Platelet refractoriness Overview

Platelet refractoriness is defined as a failure to obtain a platelet count increment (PCI) of more than 10 x 10⁹/l between 1 and 24 hours post transfusion with ABO compatible platelets on at least two separate occasions.

A patient who meets these criteria and who has had non-immune causes of platelet refractoriness excluded, needs to be tested for HLA antibodies to determine if HLA matched platelets are indicated. Note however that a patient may have immune as well as non-immune causes concomitantly. Non-immune causes include active bleeding, splenomegally, hepatomegaly, DIC, sepsis, fever and use of some anti-fungal agent.

Platelets are routinely transfused to patient with clinically significant haemorrhage and patients with severe thrombocytopenia. Normal platelet counts range from 150 – 450 x 10⁹ cells/L and a patient is considered to be thrombocytopenic when platelet counts drop to below 150 x 10⁹ cells/L. In a number of transfusion protocols, platelets are transfused prophylactically to prevent bleeding and this is in keeping with best practice guidelines.

Traditionally the main indication for prophylactic platelet transfusion is the platelet count, with haemorrhage considered likely when platelet count is 10–30 x 10⁹ cells/L. Patients with various forms of bleeding may receive therapeutic platelet transfusions. Such patients include those with petechiae (pinpoint skin haemorrhages), ecchymoses (bruising), patients with bleeding in the gums and mucosa, patients with nose bleeds, haematuria and menorrhagia and sometimes intracranial haemorrhage. However, in England prophylactic platelet transfusion remains the predominate practice even in the autologous stem cell transplant setting. Guidelines recommend transfusion when platelet counts drop to 10 x 10⁹ cells/L or 20 x 10⁹ cells/L in the presence of sepsis, fever or for patients undergoing active treatment. These thresholds may vary depending on the clinical condition. A threshold as low as 5 x 10⁹ cells/L in stable non-bleeding patients who are not in receipt of active treatment has been reported.

Platelet transfusions may also be given prophylactically as clinically required to prevent bleeding in patients with inherited thrombocytopathies such as Glanzmann’s disease or Bernard Soulier Syndrome.

The main immune cause of platelet refractoriness is the presence of antibodies directed against HLA. Alloantibodies directed against HLA class I antigens account for up to 95% of immune refractoriness. The HLA antibodies bind to the donor platelets, making them susceptible to opsonisation and Fc receptor-mediated phagocytosis primarily in the spleen by resident macrophages. Other immune factors such as antibodies directed against human platelet antigens (HPA) in combination with HLA antibodies account for < 10% of immune refractoriness, while anti-HPA antibodies alone account for <1% of immune refractoriness. In a small number of cases, high titre anti ABO antibodies, autoantibodies, some drug dependent antibodies and immune complexes have also been implicated.

HLA matched platelets are indicated for patients who have an immune cause of their platelet refractoriness. The patient should have a sample taken for platelet count pre and 15min – 1 hour (or no more than 24hours) post transfusion to determine the platelet count increment.

Non Response to Matched Platelets

If the last HLA antibody test was carried out a month or more ago, then request a new sample to see if the patient has developed new HLA antibodies. If the patient has been receiving A grade, ABO matched, fresh platelets, then also test for HPA antibodies. If the patient has HPA antibodies I will match for HPA and HLA where possible but prioritise HPA matching over HLA matching if there were limited options. Discuss with the clinicians again to ensure non-immune causes were definitely excluded. Even if evaluated before, new non-immune causes such as bleeding, sepsis, splenomegally, hepatomegaly, DIC and use of medication such as Amphotericin may now be present.

Epitope matching in platelet transfusion

An epitope is the part of an antigen that is recognized by the immune system, specifically by antibodies, B cells or T cells and is therefore also known as an antigenic determinant. Early attempts at epitope definition of HLA were based on direct analysis of amino differences between patients and donors. The HLAMatchmaker system was introduced by Rene Duquesnoy as an in silico system for assessing patient – donor compatibility. This original version of the HLAMatchmaker system treated HLA class I antigens as linear sequences of Triplets of amino acids in antibody accessible regions of the HLA molecule. A later, Eplet version of HLAMatchmaker was introduced which was based on an analysis of the three-dimensional structure of the HLA molecule rather than on linear sequences of amino acids. Instead of epitopes being defined by three amino acids in a linear sequence, these new epitopes were defined as all the amino acids within a 3 to 3.5Å radius of each polymorphic residue position.

In many instances, the Eplet epitopes are identical to Triplets. However, Eplets remove the artificial limitation to three amino acids. Eplet epitopes could consist of any number of amino acids that fall within the radius. In addition, Eplet epitopes extended the repertoire of epitopes to include epitopes formed by the folding of the molecule bringing together discontinuous amino acid sequences to form new epitopes.

Epitope matching has two key concepts:

• Epitopes in a given location on a HLA antigen at one locus are equivalent to the same epitope in the same position on another HLA antigen on a different locus. The definition of epitopes requires that HLA typing be carried out at high resolution so that the exact sequence can be used to identify the epitopes present
  • A patient cannot make antibodies against their own Triplets

Epitope matching gives the potential of increasing the donor pool available to any given patient. A platelet unit which is poorly matched by standard HLA antigen matching for instance could prove to actually have only a few epitope mismatches and may be suitable for that patient. In addition, where there was a choice of equally matched donor platelets, there is some evidence that the one with the fewer epitope mismatches may result in a higher increment. There is also some evidence that Epitope matching may be associated with a lower level of de-novo HLA antibody formation.