Just as HIV disease progression is highly variable, so too is the aftermath of successful suppression with the new highly active antiviral therapies (HAART). One disturbing newly emerging syndrome, commonly referred to as HIV-associated lipodystrophy (HAL) includes a vast constellation of body shape and metabolic abnormalities. Most symptoms fall into three main categories: The most obvious are altered fat deposits. Subcutaneous fat in the arms, legs and face shrinks, resulting in thin limbs with bulging veins and marked facial wrinkling with hollow cheeks. These losses are frequently accompanied by new fat bulges between and above the shoulder blades (dorsal cervical fat pads or "buffalo hump") and in the abdominal cavity, surrounding the internal organs (truncal adiposity or "protease paunch"). Breast enlargement also occurs, mostly in women.
More common are changes in fat metabolism, resulting in hyperlipidemia, or high levels of blood lipids (triglycerides and cholesterol). Alterations in sugar processing may accompany these perturbations in fat storage and utilization. Blood sugar may rise simply because fat cells are processing less glucose into fat stores. Insulin production may rise significantly to increase the removal of sugar from the blood, and this elevated production may or may not successfully keep sugar levels below abnormal, hyperglycemic levels. Elevated insulin also tends to lose its effect over time as cells decrease their response to this hormone, a phenomenon called "insulin resistance."
Some of lipodystrophy's manifestations appeared before protease inhibitors became available and may be associated with other antivirals, genetic host factors, or even underlying metabolic problems inherent to HIV disease itself. The deluge of reports flooding in since protease inhibitors came into common use offers compelling evidence of some linkage to lipodystrophy. It could be exacerbated by direct action of the protease inhibitors themselves or as an effect of the immune reconstitution that occurs following successful viral suppression by protease inhibitor-containing regimens.
At last winter's 6th Conference on Retroviruses and Opportunistic Infections in Chicago, there was no consensus on the syndrome's manifestations, and lipodystrophy's frequency was widely debated. Disagreement also arose as to the cause and relationship of the associated disorders, while strategies for preventing or reversing them are still in their infancy. Obviously, much work remains to be done in understanding and reversing lipodystrophy. Still, the Conference proceedings marked a watershed in this area: The widespread debate showed that lipodystrophy has attracted researchers' attention, and their disagreements indicated that they had already taken the first steps in thinking and working on the problem.
Prevalence of the Signs and Symptoms
The reported prevalence of lipodystrophy varies greatly, depending on the definition of lipodystrophy and the methods used for evaluating it. The observed rates range from as low as 3% to as high as 80% and appear to be increasing with prolonged use of HAART. At the Retrovirus Conference, Andrew Carr from Sydney, Australia, reported that 83% of 116 patients on protease inhibitors (PIs) complained of having fat redistribution, as opposed to only 4% of PI-naïve patients (abstract 641). This was after an average time of 21 months on therapy. Just 11% of the 116 rated their lipodystrophy as severe, while another 30% deemed their lipodystrophy "moderate." The investigators defended themselves from previous criticism of these self-reported evaluations by noting that doctors agreed with their patients' assessments 98% of the time after conducting physical examinations. Two factors were independently associated with loss of fat mass: lower weight when starting pre-protease inhibitors and duration of protease inhibitor therapy. Increases in blood serum triglycerides and C-peptide (a byproduct of insulin production) during the first year of antiviral therapy predicted the eventual severity of lipodystrophy. Glucose processing was on average little changed even in persons with severe lipodystrophy, but insulin production (and insulin resistance) increased sharply along with lipodystrophy. Factors that did not affect the results included CD4 count, viral load -- whether pretherapy or current -- and duration of HIV infection.
Presenters at the Retrovirus Conference also described several of the newer, larger and more rigorous studies of lipodystrophy's frequency. Results were similar to the Australian study. A French evaluation of 624 patients on protease inhibitors at Paris's Hôpital Rothschild found that 78% bore at least one physical sign of fat redistribution (S. Gharakhanian et al., poster 642). About 42% had a complete set of symptoms including atrophy in subcutaneous fat and swelling of truncal or dorsal fat masses.
The Centers for Disease Control's Adult and Adolescent Spectrum of Disease project was able to sift through enough data to present the experience of 745 of its 2,892 enrollees (M. Thompson et al., poster 649). The rates of triglyceride abnormality were 15.1% in those taking a protease inhibitor plus nucleoside analogs; 6.1% in those on a nonnucleoside reverse transcriptase inhibitor (NNRTI) and nucleoside analogs; 6.5% in those on just nucleoside analogs; and only 3.0% in those receiving no treatment at all. All these differences were statistically significant (except the one between nucleoside analogs only and no treatment).
In the CDC study, instances of high triglycerides were defined as the first time blood serum levels rose to above 500 mg/dl. Usually, triglyceride levels above 200 mg/dl of serum are considered excessive. The CDC investigators employed a much higher level to define abnormal triglyceride levels because blood samples had been taken without regard as to when the study participants had last eaten. Fasting samples (typically taken in the morning) are highly preferable, as recent meals will greatly affect the results of blood tests for lipids and glucose.
The Australian study did use fasting samples, but lack of such samples has been a major problem, as many studies are retrospective "look backs" planned once the lipodystrophy syndrome became recognized. Another analysis at the 6th Retrovirus Conference that did utilize blood samples taken while fasting was reported by a researcher from Montreal (J. Falutz, poster 646). Dr. Falutz checked the body composition and lipid levels of 86 men on stable HAART regimens. About half the men had viral loads below 400. Cholesterol levels were above the normal range in 33% of the group without lipodystrophy and in 52% of the group with just fat depletion. Seventy-five percent of the group with both fat depletion and abnormal fat accumulations had high cholesterol. For excessive triglycerides, the percentages were 26%, 65% and 69%, for the no lipodystrophy, fat depletion only and depletion plus accumulation groups, respectively. The differences in average lipids between the two lipodystrophy subgroups were not clear-cut. The existence of fat deposits in the second lipodystrophy stratum might reflect successful, if distorted, uptake of blood lipids by fat cells in the truncal and dorsal bulges. Without those deposits, blood lipids might have been much higher.
A similar small study in 33 women on protease inhibitors found that waist-to-hip ratios correlated with lipid and glucose levels in fasting blood samples (L. Bausserman et al., 6th Retrovirus Conference, poster 659). As with Dr. Falutz's report, average elevations were only moderately above normal. An association of this nature, observed after the fact at one point in time, is not proof of causality: The women with high pretreatment lipids and glucose may have been more susceptible to truncal gains and peripheral shrinkage. Since baseline measurements are unavailable, it is hard to say what the role of protease inhibitors is in upsetting the metabolic balance.
How Those Signs and Symptoms Arise
David A. Cooper, M.D., Andrew Carr, M.D., and colleagues from St. Vincent's Hospital in Sydney were among the first to report on altered body composition. At the 5th Conference on Retroviruses and Opportunistic Infections, held in February 1998, they reported that 64% of their patients on protease inhibitors had self-reported changes in body shape. The Australians published a provocative hypothesis in The Lancet (June 20, 1998, pages 1881-3), which was later presented at the World AIDS Conference in Geneva. According to their research, the region of the HIV protease enzyme to which PIs bind is structurally similar to regions on two human proteins (CRABP-1 and LRP) involved in the breakdown of fats or lipids. Inhibition of these two proteins may lead to increased death of fat cells in the extremities and impaired clearance of fats from the blood. This theory has many critics, but it does provide an explanation for fat loss in the limbs, abdominal girth, fat deposits, lipid increases, insulin resistance, premature coronary artery disease and possibly even the skin changes noted by many protease-treated patients.
Glaxo Wellcome's obesity/diabetes research unit has further investigated this hypothesis. Its scientists, led by James Lenhard, Ph.D., and Eric S. Furfine, Ph.D., presented an intriguing new hypothesis last winter at the 6th Retrovirus Conference (posters 665 and 666), zeroing in on the possible mechanism for fat redistribution, metabolic dysfunction, and insulin resistance. They believe ritonavir, nelfinavir and saquinavir may be causing symptoms by one particular mechanism, indinavir by another. Some of the mesenchymal stem cells in the body mature into fat cells, or adipocytes, via a process called adipogenesis. The Glaxo researchers found that ritonavir, saquinavir and nelfinavir interfered with adipogenesis, but indinavir did not. Saquinavir, nelfinavir and, to a lesser extent, ritonavir also slowed fat formation while increasing fat breakdown in mature adipocytes.
Another set of assays tested the influence on the production of 9-cis retinoic acid, a major part of the hypothesis advanced by Drs. Carr and Cooper. CRABP-1 works with the liver's cytochrome P450 3A4 enzyme, which itself is well known to be inhibited by protease inhibitors, to transform "all-trans" retinoic acid (a vitamin A derivative) to 9-cis retinoic acid. This molecule promotes the maintenance, proliferation and maturation of fat cells, but its influence is less in fat tissues around the abdominal organs and between the shoulder blades. The Glaxo team found that indinavir, but not the other protease inhibitors, blocked 9-cis retinoic acid production.
Disruptions in fat formation and storage could lead to any number of symptoms -- fat redistribution, increased blood lipids and insulin resistance. This is just a guess at this point, based on in vitro lab results. Glaxo is now investigating the exact mechanisms by which the protease inhibitors affect fat formation and retinoic acid signaling, by binding to CRABP-1 in the case of indinavir and to 9-cis retinoic acid receptors within fat cells in the case of the other three.
Donald P. Kotler, M.D., of St. Luke's/Roosevelt Hospital in New York City, a leading expert on HIV-wasting/metabolic disorders, believes the problem may be more complex and not necessarily directly related to PIs. Recognition of the lipodystrophy syndrome coincided with the introduction of protease inhibitors, but he thinks that researchers may have drawn premature conclusions. He had occasionally seen similar body composition and metabolic abnormalities in patients treated before the prevalent use of PIs. Additionally, some patients with non-HIV conditions manifest comparable symptoms.
In an analysis that Dr. Kotler conducted in 77 men and 19 women, CD4 cell count and type of therapy did not appear to influence fat distribution when other factors were taken into account (D.P. Kotler et al., Journal of Acquired Immune Deficiency Syndromes, March 1999, pages 228-37). Low viral load and female gender were the only independent correlations that turned up.
Dr. Kotler has noted that many of his patients with lipodystrophy have high levels of cortisol secretion. He points out that a syndrome similar to HIV-associated lipodystrophy occurs in other chronic infections. This so-called "Syndrome X" includes fat redistribution along with elevated blood levels of glucose, insulin and lipids. The condition seems to be a result of chronic stress and includes elevated cortisol secretion as well as reduced sensitivity to insulin. Dr. Kotler believes that protease inhibitors merely uncover this underlying phenomenon by drastically reducing viral load and permitting renewed immune system function.
It is true that several studies have found individuals who had signs of lipodystrophy without having taken protease inhibitors. DuPont's pivotal 006 trial that included 154 persons on AZT/3TC/efavirenz observed 25% increases in nonfasting cholesterol on this nucleoside analog/NNRTI regimen (S. Staszewski et al., 4th International Conference on Drug Therapy in HIV Infection, Glasgow; Nov. 8-12, 1998, oral presentation OP5.1). At the 6th Retrovirus Conference, Drs. Carr and Cooper mentioned that 4% of their patients who were not on protease inhibitors reported that they had experienced physical changes in their fat tissues. Other presentations at the conference correlated signs of lipodystrophy with d4T (T. Saint-Marc et al., 6th Retrovirus Conference, poster 653) or 3TC (C. Gervasoni et al., poster 660).
Considering these differing observations, Dr. Kotler has argued, "The syndrome, as currently viewed, may not be a single entity, and may represent a spectrum of abnormalities. HIV infection and its treatment may act in several ways to promote the observed changes."
One such additional mechanism was suggested by some of Drs. Carr and Cooper's observations. They noted that leptin levels in patients on protease inhibitors were a third of the levels in similar HIV-positive persons not on PIs. Fat cells unable to accumulate normal amounts of lipids reduce their production of leptin and other hormones as a way of asking the body to increase glucose as well as triglyceride levels in the blood, first of all by eating more. Normally, this hormonal signaling leads in the end to restored fat mass, but in persons with lipodystrophy, it would be futile. The decline in leptin would exert further upward pressure on blood lipid, sugar and insulin levels and exacerbate the lipodystrophy.
It may be just a matter of time before the prevalence of diabetes increases among people with glucose processing abnormalities. Similarly, high lipid levels are a major risk factor for eventual coronary heart disease. This condition starts out as a buildup of fatty plaque in the arteries. The plaque forms lesions on the walls of arteries that may ultimately rupture, leading to blood clotting and further blockage of the artery. A heart attack will ensue when the oxygen supply to a section of heart muscle declines to a critical point. Cholesterol complexed with proteins into low-density lipoprotein (LDL) particles for transport through the blood stream is considered the major source of this plaque. In contrast, high-density lipoprotein (HDL) cholesterol keeps the LDLs from collecting and is protective. (HDL's most studied function is to transport excess lipids back to the liver for degradation; this may or may not be the process at work here.) High triglyceride levels also seem to have a role in promoting coronary artery disease, but the extent of the increased risk such levels present is not yet well defined.
People on protease inhibitors generally have raised levels of LDL cholesterol and triglycerides while HDL cholesterol remains constant, resulting in a decreased HDL-to-LDL ratio. This ratio is the most commonly used standard to gauge heart attack risk. The DuPont 006 trial did note a 15% increase in HDL for the AZT/3TC/efavirenz combination along with a 25% increase in total cholesterol (blood samples were nonfasting). Here, too, then, there was a decrease in the HDL/LDL ratio, though perhaps not as great as observed with protease inhibitors.
A year ago, Dr. Keith Henry and his associates at Regions Hospital in St. Paul, Minnesota published a description of two patients with severe blockage of heart arteries (K. Henry et al., The Lancet, May 21, 1998, page 1328), and many other anecdotal descriptions have since appeared. But the incidence of heart attack remains low. Dr. Carl Grunfeld of the San Francisco Veterans Administration Hospital estimates that the lipid changes seen in patients on protease inhibitors at the VA clinic would lead to 1.41 additional cases of coronary artery disease per 100 patients over 10 years. This is not a large amount given the drastic fall in deaths from AIDS observed up to this point.
This calculation of increased risk from protease inhibitors did not take into account the other risk factors that persons with HIV may have. Risk factors are more than additive. They work together synergistically to multiply an individual's risk of heart disease. HIV infection by itself leads to decreased HDL and heart muscle damage. Cardiac disease was the primary cause of death in 9.1% of persons with HIV in the pre-protease inhibitor era according to one study (R.C. Patel and W.H. Frishman, Medical Clinics of North America, November 1996, pages 1493-512). Much of this mortality was due to unrecognized infections that weakened heart muscle, and there is suspicion that HIV directly damages the heart.
Heart problems will increase further as more people with HIV advance into their fifties. Besides HIV and aging, they may have other risk factors, including hyperglycemia and diabetes, family history of heart disease, sedentary lifestyle, and smoking. Use of testosterone, anabolic steroids and certain other medications that increase blood clotting also adds to the danger. All these elements work together synergistically to multiply an individual's chance of heart disease. (See Treatment Issues, May 1998)
Treatments -- Fight or Switch?
Another report by Dr. Keith Henry described treating people on HIV protease inhibitors according to the guidelines established by the National Cholesterol Education Program (6th Retrovirus Conference, poster 671). These guidelines were formulated to reduce lipid levels and heart attack risk in the general population. Twenty-two patients in his clinic with only moderately excessive lipid levels participated in a diet and exercise program. Another 22 received lipid-lowering drugs -- atorvastatin (Lipitor), gemfibrozil (Lopid) or the two together. The diet/exercise program was considered insufficient, although it did decrease average cholesterol levels by 17% and triglycerides by 26%. Those on lipid-lowering drugs experienced an average 40% drop in cholesterol levels while their triglyceride levels fell by 56%.
Several early reports suggest mild to moderate effects on lipids using an agent that improves sensitivity to insulin. Both metformin (Glucophage) and troglitazone (Rezulin, a drug now embroiled in controversy over its liver toxicities) reduced insulin resistance to some degree but had minimal change in the body shape. (For metformin, see T. Saint-Marc et al., 6th Retrovirus Conference, poster 672 and this issue for further followup. For troglitazone, see R.K. Walli et al., 6th Retrovirus Conference, poster 673.) The metformin trial recruited persons on protease inhibitors who had impaired glucose tolerance. All but one also had increased abdominal girth. Fourteen persons on metformin averaged a 60% decrease in insulin levels over eight weeks. An average 13% decrease in the visceral-to-total fat ratio accompanied this decrease in insulin production. A 13-person control arm remained essentially stable.
Another therapeutic strategy, administration of standard doses of recombinant human growth hormone (Serostim, HGH), was effective in reducing central adiposity and buffalo humps but not in reversing hyperlipidemia or peripheral lipodystrophy (R. Torres et al., 6th Retrovirus Conference, poster 675). Side effects of HGH included increased tissue turgor, joint pains, carpal tunnel syndrome and flu-like symptoms. Of 10 patients treated with HGH, eight of nine cases of truncal adiposity improved or resolved, as did five or six cases of buffalo hump. Relapse of buffalo humps and truncal adiposity occurred with dose reduction or permanent discontinuation of HGH.
But the obvious way to reduce lipid levels is to stop therapy, at least the protease inhibitor portion. In Dr. Henry's treatment study described above, 11 patients eventually did just that. Although most of the patients had had persistently high blood lipids despite lipid-lowering drug therapy, the group's mean lipid levels dropped by half during the first month off PIs and continued to decline slowly. The price was loss of control of HIV levels. Viral load went from a mean of 13,000 to 97,000 after four months followup.
This experience suggests another strategy, that of switching from protease inhibitors to NNRTIs, which would help keep HIV suppressed. Several studies have investigated this switch. Graeme Moyle and colleagues in London have presented early data from a 12-person study in which efavirenz replaced indinavir in antiviral combinations (6th Retrovirus Conference, poster 669). Participants had overt body shape changes and, with one exception, viral loads suppressed to below 500 copies/ml. After 12 and 24 weeks of therapy, those who switched to efavirenz had a significant increase in weight, but no notable changes in body composition. In addition, elevations in serum triglycerides and cholesterol accompanied the switch to efavirenz. HIV inhibition held constant for all study participants.
Three similar studies presented at the 6th Retrovirus Conference dealt with replacing a protease inhibitor with the NNRTI nevirapine (L. Ruiz et al., latebreaker oral presentation 14; A. Carr et al., poster 668; and E. Martinez et al., poster 670). Preliminary data for up to six months' observation were available. All three studies recorded reductions in lipid levels and a tendency toward more normal body shape. HIV suppression continued after the switch, except for 3 of 16 volunteers in one of the studies.
These early partial improvements suggest that lipodystrophy could be reversible, but it will take larger and longer trials to delineate the extent of recovery. If factors other than protease inhibitors contribute to lipodystrophy, improvement will never be more than partial without specific therapeutic interventions to correct the underlying metabolic or immune disorders.
Back to the GMHC Treatment Issues March 1999 contents page.