icon_folder.gif   Conference Reports for NATAP  
  8th European Conference on Clinical Aspects and treatment of HIV-Infection (ECCATHI)
Athens, Greece - October 2001
Back grey_arrow_rt.gif
Summaries from the ECCATH Meeting:
Written for NATAP by Graeme Moyle, MD, Chelsea & Westminster Medical Center, UK, London
Interferon Monotherapy in HCV/HIV Coinfected Patients
Body Changes in Italian Women: risk factors
Drugs & the Liver
Lipodystrophy & Metabolic Abnormalities: mitochondrial toxicity & NRTIs, fat cells
Lipids & Medication Choices: the Atlantic Study
HAART and the Heart
The Athens conference was disappointing with regards to HIV co-infection with Hepatitis, despite the high and rising rates of co-infection in Europe. Much of the presented data was retrospective, included therapy approaches that are now no longer standard of care or included few patients.
Drugs and the Liver
Evaluation of hepatotoxicity of antiretroviral therapy was reported from a number of clinical cohorts. In a study of 182 individuals from Spain who were antiretroviral therapy-naļve and commenced on efavirenz-based regimens, follow-up over six months was reported. Of the study population, 34% were co-infected with hepatitis C and 53% had received a diagnosis of AIDS. 15% of patients discontinued efavirenz during the follow-up but in no cases was this due to liver-toxicity. Five-fold or greater elevations in liver function tests (transaminases) occurred in 6.1% of patients with 9 of the 11 events occurring in individuals who were Hepatitis C. In a multi-variate analysis, Hepatitis C was the only significant risk factor for transaminase elevation [abstract P129]. A similar evaluation of 427 patients in Italy, initiating non-nucleoside reverse transcriptase inhibitor (NNRTI)-based regimens between March 1997 and March 2000 at a single centre evaluated the frequency of transaminitis to 5 times above the upper limit of normal or an increase of at least 2.5 fold above base-line values in those with elevated values at baseline. The cohort was 65% male, 62% previous or current IDU, with 6% hepatitis B surface antigen positive and 54% hepatitis C positive. Median follow was 8 months, during which 17 (4%) of individuals experienced transaminitis, an incidence of 5 per 100 person years of treatment. Only 5 of the 17 individuals reported a more severe transaminitis to greater than 10 times the upper limit of the normal range. The median time to onset of transaminitis was 22 weeks and was observed in 4% person years in 295 patients treated with nevirapine and 6% person years in 132 patients treated with efavirenz. These differences were not statistically significant. 14 of the 17 patients (82%) with transaminitis were hepatitis C co-infected and 12 (71%) received AZT as their thymadine analogue. In a multi-variant model, use of AZT was associated with 3.2-fold greater relative risk of transaminitis compared to regimens that did not contain AZT and hepatitis C co-infection increased the relative hazard 4.8-fold. Choice of non-nucleoside was not associated with differences in relative risk.
HIV, Hepatitis C and Survival
A retrospective analysis of patients in the Bonn HIV Cohort, followed since 1990 was evaluated for total and liver-related deaths and assessed the impacts of triple therapy (HAART) introduced in 1996. Of 285 patients included in the study, 94 of whom received HAART, 107 deaths were recorded, 81 in the years 1990-1995 and 26 in the years 1996 to 2000 after the introduction of HAART. Of these 107 deaths 25 were thought related to liver disease, 15 in the pre-HAART era and 10 in the post-HAART availability era. Only two liver-related deaths, however, actually occurred in individuals who had received HAART and this was significantly lower, relative to the proportion of the population who never received HAART. Kaplan-Meier analysis confirmed the benefit for HAART in individuals who were co-infected, with survival increasing significantly from 1679 days post HIV-diagnosis without HAART to 2658 days with HAART. This advantage remained when liver-related deaths alone were considered although it was not significant in this smaller sub-analysis. [Abstract P33].
Treatments Studies
As the availability of pegylated interferons and ribavirin increased within Europe, so the end of the era of studies looking at standard interferon preparations is drawing to a close with the general conclusion that the efficacy of interferon alpha monotherapy in people with HIV is not impressive for Hepatitis C.
A multi-centre open randomised controlled trial comparing interferon alpha 2A (6,000,000 units three times a week) to no treatment for Hepatitis C in persons with HIV was reported from an Italian cohort. Patients entering the study had established hepatitis C infection and a CD4 of 300cells/mm3. Assessment of response was an undetectable Hepatitis C RNA (less than 100cps/ml) after six months of treatment. These responders were offered continuation of treatment for an additional six months. 83 patients were included in the study between January 1997 and January 2000 with 53% of patients having hepatitis C genotype 1. 84% of the patients were male and 84% had a history of injection drug-use with 50% receiving antiretroviral therapy with at least three drugs at study entry. 40 patients received therapy and 43 acted as controls with no difference in clinical or demographic characteristics between the two groups. A treatment response at six months was observed for 16% of treated patients but none of the untreated patients, and a sustained response with normalised ALT and undetectable hepatitis C RNA at one year observed in 7% of treated individuals. No difference in disease markers for HIV disease were observed over the course of Hepatitis C treatment or non-treatment with interferon. Editorial note from Jules Levin: a 7% SVR is not unusual with inteferon monotherapy, although only 53% had genotype 1. In Europe its estimated that about 50% have genotype 1 while in the US 70% have genotype 1. 80% who contracted HIV through IVDU have genotype 1 anf over 90% of African-Americans have genotype 1. As is the case with many of these small studies I'm not sure how much they tell us about response to therapy by coinfected patients. In addition, there is no data reported here on adherence and discontinuation rates. [Abstract P292]
Poor responses, such as in this study, underline the need for a shift in treatment for Hepatitis C to include combination approaches such as pegylated interferon's with ribavirin and the development of new agents which may improve the efficacy of therapy in persons with Hepatitis C and HIV co-infection. Unfortunately the conference did not include any studies that evaluated combination treatment approaches in randomised studies.
Lipodystrophy and Metabolic Disturbances
Fat cells (adipocytes)
Contributions to the first report made by Jules Levin and reports from paper's author David Nolan
The aetiology (cause) of lipodystrophy (body changes) remains unclear although there is evidence that problems may exist both at the hepatic and at the level of the fat cell or adipocyte.
A biopsy study was presented during the late breakers, evaluating mitochondrial content in samples of fat excised from the super-iliac ('love handle') region. These included both HIV negative controls and HIV positive individuals who had not received therapy as well as treated individuals with or without clinical lipodystrophy. No differences in the mitochondrial DNA content were found between the two control groups of HIV negative and HIV positive but untreated folks. Amongst individuals receiving antiretroviral therapy the mitochondrial DNA content was significantly lower relative to controls. There were more d4T-treated than AZT-treated study patients who had clinical lipodystrophy: 9 of 12 individuals biopsied whilst receiving D4T had clinical lipoatrophy, 4 of 10 individuals biopsied whilst receiving AZT had peripheral lipoatrophy. 2 of the 4 individuals on AZT developed lipodystrophy while on AZT and had not had d4T previously. 2 developed lipodystrophy while on d4T (and had a biopsy before switching to AZT) ­ and then had another biopsy 4-5 months later while on AZT. In these 2 latter cases fat wasting did not improve, although their mitochondrial DNA levels increased markedly in the time period. Previous studies have demonstrated that individuals with lipoatrophy regardless of therapy have reduced levels of mitochondrial DNA. However, the investigators reported results for patients who had only received d4T without prior AZT (n=9) and for patients who had only received AZT without prior d4T (n=8). Overall, in the individuals who had only taken D4T, there were lower levels of mitochondrial DNA relative to the AZT group and relative to the controls. The group receiving AZT had also significantly lower mitochondrial DNA content relative to control patients. The mtDNA levels/cell were 759 copies/cell for the controls, 339 copies/cell for the AZT group (44% of control group), and 90 copies/cell for the d4T group (12% of control group)-- AZT vs d4T, p=0.001. The authors commented that mtDNA depletion appeared quite early on in the course of d4T therapy (within 5 months and before any evidence of lipodystrophy), and that lower levels were maintained after that early decrease. Switching to AZT was also associated with a rapid increase of mtDNA levels (again, within 5 months) without any changes in lipodystrophy.
When including all the study patients in the analysis including patients who had prior AZT experience but were now on d4T and patients with prior d4T experience but were now on AZT, the results were similar‹control groups mtDNA level/cell 759 copies/cell vs AZT 389 copies/cell (51% of controls) vs d4T 90 copies/cell (12% of controls). This is a small study that looks at the affect of d4T and AZT on mitochondrial depletion. The study does not find mitochondrial depletion is related to mitochondrial dysfunction, cell pathology, and lipodystrophy. These remained unanswered questions that need to be explored. Another question is whether early mtDNA depletion associated with d4T therapy can predict the later development of lipodystrophy. Using only lipodystrophy as the outcome in these types of studies, without also looking at cell pathology and mitochondrial dysfunction, is likely to give only a crude measure of the underlying process. Nolan commented further that there didn't seem to be a slow, steady decline in mtDNA levels over time that might be expected if the development of lipodystrophy and the development of mtDNA were closely correlated. I would speculate that perhaps a certain threshold of mtDNA depletion has to occur before lipodystrophy might start to develop. Nolan said mtDNA depletion was much more strongly correlated with d4T treatment than it was with the presence of lipodystrophy. [abstract LB03]
Comments from Dr. Moyle: Several studies evaluating individuals who have only ever received AZT or only ever received D4T have not found differences between these two drugs in first-line therapy although the Novavir study, which randomised individuals pre-treated with AZT to either continue AZT or take D4T in addition to 3TC and a PI did find that lipoatrophy was more common amongst individuals randomised to D4T. It is possible, therefore, that there may be a "priming" effect from AZT that leads to lipodystrophy being more common when D4T is instituted within a regimen. This hypothesis requires further evaluation in in-vitro and clinical studies.
Edit note from Jules Levin: a number of studies have d4T is associated with lipoatrophy but to the best of my knowledge no study has found d4T causes lipoatrophy. The types of studies needed to be done to answer the question conclusively are hard to design and implement. That's why these studies have not yet been conducted.
An in-vitro cell culture looked at the effect of antiretroviral therapy on brown pre-adipocytes. These cells are present in mice and are responsible for heat production (thermogenesis) in these animals. These cells are also seen in human infants but it is unclear where they remain present into adulthood, although they may be representative of visceral fat cells. The cells are brown because of the predominance of mitochondria within the cells. The in-vitro cell culture studied the differentiation and gene expression of the UCP1 gene when the cultures were treated with either stavudine, nevirapine or indinavir during differentiation from pre-adipocyte to mature adipocyte. This process normally occurs over 8-10 days in the culture, which the first four days spent in a proliferation phase and then the subsequent 4-6 days spent in a differentiation phase with the cells acquiring increased amounts of lipid and expressing the UCP 1 gene product. When stavudine was included in the culture, cell morphology and differentiation was unaffected although the expression of UCP1 and other markers of adipogenesis and mitochondriogenesis were up-regulated. Similar modest positive benefits on maturation and differentiation were observed when nevirapine was included in the culture. Treatment with indinavir, on the other hand, decreased the expression of UCP1 and other markers of adipogenesis and mitochondriogenesis leading to abnormal adipocyte morphology and differentiation (Abstract P95). Similar data had previously been presented looking at peripheral fat cell lines that indicate that whilst nucleoside analogues have limited impact on peripheral adipocytes maturing in culture, protease inhibitors appear to affect fat cell maturation and this may play a role in reduced replacement of lost fat cells. It is interesting to observe that, in the proliferation phase of pre-adipocytes, when these cells are undergoing mitosis, and are most likely to express enzymes responsible for phosphorylation and activation of thymidine nucleoside analogues, no impact of stavudine was seen with regards to either proliferation cell morphology or adipogenesis mitochondriogensis.
A study looking at mitochondrial DNA content in peripheral blood lymphocytes from HIV positive children with or without lipodystrophy was also reported to see whether blood markers may be useful to see whether mitochondrial problems were occurring in these individuals to act as a blood marker of risk of lipodystrophy or diagnosis of lipodystrophy. 18 children were studied, including 6 with clinically evident lipodystrophy and 12 without lipodystrophy and, in addition, 10 healthy HIV negative children. The study found that mitochondrial function, lymphocyte apoptosis and mitochondrial DNA content were similar in the individuals with lipodystrophy and without lipodystrophy relative to healthy controls. The results suggest that, at least in peripheral blood lymphocytes, mitochondrial DNA content and function are normal and this will not represent a means to assess lipodystrophy. As lymphocytes in children may be different to adults it would seem appropriate that this study were repeated with adults with or without lipodystrophy (abstract P98).
A study of muscle biopsies in 7 individuals with clinical lipodystrophy, 10 individuals (including 3 treatment-naļve patients) without lipodystrophy and 12 HIV negative control patients reported morphological and mitochondrial DNA changes within samples. Patients with lipodystrophy presented with morphological change in the muscle mitochondrial with structural and ultra-structural abnormalities and mitochondrial respiratory chain dysfunction with some areas of mitochondrial DNA deletion. Abnormalities in the structure of mitochondria were also observed in patients without lipodystrophy, although mitochondrial DNA depletion was not present. The healthy volunteer group showed no abnormalities of mitochondrial appearance or function. The authors suggest that muscle biopsies may be a useful way of evaluating the presence of absence of lipodystrophy. The types of antiretroviral therapy used within the study were not reported (abstract P133).
Possible risk factors
Risks of development of morphological alterations were respectively assessed in women in a study reported from Italy. The study included 264 women who commenced follow-up in 1998. The prevalence of morphological abnormalities increased from 11.7% at baseline to 40.9% one year later and 58.3% two years later. Risk factors for the new onset on morphological change included the duration of antiretroviral therapy but did not include choice of antiretroviral agent. Individuals who had detectable viral loads during follow-up had a non-significant reduction in the risk of developing lipodystrophy consistent with previous studies. In individuals who saw resolution of breast enlargement between assessments in 1999 and 2000, detectable viral load was the only variable significantly associated with this resolution. These data are suggestive that whilst a high and increasing frequency of morphological change occurs in individuals on antiretroviral therapy this risk is related more to the duration and effectiveness of treatment rather than to specific drug choices (Abstract P105).
[editorial note from Jules Levin: perhaps patients had used so many different HIV drugs that no association could be found with any particular drug(s). But an association with duration of therapy makes sense. The association with detectable viral load could mean the patients were not fully adherent, were taking period interruptions, or that full suppression of HIV is a factor in developing lipodystrophy.]
A second study from Italy of 655 patients on first-line therapy assessed statistical associates with development of morphological change over time. Median follow-up of patients was 86 weeks during which 128 patients developed morphological alterations. Female gender and Hepatitis C positivity were associated with an increased risk of developing morphological change. Again in this study no drug choice was associated with significant alteration in risk. The authors suggested that genetic and acquired factors independently influence the risk of developing morphological change (Abstract P108).
Lipids and Medication Choices
The comparative effects of different antiretroviral therapy choices on lipids were reported from a sub-study of the Atlantic Study (abstract P126). This study looked at treatment-naļve patients at changes in cholesterol, triglycerides and other lipo-proteins of the impact of initiating either nevirapine, 3TC or indinavir on the back of D4T and DDI. Changes in cholesterol throughout 96 weeks of follow-up favoured the nevirapine group. At week 24, whilst LDL rose modestly in all groups, HDL the "protective" cholesterol rose 49% in individuals who received nevirapine relative to just 9% in those who received indinavir. This effect was sustained through to 96 weeks when the HDL (good) cholesterol in the nevirapine group remained 40% above the baseline, whereas in indinavir it was 6% higher and with 3TC 20% higher. The LDL (bad) cholesterol at 96 weeks had risen 3% with 3TC and, 16% with indinavir and 19% with nevirapine but, overall, there was an improvement in the HDL to total cholesterol ratio, a marker of future risk of cardiac disease, whereas the ratio had worsened in the group receiving indinavir and was essentially unchanged in the group who received 3TC. Triglycerides also rose in all three groups but by 96% in indinavir, 61% in 3TC and only 43% with nevirapine. Triglycerides are a known independent risk factor for cardiac disease.
Larger ongoing studies are evaluating whether changes in lipids in people with HIV will lead to an increased risk of heart disease, for example, the DAD study (Abstract 018) is looking at the development of cardio-vascular disease in 20,00 individuals across 11 databases. Baseline data in the first 17,850 patients within this cohort indicate that individuals naļve to therapy or not currently receiving therapy had the lowest cholesterol values, whereas elevations of cholesterol above 220mg/dl were present in 22% of NNRTI treated patients, 25% of protease inhibitor patients and 44% of those taking both drug classes in combinations including nucleoside analogue therapies. Other factors were also noted to play a role, with those with more advanced HIV infection having lower HDL (good cholesterol) levels and higher triglyceride levels.
HAART and the Heart
A study of 70 patients evaluated carotid intimal thickness using ultrasound was reported. The thickness of the carotid intima is a known marker of risk of cardio-vascular disease in the general population. The study of 70 patients included 50 men and 20 women with a mean age of 42 years, a CD4 of 507/mm3 with 50% of individuals having a viral load <50 cps/ml. Only four individuals were treatment-naļve with 25 patients receiving protease inhibitors and 41 patients receiving protease-sparing regimens. 23 of the patients had clinical lipodystrophy (not defined) and lipid changes including hypercholesterolaemia (greater than 5.2mmol/l or 200mg/dl) in 54 individuals and hypertriglyceridaemia (greater than 1.95mmol/l) in 54% of individuals. Carotid intimal thickening was noted 16% more of HIV positive individuals relative to the known HIV negative age matched controls. Increases in the risk carotid intimal thickening were noted in a univariant model to be higher in persons with lipodystrophy and with a multi-variant analysis specifically with increase in abdominal circumference (abstract P125). This risk is consistent with risk in the general population.
As more data on heart disease risk and HIV emerge it is becoming increasingly evident, to no ones surprise, that the same risk factors for heart disease in the general population increase risk in the HIV population. With improved survival from HIV the infectious disease doctors are going to soon need to start updating their cardiology skills.