Leendert Porcelijn’s research while affiliated with Sanquin Blood Supply Foundation and other places

Background and Objectives Foetal and neonatal alloimmune thrombocytopenia (FNAIT) results from maternal platelet‐directed antibodies and can result in severe intracranial haemorrhage (ICH) in foetuses and newborns. Screening for human platelet antigen‐1a (HPA‐1a)‐directed antibodies during pregnancy could allow timely intervention with antenatal treatment and prevent ICH. We assessed the cost effectiveness of HPA‐1a typing and anti‐HPA‐1a‐screening as part of the prenatal screening programme. Materials and Methods Different HPA‐1a screening scenarios were tested in a decision analysis model and assessed for diagnostic, treatment, intervention and lifetime costs and prevention effects compared to the current situation without screening in the Netherlands. Model parameters were based on available data, literature and expert opinions. One‐way sensitivity analysis and probabilistic sensitivity analysis were performed. Results Adding screening for anti‐HPA‐1a antibodies to the current antenatal screening programme of the Netherlands will lead to an additional cost of €4.7 million per year and a gain of 226 quality‐adjusted life years (QALYs) per year, indicating an incremental cost–effectiveness ratio (ICER) of €20,782 per QALY gained. One‐way sensitivity analysis showed that the uncertainty around the incidence of ICH, lifetime costs of disabled children and the probability of having antibody quantitation >3.0 IU/mL at 20 weeks had the highest effect on the ICER. Conclusion Antenatal anti‐HPA‐1a screening might be cost effective. To obtain more knowledge and thereby to improve risk stratification, a pilot screening programme is warranted.

Introduction The spleen plays a critical role in immune thrombocytopenia (ITP), because it is a major site of platelet destruction in many patients and the major secondary organ of autoantibody production, with expansion of germinal centers (GC). After the age of 60, the incidence of ITP peaks. Recent evidence shows striking shifts in B cell subsets in the elderly, suggesting an important role in the development of age-related immune dysfunction. We aimed to investigate the distribution of B cells in patients with ITP compared to healthy controls (HC) of the same age range. Methods We included 40 patients with primary ITP ≤ 65 years old (yo), 62% female, 58.5±12 yo; platelet count 102 (5-502) platelets/µL; and 40 patients >65 yo, 65% female, 76.5±7 yo, platelet count 117(3-556) platelets/µL; who had not received prior immunosuppressive treatment or B-cell depletion therapy. Sixty HC ≤ 65 yo, 51% female, 54±15 yo, 245 (105-431) platelets/µL; and 40 >65 yo, 57% female, 78±9 yo, 244 (124-663) platelets/µL were included. Anti-platelet antibodies (anti (a)-GPIIb/IIIa, a-GPV and a-GPIb/IX) were measured by MAIPA. We characterised the pre-CG compartment of B cells, immature/transitional B-cells [including 3 maturation-associated populations of CD5- CD38++ CD21het CD24 ++ (CD5+ CD38+/++ CD21het CD24++, and CD5+ CD38het CD21+ CD24+ immature B-cells)]; 3 mature naïve B cells (CD21+ CD24+, CD21- CD24++, and CD21- CD24- subsets), together with unswitched IgMD+ memory B cells, switched memory B cells and plasmablasts/plasma cells] using standardized EuroFlow protocols and Infinicyt software. B cells activating factor (BAFF) was measured in plasma with an ELISA kit (R&D, Spain). Statistical analyses were performed using Graph Pad Prism 7.0 and two-way Kruskal-Wallis was used for multiple analyses. Results We observed significant differences between HC groups. Compared to HC ≤65 yo, HC >65 yo had decreased immature B cell subsets (cells/µL, HC≤65: 34.4±19.7; HC>65: 8.9±10.1, p<0.001); in naïve B cell subsets (cells/µL, HC≤65: 93.6±53.8; HC>65: 30.2±47.4, p<0.001) and memory B cells (cells/µL, HC≤65: 76.8±39.1; HC>65: 31.5±21.0, p<0.001). A full analysis of the pre-CG LB compartment showed distinct patterns for the subtypes of immature B cells when comparing ITP patients ≤65 yo with age-matched HC: there was no difference in the amount of CD5-CD38++CD21hetCD24+++ but reduced CD5+CD38+/++CD21het CD24++ (cells/µL, HC≤65:8.4±7.6; ITP≤65: 4.4±4.2, p<0.01) and CD5+CD38+CD21+CD24+/++ (cells/µL, HC≤65:24.8±21.5; ITP≤65: 11.7±9.8, p<0.001). A different distribution of naïve mature B cells was also observed: a decreased CD21+CD24+ compartment (cells/µL, HC≤65:92.2±56.1; ITP≤65: 55.3±47.8, p<0.005) and an increase in CD21-CD24- (cells/µL, HC≤65:1.0±0.9; ITP≤65: 2.5±2.0, p<0.001), while the number of CD21-CD24++ did not differ. For post-CG B cells, ITP patients ≤65 yo had significant reductions in memory LBs (cells/µL, HC≤65:76.8±39.0; ITP≤65: 44.4±33.6, p<0.005), with no differences in plasma cells. There were no differences between HC>65 and ITP patients >65 in distribution of immature and in post-CG B cells, but a significant increase in CD21-CD24- (cells/µL, HC>65: 1.7±1.3; ITP>65: 2.6±1.9, p<0.01) and CD21-CD24++ (cells/µL, HC>65: 0.1±0.1; ITP>65: 0.4±0.4, p<0.05) in naïve mature B cells. BAFF, essential to promote B cells maturation, was increased in both ITP groups (pg/mL; HC ≤65 yo: 405.4±78.8, ITP≤65 yo: 674.3±356.9, p<0.001; HC >65 yo: 500.4±193.3, ITP>65 yo: 1100.1±675.0, p<0.001); and no difference between both ITP groups. Anti-platelet antibodies were detected in 44.0% of patients with ITP ≤65 yo (4.9% a-GPII/IIIa, 4.9% a-GPV, 4.9% a-GPIb/IX; 22.0% with 2 and 7.2% with 3 antibodies); and in 33% of those >65 yo (3.2% a-GPII/IIIa, 6.6% a-GPV, 10.0% a-GPIb/IX; 3.2% with 2 and 10.0% with 3 antibodies). Conclusion Age modifies the distribution of B-cells in HC, so it is important to compare immunological data of patients with those from an age-matched HC. Naïve CD21-CD24- were higher in both ITP groups than in their respective HCs. LB subset with low CD21 expression is also expanded in other autoimmune diseases (Saadoun et al, PMID: 23279883). Our results highlight the need to identify patients who are more likely to respond to B-cell targeted therapies in order to tailor their individual treatment options. Funded by ISCIII-European Union (PI22/01489), and by Beca Luis Álvarez 2022-IdiPAZ.

Immune thrombocytopenia (ITP) is an auto-immune bleeding disorder characterized by isolated thrombocytopenia due to an accelerated platelet destruction and impaired platelet production. Thrombopoietin receptor agonists (TPO-RAs, e.g. romiplostim) are recommended as a subsequent treatment option for patients who failed first-line therapy with corticosteroids. Long lasting use of TPO-RAs, however, is associated with high costs and a persisting burden for patients. Strikingly, sustained remissions after tapering of TPO-RAs have been reported in 10-50% of the patients with primary ITP. These findings suggest that TPO-RAs, besides stimulating platelet production, may also have an immune modulatory effect via a hitherto unknown mechanism. No distinct clinical factors or biomarkers have been identified that can reliably predict successful tapering of TPO-RAs. The STIP study is a prospective multi-center single-arm intervention trial in the Netherlands aimed to determine the rate of sustained remission off-treatment (SROT) after 1 year of romiplostim treatment. Thirty-nine adult patients with primary ITP lasting for more than 3 months and an indication for second-line treatment received romiplostim for 1 year. Subsequently, romiplostim was tapered within 6 weeks and patients were followed for 1 year. SROT was defined as platelet counts > 30 x 109/L, no bleeding symptoms and no need for treatment. Data about clinical parameters were gathered to identify potential predictors for SROT. Moreover, blood samples were taken at six different time points before, during and after romiplostim treatment. These samples were assessed for anti-glycoprotein (anti-GP) antibodies by monoclonal antibody immobilization of platelet antigen tests and for endogenous TPO levels by an enzyme-linked immunosorbent assay. Indium-111 scans were performed before and after +/- 1 year of romiplostim treatment to assess changes in platelet sequestration sites and clearance rate. Seventy-seven percent (30/39) of the included patients had chronic ITP and 41% (16/39) had received 2 or more treatment lines prior to inclusion. Twenty-nine patients completed the first year of treatment of which 86% (25/29) started the tapering phase. Censored survival analysis showed that 23.6% (95% CI: 11.0-50.5%) had a SROT at 1 year after tapering (n=5, uncensored: 20%). The median time to relapse was 58 days. Only mild bleeding symptoms (WHO scale ≤ 1) were observed in 41% (7/17) of the patients who relapsed. Five of the 17 patients who relapsed did not restart therapy. Hence, romiplostim tapering resulted in a treatment-free response of at least 12 months in 40% (10/25). Patients with SROT at 1 year had higher platelet levels during romiplostim treatment (median: 332.5 vs. 84.5 x109/L) and required lower doses at start tapering (median: 1.0 vs. 4.5 μg/kg) compared to patients who relapsed. Anti-GP antibodies were detected in 8/25 patients at baseline, of which 5 (2/3 patients with SROT and 3/5 with relapse) showed an substantial decrease (OD value ≥ 0.10) in antibody levels. We found that the endogenous TPO levels were similar for patients with SROT and relapse at the start of the tapering (median: 13 vs. 10.5 IE/ml). The indium-111 scan showed for 75.0% (3/4, missing: n= 1) of the patients with SROT compared to only 7.1% (1/14, missing: n= 3) of those who relapsed clearance rates within a normal range at the start of tapering (>62% circulatory platelet-associated radioactivity at 48hrs). The median difference in splenic: liver ratio between scans in patients who relapsed was 0.0 (range: -1.8 - 1.4, data evaluable in n= 13), while in those with SROT a median increase of 0.6 was observed (range: 0.1 - 1.9, n= 5). The STIP study demonstrates a sustained remission rate after romiplostim treatment of 23.6% in a diverse and largely unselected group of patients with ITP. Moreover, we have observed that tapering romiplostim using a standardized protocol is relatively safe, if closely monitored. We observed that patients with SROT tended to use lower romiplostim doses during treatment resulting in higher platelet counts. Normalization of the platelet clearance rate before tapering may be indicative of reaching remission in individual cases. Although more research into the underlying mechanisms leading to immune modulation is needed, these observations can be useful to identify patients who may benefit from TPO-RA treatment and subsequent tapering.

Vaccine-induced thrombotic thrombocytopenia (VITT) is a rare but severe complication following COVID-19 vaccination, marked by thrombocytopenia and thrombosis. Analogous to heparin-induced thrombocytopenia (HIT), VITT shares similarities in anti-platelet factor 4 (PF4) IgG-mediated platelet activation via the FcγRIIa. To investigate the involvement of platelet-antibodies in VITT, we analyzed the presence of platelet-antibodies directed against glycoproteins (GP)IIb/IIIa, GPV and GPIb/IX in the serum of 232 clinically suspected VITT patients determined based on (suspicion of) occurrence of thrombocytopenia and/or thrombosis in relation to COVID-19 vaccination. We found that 19% of clinically suspected VITT patients tested positive for anti-platelet GPs: 39%, 32% and 86% patients tested positive for GPIIb/IIIa, GPV and GPIb/IX, respectively. No HIT-like VITT patients (with thrombocytopenia and thrombosis) tested positive for platelet-antibodies. Therefore, it seems unlikely that platelet-antibodies play a role in HIT-like anti-PF4-mediated VITT. Platelet-antibodies were predominantly associated with the occurrence of thrombocytopenia. We found no association between the type of vaccination (adenoviral vector vaccine versus mRNA vaccine) or different vaccines (ChAdOx1 nCoV-19, Ad26.COV2.S, mRNA-1273, BTN162b2) and the development of platelet-antibodies. It is essential to conduct more research on the pathophysiology of VITT, to improve diagnostic approaches and identify preventive and therapeutic strategies.

Background Glanzmann thrombasthenia (GT) and Bernard‐Soulier syndrome (BSS) patients require frequent platelet transfusions and hence have an increased risk for alloimmunization against donor Human Leukocyte Antigens (HLA) when no HLA‐matching is performed. Knowing that Human Platelet Antigens (HPA) are located on the platelet glycoproteins that can be absent in these patients, preventive HPA‐matching may also be considered. Uniform recommendations on this topic lack in transfusion guidelines making standard practice unclear, therefore, we aimed to provide a framework for matched platelet transfusions. Study Design and Methods We conducted a targeted literature search and a national survey of Dutch (pediatric) hematologists from July to September 2021. Results We found 20 articles describing platelet transfusion policies in 483 GT‐patients and 29 BSS‐patients, both adults and children. Twenty surveys were returned for full analysis. All responders treated patients with platelet disorders, including GT ( n = 36 reported) and BSS ( n = 29 reported). Of respondents, 75% estimated the risk of antibody formation as “likely” for HLA and 65% for HPA. Formation of HLA antibodies was reported in 5 GT and in 5 BSS‐patients, including one child. Fifteen respondents gave preventive HLA‐matched platelets in elective setting (75%). Three respondents additionally matched for HPA in GT‐patients (15%). Main argument for matched platelet transfusions was preventing alloimmunization to safeguard the effectivity of ‘random’ donor‐platelets in acute settings. Conclusion Elective HLA‐matching for GT and BSS‐patients is already conducted by most Dutch (pediatric) hematologists. HPA‐matching is mainly applied when HPA‐antibodies are formed. Based on the current literature and the survey, recommendations are proposed.

Background and Objectives Vaccine‐induced thrombotic thrombocytopenia (VITT) is a rare adverse effect characterized by thrombocytopenia and thrombosis occurring after COVID‐19 vaccination. VITT pathophysiology is not fully unravelled but shows similarities to heparin‐induced thrombocytopenia (HIT). HIT is characterized by the presence of antibodies against platelet factor 4 (PF4)/heparin complex, which can activate platelets in an FcγRIIa‐dependent manner, whereas IgG‐antibodies directed against PF4 play an important role in VITT. Materials and Methods We characterized all clinically suspected VITT cases in the Netherlands from a diagnostic perspective and hypothesized that patients who developed both thrombocytopenia and thrombosis display underlying mechanisms similar to those in HIT. We conducted an anti‐PF4 ELISA and a functional PF4‐induced platelet activation assay (PIPAA) with and without blocking the platelet‐FcγRIIa and found positivity in both tests, suggesting VITT with mechanisms similar to those in VITT. Results We identified 65 patients with both thrombocytopenia and thrombosis among 275 clinically suspected VITT cases. Of these 65 patients, 14 (22%) tested positive for anti‐PF4 and PF4‐dependent platelet activation. The essential role of platelet‐FcγRIIa in VITT with mechanisms similar to those in HIT was evident, as platelet activation was inhibited by an FcγRIIa‐blocking antibody in all 14 patients. Conclusion Our study shows that only a small proportion of clinically suspected VITT patients with thrombocytopenia and thrombosis have anti‐PF4‐inducing, FcɣRIIa‐dependent platelet activation, suggesting an HIT‐like pathophysiology. This leaves the possibility for the presence of another type of pathophysiology (‘non‐HIT like’) leading to VITT. More research on pathophysiology is warranted to improve the diagnostic algorithm and to identify novel therapeutic and preventive strategies.

Objective Systemic Lupus Erythematosus (SLE) is characterized by arthritis, rash, glomerulonephritis, and hematologic manifestations, including secondary immune mediated thrombocytopenia (ITP). Presence of ITP type, platelet specific glycoprotein antibodies have been detected in SLE, but their exact role and frequency is not known although thrombocytopenia can be of prognostic importance. The objective of this study was to determine the frequency of platelet specific antibodies in patients with SLE and if the presence of these antibodies is associated with disease activity. Methods We collected serum samples from 74 patients with SLE (≥ 4 American College of Rheumatology 1982 criteria) during high (H) and lower (L) SLE Disease activity index 2000 (SLEDAI-2K) score (median SLEDAI-2K, H = 8.5 and L = 0.5). Mean age at the first sampling was 43,6 and the second 48,6 years. Antibodies towards platelet glycoproteins (GP) IIb/IIIa, GP V and GP Ib/IX were detected in H and L serum samples, using the direct monoclonal antibody immobilization of platelet antigens assay (MAIPA). We also analyzed the presence of antiphospholipid criteria and non-criteria antibodies: Anti-CL (IgG and IgM) Anti-B2GP1 (IgG and IgM) Anti-PS/PT (IgG and IgM) Anti-AnnexinV (IgG and IgM), using ELISA. Results During H and L SLEDAI-2K we detected anti-GP IIb/IIIa antibodies in 29.7% vs 28.4%; anti-GP V antibodies in 48.6% vs 27.0% and anti-GP Ib/X antibodies in 58.1% vs 31.1%. The anti-GP V and anti-GP Ib/X levels were significantly higher in SLEDAI-2K H vs L and positively correlated with SLEDAI-2K score (r=0,322, p<0,0001 and r=0,265, p=0,001) using spearman correlation. During active disease, 66% had at least one positive GP-antibody, compared to 44% in the less active disease. All antiphospholipid antibodies of IgG type had significantly higher levels during SLEDAI-2K H. In total, 45% of patients had at least one positive antiphospholipid antibody during SLEDAI-2K H and 26% during L. Conclusion This study shows that platelet autoantibodies are prevalent in SLE and their increased presence may be modulated by treatment and/or mechanisms regulating disease activity. Future studies will assess the functional contribution from these antibodies on platelet activation, thrombocytopenia and other aspects of SLE pathophysiology.

Background Fetal and neonatal alloimmune thrombocytopenia (FNAIT) is a condition during pregnancy, which can lead to thrombocytopenia and a bleeding tendency with intracranial hemorrhage (ICH) being the most concerning complication in the fetus or neonate. An incompatibility between human platelet antigen (HPA)‐1a accounts for the majority of FNAIT cases. Binding of HPA‐1a‐specific alloantibodies to their target on fetal platelets and endothelial cells can induce apoptosis of megakaryocytes, disrupt platelet function, and impair angiogenesis. Currently, there is no screening program to identify pregnancies at risk for severe disease. A better understanding of HPA‐1a‐specific antibody heterogeneity in FNAIT could aid in identifying pathogenic antibody properties linked to severe disease. Study Design and Methods This study aimed to isolate HPA‐1a‐specific B‐cells from an HPA‐1a‐alloimmunized pregnant woman. Using fluorescently labeled HPA‐1a‐positive platelets, single B‐cells were sorted and cultured for 10 days to stimulate antibody production. Subsequently, supernatants were tested for the presence of antibodies by enzyme‐linked immunosorbent assay and their reactivity towards HPA‐1a‐positive platelets. Amplification and sequencing of variable regions allowed the generation of monoclonal antibodies using a HEK‐Freestyle‐based expression system. Results Three platelet‐specific B‐cells were obtained and cloned of which two were specific for HPA‐1a, named D‐ and M‐204, while the third was specific for HLA class I, which was named L‐204. Discussion This study outlined an effective method for the isolation of HPA‐1a‐specific B‐cells and the generation of monoclonal antibodies. Further characterization of these antibodies holds promise for better understanding the pathogenic nature of alloantibodies in FNAIT.