key: cord-0291766-5mz3uf24
authors: Kanack, A.; Jones, C.; Singh, B.; Leger, R.; Heikal, N.; Chen, D.; Pruthi, R.; Padmanabhan, A.
title: Long-term Stored Platelets Coupled to Thrombospondin-1 Detection for Rapid and Accurate Identification of Pathogenic HIT Antibodies
date: 2021-09-05
journal: nan
DOI: 10.1101/2021.09.01.21262974
sha: 1813f44799da179521cec0e19d1189b6dd0527bf
doc_id: 291766
cord_uid: 5mz3uf24

Heparin-induced thrombocytopenia (HIT) is a potentially life-threatening disorder characterized by antibodies to Platelet Factor 4 (PF4)-polyanion complexes which cause thrombocytopenia and thrombosis. Currently used technically-simple frontline assays such as the PF4-polyanion enzyme-linked immunosorbent assays (ELISAs) lack specificity, and more accurate functional assays such as the serotonin release assay (SRA) and PF4-dependent P-selectin expression assay (PEA) have long turnaround times due to technical complexity and availability only in the reference laboratory setting. There is a critical need for accurate near-patient functional testing to guide patient management, but a key barrier to attaining this goal is the short-term viability of platelets. Here, we detail a process of platelet cryopreservation that renders them viable for at least one year and show that PF4-treated cryopreserved platelets, when coupled with ELISA-based measurement of thrombospondin-1 release (a platelet -granule protein), detects pathogenic HIT antibodies with high accuracy. Furthermore, testing of a cohort of non-pathogenic HIT antibodies that were strongly reactive in PF4/polyanion ELISA but negative in functional assays demonstrated negative results in the thrombospondin-1 release assay, confirming high specificity of this technique. These findings have the potential to transform the diagnostic testing paradigm in HIT by making in-hospital functional testing available for rapid and accurate diagnosis.

Heparin-induced thrombocytopenia (HIT) is a potentially life-threatening disorder characterized by antibodies to Platelet Factor 4 (PF4)-polyanion complexes which cause thrombocytopenia and thrombosis. Currently used technically-simple frontline assays such as the PF4-polyanion enzyme-linked immunosorbent assays (ELISAs) lack specificity, and more accurate functional assays such as the serotonin release assay (SRA) and PF4-dependent P-selectin expression assay (PEA) have long turnaround times due to technical complexity and availability only in the reference laboratory setting. There is a critical need for accurate near-patient functional testing to guide patient management, but a key barrier to attaining this goal is the short-term viability of platelets. Here, we detail a process of platelet cryopreservation that renders them viable for at least one year and show that PF4-treated cryopreserved platelets, when coupled with ELISAbased measurement of thrombospondin-1 release (a platelet α -granule protein), detects pathogenic HIT antibodies with high accuracy. Furthermore, testing of a cohort of nonpathogenic HIT antibodies that were strongly reactive in PF4/polyanion ELISA but negative in functional assays demonstrated negative results in the thrombospondin-1 release assay, confirming high specificity of this technique. These findings have the potential to transform the diagnostic testing paradigm in HIT by making in-hospital functional testing available for rapid and accurate diagnosis.

HIT is a disorder characterized by antibodies to PF4/polyanion complexes that cause thrombocytopenia and thrombosis [1] [2] [3] [4] . A recent large epidemiological study suggests high morbidity and a mortality rate of ~10% in HIT patients 5 . Currently, two classes of assays are used for the HIT diagnosis. PF4/polyanion enzyme-linked immunosorbent assay (ELISA)-based tests are simple from a technical standpoint but are highly non-specific 6 . For example, in a recent 400+ patient prospective HIT diagnostic study, only ~50% of ELISA-positive samples harbored platelet-activating antibodies 7 . Thus, reliance on these assays results in an "epidemic of overdiagnosis" 8 and excessive use of alternative anticoagulants that have a higher risk of bleeding 6, 9 . In addition to the medical risks associated with excessive use of non-heparin anticoagulants, they are expensive 10 .

On the other hand, more accurate functional (platelet-activation based) assays such as the serotonin-release assay (SRA) 11 and heparin-induced platelet activation assay (HIPA) 12 require complex techniques and handling of radioactive reagents or aggregometry equipment. Thus, functional assays are available only in the context of reference laboratories. Recent studies suggest that the use of PF4-treated platelets in functional HIT testing is equivalent or superior to the use of heparin-treated platelets for HIT diagnosis. 7, 13, 14 A new assay developed based on these findings, the PF4-dependent P-selectin expression assay (PEA), although simpler, still requires flow cytometry 7 and is currently offered by only one laboratory in the United States.

Thus, functional testing is associated with long turnaround times, making their use impractical for immediate patient management.

Therefore, there is a critical need to develop rapid and accurate (functional) diagnostic assays for HIT that can be implemented in the near-patient, in-hospital environment. A key challenge that has prevented the development of self-contained functional HIT in vitro diagnostic assays (IVDs) is the extremely short viability of platelets. Here, we detail the development of a process of cryopreservation that extends platelet viability for at least one year. We also demonstrate that PF4-treated cryopreserved platelets can support the detection of pathogenic, platelet-activating HIT antibodies by measuring the release of an abundant platelet α -granule protein, thrombospondin-1 (TSP1) in a simple ELISA-based format.

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted September 5, 2021. ; https://doi.org/10.1101/2021.09.01.21262974 doi: medRxiv preprint Platelet-rich plasma (PRP) units sourced from whole blood donations anticoagulated with citrate phosphate dextrose from healthy blood donors were obtained two days after draw, upon completion of infectious disease testing. Anticoagulant citrate dextrose-A (1.2mM sodium citrate; 1.4mM dextrose, pH 5.9) and prostaglandin E-1 (50µg/mL stock, Millipore Sigma, St.

Louis, MO) and was added 1:10 (v/v) and 1:1000 (v/v) respectively to the two day old PRP units. The PRP was centrifuged at 100g, 15 minutes. Supernatant was collected and centrifuged at 1000g 15 minutes to pellet platelets. Supernatant was discarded, and platelets were resuspended at a concentration of 1x10 6 /µL and incubated at 37°C for 2 hours in a buffer containing the cryoprotectant, trehalose (9.5 mM HEPES; 100 mM NaCl; 4.8 mM KCl; 5.0 mM glucose; 12 mM NaHCO 3 ; 50 mM trehalose, pH 6.8). 30% (w/v) bovine serum albumin (BSA, Millipore Sigma, St. Louis, MO) was added to a final 4% (w/v) concentration. This platelet solution was then aliquoted in 500µL volumes per tube. Platelets were cryopreserved in a Thermo Scientific CryoMed 7450 controlled rate freezer at a range of freezing rates. Platelets cryopreserved at an optimal rate of 4°C/min to the target of -80°C are presented in this study.

Platelets were stored at -80°C for various periods before testing, as noted in the Results and Discussion section.

Thrombospondin-1 (TSP1), a highly expressed protein in platelet α -granules 15 released upon HIT-antibody mediated platelet activation, was measured by ELISA. Briefly, after method optimization (see Supplementary Appendix for initial conditions), cryopreserved platelet tubes were thawed at 37°C and centrifuged at 1000xg for 5 minutes. Supernatants were discarded, and platelets were resuspended in a modified Ringer's Citrate Dextrose (RCD) buffer containing trehalose (108mM NaCl; 3.8mM KCl; 1.7mM NaHCO 3 ; 22.9M sodium citrate; 27.8mM Glucose; 50mM trehalose, pH 6.5) and centrifuged at 1000xg for 5 minutes. Supernatants were discarded. Platelets were resuspended in 80µL of a phosphate-buffered saline-based (PBS) 7.4 ) and incubated at room temperature for 20 minutes. 20µL of patient serum was added to the PF4-treated platelets, incubated for 30 minutes at room temperature, and supernatants were collected after centrifugation at 1000xg for 5 minutes. In experiments using thrombin receptor activating peptide (TRAP), platelets were prepared similarly but were not treated with PF4.

TRAP (125µg/mL stock, AnaSpec, Fremont, CA) or buffer control was added in lieu of serum to a final concentration of 25µg/mL. TSP1 was quantified from platelet supernatants using a . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

In initial studies, we evaluated the ability of one-week-old refrigerated (4°C) platelets to support detection of HIT antibodies in functional testing using PF4-treated platelets, compared to 2-day old platelets obtained from room temperature stored PRP bags (Supplementary Fig S1) .

Results demonstrated strong baseline platelet activation due to refrigeration, as evidenced by increased P-selectin (CD62p) expression even upon incubation with healthy donor sera ( Supplementary Fig S1) , suggesting that long-term refrigeration of platelets is not a viable option for platelet storage. Our observation of increased activation with cold-stored platelets agrees with previously published observations that refrigerated platelets undergo structural, metabolic, and functional changes that impact their activation state 16, 17 . As an alternative to refrigeration, we explored the possibility of using cryopreserved platelets in functional testing.

Cryopreserved platelets, developed as described in Methods, were initially treated with a strong platelet agonist, TRAP, which resulted in robust activation as evidenced by high TSP-1 release (190 µg/mL vs. 26 µg/mL with buffer control; Supplementary Fig S2) .

Next, we assessed the feasibility of using PF4-treated cryopreserved platelets to detect plateletactivating HIT antibodies. Initially, 20 platelet lots, each obtained from a different healthy donor were used in the study (Figure 1A) . Platelets were cryopreserved as described, thawed, treated with PF4, and incubated with three platelet-activating HIT or normal control samples to assess their ability to elicit platelet activation in the thrombospondin-1 release assay (TRA). Despite some variability, HIT samples stimulated higher TSP1 release relative to controls for each of the 20 platelet lots tested (Figure 1A) . To assess the longer-term viability of cryopreserved platelets . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted September 5, 2021. ; https://doi.org/10.1101/2021.09.01.21262974 doi: medRxiv preprint for HIT antibody detection, the TRA was repeated using seven platelet lots that had been frozen for more than four weeks (Figure 1B) , and demonstrated that platelets remained viable and capable of degranulation upon activation even after storage. Of note, the reactivity of platelets at these two time points was highly reproducible within each platelet lot. Together, this data suggests that while all platelets were "reactive" in the TRA, some may be more reactive than others, a phenomenon that has been documented with other HIT functional assays such as the SRA 18 . Critically, longer-term storage of platelets for 12 months supported HIT antibodymediated, PF4-dependent platelet activation resulting in the release of significant amounts of TSP1 ( Figure 1C) .

To assess the diagnostic accuracy of the cryopreserved platelet-coupled TRA, a cohort of 18 PEA-positive and 18 PEA-negative HIT-suspected patient samples were selected for the study (Figure 2A) . The cohort was specifically selected to include several cases with "false-positive" IgG-specific ELISA results, i.e., ELISA-positive but negative in functional testing. In this cohort, PEA-SRA concordance was demonstrated in 34 of 36 samples (Figure 2A) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted September 5, 2021. ; https://doi.org/10.1101/2021.09.01.21262974 doi: medRxiv preprint conducted using cryopreserved platelets that were shipped on dry ice for ~30 hours, demonstrating the viability of platelets with cold chain-compatible transport. Proof-of-concept studies presented in Supplemental Figure S3 also demonstrate that heparin-treated cryopreserved platelets, like PF4-treated platelets, can be used to detect platelet-activating HIT antibodies.

In summary, using a heterogeneous group of patient samples, including those that were strongly false-positive in the ELISA, cryopreserved platelets coupled to a TSP-1 ELISA endpoint demonstrated 100% sensitivity and specificity for the detection of platelet-activating HIT antibodies. While platelet activatability after long-term storage (12-months) has been demonstrated, studies to evaluate platelet stability at 18 and 24-months are ongoing. The ability to use 2-day old platelets from PRP units, which are available in abundance from the blood banking industry, should facilitate large-scale production of platelets for in vitro diagnostic kit use. Pooling of platelets, a strategy that we have used successfully for a recently deployed functional HIT diagnostic assay, the PEA 7 , will be employed in this setting to minimize lot variation in platelet activatability. These findings have the potential to transform the diagnostic testing paradigm in HIT by making in-hospital functional testing available for rapid and accurate diagnosis by the deployment of a self-contained HIT IVD assay that will not require the user to obtain donor platelets. Emerging data suggest that anti-PF4 antibodies in Vaccine-induced immune thrombocytopenia and thrombosis (VITT) seen after some COVID-19 vaccines are consistently detected using PF4-treated platelets as opposed to heparin-treated platelets [20] [21] [22] .

Thus, findings reported here may be relevant for the detection of VITT antibodies, and studies are planned.

We would like to thank Stephanie Hafner, BS from Mayo Clinic's Research Innovation Office for exceptional research coordination support. This work was supported, in part, by National presented. Seven patients had platelet-activating antibodies (closed circles) while four were did not (red squares). Y-axis depicts TSP1 concentration (in µg/mL) measured in platelet supernatant and X-axis depicts the ELISA optical density. Each data point represents the mean of technical triplicates. A two-tailed, unpaired Student's t test was used for comparison. ****p < 0.0001.

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted September 5, 2021. . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 1 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted September 5, 2021. ; https://doi.org/10.1101/2021.09.01.21262974 doi: medRxiv preprint

Heparin-Induced Thrombocytopenia

Laboratory diagnosis of heparin-induced thrombocytopenia

Clinical and laboratory diagnosis of heparin-induced thrombocytopenia: an integrated approach

Pathogenesis of heparin-induced thrombocytopenia

Disease burden, complication rates, and healthcare costs of heparin-induced thrombocytopenia in the USA: a population-based study

Five hematologic tests and treatments to question

A prospective, blinded study of a PF4-dependent assay for HIT diagnosis

Heparin-induced thrombocytopenia (HIT) in 2011: an epidemic of overdiagnosis

The risk of major bleeding in patients with suspected heparin-induced thrombocytopenia

Assessing the clinical and cost impact of on-demand immunoassay testing for the diagnosis of heparin induced thrombocytopenia

A diagnostic test for heparin-induced thrombocytopenia

A rapid and sensitive test for diagnosing heparin-associated thrombocytopenia

Beneficial effect of exogenous platelet factor 4 for detecting pathogenic heparin-induced thrombocytopenia antibodies

A Novel PF4-Dependent Platelet Activation Assay Identifies Patients Likely to Have Heparin-Induced Thrombocytopenia/Thrombosis

Role of thrombospondin in platelet aggregation

The clearance mechanism of chilled blood platelets

Mechanism of shape change in chilled human platelets

Determinants of donor platelet variability when testing for heparin-induced thrombocytopenia

Serotonin release assay (SRA)-negative HIT, a newly recognized entity: Implications for diagnosis and management

Successful treatment of thrombotic thrombocytopenia with cerebral sinus venous thrombosis following Ad26.COV2.S vaccination