Abstracts Vol. 1 [International Conference on AIDS (11th: 1996: Vancouver, Canada)]

Tu.C.343 - Tu.C.43 I1

Tuesday, July 9, 1996

Tu.C.343
HIGH VIRAL LOAD PREDICTS PERINATAL HIV-I SUBTYPE E TRANSMISSION,
BANGKOK,THAILAND
Shaffer Nathan1,2 Bhiraleus P3, Chinayon P4, Kalish M2, Roongpisuthiponi, Sr irwasin
W" Chearskul 53, Chotpitayasunondh T5, Sangkarat S3,Young N2, Brown 2 Batter V2,
Mastro TDI.2 I HIV/AIDS Collaboration, Nonthaburi,Thailand; 2CDC, Atlanta, USA; 3Sinraj
Hospital, Mahidol University Bangkok: 4Rajvithi Hospital, MOPH. Bangkok; 5Childrens
Hospital, MOPH. BangkokThailand.
Objectives: It.has been proposed that subtype E may be more transmissible heterosexually
thar other HIV- I subtypes.We investigated the rate of and risk factors for perinatali HIV- I,
subtype E trianrsmission.
Methods: At 2 large hospitals in Bangkok, HIV-infected (HIV +) pregnant women and their
bottlefed newborn children were enrolled and followed prospectively HIV infection status
of the children was deterrined based on PCR testing and serology at 15 nmonths. HIV-I
sublype was deterrmined by specificV3 loop peptide EIAs. Quantitative HIV RNA testing
(NASBA, Organon Teknika) was pedrformed on rmaternal delivery specimens. Histology for
choiioanronitis was evaluated on a sample of placentas.
Results: From Nov 92 through Mar 94, 342 HIV + women were enrolled from antenatal
clin:s with a background HIV I seroprevalence of 2%. Enrolled mothers tended to be
young (median age 22 yrs), primipara (57%) and asymptomatic (95%). Most (93%) were
seroreactive to FHIV- I subtype E. Infection status could be determined for 28 I (95%)
infants of 295 mothers who rermained in the studyThe transmission rate was 24.2% (95%
CI 19-29%); 26% for subtype E and 14% for subtype B (RR 1.8; p=0.5).Transmission was
associated with: low maternal CD4 count (RR 2.6; p=0.03); vaginal delivery (RR 2.4;
p=0.07); prematurity (RR 2.4; p=0.02) and high (> 8,000 copies/mi) and viral load (RR 2.2;
p=0.004). Chornoamnionitis was more common in HIV + than HIV- control women, but
was inversely associated with transmission (p<O.01). On multivanrate analysis, prematurity
(Of 4.2; p=0.03) and high viral load (OR 3.0; p=0.007) remarined independently associated with transmission.
Conclusions: The perinatal HIV- I transmission rate in Bangkok where subtype E predominates is 24%, comparable to rates reported worldwide for other HIV- I subtypes. Higher
maternal vir al load at delivery is strongly associated with increased transmission risk, but the
risk appears to be multifactorial.
Nathan Shaffec HIV/AIDS Collaboration, 88/7 Soi Bamrasnaradura,Tivanon Road,
Norithabunr II 000,Thailand Tel: [phone], Fax: [phone],
E-mail: nas4(gbangkok.em.cdc.gov
Tu.C.344
THE EFFECT OF MATERNAL VIRAL LOAD ON THE RISK OF PERINATAL
TRANSMISSION OF HIV-I.
Thea, Donald M", Steketee R", Bornshlegel Kr *, PlinerV*, Brown T-. the NYC Perinatal
HIV Transmission Collaborative Study Medical and Health Research Assoc, NYC.,
"tCenters for Disease Control, Atlanta, GA., '*NYC Department of Health.
Objective: To determine the effect of viral load (VL) at delivery on the risk of perinatal
transmission of human immunodeficiency virus type I (HIV).
Methods: A nested case control study was pedformed within a prospectively followed
cohort of HIVrinfected pregnant women and their infants. 51 women (TR) who gave birth
to HIVinfected infants were frequency mnatched within CD4+ cell count quintiles with 54
nontransmitting mothers (NTR). Plasma obtained between 8 weeks pre- and 2 weeks post
delivery was retrospectively assayed for the quantity of HIV RNA using the NASBA assay
system. A value of 1,999 was assigned to samples with VL below the detection lirmit (2,000
RNA copres/ml).
Results: Viral RNA was detected in 73 (70%) of 105 women.The geometric mean VL was
10,300 RNA copies/mi overall; 5.300 in TR and 7,200 in NITR (p=.004). In this CD4 -nmautched sample, the transmision rate was 22% (7/32) among women with VL below
detectionr vs. 65% (17/26) among women with VL in the highest quartile (>32,000). By
logistic regression arnalysis the odds ratio for perinatal transmission of log I 0 VL, adjusted for
maternal CD4 count near delivery (AOR), was 3.1 (95% ClI.5- 6.5).Women withVL above
the threshold for detection were significantly more likely to transmit HIV (44/73) than
women with VL below detection (7/32) (AOR 5.4, 95% Cl 1.9 -15.9). A stratified analysis by
maternal CD4 count and the presence of AIDS indicated that increased VL had the greatest
effect on the likelihood of transmission among women without AIDS and with a CD4I
count > 500. In this group, the likelihood of transmission increased > 18-fold (95% CI 3.6 -90.9) for every 10 fold increase in VL versus 2.3-fold (95% CI 0.6 8.5) increased likelihood
in women without AIDS and CD4 count between 200-500.The effects of other nmaternal
and delivery factors arte being investigated anrid will be reported.
Conclusions: High maternal VL. near delivery increases the likelihood of perinatal transmissior of HIV and this effect appears to be greatest in women without AIDS or advanced
imrunosuppres ionHIV I nfected pregnant women without advanced disease, shown by
iother- ) yhave the lowest risk of perinatal transmission, may benefit the most from efforts
to identify and decrease VLI at delivery.
Donald M.Thea, M. D., 299 River-side Di:, # I0A New York City NY 10025.
ph 12)442 3483. f ix (2i2) 442-3494, email dyt4@wonder:em.cdc.gov
Tu.C.345
HIV-RNA LEVELS DURING PREGNANCY AND VERTICALTRANSMISSION OF HIV-I
Burns David N., Landesman 52, Rubinstein A3, Waters D4, Willoughby A1, Goedert jj5.
-7nCHD rod 5NCI, National Institutes of Health; 2SUNY Brooklyn, NY; 3Albert Einstein,
Bronx, NY; 15AIC, Frederick, MD.
Objective: To assess changes in HIV- I RNA copy number (Copy#) during and after pregnancy and to examine the relationship between antepartum Copy# and vertical transmission of HIV- I.
Methods: Plasma and serum obtained prior to the third trimester (prTM3), during TM3
(TMt3), and two months postpartum (2MPP) from HIV-infected women enrolled in the
Mother and linfiruts Coho t Study between 1986 and I991 were frozen and stored at -
70 until taiwed and a layzed in batch using the Roche Monitor HIV-RNA assay.
Results: here wii ttle overall change in Copy# between prTM3 and TM3 (median difference for n I2 paired data 143: p-0.45) orTM3 and 2MPP (n=87; median, 0: p--O.61).

TM3 data were!aarivie irr 138 women, only six of whom received antiretroviral therapy
during pregnos;,aryv (tvso lraismitted, four did not):

Copy#
< 1,000,000-9,999
I 0,000-99,999
>100,000

No. (%) Trans
O 0
5z 7 (13.5)
I 1 6 (33.3)
18 6 (33.3)

Estimated OR (95% CI)
0. I5 (0.08-2.72)
1.0 PM-H for trend
3.21 (1.2 I -8.52) - 0.001
3.2 I (0.94- 1.0)

A similar overall associationr was seen when these data were analyzed using the Wilcoxon
rank-sum test (p=0.004) and logistic regression analysis (p=0.002).The association persisted
when CD4+ level and duration of ruptured membranes (DROM) were included in the
logistic regression model (p=0.027).
Conclusions: HIV RNA Copy# was stable during pregnancy and the early postpartum perod, suggesting that pregnancy has little general effect on this viral measure.There was a
strong overall association between TM3 HIV-RNA Copy# and vertical transmission of HIVI.This association was independent of CD4+ level and DROM.
D.N. Burns, 6100 Executive Blvd, 4BI I, Bethesda, MD 20892-75 10. USA Tel: [phone]
Fax: [phone] email: burnsd@hd0 l.nichd.nih.gov
Tu.C.430
COMPARISON OF HIV DISEASE PROGRESSION AMONG HOMOSEXUAL MEN,
HETEROSEXUAL MEN AND HETEROSEXUAL WOMEN.
N. Carre*, L. MCeXer, A.Wade*, N. Bialowons*, D.Vittecoq* 1 M. Bary"*, and the SEROCO
Study Group. INSERM U-292, Le Kremlin-Bicetre, France. Hopital P Brousse,Villeuf.
France. **ACCTES, Pans, France
Objective:To compare HIV disease progression in homosexual men. heterosexual men, and
women enrolled in the same cohort.
Methods:The occurrence of HIV-related clinical disease (group IV or AIDS, 1987 revised
Centers for Disease Control and Prevention classification) was compared in subjects with a
known date of infection (Logrank test, Cox model for estimation of relative risks (RR)).
Kaposi's sarcoma was excluded.
Results: After adjustment for age and symptomatic primary infection (SPI). the RR of HIVrelated disease was 1.9 [95%C1: 1.1 - 3.5] in homosexual men compared to heterosexual
men, and I. I [0.5 -2.5] in women compared to heterosexual men. Age and SPI were also
predictive of disease progression.These findings were not explained by differences in followup, calendar year, or higher frequency of a particular HIV-related condition........ H me...,,....... n, -ti ed,,..,...unm n 121
Hetcri e un mn (na 42
Conclusion: Urdefined co-infections, phenotypic or genotypic characteristics of HIV- I
strains, could explain the rapid progression of HIV disease in homosexuals.
L. Meyer, INSERM U292 HOpital de Bicetre. 82, Rue du General Leclerc, 94276 Le KremlinBicetre. Cedex, France
Tu.C.43 I
A UNIQUE COHORT OF IDU INFECTED WITH AN IDENTICAL VIRAL STRAIN OF
HIV: MARKED VARIATION IN CLINICAL PROGRESSION
McMenamin j, Pithie A, Goldberg D, Allardice G. Codere G.Taylor A, Camneron S, Williams
F*,Yirrell D*, Leigh-Brown A, Galbraith* I *, Henderson N"*. Ruchill Hospital, Glasgow
Scotland, *Northern Ireland Regional Tissue Typing Service, Belfast. "Edinburgh University,
***Glasgow Royal Infirmary Scotland, United Kingdom.
Background: Following an outbreak of HIV infection rin HM Prison Glenochil in central
Scotland, UK, in 1993, 14 inmates were identified as HIV positive. The cohort were
remarkably similar; mean age 24 (range 22-28); all male caucasians: all admitted IDU activity
within Glenochii; all were RIBA-2 confirmed HCV Ab positive. In addition 13 of the 14 had
an identical virus as evidenced by results of nested PCR of the p 17 region of gag and V3/V4
region of envy gene; they cluster together as a unique phylogenetic branch. Further the high
degree of sequence homology combined with epidemiological evidence previously presented byTaylor et alI supports the view that 13 of the 14 IDUs were HIV infected within a
few months of each other whilst in HM Prison Glenochil.
Objectives: I.do determine if identical viral strain, time of seroconversion and patient
demography results in similar clinical and immunological progression. 2. Compare rate of
clinical and immunoiogical progression in the Glenochil cohort with other Scottish
IDU seroconvertors
Methods: Prospective cinical and immunological information was recorded for the
Glenochil cohort (I I of original 14) and 25 age and sex matched controls were selected
from other Scottish IDU seroconvertors. An estimated date of seroconversion was calculated for each ind vidual.
Results: Two of the I I Glenochil cohort patients under clinical follow up have since developed an AIDS defining illness (57 and 426 days) and died, while two others have CD4
counts consistently below 200. Interestingly all 4 have a A I B8 HLA class I haplotype (type
11I being processed). Of the 25 IDU seroconvertor control group only one has developed
AIDS (24 I days) while another four have had CD4 counts consistently below 200.
Progression in the Glenochil group to a first CD4 count less than 200 is significantly quicker
(p=O.03,Wilcoxon rank tests). Small numbers preclude analysis of time to AIDS.
Conclusions: Despite identical viral strain, uniform ethnicity IDU risk category gender and
age band, the c inical progression of HIV among the Glenochil cohort varies considerably.
This observation supports the importance of the host immunological response to HIV
infection as does the results of their preliminary HLA typing. However among those individuals who rapidly progress the rate of CD4 decline to a level below 200 cells/crmn is signficantly faster in the Glenochil cohort than that of the IDU seroconvertor group. It is

247