Abstract Book Vol. 2 [International Conference on AIDS (14th: 2002: Barcelona, Spain)]

XIV International AIDS Conference

Abstracts ThPeC7436-ThPeC7439 447

women using injectable progestins (DMPA), low-dose combined oral contraceptives (COC), and those using non-hormonal or no contraceptive methods.
Results: From 11/1999 to 11/2001, we screened 8,893 women for HIV-1 and
enrolled 5,397 HIV-uninfected women, 88% of the target enrollment number. Enrolled women undergo a pelvic examination, testing for sexually transmitted infections, and complete an interviewer-administered behavioral and contraceptive
history questionnaire. Women are followed at 3-month intervals, and undergo
similar procedures at follow-up visits including HIV-1 testing, for a maximum of
24 months. The study is designed to detect a 2-fold effect of HC use on HIV-1
incidence with 90% power. Enrollment will be completed by the second quarter of
2002, and follow-up will be completed in late 2003.
Conclusions: While a randomized controlled trial would provide the most rigorous design to test our hypothesis, we instead chose a design where women
could select their method of choice. This reduces adherence issues, as well as
the chance ethical issues could result from randomizing women to a method that
might increase risk. We minimized biases by enrolling nearly equal groups of
women in each country who use DMPA, COC, and non-hormonal methods, and
by collecting detailed information on potential confounding factors. This study will
provide critical information regarding the relationship between HC use and HIV-1
acquisition.
Presenting author: Barbra Richardson, Harborview Medical Center, Box
359931, 325 Ninth Avenue, Seattle, WA 98104-2499, United States, Tel.: +1 206
731 2425, Fax: [phone], E-mail: [email protected]
ThPeC7436 The effect of hormonal contraception on genital
tract shedding of human immunodeficiency virus
type-1
C.C. Wang1, R.S. McClelland1, J. Overbaugh2, M. Reilly3, D.D. Panteleeff2,
K. Mandaliya4, B. Chohan1, J. Ndinya-Achola5, J. Bwayo6, J.K. Kreiss.
SUniversity of Washington, Seattle, United States; 2Fred Hutchinson Cancer
Research Center, Seattle, United States; 3University College, Dublin, Dublin,
Ireland; 4 Coast Provincial General Hospital, Mombasa, Kenya; 5 University of
Nairobi, Nairobi, Kenya; 6 University of Nairobi, Nairobi, United States
Background. Previous studies have reported that use of hormonal contraception
(HC) by women may be associated with increased infectivity to uninfected partners by HIV-1 + women. In developing parts of the world, -.25% of women using
family planning choose oral contraceptives (OCs) and -30% use depot medroxyprogesterone (Depo). Because most HIV-1+ women are of reproductive age,
health care providers must provide accurate information about the effect of HC
on the infectivity of HIV-1+ women.
Methods. HIV-1+ women were recruited from a family planning clinic in Mombasa, Kenya. At baseline, samples were collected for detection of HIV-1 RNA and
DNA by polymerase chain reaction. Cervical samples were collected by inserting
a dacron swab into the cervical os and rotating between the fingertips 3 times.
Women were started on the HC method of their choice, and follow-up sample
collections were performed at one and two months.
Results. One hundred one women were started on Depo, 52 on low dose OCs,
7 on high dose OCs, and 50 on progesterone-only OCs. After a mean of 61 days,
plasma and swab HIV-1 RNA concentration did not change significantly, nor did
prevalence of HIV-1 DNA detection. When the analysis was limited to women
with baseline swab RNA levels of at least 1000 copies, the concentration of HIV
RNA detected on cervical swabs decreased significantly for women on Depo and
progesterone OCs. (Depo p=<.001, progesterone OCs p=.006) The prevalence
of cervical ectopy decreased significantly in women after initiation of Depo (41%
to 15%, p=.001), but increased significantly for women after initiation of progesterone OCs (60% to 85%, p=.025).
Conclusions. No change in cervical shedding of HIV-1 infected cells or free virus
was detected after initiation of HC. In the subgroup of women with > 1000 viral copies/cc on cervical swabs, a significant decrease in cervical shedding was
noted for progesterone-containing HC.
Presenting author: Chia Wang, 325 Ninth Avenue, Box 359-908, Harborview
Medical Center, Seattle, WA 98104, United States, Tel.: [phone], Fax:
[phone], E-mail: [email protected]
ThPeC7437 The association of polymorphisms in HLA class I
and TAP genes with resistance to HIV-l infection
C. Liu', M. Carrington2, R. Kaslow3, X. Gao2, C. Rinaldo4, L. Jacobson5,
J. Margolick5, J. Phair6, S. O'Brien2, R. Detels'. 1University of California, Los
Angeles, Los Angeles, California, United States; 2National Cancer Institute,
Frederick, Maryland, United States; 3University of Alabama at Birmingham,
Birmingham, Alabama, United States; # University of Pittsburgh, Pittsburgh,
Pennsylvania, United States; 5Johns Hopkins University Baltimore, Maryland,
United States; 6Northwestern University Chicago, ///inois, United States
Background: Human leukocyte antigen (HLA)-restricted immune responses are
reportedly involved in resistance to HIV infection. This study examined relation
ships of HLA class I and the transporter associated with antigen processing (TAP)
genes with resistance to HIV-1 infection.
Methods: 100 persistently seronegative men repeatedly exposed to HIV-1 and
184 seroconverters with lower risk according to an algorithm based on the number of different insertive anal sexual partners were included. HLA class I genes
were typed by sequence-specific oligonucleotide and sequence-specific primer

methods supplemented by direct sequencing. TAP genes were typed with DNA
single strand conformation polymorphism technique and/or oligonucleotide hybridization. Associations between genetic variants and HIV-1 resistance were explored with univariate and multiple logistic regression techniques.
Results: Among 9 HLA class I supertypes only the A2/6802 supertype excluding A*0201 (OR, 4.4; 95% CI, 1.3-4.8, P=0.009), due completely to the A*0205
subgroup, was associated with HIV-1 resistance. Susceptibility was associated
with the previously implicated Px subfamily (OR, 0.3; 95% CI, 0.1-1.0; P=0.04) of
B*35, suggesting that differential preferences for amino acids at the C terminus influence binding capacity or selection of peptides with proline at position 2 by TAP
gene products. TAP2 665 ala was also associated with resistance (OR, 2.3; 95%
CI, 1.4-3.80; P=0.002), perhaps because of its greater efficiency in transporting
peptides eliciting a greater CD8+ response or because of linkage disequilibrium.
In multivariate analysis, only the A*0205 subgroup (OR, 5.8; 95% CI, 1.4-23.9;
P=0.016) and the TAP2 665 ala (OR, 2.1; 95% CI, 1.2-3.7; P=0.009), remained
associated with resistance.
Conclusions: The HLA A*0205 subgroup and the TAP2 665 ala variant appear
to be associated with resistance to HIV-1 infection.
Presenting author: Roger Detels, UCLA Department of Epidemiology, School
of Public Health, Box 951772, Los Angeles, CA 90095-1772, United States, Tel.:
[phone], Fax: [phone], E-mail: [email protected]
ThPeC7438 High level of HIV-1 infection from anal intercourse:
a neglected risk factor in heterosexual AIDS
prevention
D.T. Halperin, S.C. Shiboski, J.M. Palefsky, N.S. Padian. University of California,
USAID, GH/OHIV 3.6-82, 1300 Penn. Ave., Washington, DC 20523-3700, United
States
Background: Most AIDS prevention programs targeted at the general population continue to emphasize the risks of vaginal and, increasingly, oral sex. Clinical
studies and population-based surveys in countries such as Brazil have found up
to 60% of adults report practicing anal sex. In absolute numbers, it is estimated
that in the United States approximately seven times more women than men-whohave-sex-with-men engage at least occasionally in unprotected receptive anal
intercourse. The practice is relatively common in many other regions of the world,
and is more frequently reported among higher-risk subpopulations such as bisexual men, injecting drug users, sex workers, adolescents and serodiscordant
couples. Reported condom use is consistently lower for heterosexual anal than
for vaginal sex.
Methods: Data from the California Partners Study of heterosexual transmission
in monogamous female partners of HIV-pos. males was used to estimate the percontact risks for anal and vaginal intercourse, and the fraction of observed HIV
infections attributable to anal sex among couples practicing both forms of intercourse.
Results: Adjusted per-act transmission risk (infectivity) for anal contact was 10.3
times higher than the per-act risk for vaginal contact (95% C.I., 2.1-51.3). Among
couples reporting both anal and vaginal intercourse (30% in this sample), an estimated 62% of female HIV infections were due to anal contact (95% C.1, 38%77%).
Conclusions: Our results confirm and quantify previous findings that anal intercourse is a far more efficient mode of transmission than vaginal intercourse, and
suggest that, especially in populations where anal sex is prevalent, a significant
proportion of heterosexual HIV transmission results from this practice. This typically stigmatized and taboo sexual practice must be given greater attention in HIV
risk reduction, microbicide development, condom promotion and other AIDS/STD
prevention programs.
Presenting author: daniel halperin, USAID, GH/OHIV 3.6-82, 1300 Penn. Ave.,
Washington, DC 20523-3700, United States, Tel.: [phone], Fax: +1202
216 3046, E-mail: [email protected]
ThPeC7439 Impact of HIV-1 subtype on viral transmission
X.F Yu', W. Liu2, J. Chen2, O. Laeyendecker3, T. Quinn3, S.H. Lai.
' Department of Molecular Microbiology and Immunology, Johns Hopkins
Bloomberg School of Public Health, Department of Molecular Microbiology and
Immunology Johns Hopkins Bloomberg School of Public Health, 615 N Wolfe St,
Baltimore, MD 21205, United States; 2Guangxi Centers for Diseases Control
and prevention, Nanning, China; 3NIAID, Bethesda, United States
Background: CRF-BC recombinant HIV-1 are rapidly spreading in China, while
other HIV subtypes have been restricted in their original areas. Whether CRF-BC
HIV-1 strains have been adapted to a new environment and became relatively
more infectious than other subtypes is an open question.
Methods: Seroincident cohorts with isolated CRF-AE and CRF-BC prevalent
population were established. HIV and HCV prevalence and incidence were determined. Viral loads and HIV-1 sequences were obtained from HIV-1 prevalent
and incident cases.

Results: HIV and HCV prevalence were comparable among high risk subjects
in the CRF-AE or CRF-BC HIV cohorts. Risk factors contributing to HIV-1 infection were also comparable for these two cohorts. The HIV-1 incidence for the
CRF-AE cohort is 4.9 cases/100 person year (95% CI: 2.1-7.6) after one year
follow-up compared to 9.5 cases/100 person year (95% CI: 6.1-12.9) for the CRFBC cohort. At the same time, the HCV incidence for the CRF-AE cohort is 37.8