Statistical approaches for anti-drug antibody bioassays
Schaarschmidt, Frank, Hofmann, Matthias, Jaki, Thomas, Grün, Bettina and Hothorn, Ludwig A (2009) Statistical approaches for anti-drug antibody bioassays. Biologicals, 37 (5). pp. 323-330. ISSN 10451056
Abstract
Cut points in immunogenicity assays are used to classify future samples into
positive or negative. To determine a cut point, drug-naive samples are analyzed
on multiple microtiter plates taking sources of future variability, such as
runs, days, analysts, gender, spiked/un-spiked samples and replicated wells
into account. Three phenomenons may complicate the statistical cut point
estimation: i) drug-naive samples may contain already ADA-positives, ii)
between-plate heterogeneities may remain after normalization, and iii) complex
small sample size design with crossed and hierarchical factors cause low
power for pre-tests on distribution, outliers or variance structure.
Complex statistical methods for both screening and con�rmation cut point
estimation are provided: i) prediction interval in a mixed e�ects model, ii)
mixture approach in a mixed e�ects model, and iii) distribution diagnostic
in a mixed e�ects model. Related R-programs and an user-friendly program
mixADA are available.
Item Type: | Article |
---|---|
Date Deposited: | 29 Apr 2016 23:45 |
Last Modified: | 29 Apr 2016 23:45 |
URI: | https://oak.novartis.com/id/eprint/22729 |