Abstract
This article discusses key considerations regarding ticagrelor’s reported effect on heparin-induced thrombocytopenia functional assays, such as literature gaps and possible management strategies.
Limited data indicate that ticagrelor may induce false-negative results in functional assays used in the diagnosis of heparin-induced thrombocytopenia. False-negative functional assays for heparin-induced thrombocytopenia could have catastrophic consequences. The manufacturer labeling of ticagrelor now includes a warning for this potential drug-laboratory interaction. This article suggests areas that would benefit from further research and strategies in navigating this possible interaction.
Clinicians should exercise caution when evaluating functional assays for heparin-induced thrombocytopenia in patients receiving ticagrelor. This article offers suggestions for future areas of research and potential management strategies.
In the setting of limited data, the ticagrelor–functional assay interaction presents a clinical conundrum to clinicians in diagnosing heparin-induced thrombocytopenia (HIT).
Strategies to manage this potential interaction include evaluating the magnitude of the 4Ts score and HIT immunoassay result, changing to an alternative P2Y12 inhibitor, and holding ticagrelor for 48 hours if clinically appropriate before testing.
Research elucidating whether this interaction is generalizable to additional HIT functional assays and/or P2Y12 inhibitors is an unmet scientific need that warrants further study.
Type II heparin-induced thrombocytopenia (HIT) is a serious prothrombotic adverse drug reaction characterized by the formation of antibodies that recognize platelet factor 4 (PF4)–heparin complexes and bind to platelet FcγRIIA receptors.1-3 The overall incidence of HIT has been estimated to be 1 in 5,000 hospitalized patients.2 Use of unfractionated heparin (UFH) has been found to carry a greater than 10-fold–higher risk of HIT compared to low-molecular-weight heparin.4 Additional risk factors for HIT include therapeutic doses of UFH, prolonged duration of UFH therapy, and undergoing major surgery.1,4,5 Severe and life-threatening venous and arterial thromboses, including ones requiring amputation, have occurred in patients with HIT.1,6 Adverse outcomes such as increased mortality, prolonged length of hospitalization, and greater hospitalization costs have also been associated with HIT.7
Clinical guidelines recommend utilizing the 4Ts scoring system (range: 0 to 8 points), a validated tool for predicting the probability of HIT, as an initial risk-stratification tool in the diagnosis of HIT.1,8 The score is based on 4 criteria: degree of thrombocytopenia, timing of platelet decline, presence of a new or worsening thrombosis, and presence or absence of other etiologies of thrombocytopenia.2 The diagnostic criteria for HIT include an intermediate or high risk of HIT (≥4 points) as determined with the 4Ts tool, an elevated immunoassay value, and disease confirmation with a HIT functional assay.1,2,9-12
The enzyme-linked immunosorbent assay (ELISA), a highly sensitive, widely used screening immunoassay, measures the concentration of anti–PF4-heparin antibodies.9 The test detects color changes within a sample, which are proportional to the concentration of anti–PF4-heparin antibodies and are reported in units of optical density (OD).9,13 However, this ELISA has poor specificity in patients with low suspicion of HIT due to the detection of clinically insignificant antibodies, leading to false positives.9 A more recently developed HIT immunoassay, the latex immunoturbidimetric assay (LIA) (HemosIL HIT-Ab(PF4-H)), uses anti-PF4–heparin murine monoclonal antibody coated onto polystyrene latex nanoparticles.14 In the absence of plasma containing HIT antibodies, PF4-polyvinylsulfonate complexes agglutinate with the coated monoclonal antibody, leading to higher absorbance and lower light transmission.14 In the presence of plasma containing HIT antibodies, agglutination is inhibited and light is not absorbed by the sample.14 Results are expressed in units of concentration (units/mL), with values of ≥1 unit/mL interpreted as positive for HIT.14 Considered “a functionalized immunoassay,” LIA may have a higher specificity than the ELISA.14
Functional tests, such as the heparin-induced platelet activation (HIPA) test and the serotonin release assay (SRA), confirm the diagnosis by detecting molecular binding between pathogenic antibodies and platelets, which explains their greater specificity compared to immunoassays.2,10 Functional assays for HIT combine donor platelets with heparin and the patient’s serum.10 The HIPA test, which is commonly used in Europe, relies on a subjective evaluation of platelet aggregation.10,12 The SRA test measures 14C-radiolabeled serotonin release from dense granules of activated platelets.10
In vitro studies have demonstrated that the reversible ADP P2Y12 inhibitor ticagrelor induces false-negative HIPA results.15 In a recent case report, a patient with a high 4Ts score (7 points) and a positive immunoassay (HemosIL HIT-Ab(PF4-H) reading of >4.8 U/mL) initially had a negative SRA result while receiving ticagrelor.16 Ten days after ticagrelor discontinuation, the SRA test was repeated with a positive result.16
Ticagrelor, an antiplatelet used for patients with coronary stents, impairs platelet ADP receptor function and prevents FcγRIIA cross-linking and platelet activation and aggregation.17 Ticagrelor is not expected to interfere with anti-PF4 antibody testing because only the serum concentration of antibody is measured rather than platelet activation.18 False-negative results are considered to be 10 times worse than false-positive ones, underscoring the catastrophic complications (eg, thrombosis or death) when failing to identify HIT.1 Moreover, false-negative results could lead to the re-initiation of heparin products, potentially resulting in devastating outcomes.
Ticagrelor and functional assays for HIT: new precautions in a landscape of scant data
In vitro studies with HIPA testing prompted a change in ticagrelor’s manufacturer labeling, which currently states: “Ticagrelor has been reported to cause false-negative results in platelet functional tests (to include, but may not be limited to, the heparin-induced platelet aggregation [HIPA] assay) for patients with heparin-induced thrombocytopenia [HIT]).”18 Although the manufacturer labeling does not specifically list the SRA test, which is commonly used in the US, a possible interaction is implied. Additionally, a laboratory that conducts the SRA test now includes this interaction in its instructions for blood sample collection and advises against the concurrent use of ticagrelor.19
As the evidence for this interaction is based on scant data, specific knowledge gaps exist that, if elucidated, may provide guidance in best managing this interaction. In vitro studies to determine ticagrelor’s effect on the SRA test would assist US-based practitioners. Additionally, in vitro studies investigating the effect of non-ticagrelor P2Y12 inhibitors on HIT functional assays are also needed. Lastly, further development of diagnostic algorithms without the use of functional assays may provide clinicians diagnostic confidence in the treatment of patients with suspected HIT who require uninterrupted use of ticagrelor.
Because of the serious nature of this interaction, clinicians may be required to change their usual approach to diagnosing HIT. This article discusses various aspects of this interaction, including crucial areas of research that would benefit from further investigation and strategies to navigate this potential interaction.
Educating the medical community
Prescribers, pharmacists, laboratory personnel, and other relevant healthcare professionals should be educated about this potential drug-laboratory interaction. The electronic health record may be utilized to alert clinicians of this possible drug-laboratory interaction. Institutional policies and procedures, with coordination between pharmacy and the laboratory, should be developed to address this possible interaction. In one study, the laboratory contacted the pharmacy when HIT immunoassays were ordered.20 Pharmacists then calculated the 4Ts score and provided recommendations to prescribers based on the probability of HIT.20 As a result, this program led to cost savings through the elimination of unnecessary tests, without an increase in the incidence of thrombotic or hemorrhagic events.20
Judicious ordering of HIT tests
Clinicians should order HIT tests when this is clinically appropriate.1 Unless a concern for incomplete or missing data exists, HIT testing should be avoided for low-probability 4Ts scores (≤3 points).1 False-positive immunoassays may lead to unwarranted treatment and testing. Unnecessary HIT testing has been minimized by the implementation of pharmacist-led stewardship programs, which involved a clinical support tool sending an alert to the pharmacists’ pager, aimed at discontinuing immunoassays in low-risk patients.21
Clinicians should also assess patients for other causes of thrombocytopenia, such as malignancy, liver disease, autoimmune disorders, disseminated intravascular coagulation, medications (eg, chemotherapy), extracorporeal devices, and coagulation disorders.22,23 Of note, absolute thrombocytopenia has been reported with ticagrelor.24
Evaluation of novel diagnostic tests
Multiple novel laboratory tests have been described in attempts to expedite the diagnosis of HIT.11,25 The heparin neutralization assay (HNA) has been published and validated within a HIT diagnostic framework.25 An HNA score of <70% inhibition is associated with a high negative predictive value of HIT.25 When incorporated into a diagnostic algorithm with the 4Ts score and ELISA OD measurement, the HNA score may provide high sensitivity and specificity for the diagnosis of HIT and may obviate the need for functional assays in 90% of HIT diagnoses.25
Evaluation of the magnitude of immunoassay results
For HIT antibody immunoassays, the likelihood of HIT correlates with the magnitude of the positive result.9,11 Therefore, the probability of HIT increases with higher 4Ts scores and antibody concentrations.1 For the anti–PF4-heparin ELISA, the OD value is correlated with the probability of HIT. An OD of ≥2.00, 1.50-1.99, 0.6-1.49, and <0.6 indicates a high probability, increased chance, indeterminant chance, and low likelihood of HIT, respectively.9 Other OD cutoffs have been suggested.25 We propose a resultant algorithm to diagnose HIT in the event of functional assay unreliability, which combines the OD concentration, HNA measurement, and current HIT treatment guidelines (Figure 1).
Heparin-induced thrombocytopenia (HIT) diagnosis in the absence of reliable functional testing.1,11,25 The 4Ts score should be based on accurate and complete data and should be reevaluated often.1 For questionable 4Ts scores, clinicians may empirically assume a higher 4Ts score to prevent misdiagnosis of HIT.1 ELISA indicates enzyme-linked immunosorbent assay; HNA, heparin neutralization assay; LIA, latex immunoturbidimetric assay; OD, optical density; SRA, serotonin release assay.
Furthermore, the aforementioned LIA assay also provides prognostic value in risk stratifying an individual patient’s likelihood of having HIT based on the degree of elevation in the LIA.14 Previously defined laboratory cutoffs of weakly, moderately, and strongly positive LIA values are associated with a progressively increased likelihood of HIT truly being present.14 We propose that this assay may be used to further refine clinical suspicion for HIT in situations where functional assays may be unable to provide reliable results (Figure 2).
Heparin-induced thrombocytopenia (HIT) risk stratification in diagnostic uncertainty.14 LIA indicates latex immunoturbidimetric assay.
Temporary withdrawal of ticagrelor if clinically appropriate
Eekels et al15 found that at clinical steady-state concentrations (227-770 ng/mL), ticagrelor inhibited the HIPA test in 4 of 5 HIT-positive sera samples. Ticagrelor’s ability to inhibit the HIPA test appears to be concentration dependent.15 In the single case report describing the possible ticagrelor-SRA interaction in a patient with a high likelihood of acute HIT, the SRA result was positive 10 days after ticagrelor discontinuation.16 In scenarios where withholding ticagrelor is reasonable, the optimal time required to diminish ticagrelor’s effect on HIT functional assays may depend on its half-life. The mean half-life of ticagrelor is 7 hours, while that for its active metabolite is 9 hours.18 The effect of ticagrelor should be minimal 48 hours after discontinuation;26 however, organ dysfunction and drug interactions could alter ticagrelor’s half-life and its duration of action.18 In patients receiving ticagrelor with a moderate to high probability of HIT (4Ts score of ≥4 points) despite a negative functional assay, avoiding heparin products and monitoring for thrombosis may be warranted.
Use of alternative P2Y12 inhibitors
Although there have not been any studies investigating the effect of irreversible P2Y12 inhibitors such as clopidogrel and prasugrel on HIT functional assays, an interaction appears less likely than with ticagrelor.15 Clopidogrel and prasugrel are both prodrugs that are converted to active metabolites with thiol groups.15,27-29 It has been postulated that the inherent instability of the thiol group decreases the likelihood of an interaction with HIT functional assays.15 The case report demonstrating a possible ticagrelor-SRA interaction reported that the SRA result was positive when the patient was on clopidogrel, indicating that clopidogrel likely did not affect the SRA result.16 However, definitive studies are needed to determine whether clopidogrel or prasugrel interferes with HIT functional assays.
Cangrelor, an intravenous, reversible P2Y12 inhibitor with a short half-life (3 to 6 minutes),30 may interact with HIT functional assays;15 however, conclusive studies are needed to confirm an interaction. Discontinuation of cangrelor 30 minutes before blood sample collection has been proposed as an option to avoid interference with HIT functional assays.15 Bridging with cangrelor typically involves discontinuation of ticagrelor for 48 hours before initiating cangrelor.30 However, at 48 hours, the likelihood of an interaction between ticagrelor and HIT functional assays diminishes, which would preclude the need for cangrelor, a costly drug.31
When changing to an alternative P2Y12 inhibitor, clinicians must also assess the clinical appropriateness. Clopidogrel is a less potent antiplatelet agent relative to the other P2Y12 inhibitors, with marked inter-individual response due to variable metabolic activation through CYP2C19.27,32 Prasugrel is contraindicated in patients with a history of stroke and transient ischemic attack.33 Clinicians should weigh the benefits and risks of withholding or switching between P2Y12 inhibitors.
Laboratory techniques to counter ticagrelor’s effect
Laboratory techniques to remove ticagrelor from serum have been suggested, including charcoal powder, an immunoglobulin G fraction, or a ticagrelor reversal agent.10 Because these laboratory techniques are currently not approved by the Food and Drug Administration, these methods would be classified as moderate- to high-complexity testing.34 At this time, laboratories may not be equipped with the necessary materials and personnel training to perform these techniques. Further studies are needed to validate techniques to counter the effects of ticagrelor.
Conclusion
Ticagrelor’s ability to cause false-negative results in HIT functional assays is based on limited evidence. Considering the current available data and the potential detrimental consequences of this interaction, clinicians may be required to adjust their usual practice. Robust research is needed to further explore the ideas presented in this article.
Acknowledgments
The authors would like to thank the following individuals for their contributions to this manuscript: John Sto. Domingo (lead medical laboratory scientist at Providence St. Jude Medical Center, Fullerton, CA), Puja Patel, PharmD (oncology pharmacist), and Dev Udani, PharmD.
Data availability
No new data were generated or analyzed in support of this article.
Disclosures
The authors have declared no potential conflicts of interest.
Additional information
Dr. Salman and Dr. Johannesmeyer contributed equally to this work.
Comments