893
Multi-center Evaluation of Explant Models for Topical Microbicide Development
James Cummins*1, N Richardson-Harman2, J Bremer3, P Anton4, C Dezzutti5, P Gupta6, N Lurain3, L Margolis7, R Shattock8, and P Reichelderfer7
1Southern Res Inst, Frederick, MD, US; 2BioStat Solutions, Mt Airy, MD, US; 3Rush Med Ctr, Chicago, IL, US; 4David Geffen Sch of Med, Univ of California, Los Angeles Med Ctr, US; 5CDC, Atlanta, GA, US; 6Univ of Pittsburgh, PA, US; 7Natl Inst Child Hlth and Devt, NIH, DHHS, Bethesda, MD, US; and 8St George's Hosp Med Sch, London, UK
Background: Currently
no standard methodology for the evaluation of topical microbicides in ex vivo explant models is available. The reliable assessment of these products
requires reproducible assays that accurately assess virus inhibition. In
an effort to better standardize these models for the preclinical evaluation of
microbicides, this multi-center study was performed to evaluate several explant
models by comparing different virus stocks, in-house vs. centralized p24
determinations, and maximum virus growth as an endpoint.
Methods: Growth
of HIV-1Ba-L in three explant models (cervical, rectal, and tonsil)
from five laboratories was examined. A common stock of virus was distributed,
and each laboratory evaluated both the common and in-house HIV-1Ba-L
in their respective explant models. Supernatants were taken every 2-3 days for 15
days, and virus growth was determined by measuring HIV-1 p24 antigen in-house
and at a centralized laboratory.
Results: Since
virus growth can vary according to the day of observation, comparisons were
made when exponential virus growth was achieved. Optical Density (OD) data from
all p24 assays (n=92) were analyzed with a universal, standard curve for OD values
limited to those within a +/- 95% confidence interval of the plate standards. When
used as the endpoint, a binary classification system for virus growth determined
that maximum virus growth occurred most often at day 15 in 3 of the 5
laboratories. Explant tissues infected with the common HIV-1Ba-L
stock resulted in higher p24 levels compared to in-house virus stocks
(p<0.01). The majority of laboratories (3/5) showed no difference in p24
values that were determined in-house vs. the centralized laboratory.
Conclusions: For standardization of ex vivo explant models among multiple laboratories, these study
results suggest the use of a common virus source but not a central laboratory
for p24 analysis. In addition, the developing statistical algorithm indicates a
binary classification system of virus growth as the primary endpoint as well as
an assay duration of at least 15 days. These parameters will be used in future explant
studies to reliably compare topical microbicide activity between different explant
models and laboratories.
|
