- Metabolic Complications of HIV-1 Disease (State-of-the-Art Summary)
Authored by A. Carr and S. Grinspoon
This was a thorough discussion of a complex topic that left me with the understanding that the syndrome is made up of different physical and physiologic components. We are trying to group them together into a cohesive understanding, and we just don't know enough about the science yet to explain it clearly. There has been a lot of progress made in our understanding and it would be foolish to look from afar and just say it is too complicated to understand -- so let's review what was clear from both presentations. The first section by Carr dealt with morphological changes.
Elevations in Lactate
Dr. Carr began with a discussion of lactate, a metabolic byproduct of cell life that accumulates in the body and the blood to sometimes dangerous levels and may be attributable to use of NRTI. "Severe" lactic acidosis is defined as greater than level 10, is associated with significant symptoms, and is 80% fatal. The symptoms are fatigue, abdominal pain, distension, and respiratory problems. This is very uncommon but definitely occurs. The next gradation, "moderate," is defined as a level of 5-10 which will usually present with similar symptoms, is not usually fatal, and requires some action (i.e., supportive care and cessation of NRTI). The third level, "mild," is the most common with levels between 2-5. The appropriate course of action for this level is not clear.
These elevations in lactate levels have been reported with all NRTIs. However, there are slightly more reports with stavudine (d4T) than others, and these elevations are associated with increased incidence in females, liver disease, pregnancy, and advancing age. The incidence may be as low as 1/1000, but this is an estimate. Both AZT and d4T have been associated with elevations in lactate that rise over six months and then plateau with d4T having a slightly higher level than AZT. There is a concern that the elevations in lactate may be caused by chronic use of NRTI as well as the use of specific agents.
NRTI also cause loss of fat, commonly seen in the thigh, face, and upper extremities. The physiology behind these losses is unclear but may result from adipolysis, a dissolving of fat due to hormonal or structural changes in fat cells. It is hypothesized that mitochondrial DNA may be damaged in these cells and cause fat loss as a result of their dysfunction and subsequent effects on other cells in the body. These changes are not seen with protease inhibitors, have no clear relation with NNRTI, and have a clear association with NRTI. The ratio of these effects from a Swiss study were increased with d4T to 2.5 and, with combined d4T/ddI, were 5.1 (which reflects increased risks with these agents). There is speculation that these effects may be related to increased intracellular levels of metabolites of d4T which are exacerbated by concomitant protease inhibitor use (which had been shown to raise these intracellular levels). Further investigation is needed to clarify this relationship.
d4T has been associated with this condition. Brew reported on elevations of lactate associated with peripheral neuropathy (see abstract 9
) and Dr. Carr reported on a clinical trial that withdrew d4T and subsequently found improvement in both lactate and symptoms.
Indinavir was found to inhibit in vitro bone formation without toxicity, through its suppressive action on osteoblasts, the cells that make bone (see abstract 541
). It is unclear what dosing is required to cause this effect but there was speculation that this is occurring at current dosages due to the clinical presentation. Ritonavir suppresses bone reabsorption (osteoclasts) and when used in combination with indinavir may have a counter-balancing effect on bone loss. Other proteases are not associated with these effects, and there was a discussion about the presence of lactic academia as associative or causal with bone loss (see abstract 631
) versus other reports (see abstract 634
) that lactic acidosis is not associated with bone loss. Stay tuned for the final story.
Grinspoon focused his comments on research surrounding endocrine and metabolic toxicities. He summarized the main areas of concern as loss of fat, fat deposition, dyslipidemias, insulin resistance, and the subsequent risk of coronary artery disease. These syndromes may be a direct result of the drugs or an indirect effect mediated through the metabolic effects of increased fat changing hormonal activity.
The prevalence of elevations in glucose and insulin with peripheral resistance to insulin is quite common with HAART. Fasting glucose may be normal in many patients, but two-hour glucose tolerance tests must be used to tease out this abnormality, and these were abnormal a significant amount of time, upwards of 40%. Fasting insulin levels were sometimes elevated but were not as sensitive as a two-hour insulin measure, which was significantly abnormal. Patients with abnormal morphology diagnosed as having mixed lipoatrophy and fat deposition has markedly abnormals for glucose, >200 8.9%, >140 35%, with 55% elevated cholesterol and triglycerides. Indinavir and other protease inhibitors were shown to decrease glucose uptake and decrease insulin sensitivity over time, ritonavir raised triglycerides markedly over two weeks alone in HIV-negative subjects as well.
Grinspoon presented anatomic data showing that carotid wall thickness was greatly increased with protease inhibitors with a value of 22.5mm thickness for protease inhibitors, 14.9mm for non-protease-inhibitor regimen, versus 6.4mm for normal patients. Another study showed decreased distensibility or increased rigidity reflecting early vascular disease in patients on HAART. Epidemiologic data presented by Klein (see abstract 655
) showed no increase in hospitalizations in a large cohort of patients on meds, but increased number of cardiovascular events suggesting that subclinical disease may be accumulating. Krause (see abstract 657
) reported a dose-related increase with higher incidence of MI, though this was only an 18-month follow-up. Over the last two years there are increasing number of studies beginning to detect coronary disease and it is my impression that this will become more apparent in coming years.
The role of diet and exercise was discussed with documented reductions of cholesterol by 11% and 21% for triglycerides. Combination therapy with gemfibrazole, atorvastatin, and pravastatin were discussed with mild success. The role of insulin-sensitizing agents such as metformin was discussed and showed some preliminary success using 500mg BID versus a placebo for twelve weeks. There was great care to follow lactates as a measure of toxicity and the treatment was well tolerated. Increases in insulin were seen and certain markers of coronary vascular disease TPA and PAI-1 were reduced as well as improvements in morphology with decreases in weight, waist circumference, and diastolic blood pressure. These effects are synergistic for reducing coronary vascular risk and could be a positive approach. Further research is needed.
The use of PPAR GAMMA activating agents to increase glucose transport into cells and increase peripheral glucose uptake was also discussed. Use with HIV-negative patients with increased abdominal fat resulted in decreases in visceral fat and increased fat deposition in the thigh and subcutaneous tissue of the abdomen. Rosiglitazone and pioglitazone were discussed and will be used in an upcoming ACTG trial.
The PI-to-NNRTI switch trials have been virologically successful and have differed in their cholesterol reductions. With nevirapine, there was some lowering of cholesterol in trials; with efavirenz, there was little cholesterol change and some elevation of HDL. There have been few improvements reported in morphological changes. Walli reported improvements in insulin sensitivity and a decrease in both total cholesterol and triglycerides when switching abacavir for a protease (see abstract 672
). Estrada showed no improvement in the same parameters when switching efavirenz for a protease inhibitor (see abstract 671
). Jolly reported a prospective randomized switch trial for metabolic and morphologic changes with patients who were failing NRTI-only regimen (see abstract 539
). The patients were randomized to d4T/3TC/IDV or AZT/3TC/IDV and had no difference virologically or metabolically, but patients on d4T had and increased incidence of clinical lipoatrophy as measured by exam and skin fold thickness.
Dr. Grinspoon stressed the need for further studies and individual discussion for any given treatment plan. The information and mechanisms for the changes in body composition and metabolism are just beginning to emerge. As we try to balance the risk of these side effects versus no side effects, there are no easy answers. These toxicities are changing the paradigm of HIV medicine. They are focusing our efforts on starting treatment later and considering different strategies to use less drugs and to use them in synergistic combinations in order to preserve quality of life as we expand longevity. It will be a long dance, hopefully, and many of us are forced to make decisions before we have all the answers. Hopefully, the next wave of research will shine more light on these questions. Until the answers arrive, education and communication will be our best tools.