icon star paper   Articles  
Back grey_arrow_rt.gif
Lipoatrophy and Mitochondrial Toxicity Issues at the 5th Lipodystrophy Conference
  Reported by Cecilia Shikuma, MD, University of Hawaii and the ACTG July 8, Paris 2003
Data from 2 prospective longitudinal studies were presented that shed light on the degree and course of lipoatrophy occurring with specific NRTI therapy over an approximate 3 year period. The results from a 182 person NRTI substudy of CPCRA's FIRST study (CPCRA 058) [Abs 13] were reported which analyzed difference in subcutaneous fat composition by backbone NRTI - ddI/d4T vs ABC/3TC - utilizing serial anthropometric measurements conducted at 4 months intervals over a median follow-up period of 33 months. Low but steady decrease in subcutaneous fat in both peripheral and central sites was seen over this close to 3 years of follow-up with significantly greater loss found in those treated with ddI/d4T as compared to ZDV/3TC. While it is well known that the use of anthropometric measurements as a research tool is usually hampered by intra- and inter-operator variability, the investigators mentioned that the methodology for anthropometric measurements was rigorously controlled in this study to minimize such problems.
Utilizing whole body DEXAs, usually perceived as a more objective modality for the assessment of fat composition, similar findings on loss of subcutaneous fat loss by specific NRTIs were also reported by the Western Australian group [Abs 95] in subjects (n=72) initiating ZDV vs d4T in their Western Australian HIV cohort participants. They found that the choice of d4T compared to ZDV as part of 1st time HAART regimen, or as a subsequent switching strategy, was associated with an increase in fat wasting over time. After 36 months, the estimated decrease from baseline in % leg fat was 30% for ZDV and 45% for d4T, with younger (-age 35) subjects experiencing less fat loss. Interestingly, assessment of subcutaneous fat mtDNA content from a subcohort of their patients [Abs 18] demonstrated significant decrease in mtDNA levels within 2 - 12 months following such ARV therapy in sequential biopsy samples (n=19, 38 biopsies). The degree of mtDNA depletion fully explained the subsequent fat loss experienced by those on d4T or ZDV, leading the investigators to conclude that mtDNA depletion and mitochondrial toxicity provided a pathophysiological basis for the observed peripheral fat loss. An extended implication of this assessment may be that the newer more "mitochondrial friendly" medications which in vitro assessment in fat demonstrate less mitochondrial depletion may be less likely to result in peripheral fat wasting.
Utilizing data from the WIHS cohort, K. Mulligan presented information [Abs 15] on body composition by DEXA in HIV+ (divided into noART, HAART/noPI, and HAART/PI groups) vs HIV- non-pregnant, predominantly minority women in the U.S. A concerning high degree of obesity was found in this population both among HIV+ as well as HIV- women, with BMI > 30 kg/m2 found in 35% of HIV+ and in 51% of HIV- women. BMI as well as total and regional fat did not differ significantly between HIV+/no ART and HIV- women, suggesting that there was no effect of HIV infection per se. While decrease in both truncal fat and leg fat was seen in the HIV+ HAART/no PI group, an interesting preservation of truncal fat (with similar loss of leg fat) was seen in the HAART/PI group. They concluded that consistent with reports in men, lower levels of peripheral leg fat were seen in HIV+ women despite the high prevalence of obesity in this population. Because this study did not conduct abdominal CTs or MRIs, it is unknown at this time whether the conservation in truncal fat seen in the HAART/PI group represented an increase in the visceral fat component of the abdomen. This study demonstrates that gender specific studies are important in understanding the body composition changes associated with HAART.
Several groups including the MITOX Study Group from Australia and TARHEEL Study Team from the US had previously presented short term data suggesting small improvement in subcutaneous fat as assessed by DEXA and mid-abdominal CT scans following NRTI switch to a "more mitochondrial friendly" regimens. Much concern was voiced at that time as to whether such small improvements were clinically relevant and whether these improvements would continue to occur as time progressed. Andrew Carr, presenting for the MITOX Study Group, presented 128 wk data on their subjects, 50 of whom had originally switched from d4T- or ZDV-containing HAART to ABC and 56 of whom had remained on d4T or ZDV in the original control arm but had been allowed to switch to ABC after wk 24 of study. Utilizing whole body DEXA, the original switch group continued to gain peripheral fat over the course of 2 years achieving a mean increase in limb fat of 1.26 kg from the original baseline of 3.7 kg. The sole positive predictor of recovery found by these investigators was high BMI. Interestingly, the gain in peripheral fat was primarily evident as leg fat with arm fat initially improving but then decreasing over this 2 year period. It should be noted that in the TARHEEL study [Abs 91], data at wk 48 in their 118 subjects who replaced d4T with either ABC or ZDV suggested that the most gain in peripheral fat, also assessed by DEXA, was in the arms rather than legs - median increases in actual (grams) and % change from baseline: arms (239, 35.3%), legs (269, 12.0%) and trunk (859, 16.4%). Neither group demonstrated increases in visceral fat.
The encouraging message in these 2 studies is that there appears to be continued improvement in subcutaneous fat gain over 2 years of followup. However this gain was modest and concerns remain as to whether full recovery is possible. Even if possible, it was suggested that recovery may perhaps take as long as it took to develop the lipoatrophy. Whether the favorable return in fat tissue will be seen over longer period of follow-up, and whether full reversibility will be possible in all patients or whether there is a "point of no return" are important future research questions. As in all disease states, prevention of lipoatrophy, rather than treatment or switch schemes, is likely to be the most beneficial strategy for patients in clinical care. Whether it might be more appropriate to "reserve" d4T for special circumstances rather than utilizing it blanketly as part of 1st time HAART was debated. Furthermore, prevention strategies (such as perhaps the use of rosiglitazone together with d4T to delay the onset of lipoatrophy) may warrant investigation.
The lack of effective treatment modalities for NRTI-induced toxicity has been a very frustrating issue in the HIV metabolic field. While the use of PPARg agonists has shown some promise, as reported already from this conference by Jules Levin, Ulrich Walker from University of Freiburg, Germany [Abs 19] presented very preliminary data on the possible efficacy of uridine to prevent and treat NRTI-mediated mtDNA depletion. Uridine is a compound occasionally utilized in patients with congenital mitochondrial diseases and is the standard treatment for hereditary orotic aciduria. In an in vitro model of human HepG1-hepatocytes observed for 25 days, ddC induces a severe depletion of mtDNA to 8% of control, a drop in respiratory chain subunit COX II to 8% of control, a severe intracellular steatosis and an increase in lactate to 350% of control. The addition of uridine with ddC resulted in repletion of mtDNA to 65% of control levels as well as full normalization of cell proliferation, lactate levels and intracellular lipid levels. Even more impressive was the fact that uridine could be used in a "simulated" in vitro "treatment" mode to substantially ameliorate such evidence of mitochondrial toxicity when it was added following 15 days of ddC exposure. Similar results were obtained when ZDV or d4T was used instead of ddC. Uridine did not alter the IC50 or IC90 of currently licensed NRTIs in HIV resistance assays.
A uridine compound extracted from sugar cane named Mitocnol (NucleomaxX¨) is available for clinical use. The chief known side effect of this medication is diarrhea when given orally. Dr. Walker reported use of this agent in a single 50 yr old subject on d4T containing ARV regimen who exhibited evidence of hepatitis, steatosis, CK elevation with a lactate level of 4.2. Treatment for 4 days with Mitocnol resulted in a substantial improvement in these parameters. A phase I clinical trial is planned. While the in vitro data presented by Dr. Walker requires independent confirmation and treatment in one subject may have little value in determining efficacy, further research on the potential use of uridine will certainly be watched with interest.
  icon paper stack View Older Articles   Back to Top   www.natap.org