The XIIIth International AIDS Conference was held this year in Durban, South Africa from July 9-14. While there were no startling scientific discoveries, we are all more enlightened as a result of this conference. A number of presentations discussed treatment strategies with a focus on improving HIV outcomes for adults and children and managing the costs of care. The following are capsule summaries of those presentations that should be of interest to providers and patients alike. Relevant papers are cited at the end of each section.

The immune-system debate
Dr. Jay Levy (University of California at San Francisco) argued that innate immunity is as important as adaptive immunity, the immune system response that occurs as the result of antiretroviral therapy. Innate immunity is a natural defense mechanism, and is the body's first line of defense against pathogens. Furthermore, innate immunity provides a quick response (minutes to days), whereas adaptive immunity occurs on the order of days to weeks. In HIV-infected individuals, an absence of opportunistic infections (OI) is observed in individuals expressing interferon-a (IFN-a). Dr Levy concluded by stating that innate immunity and adaptive immunity are linked: innate immunity helps to induce the adaptive immune system and protects many individuals from HIV infection-without the innate immunity system, the adaptive immunity system could not exist; it should therefore be considered equally as important.

Arguing "against" Levy was Dr. Philip Goulder, a U.S. researcher, who countered that the responses generated through adaptive immunity would be more effective than innate immunity in prevention and treatment. A critical difference between the two systems is that the adaptive immunity system has no memory. This allows for the continuous suppression of plasma HIV RNA. Dr Goulder concluded that the more the scientific community knows about adaptive immunity, the more likely it is that the community will create good vaccine strategies.

In summary, Drs. Levy and Goulder both agreed that the two systems could not exist without each other. Dr. Pepe Alcami ended the debate by noting that the take-home message is to understand the pathogenesis of HIV, as the individual immunity of HIV-infected individuals is highly complex. Ultimately, both innate immunity and adaptive immunity fail because neither prevents HIV replication.

Mackewicz, C et al. CD8 non cytotoxic anti-HIV immune response inhibits viral transcription and not early steps in virus infection. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPpA1151.

Altfield M et al. Enhancement of tHIV-1 specific CTL responses during structured treatment interruptions (STI) following treated acute HIV-1 infection associated with the control of HIV-1 viremia. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract ThOrB750.

Mother-to-child transmission
There was considerable discussion about mother-to-child transmission (MTCT) at the Durban Conference. This problem has been all but eliminated in the U.S. by more widespread HIV testing in general, testing of women in obstetrical settings, and the three-part standard of care that uses antiretrovirals during pregnancy, at the time of delivery, and during the perinatal period.

Several studies presented in Durban have relevance for Africa and the developing world, where MTCT is still a significant health issue.

The SAINT Trial was a large, multicenter study designed to evaluate the safety and efficacy of a simple regimen of Viramune® versus ZDV+3TC in reducing MTCT of HIV. This randomized, open-label, comparative trial enrolled more than 1300 women who tested HIV- positive in labor or late in pregnancy (>38 weeks) who were not previously received and had not currently receiving other antiretroviral therapy for HIV-1.

Nevirapine (Viramune®) was shown to be well tolerated and effective in the prevention of MTCT of HIV. While nevirapine is not currently indicated for the prevention of MTCT, the use of this drug to prevent MTCT is particularly intriguing because of its long half-life. Study findings show that an inexpensive, simple regimen of nevirapine is as effective as a seven-day course of two drugs, ZDV+3TC, at reducing MTCT of HIV. In the nevirapine arm, one dose of Viramune® was administered to HIV-infected mothers in labor, followed by a second dose 24 to 48 hours after delivery; one dose was given to their infants immediately after birth.

The findings show that there was no significant difference in efficacy between the nevirapine regimen and the ZDV+3TC regimen; both arms appeared to be well tolerated. The overall rates of MTCT of HIV were 14.0% and 10.8% for the nevirapine and ZDV+3TC arms, respectively. The overall rates included the incidence of MTCT of HIV occurring during pregnancy (intrauterine), around the time of birth (perinatal), and shortly after birth (early postpartum) by 8 weeks of age, as confirmed by two positive viral tests (DNA-PCR). The rate of MTCT occurring intrapartum or early postpartum was 6.3% and 4.3% for the nevirapine and ZDV+3TC arms, respectively.

Preliminary data suggested that ddI and d4T are safe, effective, and tolerable alternatives to AZT or nevirapine in preventing MTCT of HIV.

The AI pilot study conducted in Soweto, South Africa randomized 373 HIV-positive pregnant women to an orally administered regimen of ddI, d4T, ddI + d4T, or AZT. Treatment was initiated at 34-36 weeks of gestation and continued through labor. Formula-fed newborns were administered the same therapy as their mothers for 6 weeks postpartum (after delivery). The effectiveness of each regimen was measured by the presence or absence of HIV infection using DNA-PCR testing in infants at birth through 24 weeks and measuring antiviral activity in HIV-infected pregnant women.

Preliminary results from an interim analysis on the first 204 women randomized showed that the ddI, d4T, and ddI + d4T treatment arms were as effective in reducing MTCT as AZT. Preliminary 6-week data for the infants showed that MTCT rates for ddI (1.9%), d4T (4.2%), and ddI + Zed4Trit (2.0%) were similar to rates for AZT (6.3%).

McIntyre J, et. al. Evaluation of safety of two simple regimens for prevention of mother to child transmission (MTCT) of HIV infection [Nevirapine (NVP) vs. lamivudine (3TC) + ZDV] used in a randomized clinical trial (SAINT Trial). Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuOrB356.

Gray G, et al. Preliminary efficacy, safety, tolerabil ity and pharmacokinetics of short course regimens of nucleoside analogues for the prevention of mother-to-child transmission (MTCT) of HIV. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuOrB355.

HIV-infected children
Dr. Maurizio de Martino and colleagues of the University of Florence, Italy, followed 1142 HIV-infected children born between 1980 and 1997. Cumulative survival probabilities of the birth cohort and the birth cohort were similar. However, this measure increased significantly in the birth cohort.

The improved survival of HIV-infected children coincides with the period when triple combination therapy became widespread. The risk of death decreased by 30% for those receiving two antiretroviral drugs, and by nearly 70% for those receiving triple-drug therapy.

The effectiveness of triple therapy was significantly greater than double therapy, but double therapy and monotherapy were not significantly different.

The authors concluded that the effectiveness of triple combination therapies in infants and children is at least similar to, and may even exceed, that demonstrated in HIV-infected adults.

During a briefing in Durban, Dr. de Martino's colleague Dr. Patrizio Pezzotti made the point that important problems in treating pediatric HIV infection still exist. These include "the lack of pediatric formulations and poor palatability" of antiretroviral medications, along with the problem of "non-adherence and poor compliance."

HIV and breast-feeding
The benefits of the short-course zidovudine regimen did not appear to be lost by breast-feeding at the 6-month follow-up in PETRA and other similar double-blind, placebo-controlled trials performed in developing countries.

Conducted under the auspices of UNAIDS, the so-called PETRA trial followed the outcome of 1754 HIV-infected women and their infants after they were randomized to one of four arms. These included zidovudine and 3TC administered at 36 weeks gestation and continued through 1 week after delivery (arm A); zidovudine and 3TC intrapartum through 1 week after delivery (arm B); zidovudine and 3TC intrapartum only (arm C); or placebo.

While both neonatal HIV transmission rates and mortality were significantly reduced after 6 weeks of life in treatment arms A and B compared with placebo, no significant difference in any of the study arms was seen after 18 months. During the follow-up period, most of the participants were breast-feeding, which the researchers believe accounts for the high number of HIV infections.

"We know what to do to reduce the risks of intrapartum and peripartum transmission," Dr. James McIntyre of the Chris Hani Baragwanath Hospital in Johannesburg commented at a press briefing. "We now have to look at the next phase...how do we either make breast-feeding safe or...make replacement feeding safe?"

Breast-feeding
Coutsoudis et al reported on new data available from the Durban study. Previously reported data described HIV transmission rates in 3-month-old infants as 18.8% in infants fed formula (no breast milk), 14.6% in infants exclusively breast-fed (breast milk only), and 24.1% in infants who had mixed feeds (breast milk and other liquids/solids). The new data were the HIV transmission rates for these infants at 15 months.

These new data appear to negate the effects of exclusive breast-feeding over formula, but emphasized the superiority of exclusive breast-feeding over mixed feeding. The HIV transmission rates at 15 months were 19.4% in formula-fed infants, 24.7% in infants exclusively breast-fed, and 35.0% in mixed-fed infants. Whether these differences were significant or not was not mentioned. The limitations of the study (i.e., it was not a randomized controlled trial) were stressed again, although Dr. Coutsoudis presented analyses to suggest that the initial findings were not due to the effects of reverse morbidity. Dr. Coutsoudis went on to recommend that exclusive breast-feeding be the feeding option of choice for HIV-infected women in developing and developed countries.

Fowler et al reported that the Durban study was the only study that had found lower HIV transmission rates in breast-fed infants compared with formula-fed ones, but he conceded that other studies had not analyzed the effects of exclusive vs. mixed breast-feeding (or had analyzed them only to a limited extent). Thus an important issue is the acceptability of exclusive breast-feeding to women in developing countries. The participants agreed that exclusive breast-feeding is uncommon, and that social and cultural norms promote the use of mixed feeding. For example, in the Durban cohort, only 26% of the women who chose to breast-feed fed their infants with breast milk exclusively. See these abstracts on breastfeeding from the Durban Conference.

Lamah, HV. Sponsoring breast feeding of little children of women living with AIDS. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000; Durban, South Africa. Abstract WePeE4884.

Bland RM et al. Logitudinal infant feeding study: constraints to exclusive breast feeding. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000; Durban, South Africa. Abstract WeOrC497.

Wikto, SZ et al. Twenty-four-month efficacy of short-course maternal zidovudine for the prevention of mother-to-child HIV-1 transmission in a breast feeding population: A pooled analysis of two randomized clinical trials in West Africa. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000; Durban, South Africa. Abstract TuOrB354.

Viral blips and their significance
Some individuals who achieve viral load suppression to <50 copies/mL may have episodes of detectable viral load, generally in the range of 50 to 400 copies/mL. The significance of this detectable virus is not known: it could be a simple "blip," perhaps for technical reasons, or it may be an indicator of virologic rebound and loss of treatment response.

Data from a large office-based HIV practice in Washington, DC, were presented on the first day of the meeting. Of 342 patients followed, 32 had been on a stable antiretroviral regimen with viral suppression of < 50 copies/mL on two separate measurements during a 4-month period, but subsequently had a detectable viral load between 50 and 400 copies/mL. Detectable samples were not retested to rule out laboratory error. Individuals who changed antiretroviral therapy were excluded from the study, and adherence data were not available.

At subsequent follow-up, 24 (75%) of 32 patients reestablished a viral load of < 50 copies/mL; in 22 cases, this was achieved at the next test, within 1 month of the detectable test result. Eight patients failed to resuppress HIV

RNA and progressed to have higher viral load levels or to require a change in treatment. Patients who achieved resuppression had a mean rebound viral load of 96.4 copies/mL compared with 131 copies/mL in those with subsequent viral rebound (P = .30). CD4+ counts at the time of initial viral rebound in the resuppression group and the subsequent viral rebound group were 625 cells/mm³ vs. 540 cells/mm³, respectively (P = .58).

Of the patients in this cohort who were on their first-ever regimen, all four were noted to achieve resuppression, and there were no differences in regimen choice in the two groups. The authors concluded that a rise from an "undetectable" viral load to a measurement between 50 and 400 copies/mL is not necessarily an indication of actual or impending virologic failure of an antiretroviral regimen, and they suggested that the most appropriate response is to retest the patient's viral load.

Suggested reasons for blips include ongoing rounds of replication; release of virus from latently infected cells or from sanctuaries including lymphoid tissues, the genital tract, and the central nervous system; and variations in adherence. Because evidence exists for persistence of viral replication below assay detection limits, variations in adherence is perhaps the most likely explanation. Of course, some of these blips may simply represent assay variation or intercurrent illnesses or vaccination, which could push a low level of ongoing replication above the arbitrary cut-off of the assay.

For the time being, calling the patient back for a repeat sample, or maybe just retesting the same sample, should be the first response to a modest rise in viral load.

Ward D, Sklar P. The significance of low-level viremia in patients with previously "undetectable" HIV-1 RNA levels. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract MoPpB1019.

Havlir D, Hirsch MS, Richman DD, et al. Prevalence and predictive value of intermittent viremia in patients with viral suppression. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPeB3195.

Structured intermittent therapy
Dr. Anthony Fauci of the U.S. National Institutes of Health (NIH) stated that while eradication of HIV infection will not be possible with the currently available antiretrovirals, prolonged control is a viable treatment goal. He concluded that the current strategy of continuous combination therapy will not be possible due to short- and long-term toxicities (including metabolic abnormalities, lipodystrophy syndromes, mitochondrial toxicity, and liver toxicity), adherence challenges, development of resistance, and cost.

The goals of Dr. Fauci's intermittent treatment strategy are to (1) induce a "permanent" new immunologic set point to control virus replication without HAART, similar to that seen in an HIV-infected long-term nonprogressor, and (2) provide intermittent drug-free periods to decrease toxicity, improve adherence, and decrease cost.

He described two studies in progress at the NIH Clinical Center. In one study, 7 patients are receiving 2 months of HAART followed by 1 month off treatment. For the first 3 cycles of this treatment, viral load has returned to undetectable levels when HAART is restarted.

n the second study, 5 patients are receiving a "7 days on/7 days off" cycle of treatment. After 7 cycles of treatment (14 weeks), none of the 5 patients has had a detectable viral load.

These studies are designed to examine the relationship between the CD8+ HIV-specific immune response and the control of viremia. Prior studies have demonstrated that HIV-specific CD8+ memory cells increase when patients have stopped HAART and decrease when it is reinitiated.

If this strategy is shown to be effective, patients may be able to reduce total time on HAART while maintaining a tolerably low level of viremia. As a part of the "structured" approach, if viral load rebounds in excess of a predetermined level during an interruption period, HAART will be reinitiated.

Other benefits of such a strategy, according to Dr. Fauci, are that patients will need only half the amount of drugs that they would have taken had they remained on HAART for the entire year. Dr. Fauci said that 50% less drug probably would be associated with less drug toxicity and lower drug costs.

Comment: A note of caution-Dr. Fauci did caution that this strategy is not ready for prime time and requires more study in larger numbers of patients to determine the optimal duration of the interruptions and the intervening periods of being on therapy. Outcomes to be measured include control of viral load and CD4 count stability. On the other hand, since many of our patients have treatment interruptions for a variety of reasons, it is important to be thinking about the "structured" part of any treatment interruption. Some considerations include an agreement between the patient and provider that even though treatment may be interrupted, follow up of clinical and laboratory parameters needs to continue on a predetermined schedule as a part of this revised treatment plan. This new treatment plan should be written in the medical record and should outline the frequency of visits and laboratory monitoring, the viral load threshold for restarting therapy, and any anticipated changes in the drug regimen once treatment is restarted. Fauci A. Host factors in HIV disease. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Session TuOr36.

Induction/maintenence
New data from the ADAM study reported on the effect of a longer period of induction therapy on the feasibility of treatment deintensification. Antiretroviral-naive patients were treated with a 4-drug induction regimen consisting of stavudine (d4T) + lamivudine (3TC) + saquinavir (initially as hard gel; SQV) + nelfinavir (NFV) for either 26 or 50 weeks. Randomization at week 26 was discontinued following an interim analysis that showed inferior suppression of viral replication during maintenance therapy. At week 50, patients whose plasma HIV-1 RNA levels had been below the limit of quantification (<50 copies/mL) at both weeks 48 and 49 were randomized to maintenance therapy (either d4T + NFV or SQV + NFV) or to continued quadruple therapy. After randomization, monthly monitoring of viral load in plasma was performed. Treatment failure was defined as two consecutive HIV-1 RNA measurements greater than 100 copies/mL.

Of the 65 patients who initially started in the study, 16 patients had been randomized at week 26. Of the remaining 49 patients, 17 patients were randomized at week 50, of whom 6 received d4T + NFV and 4 received SQV + NFV. In each of these arms, 1 patient withdrew after randomization. Treatment failure was observed in 4 of 8 patients who deintensified, compared with 1 of 5 evaluable patients on quadruple therapy (P = .56). Kaplan-Meier analysis indicated that the time to a plasma HIV-1 RNA increase of 400 copies/mL was comparable among those who deintensified at week 26, and those who deintensified at week 50. Thus, the authors concluded that a longer period of induction does not postpone recurrence of viral replication following deintensification.

Comment: The induction maintenance strategy of antiretroviral use has not been shown to be an effective way to maintain viral load suppression. An earlier study by Havlir using triple drug induction and 2-drug maintenance failed to show long-term suppression.

Reijers MH, Weverling GJ, Jurriaans S, et al. The ADAM study: Maintenance therapy after 50 weeks of induction therapy. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPpA1154.

See also:

Reijers MH, Weverling GJ, Jurriaans S, et al. Maintenance therapy after quadruple induction therapy in HIV-1 infected individuals: Amsterdam Duration of Antiretroviral Medication (ADAM) study. Lancet. 1998;352:.

Havlir DV, Marschner IC, Hirsch MS, et al. Maintenance antiretroviral therapies in HIV-1 infected subjects with undetectable plasma HIV RNA after triple-drug therapy. N Engl J Med. 1998;339:.

Simpler, more compact treatment regimens

Several compact regimens were discussed. These various combinations used once or twice daily show some promise in suppressing viral load between 16 weeks and 12 months of therapy, depending on the study. Listed below are the capsule summaries of these presentations.

Patients on an indinavir-containing regimen who had a viral load of less than 50 copies/mL were offered the opportunity to switch to once-daily indinavir/ritonavir (1200 mg/100 mg). Matched controls were selected from the clinic's database.

Twelve patients have completed 16 weeks of follow-up. Their baseline characteristics included viral load of less than 50 copies/mL for a mean of 16 months and a mean CD4+ count of 369 cells/mm³. Viral rebound to more than 400 copies/mL has occurred in 1 control case, but in none of the simplification patients. However, 2 simplification and 3 control patients had a viral measurement of more than 50 copies/mL at week 16. It is not known whether these represent transient but modest increases ("blips") of viral replication. Two simplification patients and 1 control patient have reported renal calculi. This may have been expected because the peak indinavir levels reported for this regimen are around 80% higher than with 3-times-daily dosing.

Maggiolo F, Rizzi M, Finazzi G, Suter F. Once-a-day indinavir therapy in virologically controlled HIV+ persons. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPpA1155.

A study was presented evaluating a once-daily 400 mg dose each of didanosine (ddI) and nevirapine vs. twice-daily dosing of these agents, in each case with twice-daily stavudine (40 mg), in asymptomatic antiretroviral-naive patients with CD4+ counts of more than 500 cells/mm³ and HIV RNA higher than 5000 copies/mL. Using intent-to-treat analysis, the proportions of patients with HIV RNA levels less than 200 copies/mL at 12 months were 73% and 68% in the once- and twice-daily arms, respectively, while the proportions with less than 5 copies/mL were 40% and 45%, respectively. No differences in CD4+ count increase were observed.

Felipe G, Hernando K, Maria Antonia S, et al. An open randomized study comparing d4T + ddI and nevirapine (Qd) vs. d4T + ddI and nevirapine (Bid) in antiretroviral naive chronic HIV-1 infected patients in very early stages: Spanish scan study. Final results. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPpA1156.

Another study of once-daily dosing combined two NNRTIs, efavirenz (EFV), and nevirapine (NVP) with ddI. The rationale for combining NNRTIs is that the total NNRTI exposure will be increased, potentially leading to additive antiretroviral effects. In this study, efavirenz is dosed at 800 mg/day with standard-dose NVP and ddI 400 mg/day.

Among the treatment-naive patients, the mean baseline viral load was 4.59 log10 copies/mL, and the mean baseline CD4+ count was 351 cells/mm³; after 9 months of therapy, 12 of 12 evaluable patients had HIV RNA levels of less than 400 copies/mL, and the mean CD4+ count had increased by 351 cells/mm³. At baseline, 9 of 11 treatment-experienced patients had viral load of less than 400 copies/mL, and a mean CD4+ count of 368 cells/mm³; after 9 months, 9 of 9 had HIV RNA levels less than 400 copies/mL , and the mean CD4+ count had increased by 203 cells/mm³.

Five (19%) of the 26 patients discontinued therapy, 2 because of rash, 3 because of central-nervous-system effects. This pilot study suggests that double NNRTI therapy is a potentially attractive approach, and further evaluation of once-daily NVP + EFV + ddI is warranted.

Jordan W, Jefferson R, Yemofio F, et al. Nevirapine (NVP) + efavirenz (EFV) + didanosine (ddI): A very simple, safe, and effective once-daily regimen. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPeB3207.

Lopinavir/ritonavir (Kaletra®) in treatment-experienced patients.
At the conference, the first results were presented of lopinavir/ritonavir (ABT-378) in patients who have been exposed to multiple PIs. Fifty-seven subjects who were nonnucleoside reverse transcriptase inhibitor (NNRTI)-naive and had been treated with an average of 7 antiretrovirals, including an average of 3 PIs, were treated with 1 of 2 doses of lopinavir/ritonavir (400 mg/100 mg or 533 mg/133 mg twice daily) plus efavirenz and the NRTI of the investigators' choice.

At baseline, 68% had 4-fold or greater decreases in susceptibility to 3 or more PIs, and 43% had a more than 10-fold decrease in susceptibility to lopinavir/ ritonavir when compared with wild-type virus. Despite this fact, 80% of subjects treated with lopinavir/ritonavir 400 mg/100 mg and 96% of subjects treated with 533 mg/133 mg who continued to receive study therapy achieved HIV RNA levels of less than 400 copies/mL; when missing values were considered treatment failures, these proportions were 69% and 82%, respectively.

These study results compare favorably with any previous study in this patient population. Clearly, efavirenz contributed to treatment success in these patients, but without the substantial PI activity of lopinavir/ritonavir, the effects of the NNRTI would likely have been relatively transient. Of note, efavirenz lowered lopinavir/ritonavir trough levels by approximately 33%, which might explain the better success rates in the higher dose group.

Thompson M, Murphy R, Gulick R, et al. Analysis of duration of virologic response in two Phase II studies of ABT-378/ritonavir (ABT-378/r) at 72 weeks. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPeB3197.

Kessler H, Eron J, Hicks CB, et al. Analysis of safety data from ABT-378/ritonavir (ABT-378/r) in two Phase II clinical trials. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPeB3198.

Clumek N, Girard PM, Telenti A, et al. ABT-378/ritonavir (ABT-378/r) and efavirenz: 16-week safety/efficacy evaluation in multiple PI experienced patients. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPeB3196.

Withdrawl of secondary prophylaxis for pneumocystis
Following up on their previous finding that primary PCP prophylaxis can be safely stopped in patients with sustained increases in absolute CD4+ counts to above 200 cells/mm³ and in CD4+ percentage to greater than 14%, the Swiss HIV Cohort investigators have embarked on a similar study of secondary PCP prophylaxis. The CD4+ count criteria were identical to the earlier study and had to be sustained for at least 12 weeks. Preliminary data on 55 subjects with a previous history of PCP were reported. Median nadir CD4+ count was 23 cells/mm³, and median CD4 count at study entry was 394 cells/mm³. Cotrimoxazole was the agent discontinued in 82% of the patients. After a median follow-up of 10 months, there have been no cases of PCP or toxoplasmosis.

Likewise, in an Italian study in which 124 HAART responders with CD4+ counts of at least 200 cells/mm³ were randomized to discontinue secondary PCP prophylaxis or not, no cases of PCP were diagnosed after a median of approximately 8 months. A similar study conducted in Spain involving 113 randomized subjects also detected no PCP, AIDS progression, or death among the subjects after a median 12 months of withdrawal of PCP prophylaxis.

It is important to emphasize that each of these studies and those of primary PCP prophylaxis discontinuation targeted patients with sustained and usually robust CD4+ cell count elevations. Median CD4+ counts were well above 200 cells/mm³, and in the Swiss studies, subjects were required not only to have a high absolute count but a CD4+ percentage of more than 14%. Drops in CD4+ cell counts below 200 cells/mm³ or 14% should trigger reintroduction of PCP prophylaxis, particularly in those with a history of PCP.

Furrer HJ, Opravil M, Bernasconi E, et al. The Swiss stopcox2 study: Is it safe to discontinue secondary PCP prophylaxis? Experience of the Swiss HIV cohort study. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract MoPeB2276.

Mussini C, Pezzotti P, Borghi V, et al. An open, controlled, randomized study on discontinuation of secondary prophylaxis for Pneumocystis carinii pneumonia in patients with AIDS. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract MoPeB2278.

Lopez JC, Miro JM, Pena JM, et al. Discontinuation of secondary Pneumocystis carinii prophylaxis in HIV-1 infected patients after immunological recovery with HAART: Results of a prospective, randomized and multicentric clinical trial (the GESIDA 04/98 study). Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract MoPeB2288.

Increased lactic acid levels are more common than lactic acidosis
Two presentations concluded that increased lactic acid levels in patients taking NRTIs is common, usually mild, and reversible. It more likely results from hepatic rather than muscle dysfunction. There is still no consensus about whom to monitor or how often to perform monitoring. There is agreement, however, that serious abnormalities should prompt discontinuation of NRTIs. It is not clear what should be done in the case of mild abnormalities, since mild disease may be seen in up to one third of patients, but lactic acidosis is very uncommon.

There was also some discussion about potential therapies for hyperlactatemia. A 1000 mg twice-daily dose of carnitine was suggested. The question of using acetyl carnitine, which is said to be more bioavailable than carnitine, was raised, but no data exist to recommend this compound.

In another poster presentation, lactic acid levels were measured in all patients in an outpatient clinic over 3 months. The 211 patients were divided into 3 groups based on treatment: no treatment, AZT-based treatment, and d4T-based treatment. Normal lactic acid levels were seen in 92%, 81%, and 72% of the patients, respectively. All but one patient with abnormal lactic acid levels had mild elevations.

Harris M, Tesiorowski A, Montaner JSG. Screening for nucleoside-associated lactic acidosis. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPpB1233.

Vrouenraets S, Treskes M, Regez RM, et al. The occurrence of hyperlactatemia in HIV-infected patients on NRTI treatment. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa. Abstract TuPpB1234.

Lipodystrophy
In an attempt to uncover the cause(s) of lipodystrophy (fat redistribution), researchers have found that skin biopsies from the hip or "buffalo hump" area of HIV-positive patients taking NRTI (nucleoside reverse transcriptase inhibitor) drug therapy with or without PI (protease inhibitor) drug therapy have abnormal appearances of fat cells (adipocytes).

Electron micrographs showed "abnormal adipocyte ultrastructure with mitochondrial abnormalities, cytoplasmic lipid accumulation, and loss of adipocyte volume." In addition, there was associated fat cell loss and "lipogranuloma" formation. Also, Dr. S. Mallal, from the Royal Perth Hospital in Australia, found that these changes were seen in regions of fat loss (lipoatrophy) and accumulation ("buffalo hump"). HIV-positive patients without any anti-HIV treatment had no abnormalities on electron microscopy, but occasional "lipogranulomata" were seen on regular microscopy. Mitochondrial toxicity has been proposed as a common mechanism of side effects from certain drugs in the NRTI class. This is the first visual evidence of possible anti-HIV drug toxicity associated with lipodystrophy or fat redistribution syndrome.

Mallal S, et al. Light and electron microscopy findings in subcutaneous fat in anti-retroviral treated and naïve HIV-infected patients. Program and Abstracts of the XIIIth International AIDS Conference and Late Breaker poster presentation LbPeB7054.

Antiretroviral News from the XIIIth International AIDS Conference

Switching from protease inhibitors to efavirenz (Sustiva) maintains viral load suppression in HIV-positive patients
Efavirenz, a non-nucleoside reverse transcriptase Inhibitor (NNRTI), may provide an effective and safe alternative to long-term protease inhibitor treatment. Efavirenz was potent enough to maintain a viral load of less than 50 copies/mL in patients who switched from protease inhibitor-containing regimens.

Abacavir, a nucleoside reverse transcriptase inhibitor (NRTI), was added to the existing regimen of a protease inhibitor and two NRTIs, and Sustiva was substituted for the protease inhibitor after 6 weeks. The first 16 patients evaluated after 24 weeks of treatment maintained a viral load of less than 50 copies/mL. Fasting triglyceride, HDL-, and LDL-cholesterol levels in the 16 patients improved significantly compared to baseline.

Switching to efavirenz reduces triglyceride levels, increases viral load reduction
Results from a study presented by Franco Maggiolo, MD, Division of Infectious Diseases, Ospedali Riuniti di Bergamo, Italy, indicate that switching patients with lipodystrophy and metabolic changes taking protease inhibitors to a protease inhibitor- sparing regimen including efavirenz may lead to a significant reduction of triglycerides and enhance virological response.

After one month of using efavirenz in combination with their previously prescribed NRTIs, the mean triglyceride level of the patients studied was reduced to 362 ± 233 mg/dL (p < .004); at four months it was 325 ± 157 mg/dL (p < .008) and at 8 months it was 359 ± 149 mg/dL (p < .03).

First once-daily nelfinavir (Viracept®) data presented
In a study in healthy volunteers, a once-daily dosing of the HIV protease inhibitor nelfinavir (Viracept) in combination with small doses of ritonavir resulted in comparable pharmacokinetics xe.g., area under the curve (AUC)xof nelfinavir relative to twice-daily dosing (bid) of nelfinavir alone. The three study arms used the following doses: nelfinavir 1875 mg, 2500 mg, and 2500 mg, with respective ritonavir doses of 200 mg, 100 mg, and 200 mg. The combined concentrations of nelfinavir and its active metabolite, so-called M8, showed similar or higher results for all three pharmacokinetic parameters in these groups. No serious adverse events were observed in this study.

Other compact regimens reviewed at the conference include once-daily indinavir plus ritonavir (1200 mg and 100 mg, respectively; Maggiolo et al, Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa, Abstract TuPpA1155); a once-daily 400 mg dose each of didanosine (ddI) and nevirapine (NVP) versus twice-daily dosing of these agents, in each case with stavudine, 40 mg bid (Felipe et al, Abstract TuPpA1156); efavirenz, 800 mg/d, with standard-dose NVP and ddI, 400 mg/d (Jordan et al, Abstract TuPeB3207); once-daily nelfinavir plus ritonavir (2500 mg and 200 mg, respectively; Hsyu et al, Abstract LbPeB7049); and a compact four-drug regimen of co-formulated zidovudine-lamivudine (Combivir twice a day, abacavir 300 mg twice a day, and efavirenz, 600 mg/d; deTruchis et al, Abstract TuPeB3208).

Comment: the trend toward more convenient drug dosing with twice daily regimens or even daily regimens can help improve quality of life for patients and improve adherence. These studies are a good start in terms of assessing the pharmacokinetics of simpler regimens that use ritonavir to boost PI levels. Before these regimens are ready for prime time, however, most experienced providers would like to see more data on viral load suppression.

Adherence
A study comparing adherence with twice-daily nelfinavir and once-daily non-nucleosides (NNRTI) (nevirapine, efavirenz) showed that people on the once-daily NNRTI were as likely to miss doses as patients taking twice-daily neflinavir. For example, when asked if they missed all doses of medication at least one day in the past month, 37% of the nelfinavir group and 38% of the NNRTI group said they had.

Luber et al. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000, Durban, South Africa, Abstract ThPeB4991.

In a study of patient self-report, Mannerheimer and associates (Abstract TuOrB421) reported the results of a self-completed questionnaire and noted that changes in viral load and having a viral load of below 50 copies/mL correlated significantly with self-reported adherence.

Nakamisha, in a presentation of interviews of 4200 patients on antiretrovirals, from the CDC, (Abstract TuPpD1201) reported that 68% always, 25% usually, 5% sometimes, and 2% never or rarely adhered to therapy. Factors associated with nonadherence included white/ non-Hispanic identity, AIDS diagnosis, time since HIV diagnosis more than one year, use of PIs, treatment duration more than two years, taking more than three medications, being in jail, and crack cocaine use in the past year. Interestingly, factors not predictive of nonadherence included CD4 cell count, black ethnicity, injection drug use, age, gender, homelessness, and receipt of public assistance.

Triple Nucleoside Combination (Trizivir) Introduced
The triple nucleoside combination of AZT (300 mg), 3TC (150 mg), and abacavir (300 mg) has approved for use in Europe. A new drug application has been submitted to the U.S. Food and Drug Administration.

This combination, called Trizivir, was reviewed in two papers presented at the recent Durban Conference.

In a pharmacokinetics study, Cahn et al showed that the area under the curve (AUC) for Trizivir was similar to a triple nucleoside combination of combivir + abacavir.

Cahn P, et al. Preliminary efficacy, adherence and satisfaction with COM/ABC versus COM/IDV, an open-label randomised comparative study (CNAB3014). Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000. Abstract WeOrB606.

In an open-label study comparing efficacy and safety of Trizivir to a protease inhibitor (PI)-containing regimen, Matheron et al concluded that the triple nucleoside combination at 48 weeks was safe with viral suppression equal to the PI-containing regimen in antiretroviral-naïve patients.

Matheron S, et al. An open label study to compare efficacy and safety of the triple nucleoside analog combination Combivir/abacavir versus a protease inhibitor containing regimen in antiretroviral naïve adults. CNAF3007/Ecureuil. Program and Abstracts of the XIIIth International AIDS Conference, July 9-14, 2000. Abstract WeOrB605.

Still to be studied is the role of Trizivir in treatment adherence; additional studies will also be conducted on this combination in treatment-experienced patients.

Items of Interest from the CDC Prevention News Mailing List

Hepatitis C and HIV in healthcare workers
Research reported at the International Conference on Emerging Infectious Diseases in Atlanta shows that healthcare workers face a 20 to 40 times greater risk of contracting hepatitis C virus (HCV) from an accidental needle stick compared to the risk of contracting HIV. Dr. Robert Ball polled 66 facilities in South Carolina and found that in the 53 hospitals that responded, a total of 1,668 healthcare workers were exposed to either HCV or HIV in 1998. Of the source patients, 1,451 had been tested for HCV and 1,508 had been tested for HIV, with 5.2 percent overall testing seropositive for HCV and 2.3 percent testing seropositive for HIV; the rates for the general population are 1.5 percent for HCV and 0.3 percent for HIV. Ball concluded that increased efforts to highlight the risk of HCV infection to healthcare workers.

See the conference summary at: <http://www.cdc.gov/iceid/> and http://view.abstractonline.com/asm/ abstractviewer.asp

HIV risks in young men who have sex with men
Researchers from the Centers for Disease Control and Prevention used the Young Men's Survey (YMS), a cross-sectional, multisite survey conducted between 1994 and 1998, to estimate the prevalence of HIV and related risk behaviors in young men who have sex with men (MSM). The researchers visited a total of 194 bars, dance clubs, and social groups frequented by young MSM in seven major cities. In all, 3,492 MSM between the ages of 15 and 22 were interviewed and tested for HIV. According to the CDC, overall HIV prevalence for the group was 7.2 percent; the rate was lowest in Seattle and highest in New York City. MSM aged 20 to 22 had the highest HIV prevalence; none of the 15-year-olds were infected with HIV. The multivariate-adjusted HIV prevalence was higher among African Americans, young men of mixed or other race, and Hispanics compared to whites, Asian Americans, and Pacific Islanders. Factors strongly linked to HIV infection among these men include being African American, mixed, or other race; having ever had anal sex with a man; or having had sex with at least 20 men. The prevalence of unprotected anal sex in the last six months was 41 percent overall. The results indicate "a critical and widespread public health problem," the researchers conclude, "and underscore a need to evaluate and intensify prevention efforts for young MSM," especially for African Americans, men of mixed race or ethnicity, Hispanics, and young adolescents.

Valleroy LA, MacKellar DA, Karon JM, et al. Journal of the American Medical Association (JAMA). 2000;284:198. Available at <www.jama.com>.

AIDS impacts on African school children
A sobering report on how AIDS impacts on the education of children in Africa was just published. Last year, over 2,000 teachers in Kenya died from HIV or AIDS, an East African newspaper reported. The newspaper, quoting the recent UNICEF Progress of Nations 2000 report, noted that 95,000 primary school students in Kenya had no teachers due to the AIDS epidemic last year. The report estimated that 860,000 students in sub-Saharan Africa lost their teachers to the disease.

Impact of antiretroviral therapy on children with HIV
Researchers studied data from the Italian Register for HIV Infection in Children to determine the effect of availability of antiretroviral therapy on reducing mortality in perinatally HIV-infected children. The study involved more than 1,100 children with perinatally acquired HIV infection who were born between November 1980 and December 1997. The data show that 52.5 percent of the infected children were female, and 57 percent were born to a mother who injected drugs or had sex with an intravenous drug user.

The median follow-up of 5.9 years saw 421 children die from HIV-related causes. Treatment of the children with antiretroviral therapy (AZT monotherapy) began in 1987, and double therapy started in 1993. By 1998, triple combination therapy was the most common treatment. The most commonly used triple therapies were stavudine, lamivudine, and ritonavir or nelfinavir or indinavir; and zidovudine, lamivudine, and ritonavir or nelfinavir or indinavir. According to the authors, "Our study shows that although the survival of HIV-infected children in Italy remained unchanged up to 1995, it has significantly improved since 1996, with a more than 30 percent reduction in the adjusted risk of death for children at risk in the period vs. those at risk in the period ."

De Martino M, Tovo P, Balducci M, et al. Reduction in mortality with availability of antiretroviral therapy for children with perinatal HIV-1 infection. Journal of the American Medical Association (JAMA). 2000;284:190. Available at:<www.jama.com>.

A review of restoration of immune function in HIV
In another JAMA article in the same issue devoted to HIV, Lederman and Valdez note that the magnitude of immune restoration seen after treatment with HAART varies substantially among treated persons and that restoration is generally incomplete. Nonetheless, even partial immune restoration is sufficient to provide protection from most major opportunistic infections; these risks can be largely predicted by the number of circulating CD4 cells. Limited data suggest that treatment earlier during the course of HIV infection may result in greater preservation of immune function, though this has not been studied in great detail. How much immune restoration will be enough to ensure long-term survival in persons with HIV infection remains an open question, according to the authors, who review the subject in detail in terms of our current understanding and of future directions.

Lederman M, Valdez H. Immune restoration with antiretroviral therapies: Implications for clinical management. Journal of the American Medical Association (JAMA). 2000;284;. Available at:<www.jama.com>.

To subscribe to the CDC Prevention News mailing list, send a blank message to: .