Background/Purpose: Several immunoassays are available for the detection of anti-cardiolipin (aCL) and anti-beta2glycoprotein I antibodies (anti-B2GPI), but standardization and harmonization of these tests is an ongoing process. Here we aimed to comparing the performance of an automated chemiluminescence assay with a validated in-house ELISA method, using clinically characterized samples derived from a routine diagnostic workup for the detection of antiphospholipid antibodies (aPL).
Methods: A total of 400 samples with a complete aPL panel (aCL IgG/IgM and anti-B2GPI IgG/IgM determined by an in-house ELISA, Lupus Anticoagulant (LA) tested with a DRVVT and aPTT based method) were collected. All the samples were tested for aCL IgG/IgM and anti-B2GPI IgG/IgM by QUANTA Flash CIA (INOVA). The two assays were compared using a definition of aPL profile for each isotype (single vs double positive samples). The clinical diagnosis/reason for aPL detection was retrieved. Subjects were grouped upon the clinical situation: A) clinical features compatible with Antiphospholipid Syndrome (APS) and/or other systemic autoimmune rheumatic diseases (SARD) (patients considered at risk of aPL-related manifestations); B) clinical features other than A; C) unresolved clinical picture.
Results: The two assays displayed a good overall agreement for both IgG and IgM detection (81% and 84% respectively). Such agreement can be estimated to be higher after reconciliation of discrepant results using clinical features and degree of aPL positivity. The majority of discrepant results for both IgG and IgM assays were due to low titer values. No significant difference was found in the rate of group A patients in the two subgroups of discrepant samples (ELISA pos – CIA neg and viceversa) (Chi squared test). LA was positive in 11/75 (15%) for IgG assays and in 9/63 (15%) for IgM assays discrepant samples.
IgG assays |
QUANTA Flash CIA |
||||
aCL pos and anti-B2GPI pos (double pos) |
aCL pos or anti-B2GPI pos (single pos) |
aCL neg and anti-B2GPI neg (double neg)
|
Total |
||
In-house ELISA |
aCL pos and anti-B2GPI pos (double pos) |
13 (3.3%) |
0
|
0
|
13 (3.2%) |
aCL pos or anti-B2GPI pos (single pos) |
6 (1.5%) |
14 (3.5%) |
35 (8.8%) |
55 (13.8%) |
|
aCL neg and anti-B2GPI neg (double neg) |
8 (2%) |
26 (6.5%) |
298 (74.5%) |
332 (83%) |
|
Total |
27 (6.8%) |
40 (10%) |
333 (83.2%) |
400 (100%) |
IgM assays |
QUANTA Flash CIA |
||||
aCL pos and anti-B2GPI pos (double pos) |
aCL pos or anti-B2GPI pos (single pos) |
aCL neg and anti-B2GPI neg (double neg)
|
Total |
||
In-house ELISA |
aCL pos and anti-B2GPI pos (double pos) |
12 (3%) |
5 (1.3%) |
6 (1.5%) |
23 (5.8%) |
aCL pos or anti-B2GPI pos (single pos) |
3 (0.8%) |
7 (1.8%) |
46 (11.5%) |
56 (14%) |
|
aCL neg and anti-B2GPI neg (double neg) |
0 |
3 (0.8%) |
318 (79.5%) |
321 (80.2%) |
|
Total |
15 (3.8%) |
15 (3.8%) |
370 (92.4%) |
400 (100%) |
Discrepant results |
IgG assays (n=75) |
IgM assays (n=63) |
ELISA pos – CIA neg IgG n=35/75 (47%) IgM n=57/63 (90%) Positive LA n=11 (15%) |
Group A: 23/35 (66%) Group B: 87/35 (23%) Group C: 4/35 (11%) Low titer: 27/35 (77%) |
Group A: 40/57 (70%) Group B: 13/57 (23%) Group C: 4/57 (7%) Low titer: 45/57 (79%) |
ELISA neg – CIA pos IgG n=40/75 (53%) IgM n=6/63 (10%) Positive LA n=9 (15%) |
Group A: 30/40 (75%) Group B: 7/40 (18%) Group C: 3/40 (7%) Low titer: 34/40 (85%) |
Group A: 5/6 (83%) Group B: 1/6 (17%) Group C: 0 (11%) Low titer: 4/6 (67%) |
Conclusion: In-house ELISA assays and the CIA assays displayed a comparable performance in the detection of aCL and anti-beta2GPI. Discrepant results were mostly due to values in the low positive range. The clinical features of the subjects are crucial information to interpret results when comparing different assays for aPL detection.
Disclosure:
A. Zanola,
None;
L. Andreoli,
None;
C. Nalli,
None;
F. Allegri,
None;
M. Mahler,
Inova Diagnostics, Inc., San Diego, CA,
3;
G. Norman,
Employee of INOVA Diagnostics,
3;
A. Tincani,
None.
« Back to 2014 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/performance-of-an-automated-chemiluminescence-assay-for-anti-cardiolipin-and-anti-beta2glycoprotein-i-antibodies-detection-in-a-cohort-of-400-clinically-characterized-consecutive-routine-samples/