Published in the Bulletin of Experimental Treatments for AIDS Winter 2001 issue, by the San Francisco AIDS Foundation.

Winter 2001 Table of Contents

Main Page

beta@sfaf.org

Conference Coverage

New Research Findings

Osteopenia, Osteoporosis Found in Antiretroviral-Naïve HIV Positive Persons

At the 7th Conference on Retroviruses and Opportunistic Infections (CROI), held in San Francisco on January 30-February 2, 2000, the results of two studies suggested that loss of bone minerals (osteopenia, osteoporosis) was associated with protease inhibitor (PI) drug therapy. The reports concluded that it was very common in HIV positive persons taking highly active antiretroviral therapy (HAART). Now, researchers from the Hôpital Hôtel-Dieu in Nantes, France, have reported that bone mineral loss also occurs in HIV positive persons who are not receiving treatment for HIV. E. Billaud, MD, the lead author also reported that bone mineral loss was associated with the total time of HIV infection and not with the presence of so-called lipodystrophy (fat redistribution). Severe bone mineral loss increases the risk of bone fractures. Outside of HIV/AIDS, bone mineral loss commonly occurs in elderly persons and represents a significant cause of bone fractures.

In the study, a total of 85 HIV positive subjects were included, with 18% women and 3% African American. Bone mineral density was measured by DEXA (dual energy X-ray absorptiometry) testing. Definitions for osteopenia and osteoporosis (severe loss of bone mineral density or concentration) were similar to those used by the World Health Organization (WHO). Among the 15 subjects who never took any drugs to treat HIV, 20% had osteopenia. Among the subjects taking combination therapy to treat HIV, osteopenia was present in 40% of those taking a PI drug and 45% of those not taking a PI drug. Severe bone mineral loss (osteoporosis) was present in none of those who had not taken anti-HIV therapy and in 7% of those who were. In a univariate statistical analysis, factors associated with bone mineral loss included total length of HIV infection, daily calcium intake, and osteocalcine levels (a marker of bone formation).

The researchers concluded that "osteopenia and osteoporosis are more frequent in patients with prolonged duration of HIV infection with no clear association with use of potent ART [antiretroviral therapy]." These results will need to be confirmed in larger studies. Regardless of the cause (HIV and/or ART), HIV positive persons may ultimately need specific treatment for bone mineral loss. Limitations of the study include a small number of participants, other factors among the women that might have contributed to bone mineral loss, and no multivariate statistical analysis.

References

Billaud E and others. Osteopenia and osteoporosis in HIV-infected patients: role of antiretroviral therapy? Abstract and poster 1304.

Hoy J and others. Osteopenia in a randomized, multicenter study of protease inhibitor substitution in patients with the lipodystrophy syndrome and well-controlled HIV viremia. Abstract 208 at the 7th Conference on Retroviruses and Opportunistic Infections; January 30-February 2, 2000; San Francisco, California.

Tebas P and others. Accelerated bone mineral loss in HIV-infected patients receiving potent antiretroviral therapy. Abstract 207 at the 7th Conference on Retroviruses and Opportunistic Infections; January 30-February 2, 2000; San Francisco, California.

Low-Dose Ritonavir (Norvir): Does It Cause PI Drug Resistance?

Many persons with HIV infection are taking low-dose ritonavir to boost the blood level of a second PI drug as a part of HAART. One concern physicians have had about using low-dose ritonavir is whether this could possibly lead to developing ritonavir resistance mutations. Such mutations, it is feared, may lead to cross-resistance to the second PI drug, and eventual HIV viral breakthrough (rebound). A significant concern would be if the second PI drug were indinavir (Crixivan), since ritonavir and indinavir share genotypic resistance mutations. The most common argument against this concern is that if complete viral suppression (undetectable viral load) is achieved through successful HAART, HIV would not reproduce and therefore resistance mutations would not be an issue. However, Diane Havlir, MD, of the University of California at San Diego and others have reported that viral blips (transient, often very small, increases in viral load) commonly occur, even with an undetectable viral load due to HAART. Experts do not know whether all HAART medications penetrate every compartment and every cell in the body (assuming 100% adherence to an anti-HIV drug regimen).

S. Chaillou, MD, from Nice University Hospital in France has now reported that ritonavir resistance mutations have developed in HIV positive persons taking low-dose ritonavir with saquinavir (Fortovase) combination "salvage" therapy. Thirty-four participants of the VIRADAPT study were switched to low-dose ritonavir (100 mg twice daily) plus saquinavir combination therapy. Some participants developed genotype resistance to ritonavir during the next 12 months; the V82A/F/T ritonavir mutation developed in 23% and the M46I/L ritonavir mutation developed in 15%. The researchers also measured the lowest (trough) blood levels of ritonavir and found a median (average) concentration of 0.47 micrograms per milliliter. This is not far from the EC50 (effective concentration that inhibits 50% of wild-type HIV, without mutations) of ritonavir. The authors conclude that low-dose ritonavir could be a threat in [treatment-]naïve patients as it could select ritonavir/indinavir resistance associated mutations." And, those "could be added to the specific resistance associated mutations of the boosted PI [drug]." Potential confounding factors in the study include other baseline PI drug mutations (all had a previous PI drug regimen that led to viral rebound), dosing adherence and the long-term significance (long-term viral load suppression). Until the findings and significance of this report are confirmed (or not) by other studies, no one should stop or change their anti-HIV medications without first talking with a physician.

References

Chaillou S and others. Does ritonavir "baby dose" induce specific mutations in salvage combination therapy? A VIRADAPT sub-study. Abstract and poster 1267.

HIV Clinical Studies

Substitution Studies

There were several "switch" or drug substitution studies reported at the 40th ICAAC and at the 2nd Annual Workshop on Adverse Drug Reactions. Most of them involved substituting a PI drug with a non-nucleoside reverse transcriptase inhibitor (NNRTI) drug or the NRTI drug abacavir (Ziagen). Reasons for switching varied between studies and are described in each report. In general, the following trends were reported:

1. Some metabolic changes (blood cholesterol,
   triglycerides [fats]) tended to improve
   
   
2. Fat redistribution was not usually improved
   
   
3. HIV RNA viral load remained undetectable for the vast majority (a few subjects had rebound, as did a few in the control arms that continued their PI drug and did not switch) and in one study remained undetectable for a higher percentage of subjects who switched than those who did not switch.
   
   
4. CD4 cell counts continued to increase when the viral load remained undetectable.

Those who had had prior experience with a regimen containing only one or two NRTI drugs did not fare as well when switching to abacavir and therefore switching to a triple NRTI combination.

The interim 24-week results of Study DMP 266-049 were presented by Anita Rachlis, MD, of Sunnybrook and Women's College Health Sciences Centre in Toronto, Canada. In this trial efavirenz (Sustiva, an NNRTI drug) was substituted for one to two PI drugs in a triple regimen with two NRTI drugs. At baseline, all participants were required to have an undetectable viral load (with a lower limit of 50 copies/mL). A total of 226 subjects (6% had a history of injection drug use, 10% women, 20% African American, and 8% Latino) were randomized to the switch arm. The mean baseline CD4 cell count was 578 cells/mm3. The mean duration of the prior PI drug regimen was 20 months. The prior PI drug was indinavir in 44% and nelfinavir (Viracept) in 30%, and 12% had taken two PI drugs in the prior regimen. The most common dual NRTI drug combinations (used by 90% of subjects) were 3TC (Epivir) with either AZT (Retrovir) or d4T (Zerit).

Results after 24 weeks showed that a significantly greater percentage maintained viral undetectability (89%) than those who remained on their PI drug regimen (81%) (on a strict "noncompleter equals failure" analysis, indicating that all enrolled participants are included). Also, the time to virologic rebound was significantly longer in the switch arm than in the no-switch arm. There were no significant differences in CD4 cell count increases, approximately 40 cells/mm3 in both arms. The only significant difference in blood cholesterol between the two arms was a mild increase in the HDL ("good" or high density lipoprotein) cholesterol in the switch group. However, those measurements were not taken after fasting for at least 12 hours. The only significant change in liver enzymes was an increase in the GGT (gammaglutamyl transpeptidase) in the switch arm. Adverse events (moderate or worse) were higher in the switch arm (30%) than in the no-switch arm (17%). Most of those were known side effects due to efavirenz, including CNS (central nervous system or brain) side effects and rash. Potential changes in fat redistribution were not reported. Adherence questionnaires completed by participants revealed a significantly lower percentage who reported missed doses at several clinic visits in the switch arm (8%) than in the no-switch arm (20%). Discontinuation rates were slightly lower in the switch arm (7%) than in the no-switch arm (12%). The authors concluded that substituting efavirenz for one to two PI drugs in combination maintains viral undetectability in a significantly greater proportion than those who continue their PI drug-based regimen. Also, CD4 cell counts continue to increase, and the improved benefits are likely associated with improved adherence.

References

Rachlis A and others. Successful substitution of protease inhibitors with Sustiva (efavirenz) in patients with undetectable plasma HIV-1 RNA levels: results of a prospective, randomized, multicenter, open-label study (DMP 266-049). Abstract and presentation 473.

The interim results of a similar study to DMP 266-049 above were presented by Bonaventura Clotet, MD, of Germans Trias i Pujol Hospital in Barcelona, Spain. However, in this study, participants were randomized to one of three arms: (1) switch to nevirapine (Viramune, an NNRTI drug); (2) switch to efavirenz (Sustiva, an NNRTI drug); or (3) maintain the PI drug-based regimen. All 77 subjects (29% women) had an undetectable viral load (with a lower limit of 80 copies/mL) for at least nine months. None had previously taken an NNRTI drug. The mean baseline CD4 cell count was rather similar in all three arms, 595-660 cells/mm3. The mean time on prior anti-HIV therapy was just over five years; one-quarter had taken HAART as their first regimen. The results after nine months showed the following. Viral rebound occurred at the same low rate: one or two participants in each arm. CD4 cell counts continued to increase in all arms.

Adverse events occurred among two to three subjects in each arm due to known side effects of the drugs used. Arm 1 (switched to nevirapine) had a significant decrease in the total cholesterol level from an abnormally high level at baseline and had a significant increase in liver enzymes (ALT and GGT). However, the other two study arms also had mild increases in those liver enzymes (only an increase in the GGT was significant in the efavirenz arm). Increases in liver enzymes occured among those with hepatitis C virus (HCV) coinfection. Total cholesterol, LDL (low-density lipoprotein or "bad") cholesterol and triglycerides (fats) decreased significantly in the nevirapine arm after nine months. Switch medications were stopped in the nevirapine arm in two participants (due to liver enzymes or rash) and in the efavirenz arm in three participants due to CNS side effects that occurred among 32%. Fat redistribution was unchanged, and was measured by DEXA (dual energy x-ray absorptiometry). Quality-of-life questionnaires indicated a significant improvement in both switched arms after nine months.

The conclusions of this small study indicate that viral undetectability is maintained in nearly all participants, but there is a somewhat different side effect profile when comparing persons who switch to either nevirapine or efavirenz with those who remain on a PI drug regimen. Dr. Clotet said that monthly testing of liver enzymes among persons starting nevirapine "is required mainly in HCV-coinfected patients.

References

Negredo E, and others. Impact of switching from protease inhibitors to nevirapine or efavirenz in patients with viral suppression. Abstract and presentation 473.

During the same session, a similar study, called the Maintavir Study, was presented by Francois Raffi, MD, of University Hospital in Nantes, France. A total of 73 subjects (21% women) switched to either nevirapine (86%) or efavirenz (14%) after having an undetectable viral load (with a lower limit of 400 copies/mL) for at least six months while taking a PI drug-based regimen. NRTI drugs were continued. None had ever taken an NNRTI drug. The desire to switch was due to a number of reasons, but simplifying the regimen was the most common. The median baseline CD4 cell count was 473 cells/mm3 with a median 22 months of prior PI drug therapy. Ninety-five percent had a baseline viral load below 50 copies/mL. There were 20 months of follow-up results. Virologic rebound occurred among 14% of the nevirapine arm and 10% of the efavirenz arm, a nonsignificant difference. The rate of virological rebound was also reported in relation to the participants' history of anti-HIV drug therapy prior to the HAART regimen taken before entry into the current study. For those with no prior therapy, viral rebound occurred among 7% (58% of the 73 participants), compared with 19% among those with prior anti-HIV therapy (42% of all subjects). All ten participants with viral rebound later became undetectable after switching back to a PI drug regimen within six weeks. CD4 cell counts continued to increase in both arms. Adverse events at six months showed a significant decrease in blood triglycerides and no significant change in cholesterol, unlike the results reported by Dr. Clotet above. Fat redistribution improved in some by subjective assessment. Discontinuations occurred among 7%.

References

Raffi F and others. The Maintavir Study, substitution of a non-nucleoside reverse transcriptase inhibitor (NNRTI) for a protease inhibitor (PI) in patients with undetectable plasma HIV-1 RNA: 18 months follow-up. Abstract and presentation 474.

Switching from a PI drug-based HAART to an abacavir-based triple NRTI drug regimen was the strategy in two presentations. The first was presented by Milos Opravil, MD, of University Hospital in Zurich, Switzerland. In that study, the switch arm took abacavir plus AZT and 3TC as the one-pill formulation known as Combivir. When the new Trizivir formulation was available (providing all three drugs in one pill), it was substituted. Trizivir represents a regimen of one pill every twelve hours. A total of 163 subjects (20% women) were randomized to abacavir/AZT/3TC or to continue their current PI-based regimen. Requirements for entering the study included an undetectable HIV viral load (with a lower limit of 50 copies/mL) for at least six months and not having the AZT 215 gene mutation in DNA of blood immune cells. The mean baseline CD4 cell count was 512 cells/mm3. The mean duration of the previous PI-based HAART regimen at baseline was 20 months. Pre-HAART treatment with one or two NRTI drugs had occurred among 46% of study subjects.

The median (average) follow-up was 68 weeks. Viral rebound occurred among 15% of the swtch arm, compared with 6% of the no-switch arm (in a strict intent-to-treat analysis, i.e., in which all study participants were included). However, the total "treatment failure" rate was slightly higher in the no-switch arm (29%) than in the switch arm (25%). This opposite outcome was due to a higher rate of changing drug(s) due to adverse events and a higher rate of participants lost to follow-up in the no-switch arm. In a separate analysis, Dr. Opravil reported that viral rebound was five times more likely to occur in either treatment arm if there had been exposure to AZT before HAART. This finding was statistically significant for the switch arm. Virologic rebound also was associated with one- or two-drug NRTI treatment before HAART. There was a significant decrease in the blood cholesterol in the switch arm. Virologic rebound in the switch arm still allowed for constructing a new regimen with an NNRTI or PI drug(s) in combination therapy. The authors concluded that "simplified maintenance therapy with abacavir/Combivir (Trizivir) is an effective option if prior treatment history indicates [an] absence of archived [saved from the past] resistance mutations [to NRTI drugs] and adherence is maintained." The study also suggests that this triple NRTI combination might be less effective among those subjects with prior one- or two-drug NRTI treatment regimens.

References

Opravil M and others. Simplified maintenance therapy with abacavir + lamivudine + zidovudine in patients with HAART-induced long-term suppression of HIV-1 RNA: final results. Abstract and presentation 476.

Julio Montaner, MD, of St. Paul's Hospital in Vancouver, Canada, presented the results of a similar study (CNA 30017). Due to a much lower rate of prior NRTI drug experience at baseline (9%), this study had different results than the one above. In this larger study of 211 persons (approximately 18% women), participants must have had an undetectable viral load (with a lower limit of 50 copies/mL) for at least six months on a triple regimen with a PI plus two NRTI drugs. Participants were randomized to continue their regimen or switch the PI drug to abacavir in a simplified regimen. Note that in this study, the switch arm could have continued with any two NRTI drugs, not necessarily AZT plus 3TC as Combivir that occurred in the above study. The median CD4 cell count was approximately 505 cells/mm3. The follow-up period in this study was somewhat shorter at 48 weeks.

The results revealed that 11% in both arms had virologic rebound at 48 weeks using a lower limit of 50 copies/mL (in a strict intent-to-treat analysis). Using a lower limit of 400 copies/mL, 4% in the switch arm and 2% in the no-switch arm had virologic rebound. Among those with rebound in the switch arm who did not have drug resistance at baseline, two of three study subjects developed only one mutation, allowing for a wide range of options for the next drug regimen. However, treatment failure (discontinuation due to any reason) was twice as high in the no-switch arm (26%) than in the switch arm (13%), a statistically significant difference. Most of the difference was due to discontinuation resulting from adverse events in the no-switch arm. The switch arm also had a significantly longer time until treatment failure than the no-switch arm. Serious adverse events were nearly equal in both arms, 10% in the switch and 12% in the no-switch arms. There was a significantly greater decrease in blood cholesterol in the switch arm than in the no-switch arm. CD4 cell counts increased mildly in both arms.

These two studies about switching to an abacavir, triple NRTI regimen indicate significant potency with this simplified regimen. However, this may not be the best regimen for those with prior single or dual NRTI regimens due to a higher rate of virologic rebound. The impending availability Trizivir with all three drugs provided in a regimen of only one pill twice daily should improve adherence and this may increase the likelihood of maintaining virologic suppression.

References

Montaner J and others. A novel use of abacavir to simplify therapy and reduce toxicity in PI experienced patients successfully treated with HAART: 48-week results (CNA30017). Abstract and presentation 477.

Three-Drug Versus Four-Drug HAART

An interesting presentation about the number of anti-HIV drugs in a regimen was authored by W. Jeffrey Fessel, MD, of Kaiser Foundation Hospital in San Francisco. In that study, 53 subjects had a stable, undetectable viral load (with a lower limit of 50 copies/mL) with a triple drug regimen of a PI and two NRTI drugs. Participants were then randomized to continue the regimen or add efavirenz. The two groups were comparable in baseline characteristics. After follow-up of 20 weeks, 17% in the three-drug arm and none in the four-drug arm had viral rebound to greater than 50 copies/mL, a significant difference. There were another five subjects with a baseline viral load greater than 50 but less than 500 copies/mL that were a part of the arm that maintained three drugs. Two of them decreased their viral load to less than 50 copies/mL while the other three maintained a level greater than 50 copies/mL. There was a 14% discontinuation rate in the four-drug arm, due to efavirenz side effects. While adding efavirenz did show virologic benefits, there are other considerations. First is the fact that the four-drug arm included all three-drug classes which limits future options if virologic rebound occurs. Second, as Dr. Fessel points out, the potential negative long-term adverse events associated with four drugs must be weighed against the potential increase in sustained virologic benefits.

References

Fessel WJ and others. Four drugs are better than three drugs to maintain existing HIV suppression and reduce productive infection. Abstract 535.

Comparing NNRTI-Based with PI-Based Regimens and Contrasting Different NNRTI-Based Regimens

At the 40th ICAAC, several presentations compared drug regimens in randomized studies with HIV positive subjects who had not had previous anti-HIV treatment. In the Spanish SENC trial, nevirapine was compared with efavirenz when combined with the two NRTI drugs ddI (Videx) and d4T. The small study of 54 HIV positive persons (approximately 25% women) was randomized and prospective (planned beforehand). The two study arms were comparable at baseline. The median baseline HIV RNA viral load was approximately 4.3 log (21,000) copies/mL and the CD4 cell count was approximately 360 cells/mm3. Approximately one out of three was HCV positive.

After a median follow-up of nine months, the following interim results were presented. Outcome measurements were generally similar in both arms. An undetectable viral load (with a lower limit of 50 copies/mL) was achieved by 79% of the nevirapine arm and 85% of the efavirenz arm (in a strict intent-to-treat analysis). Undetectability was achieved at the same rate in both arms. Rash, a common side effect in the NNRTI drug class occurred among 17% of the nevirapine arm and 12% of the efavirenz arm, however no one discontinued due to rash. Increases in liver enzymes (ALT, AST) were generally mild and were more common in the nevirapine arm (59%) than in the efavirenz arm (35%); although, the difference was not statistically different. Liver toxicity was significantly associated with HCV positivity. The total discontinuation rate was 22%, although only two subjects did so due to adverse events related to study drugs (both efavirenz).

The lead author, M. Nunez, MD, concluded that there was a similar virologic outcome when ddI and d4T are combined with either nevirapine or efavirenz. A much larger study called 2NN with 1,200 participants has begun and will compare the two NNRTI drugs in combination therapy. The four-arm study will have a NRTI backbone of d4T plus 3TC. The other drugs will be: (1) nevirapine once daily; (2) nevirapine twice daily; (3) efavirenz once daily; or (4) efavirenz plus nevirapine.

In the Combine Study, the results of a nevirapine-based regimen were compared with a nelfinavir-based regimen in a randomized fashion. Daniel Podzamczer, MD, of Hospital de Bellvitge in Barcelona presented the interim nine-month results. Either drug was combined with the double NRTI drug formulation Combivir in HIV positive persons without previous anti-HIV treatment. The twice-daily dose of nelfinavir was used (1,250 mg). Of the 142 subjects, 25% were women overall, although there were significantly fewer women randomized to the nelfinavir arm (18%) than the nevirapine arm (33%). Race-ethnicity was not reported. The median baseline viral load was 4.8 log (63,095) copies/mL with a CD4 cell count of approximately 356 cells/mm3.

The interim results were presented after nine months. Using a lower limit of 20 copies/mL, significantly more participants achieved an undetectable viral load in the nevirapine arm (67%) than in the nelfinavir arm (39%, in a strict intent-to-treat analysis). Less of a difference was found when using a lower limit of 200 copies: 71% in the nevirapine arm and 56% in the nelfinavir arm. Greater virologic suppression occurred among those with a high baseline viral load, greater than 100,000 copies/mL (approximately one out of three study subjects). While the nelfinavir arm achieved a greater increase in the CD4 cell count (increase of 172 cells/mm3) than the nevirapine arm (increase of 116 cells/mm3), this difference was not statistically significant. Discontinuation due to adverse events was similar in both arms: 22% in the nevirapine arm and 19% in the nelfinavir arm. Adverse events were similar to what has been reported for these drugs in the past. The study did have a somewhat high rate of participants lost to follow-up: 24% in the nelfinavir arm and 14% in the nevirapine arm. The interim results suggest that a combination of nevirapine/Combivir shows virologic benefits over a nelfinavir/Combivir combination. The higher rate of people lost to follow-up in the nelfinavir arm is one limiting factor of the results. The study is ongoing for 48 weeks.

References

Podzamczer D and others. A randomized, open, multicenter trial comparing Combivir plus nelfinavir or nevirapine in HIV-infected naïve patients (The Combine Study). Abstract and presentation 694.


In another study, nevirapine combination therapy was compared with indinavir combination therapy. The results after 36 weeks revealed similar outcomes. J. Guardiola, MD, of Sant Pau Hospital in Barcelona presented the study. The dual NRTI backbone was d4T and ddI. A total of 50 HIV positive persons (27% women) with no previous anti-HIV therapy were enrolled. Nevirapine was dosed twice daily, 200 mg, while indinavir had a standard dose of 800 mg three times daily. The median baseline viral load was approximately 5.3 log (199,526) copies/mL with a CD4 cell count of 370 (nevirapine) and 337 cells/mm3 (indinavir). The results after nine months revealed an equal percentage with an undetectable viral load (50%, with a lower limit of 50 copies/mL) for both arms (in a strict intent-to-treat analysis). There were also similar results for those with a high baseline viral load. The CD4 cell count increase was 223 (nevirapine) and 166 cells/mm3 (indinavir) arm. Adverse events likewise occurred at similar rates in the two arms, not all of which were due to nevirapine or indinavir. Changes in lipids and cholesterol also were very similar when comparing the two arms. This small study reveals a near equivalence comparing nevirapine to indinavir in combination therapy, a different result than the study above, comparing nevirapine to nelfinavir. (Note the NRTI backbone was different in the two studies.)

References

Guardiola J and others. An open-label, randomized, comparative study of stavudine (d4T) + didanosine (ddI) + indinavir versus d4T + ddI + nevirapine in treatment of HIV-infected naïve patients. Abstract 539.

New Reports about
Lopinavir/Ritonavir (Kaletra)

At the 40th ICAAC, there were several presentations about lopinavir/ritonavir. This double-PI drug formulation named Kaletra received U.S. Food and Drug Administration (FDA) approval a few days before the conference began.

Treatment-Naïve Persons

Constance Benson, MD, of the University of Colorado at Denver authored a 96-week update of lopinavir/ritonavir in Study M97-720. The study enrolled 100 treatment-naïve HIV positive persons (i.e., who had never taken anti-HIV therapy). Four percent of subjects were women, 29% Black, and 6% Latino. The dosing was lopinavir 200-400 mg plus ritonavir 100-200 mg twice daily plus d4T and 3TC. After week 48, all study subjects took the FDA-approved dose of lopinavir 400 mg plus ritonavir 100 mg in a fixed formulation twice daily. The median baseline HIV RNA viral load was 4.8 log (63,095) copies/mL with a CD4 cell count of 326 cells/mm3. The results after 96 weeks showed that 78% had an undetectable viral load (with a lower limit of 50 copies/mL) using a strict intent-to-treat analysis. The rate was nearly the same whether subjects had a baseline viral load greater than or less than 100,000 copies/mL. However, 28% of those with a baseline viral load greater than 100,000 copies/mL who achieved undetectability required more than nine months to reach that level. The mean increase in the CD4 cell count was 290 cells/mm3. The most common adverse events (side effects, moderate or worse) were diarrhea 23%, nausea 15%, stomach-area pain 8%, weakness 7%, headache 7%, and vomiting 5%. The most common abnormal laboratory tests were increased cholesterol (greater than 300 mg per deciliter) in 14%, increased triglycerides (fats, greater than 750 mg per deciliter), increased liver enzymes (greater than five times the upper normal limit) in 10%, and increased amylase (pancreas gland enzyme, greater than twice the upper normal limit) in 4%. Yet only 2% of participants discontinued due to adverse events. The total discontinuation rate was 14%, due to dosing nonadherence, medical problems unrelated to study drugs, becoming lost to follow-up, and other reasons. The overall results indicate that the combination of lopinavir/ritonavir, stavudine, and lamivudine is quite potent in treatment-naïve persons.

Treatment-Experienced Persons

Steven Becker, MD, of the Pacific Horizon Medical Group in San Francisco presented an update of the 24-week results of Study M98-957. A total of 57 HIV positive persons with a detectable HIV RNA viral load after at least two PI drugs (at least three months each) were enrolled. None had ever taken an NNRTI drug. Women represented 21% and non-Caucasians 12%. The median baseline viral load was 4.5 log (31,622) copies/mL with a median CD4 cell count of 218-271 cells/mm3. Subjects were randomized to receive the FDA-approved dose of lopinavir 400 mg/ritonavir 100 mg twice daily or 533 mg/133 mg twice daily. Also in the regimen were efavirenz and two NRTI drugs. Participants had taken a mean three previous PI drugs and seven total anti-HIV drugs. The mean baseline susceptibility to lopinavir was decreased 16-fold (relative resistance) when compared with wild-type virus without mutations.

After 24 weeks, the results were as follows. An undetectable HIV RNA was achieved by 62% in the 400 mg arm and 64% of the 533 mg arm (in a strict intent-to-treat analysis, with a lower limit of 50 copies/mL). The mean CD4 cell increase was 46 cells and 41 cells/mm3 in the 400 mg and 533 mg arms, respectively. The most common adverse effects were quite similar to Study M97-720 above: diarrhea, weakness, increased cholesterol, triglycerides, amylase and liver enzymes. Seven subjects discontinued (12%, four subjects in the 400 mg and three in the 533 mg arms), including four due to adverse events and three due to virologic failure. Both categories of discontinuation reasons occurred among both dosing arms. The participants are still being followed, although they are taking the higher dose of lopinavir/ritonavir. The results show that the five-drug combination of lopinavir/ritonavir, efavirenz and two NRTI drugs is quite effective as a third-line regimen for those who have taken at least two PI drugs in the past but not any NNRTI drugs-even with significant PI drug resistance at baseline. Dr. Becker also concluded that lopinavir/ritonavir "should be increased to 533/133 mg BID [four capsules twice daily] when coadministered with efavirenz in patients with extensive prior PI [drug] use."

Cholesterol-Lowering Drug Interaction with Lopinavir/Ritonavir

High cholesterol levels represent a common side effect of most PI drugs and some NNRTI drugs. High cholesterol is one risk factor for blood vessel disease that can lead to a heart attack or stroke. Both atorvastatin (Lipitor) and pravastatin (Pravachol) are members of the "statin" drug class to lower blood cholesterol levels. Kaletra should not be combined with atorvastatin due to five-fold increased blood levels of the latter drug. However, it is safe to combine lopinavir/ritonavir with pravastatin. The drug levels of pravastatin are increased by approximately 30% and are unlikely to represent a problem for most people. Neither of the two statin drugs affected the blood levels of lopinavir/ritonavir.

Efavirenz Drug Interaction

Kaletra dosing might need to be increased when combined with the NNRTI drug efavirenz. The blood concentrations of lopinavir were reduced when lopinavir/ritonavir and efavirenz were combined in HIV positive persons. Specifically, the minimal concentration was reduced 44% and the total concentration (area-under-the-curve, or AUC) was decreased by 25%. In healthy volunteers, the combination decreased all blood concentration measurements of efavirenz by less than 16%, suggesting that no dose adjustment for efavirenz is needed. Ritonavir blood levels were not affected. If the twice daily dose of lopinavir 400 mg/ritonavir 100 mg is increased to 533/133 mg twice daily when combined with efavirenz, then the blood levels of lopinavir are nearly the same as when lopinavir/ritonavir is taken alone. This would mean four capsules twice daily of lopinavir/ritonavir, plus the other medications in the regimen.

Baseline genotypic resistance testing in this study was presented by Dale Kempf, PhD, of Abbott Laboratories. He expanded his previous findings and proposed a new categorization for lopinavir resistance. Zero to five genotype mutations to lopinavir would be called sensitive (no resistance); six or seven mutations would be called intermediate resistance. Only if there were eight or more mutations would this be called resistant to lopinavir.

Lopinavir-Based HAART Compared with Nelfinavir-Based HAART

Interim results of the first Phase III study comparing lopinavir/ritonavir combination therapy to another PI drug-based combination were presented. Sharon Walmsley, MD, of Toronto General Hospital in Canada discussed the results of the randomized, blinded (i.e., medications were unknown to participants) study of 653 subjects. Women represented 20% of participants, African-Americans 27%, and Latinos 13%. No participant had ever taken more than 14 days of anti-HIV drug, without any previous d4T or 3TC. The mean baseline HIV RNA viral load was 4.9 log (79,432) copies/mL with a CD4 cell count of 259 cells/mm3. Subjects were randomized to receive lopinavir 400 mg/ ritonavir 100 mg twice daily or nelfinavir 750 mg three times daily. (The nelfinavir dose was changed later to 1,250 mg twice daily.) In addition, all subjects took d4T and 3TC, both twice daily NRTI drugs. For the study to be blinded, participants also took placebo (inactive) pills of the PI drug to which they were not randomized. This led to a high number of daily pills.

After 40 weeks, the results showed that the lopinavir arm had a significantly higher rate of viral undetectability (79% were lower than 400 copies/mL) compared with 64% in the nelfinavir arm (in a strict intent-to-treat analysis). Using a lower limit of 50 copies/mL, the undetectability rates were 70% and 54%, respectively. The CD4 cell count increases were 190 and 177 cells/mm3 for the lopinavir and nelfinavir arms, respectively. Adverse events (side effects) were similar in both groups: diarrhea, nausea, weakness, stomach-area pain, vomiting, and headache. The lopinavir arm had a higher rate of grade 3-4 (severe or life-threatening) increase in blood cholesterol (8%) than the nelfinavir arm (4%). Similarly, the lopinavir arm had a higher rate of grade 3-4 increase in blood triglycerides (fats, 7%) when compared with the nelfinavir arm (1%). The discontinuation rate after 40 weeks was 15% in the lopinavir arm and 20% in the nelfinavir arm; although only 2-3% in each arm was due to adverse effects related to study drugs. These preliminary results suggest that the lopinavir combination therapy arm may be more potent than the nelfinavir combination therapy arm. However, the number of pills required to make this a placebo-controlled study and the fact that a middle of the day dosing was required initially both limit the results. This study is ongoing.

Food Interactions with Lopinavir/Ritonavir

The total blood concentration (AUC) of lopinavir from the soft capsule formulation was increased 26% when taken with a high fat meal (56% of calories from fat) than when taken with a moderate fat meal (25-30% of calories from fat). The same experiment using the liquid formulation led to a 37% increase in blood levels of lopinavir. When taken without food (fasting), the capsule formulation led to 36% lower blood levels than with a moderate fat meal. The same test with the liquid formulation led to a 44% lower level. The results indicate that lopinavir/ritonavir should be taken with food, preferably a high fat meal.

References

Becker S and others. ABT-378/ritonavir and efavirenz: 24 week safety/efficacy evaluation in multiple PI experienced patients. Abstract and oral presentation 697.

Benson C and others. ABT-378/ritonavir (ABT-378/r) in antiretroviral naïve HIV + patients: 96 weeks. Abstract and poster 546.

Bertz R and others. Assessment of the pharmacokinetic interaction between ABT-378/ritonavir and efavirenz in healthy volunteers and HIV+ subjects. Abstract 424.

Carr RA and others. Concomitant administration of ABT-378/ritonavir results in a clinically important pharmacokinetic interaction with atorvastatin but not pravastatin. Abstract and presentation 1644.

Gustavson LE and others. Assessment of the bioequivalence and food effects for liquid and soft elastic capsule co-formulations of ABT-378/ritonavir in healthy subjects. Abstract and presentation 1659.

Kempf D and others. Definition of genotypic breakpoints for ABT-378/ritonavir for use in the interpretation of HIV resistance testing. Abstract and presentation 1264.

Walmsley S and others. Efficacy of ABT-378/ritonavir versus nelfinavir in antiretroviral-naïve subjects: results of a phase III blinded randomized clinical trial. Abstract and presentation 693.

Experimental Treatments Under Development for HIV

PI Drugs

BMS-232632, Mozenavir (DMP-450), Tipranavir

I. Sanne, MD, of Johannesburg Hospital in South Africa presented the interim results of a Phase II study about BMS-232632. This drug is an experimental protease inhibitor (PI) that has a once daily dosing, without "boosting" from a second PI drug. A total of 422 HIV positive persons were enrolled; 32% were women, 40% were not Caucasian, and 13% had a history of injection drug use. Less than four weeks of previous NRTI drug experience and less than one week of previous NNRTI or PI drug experience was allowed. The median baseline CD4 cell count was 318 cells/mm3, with an HIV RNA viral load of 4.7 log (50,118) copies/mL. In this double-blind study (in which the medication type and dose was not known by either participants or study physicians), subjects took BMS-232632 as the only drug for two weeks. They were randomized to a dose of 200, 400, or 500 mg once daily. A control arm took nelfinavir 750 mg three times daily. After two weeks, d4T and ddI were added. A sub-group of 98 participants have completed 24 of the planned 46 weeks, while the remaining have completed only 16 weeks.

Interim results in the first sub-group after 24 weeks showed that up to 35% had an undetectable viral load, with a lower limit of 50 copies/mL (in a strict intent-to-treat analysis). In the larger subgroup, the three doses and their respective percentages with an undetectable viral load after 12 weeks were: 200 mg (39%), 400 mg (31%) and 500 mg (50%). In general, the viral load decreased by approximately 2.5 log (316-fold) at 12-16 weeks. In the larger subgroup, the CD4 cell count increased by approximately 75-100 cells/mm3 after 12 weeks. The most common side effects were nausea, diarrhea, stomach area (gastrointestinal) pain, infection and headache. As has been previously reported, the blood bilirubin (bile pigment) increased in all three dosing arms - this was treated by reducing the dose of BMS-232632. In seven of these cases, the subjects had jaundice (yellow appearance to skin and whites of eyes). There were five cases of life-threatening (grade 4) increases. On the other hand, there were only minimal increases or no changes in total blood cholesterol, LDL ("bad" or low density lipoprotein) cholesterol or triglycerides (blood fats). Whereas, the nelfinavir control arm did have increases in those measurements of cholesterol and triglycerides.

There were two other presentations about this drug. In a second presentation about BMS-232632, Edward O'Mara, MD, of Bristol-Myers Squibb showed that there were no significant changes in blood levels of BMS-232632 when it is taken at a 400 mg dose with 200 mg of ketoconazole (Nizoral, an antifungal antibiotic). In a third presentation about BMS-232632, Dr. O'Mara reported initial results suggesting a genetic marker increases the risk of increased bilirubin (see side effects above) from the drug. The marker is called UDP-GT 1A1 (see reference below). This marker also should be tested for the same side effect that occurs among some people when they take indinavir. In a fourth presentation about BMS-232632, Richard Colonno, PhD, of Bristol-Myers Squibb showed that the drug has "no obvious mutational pattern…other than a requirement of several [PI drug] mutations to be present."

References

Colonno RJ and others. Efficacy of BMS-232632 against a panel of HIV-1 clinical isolates resistance to currently used protease inhibitors. Abstract 2114.

O'Mara E and others. Relationship between uridine diphosphate-glucuronosyl transferase (UDP-GT) 1A1 genotype and total bilirubin elevations in healthy subjects receiving BMS-232632 and saquinavir. Abstract and poster 1645.

O'Mara E and others. Steady-state pharmacokinetic interaction study between BMS-232632 and ketoconazole in healthy subjects. Abstract and poster 1646.

Sanne I and others. Safety and antiviral efficacy of a once-daily HIV-1 protease inhibitor BMS-232632: 24 weeks results from a phase II clinical study. Abstract 691.

There were three presentations about mozenavir, an experimental PI drug under development by Triangle Pharmaceuticals. A Phase I/II study included 40 HIV positive subjects. Women represented 13%; race-ethnic background was not reported. None had previously taken anti-HIV drugs. The median baseline viral load was 4.6-5.3 (39,810-177,827) log copies/mL with a median CD4 cell count of 234-285 cells/mm3. The dose of DMP-450 was 750 mg three times daily, 1,250 mg three times daily, or 1,250 mg twice daily. The two NRTI drugs also taken by all participants were d4T and 3TC.

After up to 48 weeks, 87% had an undetectable viral load using a lower limit of 50 copies/mL (in a strict intent-to-treat analysis). Side effects included diarrhea, nausea, muscle aches, headache and rash. There were mild increases in blood cholesterol. The authors concluded that DMP-450 was tolerated well, had significant anti-HIV benefits with few differences when comparing the three doses. Interestingly, the 1,250 mg three times daily dosing led to equal or marginally better drug concentrations than the 1,250 mg twice daily dosing. This included the maximal, minimal, and AUC concentrations. This suggests that DMP-450 might be dosed twice daily, which is associated with better adherence to dosing in other studies. Dr. G. Moralles of Triangle Pharmaceuticals has reported previously that this PI drug has a relatively lower cost of production that could lead to marketing in resource-poor countries.

Information about mozenavir also was presented by Dr. Laurene Wang of Triangle Pharmaceuticals. Dr. Wang reported drug interaction data when it was added to either indinavir or saquinavir. Healthy HIV negative volunteers were enrolled. Neither of those two PI drugs affected the blood concentrations of DMP-450. However, 1,000 mg of DMP increased the minimal and total (AUC) concentration of indinavir by approximately 40%. Yet, the maximal concentration and half-life (time until an original amount is metabolized by one-half) were not changed. When DMP-450 was added to saquinavir, the maximal concentration increased five-fold, the AUC concentration ten-fold, and the half-life by 2.5-fold. This information is important before clinical studies could be devised that would combine DMP with either indinavir or saquinavir.

A third presentation about side effects of DMP-450 was authored by Dr. Gregory Chitick of Triangle Pharmaceuticals. The results were a lesser increase in total blood cholesterol when compared with indinavir and similar increases in triglycerides (fats) and glucose (sugar). Dr. Chitick concluded that 2,000 mg twice daily of DMP-450 "appears to be the maximum tolerated dose."

References

Chittick GE and others. Effects of a novel HIV-1 protease inhibitor, DMP 450, on cardiac tracing, serum lipids and glucose tolerance, as compared to indinavir. Abstract and presentation 1648.

Sierra J and others. Preliminary profile of the antiviral activity, metabolic effects, safety and pharmacokinetics of DMP-450, a novel cyclic urea protease inhibitor. Abstract and presentation 540.

Wang LH and others. Effects of a novel HIV-1 protease inhibitor, DMP 450, on the pharmacokinetics of indinavir and saquinavir. Abstract and presentation 1647.

Information about resistance to tipranavir was presented by Dr. Sharon Kemp of Virco. This experimental PI drug has activity against numerous HIV strains (isolates) that are resistant to PI drugs currently on the market. Dr. Kemp concluded that "there does not appear to be any obvious common combinations of the known PI mutations that clearly confer tipranavir resistance." She was able to generate moderate resistance with eleven primary and secondary mutations seen with other PI drugs.

Reference

Kemp S and others. Site-directed mutagenesis and in vitro drug selection studies have failed to reveal a consistent genotypic resistance pattern for tipranavir. Abstract 2113.

NtRTI Drug

Tenofovir DF

Three presentations at the 40th ICAAC focused on tenofovir disoproxil fumarate (TDF, previously known as oral PMPA). This drug is an experimental nucleotide reverse transcriptase inhibitor drug (NtRTI) under development by Gilead Sciences. It is dosed once daily and has activity against HIV strains (isolates) with specific resistance to several NRTI drugs (see two paragraphs below). The NtRTI drugs differ from the NRTI drugs in that they only require two steps for activation rather than three that are required for the NRTI drugs. An update of tenofovir in persons with advanced disease was presented by Robert Schooley, MD, of the University of Colorado at Denver. The study was double-blind and placebo-controlled (i.e., one arm took a placebo). Enrolled participants must have had a detectable viral load on a stable anti-HIV regimen for at least two months. Tenofovir was added at a dose of 75 mg, 150 mg, or 300 mg. All participants, including those randomized to placebo took 300 mg of tenofovir after week 24. The 189 enrolled subjects had taken anti-HIV medication for a mean of 4.6 years with a mean baseline viral load (when starting this study) of 375 cells/mm3. At that time, 97% had genotypic resistance mutations to NRTI drugs.

After 48 weeks, the results showed the following viral load reductions: 300 mg (decrease of 0.7 log copies/mL or five-fold); 150 mg (decrease of 0.6 log copies/mL or four-fold); 75 mg (decrease of 0.4 log or 2.5-fold); and placebo changed to 300 mg (decrease of 0.7 log or five-fold). The CD4 cell count increases were: 300 mg (11 cells/mm3); 150 mg (16 cells/mm3); 75 mg (11 cells/mm3); and placebo changed to 300 mg (25 cells/mm3). At 48 weeks, 20% had stopped tenofovir, compared with 25% in the 300 mg arm up to 24 weeks. Dr. Schooley reported that there was no dose-related toxicity. The percentage of participants with an undetectable viral load was not presented.

Michael Miller, PhD, of Gilead presented information about tenofovir resistance. HIV strains (isolates) with the common 3TC-induced mutation M184V are hypersensitive (i.e., less drug is needed to achieve the same effect as with wild-type virus without mutations) to tenofovir. Almost all strains with high-level resistance to AZT also were sensitive to tenofovir. Adding M184V to those strains increased tenofovir sensitivity somewhat. A combination of NRTI mutations (Q151M) that lead to multidrug NRTI resistance did not decrease tenofovir's effectiveness, with or without M184V. However, another group of multidrug resistance mutations ("69 insertions") were resistant to tenofovir, but became more sensitive ("intermediate resistance, six-fold decrease") when M184V was added. The K65R resistance mutation (due to abacavir, ddI, or ddC) decreased sensitivity to tenofovir somewhat, but this was nearly reversed when M184V was added. These results indicate the unique effectiveness of tenofovir against HIV with several types and groupings of mutations that cause NRTI drug resistance.

Dr. L.K. Naeger of Gilead presented other information that might explain part of the reason for tenofovir's effectiveness against HIV strains with various NRTI mutations. Resistance to zidovudine and d4T has been partly explained by two different mechanisms. They are called "pyrophosphorolysis" and "dinucleotide polyphosphate synthesis." In a series of experiments, Dr. Naeger showed that tenofovir is much less susceptible to these resistance mechanisms than zidovudine.

References

Miller MD, and others. Antiviral activity of tenofovir (PMPA) against nucleoside-resistant HIV samples. Abstract and presentation 2115.

Naeger LK and others. Tenofovir (PMPA) is less efficiently removed through pyrophosphorolysis and dinucleotide polyphosphate synthesis than zidovudine by HIV-1 wild-type RT and RT mutations. Abstract and presentation 1265.

Schooley R and others. Tenofovir disoproxil fumarate (TDF) for the treatment of antiretroviral experienced patients, a double blind, placebo-controlled study. Abstract and presentation 692.

NRTI Drugs

Emtricitabine (FTC, Coviracil)

There are several NRTI drugs under development that were presented at the 40th ICAAC. Emtricitabine (FTC, Coviracil) is an NRTI drug with activity against HIV and HBV (hepatitis B virus). Safety results were presented by Dr. T.B. Grizzle of Triangle Pharmaceuticals. In tests involving small animals, essentially no toxicity (including cancers and birth defects) was detected. At a dose more than 100 times that used in humans, anemia (low red cell count) occurred in female monkeys and resolved after the drug was stopped. At a dose almost 1,000 times that used in humans, the same side effect occurred in mice. No mitochondrial toxicity was found. (Mitochondria are the energy producers of cells; several NRTI drugs are toxic to them.) In a small study of 40 treatment-naïve subjects who combined FTC with ddI and efavirenz, 95% achieved an undetectable HIV viral load (with a lower limit of 400 copies/mL) after 48 weeks (see reference). This strongly suggests very high potency of that triple drug combination.

References

Grizzle TB and others. Emtricitabine: summary of toxicology and nonclinical pharmacology evaluations. Abstract and poster 1631.

Molina J-M and others. Once-daily therapy with emtricitabine, didanosine and efavirenz in treatment naïve HIV-infected adults: 48-week follow-up of The ANRS 091 Trial. Late-breaker abstract 648 at the 38th Annual Meeting of the IDSA (Infectious Diseases Society of America); September 7-10, 2000; New Orleans, Louisiana.

DAPD

Joseph Eron, MD, of the University of North Carolina at Chapel Hill presented the results of a Phase I/II study of DAPD. DAPD is a NRTI drug and has activity against HIV and HBV (hepatitis B virus). It also has activity against HIV strains with selected resistance to AZT, 3TC, and/or abacavir. HIV positive persons without previous treatment for HIV were enrolled. The DAPD dose was twice daily of 25 mg, 100 mg, 200 mg, 300 mg, or 500 mg. The maximum HIV RNA viral load reduction after 15 days of the drug was 1.5 log (31-fold) copies/mL at the 300 mg dose. The drug was tolerated well, without any discontinuations due to drug adverse events (side effects). No genotypic resistance associated with NRTI drugs was detected at the end of the 15-day study.

Experimental Drugs in Preclinical Testing

There were a few other drugs or compounds presented at the 40th ICAAC that are in preclinical (before use in humans) development. The first group of integrase inhibitors called "butanoic acids" was described by Dr. John S. Wai from Merck. PRO 140, an entry inhibitor drug, is an antibody (type of protein) to the CCR5 receptor on human immune cells to which HIV binds. In lab tests, low nanomolar (extremely low) concentrations of PRO 140 blocked infection of blood mononuclear cells by HIV strains that use CCR5 to enter. Dr. W.C. Olson of Progenics presented the study. In a separate presentation, Dr. Olson found that combining the two HIV entry inhibitor drugs PRO 542 and T-20 led to synergistic (enhanced combination) benefits in the laboratory. (T-20 is already in Phase III human studies and is also called an HIV fusion inhibitor.)

In yet another presentation, Dr. C.L. Tremblay of Massachusetts General Hospital in Boston reported anti-HIV synergy (enhanced effects) when T-20 was combined with the CCR5 receptor inhibitor TAK-779. Previously, Dr. Tremblay has reported a similar synergistic effect in the laboratory when T-20 was combined with AMD-3100, an entry inhibitor of the CXCR4 receptor on human T-cells that HIV uses to enter. AMD-3100 was reported to be absorbed into the blood of rabbits when they took the experimental medication by mouth. Dr. R.T. MacFarland of AnorMED Inc. presented the report at ICAAC. Two experimental ribonucleotide reductase inhibitors showed anti-AIDS benefits in an animal ("murine") AIDS model, either alone or when combined with ddI. Both Trimidox and Didox are similar to hydroxyurea (Hydrea), but showed much less toxicity than hydroxyurea. The report was presented by Dr. C.N. Mayhew of the University of Wolverhampton in the UK.

References

MacFarland RT and others. An orally bioavailable CXCR4 antagonist for inhibition of HIV replication. Abstract 1845.

Mayhew CN and others. Comparison of novel ribonucleotide reductase inhibitors, Didox and Trimidox, to hydroxyurea regarding antiretroviral activity and toxicity in murine AIDS. Abstract 553.

Olson WC and others. Potent, broad spectrum inhibition of HIV-1 by the CCR5 antibody PRO 140. Abstract 550.

Olson WC and others. Potent, synergistic inhibition of HIV-1 by combinations of the viral entry inhibitors PRO 542 and T-20. Abstract 549.

Tremblay CL and others. In vitro synergy observed between the fusion inhibitor T-20 and a CCR5 inhibitor TAK-779. Abstract 1164.
Wai JS and others. 4-aryl-2,4-dioxobutanoic acid inhibitors of HIV-1 integrase and viral replication in cells. Abstract and presentation 1844.

Coinfection with HIV and HCV

Approximately one-third of HIV positive persons in developed countries are also infected with HCV, due to somewhat similar routes of transmission. As more such persons have so-called stable HIV disease due to HAART, liver disease due to HCV has become an increasing problem. A test for abnormal liver enzymes (ALT, alanine aminotransferase) would suggest liver disease among those infected with HCV. Some physicians may feel that if the ALT were normal, a liver biopsy sample (piece viewed under the microscope) would not be necessary. Much less information is available about persons coinfected with HIV and HCV. Now, in a small study from Allentown, PA, researchers have found that 80% of HIV-HCV coinfected persons had an abnormal liver biopsy, even though their liver enzyme test (for ALT) was normal. Even though the study was small, the results suggest that among persons coinfected with HIV and HCV, liver damage might be present even when the liver enzyme test is normal. An abnormal liver biopsy would be one requirement for considering treatment for chronic hepatitis C.

In this study, there were 24 subjects with HIV-HCV coinfection (33% women, 58% Latino, and 25% African-American). Therapy for HIV was being taken by 83%: almost all of them were taking three-drug HAART. An abnormal ALT was present in 58%, while the remaining 42% had a normal level. Women and Latinos were evenly distributed among those with or those without a normal ALT. At the time of liver biopsy, the mean (average) CD4 cell count was 442 cells/mm3, 58% had an undetectable HIV RNA viral load (with a lower limit of 50 copies/mL) and 71% had an HCV RNA viral load greater than 1 million copies/mL. All 24 participants indicated they were not actively drinking alcohol or using intravenous (IV) drugs. The HCV genotype was not reported.

The following results were presented. All 14 subjects with an abnormal ALT had an abnormal liver biopsy showing inflammation and fibrosis (scarring). And, 80% of the ten subjects with normal ALT levels had an abnormal liver biopsy. Both inflammation and fibrosis were present among 60% of those with a normal ALT, and only inflammation was present among 20%. Moderate or severe inflammation and fibrosis was present in 20% of those with a normal ALT.

HIV-HCV coinfected persons with an abnormal liver biopsy due to HCV are at risk for progression of liver disease. A sustained virologic response to dual drug therapy (interferon alfa plus ribavirin) for chronic hepatitis C occurs among approximately 39%. Response rates will likely be higher when ribavirin is taken with a long-acting, pegylated interferon alfa that is not yet FDA-approved (Pegasys, PEG Intron). There are very limited studies evaluating the benefits of dual therapy for HCV among HIV-HCV coinfected persons with stable HIV disease resulting from HAART. In the mean time, even though this study was small, the results suggest that a normal ALT test among HIV-HCV coinfected persons does not necessarily mean that the liver biopsy will be normal. An abnormal liver biopsy indicates increased risk of future liver disease, including possible death from liver failure that could take several years to decades.

Reference

Hoffman-Terry M and others. Correlation of ALT with degree of liver damage by biopsy in HIV/HCV coinfected adults. Abstract and poster 175.

Liver Disease

With Mild Liver Disease, Abacavir Dose Should Be Decreased

Sometimes people with kidney or liver problems require a lower dose of certain medications. François Raffi, MD, from France and Dr. G. Yuen of Glaxo Wellcome presented a study showing that the dose of abacavir should be decreased in persons with mild liver disease. The researchers defined mild liver disease as a Child-Pugh score of five on liver biopsy samples. The study of HIV positive subjects included nine with mild liver disease and nine control subjects without liver disease matched for age and weight. Women represented 22% of both groups, while all of those with liver disease were White. The cause of liver disease was not reported. After one dose of abacavir 600 mg, blood and urine tests were sampled for up to 24 hours. The results showed that the blood concentrations of abacavir were higher among those with liver disease. This included the highest blood concentration (maximum) and the AUC. The maximum level increased by a mean (average) of 26%, while the AUC increased by 89%. The half-life also increased 58% among those with liver disease. There were seven emergent side effects (adverse events) that occurred, and 71% of these were among those with liver disease. Yet, all but one was mild or moderate and nonserious. The authors concluded that "patients with mild hepatic [liver] impairment [disease] should receive 150 mg [of] abacavir twice daily" instead of the usual dose of 300 mg twice daily.

Reference

Raffi F and others. Pharmacokinetics of, and tolerability to, a single, oral, 600 mg dose of abacavir in HIV-positive subjects with or without liver disease (CNAB1006). Abstract and poster 1630.

Liver Disease Increases Nelfinavir Levels

A preliminary study to evaluate blood levels of nelfinavir, a PI drug, in persons with varying degrees of liver disease was presented. All of the 24 subjects were HIV negative. One 750 mg dose was used. The results revealed moderate increases in blood concentrations of the drug. This included the peak (maximal) concentration and the AUC. Also, the half-life increased. As a result of this study, Agouron Pharmaceuticals plans to do a similar study in HIV positive persons with liver disease. Note that a finite increase in the blood concentration of nelfinavir would likely be a benefit in terms of greater anti-HIV potency. Therefore, a dose adjustment might not be needed. This assumes that no increased toxicity or side effects occur when future studies are completed.

Reference

Hsyu PH and others. Pharmacokinetics of nelfinavir and metabolite M8 in patients with liver impairment after a single oral 750 mg dose. Abstract 1657.

Portions of Dr. Bartnof's autumn 2000 coverage of the 40th ICAAC can be viewed on the NATAP Web site at www.natap.org.

Harvey S. Bartnof, MD, is a staff physician at the AIDS Virus Education and Research Institute (AVERI) in San Francisco, and the former Medical Editor of HIVandHepatitis.com.

Page last updated 20 March 2001