Approximately 12,500 people attended the XIII International AIDS Conference in Durban, South Africa, on July 9-14, 2000. The title was "Break the Silence," a theme that was woven throughout the conference. Attendees included physicians, researchers, other health professionals, policy makers, and persons living with HIV or AIDS representing more than 178 countries. In addition, there were more than 1,000 volunteers.
This was the first time that the biannual international AIDS conference occurred in a resource-poor or developing country; all others have been hosted in North America, Europe, or Japan. The location is particularly important, given that approximately three-fourths of the estimated 34 million persons with HIV or AIDS worldwide live in sub-Saharan Africa (figures are as of December 1999). The HIV/AIDS epidemic is particularly severe in South Africa, where approximately four million people are HIV positive, including one-fourth to one-third of all pregnant women.
More than 5,000 abstracts were presented, including 14% in oral presentations, 10% in oral poster presentations, and the remaining 76% in standard poster presentations.
There were five scientific tracks: Basic Science; Clinical Science; Epidemiology, Prevention, and Public Health; Social Science; and Rights, Politics, Commitment, and Action. At any one time, up to 13 concurrent sessions took place. Many oral sessions are available at www.webcast.aids2000.com. This program is sponsored by Gilead Sciences.
The conference was organized by five groups: the Global Network of People Living with HIV/AIDS (GNP+), the International AIDS Society (IAS), the International Council of AIDS Service Organizations (ICASO), the International Community of Women Living with HIV/AIDS (ICW), and the Joint United Nations Programme on HIV/AIDS (UNAIDS). There were 13 principal sponsors: Abbott Laboratories, Agouron Pharmaceuticals, Boehringer Ingelheim, Bristol-Myers Squibb, Glaxo Wellcome, DuPont Pharmaceuticals, Merck Sharp & Dohme, Roche, Pharmacia & Upjohn, Dimension Data, Cisco Systems, Dell E Com, and Microsoft. Other sponsors included the Department of Health of South Africa, the U.S. National Institutes of Health (NIH) Office of AIDS Research (OAR), Gilead Sciences, the Ford Foundation, Trimeris, and Bayer Diagnostics.
There was much controversy surrounding the Durban conference in the months before it took place. Issues included safety concerns for attendees, a threatened boycott by potential attendees, and South African President Thabo Mbekis statements denying that HIV is the cause of AIDS. While the first two concerns essentially did not materialize, President Mbeki and his statements continue to detract from much of the real issues of this pandemic in the year 2000. Notwithstanding, the conference overall was remarkable, given the perspectives and inspiration from African participants.
Events that took place before the conference added to the controversy. Several thousand physicians, scientists, and other researchers signed the so-called Durban Declaration affirming that HIV is the cause of AIDS; the document was published in the journal Nature. Also, in the afternoon before the opening ceremony, approximately 4,000 protestors, including persons with HIV/AIDS, participated in a rally in downtown Durban demanding access to anti-HIV drugs in resource-poor countries. Former South African First Lady Winnie Madikizela-Mandela was a speaker and participant.
The keynote speaker at the opening ceremonies was South African President Thabo Mbeki. Even though he did not claim or deny that HIV is the cause of AIDS, he did use the terms "HIV" and "HIV/AIDS" several times during his speech. Having encountered fringe scientists on the Internet as well as so-called HIV deniers, President Mbeki decided that other social and health issues need to be considered in countries like South Africa. Such concerns include the poverty and malnutrition that continue to separate relatively wealthier countries of the north (and west) from those in sub-Saharan Africa. Many are concerned because the South African presidents stance allows him to deny access of anti-HIV therapy to many South Africans with HIV, including pregnant women.
One of the high moments of the conference was the closing speech by former South African President Nelson Mandela, who received a standing ovation. Mr. Mandela emphasized, "We need to break the silence, banish stigma and discrimination, and ensure total inclusiveness within the struggle against AIDS. Let us combine our efforts to ensure a future for our children. The challenge is no less."
Nelson Mandela Foundation. Closing address at the XIII International AIDS Conference.
President Thabo Mbeki. Opening ceremony at the XIII International AIDS Conference.
One of the more interesting groups of presentations in Durban was in the area of highly active antiretroviral therapy (HAART) interruptions. The research included presentations by NIH researchers Mark Dybul, MD, and Anthony Fauci, MD, Head of the U.S. National Institute of Allergy and Infectious Diseases (NIAID). Dr. Dybul describes their research in this issue of BETA.
Dybul, M. and others. Short cyclic intermittent HAART: a pilot study. Abstract and late-breaker oral presentation LbOr12.
Fauci, T. Host factors in HIV disease. Abstract and oral plenary presentation TuOr36.
Several presentations addressed potential mechanisms for decreasing HIV disease progression in resource-poor countries. Strategies are based upon the observations that many AIDS-related opportunistic infections (OIs) increase HIV RNA viral load, which in turn may increase HIV disease progression and AIDS death rates. Therefore, treating OIs -- which is often not done in resource-poor countries -- would be expected to decrease HIV disease progression, decrease AIDS death rates, and likely decrease HIV transmission rates. Mauro Schechter, MD, from Brazil discussed one example: TMP-SMX (Bactrim, Septra, cotrimoxazole) to treat and help prevent PCP (Pneumocystis carinii pneumonia) and several other bacterial infections. In his plenary presentation called "Advances in Treatment of AIDS," Dr. Schechter reviewed statistics from several countries and regions that showed a decline of AIDS-related deaths before the advent of HAART that included a protease inhibitor (PI) or non-nucleoside reverse transcriptase inhibitor (NNRTI) drug. That decline is at least partly due to wider use of antibiotics to treat and prevent OIs in AIDS.
In another presentation, Dr. Z. Bentwich and colleagues from Kaplan Medical Center in Rehovot, Israel, expanded their previous reports about "deworming." Dr. Bentwichs team evaluated Ethiopians who had immigrated to Israel, many of whom had chronic worm infection (helminth infestation) of their intestinal tracts (colon), HIV infection, or both. They found that immune cells from Ethiopians with intestinal worms and without HIV infection were more susceptible to HIV infection in laboratory tests. This was due to immune system activation associated with a higher level of CCR5 receptors (which HIV uses to enter cells) and a lower level of beta chemokines (intercellular messengers that decrease the risk of cellular HIV infection). Eggs in the stools of Ethiopians with worm infestation and HIV infection significantly correlated with higher HIV viral loads in blood plasma. In addition, HIV RNA decreased significantly when the worms were effectively treated. The authors concluded that helminthic infestation and immune activation "account for rapid spread [transmission] and progression of HIV infection in helminthic-infested regions and will probably undermine HIV vaccine success." They also claimed that "eradication of helminths is feasible, can be applied in Africa and the developing world, and [may] have far-reaching impact on the AIDS epidemic." Treating intestinal worms would likely decrease HIV transmission rates by the sexual route and from mothers to their newborns. Most of the drugs used to treat intestinal helminths are generic and relatively inexpensive.
Bentwich, Z. and others. Helminthic infections have a major impact on pathogenesis of AIDS and development of HIV protective vaccines in Africa and the developing world. Abstract and late-breaker LbOr29.
Schechter, M. Advances in treatment of HIV/AIDS. Abstract and plenary oral presentation TuOr37.
Results of the first government-sponsored study of marijuana effects in persons with HIV were presented in Durban. The results showed that three weeks of smoking a marijuana cigarette three times daily does not increase HIV viral load in persons taking HAART. The randomized, placebo-controlled, and partially blinded study compared inhaled (smoked) marijuana with oral dronabinol (Marinol 2.5 mg, a synthetic pill with active marijuana derivative) and oral placebo (inactive drug), each three times daily before meals. The marijuana was supplied by the U.S. government, which also funded the study. The lead author was Donald I. Abrams, MD, of the University of California, San Francisco (UCSF). The PI used was either indinavir (Crixivan) or nelfinavir (Viracept).
Sixty-seven subjects (4% women, 6% male-to-female transgender, 52% non-White) enrolled and remained in a clinical research center for 21 days of the study. Two-thirds were older than 40. The median baseline CD4 cell count was 300 cells/mm3, with a median HIV RNA less than 50 copies/mL. A prior AIDS OI had occurred in 49%.
After the three-week study, the inhaled marijuana group had a mean viral load decrease of -0.15 log, the dronabinol group had a -0.19 log decrease, while the placebo arm had a 0.06 log copies/mL increase (in an "as treated" analysis, including only those participants who completed the study). None of these changes are statistically significant. In 50-55% of each arm, there was no significant change in viral load. CD4 cell count changes and other parameters are pending. There was a 93% completion rate. Weight gains occurred in all three arms: inhaled marijuana (3.5 kilograms, or kg); dronabinol (3.2 kg); and placebo (1.3 kg). (Kilogram weight divided by 0.454 equals weight in pounds, or lbs.) This is the first study that has shown a weight gain resulting from dronabinol.
There were three adverse events in the inhaled marijuana arm: one subject had moderate neuropsychiatric symptoms (this person chose to leave the study), one subject developed moderate tachycardia (fast heart beat; the subject remained on study), and one subject developed severe neuropsychiatric symptoms (treated with lorazepam [Ativan] antianxiety drug; the subject remained on study). In the drona-binol arm, there were four adverse events: one subject developed moderate neuropsychiatric symptoms (the subject chose to leave the study), one person had moderate headache and nausea (the subject chose to leave the study), one subject had kidney-area pain, possibly related to indinavir (possible a kidney stone; the event was not related to dronabinol and the subject remained on study), and one subject died two months after the study from hemorrhage (bleeding) due to portal hypertension (high blood pressure in the portal vein leading to the liver, usually due to cirrhosis or liver scarring; all unrelated to dronabinol). There were no adverse events in the placebo arm.
Other analyses about participants in this study are pending. The viral load results are only short-term. It is possible that longer-term effects might not be as favorable in terms of viral load. The benefits of inhaled marijuana (decreased nausea, increased appetite, and weight gain) must be balanced against the documented side effects. These include an increase in respiratory infections, decreased immune system function, and a decrease in short-term memory.
Abrams, D.I. and others. Short-term effects of cannabinoids on HIV-1 viral load. Abstract and late-breaker poster presentation LbPeB7053.
Forty-eight-week results of T-20 in treatment-experienced subjects were presented in Durban by Cal Cohen, MD, of the Community Research Initiative in Boston. The late-breaker poster showed that one-third had at least a 1 log reduction in viral load and that no participant withdrew due to side effects of T-20. A total of 71 persons (8% women, 20% non-White) were enrolled. Subjects added T-20, 50 mg self-injected twice daily, to a new anti-HIV drug regimen that was based on drug resistance testing.
Study T20-205 was a Phase II, open-label, "rollover" (from other T-20 studies) trial. The baseline, median number of prior anti-HIV medications was ten, and 80% of participants had previously taken at least one drug from each of the first three anti-HIV drug classes. Prior PI therapy was taken by 79% of enrollees. The median baseline viral load was 4.8 log (63,095) copies/mL with a CD4 cell count of 133 cells/mm3. Forty-one subjects (58%) remained on study at the end of 48 weeks. Yet none of the 42% who discontinued did so due to adverse events related to T-20.
The results showed the following, using a strict "intent-to-treat" analysis (all enrollees included). An undetectable HIV RNA viral load (limit 50 copies/mL) was achieved by 13%. Another 10% had between 50 and 400 copies/mL, while yet another 10% had greater than a 1 log (10-fold) reduction in viral load and were at a level greater than 400 copies/mL. Among those still remaining on study, the mean decrease in viral load at 48 weeks was approximately -1.3 log copies/mL (note that a decrease is associated with decreased HIV disease progression). Among the 30 subjects who discontinued, 14 (20% of total) did so due to minimal HIV RNA reduction (less than -0.5 log copies/mL), seven (10% of total) voluntarily withdrew from study, three (4%) withdrew due to adverse events unrelated to T-20, and two (3%) were nonadherent to dosing. CD4 cell count changes were not reported.
Reported adverse events included mild-to-moderate injection site reactions in 71%, while 21% had severe adverse events possibly related to T-20. Serious adverse events possibly related to T-20 occurred in 10%. These included increased liver enzymes (alanine aminotransferase [ALT], alanine aspartate transferase [AST], and/or gamma glutamyl transpeptidase [GGT]), altered mental state, anemia (low red cell count), increased amylase (a pancreatic enzyme) or neutropenia (low white cell count). In a separate presentation, Joe Eron, MD, of the University of North Carolina at Chapel Hill reported that participants in the study viewed injection of T-20 as convenient, with little distraction from their daily activities.
T-20 is the first drug in the class of fusion/entry inhibitors. Even though the viral load results in this study are less than stellar, this drug appears promising for highly treatment-experienced participants. T-20 is under development by Trimeris and Roche.
Cohen, C. and others. 48-week analysis of patients receiving T-20 as a component of multidrug salvage therapy. Abstract and poster presentation LbPp116.
Eron, J. Entry inhibitors: the next frontier in HIV therapy. Satellite symposium.
Several presentations focused on the double-PI drug formulation lopinavir/ritonavir (ABT-378/r, or Kaletra) in combination therapy. [Ed. note: Kaletra was approved by the FDA this past September.]
The benefits of lopinavir/ritonavir combination therapy in so-called salvage therapy were presented for persons who had a detectable HIV viral load despite taking a second PI drug. The doses of lopinavir/ritonavir twice daily were 400/100 mg (three capsules) or 533/133 mg (four capsules). The regimen also included efavirenz (a non-nucleoside reverse transcriptase inhibitor) plus two new NRTI drugs. Study subjects (21% women and 12% non-White) had never taken NNRTI drugs. The lead author of Study M98-957 was Nathan Clumeck, MD, from St. Pierre University Hospital in Brussels, Belgium.
For the 29 participants taking the 400/100 mg lopinavir/ritonavir dosing, the mean baseline HIV RNA was 4.6 log (39,810) copies/mL with a CD4 cell count of 230 cells/mm3. For the 28 participants taking the 533/133 mg dosing, the mean baseline viral load was 4.4 log (25,118) copies/mL, with a CD4 cell count of 325 cells/mm3.
Results of the Phase II study were available up to 24 weeks. Even though the 57 participants had taken an average (mean) of three prior PIs at study entry, 69% (400/100 mg arm) to 82% (533/133 mg arm) achieved an undetectable HIV RNA viral load. (Results were with a lower limit of 400 copies/mL and a strict intent-to-treat analysis, which includes all enrolled subjects.) The mean CD4 cell count increase was 48 cells/mm3 in the 400/100 mg arm and 41 cells/mm3 in the 533/133 mg arm.
At baseline, 43% of the 57 participants had an HIV isolate (strain) that had at least a 10-fold decreased susceptibility (resistance) to lopinavir, when compared with the EC50 (effective inhibitory concentration for 50% of viruses) of wild-type, or nonresistant, HIV. Baseline HIV isolates from 68% of participants had at least a 4-fold increase in the EC50 (i.e., decreased susceptibility or resistance) to at least three PI drugs. Phenotypic resistance testing was performed using the PhenoSense HIV test from ViroLogic. (A phenotypic drug resistance test measures the ability of a persons HIV isolate to grow in the presence of each anti-HIV drug. The protease and reverse transcriptase genes from the individuals isolate are inserted into a laboratory isolate before testing.)
The most common adverse events that were at least moderate (grade 2 or higher) included diarrhea (7% in the 400/100 mg arm and 14% in the 533/133 mg arm) and weakness, or asthenia (also 7% in the 400/100 and 14% in the 533/133 mg arm). In the 400/100 and 533/133 mg arms, respectively, severe or life-threatening (grade 3 or 4) laboratory events included increases in the following: glucose greater than 250 mg per deciliter, or mg/dl (10% and zero); ALT greater than five times the upper normal limit (zero and 4%); total cholesterol greater than 300 mg/dl (28% and 36%); triglycerides (lipids or fats) greater than 750 mg/dl (31% and 36%); and amylase greater than twice the normal upper limit (zero and 7%). However, the ratio of total cholesterol to HDL (high density lipoprotein or "good") cholesterol was not significantly changed in either arm, when compared with baseline (another measurement of heart artery disease risk). In each of the two study arms, a grade 3 or 4 decreased neutrophil (white blood cell) count (below 750 cells/mm3) occurred in 7%.
There were seven total discontinuations (12%) before 24 weeks in both arms. Two subjects in each arm discontinued due to adverse events, including central nervous system (CNS) side effects in two persons, stomach-colon ("GI") problems in one subject, and lactic acidosis (likely due to the NRTI drug[s]) in another subject. Two participants in the 400/100 mg arm and one in the 533/133 mg arm discontinued due to virologic "failure" (detectable HIV RNA).
Dr. Clumeck concluded that the five-drug combination of lopinavir/ritonavir, efavirenz, and two NRTIs showed significant potency after 24 weeks in this small study of persons who had prior multi-PI-drug experience and who were naive to NNRTIs. It is noteworthy that the percentage of persons in each study arm that achieved viral undetectability is higher than in many studies of treatment-naive persons. Part of this would be due to the high inhibitory quotient of lopinavir, defined as the trough (lowest) drug level (of lopinavir) achieved divided by the EC50 (see above) of the individuals HIV isolate.
Dr. Clumeck also reported that lopinavir drug levels are decreased when the Kaletra 400/100 mg formulation is combined with efavirenz, as compared with lopinavir/ritonavir (same dose) without efavirenz. Specifically, the trough lopinavir concentration is reduced by approximately 33%, while the AUC (area-under-the-curve, or total drug exposure) concentration is decreased approximately 25%. The maximal, or peak, concentration change was not reported. However, when the experimental 533/133 mg dose of lopinavir/ritonavir was combined with efavirenz, the lopinavir concentrations achieved were similar to those taking the 400/100 mg dose without efavirenz. The drug levels of efavirenz in the current study were reported to be the same in both study arms. Due to the viral load results and drug interaction studies, the authors reported that all ongoing participants subsequently have been placed on the 533/133 mg dose of lopinavir/ritonavir twice daily.
There are some limitations in these results. It would have been helpful to see the rates of HIV undetectability using an ultrasensitive test. The CD4 cell increase was not spectacular, but will likely continue to increase among those who remain in the study and who have persistently undetectable viral loads or who maintain lower levels than baseline. Of some concern were the rates of severe or life-threatening increases in cholesterol and triglycerides in approximately one-third of subjects, even though there was little change in the overall total cholesterol/HDL cholesterol ratio. It would be helpful if drug interaction studies could be performed using one or more of the "statin" drugs (used to treat high cholesterol) and lopinavir/ritonavir, with and without efavirenz. The intent would be to ensure that drug levels are not altered significantly.
Study M97-765, also presented in Durban, concerned lopinavir/ritonavir in persons with a detectable viral load after taking their first PI combination. The regimen also included nevirapine and two NRTIs. The lead author was Melanie Thompson, MD, from the AIDS Research Consortium in Atlanta.
Participants, comprising 70 HIV positive persons (10% women, 34% non-White), had to have been naive to NNRTIs and to at least one NRTI that was included in the new regimen. Dosing was lopinavir 400 mg/ritonavir 100 mg (for 36 subjects) or 400/200 mg (for 34 subjects). The median baseline HIV RNA for the 70 participants was 4.0 log (10,000) copies/mL with a CD4 cell count of 349 cells/mm3.
The 72-week results were as follows. Using a strict intent-to-treat analysis (including all study enrollees), 57% achieved an undetectable viral load (limit 50 copies/mL). Using a lower limit of 400 copies/mL, 73% had an undetectable viral load. However, it was reported that among the 24 participants who had a detectable viral load at week 72, half of them could be explained by "documented noncompliance, early discontinuation, or interruption of therapy." Five additional subjects who had a detectable viral load (limit 50 copies/mL) did have fewer than 400 copies/mL. The mean CD4 cell increase was 154 cells/mm3.
Baseline phenotypic drug resistance information was available for 81% of participants. Among those, 63% had at least a 4-fold decreased susceptibility (resistance) to their previously used PI drug. Phenotypic drug resistance testing was performed using the Antivirogram test from Virco.
The most common adverse events classified as moderate or worse were similar to those in the first lopinavir report above. Those included diarrhea (more than three loose stools daily) in 23%, abnormal stools (three or fewer loose stools daily) in 4%, nausea in 4%, asthenia (weakness) in 6%, headache in 1%, and vomiting in 1%. Laboratory abnormalities that were severe or life-threatening (grade 3-4) included increases in the following: cholesterol greater than 300 mg/dl (28%); increased triglycerides greater than 750 mg/dl (25%); GGT greater than five times the upper normal limit (28%); ALT greater than five times the upper normal limit (14%); and AST greater than five times the upper normal limit (9%). Cofactors that might have contributed to the increase in liver enzymes, including infection with HCV or HBV, were not reported.
The discontinuation rate was 17%, including 4% due to adverse events and 13% due to other reasons (unrelated to viral load). The three participants who discontinued due to adverse drug-related events included two due to gastrointestinal (stomach-colon) symptoms and one due to rash that might have been caused by efavirenz. Nine participants discontinued for other reasons, including myocardial infarction (heart attack) in one subject on day 1 of the study (therefore unrelated to the study medications), personal reasons, nonadherence, lack of participation in follow-up, and death due to lung cancer or rhabdomyolysis (destruction of muscle cells) leading to kidney failure.
These results demonstrate significant potency of the regimen containing lopinavir/ritonavir plus nevirapine and two NRTI drugs for those participants with a detectable viral load who were taking their first PI drug combination and who had never taken an NNRTI. As with all regimens, maximal adherence to dosing is a key factor in viral load undetectability.
Clumeck, N. and others. ABT-378/ritonavir (ABT-378/r) and efavirenz: 24-week safety/efficacy evaluation in multiple PI-experienced patients. Abstract and poster presentation TuPeB3196.
Kempf, D. and others. Identification of clinically relevant phenotypic and genotypic breakpoints for ABT-378/Ritonavir in multiple PI-experienced, NNRTI-naïve patients. 4th International Workshop on HIV Drug Resistance and Treatment Strategies. Sitges, Spain. June 12-16, 2000. Abstract and oral presentation 89.
Sun, E. New frontiers in antiretroviral therapy. Satellite symposium sponsored by Abbott Laboratories.
Thompson, M. and others. Analysis of duration of virologic response in two phase II studies of ABT-378/ritonavir (ABT-378/r) at 72 weeks. Abstract and poster presentation TuPeB3197.
The final, 48-week results of the Atlantic Study were presented in a late-breaker poster in Durban. This study combined a backbone of two NRTIs, d4T (Zerit) and ddI (Videx), with a third drug. The third drug was either nevirapine (an NNRTI), indinavir (a PI), or 3TC (Epivir, an NRTI ). Overall, the results showed a better outcome in the nevirapine- and indinavir-containing arms, as compared with the 3TC-containing arm. The lead author was Kathleen Squires, MD, of the University of Alabama at Birmingham.
This open-label and randomized study enrolled 298 (20% women) HIV positive, treatment-naive persons. Entry criteria were a CD4 cell count greater than 200 cells/mm3, a viral load greater than 500 copies/mL, and no prior HIV-related symptoms. Standard dosings were used for all medications except ddI and nevirapine, which were dosed once daily.
At baseline, the median HIV RNA viral load was 4.3 log (17,782) copies/mL with a CD4 cell count of 406 cells/mm3. A high baseline viral load (greater than 100,000 copies/mL) was detected in 13% of participants. Baseline characteristics were relatively equal among the three study arms.
Results after 48 weeks revealed the following. Using a strict intent-to-treat (ITT) analysis, an undetectable viral load (limit 50 copies/mL) was achieved in 49% of both the nevirapine- and indinavir-containing arms and in 40% of the 3TC-containing arm. The differences when comparing any two of the arms were not statistically significant. However, using an on-treatment analysis (excluding drop-outs), there was a significant difference when comparing the indinavir- and 3TC-containing arms, with a trend towards a significant difference when comparing the nevirapine and 3TC-containing arms. There was no significant difference when comparing the nevirapine- and indinavir-containing arms. The median CD4 cell count increase was approximately 140 cells/mm3 in the nevirapine- and indinavir-containing arms and approximately 160 cells/mm3 in the 3TC-containing arm.
In a separate analysis, Dr. Squires evaluated the viral load response for those with a high baseline level in the upper quartile, or greater than 58,519 copies/mL. In an ITT analysis, the percentage with an undetectable viral load (limit 50 copies/mL) was 48%, 28%, and 26% in the indinavir-, nevirapine- and 3TC-containing arms, respectively; however, the differences were reported as insignificantly different. In an on-treatment analysis, the percentage undetectable was 80%, 56%, and 42%, respectively for the same three arms, with a trend towards a statistically significant difference. There was no reported information about adhering to dosing of medication.
Adverse events were reported as including all grade 1-4 effects (mild to life-threatening) that occurred among at least 5% of enrollees. (Note that most other studies usually do not report mild adverse events, so the following percentages may appear to be high.) Gastrointestinal side effects occurred among 74%, 48%, and 53% of the indinavir-, nevirapine- and 3TC-containing arms, respectively. These included nausea/vomiting in 31%, 19%, and 17%, and diarrhea in 25%, 15%, and 13%. Rash or redness of the skin occurred in 4%, 12%, and 0%. Numbness or tingling (neuropathy, paresthesia) occurred in 10%, 15%, and 15%. Systemic effects (assumed to be malaise, weakness, and/or flu-like symptoms) occurred in 32%, 24%, and 21%. Discontinuations due to drug toxicities were reported to be no different in the three arms, without percentages given.
Abnormal laboratory tests were not reported. However, an interim analysis of the first 235 participants (79%) who completed 48 weeks was presented at last years 39th Intersci-ence Conference on Antimicrobial Agents and Chemotherapy (ICAAC). Severe or life-threatening adverse events at that time revealed nothing surprising that had not been reported previously for each of the study drugs. In the indinavir-containing arm, 5% of subjects developed a kidney stone(s). Severe or life-threatening laboratory toxicities were as follows: increased liver enzymes occurred in 5%, 13%, and 6% of the indinavir-, nevirapine-, and 3TC-containing arms, respectively. Increased amylase occurred in 4%, 4%, and 7%.
Twenty-five percent of participants in the indinavir arm, 17% in the nevirapine arm, and 29% in the 3TC arm discontinued the study early as of that preliminary analysis. Adverse events were responsible for 42% of study discontinuations. Other common causes for discontinuation included participant request or lack of participation in follow-up. The interim results had been presented by Robert L. Murphy, MD, of Northwestern University in Chicago.
Results of the Atlantic Study after one year indicate that a regimen with three NRTIs that includes 3TC, ddI, and d4T does not perform as well as triple therapy arms that included ddI and d4T with either nevirapine or indinavir. Those two latter arms performed rather similarly, with a trend towards a better outcome for those with a high baseline viral load (greater than 58,519 copies/mL) that favored the indinavir-containing arm. The adverse event profile of each arm was somewhat different, reflecting known side effects for the drugs used. All three regimens were similarly well tolerated with no significant differences in discontinuation rates when comparing the three arms. This study is ongoing and will continue for a total of roughly three years (144 weeks).
Murphy, R.L. and others. The Atlantic Study: a randomized, open-label trial comparing two protease inhibitor (PI)-sparing antiretroviral strategies versus a standard PI-containing regimen, 48 week data. 39th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Francisco. September 26-29, 1999. Abstract and late-breaker oral presentation LB-22.
Squires, K. and others. The Atlantic Study: a randomized, open-label study to evaluate the efficacy and safety of three triple-combination therapies aimed at different targets in anti-retroviral naïve HIV-1 infected patients, final 48 week analysis. Abstract and late-breaker poster presentation LbPeB7046.
Two hundred fifty-nine abstracts (5% of the total) addressed adherence to dosing of anti-HIV medications. This is an extremely important issue in terms of achieving and maintaining an undetectable viral load. However, 100% adherence is very difficult for many people, particularly after a prolonged period of taking medications.
Interim results of one presentation highlighted the ease and preference of twice-daily dosing of medications, as compared with three-times-daily dosing. Specifically, researchers from the NICE study (M99-047) have reported that persons who switch to a double PI combination with indinavir and ritonavir (400 mg of each twice daily) plus two NRTIs prefer that regimen and are more adherent to medication dosing. Before switching, participants were taking standard indinavir dosing of 800 mg three times daily plus two NRTI drugs. When compared with the nonswitched arm of the randomized, controlled study, those who switched were significantly more likely to report that 1) it was easier to take their PIs as prescribed, 2) they missed fewer doses, 3) the new regimen was easier to take at the same time each day, and 4) it was easier to coordinate dosing with meals. The presenting author was Edwin DeJesus, MD, from the IDC Research Initiative in Altamonte Springs, FL.
The NICE study has planned to enroll a total of 500 participants who are taking standard three-times-daily indinavir (800 mg) plus two NRTIs and who have an undetectable HIV RNA viral load (using a test used at a practitioners local laboratory). Four hundred subjects will be randomized to switch to twice-daily indinavir/ritonavir (400/400 mg) plus two NRTIs, while the other 100 subjects will be randomized to maintain their current regimen. Those in the switching arm may take their medications without meal restrictions or fluid requirements. After 12 weeks of the study, participants may cross over to either arm, depending upon individual or physician preference. The 400 mg twice-daily dosing of indinavir/ritonavir has been associated with an extremely low rate (i.e., only in rare anecdotal cases) of kidney stones -- a significant side effect of indinavir that still occurs with other dosing (indinavir 800 mg/ritonavir 100-200 mg, each twice daily, or mono-PI drug therapy with indinavir).
At the time of interim analysis, 215 participants randomized to the switching arm and 60 participants randomized to the nonswitching arm had completed four weeks of the study. Among all randomized subjects, 7-17% were women and 30-40% were non-White. The mean baseline CD4 cell count was 558-572 cells/mm3. As indicated, all participants had an undetectable viral load at baseline.
Those who completed four weeks were given a standardized "Medication Taking Survey." One question in the survey was "In the past two weeks, on a [Likert] scale of 1 ("very easy") to 5 ("not easy"), how easy has it been for you to take your PI medicine as prescribed?" Significantly more persons in the switching arm reported "very easy" after four weeks (59%) compared with baseline (48%), as compared with the nonswitching arm (52% after four weeks and at baseline, i.e., no change).
Another question was "In the past two weeks, how many times do you think you missed taking your PI medication?" Significantly more persons in the switching arm reported "none" after four weeks (73%) compared with baseline (62%), as compared with the nonswitching arm (57% after four weeks and 52% at baseline, i.e., a 5% increase). Also, after four weeks 98% of the switching arm reported zero, one, or two doses missed, as compared with 90% of the nonswitching arm.
Yet another question was "In the past two weeks, when did you miss at least one dose of your PI medication?" The three choices were morning, midday, and bedtime. At baseline, 23% of the switching arm and 30% of the nonswitching arm indicated that they missed the midday dose. After four weeks, 4% of the switching arm indicated midday (on a twice-daily regimen, none should have had a midday dose) with 30% of the nonswitching arm still indicating midday (i.e., no change). After four weeks, 26% of the switching arm admitted to missing at least one dose anytime (in the morning, midday, or at bedtime), compared with 47% of the nonswitching arm.
There were no interim results for viral loads or CD4 cell counts to correlate with the study questionnaire results. However, the entire enrollment had not yet been filled.
One potential limitation of the study is that the results are based upon information reported by participants; this is the case with many studies measuring adherence. Other studies have shown that reported adherence by participants is not always the same as adherence measured by MEMS caps (computerized medication bottle caps that record each time the container is opened). One way to validate peoples self-reports would be to measure drug levels in blood (see "Therapeutic Drug Monitoring"). Regardless, the statistically significant improvements in adherence issues that were measured in the switching arm (to twice-daily indinavir/ritonavir) when compared with little or no change in the nonswitching arm of the NICE study strongly suggests that participants may have been quite truthful in their responses. These results may also apply to other twice-daily drug regimens (see below). Of course, the significance will be underscored when viral load and CD4 cell counts are reported after longer follow-up.
Several other studies at Durban indicated easier and improved adherence with twice-daily dosing of PI drugs, many of which included twice-daily nelfinavir after participants were switched from three-times-daily dosing. One study found an equivalent level of adherence when comparing twice-daily nelfinavir in combination therapy with once-daily efavirenz combination therapy.
DeJesus, E. and others. A randomized, controlled, open-label study comparing the adherence and convenience of continuing indinavir Q8H vs. switching to Norvir/indinavir 400 mg/400 mg BID (The NICE Study). Abstract and oral presentation WeOrB482.
Dhingra, R. and others. Twice daily nelfinavir results in greater durability and higher CD4 counts than TID dosing. Abstract WePeB4118.
Hugen, P. and others. Compliance to HIV protease inhibitors is more accurately measured by combining various methods, including MEMS (electronic monitoring) and therapeutic drug monitoring [measuring drug levels in blood]. Abstract ThPeB5029.
Morris, A.B. and others. The efficacy and durability of BID nelfinavir in two clinics. Abstract WePeB4153.
Novak, R. and others. Effect of change from TID to BID nelfinavir on viral load, CD4 count and adherence. Abstract WeOrB483.
A meta-analysis of 23 studies with HAART in treatment-naive (not previously treated) persons was presented by John A. Bartlett, MD, of Duke University Medical Center in Durham, NC. Dr. Bartlett showed that the total number of daily pills is significantly inversely associated with HIV viral load undetectability after 24 or 48 weeks. This means that a lower number of daily pills is associated with a higher likelihood of achieving an undetectable viral load. It also means that a higher number of daily pills or capsules is associated with a lower chance of achieving viral load undetectability. Dr. Bartlett had presented a similar meta-analysis at this past Februarys 7th Conference on Retroviruses and Opportunistic Infections (CROI). The study suggests that adherence might be easier with a lower number of daily pills or capsules.
Bartlett, J.A. Correlation between antiretroviral pill burden and durability of virologic response -- a systematic overview. Abstract and poster presentation ThPeB4998.
Researchers from Perth, Australia, presented an interesting late-breaker poster that provided some evidence to link lipoatrophy (fat loss under the skin) with anti-HIV drug therapy and changes in the mitochondria (energy producers) of cells. Skin biopsies from the hip or "buffalo hump" area of HIV positive persons taking nucleoside reverse transcriptase inhibitors (NRTIs) with or without PIs have abnormal appearances of fat cells (adipocytes). These cells under the microscope or electron microscope showed "abnormal adipocyte ultrastructure with mitochondrial abnormalities, cytoplasmic [outside the nucleus] lipid [fat] accumulation, and loss of adipocyte volume." In addition, there was associated fat cell loss and lipogranuloma formation. Those who were taking both NRTIs and a PI drug had more severe changes than those taking only NRTI drugs. Also, S. Mallal, MD, from Royal Perth Hospital in Australia found that these changes were seen in regions of fat loss (lipoatrophy) and accumulation ("buffalo hump"). (Note, however, that fat loss under the skin can still be present when there is deeper fat accumulation, as has been shown in the abdominal area.) HIV positive subjects 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 so-called lipodystrophy (body fat changes).
Mallal, S. and others. Light and electron microscopy findings in subcutaneous fat in antiretroviral treated and naive HIV-infected patients. Abstract and late-breaker poster presentation LbPeB7054.
In a late-breaker poster presentation, Dr. A. Alfirevic and colleagues from the University of Liverpool in the UK found evidence for a genetic predisposition to lipodystrophy. A specific gene variation (i.e., 238 polymorphism genotype) in the "promoter" region of tumor necrosis alpha (TNF-alpha) was statistically associated with lipodystrophy in HIV positive persons taking anti-HIV therapy. Dr. Alfirevic concluded that the gene variation represents a susceptibility factor for developing HIV-related lipodystrophy. These results need to be confirmed in a larger study.
Alfirevic, A. and others. Tumor necrosis factor-alpha (TNF-alpha) promoter region gene polymorphism in patients with HIV-1 associated lipodystrophy. Abstract and late-breaker poster presentation LbPp113.
Attendees were pleased to hear additional follow-up of the landmark HIVNET 012 study that had used only two doses of nevirapine to help prevent HIV transmission from mother to newborn. Among the 311 mother-infant pairs who received nevirapine (one dose for mother at onset of labor and one dose for the infant within three days after birth), the HIV transmission rate to those infants was 16%. Among the 308 mother-infant pairs who received AZT (Retrovir) during labor and for the infants first week of life, the transmission rate to those newborns was 24%. The difference between the two groups was statistically significant. Among those mother-infant pairs who completed 12 months of the study, 95% of the infants were breast-fed. Among the 65% of mother-infant pairs who completed 18 months, there were no additional infants who became HIV positive in the nevirapine arm, while an additional four subjects in the AZT arm became infected. The presenting author was M. Owor, MD, from Mulago Hospital in Kampala, Uganda. However, attendees were disappointed by the results of the PETRA trial (AZT plus 3TC) to prevent mother-to-child transmission (see below).
Immediately prior to the Durban conference, Boehringer Ingelheim (BI) Pharmaceuticals announced that the company will offer free nevirapine for five years to resource-poor countries to help prevent HIV transmission from mother to newborn. In a press release, BI said, "The initiative is part of [the companys] commitment to the collaborative effort with five companies (BI, Bristol-Myers Squibb, Hoffmann-La Roche, Glaxo Wellcome, and Merck & Co.), United Nations Agencies (World Health Organization [WHO], World Bank, UNICEF, UNFPA, and UNAIDS), and committed governments to explore practical ways of working together to make HIV/AIDS care available and affordable to a significantly greater number of people in developing countries." This donation by BI is rather generous. Countries with the greatest need for the medication usually do not have the infrastructure to allow for appropriate HIV-related education, counseling, blood testing, drug distribution, and follow-up of HIV positive women, their infants, and sexual partners.
Boehringer Ingelheim press release. Boehringer Ingelheim offers Viramune (nevirapine) free of charge to developing countries for the prevention of HIV-1 mother-to-child transmission. July 7, 2000.
Owor, M. and others. The one year safety and efficacy data of the HIVNET 012 trial. Abstract and late-breaker oral presentation LbOr1.
One of the more disappointing studies presented in Durban was the 18-month follow-up of the PETRA trial. Previously, the PETRA trial showed that short-course AZT plus 3TC was able to reduce HIV transmission by 52% to newborns at age six weeks. The dosing started at week 36 of pregnancy, then continued during delivery and for one week after delivery for mother and infant. Unfortunately, those benefits essentially are lost by the time the infant reaches 18 months and is breast-feed without any continuous anti-HIV therapy taken by infant or mother. Continuing to drink breast milk infected with HIV outweighs the six-week benefits of short-course AZT plus 3TC. The results were not necessarily unexpected given what is known about HIV transmission via breast-feeding.
The double-blinded (i.e., medications were not disclosed to treating physicians or study subjects), randomized, and placebo-controlled PETRA study enrolled a total of 1,457 mothers and 1,501 infants from Uganda, Tanzania, and South Africa (Durban and Johannesburg). The trial was sponsored by UNAIDS (United Nations Programme on AIDS), with researchers from the Netherlands and the three African countries. One-third of the infants were delivered by cesarean ("C") section and 71% were breast-fed.
Six-week results were as follows. HIV infection or death occurred in 9% of the treatment arm and 21% of the placebo arm. (Shorter durations or dosing combinations led to intermediate rates of HIV infection or death.) The decreased relative risk was 0.48 for the treatment arm, when compared with placebo; the results were statistically significant. After 18 months, however, the difference was no longer statistically different. The rate of HIV infection or death in the treatment arm was 21%, compared with 27% in the placebo arm. The relative risk in the treatment arm was 0.78, i.e., not statistically significant. The results show that AZT/3TC treatment around the time of infant delivery decreases HIV infection rate or death at 6 weeks of age, but not at 18 months, in a population in which a majority of women breast-feed. Additional analyses may show benefits for those who did not breast-feed. The results were presented by Glenda Gray, MD, from Baragwanath Hospital in Johannesburg during a late-breaker session. Note that these results are different from short-course nevirapine to prevent HIV transmission to newborns (see above).
Potential solutions will be to find a way to make breast-feeding safer or to allow the substitution of safe, nutritious alternatives. Both will prove challenging in many of the countries where the infrastructure for sanitary food preparation and clean water is a problem.
Gray, G. and others. The PETRA study: early and late efficacy of three short ZDV/3TC combination regimens to prevent mother-to-child transmission of HIV-1. Abstract and late-breaker oral presentation LbOr5.
As a result of a study of female sex workers from Africa and Thailand, UNAIDS has recommended that women at risk for HIV infection not use N-9 products in the vagina. N-9 is a common spermicide (sperm-killing agent) used in the vagina to help prevent pregnancy. It also has some microcidal effects (i.e., it kills certain bacteria and/or viruses), including HIV and other sexually transmitted infections, in laboratory tests. In the U.S. and China, N-9 is available under the trade name "Advantage S." The results of the triple-blind, randomized, multicenter study of N-9 were discussed by Dr. Joseph Perriens and reviewed by Dr. Isabelle de Zoysa during a plenary session in Durban.
The study enrolled 990 HIV negative women from Thailand, South Africa, and Côte dIvoire (Africa). They were given HIV counseling, condom promotion information, free condoms, and free treatment of sexually transmitted diseases (STDs). The women were randomized to receive N-9 gel or the same gel vehicle without N-9 (a placebo) used intravaginally before heterosexual vaginal intercourse.
The rate of new HIV infections per 100 "person-years"* was 10.2 in the placebo arm and 15.5 in the N-9 arm. During the observation period, 41 women in the placebo arm and 59 in the N-9 arm became HIV positive. However, the two study arms had a lower incidence (rate) of new HIV infections than did control women who were not in the study. Interestingly, the N-9 arm had higher rates of both gonorrhea and chlamydiae cervical infections, but the differences were not statistically significant. The trial was sponsored by UNAIDS.
In previous studies, N-9 has been shown to increase vaginal inflammation (swelling with redness due to increased blood flow with increased white blood immune cells). Sometimes, the top layer of cells was eroded or no longer present (due to inflammation) which would mean an absence or decrease of the normal protective barrier to infection.
A prior study (known as the VCF trial) published by Dr. Roddy and colleagues found similar results in 1998. As a result of these studies, "UNAIDS believes that women at high risk of HIV infection should not use N-9 [products], given that the bulk of the data now suggests that it is either ineffective or harmful as an anti-HIV agent [in heterosexual vaginal transmission]." Meanwhile, at least 13 other vaginal microbicides are in the research pipeline as possible female-controlled products to help decrease HIV transmission through the vagina. Some examples are alkyl sulfates, Buffergel, and others.
* A person-year is the number of persons multiplied by the years (or fractions of years) of follow-up, to allow for fair or equal comparisons. For example, one person followed for two years equals two person-years, as is two people followed for one year or four people followed for one-half year each.
de Zoysa, I. Track C plenary rapporteur session.
UNAIDS press release. UNAIDS calls for continued commitment to microbicides. July 12, 2000.
Previously, there had been conflicting information about HIV viral load differences when comparing women with men. Now, a meta-analysis of six clinical trials confirms that women have no significant differences in HIV RNA viral load when compared with men. Women with advanced HIV infection had insignificantly lower viral loads than men with the same CD4 cell counts. Conversely, at higher CD4 cell counts, women had insignificantly higher viral loads than men with similar CD4 cell counts. The lead author was Dr. Starley B. Shade from the Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA) and the University of California, San Francisco.
Shade, S.B. and others. Gender differences in viral load: results from six clinical trials. Abstract and poster presentation MoPeE2946.
A randomized, placebo-controlled trial of 741 persons with HIV/hepatitis C virus (HCV) coinfection began this past summer. The purpose of the study is to compare the potential benefits of adding anti-HCV combination therapy in coinfected persons who are already taking HAART for HIV.
Subjects in the three study arms taking HAART will add peginterferon alfa-2a (Pegasys) with or without ribavirin (Rebetol), or interferon alfa-2a (Roferon-A) with ribavirin. The pegylated form of interferon alfa is an experimental, long-acting version that is dosed once weekly by injection. The dose of ribavirin will be 800 mg, which has been used more commonly in HIV-HCV-coinfected persons. The standard dose is 1,000 mg (in persons 165 lbs or less) or 1,200 mg (in people greater than 165 lbs) in HCV positive persons without HIV infection. Therapy will be taken for 48 weeks with additional follow-up 24 weeks later to measure the sustained virologic response.
Due to in vitro antagonism (canceling effects), there have been some concerns about adding ribavirin (a nucleoside analog) to HAART regimens including either d4T or AZT. Francesca J. Torriani, MD, of the University of California, San Diego, described the new study during a satellite symposium called "Fabric of the Future." Dr. Torrianis presentation was entitled, "New Developments in the Management of HIV Co-Infections: HCV and CMV."
Torriani, F.J. New developments in the management of HIV co-infections: HCV and CMV. Satellite symposium.
While HIV has been shown to increase HCV disease progression, the effects of HCV upon HIV disease are somewhat less clear. Now, results of a large cohort study from Switzerland show a significant progression of HIV disease after HAART was started in coinfected subjects. Anti-HIV therapy was started in 1,157 subjects (34% women), who were compared with 1,954 persons with HIV monoinfection (without HCV) who started treatment. The large Swiss Cohort was followed for more than four years. Results after three years of treatment showed that the proportion who were alive and without clinical progression of HIV disease was 80% among HIV-HCV coinfected persons who continued injecting drugs, 85% among HIV-HCV coinfected persons who stopped injecting drugs, and 90% among those who were only infected with HIV and not injecting drugs. These differences were highly statistically significant. HIV-HCV coinfected participants had a 1.7 "relative hazard" of HIV/AIDS progression or death when compared with persons infected only with HIV in a "regression (statistical) analysis" that included other potential cofactors, including HCV viral load. The researchers found that coinfected persons had a reduced ability to increase CD4 cell counts after achieving HIV RNA viral undetectability, when compared with HIV monoinfected subjects. The lead author was G. Greub, MD, of the University Hospital in Lausanne, Switzerland.
Greub, G. and others. Negative impact of HCV infection on HIV progression, survival, and immune restoration in the Swiss HIV cohort studies. Abstract and poster presentation MoPeB2139.
The Centers for Disease Control and Prevention (CDC) has reported that those with HIV infection have a significantly higher rate of acute infection with HBV when compared with the general population. In people with HIV infection (and much less commonly among those without HIV infection), HBV infection can be lethal. HBV is transmitted by the same mechanisms as HIV.
More than 11,000 HIV positive persons from 11 U.S. cities in the Adult and Adolescent Spectrum of HIV Disease Project (ASD) were followed between 1997 and 1998. Only 11% of enrollees with no history of acute or chronic HBV had one or more doses of HBV vaccine. Current guidelines by the U.S. Department of Health and Human Services/Infectious Disease Society of America (USPHS/IDSA) recommend the HBV vaccine series for all HIV positive persons with a negative test for HBV core antibody. Among participants, 74% had taken 3TC as a part of anti-HIV therapy. (3TC also has anti-HBV activity and is one of two drugs that is FDA-approved to treat chronic HBV infection.) There was a 54% decreased risk of acute HBV infection among persons without previous HBV infection who had taken 3TC.
Kellerman, S.E. and others. Incidence of acute hepatitis B in HIV infected individuals and the protective effect of hepatitis B vaccination and lamivudine. Abstract and poster presentation MoPpA1082.
Harvey S. Bartnof, MD, has been a member of the Scientific Advisory Committee of the San Francisco AIDS Foundation since 1987.
Back to the SFAF BETA Autumn, 2000 contents page.