measured by MRI did not change significantly in the uridine group (1350 ±390 cm 3 versus 1480 ±440 cm 3 at baseline versus at 3 months; NS). In contrast to the significant increase in total body fat mass in the uridine group, there was no significant change in total lean body mass in either group (Figure 1). Liver fat content did not change significantly in either group (Table 2). In the uridine group, five patients reported improvement in the self-assessed lipoatrophy score versus three patients in the placebo group, Table 1. Comparison of the HIV-1-related characteristics of the uridine versus the placebo group at baseline Characteristics Uridine group (n=10) Placebo group (n=10) P-value Duration of HIV-1 infection, years 11.1 ± ±1.4 NS Duration of combination antiretroviral therapy, years 6.0 ± ±0.6 NS Patients with HIV-1 RNA below 50 copies/ml, n 9 8 CD4 + T-cell count, cells/mm ± ±88 NS Patients using stavudine, n 5 5 Patients using zidovudine, n 5 5 Patients using protease inhibitor, n 10 9 Patients using non-nucleoside reverse transcriptase inhibitor, n 2 2 Data are means ±SEM. NS, not significant. Table 2. Effects of uridine and placebo on body composition in patients with HAART-associated lipoatrophy Uridine group Placebo group Baseline (n=10) 3 months (n=9) P-value* Baseline (n=10) 3 months (n=9) P-value* P-value Age, years 47 ±2 49 ±5 Male/female, n 9/1 8/2 Weight, kg 76.7 ± ±5.4 < ± ±3.4 NS NS Body mass index, kg/m ± ±1.3 < ± ±0.9 NS NS Total limb fat, g 3080 ± ±940 < ±800 3,340 ±880 NS <0.05 Total limb fat, % 20.6 ± ±6.0 < ± ±6.4 NS NS Total truncal fat, g 9,980 ±1,730 11,570 ±1,910 < ,660 ±1,350 8,490 ±1,390 NS <0.05 Total truncal fat, % 24.1 ± ±3.5 < ± ±2.9 NS <0.05 Total fat, g 13,490 ±2,580 16,330 ±2,810 < ,140 ±2,090 12,280 ±2,250 NS <0.01 Total fat, % 17.3 ± ±3.0 < ± ±2.7 NS NS Intra-abdominal fat, cm ± ±370 <0.05 2,180 ±470 1,940 ±460 NS <0.05 Liver fat content, % 5.9 ± ±3.9 NS 12.4 ± ±3.1 NS NS Total fat values measured by dual energy X-ray absorptiometry, the percentages denote % fat content of total limb, trunk and body masses. Intra-abdominal fat measured by magnetic resonance imaging. Liver fat content measured by proton spectroscopy. P* for comparison between baseline and 3 months within uridine or placebo group. P for comparison of change in the uridine versus placebo groups. Data are mean ±SEM. HAART, highly active antiretroviral therapy; NS, not significant International Medical Press

5 Uridine in HAART-associated lipoatrophy Table 3. Effects of uridine and placebo on metabolic variables in patients with HAART-associated lipodystrophy Uridine Placebo Baseline (n=10) 3 months (n=9) P-value* Baseline (n=10) 3 months (n=9) P-value* P-value Venous blood ph 7.33 ± ±0.01 < ± ±0.01 NS NS Venous blood base excess 0.15 ± ±0.33 < ± ±0.29 NS NS Plasma lactate, mmol/l 1.53 ± ±0.11 NS 1.50 ± ±0.09 NS NS Anion gap, mmol/l 14.5 ± ±0.7 NS 13.9 ± ±0.5 NS NS Serum insulin, mu/l 12.3 ± ±5.7 NS 10.8 ± ±2.4 NS NS HOMA-IR index 3.2 ± ±1.5 NS 2.8 ± ±0.6 NS NS Hb A1c, % 5.4 ± ±0.1 NS 5.2 ± ±0.3 NS NS Serum triglycerides, mmol/l 2.9 ± ±0.4 NS 3.4 ± ±0.9 NS NS Serum cholesterol, mmol/l 5.6 ± ±0.4 NS 4.8 ± ±0.5 NS NS Serum HDL cholesterol, mmol/l 1.23 ± ±0.08 NS 1.14 ± ±0.09 NS <0.05 Plasma ALT, units/l 41 ±9 42 ±12 NS 36 ±8 32 ±8 NS NS Serum CRP, mg/l 1.5 ± ±0.4 NS 1.4 ± ±0.3 NS NS Reference values: venous blood ph ; venous blood base excess 2.5 to 2.5; plasma lactate mmol/l; anion gap mmol/l; haemoglobin A1c (Hb A1c ) 4 6%; plasma alanine aminotransferase (ALT) units/l for men and units/l for women. Homeostasis model assessment of insulin resistance (HOMA-IR) calculated from the formula: fasting glucose (mmol/l) fasting insulin (mu/l)/22.5 [29]. P* for comparison between baseline and 3 months within uridine or placebo group. P for comparison of change in the uridine versus placebo group. Data are mean ±SEM. CRP, C-reactive protein; HAART, highly active antiretroviral therapy; HDL, high-density lipoprotein; NS, not significant. the difference between the groups was not statistically significant (data not shown). Although there were no statistically significant changes in blood gas analyses between the study groups, venous blood ph and base excess increased significantly within the uridine group. In the uridine group, a trend was observed in terms of an increase of the bicarbonate concentration (24.0 ±0.3 versus 24.9 ±0.3 mmol/l cm 3 at baseline versus at 3 months, P=0.052). Plasma lactate concentrations and anion gap did not change significantly in either group (Table 3). Serum insulin concentrations, other markers of insulin resistance and plasma ALT concentrations remained unchanged in both groups (Table 3). Serum HDLcholesterol non-significantly decreased in the uridine group and increased in the placebo group, the mean difference between the groups was statistically significant (Table 3). None of the patients who had an HIV-1 viral load <50 copies/ml at baseline lost virological control during the study. HIV-1 viral load remained on a similar level among patients who had measurable viral load at baseline (one patient in the uridine group and two in the control group). CD4 + T-cell counts remained stable in both groups (594 ±81 versus 613 ±92 cells/mm 3 in the uridine group at baseline versus 3 months, NS; and 517 ±88 versus 503 ±93 cells/mm 3 in the placebo group; NS). Both uridine and placebo were well tolerated. No subjective side effects were reported in either group during the 3 months of treatment. Total blood count, plasma ALT, creatinine, sodium and potassium concentrations (which were measured additionally on days 11, 41 and 71 to assess the safety of the intervention), remained unchanged in both groups (data not shown). Discussion In this randomized, double-blind, placebo-controlled trial, uridine supplementation increased limb fat mass significantly in HIV-infected lipoatrophic patients during unchanged HAART. Although the sample size was small and the data must, therefore, be considered preliminary and confirmed in a larger trial, the results are encouraging. The changes in body composition were observed using three independent methods (MRI, DEXA and by weighing). Fat mass increased in the limbs and the trunk as measured by DEXA. The increase in limb fat was greater than that of truncal fat since limb fat accounted for 18% of the total body fat mass at baseline and 25% after 3 months of treatment with uridine, while the proportion of truncal of total fat decreased from 78 to 72%. Since truncal fat measured by DEXA is a composite of both intra-abdominal and truncal subcutaneous fat, we performed MRI scans to quantify intra-abdominal adipose tissue. With the 16 slices covering the area from 8 cm above to 8 cm below the 4th and 5th lumbar interspace, we estimate to have captured approximately 60% of the total intra-abdominal fat. In this region we observed a small, but statistically significant, 200 cm 3 increase in intra-abdominal fat volume in the uridine group. This increase could have counteracted changes in insulin sensitivity and could have a potentially harmful effect on cardiovascular risk. Antiviral Therapy 12:1 101

6 J Sutinen et al. Figure 1. Mean change from baseline in (A) total limb fat mass, (B) intra-abdominal fat volume, (C) total body fat mass and (D) lean body mass A Change in limb fat mass from baseline, g 1,500 1, ** Uridine Placebo B Change in intra-abdominal fat fat volume from baseline, cm * Uridine Placebo C 2,500 ** D 2,000 Change in total body fat mass from baseline, g 1, Change in total body lean mass from baseline, g 1, ,000 2,000 1,500 Uridine Placebo 3,000 Uridine Placebo Error bars represent SEM. P<0.05 and P<0.01 for comparison of the change in the uridine versus placebo group; *P<0.05 and **P<0.001 for comparison between baseline and 3 months within the uridine group. Switching from d4t (or AZT) to abacavir or tenofovir, or to an NRTI-sparing regimen, remains the best-documented treatment option for those lipoatrophic patients who can do so safely [32 35]. The average increase in total limb fat with uridine in the present study (890 g in 3 months) was greater than increases (an average of g after 6 12 months) described in switch studies [32 37]. Despite this significant improvement, the achieved limb fat mass remained far below that of healthy men (7 8 kg) [38,39]. Nevertheless, uridine exerted a beneficial effect on lipoatrophy in patients with unchanged HAART and, therefore, could also be used in patients who cannot safely switch their antiretroviral therapy due to toxicity or viral resistance. Limitations of the current study include its small sample size and short duration, which must be taken into account especially with regard to the safety of uridine in HAART-treated patients. These issues and the potentially adverse effects of increased intra-abdominal fat must be reevaluated in a larger trial. An important inclusion criterion of the present study was current use of a thymidine analogue (AZT or d4t), based on the in vitro effects of uridine [12,13]. Therefore, these results cannot be extended to patients who are currently not receiving thymidine analogues or to patients who are using the purine analogue ddi, which was an exclusion criterion of the current study. In keeping with the presumed mechanism of action, that is, improvement of mitochondrial function, there was a statistically significant increase from baseline in venous ph and base excess in the uridine group. The changes in acid base balance with uridine contrast with switch studies in which no such changes were observed [32,36]. Yet, it should be emphasized that venous blood gas analysis and lactate concentration are only surrogate markers of mitochondrial function. No International Medical Press

7 Uridine in HAART-associated lipoatrophy direct measurements of mitochondrial function in tissue biopsies were performed. Despite its significant beneficial effects on body composition, uridine caused neither a change in laboratory markers of insulin resistance nor in liver fat content, although we contend that the group sizes might have been too small. Recent data from a murine model of NRTI-related steatohepatitis suggest that NucleomaxX can also attenuate mitochondrial steatohepatitis [40]. This raises the possibility that insulin resistance and hepatic steatosis are not directly linked with mitochondrial dysfunction in adipocytes of patients receiving HAART. Alternatively, duration of treatment and the increase in adipose tissue mass may have been too small to change insulin sensitivity significantly. Other pharmacological interventions to treat HAART-associated lipoatrophy include thiazolidinediones (glitazones). Most [27,41,42], but not all [43] placebo-controlled studies with rosiglitazone have failed to show a benefit on HAART-associated lipoatrophy. Pioglitazone was recently shown to increase limb fat in patients with HAART-associated lipoatrophy, but this effect was not seen in those patients who continued their d4t-containing regimen [44]. In contrast to uridine, a potential benefit with glitazones may be a reduction in liver fat and improvement in insulin sensitivity [27,43]. However, rosiglitazone, but not pioglitazone treatment, has also been shown to worsen hyperlipidaemia in HAART-treated patients [27,41,44]. In a recent placebo-controlled study, pravastatin was shown not only to modestly decrease serum cholesterol, but also to increase limb fat in hypercholesterolaemic HAART-treated patients [45]. Of note, the presence of lipoatrophy was not an inclusion criterion in that study. Although uridine improved lipoatrophy in the present study and had a neutral effect on markers of insulin sensitivity and liver fat content, it had a potentially harmful effects on visceral fat and HDL-cholesterol, the cardiovascular consequences of which must be carefully evaluated in future trials. Theoretically, supplementation with uridine could compromise the antiretroviral effect of pyrimidine NRTIs by increasing the concentration of intracytoplasmic natural pyrimidine triphosphates to compete with NRTIs for the viral reverse transcriptase enzyme. However, phenotypic HIV-1 resistance assays and animal and limited human data do not indicate that uridine or its metabolites interfere with NRTIs at HIV- 1 reverse transcriptase and, thus, with the antiretroviral efficacy of nucleoside analogues [11,14 16]. In the present study, we also did not observe any negative effect of uridine on the antiviral activity of NRTIs, as none of the patients with baseline viral load below 50 copies/ml lost virological control. In the current study, no side effects were reported by the patients in either the uridine or the placebo group; one patient, however, disliked the taste of NucleomaxX. We also did not observe any changes in laboratory safety parameters including liver function tests, blood count, serum creatinine and electrolyte concentrations. In conclusion, uridine supplementation represents a novel approach based on in vitro data to treat HAARTassociated lipoatrophy. Uridine predominantly and significantly increased subcutaneous fat in patients with HAART-associated lipoatrophy during unchanged antiretroviral therapy. Although encouraging, these data should be viewed as preliminary. The longer term efficacy, safety and optimal dosing of uridine warrant larger trials of longer duration, some of which are presently under way ( show/nct ). Acknowledgements We gratefully acknowledge Katja Tuominen, Mia Urjansson and Diana Schirmer for excellent technical assistance. Role of funding source The present study was fully sponsored by medical research subsidies of the Finnish government. The study sponsor and Pharma Trade Healthcare did not participate in the study design, site selection, data collection, analysis or interpretation, or in the decision to submit for publication. Conflict of interest UA Walker has applied for patents regarding the use of uridine or its precursors in lipodystrophy. He also serves as a consultant for the company that produces NucleomaxX. Other investigators had no conflicts of interest. References 1. Palella FJ Jr, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998; 338: Jaggy C, von Overbeck J, Ledergerber B, et al. Mortality in the Swiss HIV Cohort Study (SHCS) and the Swiss general population. Lancet 2003; 362: Brinkman K, Smeitink JA, Romijn JA, Reiss P. Mitochondrial toxicity induced by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-therapy-related lipodystrophy. Lancet 1999; 354: Duran S, Saves M, Spire B, et al. Failure to maintain longterm adherence to highly active antiretroviral therapy: the role of lipodystrophy. AIDS 2001; 15: Antiviral Therapy 12:1 103

8 J Sutinen et al. 5. Lichtenstein KA. Redefining lipodystrophy syndrome: risks and impact on clinical decision making. J Acquir Immune Defic Syndr 2005; 39: Joly V, Flandre P, Meiffredy V, et al. Increased risk of lipoatrophy under stavudine in HIV-1-infected patients: results of a substudy from a comparative trial. AIDS 2002; 16: Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clin Ther 2000; 22: Cote HC. Possible ways nucleoside analogues can affect mitochondrial DNA content and gene expression during HIV therapy. Antivir Ther 2005; 10(Suppl 2):M Löffler M, Jöckel J, Schuster G, Becker C. Dihydroorotatubiquinone oxidoreductase links mitochondria in the biosynthesis of pyrimidine nucleotides. Mol Cell Biochem 1997; 174: Keilbaugh SA, Hobbs GA, Simpson MV. Anti-human immunodeficiency virus type 1 therapy and peripheral neuropathy: prevention of 2,3 -dideoxycytidine toxicity in PC12 cells, a neuronal model, by uridine and pyruvate. Mol Pharmacol 1993; 44: Sommadossi JP, Carlisle R, Schinazi RF, Zhou Z. Uridine reverses the toxicity of 3 -azido-3 -deoxythymidine in normal human granulocyte-macrophage progenitor cells in vitro without impairment of antiretroviral activity. Antimicrob Agents Chemother 1988; 32: Walker UA, Venhoff N, Koch EC, et al. Uridine abrogates mitochondrial toxicity related to nucleoside analogue reverse transcriptase inhibitors in HepG2 cells. Antivir Ther 2003; 8: Walker UA, Auclair M, Lebrecht D, et al. Uridine abrogates the adverse effects of antiretroviral pyrimidine analogues on adipose cell functions. Antivir Ther 2006; 11: Calabresi P, Falcone A, St Clair MH, et al. Benzylacyclouridine reverses azidothymidine-induced marrow suppression without impairment of anti-human immunodeficiency virus activity. Blood 1990; 76: Koch EC, Schneider J, Weis R, Penning B, Walker UA. Uridine excess does not interfere with the antiretroviral efficacy of nucleoside analogue reverse transcriptase inhibitors. Antivir Ther 2003; 8: McComsey GA, O Riordan M, Setzer B, et al. Effect of NucleomaxX on fat and blood mitochondrial DNA in d4ttreated subjects with clinical lipoatrophy. Antivir Ther 2005; 10(Suppl 3):L Kelsen DP, Martin D, O Neil J, et al. Phase I trial of PN401, an oral prodrug of uridine, to prevent toxicity from fluorouracil in patients with advanced cancer. J Clin Oncol 1997; 15: van Groeningen CJ, Leyva A, Kraal I, Peters GJ, Pinedo HM. Clinical and pharmacokinetic studies of prolonged administration of high-dose uridine intended for rescue from 5-FU toxicity. Cancer Treat Rep 1986; 70: van Groeningen CJ, Peters GJ, Nadal JC, Laurensse E, Pinedo HM. Clinical and pharmacologic study of orally administered uridine. J Natl Cancer Inst 1991; 83: Becroft DM, Phillips LI, Simmonds A. Hereditary orotic aciduria: long-term therapy with uridine and a trial of uracil. J Pediatr 1969; 75: Girot R, Hamet M, Perignon JL, et al. Cellular immune deficiency in two siblings with hereditary orotic aciduria. N Engl J Med 1983; 308: Yamamoto T, Moriwaki Y, Takahashi S, et al. Effect of beer on the plasma concentrations of uridine and purine bases. Metabolism 2002; 51: Venhoff N, Zilly M, Lebrecht D, et al. Uridine pharmacokinetics of mitocnol, a sugar cane extract. AIDS 2005; 19: Darnowski JW, Handschumacher RE. Tissue uridine pools: evidence in vivo of a concentrative mechanism for uridine uptake. Cancer Res 1986; 46: Scott NW, McPherson GC, Ramsay CR, Campbell MK. The method of minimization for allocation to clinical trials. a review. Control Clin Trials 2002; 23: Taves DR. Minimization: a new method of assigning patients to treatment and control groups. Clin Pharmacol Ther 1974; 15: Sutinen J, Häkkinen AM, Westerbacka J, et al. Rosiglitazone in the treatment of HAART-associated lipodystrophy a randomized double-blind placebocontrolled study. Antivir Ther 2003; 8: Sutinen J, Häkkinen AM, Westerbacka J, et al. Increased fat accumulation in the liver in HIV-infected patients with antiretroviral therapy-associated lipodystrophy. AIDS 2002; 16: Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: Zilly M, Langmann P, Winzer R, et al. Liquid chromatographic method for the determination of uridine in human serum. J Chromatogr B Analyt Technol Biomed Life Sci 2004; 803: Moyle GJ, Baldwin C, Langroudi B, Mandalia S, Gazzard BG. A 48-week, randomized, open-label comparison of three abacavir-based substitution approaches in the management of dyslipidemia and peripheral lipoatrophy. J Acquir Immune Defic Syndr 2003; 33: Carr A, Workman C, Smith DE, et al. Abacavir substitution for nucleoside analogs in patients with HIV lipoatrophy: a randomized trial. JAMA 2002; 288: Moyle G, Sabin C, Cartledge J, et al. A randomized, openlabel comparative trial of abacavir or tenofovir DF as replacement for a thymidine analogue in persons with lipoatrophy and suppressed HIV RNA on HAART: the RAVE study. 12th Conference on Retroviruses & Opportunistic Infections. Boston, MA, USA February Abstract Murphy R, Zhang J, Hafner R, et al. Switching to a thymidine analogue-sparing or a nucleoside-sparing regimen improves lipoatrophy: 24-week result of a prospective, randomized clinical trial: AACTG A th Conference on Retroviruses & Opportunistic Infections. Boston, MA, USA, February Abstract Tebas P, Zhang H, Yarasheski K, et al. Switch to protease inhibitor-containing/nucleoside reverse transcriptase inhibitor-sparing regimen increases appendicular fat and serum lipid levels without affecting glucose metabolism or bone mineral density. The results of a prospective randomized trial, ACTG 5125S. 12th Conference on Retroviruses & Opportunistic Infections. Boston, MA, USA, February Abstract John M, McKinnon EJ, James IR, et al. Randomized, controlled, 48-week study of switching stavudine and/or protease inhibitors to combivir/abacavir to prevent or reverse lipoatrophy in HIV-infected patients. J Acquir Immune Defic Syndr 2003; 33: Sutinen J. Interventions for managing antiretroviral therapy-associated lipoatrophy. Curr Opin Infect Dis 2005; 18: Carr A, Samaras K, Burton S, et al. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998; 12:F51 F Engelson ES, Kotler DP, Tan Y, et al. Fat distribution in HIV-infected patients reporting truncal enlargement quantified by whole-body magnetic resonance imaging. Am J Clin Nutr 1999; 69: International Medical Press

9 Uridine in HAART-associated lipoatrophy 40. Lebrecht D, Vargas-Infante YA, Setzer B, et al. Uridine supplementation antagonizes zalcitabine-induced microvesicular steatohepatitis in mice. Hepatology 2007; 45: Carr A, Workman C, Carey D, et al. No effect of rosiglitazone for treatment of HIV-1 lipoatrophy: randomised, double-blind, placebo-controlled trial. Lancet 2004; 363: Cavalvanti R, Kain K, Shen S, Raboud J, Walmsley S. A randomized placebo controlled trial of rosiglitazone for the treatment of HIV lipodystrophy. 12th Conference on Retroviruses & Opportunistic Infections. Boston, MA, USA, February Abstract 854. Accepted for publication 13 September Hadigan C, Yawetz S, Thomas A, et al. Metabolic effects of rosiglitazone in HIV lipodystrophy: a randomized, controlled trial. Ann Intern Med 2004; 140: Slama L, Lanoy E, Valentin MA, et al. Effect of pioglitazone on HIV-1 related lipoatrophy: a randomized double-blind placebo-controlled trial (ANRS 113) with 130 patients. 13th Conference on Retroviruses & Opportunistic Infections. Denver, CO, USA, 5 8 February Abstract 151LB. 45. Mallon PW, Miller J, Kovacic JC, et al. Effect of pravastatin on body composition and markers of cardiovascular disease in HIV-infected men a randomized, placebocontrolled study. AIDS 2006; 20: Antiviral Therapy 12:1 105

10