Published in the Bulletin of Experimental Treatments for AIDS 1999 Year-End issue, by the San Francisco AIDS Foundation.

Main Page

HAART and Experimental Approaches to Treatment

Anti-HIV Drug Resistance Testing

Anti-HIV Therapies Early in the Pipeline

Anti-HIV Therapy Trials: New Studies and Follow-Up of Existing Trials

New Anti-HIV Drug Interactions, Toxicities, and Dosing Options

HIV Transmission

OIs and Cancers

Year-End 1999 Table of Contents

Main Page

beta@sfaf.org

Conference Notes -- New Anti-HIV Drug Interactions, Toxicities, and Dosing Options

Harvey S. Bartnof, MD

Does changing a PI to an NNRTI alleviate body fat redistribution (BFR)?

A few abstracts at the 39th ICAAC addressed the issue of BFR. Since PIs were the first drugs to be linked with fat redistribution, some researchers wanted to determine whether viral load would increase if a PI was changed to an NNRTI. A second endpoint would be the potential effects on fat redistribution. Unfortunately, most of the studies presented have been too short to determine whether changes in fat redistribution occur. A potential confounding factor in these studies is that some of the NRTIs have also been linked with fat redistribution.

The longest study of this type was presented by Graeme J. Moyle, MD, from Chelsea and Westminster Hospital in London. Several of the participants indicated they wanted to change their anti-HIV therapy due to the presence of fat redistribution. Others wanted to change due to metabolic blood plasma changes (increased cholesterol, triglycerides, and/or glucose).

A total of 20 people (10% women) were enrolled in this observational study. All had been on a PI-based regimen for a mean of 19 months with an undetectable viral load (less than 50 copies/mL). The initial regimen for most was indinavir plus d4T and 3TC. Seventy percent had had a prior AIDS-defining opportunistic infection (OI). Almost all had never taken an NNRTI previously. Body fat was measured by DEXA (dual energy X-ray absorptiometry) and CT (computerized tomography) scans.

All participants were changed to an efavirenz-based regimen. While most maintained their double NRTI "backbone," 25% added abacavir (Ziagen) if they had been on a double PI regimen or had minimal prior NNRTI exposure. Among the 20 people who completed 24 weeks of the efavirenz-based combination, 95% maintained a viral load of 50 copies/mL. The median CD4 cell count increase was 31 cells/mm³. One person (5%) developed recurrent Pneumocystis carinii pneumonia (PCP), a life-threatening OI.

Interestingly, five of six people who had an abnormal glucose tolerance test (signifying diabetes or borderline diabetes) with their PI regimen had a normal test with the efavirenz-based regimen. Regarding triglycerides (fats) after 24 weeks, three maintained a normal level, while 12 displayed an abnormal level. Two who had a high baseline level reverted to a normal level at 24 weeks. Similarly, two who had a normal level developed a high level at 24 weeks. Regarding cholesterol levels, six maintained an abnormal level, while seven displayed a high level. Four who had a normal level at baseline developed a high level at 24 weeks. Similarly, two who had a high baseline level reverted to normal at 24 weeks. Neither the triglyceride nor cholesterol results were statistically significant.

Dr. Moyle also reported on nine people (11% women) who were taking the new NNRTI-based regimen for 48 weeks. At baseline (on the PI regimen), six (66%) had fat loss in the legs. Five (56%) had fat loss in the face or arms. Increased visceral (within the abdomen) fat with an increased waistline occurred in five people (56%). Only one person (11%) had all these fat redistribution changes. Two-thirds had an abnormal glucose tolerance test.

One of the nine changed efavirenz to nevirapine due to side effects. All maintained a viral load below 50 copies/mL. The median CD4 cell count increase was 114 cells/mm³. Trends in blood lipids (fats) were similar at 48 weeks when compared with 24 weeks. Triglycerides decreased somewhat, however.

The nine participants gained a mean of 4.5 kilograms (weight in pounds x 0.454 = weight in kilograms). Lean body mass increased by 1.3 kilograms. Similarly, there was a 2.5 kilogram increase in total body fat. Visceral abdominal fat decreased significantly by a mean of 12%. Fat in the arms increased by 4%, whereas fat in the legs decreased by 3%. Some, but not all, people had an improvement in the appearance of their fat redistribution (mostly due to improvements in fat accumulation with few changes in the appearance of fat loss). Food intake and overall well-being improved. In the two people who changed d4T to abacavir, blood triglycerides decreased. These people will be followed for a longer period of time to determine long-term effects.

In an interim analysis of an efavirenz 006 substudy, there appears to be a low rate of fat redistribution when efavirenz is taken with AZT and 3TC. The analysis reviewed the charts of half of those who reached 24 weeks, one-third of those who reached 48 weeks, and 13% of those who reached 72 weeks. "Probable fat redistribution" occurred in 0.7% of the triple efavirenz arm, 2.6% of the efavirenz plus indinavir arm, and 1.4% of the indinavir plus AZT and 3TC arm.

There are two limitations in this investigation. First, it is merely an interim analysis, with few people in the 72-week analysis. Second, a chart review for the terms linked with fat redistribution is likely to underestimate the true rate. Measuring the changes by DEXA, CT, and photographs (as was done in the first report above) would yield more accurate data. The 006 substudy report was authored by K. Tashima, MD, from Miriam Hospital in Providence, RI.

• Men are more likely to experience fat loss under the skin (80%) than women (73%)
• Women are more likely to experience fat accumulation (98%) than men (76%)
• Men are more likely to have increased blood fats (triglycerides 84%, cholesterol 53%) than women (triglycerides 32%, cholesterol 28%)
• Overweight people are more likely to experience fat accumulation, whereas underweight people are more likely to experience fat loss under the skin
• The risk of fat redistribution increases with increasing age
• The risk of fat redistribution increases with increasing duration of HIV infection
  

In several studies presented at the 1st International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV, demographic trends emerged regarding the risk of fat redistribution and metabolic changes. There appear to be some gender, age, and pretherapy body weight factors that influence the types of fat redistribution that occur. Some of these characteristics were described in two separate poster presentations by the Community Research Initiative on AIDS in New York City.

One report by Norma Muurahainen, PhD, and colleagues concerned the Self-Ascertained Lipodystrophy Syndrome Assessment (SALSA) cohort. A total of 270 HIV positive people (24% women) with some form of fat redistribution were enrolled in the study. HIV positive persons who were taking HAART comprised 78% (210) of the enrollees, 98 of whom were taking a PI-based regimen. Among those taking HAART, 51% had an undetectable viral load (limit of detection 500 copies/mL). HIV negative, age-matched controls were also included. Participants completed the SALSA questionnaire and their physicians were asked to record body shape changes as well as blood lipid and glucose abnormalities.

There were high rates of fat redistribution in both men and women, but the rates were higher among women than men. Abdominal fat accumulation ("protease paunch") was reported in 98% of women and 76% of men. Fat accumulation on the lower back of the neck (interscapular fat pad, or "buffalo hump") was also more common in women (46%) compared with men (37%). In addition, fat accumulation in the breasts was also more common in women (74%) than in men (31%). True female-type breasts are very rare among men, but there have been case reports. (Note that statistical analyses for all of these percentages were not reported. Further analyses are ongoing.)

Regarding fat loss under the skin (lipoatrophy), the opposite gender effect occurred. Even though rates were high for both men and women, those for men were higher. For example, fat loss in the arms and legs was higher among men (68%) than among women (54%). Facial fat loss occurred in 57% of men, compared with only 22% of women. Also, fat loss in the buttocks was reported in 59% of men, but only 44% of women.

There were also gender differences regarding the most common areas of body fat redistribution. The most commonly reported types of redistribution among men were fat accumulation in the abdomen and fat loss in the arms and legs. The most commonly reported types of redistribution among women were fat accumulation in both the breasts and abdomen. All of these corresponding rates of BFR were higher than those in age-matched, gender-matched, HIV negative controls.

Blood fat abnormalities were more common among men than among women. High triglyceride levels occurred in 84% of men, compared with only 32% of women. Similarly, high cholesterol levels were detected among 53% of men, compared with only 28% of women. High blood sugar rates were low among both genders, but were higher nonetheless among men (15%) than among women (6%). As with fat redistribution, these metabolic changes were less common in age-matched, gender-matched controls. The authors commented that a proposed case definition of fat redistribution and metabolic abnormalities must take into account gender differences.

Similar gender differences in fat redistribution and blood lipid levels were also noted in a cohort study by J. Falutz and colleagues from Montreal. Another report by J. Galli and colleagues from Milan, Italy, also found higher rates of fat redistribution among women than among men.

In a separate analysis from the SALSA cohort, various fat redistribution locations were analyzed by body mass index (or BMI, which equals weight divided by height squared). A body mass index greater than 28 kg/m² is considered overweight, while a BMI less than 21 kg/m² is considered underweight. For men, being overweight was associated with a higher percentage having fat accumulation in the belly (88%), compared with those who were underweight who had the same accumulation (57%). The opposite trend occurred when examining fat loss. For men, being underweight was associated with a higher percentage having fat loss in the face (57%), compared with those who were overweight who had facial fat loss (20%). No statistical analyses were reported.

Similar trends were noted among women. For women, being overweight was associated with a higher percentage having breast fat accumulation (83%), compared with those who were underweight who had the same change (29%). Fat accumulation in the abdomen was very high (100%) among all women sampled, including both underweight and overweight women. As with men, an opposite trend occurred when examining fat loss among women. In underweight women, a higher percentage had facial fat loss (43%), compared with overweight women who had facial fat loss (9%). The authors concluded that any proposed case definition for fat redistribution must take into account body mass index.

Several cohort studies-including two Australian cohorts, the French Aquitane cohort, and the SALSA cohort from New York-found there is an heightened risk of body fat redistribution as age increases. Several cohort studies also found that the risk increases according to duration of anti-HIV therapy, but not necessarily with all anti-HIV drugs. The U.S. HIV Outpatient Study (HOPS) found a significantly higher percentage of HIV positive hemophiliacs having moderate or severe fat redistribution (46%) compared with other HIV positive groups who acquired their infection by another route (19%). Age may be a potential factor in the analysis, however, since virtually all hemophiliacs with HIV who are still alive acquired their HIV infection before 1985. Hence, the statistic could be due to older age.

These various trends must be confirmed in other cohort groups. It is likely that such reports will be presented during the next 12-18 months at various medical conferences.

A total of 25 women (all Latina except four African-Americans) from New York were enrolled in this study presented at the 39th ICAAC by G. Santos, MD, and colleagues from Staten Island University Hospital in New York. Increased waist size was reported by 96%, while 28% reported increased breast size. Eight percent had a "buffalo hump." Though they were started on growth hormone injections at a dose of 4-6 mg injected daily, this dose had to be reduced due to side effects (e.g., joint aches, swelling). At a dose of 2-4 mg injected every other day, 56% had a decrease in their waist size; the maximum reduction was four inches. Breast size did not decrease, however. When growth hormone was stopped, waist size began to increase again. Restarting therapy led to a reduced waist size again. While using growth hormone, these women had no clear patterns regarding changes in blood fat (cholesterol and triglycerides) or sugar levels.

G.M. Levin, MD, from the University of Florida at Gainsville presented this report at the 39th ICAAC. Venlafaxine is an antidepressant that partially works in the same way as selective serotonin reuptake inhibitors, or SSRIs (e.g., Paxil, Prozac, Zoloft). Venlafaxine caused indinavir concentrations to decrease by 28-36%, which increases the risk of developing indinavir resistance. The authors concluded that the two drugs should not be used at the same time until the necessary increase in indinavir dosing is known. Levels of venlafaxine were not affected by indinavir. These findings also raise the specter that other antidepressants could influence PI blood levels. This underscores the need to have drug-drug interaction studies completed, since depression requiring medication is very common in people with HIV infection.

This report was presented at the 39th ICAAC by E. Sellers, MD, from Sunnybrook and Women's College Hospital in Toronto. In those taking methadone and the NRTI abacavir, metabolism ("clearance") of methadone increased such that a higher dose would be needed; the dose of abacavir was not affected. Some people may experience opiate withdrawal symptoms if abacavir is added to their regimen without increasing the methadone dose.

Ritonavir, a very potent PI, is associated with a high discontinuation rate due to side effects that are maximal at the full dosage. These symptoms are usually related to the stomach and intestines and are common side effects of most PIs. They include gastrointestinal pain, nausea, vomiting, and diarrhea. Researchers from the University of Genoa, Italy, attempted to determine whether there are any gender differences in ritonavir intolerance.

A total of 93 HIV positive people (40% women) at a rather advanced stage of HIV disease was enrolled in the study. Seventy-seven percent had a history of injection drug use. Median baseline viral load was 4.8 log copies/mL, while median CD4 cell count was 91 cells/mm³. Thirty-seven percent had a prior AIDS-defining illness, and 59% were coinfected with hepatitis C virus (HCV).

Ritonavir intolerance due to side effects was significantly more common in women (66%) than in men (27%). Intolerance was due to gastrointestinal symptoms or neurologic (nerve tissue) side effects. Common neurologic side effects were headaches, tingling or numbness, dizziness, and problems both with cognition and with sleeping. Median time of onset of side effects was one month after starting therapy. The only other factor associated with ritonavir intolerance was advanced HIV disease stage. Each of the following did not have a statistical association with intolerance: HCV coinfection, age, weight, CD4 cell count, HIV viral load, and time elapsed after an AIDS diagnosis.

In a substudy analysis of 21 participants, women who experienced intolerance had higher blood levels of ritonavir. This adds to the weight of evidence that favors measuring blood levels of anti-HIV drugs. Similar reports have shown an association between high blood levels of indinavir and side effects. Other studies have indicated an association between blood levels of anti-HIV drugs and HIV viral load suppression.

In a study presented at the 1998 ICAAC held in San Diego, K. Melbourne, MD, and colleagues showed that each of the following factors is also associated with ritonavir intolerance: low body fat percentage, micronutrient (vitamin and mineral) deficiencies, lack of micronutrient supplementation, and therapy with anabolic (muscle-building) steroids. Risk factors for ritonavir intolerance may overlap in these two studies. For example, people at an advanced stage of HIV disease are more likely to have low body fat and overall poorer nutritional status, including lower levels of micronutrients.

Delavirdine (Rescriptor) is associated with low rate of liver toxicity

M. Para, MD, and colleagues from Ohio State University presented this study at the 39th ICAAC. There is a very low rate of liver toxicity due to delavirdine, even among HIV positive people coinfected with HBV or HCV. The rate of noninfectious hepatitis in 5,330 HIV positive people who took delavirdine in clinical trials was compared with the rate in 1,690 controls in the same eight trials. The rate was 0.24% in those who took delavirdine, compared with a rate of 0.29% in the control arms. Liver failure occurred in 0.11% of those taking delavirdine, compared with 0.12% in the control arms. There also were no differences in severe or life-threatening increases in liver enzymes (ALT and AST).

According to H.J. Mei, MD, and colleagues from the State University of New York (SUNY) in Buffalo, indinavir can safely be taken one hour after ddI. Blood levels of both drugs remain unchanged even though the buffering (alkaline) effects of ddI persisted in the stomach for 90 minutes. Many anti-HIV drugs cannot be taken at the same time as ddI due to its buffering effects.

Rifabutin (an anti-Mycobacterium avium complex [MAC] drug) and nevirapine can be taken at the same time without any dose adjustment for either. Nineteen HIV positive people were enrolled in a study authored by S. Maldonado, MD, from Boehringer Ingelheim Pharmaceuticals and presented at the 39th ICAAC.

According to L. Veronese, MD, from Glaxo Wellcome, the presence of liver cirrhosis (scarring) means dosing of amprenavir must be adjusted. To equal the standard 1,200 mg dose of amprenavir (taken every 12 hours) in HIV positive people without cirrhosis, the following doses are recommended. Among HIV positive people with severe cirrhosis (Child-Pugh score of 9-12), the dose should be 300 mg twice daily. For those with moderate cirrhosis (Child-Pugh score of 5-8), the dose should be 450 mg twice daily. The abstract did not indicate the causes of cirrhosis (e.g., HBV, HCV, alcohol) in the 20 people tested. The study was presented at the 39th ICAAC.

In a report given by J.R. Baldwin, MD, of Pharmacia & Upjohn at the 39th ICAAC, twice-daily dosing of tipranavir will be necessary if combined with ritonavir. Tipranavir (an experimental PI) displays activity against HIV strains resistant to currently marketed PIs. A smaller, more concentrated formulation of tipranavir is in development.

Page last updated 8 January 2000