Antiphosphatidylserine/prothrombin antibodies in primary antiphospholipid syndrome

Introduction

Antiprothrombin (aPT) antibodies may be detected by an enzyme-linked immunosorbent assay (ELISA) coating a purified antigen onto irradiated plates ¹ or by using phosphatidylserine/prothrombin complex (aPS/PT) as antigen. ² Several studies investigating aPS/PT have focused mainly on patients with systemic lupus erythematosus (SLE) and/or secondary antiphospholipid syndrome (APS). ^{2 – 4}

In this study, IgG/IgM antibodies directed against PS/PT were assessed in a large cohort of patients with primary APS (PAPS).

Patients and methods

Plasma and serum samples were collected from 158 PAPS patients attending the Rheumatology Unit of Padua University. All of the subjects fulfilled the international consensus statement classification criteria for APS. ⁵ Fifty-six of these (35.4%) had a history of pregnancy morbidity alone including fetal loss, premature birth and early miscarriages (pregnancy morbidity group), and 102 (64.6%) had a history of previous thrombosis with (n = 19) or without (n = 83) pregnancy morbidity (thrombosis group). Fifty-one (50 %) had arterial thrombosis, 27 (26.5%) venous thrombosis and 24 (23.5%) both. One hundred healthy blood donors, age and sex matched with the study population, and 114 patients with autoimmune diseases were included as the control group.

IgG/IgM aPS/PT were detected using commercial ELISA kits provided by INOVA Diagnostics Inc (San Diego, California, USA). In accordance with the manufacturer’s instructions, the IgG/IgM aPS/PT cut-off values were > 30 U/mL for both isotypes. However, our cut-off was arbitrarily set at ≥40 U/mL to increase the specificity of the assay.

IgG/IgM anticardiolipin (aCL) and IgG/IgM antiβ2glycoprotein I (antiβ2GPI) antibodies were measured using an in-house ELISA, as described elsewhere.^6,7 Lupus anticoagulant (LA) was detected according to internationally recognized recommendations. ⁸

Statistical analysis was performed using Pearson’s Chi-square, Fisher’s and Mann-Whitney tests. The multivariate analysis was performed using binary logistic regression.

Results

The prevalence of IgG/IgM aPS/PT in PAPS patients along with that of conventional antiphospholipid antibodies (aPL) is reported in Table 1. The associations between IgG/IgM aPS/PT and the clinical groups of PAPS are shown in Table 2.

OPEN IN VIEWER

Table 1

Prevalence of IgG/IgM aPS/PT and conventional aPL in PAPS patients

Antibody		PAPS n = 158	(%)
aCL	IgG	99	62.7
	IgM	50	31.6
antiβ2GPI	IgG	93	58.9
	IgM	62	39.2
LA		67	42.4
aPS/PT	IgG	44	27.8
	IgM	67	42.4

OPEN IN VIEWER

Table 2

The relationship of IgG/IgM aPS/PT antibodies with clinical features of PAPS

Antibody	Thrombosis n(%) (n = 102)	Pregnancy morbidity n(%) (n = 56)	Controls n (%) (n = 214/n = 172*)	Odds ratio for thrombosis (95% CI)	p-value	Odds ratio for pregnancy morbidity (95% CI)	p-value
IgG aPS/PT	39 (38.2)	5 (8.9)	5 (2.3)/4 (2.3)*	25.7 (9.7–68.1)	<0.001	4.0 (1.0–15.8)	0.043
IgM aPS/PT	56 (54.9)	11 (19.6)	6 (2.8)/4 (2.3)*	42.0 (17.0–103.3)	<0.001	10.2 (3.1–33.5)	<0.001

aPS/PT: antiphosphatidylserine/prothrombin; PAPS: primary antiphospholipid syndrome; CI: confidence interval; *Male patients have been excluded in the statistical comparison with pregnancy morbidity.

The prevalence of IgG and IgM aPS/PT in the different subsets of PAPS are reported in Table 3 and Figure 1. OPEN IN VIEWER

OPEN IN VIEWER

Table 3

Comparison of IgG/IgM aPS/PT prevalence in the different clinical subsets of PAPS

Antibody	Thrombosis/pregnancy morbidity n = 19 n(%)	Thrombosis n = 83 n(%)	Pregnancy morbidity n = 56 n(%)	Odds ratio for thrombosis/pregnancy morbidity vs thrombosis (95% CI)	p-value	Odds ratio for thrombosis vs pregnancy morbidity (95% CI)	p-value
IgG aPS/PT	12 (63.2)	27 (32.5)	5 (8.9)	3.6 (1.3–10.1)	<0.001	4.9 (1.8–13.7)	<0.001
IgM aPS/PT	14 (73.7)	42 (50.6)	11 (19.6)	2.7 (0.9–8.3)	<0.001	4.2 (1.9–9.2)	<0.001

aPS/PT: antiphosphatidylserine/prothrombin; PAPS: primary antiphospholipid syndrome; CI: confidence interval.

The mean titres of IgG and IgM aPS/PT were significantly higher in the thrombosis group than in the group with pregnancy morbidity (62.1 U/mL vs 13.2 U/mL, p < 0.001 and 78.2 U/mL vs 27.8 U/mL, p < 0.001, respectively). When all the patients with arterial (n = 75) and venous (n = 51) thrombosis were analysed separately, IgG aPS/PT antibodies were found to be significantly associated with venous thrombosis (p = 0.023), while IgG and IgM aPS/PT antibodies were significantly associated with arterial thrombosis (p < 0.001 for both).

Logistic regression analysis showed that IgM and IgG aPS/PT were independent risk factors for the presence of thrombosis (OR 3.5 [95% CI 1.6–7.9] and OR 4.1 [95% CI 1.4–11.7], respectively). IgM aPS/PT was an independent risk factor for arterial thrombosis (OR 2.7 [95% CI 1.1–6.7]). Moreover, IgG/IgM aPS/PT were independent risk factors for the presence of LA (OR 6.0 [95% CI 2.7–13.4] and OR 6.3 [95% CI 2.4–16.7], respectively).

Discussion

In this study we assessed IgG/IgM aPS/PT in a large, homogenous cohort of PAPS patients. Taking into account the logistic regression results, high titres of IgG/IgM aPS/PT are independent risk factors for the presence of LA and for thrombosis. Moreover, high titres of IgM aPS/PT antibodies could be considered a marker for arterial thrombosis in PAPS patients.

In our recent paper, ⁹ aPT directed to a purified antigen did not show a significant prevalence in thrombotic versus obstetric PAPS patients, thus, we could affirm that aPS/PT antibodies have a higher clinical relevance when their ability to distinguish thrombosis from obstetric morbidity is evaluated.

Further laboratory and clinical studies are certainly warranted to better define the clinical utility of antiprothrombin antibodies.