Elevated plasma level of soluble triggering receptor expressed on myeloid cells-1 is associated with inflammation activity and is a potential biomarker of thrombosis in primary antiphospholipid syndrome

A cross-sectional, case-control study design was used. The cohort consisted of consecutive patients diagnosed with PAPS or asymptomatic carriers of persistently positive APLA under routine follow-up at the Rheumatology Clinic of a tertiary university-affiliated medical center. The study protocol was reviewed and approved by the local Institutional Review Board, and all participants provided written informed consent.

The diagnosis of PAPS was based on the combined presence of one positive clinical criterion and one positive laboratory criterion (aCL and/or aβ2G1 antibodies by enzyme-linked immunosorbent assay (ELISA) and/or LAC determined according to the International Society of Thrombosis and Haemostasis) on two or more occasions not less than 12 weeks apart, as stipulated in the Sydney revised Sapporo guidelines [14]. Vascular thrombosis was defined as one or more clinical episodes of arterial, venous, or small-vessel thrombosis in any tissue or organ, confirmed by objective validated criteria (unequivocal findings on appropriate imaging studies or histopathology). Pregnancy morbidity was defined as follows: one or more unexplained deaths of a morphologically normal fetus at or beyond week 10 of gestation, with normal fetal morphology documented by ultrasound or by direct examination of the fetus (intrauterine fetal death (IUFD)); or one or more premature births of a morphologically normal neonate before the 34th week of gestation because of eclampsia or severe preeclampsia defined according to standard definitions or because of recognized features of placental insufficiency (intrauterine growth restriction); or three or more unexplained consecutive spontaneous abortions before the 10th week of gestation, excluding maternal anatomic or hormonal abnormalities and paternal and maternal chromosomal causes. For the classification of PAPS, APLA positivity was defined as a moderate-to-high titer of aCL IgG/IgM antibody (> 40 GPL/MPL U/ml) and/or anti-β₂GPI IgG/IgM above the 99th percentile (IgG/IgM aCL/anti-β₂GPI ≥ 20 GPL/MPL U/ml) and/or positive LAC by either dilute Russell viper venom time test (> 1.4) and/or silica clotting time test (> 1.3). Patients with systemic autoimmune or rheumatic diseases, including but not limited to systemic lupus erythematosus, rheumatoid arthritis, and Sjögren’s syndrome, were excluded from the study as were patients with evidence of infectious disease within 3 months prior to blood sampling and/or evidence of active malignancy, in addition to pregnant women, women within 12 weeks after delivery or abortion, and patients with thrombosis attributed to a non-APLA cause. Asymptomatic APLA carriers were defined as consecutive patients with persistently positive aCL IgG and/or IgM and/or anti-β₂GPI IgG and/or IgM and/or positive LAC test on two consecutive occasions at least 12 weeks apart with no evidence of thrombosis and/or obstetric features of APS and/or inflammatory or autoimmune systemic disorder, infection, malignancy and/or pregnancy or 12 weeks or less postpartum. Healthy control subjects were recruited from the hospital staff. Exclusion criteria for the control group were history of thrombosis, systemic autoimmune or inflammatory disease, current infection, current or previous malignancy, and/or pregnancy or 12 weeks or less postpartum.

At the study encounter, participating patients were interviewed, and demographic and clinical information was systematically collected, including current age, sex, time of diagnosis, and thrombotic and/or obstetric manifestations of PAPS. Medications used at the time of the study or within the previous 3 months were recorded, specifically low-dose aspirin (75–100 mg/day), hydroxychloroquine (200–400 mg/day), warfarin (Coumadin), new oral anticoagulants, prednisone, and statins. In addition, the individual electronic charts and the hospital’s computerized database were systematically reviewed for the following comorbidities: arterial hypertension, defined as systolic blood pressure higher than 140 mmHg and diastolic pressure higher than 90 mmHg that required antihypertensive medications; diabetes mellitus, defined as hyperglycemia requiring oral drugs and/or insulin; dyslipidemia, defined as serum total cholesterol above 200 mg/dl and/or triglycerides above 150 mg/dl and/or anti-lipidemic drug use; ischemic heart disease, defined as evidence of myocardial infarction and/or evidence of coronary artery atherosclerosis obtained by either coronary catheterization or a percutaneous coronary intervention procedure; and stroke defined by evidence of neurological deficit and/or computed tomography or magnetic resonance imaging signs of brain infarct. Estimated glomerular filtration rate (eGFR) at the time of the study encounter was calculated using the Modification of Diet in Renal Disease equation [15].

Blood was collected from all patients at the study encounter and assayed in the hospital’s routine clinical laboratories for complete blood count, erythrocyte sedimentation rate (ESR), and levels of serum high-sensitive C-reactive protein (hsCRP), ferritin, and creatinine.

The APLA results used for the purpose of the study were serum aCL and anti-β₂GPI IgA/IgM antibodies and LAC. All factors were tested routinely at the hospital’s laboratory at the most recent clinic visit prior to the study or at the study encounter. A commercial ELISA kit was used according to the manufacturer’s instructions to analyze aCL IgG and IgM antibodies (Aesku Diagnostics GmbH&Co. KG, Wendelsheim, Germany) and anti-β₂GPI IgG and IgM antibodies (Inova Diagnostics, Quanta Lite, San Diego, CA, USA), and LAC was assayed according to published guidelines [16].

All blood samples were centrifuged at 1500 g for 15 min immediately after venous blood withdrawal and stored at −70 °C until assayed. The plasma level of sTREM-1 was analyzed using a commercial ELISA kit (Human TREM-1 DuoSet ELISA kit, DY1278B, Bio-Techne, Minneapolis, MN, USA) according to the manufacturer’s instructions [17]. The assay plates (CI9018, Corning Inc., Corning, NY, USA) were coated with the capture antibody overnight at 4 °C and then washed with 0.05% Tween®20 in phosphate-buffered saline (PBS) (PeproTech Inc., Rocky Hill, NJ, USA) and blocked with the assay reagent diluent (5% Tween®20 in PBS) for 2 h at room temperature, and then washed. Next, 100 μl of serially diluted standards (3000 ± 23.45 pg/ml), patient, and control plasma were added and incubated overnight at 4 °C. On the following day, the plates were equilibrated at room temperature for 2 h and then washed and incubated with a detection antibody diluted with the reaction diluent containing 2% heat-inactivated normal goat serum for 2 h at room temperature. Next, the plates were washed, and incubated for 20 min at room temperature with horseradish peroxidase-conjugated streptavidin diluted with the reaction diluent (1:200). The plates were washed and incubated with TMB substrate (PeproTech Inc., Rocky Hill, NJ, USA) for 20 min and the reaction was stopped with 1 N HCl (PeproTech Inc., Rocky Hill, NJ, USA). The absorbance was measured by a spectrophotometer (BioTek, Winooski, VT, USA) set at 450 nm with the correction wavelength set at 570 nm. The results were analyzed by creating a four-parameter logistic (4-PL) curve against the optical density of the standards (3000 ± 23.45 pg/ml) following the subtraction of the optical density of the averaged blank controls. All plasma samples and standards were assayed in duplicates. The difference between ELISA plates was monitored by including specific healthy control plasma samples in each assay.

The statistical analysis was generated using SAS Software, version 9.4. Continuous variables are presented as mean and standard deviation, and categorical variables as number and percentage. Values of continuous variables were compared among three groups using a general linear model (GLM) with Tukey adjustment for multiple comparisons, and between two groups using t tests; categorical values were compared using the chi-square test. Correlations (r) were calculated by Pearson correlation. Logistic regression was used for receiver operating characteristic (ROC) curve analysis. Two-sided p values less than 0.05 were considered statistically significant.

Plasma samples of 116 consecutively enrolled subjects were collected and analyzed for levels of sTREM-1. The study population consisted of 33 patients with PAPS (mean age 47.8 years, range 19–88; 77.7% women), 10 asymptomatic persistent positive APLA carriers (mean age 50.6 years, range 28–75; 90% women), and 73 healthy control subjects (42.65 years, range 18–67; 54.8% women). The demographic, clinical, and laboratory features of the three groups are shown in Table 1. Within the PAPS group, 26 patients (78.78%) had previous thrombotic (n = 19) or obstetric (n = 7) events (past PAPS). Seven patients (21.21%) were evaluated for plasma sTREM-1 level at the time of an acute thrombotic event (current thrombotic PAPS). The prevalence rates of the various types of past thrombotic events in the thrombotic PAPS group (n = 26) were as follows: 18 patients (69.2%) had an arterial thrombotic event, 4 patients (15.4%) had a venous thromboembolic event, and 4 patients (15.4%) had arterial and venous thromboembolic events.

Analysis of plasma sTREM-1 levels by group yielded a significantly higher level in the patients with PAPS (299.2 ± 146.7 pg/ml) compared with the healthy control group (230.2 ± 85.5 pg/ml, p = 0.006), in the patients with thrombotic PAPS-ever (327.2 ± 151.3 pg/ml) compared with the control group (p = 0.0003, Fig. 1), in the patients with thrombotic PAPS compared with the patients with obstetric PAPS-past event (195.1 ± 58.5 pg/ml, p = 0.01), and in the patients with thrombotic PAPS compared with the asymptomatic APLA carriers (215.8 ± 51.6 pg/ml, p = 0.02, Fig. 2).

The plasma sTREM-1 level was positively correlated with antinuclear antibody titer (r = 0.33, p = 0.03). However, there were no statistically significant differences between patients with single, double, or triple APLA positivity (PAPS and asymptomatic groups). Specific values were as follows: single-positive (n = 11), mean 290.2 ± 113.5 pg/ml, median 283.3 pg/ml; double-positive (n = 13), mean 267.01 ± 92.6 pg/ml, median 253.7 pg/ml; triple-positive (n = 19), mean 282.5 ± 171.7 pg/ml, median 247.06 pg/ml. Plasma sTREM-1 levels were not significantly correlated with aCL or anti-β₂GPI antibody titers.

Within the PAPS group, higher plasma sTREM-1 levels were positively associated with current thrombotic events (429.5 ± 227.5 pg/ml) compared with past thrombotic events (289.5 ± 94.65 pg/ml, p = 0.01) (Fig. 3) as well as to past obstetric events (195.1 ± 58.5 pg/ml, p = 0.0001). The association remained significant on comparison of the patients with thrombotic PAPS-ever with the asymptomatic APLA carrier group (215.8 ± 51.6 pg/ml, p = 0.0002) and the healthy controls (230.2 ± 85.5 pg/ml, p < 0.0001). In the patients with obstetric PAPS, there was a positive correlation of plasma sTREM-1 level with number of past pregnancies and fetal loss, but it did not reach statistical significance (r = 0.2, p = 0.06).

The plasma sTREM-1 level was significantly higher compared with controls in the patients with stroke-ever (307.2 ± 105.6 pg/ml, p = 0.03), myocardial infarction-ever (371.6 ± 250.04 pg/ml, p = 0.009), and venous thromboembolic event-ever (366.67 ± 203.79 pg/ml, p = 0.0012) (Table 2). Among the patients with PAPS, the plasma sTREM-1 level was significantly higher in those who had stroke-ever than in those who had not (307.2 ± 105.6 pg/ml vs. 240.5 ± 107.2 pg/ml, p = 0.007), and in those who had a venous thromboembolic event than in those who had not (366.7 ± 203.8 pg/ml vs. 239.8 ± 94.2 pg/ml, p = 0.02). There were no significant findings for myocardial infarction.

To test whether plasma sTREM-1 level was associated with inflammatory activity in the patients with PAPS, we performed a correlational analysis of plasma sTREM-1 level with ESR and serum hsCRP levels, platelet count, and serum ferritin level, all of which are clinical biomarkers of the inflammatory acute phase response. We found a significant positive correlation for plasma sTREM-1 level with higher ESR (r = 0.4, p = 0.009) and higher hsCRP level (r = 0.4, p = 0.02) (Fig. 4).

No significant positive relationship was found between plasma sTREM-1 level and the presence of comorbidities (hypertension, diabetes mellitus, dyslipidemia, or history of a malignant disorder) or current use of anticoagulants (warfarin or new oral anticoagulants), or low-dose aspirin, hydroxychloroquine, prednisone, or statins.

The plasma sTREM-1 level was significantly associated with current age in the healthy control group (r = 0.63, p < 0.0001) but not in the PAPS or asymptomatic APLA carrier groups. Since renal function based on GFR physiologically decreases with age [18], we assumed that the age-related increase in plasma sTREM-1 level in the healthy subjects was associated with decreased renal clearance of sTREM-1. Indeed, we found a negative association for plasma sTREM-1 level and eGFR (r = −0.2, p = 0.06), but it was statistically significant only in the control group (r = −0.6, p < 0.0001). Thus, the correlation observed between plasma sTREM-1 and renal function in the control group is unrelated to levels in the patients with PAPS.

A ROC analysis was performed to evaluate the value of plasma sTREM-1 in discriminating patients with current thrombosis from other patients with PAPS and from asymptomatic APLA carriers and healthy controls. The results showed that a cutoff plasma sTREM-1 level of 281 pg/ml had a sensitivity of 65.4% and specificity of 100% for thrombotic events-ever in the PAPS group. A cutoff of 284 pg/ml had a sensitivity of 57.1% and specificity of 100% for current thrombotic PAPS. In the subgroup of thrombotic PAPS-ever, the area under the curve (AUC) was 0.73 for plasma sTREM-1 (95% confidence interval (CI) 1.003–1.013, p = 0.0014; Fig. 5). No significance was found for ESR or serum ferritin level. In the subgroup of current thrombotic PAPS (n = 7), the AUC for plasma sTREM-1 was 0.88 (95% CI 0.686–0.977, p < 0.0001; Fig. 6), similar to that for the inflammatory biomarkers serum hsCRP (AUC = 0.96, 95% CI 0.788–0.999, p < 0.0001) and serum ferritin (AUC = 0.801, 95% CI 0.571–0.941, p = 0.004).