Red blood cell (RBC) transfusion in individuals with serologic complexity

INTRODUCTION — An essential goal in transfusion medicine is that red blood cell (RBC) units selected for transfusion be compatible so as not to cause hemolytic transfusion reactions. Usually this is straightforward, as discussed separately. (See "Pretransfusion testing for red blood cell transfusion".)

However, there are times when the situation becomes more complex, and it becomes challenging or impossible to identify crossmatch compatible blood. Examples of serologic complexity include panagglutination on the antibody screen or an incompatible crossmatch. This topic reviews these situations and our approach to identifying units for transfusion.

CONCEPTUAL FRAMEWORK

Standard type/screen/crossmatch — Standard pretransfusion testing consists of:

●Type – ABO and RhD typing are done using reagent antisera on the patient's RBCs (also called forward typing). (See "Pretransfusion testing for red blood cell transfusion", section on 'Blood type (ABO and RhD type)'.)

●Screen – The antibody screen is done using reagent RBCs on the patient's plasma/serum. This can be done in a test tube (liquid phase) or enzyme-linked immunosorbent assay (ELISA)-type assay (solid phase). The screen generally starts with reagent type O RBCs that express common antigens (screening cells), and, if positive, a panel of reagent RBCs is used to determine the antibody specificity/specificities (identification cells). (See "Pretransfusion testing for red blood cell transfusion", section on 'Antibody screen'.)

If a patient's plasma/serum reacts with all of the reagent RBCs on the panel, this is referred to as panagglutination. (See 'Panagglutination' below.)

●Crossmatch – Crossmatching (also called compatibility testing) is only done when blood is ordered, to match a specific RBC unit to a specific patient. It can be done electronically for selected low-risk patients with a negative antibody screen or using different serologic assays in more complex patients. (See "Pretransfusion testing for red blood cell transfusion", section on 'Compatibility testing (crossmatch)'.)

We agree that the concept and terminology of the "least incompatible blood" should be avoided; this is no longer used in most reference laboratories responsible for providing units of RBCs [1].

In vivo crossmatch procedures are of historical interest only. These approaches used radiolabeled RBCs or in vivo assessment of hemolysis following a small infusion of RBCs [2,3].

RBC antigens and the alloantibodies most likely to cause clinical hemolysis are discussed separately. (See "Red blood cell antigens and antibodies", section on 'Clinically significant (common)' and "Red blood cell antigens and antibodies", section on 'Clinically significant (rare)'.)

Reasons for positive antibody screen or incompatible crossmatch — Reasons for a positive antibody screen or incompatible crossmatch include the following, which are discussed in detail below:

●Alloantibodies – Often when there is a single or small number of alloantibodies, compatible blood can be easily identified by excluding the implicated antigen(s). However, some individuals have multiple alloantibodies and/or incompatibility due to alloantibodies that are not identified on the antibody screen (either because they are not tested or because they are tested and not detected). (See 'Alloantibodies' below.)

●Autoantibodies – Autoantibodies can be associated with autoimmune hemolytic anemia or they can be clinically silent. The main concern is that they can obscure detection of a clinically significant alloantibody when the antibody screen is performed. (See 'Autoantibodies' below.)

●Therapeutic monoclonal antibodies – Certain monoclonal antibodies that cross-react with RBC antigens will cause panagglutination. (See 'Monoclonal antibodies that affect the antibody screen' below.)

●Change in blood type – Individuals who have undergone hematopoietic stem cell transplantation (HSCT) can receive hematopoietic stem cells of a different blood type. Some recipients will have a period of overlap during engraftment and others will have persistent partial donor chimerism. (See 'Allogeneic hematopoietic stem cell transplantation recipients' below.)

●ABO discrepancies – Incompatible crossmatches can be observed in cases of A or B subtypes or in patients with the "acquired B" phenotype. (See 'ABO discrepancies' below.)

EMERGENCY RELEASE BLOOD FOR LIFE-THREATENING ANEMIA OR BLEEDING

●Never withhold transfusion if needed – Urgent or emergency situations with severe anemia (hemoglobin <6 g/dL) and/or hemodynamic compromise from a rapid decline in hemoglobin require immediate stabilization and usually transfusion. Delay in transfusion could be fatal. In these situations, patients should be transfused with emergency release blood pending the full blood bank evaluation. Patients should never be denied a lifesaving transfusion because of difficulties in finding compatible blood.

The decision to request emergency release blood is made by the clinician caring for the patient with involvement of transfusion medicine personnel. (See 'Communication' below.)

Thresholds for transfusion are discussed in detail separately. However, the use of thresholds is not appropriate in patients with active bleeding or severe hemolysis; clinical judgment must be used regarding the urgency of transfusion in such settings. (See "Indications and hemoglobin thresholds for RBC transfusion in adults".)

●Emergency release blood is typically O negative – Blood designated for emergency release is typically group O, RhD-negative. At many institutions, group O, RhD-positive RBC units may be used for all patients, including females who are beyond childbearing age or when the usage is expected to be very high. The designated RBC units are generally stored separately from other RBC units to allow rapid access and to avoid potential mistransfusion of non-group O blood. Specialized modifications may not be available (eg, there may not be irradiated units).

Patients with a confirmed ABO type on a current specimen can receive uncrossmatched, type-specific blood for emergency release.

Blood is infused slowly while carefully observing the patient for signs and symptoms of an acute hemolytic transfusion reaction (fever, chills, respiratory distress, chest or back pain). (See "Hemolytic transfusion reactions", section on 'Acute hemolytic transfusion reactions' and "Approach to the patient with a suspected acute transfusion reaction", section on 'Immediate actions (all patients)'.)

●Maximize tolerance of hypoxia – The patient's ability to withstand severe anemia while awaiting transfusion can be maximized by enforced bed rest to decrease oxygen demand and by administration of supplemental oxygen to optimize dissolved oxygen in the plasma. In critical situations, other options may be available, as described separately. (See "Approach to the patient who declines blood transfusion", section on 'The actively bleeding patient'.)

●Protocols for documentation and follow-up – Protocols to be followed during use of emergency release blood include:

•Written indication on the released unit stating that compatibility testing has not been completed.

•Written confirmation by the treating clinician that emergency release blood is required.

After the blood has been released, compatibility testing can be performed on a pretransfusion sample from the patient, if available. This is useful both for crossmatching subsequent units and for notification of the treating clinician of any unexpected results from the antibody screen and/or incompatibilities between the emergency release unit(s) and the recipient. This information is also recorded in the medical record.

●Risk of hemolytic reactions – The risk for hemolytic transfusion reactions is relatively low with emergency release RBC transfusions (released in the setting of no or incomplete pretransfusion testing), as demonstrated in the following studies:

•In a retrospective review of 1847 emergency release RBC units transfused to 935 patients, there were four acute hemolytic transfusion reactions (0.4 percent), including one fatal reaction due to anti-KEL1 [4].

•In a retrospective review of 1444 emergency release RBC units transfused to 1407 patients, seven patients developed an antibody that could be attributed to the transfusion, and 15 patients developed an antibody after additional RBC transfusions, for an overall rate of alloimmunization rate of 3 percent [5]. There were 32 reported suspected transfusion reactions, including one delayed serologic reaction; there were no acute hemolytic transfusion reactions.

•In a retrospective review of 1002 emergency release RBC units transfused to 262 patients, there was one hemolytic transfusion reaction (risk, <1 percent), which was due to preexisting non-ABO antibodies (anti-c and probable anti-Jk^a), and seven transfusions of antigen-incompatible RBCs (3 percent) [6].

●Emergency release massive transfusion – Urgent/emergency massive transfusion is a special case that may occur in a patient with a massive hemorrhage. For massive transfusion, predetermined blood ordering protocols are used to provide sufficient numbers of RBC units and other blood components. Many institutions have massive transfusion protocols for specific clinical settings such as labor and delivery. This subject is discussed separately. (See "Massive blood transfusion".)

EVALUATION/COMMON QUESTIONS

Communication — Communication between the clinical team and the transfusion medicine service/blood bank is essential to assist with:

●Emergency release blood if needed. (See 'Emergency release blood for life-threatening anemia or bleeding' above.)

●Facilitating the evaluation. (See 'Evaluation/common questions' above.)

●Setting expectations about the urgency with which transfusion is needed and the likely time frame required for the full pretransfusion evaluation.

Patient history — The history is very important and may provide clues to the cause of a positive antibody screen [7].

●Increased likelihood of autoantibodies:

•Autoimmune hemolytic anemia (AIHA)

•Conditions associated with AIHA:

-Chronic lymphocytic leukemia (CLL) or other lymphoproliferative disorder

-Systemic lupus erythematosus (SLE)

-Immunodeficiency disorder, including human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS)

•Medications that can cause drug-induced AIHA [8,9]:

-Methyldopa

-Procainamide

-Fludarabine

●Increased likelihood of alloantibodies:

•Pregnancy (even if termination or miscarriage occurred)

•Previous transfusion (at any time)

•Sharing of needles

•Racial or ethnic background that is underrepresented in the blood donor pool

•Family history of difficulty finding compatible blood (suggests possible alloantibody against a high-frequency antigen)

●Other relevant information:

•Medications used to treat the conditions associated with AIHA, even if the patient was unaware of the diagnosis.

•History of hematopoietic stem cell transplantation. (See 'Allogeneic hematopoietic stem cell transplantation recipients' below.)

•Therapeutic monoclonal antibodies that interfere with the antibody screen. (See 'Monoclonal antibodies that affect the antibody screen' below.)

Previous antibody screens should be reviewed if available.

Serologic testing

Panagglutination — Panagglutination refers to the situation when patient plasma/serum reacts with all of the screening cells and panel reagent RBCs identification cells on the initial antibody evaluation.

Causes of panagglutination are summarized in the table (table 1).

In one series involving 52,000 antibody identifications performed over a seven-year period, there were 3124 cases of panagglutination (6 percent of all identifications) [10]. Most (3002) were associated with a positive autocontrol test; 122 had a negative autocontrol.

●Bombay phenotype – The Bombay phenotype affects the ABO blood group; it is found almost exclusively in individuals from India. Bombay phenotype presents as panagglutination on the initial testing that reacts with all reagent RBCs except for the autocontrol.

Bombay phenotype is due to the absence of a fucosyltransferase enzyme; because of this, the H antigen cannot be produced. H is the antigen on which A and B blood group antigens are added. Individuals with the Bombay phenotype make alloantibodies to the H antigen on group O RBCs as well as to A and B antigens; consequently, they can only receive blood from other individuals with the Bombay phenotype (H antigen-negative). (See "Red blood cell antigens and antibodies", section on 'ABO antigens'.)

●Para-Bombay phenotype – Like the Bombay phenotype, individuals of the para-Bombay phenotype lack the fucosyltransferase necessary for the H antigen on RBCs. Depending on their secretor status (whether they have the fucosyltransferase responsible for H, A, and B antigens in secretions and plasma), there may be low level presence of H, A, and B antigens adsorbed onto their circulating RBCs. However, pretransfusion serologic evaluation of patients with the para-Bombay phenotype will often present similarly to those with the Bombay phenotype, demonstrating panagglutination on the antibody screen. These patients should be transfused with Bombay-phenotype RBCs (H antigen-negative).

Extended RBC phenotype — Use of extended RBC phenotyping is institution-dependent.

Antibody elution — The figure illustrates the antibody elution technique (figure 1).

RBC genotyping — RBC genotyping involves testing for the genetic variants that encode specific blood group antigens. RBC genotyping is not used routinely, but its role is evolving, especially to prevent alloimmunization in certain populations. Its use is institution dependent. (See "Pretransfusion testing for red blood cell transfusion", section on 'RBC genotyping'.)

Genotyping can be especially useful for individuals with one or more of the following:

●High likelihood of alloimmunization (eg, multiply transfused individuals)

●Blood types that are not well represented in the donor pool

●Confusing serologic results, such as due to a combination of alloantibodies and autoantibodies with autoimmune hemolytic anemia

Genotyping in individuals with sickle cell disease is discussed separately. (See "Red blood cell transfusion in sickle cell disease: Indications and transfusion techniques", section on 'Genetic RBC antigen typing'.)

Individuals who have been RBC genotyped still require crossmatching. (See 'Standard type/screen/crossmatch' above.)

Identifying compatible units — The extent of testing of the patient may be affected by the urgency of transfusion, the findings on initial testing, and the complexity of the serologic picture. Identification of compatible units of RBCs also depends on these factors, especially when multiple alloantibodies or an alloantibody to a high-frequency antigen has been identified. In such cases, it may be difficult to readily find compatible units in the donor pool.

The following general principles are followed by the transfusion service:

●History – If the patient has never been exposed to any foreign RBCs in the past (never pregnant, never transfused), it is highly unlikely that an alloantibody could be present. If needed urgently, transfusion can proceed more quickly with ABO matched blood, and the remainder of the work-up can be completed after issue of the unit of blood.

●Panagglutinin

•Determine whether the reactivity is similar to all reagent cells or variable (table 1). Similar reactivity suggests an antibody to a high-frequency antigen, while variable reactivity suggests multiple alloantibodies.

Extended RBC phenotyping and/or RBC genotyping can be used to determine the recipient's type and to select phenotypically matched cells for crossmatching [11,12]. (See 'Extended RBC phenotype' above and 'RBC genotyping' above.)

•Determine whether the autocontrol and direct antiglobulin test (DAT, direct Coombs test) are positive or negative. A positive autocontrol and positive DAT suggest an autoantibody. This does not rule out the possibility of an alloantibody as well, but it suggests next steps to remove the autoantibody. A negative autocontrol and negative DAT suggest an alloantibody.

●Positive DAT – A positive DAT can be nonspecific and does not necessarily mean the transfusion recipient has AIHA; however, it suggests the recipient has an autoantibody that could interfere with identification of alloantibodies. Approximately 8 to 9 percent of hospitalized patients may have a positive DAT [13]. The prevalence of AIHA is discussed separately. (See "Warm autoimmune hemolytic anemia (AIHA) in adults", section on 'Epidemiology'.)

•Use autoadsorption to remove the autoantibody from the recipient's plasma, then use the adsorbed plasma to crossmatch donor RBCs (figure 2) if the patient has not been transfused in the prior three months and has a high enough hemoglobin (>8 g/dL). (See 'Standard type/screen/crossmatch' above.)

•Use alloadsorption if the patient has been transfused in the prior three months or their hemoglobin is ≤8 g/dL.

•Use the ethylenediaminetetraacetic acid (EDTA)-glycine acid method or chloroquine method to dissociate the autoantibody from the recipient's RBCs, then perform a forward type. (See 'Standard type/screen/crossmatch' above.)

If the patient's extended RBC phenotype is not known, partially phenotypically matched RBCs for Rh and Kell may be provided to reduce the risk of alloimmunization and delayed hemolytic and/or serologic transfusion reactions.

CAUSES OF A POSITIVE ANTIBODY SCREEN

Alloantibodies — Most alloantibodies against RBC antigens are induced via exposure to foreign RBCs by transfusion or pregnancy. Particular difficulty in providing compatible blood is encountered when there are multiple alloantibodies or an alloantibody directed against a high frequency RBC antigen expressed on the majority of donor RBCs [14-16].

Common alloantibodies — Whenever the antibody screening test is positive, the blood bank will perform studies to identify the specificity of the offending antibody or antibodies (figure 1).

The most common alloantibodies detected are anti-E, anti-Le(a), anti-K, anti-D, and anti-Le(b) [17]. Only antibodies associated with hemolytic transfusion reactions or hemolytic disease of the fetus and newborn (HDFN) are considered clinically significant.

●These are common clinically significant alloantibodies [18]:

•ABO (A, B)

•Rh (D, C, c, E, e)

•Duffy (Fya, Fyb)

•Kidd (Jka, Jkb)

•Kell (K, k)

•SsU (S, s, U)

•Lutheran (Lub)

In four series, the most frequent clinically significant alloantibodies encountered included anti-E, anti-K, anti-c, anti-Jk(a), and anti-Fy(a) [16,19-21].

●These antibodies are rarely or never considered to be clinically significant:

•Lewis (Lea, Leb)

•MN

•Lutheran (Lua)

•P1

•Xga, Cartwright (Yta)

•Bg

•York (Yka)

•Chido/Rodgers (Cha/Rga)

•Sda

•HTLA (high titer low avidity)

The titer of some alloantibodies wanes over time and may fall completely below the threshold of detection by routine antibody screening techniques [22]. This is a less common finding with antibodies directed against Rh antigens but is frequent among those developing anti-Jk(a), consistent with the large relative number of delayed hemolytic transfusion reactions mediated by anti-Jk(a). As an example, in a long-term study of the transfusion records of 304 male veterans with one or more alloantibodies and at least one follow-up alloantibody screening test, anti-D and anti-C alloantibodies were persistent in 94 and 86 percent, whereas anti-Jk(a) alloantibodies were persistent in only 21 percent [23].

Alloantibodies to high frequency antigens — High frequency antigens (also called high incidence antigens) are RBC antigens that are present on a very high percentage of individuals in the population (often, >99 percent). If a patient whose RBCs lack a high frequency antigen is exposed to that antigen, an alloantibody can form. Since almost all donor RBCs will express the antigen, it becomes very challenging to find crossmatch compatible blood. A positive family history of difficulty in finding compatible blood for transfusion is a valuable clue to this diagnosis.

Some high frequency antigens are clinically significant (capable of causing a hemolytic transfusion reaction) and some are not.

Examples of high frequency clinically significant antigens include:

●U

●Vel

●KEL2

●Lu(b)

In contrast, the Cartwright antigen Yt(a) is a high frequency antigen that has variable clinical significance. (See "Red blood cell antigens and antibodies", section on 'Clinically significant (rare)' and "Red blood cell antigens and antibodies", section on 'Limited or no clinical significance'.)

Identification of antibodies to high frequency antigens requires rare reagent RBCs lacking these antigens. These reagent RBCs are not routinely available at hospital blood banks; a reference laboratory must be consulted, and this can result in significant delays (as long as days to weeks to perform testing and find a compatible unit). If transfusion is needed urgently, blood can be issued under a nonstandard release that is approved by a clinician or as an emergency release. (See "Pretransfusion testing for red blood cell transfusion", section on 'Emergency release blood for life-threatening anemia or bleeding'.)

Approaches to identifying compatible blood include:

●Tests such as the monocyte monolayer assay (MMA) and the chemiluminescence test have been used in predicting the clinical significance of RBC alloantibodies [24-26]. Both tests are only available in reference laboratories. The MMA is a highly technical and time-consuming test. If the alloantibody is not clinically significant, multiple random units can be crossmatched with the patient's plasma. The most compatible units can be safely transfused.

●A regional blood center may use a rare donor file to identify compatible donors and compatible units.

●First-degree relatives may be tested to determine if one or more of them also lack the same RBC antigen in question and may serve as a blood donor.

●For elective surgery, the patient may provide autologous donations. (See "Surgical blood conservation: Preoperative autologous blood donation".)

Multiple alloantibodies — Some individuals may develop alloantibodies to multiple RBC antigens, especially heavily transfused patients and those who are negative for RBC antigens that are present in the donor pool. Patients with sickle cell disease can receive multiple transfusions and often have multiple alloantibodies. (See "Transfusion in sickle cell disease: Management of complications including iron overload", section on 'Alloimmunization and hemolysis'.)

Alloimmunization rates differ by disease state [27]. The likelihood of forming multiple alloantibodies in previously transfused individuals with initial alloimmunization is as high as 22 to 33 percent in hematology-oncology patients and 20 to 25 percent in non-hematology-oncology patients, as illustrated in the following studies:

●A 2020 systematic review identified common alloantibodies that could be reduced with greater antigen matching [28]. Alloimmunization in sickle cell disease is discussed in more detail separately. (See "Red blood cell transfusion in sickle cell disease: Indications and transfusion techniques", section on 'Supporting evidence for RBC antigen matching' and "Transfusion in sickle cell disease: Management of complications including iron overload", section on 'Alloimmunization and hemolysis'.)

●A 2009 study involving 18,750 individuals seen at a Veterans Affairs medical center identified 433 with alloimmunization, of which 96 (22 percent) had more than one alloantibody [29]. Common pairs of alloantibodies included anti-K with anti-E, anti-D with anti-C, and anti-E with anti-c.

Autoantibodies — The main concern with autoantibodies is that they may obscure the presence of a clinically significant alloantibody during the antibody screen.

Autoimmune hemolytic anemias — Autoimmune hemolytic anemia (AIHA) can be due to a warm alloantibody, a cold agglutinin, or a drug-induced antibody. In AIHA, the direct antiglobulin test (DAT, direct Coombs test) is typically positive because the patient's RBCs are coated with the autoantibody, as illustrated in the figure (figure 3).

●Warm AIHA – Warm AIHA is the most common type of AIHA; it is typically caused by immunoglobulin G (IgG) autoantibodies that react at body temperature. It can be primary or secondary to an autoimmune, immunodeficiency, or lymphoproliferative disorder, or autoantibodies can be clinically silent. (See "Autoimmune hemolytic anemia (AIHA) in children: Classification, clinical features, and diagnosis" and "Warm autoimmune hemolytic anemia (AIHA) in adults".)

Common RBC antigens in warm AIHA include Rh antigens and glycophorins [16]. The DAT is typically positive for IgG. (See "Warm autoimmune hemolytic anemia (AIHA) in adults", section on 'Antibody and antigen characteristics'.)

●Cold agglutinins – Cold agglutinin disease is less common. It is typically caused by IgM autoantibodies that react at colder temperatures. It can be primary or secondary to an autoimmune or lymphoproliferative disorder, or cold agglutinins can be clinically silent. (See "Cold agglutinin disease".)

Common RBC antigens in cold agglutinin disease include I and i. The DAT is typically positive for complement but not for IgG. (See "Cold agglutinin disease", section on 'RBC antigens' and "Cold agglutinin disease", section on 'Cold agglutinins'.)

●Drug-induced AIHA – There are several mechanisms of drug-induced AIHA. Commonly implicated drugs are discussed separately. The DAT is typically positive for IgG. (See "Drug-induced hemolytic anemia".)

Evidence of successful transfusion in individuals with AIHA comes from studies such as the following:

●In a 1993 review of 53 patients who received blood transfusions for decompensated AIHA, there were no definite increases in hemolysis, even when the transfused RBCs were serologically incompatible [14].

●In a 1988 review of 2149 transfusion recipients with autoantibodies, concurrent alloantibodies were identified in 294 (14 percent), most frequently anti-E and anti-K [16]. Out of 7052 RBC units issued for 1685 patients, no hemolytic transfusion reactions were reported, and it was concluded that after appropriate investigation, RBC units can be safely given to patients with autoantibodies, even in serologically complex cases.

As discussed above, no patient with autoantibodies should be denied a lifesaving transfusion because of difficulties in finding compatible blood for transfusion. (See 'Emergency release blood for life-threatening anemia or bleeding' above.)

AIHA and concurrent alloantibodies — The greatest concern in transfusing patients with AIHA involves detection of concurrent alloantibodies in the patient's plasma [7]. The presence of an autoantibody makes alloantibody detection and identification more difficult. As an example of their prevalence in this setting, one study indicated that 31 percent of subjects with a warm autoantibody also had at least one alloantibody.

Several techniques can be used to detect an underlying alloantibody [30-33]. For patients not recently transfused and with a sufficiently high hemoglobin, the autoadsorption method (figure 2) can be used [15,30,34,35]. The autoantibody coating the patient's RBCs is stripped off the cells by an elution method. The patient's own eluted RBCs are then mixed with the patient's plasma. These cells will adsorb the autoantibodies from the plasma, but alloantibodies cannot be adsorbed since the patient does not have the corresponding antigen on their own RBCs. The adsorption step is repeated multiple times (typically three to five times) until all of the autoantibody is removed from the plasma/serum. If the patient has an alloantibody, it can now be detected and identified. Use of autoadsorption may be precluded by the following:

●It cannot be used in a recently transfused patient.

●The autoantibody may be difficult to adsorb, either because the antibody titer is too high or the patient's RBCs are too fragile for any manipulation.

●It is very time consuming.

●There is often a requirement for more blood samples from an already anemic patient.

When autoadsorption is not possible, an alternate method called allogeneic adsorption is performed using allogeneic RBCs of varying phenotypes to adsorb autoantibody from the patient's plasma, leaving behind the alloantibody, if present, for subsequent identification [30,33-36]. Details of these methods are beyond the scope of this discussion.

Monoclonal antibodies that affect the antibody screen — Certain monoclonal antibodies bind to RBCs and cause panagglutination.

Anti-CD38 (daratumumab, isatuximab) — CD38 is a membrane-bound enzyme that degrades nicotinamide adenine dinucleotide (NAD), which serves as a cofactor (an electron donor) in redox reactions and is involved in calcium signaling. It is present on RBCs and is also highly expressed on the surface of multiple myeloma cells. Daratumumab and isatuximab are monoclonal antibodies directed against CD38 that are used in multiple myeloma treatment.

Because all RBCs express CD38, these antibodies can react with all reagent RBCs used in antibody screening and cause panagglutination. This makes it challenging to identify compatible blood and may obscure a clinically significant alloantibody. (See 'Panagglutination' above.)

The generally accepted approach to patients who have received (or will receive) an anti-CD38 monoclonal antibody and are likely to require transfusions includes the following:

●Inform the appropriate staff – Inform the transfusion medicine service or blood bank. Many institutions have a reminder system (eg, pharmacy alert) to make sure this happens the first time the anti-CD38 agent is ordered for a patient. (See 'Communication' above.)

●Preliminary type and screen – Perform a type and screen prior to administering the anti-CD38 therapy. This can be used to determine whether any alloantibodies or autoantibodies are present and will facilitate matching for ABO, Rh, and Kell for any future transfusions. (See "Multiple myeloma: Administration considerations for common therapies", section on 'Anti-CD38 monoclonal antibodies'.)

●Crossmatching – If a transfusion is needed, notify the transfusion medicine service or blood bank. They will review the results of the baseline (pre-anti-CD38) sample and the current pretransfusion sample to determine whether anti-CD38 is interfering with the antibody screen.

•Negative antibody screen – If the pretransfusion sample has a negative antibody screen, this suggests there is no interference by the anti-CD38 monoclonal antibody. Previously documented alloantibodies should be respected.

•Panagglutination – If the pretransfusion sample demonstrates panagglutination, this suggests that the anti-CD38 monoclonal antibody is interfering with the antibody screen (table 1). The following methods can be used to circumvent the interference [37,38]:

-DTT – Dithiothreitol (DTT) is a reducing agent that breaks disulfide bonds and destroys the CD38 antigen on RBCs [39]. DTT also destroys the Kell antigen. If DTT-treated reagent RBCs are used, it is not possible to exclude anti-K1. Thus, K1-negative units should be provided unless the individual is known from previous testing to express K1 on their own RBCs [40].

-LISS – Low ionic strength saline (LISS) decreases the tonicity of the antibody screen solution and in some cases eliminates the interference by anti-CD38.

-Anti-daratumumab F(ab) fragments – For daratumumab, anti-daratumumab F(ab) fragments can be used to block CD38 and prevent free daratumumab in the recipient's plasma/serum from binding to the reagent RBCs [41]. (See "Overview of therapeutic monoclonal antibodies", section on 'Fab fragments and single-chain antibodies'.)

When tested in a gel-based antibody detection system using several samples with known clinically relevant antibodies to RBC antigens, this reagent prevented daratumumab interference and allowed detection of the clinically important anti-RBC antibodies. This approach is promising, and similar reagents could be produced for other therapeutic monoclonal antibodies; however, broad application awaits commercial availability of the relevant F(ab) fragments and the validation with other RBC antibody detection systems.

The choice among these methods is determined by the local blood bank or testing laboratory. Samples can be sent to a reference laboratory if these methods are unavailable.

The likelihood of a panagglutinin in an individual treated with anti-CD38 therapy may be high (91 of 91 patients in one report) [38]. Among 65 who also had a pre-anti-CD38 treatment sample obtained, five (8 percent) had a preexisting alloantibody, perhaps due to the high likelihood of prior transfusion in this population.

The effect of anti-CD38 antibodies can potentially last as long as six months following administration, although there is no well-established typical range.

Anti-CD47 — CD47 is an integral glycoprotein associated with the Rh protein complex and is highly expressed on RBCs [42,43]. It encodes a "do not eat me" signal that prevents macrophage-mediated phagocytosis and is thought to contribute to immune evasion by certain cancer cells. Therapeutic monoclonal antibodies targeting CD47 (eg, magrolimab, HuFF9-G4) are under investigation [44]. (See "Principles of cancer immunotherapy", section on 'Other potential targets'.)

Treatment of a transfusion recipient with an anti-CD47 monoclonal antibody may interfere with ABO typing of non-group O patients (ie, interfere with reverse typing and antibody screening) and may cause panagglutination. (See 'Standard type/screen/crossmatch' above and 'Panagglutination' above.)

Interference by an anti-CD47 monoclonal antibody may not be readily resolvable, and it is imperative for transfusion services to develop robust algorithms for serologic evaluation of patients receiving these therapies, including baseline testing and consideration for extended RBC phenotyping and/or genotyping.

Allogeneic hematopoietic stem cell transplantation recipients — Patients who have undergone ABO mismatched hematopoietic stem cell transplantation (HSCT) may have a different ABO type than is indicated in their historical medical records. This can cause confusion in ABO typing and can result in positive crossmatches depending on the degree of donor chimerism.

For example, type O patients who receive stem cell transplants from type A donors will have type A RBCs after engraftment, but they may have anti-A in their plasma/serum produced by their residual lymphocytes, either transiently or persistently, despite converting to blood type A. If the patient's plasma is crossmatched against a type A donor unit, a positive result will be obtained. In this case, type O blood should be provided. (See "Donor selection for hematopoietic cell transplantation", section on 'ABO and Rh status'.)

ABO discrepancies — Even though ABO typing is a simple test with 100 percent accuracy being required, there are occasional difficulties in determining the patient's true ABO type. Such ABO discrepancies may cause positive crossmatches with some or all units of donor blood. Most issues can be resolved after careful laboratory investigation. If the patient requires urgent transfusion, type O blood can be safely transfused. (See 'Emergency release blood for life-threatening anemia or bleeding' above.)

In addition to ABO mismatched allogeneic hematopoietic stem cell transplant (see 'Allogeneic hematopoietic stem cell transplantation recipients' above), ABO discrepancies can also be caused by:

●Type A subgroups – Among all type A patients, 80 percent are type A1 and 20 percent are either type A2 or another subtype of A. Patients who are a subgroup of A can have anti-A1 antibodies that will result in positive crossmatches with A1 donor units. Type O blood should be provided to these patients. DNA-based red cell genotyping may be of value in such cases by providing optimally matched donations [45].

●Acquired B phenotype – When group A patients are infected with certain Gram-negative bacteria, bacterial enzymes can remove an N-acetyl group from N-acetylgalactosamine on the red cell surface, resulting in galactosamine that is similar to the group B antigen's terminal galactose. ABO typing will show a weak B antigen on the patient's RBCs with strong anti-B in the plasma (normally type B patients should not have anti-B in their plasma). If the patient's plasma is crossmatched with type B donor units, the results will be incompatible. The patient's RBCs can be correctly typed by acidifying the anti-B typing reagent, which will not recognize this "acquired B antigen" and resolve the problem. (See "Red blood cell antigens and antibodies", section on 'ABO antigens'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Transfusion and patient blood management".)

SUMMARY AND RECOMMENDATIONS

●Reasons for serologic complexity – Standard pretransfusion testing involves an ABO and RhD type and an antibody screen. Crossmatching is done with a specific unit of red blood cells (RBCs) if transfusion is required. RBC genotyping is also available in limited settings. (See 'Standard type/screen/crossmatch' above and "Pretransfusion testing for red blood cell transfusion".)

Reasons for serologic complexity include alloantibodies to high-frequency (extremely common) RBC antigens, multiple alloantibodies, and autoantibodies. Less common causes of serologic complexity include therapeutic monoclonal antibodies that bind to RBCs and changes in blood type due to allogeneic hematopoietic stem cell transplant (HSCT), type A subgroups, or acquired B phenotype. (See 'Reasons for positive antibody screen or incompatible crossmatch' above.)

●Emergency release blood – For patients with life-threatening anemia or bleeding in whom compatible blood is not readily available, we recommend emergency release blood rather than delaying transfusion (Grade 1B). Patients should never be denied a lifesaving transfusion because of difficulties in finding compatible blood. (See 'Emergency release blood for life-threatening anemia or bleeding' above.)

●Evaluation – Communication among teams is essential, and the patient history is critical for determining the likelihood of alloantibodies. If the transfusion recipient has never been pregnant or transfused, alloantibodies are unlikely, and ABO matched blood can be used if needed urgently.

Serologic testing with an extended panel of reagent RBCs can determine whether panagglutination is present and its characteristics; the table lists causes of panagglutination (table 1). Other methods including direct antiglobulin testing (DAT, direct Coombs test), antibody elution, genotyping, and others can be used. If alloantibodies are possible and the patient's extended RBC phenotype is unknown, RBCs matched for Rh and Kell may be provided.  

●Causes of a positive antibody screen

•Alloantibodies – Common alloantibodies, common antigens that elicit alloantibodies, and characteristics of multiple alloantibodies are discussed above. (See 'Alloantibodies' above.)

•Autoantibodies – Not all autoantibodies indicate autoimmune hemolytic anemia (AIHA), but they can interfere with the antibody screen and may obscure clinically significant alloantibodies. Types of AIHA (warm AIHA, cold agglutinin disease, and drug-induced AIHA) are discussed above. (See 'Autoantibodies' above.)

●Other considerations – Special considerations apply to individuals treated with certain therapeutic monoclonal antibodies (mAbs), who have undergone allogeneic HSCT, who have other reasons for ABO discrepancies, or who have received intravenous immune globulin (IVIG) or anti-D immune globulin.

•Anti-CD38 or anti-CD47 mAbs – These antigens are present on RBCs; evaluation requires a pretreatment patient sample and additional testing in some cases. (See 'Monoclonal antibodies that affect the antibody screen' above.)

•Allogeneic HSCT recipients – A change in blood type may necessitate additional testing and/or use of type O blood. (See 'Allogeneic hematopoietic stem cell transplantation recipients' above.)

•ABO discrepancies – These can also occur with certain type A subgroups and in group A patients with Gram-negative bacterial infections. (See 'ABO discrepancies' above.)

•Receipt of IVIG or anti-D immune globulin – (See "Pretransfusion testing for red blood cell transfusion", section on 'Therapeutic antibodies (IVIG, anti-D, monoclonals)'.)

ACKNOWLEDGMENTS

The UpToDate editorial staff acknowledges Dennis Goldfinger, MD (deceased), who contributed to earlier versions of this topic review.

The UpToDate editorial staff also acknowledges extensive contributions of Qun Lu, MD, and Arthur J Silvergleid, MD, to earlier versions of this and many other topic reviews.