Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

624 Abstracts 33160-33164

12th World AIDS Conference

Lessons Learned: We must reframe the complex relationships between microbicides, receptive sex and anatomy, and microbicide research must adopt a much
broader field of action than simply coitus. The rationale used to justify vaginal
microbicides is that women need prevention methods they can control when men
are unable or unwilling to wear condoms. This perspective obscures a much more
important issue. People must be able to protect themselves regardless of the
receptive organ(s) with which they copulate.
33160 Should vaginal microbicides be contraceptive or
non-contraceptive?
Elizabeth McGrory. Population Council, Idaghammarskjold Plaza, New York NY
10017, US
Background: Currently available methods of protection from sexual transmission
of HIV also prevent conception. Therefore, women who wish to become pregnant
put themselves at risk of HIV infection. This is particularly problematic for women
in areas with high prevalence of HIV infection and where women's status is
closely linked to fertility and motherhood. Some advocates and scientists have
articulated the need for a female-controlled method of HIV prevention that would
permit conception. This remains a topic of strong debate.
Methodology: Focus group discussions were held in South Africa and the
United States with women of reproductive age, recruited from family planning
clinics and primary health care centers. The women's perceptions and preferences regarding HIV protection were explored, including their views on the
need for a contraceptive or a non-contraceptive vaginal microbicidal product.
The author also reviewed the limited existing evidence in both the published and
non-published literature on the issues.
Results: Preliminary results indicate that women perceive the need for a range
of HIV prevention methods that are within their personal control. To meet women's
needs throughout their reproductive lives, this would include products that protect
against both HIV and pregnancy, as well as products that protect from HIV while
permitting conception.
Conclusions: Given that many women, especially those in the developing
world, are likely to spend some of their reproductive lives at risk of HIV infection
while still desiring pregnancy, scientists should continue pursuing promising leads
for microbicidal agents regardless of their contraceptive properties. Every effort
should continue to be made to develop both contraceptive and non-contraceptive
compounds.
33161 The effectiveness of microbicides for HIV prevention
Charlotte Watts1, W. Thompson2, L. Heise3. 1Health Policy Unit LSHTM Keppel
St. London WC1E7HT; 2HPA London School of Hygiene & Tropical Medicine
London, UK; 3Health and Development Policy Project Washington DC, USA
Aims and Methods: When used correctly, condoms reduce the per sex act
probability of HIV transmission by 90-95%. Research indicates that this figure
may be lower for microbicides. A mathematical model was used to compare the
levels of protection provided to a sub-population by different products (such as
microbicides or condoms).
Findings: An important indicator of protection is the multiple of a product's
1) efficacy (how good is it at prevention HIV transmission in any single act of
intercourse); 2) average consistency of use within a partnership; and 3) the extent
it will be used (coverage). Greatest protection is obtained when the majority
use a high efficacy method (such as condoms) most of the time. Where this is
not possible, higher levels of protection can be obtained by the use of a low
efficacy method with high levels of consistency than by a high efficacy method
used in-frequently. Likewise, a lower efficacy method used by a large proportion
of a sub-population will provide greater protection than a lower coverage of a
high efficacy method. The added levels of protection obtained from the use of
microbicides, and the implications of a decrease in coverage or consistency of
condom use as a result of the introduction of microbicides were also explored.
Conclusion: Although microbicides may have a lower efficacy than condoms,
they may provide equal or greater protection in sub-populations where they can
be used by a corresponding greater proportion of the population, or far more
consistently than condoms. The use of microbicides as a fall-back method to
condoms may also avert substantial HIV infection.
| 33162 | Can zidovudine monotherapy continue to reduce
perinatal HIV transmission? The North Carolina
experience 1993-1997
Susan Fiscus1, A.A. Adimora1, V.J. Schoenbach1, C. Wilfert2, V.A. Johnson3.
NC-Children's AIDS Network NC; ' Univ. North Carolina, CB #7140, Chaper Hill,
NC, 27513; 2Duke Univ., Durham, NC; 3Univ. Alabama, Birmingham AL, USA
Objectives: To monitor identification of HIV positive women, zidovudine (ZDV) utilization, ZDV efficacy in interrupting maternal to infant transmission, and incidence
of ZDV resistance.

Methods: Population based, retrospective review of the medical records of
infants born in North Carolina from 1993-1997 who were tested for HIV infection by HIV DNA PCR (Roche Amplicor) and HIV culture. Genotypic (Affymetrix
GeneChip'") and phenotypic ZDV resistance of the infected infants' viral isolates
were determined. Data for 1997 are still being collected.
Results: Administration of the full ACTG 076 regimen was best in preventing
transmission (10/293, 3.4%), compared with partial 076 therapy (8/94, 8.5%)

or no ZDV (38/140, 27.1%). Use of any maternal ZDV was significantly (p <.001) associated with prevention of transmission (15/373, 4%) compared with no
maternal ZDV (41/154, 27%). In 1995-97, 351 of 395 (89%) HIV infected women
were identified prenatally; only 8 (2%) refused maternal ZDV. Of 31 infected
infants born in 1995-97 16 in 1995, 12 in 1996, 3 in 1997), 58% did not receive
any ZDV, 19% received some ZDV, and 23% received the complete 076 regimen.
Isolates from 54 of 65 infected infants from 1993-97 were studied for genotypic
resistance. Only 5 had any predominant ZDV-associated genotypic mutations: 3
had single K70R, one had single T215Y, and 1 had multiple mutations - M41L,
L210W, T215Y. None of the 49 isolates from infected infants tested had high level
ZDV phenotypic resistance (IC5so > I/M).
Conclusions: ZDV monotherapy continues to be effective in reducing perinatal
HIV transmission. Women transmit more often if the complete ACTG 076 ZDV
regimen is not taken. The number of infected infants born each year in NC continues to decrease and resistance to ZDV has been detected infrequently in NC to
date. Failure to identify HIV infected women is the primary reason for persistent
mother to infant transmission in the state.
30*/33163 Randomized placebo-controlled trial of short-course
oral ZDV to reduce perinatal HIV transmission,
Thailand
Nathan Shaffer12, C. Bhadrakom3, W. Siriwasin4, R. Chuachoowong1,
P. Mock1, N.L. Young2, S. Chearskul3, T. Chotpitayasunondh5, J. Karon6,
R.J. Simonds6, T.D. Mastro2. 1HIV/AIDS Collaboration PO Box 139, Nonthaburi
11000; 2CDC & HIV/AIDS Collaboration Nonthaburi; 3Mahidol University
Bangkok; 4Rajavithi Hospital Bangkok; 5Children's Hospital Bangkok, Thailand;
6CDC Atlanta G.A., USA
Background: Many developing countries with high burdens of perinatal HIV
infection have not implemented the ACTG 076 regimen because of logistical and
cost barriers. If proven safe and effective, a short-course oral zidovudine (ZDV)
regimen administered to HIV-infected women late in pregnancy could become
a widely implemented minimum alternative regimen for reducing perinatal HIV
transmission.
Methods: A randomized, double-blind, placebo-controlled trial is being conducted in Bangkok at Thailand's two largest maternity hospitals. The study
regimen is oral ZDV 300 mg taken twice daily by HIV+ women beginning at
36 weeks gestation and every 3 hours during labor. No ZDV is given to the
newborn. Mothers do not breast-feed, consistent with national guidelines. The
target sample size is 392 women (196 in each arm), and has the power (a
= 0.05, / = 0.2) to detect a 50% decrease in transmission from a presumed
background rate of 24%. Infants are tested by DNA PCR at birth, 2 and 6 months.
Efficacy is based on estimated transmission rates at 6 months by Kaplan-Meier
analysis of time to a positive PCR test. The study is monitored by an independent
DSMB.
Results: Full study enrollment was completed from May 1996 through December 1997. Through 1997, there were 382 births (2 sets of twins); 313 children
have been followed through 2 months and 222 children through 6 months. The
median age of women at enrollment is 24 yrs (range 17-39) and the median
CD4+ cell count is 412 cells//IL (range 6-1317). The median number of days
on study drug before labor is 24 (range 3-58) and the median number of labor
doses is 3. Weekly reviews of study drug, pill counts and qualitative interviews
indicate a very high level of adherence (>95%) and tolerance. No woman has
had to stop study drug. Thus far, more than 95% of expected study visits have
been completed and <5% of children have been lost to follow-up.
Conclusions: Full study enrollment was completed in 1997. The study regimen, including the oral labor doses, has been well tolerated and adherence and
follow-up have been >95%, suggesting that this short-course regimen would be
feasible to implement in Thailand, and perhaps in other developing countries.
There were two interim efficacy reviews by the DSMB. We expect to announce
nearly final results by March 1998.
S33164 Changes in plasma viral load related to short-course oral
zidovudine (ZDV) during late pregnancy
Nathan Shaffer1, Anuvat Roongpisuthipong1, W. Siriwasim3, N.L. Young2,
R. Chuachoowong1, P. Mock1, C. Bhadrakom1, S. Asavapiriyanont3, J. Karon4,
T.D. Mastro2, R.J. Simonds4. 1CDC & HIV/AIDS Collaboration, HIV/AIDS Coil
Aboration, PO. Box 139, Nonthaburi 11000; 2 Mahidol University, Bangkok;
3Rajavithi Hospital, Bangkok, Thailand; 4CDC, Atlanta, GA, USA
Background: ACTG 076 viral load data suggest that ZDV may have only a
small effect on viral load. Recent reports suggest that cessation of antiretrovirals
may produce a viral load rebound. We investigated changes in plasma viral load
related to ZDV as part of a clinical trial of short-course oral ZDV given late in
pregnancy.
Methods: Sequential plasma specimens from a consecutive subset of women
enrolled in a placebo-controlled clinical trial of oral ZDV (300 mg twice daily)
beginning at 36 weeks (36 W) gestation and every 3 hours during labor were

batch tested by RNA PCR (Roche Monitor). Normalized viral load results (log10)
at baseline (36 W), on study drug (37 W, 38 W, delivery (DEL)), and 1 month after
treatment were compared by t test and paired t test. Groups were coded as "E"
and "F", pending the unblinding of the trial.
Results: Of the first 40 women tested, 19 were in group E and 21 were in group
F. CD4 counts were comparable in the 2 groups at baseline (median 377 and
373 cell/mm3, P = 0.9) and delivery (median 370 and 444 cells/mm3, P = 0.97).