Bulletin of Experimental Treatments for AIDS, October, 1998
Issues specific to women received considerable attention at the 12th World AIDS Conference in July. Most prominent among these was treatment aimed at preventing mother-to-infant (vertical or perinatal) transmission, with major interest focused on shorter and less expensive regimens, and efforts to make those regimens available to women in developing nations. Updated information was also presented on a variety of other women’s issues, including female-controlled prevention methods and HIV-related cervical cancer.
Until recently, the only antiretroviral regimen validated for use in preventing vertical transmission was the complex, three-part approach used in Pediatric AIDS Clinical Trials Group (ACTG) study 076. This involved 1) AZT given orally at a dose of 100 mg five times per day, beginning at 14-34 weeks of gestation and continuing throughout pregnancy, 2) intravenous AZT given during labor, starting with an initial dose of 2 mg/kg of body weight administered over one hour, followed by continuous infusion of 1 mg/kg/hour until delivery, and 3) oral AZT syrup administered to the newborn at a dose of 2 mg/kg every six hours for the first six weeks of life (with intravenous dosing for infants unable to take the drug orally). This regimen reduced the rate of perinatal transmission by two-thirds (see BETA, December 1996, page 21).
In March of this year, the U.S. Centers for Disease Control and Prevention (CDC) reported successful preliminary results from a Thai study using a shorter, simpler regimen, which involved giving the mother 300 mg of AZT orally twice a day from 36 weeks of gestation until the onset of labor, followed by 300 mg every three hours from the beginning of labor until delivery. Compared to a placebo, this short-course regimen cut the risk of perinatal transmission roughly in half, from 18.6% to 9.2%.
Although not quite as impressive as the ACTG 076 results, this reduction represents a huge improvement over no treatment at all, which is what most HIV-infected pregnant women around the world receive. Much discussion in Geneva centered around use of the Thai regimen in developing countries where poverty and lack of medical infrastructure make the full ACTG 076 regimen a practical impossibility (see BETA, April 1998, page 7).
These discussions touched on public policy as much as science. The United Nations announced a pilot program to make short-course AZT available to 30,000 pregnant women in the developing world. However, the drug will no longer be available to the women after their children are born, raising concerns about an increasing number of AIDS orphans. Conference community planning committee chair Robin Gorna pointedly observed, "This is good news for their babies, but what about the women?"
Meanwhile, plans for new placebo-controlled trials of vertical transmission interventions in Africa -- in which no volunteers will receive AZT -- prompted fierce criticism from the advocacy group Public Citizen. Public Citizen’s Peter Lurie, MD, for years a University of California at San Francisco (UCSF) AIDS researcher, blasted such trials as "simply unbelievable," adding, "in this whole debate there has been very little discussion of the treatment of the women and children in these studies. As soon as the study is over they’re terminated." The researchers involved, including Hoosen Coovadia of South Africa, staunchly defended their studies as the best and most appropriate designs for areas where even short-course AZT may never be feasible.
More detailed information on short-course regimens will be available shortly, noted Lynne Mofenson, MD, of the National Institute of Child Health and Human Development (NICHD). "Results later this year from the UNAIDS PETRA trial may help to clarify the contribution of the individual components of prophylaxis to protection. PETRA compares a short three-part AZT/3TC combination regimen to intrapartum/postpartum and intrapartum only [during labor] regimens. Efficacy of an intrapartum regimen would show the critical importance of pre-exposure infant prophylaxis, and would provide an easily administered regimen able to be implemented in most of the developing world."
Even where the resources to implement effective treatment exist, prevention protocols have not necessarily been put into widespread use. One disturbing report came from Thailand, where, despite the fact that the government and the Red Cross make AZT available at no cost for the treatment of pregnant HIV positive women, over three-quarters of 480 physicians surveyed did not routinely prescribe it. Many physicians, especially those in outlying areas, were unfamiliar with the ACTG 076 data.
Several reports examined the effects of cesarean delivery on vertical transmission, consistently reporting an apparent protective effect. In the French Perinatal Cohort Study, for example, in mothers and infants receiving AZT, "elective cesareans were associated with a lower transmission rate than emergent cesareans or vaginal deliveries" (0.8%, 11.4%, and 6.6%, respectively). The researchers noted an 80% decreased risk of HIV transmission for elective cesarean delivery when compared with vaginal delivery.
Other reports were similar. Researchers following a prospective German cohort reported an overall transmission rate of 18.4% of 255 births. Elective cesarean without AZT produced a transmission rate of 10.8%, compared to a rate of 7% in vaginal deliveries with AZT and 2.5% in cesarean deliveries with AZT. Jennifer Read of the NICHD reported on a meta-analysis of 15 cohorts that indicated that cesarean delivery prior to labor and membrane rupture "is associated with a significantly decreased risk of vertical transmission, independent of antiretroviral therapy."
At the time of ACTG 076, no one knew that combination therapy would become the standard of care for HIV infection, with AZT monotherapy universally regarded as inadequate. Unfortunately, data on combination therapy during pregnancy have been sparse. The updated International AIDS Society-USA HIV treatment guidelines released at the conference advise that, "in most respects, HIV infection in pregnant women should be treated as infection in non-pregnant patients," adding that "given available data, zidovudine [AZT] should probably be included in any regimen intended to prevent perinatal transmission."
Intuitively, it seems logical that potent combination regimens would be more effective than AZT monotherapy in preventing vertical transmission. Research reported in Geneva, though far from definitive, tended to support this hypothesis.
Analysis of ACTG 185, which compared AZT to AZT plus HIV hyperimmune globulin (HIVIG) for prevention of vertical transmission, failed to find any protective effect of HIVIG, but did indicate an association between viral load, both at entry into the study and at the time of delivery, and risk of maternal-fetal transmission. At delivery, 5% of women with HIV RNA levels of 500 copies/mL or higher transmitted HIV to their newborns, compared to none with viral load levels below 500 copies/mL. This led the researchers to suggest further studies of treatments aimed specifically at producing viral load reductions.
Regarding combination therapy during pregnancy, Karen Beckerman, MD, of UCSF presented data collected during treatment of 60 women at the Bay Area Perinatal AIDS Center at San Francisco General Hospital. Beckerman described a steady shift towards combination therapy in recent years, with 10 of 12 pregnant women treated thus far in 1998 receiving 3-drug regimens. Of 60 infants born to mothers treated since May 1995, none have shown evidence of HIV infection. Combination therapies have been well-tolerated, Beckerman stated, and "no maternal or fetal complications have been noted."
A late-breaker presented by Patrizio Lorenzi describing pregnancy outcomes in three Swiss cohorts suggested possible adverse consequences of combination therapy during pregnancy. Thus far in the ongoing observational study, 37 women have been followed and 30 infants have been born. Twenty-one women took two nucleoside analogs, while 16 took two nucleoside analogs plus one or two protease inhibitors; the most commonly used protease inhibitors were indinavir (Crixivan), taken by nine women, and ritonavir (Norvir), taken by four women. Only one case of vertical transmission occurred in this sample, involving a mother who was nonadherent.
Although no unexpected or life-threatening adverse events were observed in the mothers, Lorenzi said, 33% (11) of the infants were premature, born before week 37. Lorenzi called this prematurity rate a major concern, noting that it was roughly double the 15-17% rate observed in previous studies of pregnant women receiving either AZT monotherapy or no treatment. Also of major concern, he said, were two brain hemorrhages, both occurring in newborns exposed to indinavir, one full-term and one premature. Lorenzi termed such an occurrence in a full-term infant "very surprising." Another infant whose mother took indinavir was born with extrahepatic biliary atresia, which is damage to the bile ducts that could lead to liver damage. This, he explained, is "a very rare condition which occurs in only 1 in 14,000 live births."
While the numbers do not establish causation, Lorenzi noted in response to a question that even after adjusting for possible confounding factors such as intravenous drug use, the researchers still found a disturbingly elevated prematurity rate among infants born to mothers on combination therapy. Although most of the infants were not severely premature (median term was 36 weeks for the premature births), Lorenzi said that he would be inclined to advise against using protease inhibitor-based combination regimens during pregnancy until more is known.
After listening to Lorenzi, Beckerman argued that "one case [of biliary atresia] is meaningless. [However], other groups are staying away from indinavir because of the potential for nephrolithiasis [kidney stones] in the mother and the potential for hyperbilirubinemia [high blood bilirubin levels] in the baby."
Alice Stek of the University of Southern California reported on the use of combinations involving nevirapine (Viramune), AZT, and a second nucleoside analog during pregnancy. Of 14 infants born so far whose mothers received five weeks or more of treatment, only one baby was born prematurely (at 35 weeks) and all are healthy currently, though three were significantly anemic at birth, Stek said. Thus far all of the children are HIV negative. Eleven of the mothers maintained their viral load below 400 copies/mL through delivery, and two had serious adverse events (one had severe anemia, which resolved after a change of nucleoside analogs, and one had hepatitis). A trial to determine whether two doses of nevirapine will reduce vertical transmission is underway (see BETA, March 1997, page 50).
Researchers from the University of Pittsburgh reported an interesting correlation between weight gain and pregnancy outcomes in HIV-infected women. Of 80 women followed from 1988 to 1996, weight gain was found to be insufficient in 46.3%, compared to 30.6% in a control group of HV negative pregnant women. When compared to the mothers whose weight gain was normal, those with insufficient weight gain were more likely to transmit HIV to their newborns (18.9% vs 13.9%), and had nearly double the rates of premature birth and low birth weight. "This suggests the need for nutritional support, especially in indigent populations," the researchers noted.
Finally, a handful of studies reaffirmed the risk of HIV transmission via breast-feeding. For example, in a Malawi cohort of 621 infants born to HIV-infected mothers who were HIV negative by polymerase chain reaction (PCR) test through the first seven weeks of life, 47 seroconverted during weeks 8-17. Breast-feeding was nearly universal in this population, and researchers reported that potential aggravating factors such as cracked nipples were infrequent and that none of the infants received blood transfusions.
HIV-infected women are at high risk for cervical cancer. Numerous reports at the conference focused on the relationships between HIV and human papillomavirus (HPV) infection, cervical cellular abnormalities that can be precursors to cervical dysplasia (abnormal, pre-cancerous cell changes), and cervical cancer.
Researchers at Johns Hopkins University looked at the relationships between HIV viral load, CD4 T-cell count, HPV prevalence, and cervical dysplasia as determined by Pap smear. Higher viral load and lower CD4 count were about equally associated with higher risk of dysplasia. Women with HIV RNA levels above 30,000 copies/mL and CD4 counts below 200 cells/mm3 had a 7-fold to 8-fold increased risk for abnormal Pap smears, although there was no difference in the severity of the abnormalities. Greater immune suppression and higher viral load were also associated with increased HPV prevalence.
In a follow-up study of 94 African women who had an abnormal Pap smear showing squamous intraepithelial lesions (abnormal cell changes), 76% of the HIV-infected women had persistent abnormalities an average of five months after the original diagnosis -- over four times the rate among those without HIV.
If higher viral load is associated with greater risk of cervical dysplasia, does antiretroviral treatment reduce that risk? Three presentations looked at this question, with somewhat unclear results. When French researchers reviewed 744 Pap smears from 311 HIV-infected women from 1991 to 1997, they found elevated levels of dysplasia in those whose viral load was greater than 100,000 copies/mL, but no statistically significant difference between those who were receiving antiretroviral therapy and those who were not. The scientists noted that potent combination treatments had only been available for a small portion of the time period analyzed, so a longer period of treatment might produce different results.
Another French team did see an apparent treatment effect when they prospectively followed 85 women with advanced HIV disease who began highly active antiretroviral therapy (HAART) regimens. In this group, the prevalence of cervical squamous intraepithelial lesions decreased from 66% to 49% at a median of five months follow-up, while 7 of 20 who had initially presented with high-grade lesions regressed to a lower grade. There was a trend, though not statistically significant, toward greater CD4 count increases in those whose abnormalities regressed.
Finally, a U.S. group compared 48 women who began HAART with 48 women with similar CD4 counts who did not start therapy. As expected, at 6-12 months of follow-up the HAART group showed improved CD4 counts (a mean 65 cells/mm3 increase versus a mean 14 cells/mm3 decrease in the untreated group). But the HAART and non-HAART groups showed no statistically significant difference in Pap smear changes, acquisition or persistence of HPV, or amount of HPV as measured by PCR. The researchers concluded that women with HIV should be closely followed for cervical abnormalities regardless of any HAART-associated increase in CD4 count.
Although the need for improved woman-controlled prevention methods has been discussed at every international AIDS conference this decade, data on such approaches have remained sparse. That pattern did not change dramatically this year, and much of the data presented concerned methods that have been around for some time: the female (internal) condom and various nonoxynol-9-containing products. Still, there does seem to be progress in moving promising concepts into human trials.
The news on the female condom was mixed. Several surveys of women who were given sample female condoms and asked to use and evaluate them reported a generally high level of acceptability. In a study of 504 female sex workers in Thailand, half were instructed to use male (penile) condoms consistently, while the other half were given the option of using the female condom if their partner declined to use a male condom. Although male condom use was slightly lower in the second group, use of the female condom led to an overall 17% reduction in the proportion of unprotected sex acts and a 25% reduction in the rate of sexually transmitted diseases.
The efficacy of the female condom remains somewhat unclear. A University of Alabama study looked at effectiveness of female condoms in a total of 1,464 uses by 210 monogamous male/female couples. The researchers compared reported rates of condom failure to the women’s exposure to semen, as measured by levels of prostate specific antigen (PSA), assessed by two different measurement criteria. Seven percent of women showed semen exposure by both measurements, and 21% by one measurement. For the women who reported no condom failure, the figures were 5% and 19%, respectively. The researchers argued that these relatively high figures "indicate that self-report underestimates the frequency of condom failure."
In a summary of ongoing clinical research into vaginal microbicides, Zeda Rosenberg, MD, of the National Institute of Allergy and Infectious Diseases noted that nonoxynol-9 products still constitute the majority of microbicides being studied, although some other approaches are now moving into human testing. An efficacy study of low-dose (52.5 mg) nonoxynol-9 reported as a late-breaker showed no apparent efficacy in preventing sexually transmitted disease transmission, although the product was safe and did not cause damage to vaginal tissues. Rosenberg noted that similar results had been reported last year from a UNAIDS study of a low-dose nonoxynol-9 film. Higher-dose nonoxynol-9 products are still being evaluated.
Several products that take another microbicide approach, reducing the vaginal pH to a level that should inhibit HIV replication, are being developed. Phase I safety data on one of these, called BufferGel (see BETA, March 1997, page 19), was reported in Geneva. Sixteen sexually abstinent and 11 monogamous women used the product once or twice per day for 14 days, with no serious adverse events. A number of relatively mild symptoms were noted, with vaginal itching being the most common. Three women stopped using the product, two developed vaginal candidiasis, and the third developed a hyperkeratotic lesion (both conditions may increase the risk of HIV transmission). The researchers reported that most of the adverse events did not seem to be dose-related, and survey results indicated that most of the women found the product acceptable. Further testing of BufferGel is underway.
Rosenberg noted that the first human studies are about to begin of a gel formulation of PMPA, which has shown success in protecting monkeys from SIV infection (see BETA, June 1996, page 6). A different PMPA preparation is also being studied as an antiretroviral treatment for those with HIV infection. Other products with preclinical development work presented at the conference include PRO 2000 and UC781. PRO2000 has been shown to protect cells from infection by HIV and other sexually transmitted organisms in vitro (see BETA, March 1997, page 19). UC781 is a non-nucleoside reverse transcriptase inhibitor developed by Uniroyal; it has a half-life in the body that makes it inappropriate for therapeutic use, but has chemical properties that may make it stable in the vaginal environment.
One important issue that has moved onto researchers’ radar screens is rectal use of microbicide products. Rosenberg noted, "once these products are shown effective, they will be used rectally…there may be a difference in irritation in vaginal and rectal sites."
Bruce Mirken is a freelance writer based in San Francisco.
Editor's note: all abstracts are from the 12th World AIDS Conference. Geneva Switzerland, June 28-July 3, 1998.
Administration of zidovudine during late pregnancy and delivery to prevent perinatal HIV transmission--Thailand, 1996-1998. Morbidity and Mortality Weekly Report. 47(8): 151-154. 1998.
Balzarini, J. and others. The thiocarboxanilide UC-781 is a potential virucidal drug. Abstract 33170.
Beckerman, K. and others. Control of maternal HIV-1 disease during pregnancy. Abstract 12151.
Dorenbaum-Kracer, A. and others. Antiretroviral use in pregnancy in PACTG316; phase III randomized, blinded study of single-dose intrapartum/neonatal nevirapine to reduce mother to infant HIV transmission. Abstract 23281.
Duarte, G. and others. Influence of maternal weight gain on vertical transmission and pregnancy outcome among HIV+ women. Abstract 32232.
Duerr, A. and others. The effect of highly active antiretroviral therapy (HAART) on cervical dysplasia and HPV infection among HIV-infected women. Abstract 60289.
Fontanet, A. Protection against STDs by granting sex workers the choice of using the male or female condom: results from a randomized controlled trial. Abstract 20/33608.
Heard, I. and others. Early regression of cervical lesions in HIV-seropositive women receiving highly active antiretroviral treatments (HAART). Abstract 22312.
Kramer, F. and others. Combination therapy with nevirapine, zidovudine and a second nucleoside analog during pregnancy. Abstract 463/12152.
Lambert, J. and others. Risk factors for perinatal HIV transmission in women/infants receiving standard zidovudine (ZDV) prophylaxis. Abstract 23265.
LaRuche, G. and others. Short term course of cervical squamous intraepithilial lesions (SILs): Relation with HIV and human papilloma virus (HPV) infections in Africa. Abstract 22319.
Lawson, L. and others. A prospective study of couples who use the female condom: Exposure to semen during use. Abstract 33132.
Lorenzi, P. and others. Safety of combined antiretroviral therapies with or without protease inhibitors in pregnant HIV-infected women and their offspring. Abstract 32453.
Mandelbrot, L. and others. Decreased perinatal HIV-1 transmission following elective cesarean delivery with zidovudine treatment. Abstract 23272. Full article in Journal of the American Medical Association 280(1): 55-60. July 1, 1998.
Read, J. Mode of delivery and vertical transmission of HIV-1: a meta-analysis from fifteen prospective cohort studies (the International Perinatal HIV Group). Abstract 23603.
Shah, K. and others. Relationship of HIV-1 RNA copies in plasma and CD4 counts to human papilloma virus (HPV) prevalence and cervical dysplasia. Abstract 22317.
Taha, T. and others. HIV infection due to breast-feeding in a cohort of babies not infected at enrollment. Abstract 23270.
Copyright © 1998 - Bulletin of Experimental Treatments for AIDS (BETA). Reproduced with permission. BETA is published four times a year by the San Francisco AIDS Foundation. All rights reserved. Noncommercial reproduction is encouraged. Subscription lists are kept confidential. Call 415.487.8060; FAX: 415.487.8069. Mailing Address: P.O. Box 426182, San Francisco, CA 94142-6182. beta@sfaf.org http://www.sfaf.org/beta.html
This information is designed to support, not replace, the relationship that exists between you and your doctor.