Background/Purpose:

Antiphospholipid Syndrome (APS) is characterized by recurrent thrombotic and obstetric manifestations in the presence of ‘criteria’ antiphospholipid antibodies (aPL) [anticardiolipin (aCL), anti-β2glycoproteinI (anti-β2GPI ) and lupus anticoagulant (LA)]. However, aCL assays lack specificity, while anti-β2GPI assays lack sensitivity in the classification/diagnosis of APS. The APhL ® assay (Louisville APL Diagnostics), which detects antibodies against a mixture of negatively charged phospholipids, has been shown to be more specific and as sensitive as aCL in the diagnosis of APS in small observational studies. Our aim was to examine the performance of this assay in a large group of ethnically diverse lupus patients.

Methods:

A total of 1178 serum samples from patients with lupus (1997 ACR criteria) from the HOPKINS (n=543), LUMINA (n=588) and Jamaican cohorts (n=47) were examined for IgG and IgM positivity in aCL (in-house), anti-β2GPI (INOVA) and APhL ELISA assays. Chi-squared or Fisher’s exact test univariate analysis along with odds ratios/95% confidence intervals were used to evaluate assay correlation with clinical features (SPSS® v20.0). ROC analysis was performed to determine sensitivity, specificity, and likelihood ratios for all assays in predicting clinical features. A case series analysis was also performed of patients in whom criteria aPL assays were negative but the new APhL assay was positive.

Results:

Both IgG APHL (OR2.3, 95%CI 1.5-3.7, p<0.001) and IgM APHL (OR1.9, 95%CI 1.1-3.4, p=0.027) were significantly correlated with thrombosis, while only IgG APHL (OR1.8, 95%CI 1.1-2.9, p=0.015) was significantly correlated with pregnancy morbidity. The IgG APHL assay had the greatest performance value as measured by positive likelihood ratio (2.2 for thrombosis and 1.7 for pregnancy morbidity), increased sensitivity as compared to anti-β2GPI, and improved specificity compared to aCL assays (tables 1 and 2). Approximately 2% (23/1178) of our patients tested positive for APHL and negative to criteria assays. Of these, 47.8% (11/23) had APS clinical manifestations, including thrombotic and pregnancy related morbidity.

Conclusion:

Overall, APhL antibodies, especially IgG, represent a promising biomarker for the classification of APS patients as well as for risk assessment in APS patients with regards to pregnancy morbidity and thrombotic manifestations. It has encouraging clinical value and potential to be part of the diagnostic criteria for APS.

Table 1: ROC Analysis: AUC, Sensitivity, Specificity and LR+ of Assays for Predicting Thrombosis

Any Thrombosis	AUC	AUC p-value	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	LR+
APHL_IgG	0.558	0.0040	13.8	93.8	40.2	78.3	2.2
APHL_IgM	0.531	0.1310	6.7	96.2	34.6	77.3	1.8
ACL_IgG	0.477	0.2530	25.8	80.1	28.2	78.1	1.3
ACL_IgM	0.377	< 0.0001	16.6	75.5	17	74.9	0.8
B2GPI_IgG	0.594	< 0.0001	8.9	95.9	39.3	77.7	2.1
B2GPI_IgM	0.578	0.0001	4.1	97.5	33.3	77.1	1.7

Table 2: ROC Analysis: AUC, Sensitivity , Specificity and LR+ of Assays for Predicting Pregnancy Morbidities

Any Preg Morbidity	AUC	AUC p-value	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	LR+
APHL_IgG	0.580	< 0.0001	9.8	94.2	41.7	70.9	1.7
APHL_IgM	0.523	0.2440	4.9	95.7	32.6	70.2	1.1
ACL_IgG	0.456	0.0240	19.2	77.9	27.1	69.3	0.9
ACL_IgM	0.417	< 0.0001	17.9	73.3	22.3	67.7	0.7
B2GPI_IgG	0.573	0.0002	2.9	94.2	17.6	69.4	0.5
B2GPI_IgM	0.573	0.0002	3.3	97.1	32.3	70.2	1.1

AUC: area under the curve.

PPV: positive predicting value.

NPV: negative predicting value.

LR+: positive likelihood ratio.

Pregnancy morbidity: presence of miscarriage and/or pre-eclampsia/eclampsia.

Thrombosis: PE, DVT, TIA/CVA

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/performance-evaluation-and-clinical-associations-of-the-aphl-elisa-compared-to-criteria-antiphospholipid-immunoassays-in-lupus-patients/