|
MULTICENTRE
STUDIES
FACTORS THAT INFLUENCE THE RESULTS
OF ANTI-β2GPI ELISA
COORDINATORS
Guido Reber1, Anna
Cavazzana2, Philippe de Moerloose1
1Haemostasis Unit, Geneva University Hospital,
Switzerland and 2University of Padova, Italy
PROJECT
A previous collaborative study from
the Standardisation Group of the European Antiphospholipid Forum on anti-β2GPI
antibodies home-made assays has shown that differences at the antigen level
could explain only in part the variability in assay results. In this study,
participants were asked to compare the results of selected samples obtained
with their routine β2GPI preparation with a common β2GPI
and two calibrators sent to all laboratories, using their local protocol. They
were also asked to fill a questionnaire describing their local protocol.
On the basis of
the technical data provided by the participants for each step of the local
ELISA protocol, we tested the influence on assay results of pH of the coating
solution, of some microplate brands, the blocking solution, the ionic strength
of the dilution buffer, as well as the effect of calcium ions and of various
enzyme-conjugated signal antibodies. We tested 32 patient’s samples (25 APS, 2
SLE and 5 other autoimmune diseases) that were quantified with the Forum
Calibrators, the monoclonal antibody HCAL and our local (single patient)
calibrator.
The influence of
pH of the coating buffer, either neutral (PBS) or alkaline (carbonate pH 9.6),
on patients’ OD was tested with the Stago, the Hyphen and a home-made β2GPI.
There was no difference except for the home-made β2GPI for which
slightly higher values were observed at alkaline pH. The binding of each
calibrator to the microplate varied depending on the brand. This was also the
case for some patients. In consequence, the combination of both binding
patterns (calibrator and sample) resulted in different results as a function of
the microplate brand. The use of BSA or gelatine in the blocking buffer, as
well as omitting the blocking step, did not change mean results when using HCAL
for calibration whereas the absence of blocking resulted in higher values when
using the Forum Calibrators. Sample-blanks were clearly higher when using
gelatine with both Costar and Maxisorp microplates. Increasing the ionic
strength affected more the binding of Forum Calibrators than that of HCAL whereas
the opposite pattern occurred in the presence of calcium ions. Five
enzyme-conjugated signal antibodies were studied. Their pattern of binding to
Forum Calibrators was more homogenous than to HCAL. Therefore, the mean value
of patients results varied widely according to the conjugate and the calibrator
used.
For additional
information and applications, please e-mail: guido.reber@hcuge.ch
|
Web contens: EWPSLE Info Desk. |
|
|
Last updated: 17 November 2005 |
|