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

12th World AIDS Conference

Abstracts 41143-41147 771

ratio in blood and LN (P = 0.001). Pts were divided according to median viremia
at the time of LN biopsy.

LNC RNA/106 CD4
PBMC RNA/106 CD4
Blood CD4/CD8 ratio
LN CD4/CD8 ratio

Viremia
S100 copies/ml* - 100 copies/ml*
2.85 (1.58-5.39) 5.13 (2.92-6.31)
2.37 (1.48-3.43) 2.98 (2.13-3.80)
0.73 (0.43-1.23) 0.42 (0.26-0.60)
2.63 (1.50-3.36) 1.66 (0.78-2.44)

0.009
0.02
0.003
0.02

*Median (range), #Mann-Whitney test
Conclusions: Only 3 pts on ddl/d4T/HU had viremia <20 copies/ml. Significant correlations between viremia and cell associated RNA in blood and LN, and
between CD4/CD8 ratio in blood and LN were observed. None of the pts had
undetectable LNC RNA. Subgroup analysis reveals that in pts with viremia <100
copies/ml, LNC RNA is 10X higher than PBMC RNA, whereas in pts with >100
copies/ml, LNC RNA is 100X higher than PBMC RNA.
41143 Genotyping HIV-1 reverse transcriptase to detect
AZT-resistant mutations
Robert Heimer, T.C. Kyriakides. Yale University School of Medicine 60 College
St POB 208034 New Haven Connecticut, USA
Background: In addressing the issue of drug resistance, we developed an assay
for the rapid detection of zidovudine (AZT)-resistance associated mutations in the
reverse transcriptase (RT) gene of HIV-1.
Methods: In this two-step procedure, polymerase chain reaction (PCR) with
(for viral RNA) or without (for proviral DNA) initial cDNA synthesis is followed by
a modified Ligation/Gap-filling chain reaction. Mutant detection is based upon the
extension of upstream [32P-labeled] oligonucleotide probes and the ligation to a
downstream probe. Discrimination between mutant and wild-type resides in the
3' nucleotide of the upstream probe; use of a DNA polymerase without the 3'-5'
exonuclease activity prevents extension and subsequent ligation of the probe if
the 3' end is mismatched. Ligated products are separated from unligated probes
by denaturing acrylamide gel electrophoresis and visualized by autoradiography.
Results: The assay was used to detect mutations at codons 67 and 215 in an
AZT-resistant strain (A018-G910-6), pBH10 (wild-type sequence-containing plasmid) and paired patient samples (pre and post-AZT). A mixture of wild-type/mutant
populations was detected in the AZT-resistant strain whereas purely wild-type
populations were detected in pBH10. Four of the five pre-AZT patient samples
were shown to contain only wild-type populations (the band from the fifth sample
was undetectable); all five post-AZT samples showed a shift towards a mixed
wild-type/mutant population.
Conclusions: A novel, resistance-screening assay permits the detection of
wholly wild-type, wholly mutant, or mixed populations of HIV-1 for each point mutation associated with AZT resistance. The percentages of wild-type and mutant in
the mixed populations can be quantified. This assay, combined with vital load and
other surrogate markers, can prove to be important in HIV pharmacotherapeutics.
41144 Lymphocyte-derived and plasma HIV-RNA viral loads in
patients with HIV infection
Nurith Vardinon1, I. Yust2, Y. Sela3, I. Zeldis3, M. Burke3. Clinical Immunology
Unit, Tel Aviv Medical Center, G. Weizman St. Tel Aviv; 2Medicine A, Tel Aviv
Medical Center, Tel Aviv; 3Clin. Immunology, Tel Aviv Medical Ctr, Tel Aviv, Israel
Objective: To measure the HIV-RNA viral load in samples prepared from lymphocytes and to compare it with that obtained in plasma and to the CD4 cell count in
patients with HIV infection.
Methods: 26 adult patients with HIV infection were included in the study.
Lymphocytes were isolated by density gradient centrifugation on a polysucrose -
sodium metronizoate medium, Uni-sep (Novamed, Israel). HIV-RNA viral loads in
lymphocytes and plasma were measured utilizing Nuclisens (Organon Teknika,
Belgium) technology. CD4 cell counts were estimated using the FACS (Becton
Dickinson).
Results: A high overall correlation was found between viral loads in lymphocytes and plasma (r = 0.83), although correlation with CD4 cell counts was poor (r
= -0.38, r = 0.47 for lymphocyte-derived and plasma viral loads respectively).
The plasma HIV-RNA was very low (<100 copies/ml) in a subgroup of seven
patients and in five of these moderately high HIV-RNA levels were detected in
lymphocytes. In this subgroup, the correlation between lymphocyte-derived and
plasma viral loads was poor (r = 0.16), although a fair correlation was found
between the CD4 cell count and the lymphocyte-derived viral load (r = 0.70) but
not the plasma viral load (r = 0.27).
Conclusion: In patients with HIV infection there is a good correlation between
the viral loads obtained in plasma and samples prepared from lymphocytes. However, at low levels of detection, the correlation is poor. In other words, HIV-RNA is
frequently detected in lymphocytes even when the plasma viral load is very low.
This difference in sensitivity may in part be explained by a population of latently
infected lymphocytes, which do not release the virions into the plasma.

41145 Efficiency of three versions of the Amplicor HIV-1 monitor
test with new gag primers, and a prototype automated
version to quantitate HIV-1 RNA from different genetic
subtypes
Karine Triques', J. Coste2, J.L. Perret3, J. Reynes4, S. Herman5, J. Spadoro5,
M. Peeters'. Orstom Lab. Retrovirus, 911, Av. Agropolis BP 5045, Montpellier;
2CRTS, Montpellier; 3Hopital Laveran, Marseille; 4Lab. Virologie Hopital St. Eloi,
Montpellier; 5Roche Molecular System; Branchburg, USA
Objectives: Three versions of the Amplicor HIV-1 Monitor Test (v1.0, v1.0+ and
v1.5) and a prototype automated version of the v1.5 test (Cobas Amplicor HIV-1
Monitor), with different gag primers, were evaluated for their ability to detect and
quantify HIV-1 RNA from different genetic subtypes.
Methods: Ninety-six plasma samples from 96 different HIV-1 infected patients
(55 A, 9 B, 21 C, 2 D, 7 E and 2 G) and viral supernatants from 29 HIV-1 reference
strains (3 A, 6 B, 5 C, 3 D, 8 E, 3 F and 1 G) were tested in parallel with the
4 tests, according to the manufacturer's instructions. Version 1.0 is the original
commercially available test. Version 1.0+ contains an additional two PCR primers
that bind to the same sites in the gag gene. Version 1.5 and the COBAS test
contain a new primer pair in the gag gene.
Results: The v1.0+ and v1.5 tests detected significantly more subtype A and
E samples then the vl.0 test. Versions 1.0, 1.0+ and 1.5 detected 34, 54 and 55
of the 55 subtype A patient samples and 5, 7 and 7 of the subtype E patient samples, respectively. In total 95 of 96 samples were detected by the v1.5 test. One
subtype B sample from a patient receiving antiretroviral treatment was negative
by all tests. The average viral load for the subtype A and E patient samples was
1.53 and 1.28 log greater, respectively, with the v1.5 test than the v1.0 test. For
the subtype B and C samples only a slight increase (<0.5 log) was observed. A
similar trend was observed for the detection and quantitation of HIV-1 RNA from
the cultured reference strains. The prototype automated test yielded results that
were highly correlated with the vl.5 test, however the COBAS results were slightly
lower. Of the 92 samples tested in parallel, 75% had a difference of less than 0.5
log.
Conclusion: The new gag primers used in the version 1.5 and COBAS
AMPLICOR HIV-1 MONITOR assays yield accurate measurement of HIV-1
viral load for all the subtypes tested (A-G), and should permit the accurate
assessment of disease development and the response to antiviral therapy in
patients infected with Group M subtypes of HIV-1.
41146 1Linear range of Roche Amplicor HIV-1 Monitor assay
using ultra sensitive sample preparation protocol
Jennifer Archer1, C. Shroff2, M. Brownstein2, T. Phipps2, A. Loysen2. 17855
Haskell Ave Suite 302, Van Nuys Ca 91406; 2Consolidated Laboratory, Van
Nuys, Ca, USA
Objective: To determine the linear range of the Amplicor HIV-1 Monitor'" Ultra
Sensitive Specimen Preparation Protocol which increases the sensitivity of the
FDA approved Amplicor HIV-1 Monitor'" assay 10-fold by decreasing the lower
limit of detection to 40 copy/ml.
Methods: A clinical sample serial diluted 2 fold to include 22 dilution points
was used as a panel for the determination of linearity. The Ultra Sensitive Quantitation assay utilizes the standard HIV-1 MonitorTM amplification and detection
procedure but incorporates the Ultra Sensitive Sample Preparation protocol. 500
piL of plasma is centrifuged at 25,500 xg at 4 C for an hour to concentrate the
virus. The pellet is extracted in 600 IL MonitorTM lysis buffer containing 25 IL
quantitation standard. The extracted material is precipitated, washed and dissolved in 100 pL Monitor'm diluent. 50 ItL of preparation is used for amplification.
The described modifications enhance the sensitivity by 10 fold by increasing the
plasma equivalents from 25 1iL in the standard Monitor'" assay to 250 /1L using
the Ultra Sensitive Sample Preparation protocol.
Results: Each dilution point was assayed 20 times. The coefficients of variation
over the entire dilution series range from 163% to 16%, the highest and the lowest
dilution giving the highest CV%. The range of this assay, as determined with
this series of dilutions, is 40-50,000 RNA copy/mL. The precision of the assay
averaged 25%CV within the linear range of 40-50,000 RNA copy/mL. At 20 RNA
copy/ml this assay showed a 55% detection rate and spurious results below this
level which is reflective of the high CV%.
Conclusion: The Ultra Sensitive Sample Preparation protocol enhances the
sensitivity of the Amplicor HIV-1 Monitor~" assay by 10 fold without significant
changes to the already established protocol. With the emergence of new therapies that rapidly lower the viral load below the detectable level of the standard
Amplicor Monitor'" assay, this ultra sensitive sample prep protocol allows accurate
and reproducible HIV-1 viral quantitation as low as 40 copies/ml.
411471 Comparison of proviral DNA in HIV-1 and HIV-2 infected
babies in relation to the mode of mother-to-child
transmission and survival

Abraham Sunday Alabi', K. Ariyoshi1, N. Berry2, D. O'Donovan', M. Ota1,
P.T. N'Gom1, H. Whittle'. 1MRC LAboratories, Fajara PO Box 273, Banjul,
Gambia; 2Univ College London Medical School, London, UK
Objectives: To compare proviral load in HIV-1 and HIV-2 infected babies; and
to relate proviral load to the mode of mother-to-child transmission: peri-versus
postnatal transmission, and to their survival.
Design: Community-based, prospective study.