Improving the transfusion service’s ability to provide antigen-matched RBC units for alloimmunized patients

By: Judy O’Rourke

Yazer Head Shot Mark H. Yazer, MD, is an associate professor of pathology, University of Pittsburgh, and is the medical director of the RBC serology laboratory at the Centralized Transfusion Service (a division of the Institute for Transfusion Medicine), Pittsburgh.

Blood transfusion plays a crucial role in the care of patients who have sickle cell disease, thalassemia, and anemia secondary to cancer and cancer therapy.

Between 5% to 10% of patients who receive numerous or chronic red blood cell transfusions become alloimmunized to the donor’s cells. This can pose a problem to the transfusion service, as it is of the utmost importance to provide these recipients with RBC units that lack the antigens to which the recipient has made an antibody.

New technologies based upon blood group genotyping can assist in this process.

Looking ahead to the AABB’s annual meeting, to be held from October 12 to 15 in Denver, CLP talks with Mark H. Yazer, MD, about the potential benefits of blood group genotyping in identifying rare blood types and improving the transfusion service’s ability to provide antigen-matched RBCs for alloimmunized patients.

CLP: Tell us a little bit about red blood cells…
Mark H. Yazer, MD: The most important function of red blood cells (RBCs) is to carry oxygen to tissues and transport waste products like CO2 back to the lungs. To do this, they contain hemoglobin and little else. We transfuse RBC units to recipients to improve their oxygen carrying capacity—for example, if a patient’s bone marrow is infiltrated by cancer cells, or has been ablated following chemotherapy, or if they have experienced significant bleeding during surgery or due to a trauma, they might require RBC transfusions to ensure that they have enough hemoglobin to maintain normal cellular function.

In other situations, such as patients with sickle cell disease (SCD), RBC transfusions are administered to prevent or treat life-threatening crises that occur when the RBCs adopt a sickle cell shape. These crises can involve major organs such as the lungs and brain, and cause significant morbidity or mortality. In these situations, the indication for the transfusion isn’t so much to maintain the hemoglobin level as it is to dilute out the sickling type of hemoglobin that is causing the problem. We’ll talk more about SCD patients and the challenges they can pose to the transfusion service later. Keep in mind that at the moment there aren’t any widely available hemoglobin substitutes or artificial blood—we depend on human RBCs for transfusion.

Unlike most other body cells, a mature RBC does not have a nucleus. In fact, it sheds the nucleus as it exits the bone marrow. A nucleus would just take up room inside the RBC that could otherwise be occupied with oxygen-carrying hemoglobin molecules. Although RBCs do not contain DNA, they do have an abundance of proteins and carbohydrates on their surface, and these structures (called antigens) can differ between people. The antigens that are expressed on the RBCs are collectively known as the “RBC phenotype.”

When a recipient is transfused with donor RBCs that bear an antigen that their own cells lack, their immune system can be triggered to form antibodies to this antigen. A recipient who has formed a new antibody is said to be alloimmunized against the antigen. In some cases, if an alloimmunized recipient is transfused with RBCs bearing the same antigen again, hemolysis (destruction of the antigen-bearing RBCs) can ensue. Hemolytic reactions can cause morbidity, and in some cases, mortality. Thus, it is important for transfusion medicine specialists to determine if a recipient has antibodies directed against other people’s RBCs—and if they do, RBC units that lack these antigens have to be provided for future transfusion.

CLP: What is blood group genotyping?
Yazer: The traditional way of determining a person’s RBC phenotype (the different antigens that are on the surface of their RBCs) is by using serological techniques. We’ll talk more about these techniques later. Blood group genotyping (BGG) is a newer method for predicting which antigens will be expressed on the RBCs. BGG techniques use the individual’s DNA to predict, with a high degree of certainty, which antigens will be expressed on a person’s RBCs. The protein antigens (such as Rh, Duffy) are encoded directly by a gene, while the carbohydrate antigens (like ABO and Lewis) are created by enzymes that transfer specific sugars to structures that end up on the surface of the RBC. So, in order to predict which antigens will be on a person’s RBCs, the genes for the antigens or the enzymes can be detected using a variety of either commercially available platforms, or “home brew” tests designed and used in-house by a molecular testing laboratory. Just like DNA tests are used to detect oncogenes, we can also use someone’s DNA to tell us which blood group antigens they will express on their RBCs. As the DNA that encodes blood group antigens (and the enzymes that create carbohydrate antigens) does not change over time, each donor would only have to be genotyped once in his or her lifetime.

CLP: How is BGG helpful in the transfusion service?
Yazer: This is a good question! One of the side effects of RBC transfusion is that the recipients can make antibodies (ie, become alloimmunized) to the donor RBCs. Making these antibodies can be a problem because the transfusion service must then find RBC units that lack the corresponding antigen for all future transfusions—lest a potentially hemolytic reaction occur. If a recipient makes an antibody to an antigen that is uncommon in the population, such as the K antigen in the Kell blood group system, which is present on the RBCs of about 9% of Caucasians, then finding compatible RBCs that lack the antigen won’t be a problem. If, however, the antibody is directed to a higher-incidence antigen, or if the recipient makes multiple antibodies to different antigens, then finding antigen-negative RBC units becomes more and more difficult.

The way we currently find antigen-negative RBCs is a slow and tedious process. We literally select units off the shelf at random, and then add antisera to a small sample of these units hoping to find those that are antigen-negative. It really is a hit-and-miss process, and the serological reagents are expensive. If we could perform RBC genotyping on each of the donor units, we could make an educated guess as to which of them are antigen-negative—that is, compatible with the recipient’s antibody—and limit the serological testing to those who are predicted to be antigen-negative. It’s like arriving in London, England for the first time; there’s so much to do that it can be overwhelming. But if you have a reliable guidebook that points you in the direction of the best attractions and restaurants, then you won’t waste time in searching out the things to do that are the most compatible with your interests.

CLP: Who could benefit from BGG?
Yazer: As you can see, the transfusion service expends a lot of resources trying to find compatible RBCs for patients who have made antibodies to RBC antigens. Anything we can do to make the hunt for compatible RBCs easier and faster is going to help us reduce our workload in the blood bank. And I think that patients will benefit, because they would get compatible RBC units faster. Another way to look at this question is to ask, “How many recipients make antibodies?” Luckily only 5% to 10% of RBC recipients become alloimmunized after transfusion. I would, however, say that these 5% to 10% of recipients generate more than 85% of the workload in the immunohematology section of the transfusion service! SCD patients have a much higher alloimmunization rate compared to other RBC recipients—on the order of 35% to 40% make an antibody. Sometimes it takes us hours to determine which antibodies the recipient has produced and to find compatible units for them; depending on the nature and number of the antibodies, and the number of units ordered, we occasionally can’t completely fill the order.

Sometimes we can only find 3 units when 6 or more are ordered for an exchange transfusion. If we had a large stock of genotyped units, we could save a lot of time in locating the antigen-negative units, and we would save reagent costs because we wouldn’t have to do a mass screening of the inventory looking for them. So any recipient who has made an antibody can benefit from their blood bank having a large genotyped RBC inventory.

But beyond that, BGG can help us predict which antibodies a recipient could make, and so it can help us provide them with units that lack not only the antigen to which they are already alloimmunized, but also those to which they could make an antibody if they were exposed to the antigen. Right now, it is common practice at many large tertiary care hospitals that treat SCD patients to match the donor units to the recipient’s Rh and K antigens, because these are the most common antibodies that SCD patients tend to make. If the SCD patient makes antibodies, then the transfusion service can try to match for even more antigens including, of course, the antigen to which the patient has made the antibody. This is no small task, and as I said before, it is a manual process that uses expensive antisera. If we knew the recipient’s blood group genotype then we could match them to donors at the genetic level for literally dozens of antigens, not just Rh and K, thereby preventing them from becoming alloimmunized in the first place.

Another population of recipients who could benefit from BGG are those who have been recently transfused and who have made a new antibody. Because this recipient’s own RBCs are mixed with the donor RBCs, we often cannot reliably determine the recipient’s RBC phenotype (the combination of antigens on their RBCs), and knowing the phenotype can help guide us in determining the specificity of the antibody.

Because BGG is not hampered by the presence of donor RBCs in the same way that traditional serological-based tests are, we can use this type of testing to help predict the recipient’s phenotype, thereby assisting in our serological investigation. Patients who have autoimmune hemolytic anemia, whereby an antibody to their own RBCs obscures the ready detection of alloantibodies, can also benefit from genotyping. By providing these recipients with donor RBCs that are matched to their phenotype, the risk of alloantibody mediated immune hemolysis is mitigated.

CLP: What are the obstacles to implementing BGG?
Yazer: None of the commercially available platforms are currently FDA-approved and so they are used under research use only (RUO) protocols. This makes billing for the testing somewhat tricky. In my transfusion service, we absorb the cost of the genotyping that we perform on both donors and recipients (with an emphasis on our SCD patients) using a commercially available platform. We feel that the cost of genotyping is somewhat offset by the reductions in both antisera usage and the technologist’s time spent searching for antigen-negative RBC units.

Using these techniques in an RUO manner also means that we cannot label the RBC unit as antigen-negative solely based on the genotype data; we still have to confirm the antigen status using FDA-approved reagents. Maybe one day, once genotyping methodologies are approved by the FDA, we can skip the serological confirmation step. But even so, testing a few units that are likely to be antigen-negative based on BGG testing is much better than screening units at random. Also, since BGG requires specialized knowledge and skills that are not widely taught in medical schools or technologist training programs, I think it has, up to now, been adopted mainly in large tertiary care facilities that treat patients who are at high risk of becoming alloimmunized, and in blood centers that have hired specialists in the area. Whether BGG remains in the purview of these types of institutions in the future remains to be seen, and might hinge upon the ability to get reimbursed for performing the testing, and the number of people interested in, and qualified to interpret the results of, this sort of testing.

Mark Yazer, MD,  is an associate professor of pathology, University of Pittsburgh, and medical director, RBC serology laboratory, Centralized Transfusion Service (a division of the Institute for Transfusion Medicine), Pittsburgh. Yazer was not compensated for contributing to this article, and he is on the transfusion medicine scientific advisory boards of Grífols, Novartis, Octapharma, and Ortho J & J. For more information, contact Chief Editor Judy O’Rourke, [email protected]


Serology vs. Molecular: Understanding Molecular Immunohematology
Industry Workshop sponsored by Novartis Diagnostics

Monday, October 14, 7 am to 8:15 am

Colorado Convention Center, Room 207

Serology has long been the established means of antibody identification and understanding the blood group systems. For the major ABO blood group system, it provides inexpensive antisera and automation for routine blood grouping. Molecular tools have expanded our knowledge of the over 30 blood group systems and the molecular basis for most blood group polymorphisms. In this workshop, participants will discuss serology and explore automation using standard antisera. In addition, we will examine molecular tools currently in research to predict the phenotypes from genotype results, and look at challenges between the perceived “gold standard” of serology and molecular opportunities to embrace the strengths of molecular immunohematology in transfusion science.


  • Recognizing the Strength of Serology and Embracing the Clarity of Genotyping – Susan T. Johnson, MSTM, MT (ASCP) SBB, Blood Center of Wisconsin, Milwaukee
  • Molecular Immunohematology: Understanding the Benefit to the Chronically Transfused – Patricia A. R. Brunker, MD, DPhil (Oxon), The Johns Hopkins Hospital, Baltimore

A Q&A session will follow the presentations