Shining a light on AHG “blind” spot(s)

The antihuman globulin (AHG) test was rediscovered at Cambridge University by Robin Coombs, Arthur Mourant, and Rob Race in 1945.1 The so-called “Coombs test” has since been used to identify immunoglobulin (Ig)G-reactive antibodies using the direct test for the elucidation of autoantibodies sensitizing patients' red blood cells (RBCs) in vivo and an indirect test to detect alloantibodies ex vivo. Originally termed the Coombs test, it was renamed the direct antiglobulin test (DAT) and indirect antiglobulin test (IAT) after the realization that the test had actually been first reported in 1908 by Carlo Moreschi, an Italian clinician-scientist.2, 3 The DAT and IAT were used for years, composed simply of a polyclonal anti-IgG preparation, but it became apparent that some antibodies were capable of binding complement components to RBCs,4, 5 and this led investigators and commercial companies to produce so-called “broad-spectrum” AHG that contained polyclonal anti-IgG plus anti-complement. Many companies began to market broad-spectrum AHG for use in transfusion medicine during the 1970s. The intense competition to gain market share on the “optimal” broad spectrum AHG led to seminal work by Garratty and Petz to question the make-up of AHG, in particular, the right mix of anti-IgG and anti-complement for optimal detection of alloantibodies, both in DAT and in IAT.6 This first comparative study of AHG found that the commercial products varied substantially in their ability to detect weak IgG levels, and most failed to have adequate anti-complement reactivity.6, 7 These reports led manufacturers of AHG to either greatly improve their AHG or withdraw from the competitive market. The studies by Garratty and Petz led to polyclonal AHG that was optimized for the detection of potentially clinically significant antibodies, primarily using tube methods with albumin or enzyme enhancement common at the time. Today, AHG is manufactured by fewer commercial companies compared to 40 years ago (see the cover of TRANSFUSION). Years of debate ensued over whether AHG should contain anti-complement as well as anti-IgG. Detection of RBC-bound complement in the DAT is important in the differential diagnosis of autoimmune and drug-induced immune hemolytic anemias, but it can also be a sensitive test for the detection of complement-binding alloantibodies in compatibility testing. Support for inclusion of anti-complement in AHG followed reports of some IgG complement binding antibodies, especially those in the Kidd system, causing hemolytic transfusion reactions but reacting poorly or being missed with anti-IgG and only detected by anti-complement.8 (Issitt professed “In order that the presence of these antibodies does not go undetected, it is obviously necessary that the AHG serum contain anti-complement antibodies as well as anti-IgG.”9) Anti-complement was considered a very important component of AHG in the 1970s and 1980s. Fresh complement was often added to stored serum samples before testing. Monoclonal antibodies (MoAbs) for use in blood banking were introduced in the 1980s following the work of Köhler and Milstein.10 The first antibodies were produced to the blood group A and B antigens.11 The advantage of MoAbs is the ability to produce, in culture, a reagent with identical properties and of subclass IgM for direct agglutination. However, the antibodies recognize single epitopes as opposed to polyclonal antisera that recognize multiple epitopes and this narrower and more precise specificity can be a limitation. As MoAbs to various blood group antigens became available, so did a monoclonal anti-IgG for use in AHG testing. Mouse monoclonal anti-human C3b and anti-human C3d were also developed. These led the way to the manufacture of three specific reagents: polyspecific AHG containing both anti-IgG and anti-C3d and separate anti-IgG and anti-C3. Prospective studies of complement-binding-only alloantibodies of potential clinical significance revealed that these antibodies were rare12 and the anti-complement contained in polyspecific AHG led to many unwanted “nonspecific” reactions, likely due to underlying clinically insignificant cold agglutinins. The sensitivity of the IAT had also increased with the use of enhancement media including low-ionic-strength saline, polyethylene glycol, and gel and solid-phase testing, mitigating concerns for missing clinically significant alloantibodies in the absence of anti-C3 in AHG reagents. Today, most routine antibody screening and crossmatching use only anti-IgG reagents in IAT, although broad-spectrum AHG continues to be used for DAT and in reference laboratories for complex antibody investigations. The only FDA-licensed monoclonal anti-human IgG reagent for use in blood bank testing in the United States is Gamma-clone anti-IgG. It is a murine monoclonal IgM antibody reactive with an epitope on the CH3 domain of the Fc region of human IgG. All others are pools of rabbit anti-human IgG, but differ in that some are heavy-chain specific and others have reactivity against light chains and hence may also detect IgA or IgM sensitization. Although there are four human IgG subclasses—IgG1 through IgG4—additional variation in each constant region gives rise to at least 29 isoallotypes.13 The reactivity of anti-IgG used in transfusion service laboratories with different isoallotypes has not previously been investigated. In this issue of TRANSFUSION, Howie and colleagues14 present a novel approach using recombinant antibodies of subclass IgG1, IgG2, IgG3, and IgG4 (that target the same K1 antigen epitope) and further engineer the constant region to represent human isoallotypes to assess the sensitivity and specificity of anti-IgG preparations. Gamma-clone monoclonal anti-IgG and Ortho rabbit anti-IgG were tested for the ability to detect different human IgG isoallotypes bound to RBCs expressing the K1 antigen. Testing revealed that the mouse monoclonal anti-IgG had equivalent sensitivity for subclass IgG1, IgG2, and IgG3 but no detectable reactivity with any IgG4. Indeed, the inability of the Gamma-clone anti-IgG to detect IgG4 antibodies is indicated in the product insert and is well known to experienced serologists, although may be less well known to other transfusion professionals. This inability to detect IgG4 has not been regarded as a significant limitation, as antibodies of IgG4-only subclass are uncommon, have not been associated with acute hemolytic events, and often have high-titer low-avidity characteristics associated with clinically insignificant reactivity. As with all generalizations, there are rare exceptions. As pointed out by Howie and colleagues,14 IgG4 only anti-JMH15, 16 and anti-Ch17 have caused transfusion reactions. Of greater interest, the authors found that human IgG3-03 and IgG3-13 isoallotypes, found in highest frequencies in ethnic African or Middle Eastern, were not detected. Similar to the historical discussions regarding the anti-complement component of polyclonal AHG, the report by Howie and colleagues14 brings to light a potential shortcoming of a monoclonal anti-IgG reagent. To use the words of the authors, the “findings demonstrate ‘blind spots’ in isoalloantibody detection” when using Gamma-clone monoclonal anti-IgG that does not occur when using Ortho polyclonal anti-IgG (the only rabbit anti-IgG tested). Although the lack of detection of IgG4 antibodies is not considered clinically important for transfusion medicine, failure to detect antibodies of the IgG3-03 and IgG3-13 isoallotypes could have clinically significant sequelae. The frequency of IgG3-03 and IgG3-13 antibodies in the Caucasian population is low, but it has been estimated that these isoallotypes in African populations may reach 20% to 30%.18 Although alloantibodies to blood group antigens are rarely IgG3 alone and are most often IgG1 or IgG1 and IgG3, the failure to detect these IgG3 antibodies could have clinical significance in geographical areas where there is a large population of people with an African background. The number of antibodies that may have been missed is unknown, as this report is the first to suggest that a monoclonal anti-IgG does not detect two potentially important IgG3 isoallotypes. All transfusion medicine professionals are aware that there are instances when the IAT is negative but the transfusion, nevertheless, is less than efficacious and there are also situations with overt hemolysis seen in the absence of detectable alloantibody. The importance of this work is the demonstration that monoclonal anti-IgG will not detect all isoallotypes of IgG3 and provides the incentive, as did the anti-complement findings of the 1970s, for manufacturers of anti-IgG to determine that their product contains the necessary components directed to relevant human IgG subclasses and isoallotypes. It also reinforces the concept that polyclonal anti-IgG, which consists of many antibodies targeting multiple epitopes of human IgG subclasses may still be the gold standard. The article by Howie and colleagues14 provides a way to examine the anti-IgG isoallotype specificity of anti-IgG, which may become a standardized method for manufacturers to quality control their products. Lastly, this study supports the practice, common in experienced reference laboratories, of using more than one manufacturer's anti-IgG when investigating hemolysis in the absence of detectable antibody. The authors have disclosed no conflicts of interest. Donald R. Branch, BS, MT(ASCP), SBB, PhD1 e-mail: [email protected] Connie M. Westhoff, MT(ASCP), SBB, PhD2 1Canadian Blood Services Toronto, ON, Canada 2New York Blood Center New York, NY

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.