Association between P-Selectin Autoantibody Positive and Response to Steroid Treatment in Newly Diagnosed Immune Thrombocytopenia Patients

Immune thrombocytopenia (ITP) is characterized by isolated thrombocytopenia (platelet count <100 × 10⁹/L) and mucocutaneous bleeding that affects patients of all ages [1, 2]. The pathophysiology of ITP is not completely understood. It has been reported that platelet glycoprotein-specific antibody was important for the pathogenesis of ITP [3]. The autoantibodies could bind to the circulating platelets, resulting in the reticuloendothelial system to destruct the platelet [3]. These platelet autoantibodies include anti-glycoprotein (GP) IIb/IIIa, GP Ib/IX, and P-selectin.

P-selectin or granular membrane protein 140 (GMP140) is an adhesion glycoprotein which expresses on the platelet surface to promote platelet-leukocyte adherence [4]. P-selectin participates in the thrombosis and prothrombotic states [5]. Autoantibodies against P-selectin could affect the functions of platelets and endothelial cells [5]. Hu et al. [6] measured P-selectin antibodies in 32 patients with primary Sjögren’s syndrome and found that P-selectin autoantibody positive was correlated with a low platelet count [5]. 250 ITP patients, 163 non-ITP patients, and 243 healthy controls were enrolled in Zhai’s study [6]. They found that the levels of P-selectin autoantibodies were evaluated in ITP patients as comparing to non-ITP patients [6].

Steroids and intravenous immunoglobulin (IVIG) are first-line treatments for ITP [7]. According to the 2019 edition of ASH guidelines, steroids should be used first in newly diagnosed ITP patients whose peripheral platelet count is less than 30 × 10⁹/L, asymptomatic, or with a small amount of cutaneous and mucosal spontaneous bleeding. Wei et al. [8] found that patients with high dose of dexamethasone had higher overall response and shorter reaction time. IVIG is an effective treatment for rapid increase of platelets. It is usually used in emergency treatment of patients with severe hemorrhage and preoperative preparation of patients undergoing emergency surgery. IVIG has the potential to cause serious side effects, including renal failure and thrombosis [9], and delays the initiation of other effective treatments [10].

Previous studies have suggested that the detection of platelet autoantibodies is not only helpful in diagnosing ITP but also in monitoring ITP patient responses to treatment [6, 11]. Anti-GP Ib/IX-positive ITP patients were found to be less responsive to IVIG treatment [12]. And anti-GPIbα-positive ITP patients had low steroid response [11]. Since there are few studies reporting P-selectin autoantibody-positive ITP patients receiving steroids, it is unclear whether P-selectin autoantibody expression would affect patients’ treatment response.

Thus, in this study, we retrospectively collected data from newly diagnosed ITP patients who only received first-line therapy with steroids. We analyzed the association between P-selectin autoantibody positive and response to steroid treatment.

This was a retrospectively cohort study. Adult newly diagnosed ITP patients who were hospitalized in the hematology units of our hospital from October 2016 to May 2021 were included. The study was approved by Institutional Review Board of our hospital. All patients enrolled in this study signed the informed consents.

The inclusion criteria were shown as follows: (1) patients who were aged 18 years or older; (2) patients who were diagnosed as primary ITP according to the International Working Group (IWG) diagnostic criteria [10]; (3) patients who had not received any therapy before. Patients with secondary, relapsed, or refractory ITP were excluded.

The data including age, gender, height, weight, treatment protocol, ITP-specific bleeding score, adverse reactions, platelet glycoprotein-specific antibodies expressions, and treatment responses were retrospectively collected. The routine blood count including platelet count was measured by a hematology analyzer (Mindray Biomedical Electronics Co., Ltd., Shenzhen, China).

The modified monoclonal antibody-specific immobilization of platelet antigens (MAIPA), carried out as previously described [11, 13], was used to detect the specific antiplatelet GP IIb/IIIa, GP Ib/IX autoantibodies. Venous blood (5 mL) was collected for anticoagulation with EDTA and centrifuged at 1,800 rpm for 10 min. Platelet-rich plasma was absorbed into a 10-mL tube and centrifuged at 3,800 rpm for 5 min. The supernatant was discarded and platelets were washed three times with phosphate buffered saline (PBS)/EDTA. Platelet precipitates were suspended in PBS/EDTA and the concentration was adjusted to about 200 × 10⁹/L. 100-μL platelet suspension was incubated with 50-μL plasma at 37°C for 60 min. Platelets were lysed with 130-mL lysis buffer (1% NP40, 150-mM NaCl, 50-mM Tris-HCL, pH 8.0, 2-mM EDTA, 0.1-mM phenylmethanesulfonylfluoride [PMSF]) at 4°C for 30 min, centrifuged at 14,000 rpm for 20 min, and diluted 1:2 for further testing. GP IIb–IIIa-specific antibody mAb P2 IgG (Immunotech S.A., Marseille, France) and GP Ib-specific antibody mAb SZ2 IgG (Jiangsu Institute of Hematology, Suzhou, China) were added to a microtiter plate coated with affinity-purified goat anti-mouse IgG (Immunotech S.A.) and incubated for 60 min. The microplate was washed and then 100 μL of diluted platelet lysate was added to each well and incubated at 4°C for 90 min. Alkaline phosphatase-conjugated goat anti-human IgG and p-nitrophenyl phosphate was used for color development. The optical density value was read at 405 nm, and each sample was repeated at least three times. Plasma optical density values greater than normal plus 3 standard deviations were positive.

Flow cytometry immunobead array (FCIA) is a type of antibody-coated polystyrene microbeads that uses flow cytometry to detect binding to specific antigens. In this study, we used an FCIA assay to detect P-selectin autoantibody in ITP patients. Venous blood (2 mL) was collected for anticoagulation with EDTA and centrifuged at 300 g for 5 min at room temperature. After centrifugation, platelet-poor plasma was collected and transferred into a new tube. As in MAIPA, soluble platelet lysates were collected for FCIA determination. Monoclonal antibody against human platelet P-selectin (SZ51) was prepared, as described previously [14-16]. Fluorescein isothiocyanate (FITC)-labeled goat anti-human IgG (FITC-GAH) and anti-mouse IgG (FITC-GAM) polyclonal antibodies were purchased from Beckman-Coulter (Suzhou, China). Polystyrene microbead (4 μm in diameter with different fluorescent intensities) was from Spherotech (Lake Forest, IL). Monoclonal antibody SZ51 (160 μg each) was prepared in a carbonate coating buffer (pH 9.5) and incubated with 1 × 10⁶ microbead of distinguishable fluorescent intensities on a rotator at 4°C. After 12 h, antibody-coated microbead was washed three times with PBS in the presence of 0.05% Tween-20 and then stored in PBS containing 0.02% sodium azide at 4°C. The stability of the stored bead was tested each month for 10 months. Microbead (1 × 10⁴) coated with monoclonal antibody was added to platelet lysate (150 μL) from ITP patients and incubated at room temperature for 60 min. Samples were centrifuged at 500 g for 20 min and incubated with an FITC-GAH antibody at room temperature for 30 min. The microbeads were washed once with PBS, suspended in 0.5 mL of PBS, and analyzed by flow cytometry using an FITC-GAH antibody (Cytomics FC-500, Beckman-Coulter). Fluorescent intensity from 1,500 to 2,000 microbeads was analyzed by the CXP software (Beckman-Coulter) to calculate mean fluorescent intensity values for individual samples.

All patients with ITP were only treated with steroids after the diagnosis. The patients were first treated with high-dose dexamethasone (HD-DXM, 40 mg/day, for 4 consecutive days) intravenously, followed by oral prednisone (1 mg/kg/day) for maintenance. Treatment responses were evaluated as complete response (CR), response (R), and no response (NR) within 3 months after the onset of treatment. CR was defined as the platelet count ≥100 × 10⁹/L without bleeding manifestations; R was defined as the platelet count ≥30 × 10⁹/L or a twofold increase of baseline platelet count without bleeding; patients who did not meet the above criteria during steroid treatments and were treated with other therapies were considered NR in this study. The overall response is the sum of CR and R. Time to response (TTR) was the time from the onset of treatment to response.

The data were analyzed by SPSS 22.0. Normal distribution quantitative data were described as mean ± standard deviations. The differences between groups were compared by the Student’s t test. Non-normal distribution quantitative data were described as median with interquartile range and compared with Mann-Whitney U test. Categorical data were described as numbers and percentages and compared using the χ² test. The effect of variables on treatment response was analyzed by logistic regression analysis. p < 0.05 indicated statistical significance.

Among patients attending Hulunbeir People’s Hospital hematology clinic, 105 newly diagnosed ITP patients were screened for eligibility for our study. Included patients had a median age of 56 years ranging from 26 to 77 years and comprised 49 males (46.67%) and 56 females (53.33%). The median platelet count before treatment was 11 × 10⁹/L (ranged 0–30 × 10⁹/L). In our whole population, we observed that 18 (17.14%) patients had only GP IIb/IIIa-positive antibodies; 7 (6.67%) patients had anti-GP Ib/IX positive only; 7 (6.67%) patients had anti-P-selectin positive only; 12 (11.43%) patients had anti-GP IIb/IIIa and anti-GP Ib/IX positive; 11 (10.48%) patients had anti-GP Ib/IX and anti-P-selectin positive; 17 (16.19%) patients had anti-GP IIb/IIIa and anti-P-selectin positive; 12 (11.43%) patients had all three antibodies positive; and 21 (20.00%) patients did not detect any antibody. The patient characteristics of these 105 patients are shown in Table 1.

Of the 105 eligible patients, 80% (84/105) of the patients presented with platelet glycoprotein-specific antibody positive; 44.76% (47/105) patients with anti-P-selectin positive were included in anti-P-selectin-positive group, while 35.24% (37/105) with anti-P-selectin negative were included in anti-P-selectin-negative group. The median platelet count before treatment was 10 × 10⁹/L (ranged 0–27 × 10⁹/L) in patients with anti-P-selectin positive and 11.5 × 10⁹/L (ranged 1–30 × 10⁹/L) in patients with anti-P-selectin negative. The median platelet count between the two groups of patients was not significantly different (z = −1.314, p = 0.189) (Fig. 1). Among the 105 ITP patients, 57 (54.29%) patients obtained CR, 22 (20.95%) patients obtained R, while 26 (24.76%) patients did not respond to the treatment. The overall response rate was 75.24%. Among patients who were anti-P-selectin positive, only 48.94% (23/47) of patients obtained CR and 25.53% (12/47) obtained R. No significant difference in the overall response was observed between patients who were anti-P-selectin positive and those who were anti-P-selectin negative (74.47% vs. 89.19, χ² = 2.910, p = 0.088). But patients who were anti-P-selectin negative had significantly higher CR rate, compared to those who were anti-P-selectin positive (72.97% vs. 48.94%, χ² = 4.965, p = 0.026) (Table 2). The median TTR was 6 days (ranged 1–16 days) in anti-P-selectin-negative group and 6.5 days (ranged 2–24 days) in anti-P-selectin-positive group. There was no statistical difference in TTR between the two groups (p = 0.308).

Univariate and multivariate logistic regression analysis revealed that gender, age, and anti-GP Ib/IX positive were not associated with ITP patients’ response (Table 3). Anti-GP IIb/IIIa positive and anti-P-selectin positive were two factors that could affect patients’ response. The multivariate logistic regression analysis showed that the odds ratio (OR) of response for anti-GP IIb/IIIa-positive to anti-GP IIb/IIIa-negative patients was 3.114 (p = 0.010, 95% confidence interval: 1.313–7.388, Table 3), indicating that anti-GP IIb/IIIa-positive patients were prone to better treatment response. However, the OR for anti-P-selectin-positive to anti-P-selectin-negative patients was 0.309 (p = 0.036, 95% confidence interval: 0.127–0.753, Table 3). This result indicated that anti-P-selectin-positive patients were less responsive to steroid therapy.

In this study, we found that newly diagnosed ITP patients who were platelet glycoprotein-specific antibody positive and expressed anti-P-selectin were less responsive to steroid therapy than those who did not express anti-P-selectin. Thus, detection of anti-P-selectin antibodies may be a potential method to identify patients who were less responsive to steroid treatment.

Although many new therapeutic interventions have been developed, the cornerstone therapy of ITP remains to be steroids [13]. In the present study, 105 newly diagnosed ITP patients were included to receive steroid therapy (HD-DXM with prednisone maintenance). Compared to the commonly used steroid treatment protocols (HD-DXM or long-term conventional-dose prednisone), HD-DXM with prednisone maintenance prolonged the exposure time of steroids, obtaining more sufficient immunosuppression for ITP [17]. Among the 105 enrolled patients, 80.00% of patients were platelet glycoprotein-specific antibody positive. The overall response was higher in patients who were antibody positive than those who were antibody negative. The result was consistent with other studies [11].

Published literature works have shown that platelet glycoprotein-specific antibodies were important for dictating response to steroid therapy in ITP. For example, Liu et al. [18] found that ITP patients with anti-GP IIb/IIIa had good response to steroids. Go et al. [12] found that ITP patients with anti-GP Ib/IX may be less responsive to IVIG. The study conducted by Zeng et al. [11] also found that patients with anti- GPIIb/IIIa antibodies had higher steroid response. However, there are few studies reporting the anti-P-selectin autoantibody in ITP patients with steroid treatment.

P-selectin is a platelet activation marker which mediates the accumulation of platelets [19]. Anti-P-selectin autoantibodies were reported to lead to platelet destruction and impair the platelet function [5]. Turner and Hadley [20] found that immune destruction of platelets could be enhanced by anti-P-selectin on the platelet membrane. Thus, anti-P-selectin autoantibodies were found to be associated with low platelet counts [5]. In this study, we also found that the median platelet count was lower in patients who were anti-P-selectin positive than that in patients who were anti-P-selectin negative. But the difference between the two groups was not significantly different (p = 0.189). A lack of enough sample size may contribute to this result. For response evaluation, patients who were anti-P-selectin positive had much lower CR rate, compared to those who were anti-P-selectin negative (p = 0.026), indicating anti-P-selectin-positive patients had worse response. Our logistic regression analysis further confirmed this result. The proportion of patients who were responsive to steroid treatment was 3–4 times higher in the anti-P-selectin-negative group than in the anti-P-selectin-positive group (the OR for anti-P-selectin-positive to anti-P-selectin-negative patients was 0.309). Thus, anti-P-selectin-mediated-ITP may be less responsive to steroid treatment.

A previous study found that patients with anti-GP Ib/IX autoantibodies presented worse response to ITP treatment [11]. However, in the present study, our logistic regression analysis found that anti-GP Ib/IX was not associated with ITP patients’ response. In Zeng’s study, they did not consider the expression of anti-P-selectin [11]. They only classified ITP patients into anti-GP Ib/IX and anti-GPIIb/IIIa groups [11]. But our study found that about 44.76% of patients were anti-P-selectin positive. The expression of anti-P-selectin was considered in the analysis. Thus, we obtained different results from Zeng’s study. Furthermore, our results are consistent with previous reports [17, 18] showing that anti-GP IIb/IIIa-positive patients were prone to better treatment response. The OR for anti-GP IIb/IIIa-positive to anti-GP IIb/IIIa-negative patients was as high as 3.114. This result indicated that antibody-positive patients achieved high response rate and this mainly depended on patients who were anti-GP IIb/IIIa positive.

This study has several limitations. First, the sample size of this study was very small. Second, we did not detect the mechanism by which anti-P-selectin-mediated ITP was less responsive to steroid treatment. Finally, the observation period was short; the long-term efficacy was not detected. Thus, further confirmatory studies on a larger number of patients are really needed.

In conclusion, our study found anti-GP IIb/IIIa positive and anti-P-selectin positive were two factors that could affect patients’ response. The results indicate that anti-GP Ib/IX antibodies are not associated with ITP patients’ response. Anti-P-selectin-positive patients have a lower response to steroid treatment, while anti-GP IIb/IIIa-positive patients tend to have a better response to the treatment. Thus, detection of anti-P-selectin antibodies may be a potential method to identify patients who are less responsive to steroid therapy.

This study was conducted in accordance with the declaration of Helsinki. This study was conducted with approval from the Ethics Committee of Hulunbeir people’s Hospital (No.:2021SYY-04). Written informed consent was obtained from all participants.

The authors declare that they have no competing interests.

Academician of Hematology (Ruan Changgeng) Expert Workstation of Hulunbeir People’s Hospital.

1. Conception and design: L.Wang.

2. Administrative support: D.-D. Wang and R.-Y. Jiao.

3. Provision of study materials or patients: C.-X. Liu and Y.-Q. Hou.

4. Collection and assembly of data: H. Qin and H.-J. He.

5. Data analysis and interpretation: Y.-Q. Hou, C.-X. Liu, and R.-Y. Jiao.

6. Manuscript writing: all authors.

7. Final approval of manuscript: all authors.

All data generated or analyzed during this study are included in this published article.

Li Wang and Dong-Dong Wang contributed equally to this study.