When an antibody screen is positive and the panel reactions sort by antigen, the protein blood group systems explain the pattern. Protein blood group antigens are inherited structures of the red-cell membrane. Their genes, carrier proteins, population frequencies, and susceptibility to reagent treatment help explain a patient's serologic pattern. Antibody significance then depends on thermal range, immunoglobulin class, complement activation, antigen density, prior exposure, and the survival of transfused cells.
The Rh system
Genes, proteins, and nomenclature
The Rh system, ISBT 004, contains 56 recognized antigens. Its five principal antigens are D, C, c, E, and e. RHD encodes RhD, while RHCE encodes one of four RhCE proteins carrying C or c together with E or e. The two genes lie close together on chromosome 1 and usually travel as a haplotype (one inherited set of linked alleles). RHAG on chromosome 6 encodes the associated glycoprotein required for normal Rh expression. RhD and RhCE are homologous, nonglycosylated 417-amino-acid proteins with 12 membrane-spanning domains.1-3
Four nomenclatures describe the same common antigens and haplotypes:
Scroll sideways to see all columns
| Nomenclature | Organizing principle | Example for DcE/dce |
|---|---|---|
| Fisher-Race | Names D, C/c, and E/e antigens | DcE/dce |
| Wiener | Assigns one symbol to each haplotype | R2r |
| Rosenfield | Records tested antigen presence or absence by number | Rh:1,−2,3,4,5 |
| ISBT | Uses system 004 and numbered antigen assignments | RH:1,−2,3,4,5 |
The lowercase d records the absence of D. The common Wiener mappings are R1 = DCe, R2 = DcE, R0 = Dce, Rz = DCE, r = dce, r′ = dCe, r″ = dcE, and ry = dCE. Gene symbols are italicized, protein and antigen symbols remain in plain type, and antibody names carry the anti- prefix.
Routine typing establishes a phenotype. A D+C+E−c+e+ phenotype may arise from DCe/dce or another haplotype pair that supplies the same five antigens. Population frequencies can support a probable genotype, while molecular testing can resolve an inherited allele or RHD zygosity when that result changes laboratory interpretation.1
Population haplotype frequencies also guide donor recruitment. The R0 (Dce) haplotype is more frequent among donors with African ancestry, so targeted recruitment can improve access to C-negative, E-negative units when D-positive support is appropriate. Rare-unit searches combine validated frequency data with the patient's complete antibody profile.1
D expression and variants
A complete RHD deletion is a common molecular basis for D-negative status in people of European ancestry. Other populations include RHD pseudogenes, RHD-RHCE hybrid alleles, and DEL alleles at different frequencies. The serologic label weak D therefore describes a reaction pattern with more than one possible molecular cause.1,3
Scroll sideways to see all columns
| Pattern | Molecular or expression basis | Laboratory consequence |
|---|---|---|
| Weak D types 1, 2, and 3 | Missense variants reduce D antigen density while retaining the complete D epitope set | Routine anti-D may be weak or negative, with stronger reactivity at the indirect antiglobulin phase |
| Partial D | One or more D epitopes are absent, often because of a hybrid or altered RHD allele | Exposure to conventional D-positive cells can stimulate anti-D against a missing epitope |
| DEL | D site density is below routine agglutination detection | Adsorption-elution or molecular testing may demonstrate RHD-associated expression |
| C in trans position effect | A C-bearing haplotype opposite RHD reduces D expression | Once confirmed as an isolated position effect, transfusion and Rh immune globulin decisions for the recipient follow D-positive status |
For Whole Blood donor labeling, 21 CFR 640.5(c) requires an initially negative anti-D result to receive additional testing that includes weak D. The product may be labeled Rh negative only when the further testing is negative. Either a positive direct anti-D test or a positive weak-D test supports Rh-positive labeling.4
Transfusion and obstetric Rh immune globulin decisions for recipients must distinguish serologic reactivity from genotype. The AABB and CAP joint statement recommends RHD genotyping for a pregnant patient or other female of childbearing potential with a serologic weak D phenotype. That 2015 statement supports D-positive management for weak D types 1, 2, and 3. AABB Association Bulletin 24-02 also lists weak D type 4.1 among variants not considered at risk for D alloimmunization. Because the two lists differ, type 4.1 and the other variants are handled according to the genotype report and the transfusion service's adopted policy. An unresolved variant is managed as being at risk until its status is established.5,6 A recipient with a partial D variant capable of anti-D formation is managed as D-negative and generally receives D-negative red cells.1,5
Rh antibodies and related phenotypes
Rh alloantibodies are usually immune IgG antibodies that react best at 37 °C and the antiglobulin phase. They commonly show dosage and enhanced reactions with papain- or ficin-treated cells. They usually cause extravascular red-cell clearance and can cause hemolytic transfusion reactions and hemolytic disease of the fetus and newborn. Among the common Rh antigens, the usual immunogenicity order is D, c, E, C, then e.1
Anti-G can resemble anti-D plus anti-C because G is present on nearly all D-positive cells and all C-positive cells. The distinction changes Rh immune globulin eligibility when a pregnant patient has anti-G without anti-D. Transfusion support for anti-G uses D-negative, C-negative red cells. The f antigen is a compound antigen expressed when c and e occur on the same RhCE protein, so anti-f can resolve haplotype arrangements. For example, it reacts with cells carrying the dce haplotype, as in DCE/dce. It is nonreactive with DcE/DCe cells, which carry c and e on opposite haplotypes.1
Rhnull red cells lack all Rh antigens. An RHAG defect produces the regulator type, while inactive Rh structural genes can produce the amorph type. Loss of the Rh complex changes membrane stability and can cause a compensated hemolytic anemia with stomatocytes. An immunized Rhnull patient can form anti-Rh29, an antibody to the total Rh complex, and needs Rhnull or other exceptionally rare compatible blood. Anti-Rh17 belongs to a different phenotype: D−− individuals, who express D but no C, c, E, or e, form it when immunized.1,13
Rhmod results from partial RHAG function. All Rh antigens are weakly expressed, and the red-cell membrane changes and compensated hemolysis are generally milder than in Rhnull. Broadly weakened Rh typing can therefore reflect an Rh-complex expression defect that warrants reference-laboratory investigation.1
MNS, Kell, Kidd, Duffy, and Lutheran
These systems account for many non-ABO/Rh antibodies found during pretransfusion testing. The carrier protein predicts several useful laboratory properties.
MNS
M and N are carried on glycophorin A, encoded by GYPA. S, s, and the high-prevalence U antigen are carried mainly on glycophorin B, encoded by GYPB. Homology between these genes permits crossing over and gene conversion, producing hybrid glycophorins, low-prevalence antigens, and phenotypes lacking a high-prevalence antigen.1,2
Anti-M and anti-N are commonly cold-reactive IgM antibodies and show dosage. Papain and ficin destroy their target structures. Reactivity at 37 °C or in the antiglobulin phase may indicate clinical significance. Anti-S, anti-s, and anti-U are commonly IgG and can cause hemolytic transfusion reactions and hemolytic disease of the fetus and newborn. The S−s−U− phenotype occurs most often in people with African ancestry and creates a rare-unit requirement when anti-U is present.1,2
Kell and Kx
The Kell system, ISBT 006, contains 38 recognized antigens. The KEL glycoprotein carries K, k, and other antithetical antigen pairs. K is a strong immunogen, and anti-K is usually immune IgG reactive at the antiglobulin phase. Anti-K can cause severe transfusion reactions and can suppress fetal erythropoiesis because Kell is expressed on erythroid precursors.1,2
Dithiothreitol and other sulfhydryl reagents disrupt Kell antigens. This property supports selected antibody investigations. The reagent also destroys or weakens other antigens, so every interpretation needs appropriate controls. A K0 person lacks Kell-system antigens and can form anti-Ku against an antigen present on almost all ordinary red cells.1,2
The Kx antigen belongs to the XK system and supports normal Kell expression. An XK variant can produce the X-linked McLeod phenotype, with absent Kx, weakened Kell antigens, acanthocytes, and reduced red-cell survival. K0 cells, in contrast, lack Kell antigens and retain strong Kx expression. Kx testing therefore distinguishes K0 from McLeod when Kell reactions are weak or absent. Reference testing defines the phenotype and antibody profile needed for compatible support.1,2,7
Kidd
The Kidd glycoprotein, encoded by SLC14A1, transports urea and carries Jka, Jkb, and Jk3. Kidd antibodies often show dosage, bind complement, and react more strongly with enzyme-treated cells. Their concentration can fall below routine detection and rise rapidly after re-exposure. This evanescence and anamnestic response make anti-Jka and anti-Jkb classic causes of delayed hemolytic transfusion reactions.1,2
Jk(a−b−) cells lack the Kidd urea transporter and resist rapid lysis in 2 mol/L urea. The phenotype is rare overall and occurs at a higher frequency in some Polynesian populations. A person with the null phenotype can form anti-Jk3 and require Jk(a−b−) units.1,2
Duffy
The Duffy glycoprotein is atypical chemokine receptor 1, encoded by ACKR1, and carries Fya, Fyb, and other Duffy antigens. Anti-Fya and anti-Fyb are usually IgG, react at the antiglobulin phase, and can cause acute or delayed hemolytic transfusion reactions and hemolytic disease of the fetus and newborn. Papain and ficin destroy Duffy antigen reactivity.1,2
An ACKR1 promoter variant common in people with African ancestry silences erythroid expression and produces Fy(a−b−) red cells while retaining expression in other tissues. These individuals rarely form anti-Fyb because expression in other tissues supports immune tolerance. A true Duffy-null allele abolishes expression in all tissues and permits anti-Fyb formation, so molecular context changes antibody risk.1,2
Erythroid ACKR1 silencing markedly reduces susceptibility to Plasmodium vivax. Infection remains possible. A 2024 Ethiopian study found P. vivax in Duffy-negative participants, with infection substantially more frequent among Duffy-positive participants.8
Lutheran
The BCAM adhesion glycoprotein carries Lutheran antigens. Lua has low prevalence and Lub has high prevalence. Anti-Lua is commonly a naturally occurring IgM antibody with a characteristic loose, mixed-field appearance. Anti-Lub is usually immune IgG and may shorten transfused-cell survival.1,2
A recessive BCAM null phenotype lacks Lutheran antigens and can permit anti-Lu3 formation. A dominant KLF1 regulator variant produces the In(Lu) phenotype with markedly weakened Lutheran and selected other antigens. The two mechanisms create different antibody risks.1,2
Reagent effects and antibody patterns
Scroll sideways to see all columns
| System or antigen | Reagent-treatment effect | Usual antibody pattern | Important laboratory consequence |
|---|---|---|---|
| M and N | Destroyed by papain and ficin | Often cold IgM; dosage common | Significance increases with 37 °C or antiglobulin reactivity |
| S and s | Papain and ficin may weaken or destroy antigen reactivity, depending on the method | Commonly IgG | Clinically significant examples require antigen-negative units |
| Rh | Enhanced by papain and ficin | Usually immune IgG; dosage common | Hemolytic transfusion reactions and fetal disease can be severe |
| Kell | Preserved by papain and ficin; destroyed by sulfhydryl reagents | Usually immune IgG | K is strongly immunogenic; reagent treatment can separate Kell reactivity from many other systems |
| Kidd | Enhanced by papain and ficin | IgG with frequent dosage and complement binding | Historical antibodies remain important after current reactivity fades |
| Duffy | Destroyed by papain and ficin | Usually IgG | Enzyme panels can remove Duffy reactivity while enhancing Rh and Kidd |
| Lutheran | Usually preserved by papain and ficin | Lua often IgM; Lub often IgG | Mixed-field Lua reactions and high-prevalence Lub guide the workup |
Enzyme or chemical treatment changes multiple antigens at once. Treated cells therefore provide one line of evidence within a controlled antibody investigation.1,2
Other red-cell systems and antigen prevalence
Selected systems provide recognizable clues when common panel interpretations fail:1,2
Scroll sideways to see all columns
| System | Carrier or antigen | Workup clue and clinical relevance |
|---|---|---|
| Diego | Band 3 for Dia, Dib, and Wra; the complex formed by Band 3 and glycophorin A for Wrb | Antibodies can cause hemolytic transfusion reactions and fetal disease; Wra may appear as an incompatible crossmatch because routine screen cells often lack it |
| Dombrock | ART4, including Doa, Dob, Hy, and Joa | Weak reactions and antibody evanescence can complicate identification; delayed reactions occur |
| Colton | Aquaporin 1, including Coa, Cob, and Co3 | Antibody to a high-prevalence Colton antigen can create broad panel reactivity and a rare-unit need |
| Landsteiner-Wiener | ICAM4, including LWa | Anti-LW often reacts strongest with adult D-positive cells and strongly with both D-positive and D-negative cord cells12; sulfhydryl treatment destroys LW while preserving RhD |
| Chido/Rodgers | Complement C4 fragments adsorbed from plasma | Antibody is usually clinically insignificant and can be neutralized by pooled plasma during identification |
| Vel | SMIM1 | Anti-Vel can activate complement and cause a severe hemolytic transfusion reaction |
| JR | ABCG2 carrying Jra | Anti-Jra has variable clinical behavior; compatible units may require rare-donor resources |
| Augustine | ENT1 carrying Ata and related antigens | Anti-Ata can cause a severe hemolytic transfusion reaction |
ISBT places genetically assigned antigens in numbered blood group systems. Unassigned antigens with a prevalence below 1% enter the 700 series, and unassigned antigens with a prevalence above 90% enter the 901 series.2
An antibody to a low-prevalence antigen may be absent from the routine screen because the reagent cells lack its target, then appear when a donor crossmatch is incompatible. An antibody to a high-prevalence antigen commonly reacts with nearly every panel cell while the autocontrol remains negative. The latter pattern prompts review of reagent-cell exclusions, patient ancestry and transfusion history, null phenotypes, molecular testing, and rare-donor resources. Final component selection follows the identified antibody, its clinical significance, and the laboratory's validated procedure.1,2
HLA and human platelet antigens
HLA organization and inheritance
The human leukocyte antigen region lies within the major histocompatibility complex on chromosome 6. Mature red cells lack HLA class II and carry only small, variable amounts of HLA class I. Some people express enough red-cell HLA class I to type as the Bennett-Goodspeed (Bg) antigens. Platelets express HLA class I. Class I and class II genes encode antigen-presenting molecules; the class III region contains complement and other immune-response genes.1
Scroll sideways to see all columns
| MHC region | Principal loci or genes | Product distribution | Principal role |
|---|---|---|---|
| Class I | HLA-A, HLA-B, HLA-C | Nearly all nucleated cells and platelets | Presents endogenous peptide to CD8-positive T cells |
| Class II | HLA-DR, HLA-DQ, HLA-DP | B cells, dendritic cells, macrophages, and other activated antigen-presenting cells | Presents processed extracellular peptide to CD4-positive T cells |
| Class III | C2, C4, CFB, TNF, and other immune-response genes | Complement proteins circulate in plasma; other encoded products act in tissues | Encodes complement and inflammatory mediators; antigen presentation belongs to classes I and II |
Closely linked HLA alleles inherited on one chromosome form a haplotype. Each full sibling has an approximate 25% probability of sharing both parental haplotypes, a 50% probability of sharing one, and a 25% probability of sharing neither. Recombination within the region and population linkage disequilibrium affect the observed allele combinations.1
An HLA allele name begins with the locus, an asterisk, and colon-separated fields. In HLA-B*27:05, B identifies the locus, 27 identifies the allele group associated with a serologic family, and 05 distinguishes a specific protein sequence. Additional fields can record synonymous or noncoding sequence differences. The WHO Nomenclature Committee for Factors of the HLA System assigns official names, and the IPD-IMGT/HLA Database maintains the current sequence and nomenclature records. Release 3.65 was issued in July 2026.9
Pregnancy, transfusion, and transplantation can stimulate HLA alloantibodies. In transfusion medicine, these antibodies are central to immune platelet refractoriness and selection of HLA-matched or crossmatch-compatible platelets. HLA typing and antibody testing also support donor-recipient matching and transplant risk assessment.1
Human platelet antigens
Human platelet antigens are polymorphisms on platelet membrane glycoproteins and are classified separately from HLA. HPA-1a and HPA-1b, for example, are carried on integrin β3, historically called glycoprotein IIIa. An HPA alloantibody can cause fetal and neonatal alloimmune thrombocytopenia, posttransfusion purpura, or platelet refractoriness. The ISBT Platelet Immunology Nomenclature Committee assigns HPA names, and the Versiti HPA Gene Database maintains the current antigen and gene record.1,10
A poor posttransfusion platelet-count increment may reflect HLA antibody, HPA antibody, or a nonimmune cause. The distinction directs the subsequent antibody test and the choice of compatible platelet support.1,10
Immunogenicity and compatible-unit availability
Two questions follow every newly identified antibody: how likely was it to form, and how hard will compatible units be to find? An antigen is a molecular structure that an antibody or B-cell receptor can bind specifically, or that a T-cell receptor can recognize after appropriate presentation. An immunogen is an antigen that induces an immune response. A hapten is a small molecule that becomes immunogenic after conjugation to a carrier protein. Immunogenicity depends on structural difference from the recipient's own proteins, antigen density and accessibility, dose and route of exposure, inflammatory context, and the recipient's immune-response genes.1
Classic calculated estimates of the probability that an antigen-negative transfusion recipient will form the corresponding antibody after a single exposure to antigen-positive red cells rank several antigens as follows:11
Scroll sideways to see all columns
| Antigen | System | Estimated probability of antibody formation |
|---|---|---|
| D | Rh | 50% |
| K | Kell | 5% |
| c | Rh | 2.05% |
| E | Rh | 1.69% |
| Fya | Duffy | 0.23% |
| Jka | Kidd | 0.07% |
These values are historical model estimates derived from antibody occurrence and antigen exposure. Study population, number of exposures, pregnancy history, testing sensitivity, antibody evanescence, and recipient biology change the observed rate. Their durable teaching value is the comparative order: D is highly immunogenic, K follows among common non-ABO antigens, and several antibodies with lower formation rates remain clinically important once present.11
Antigen prevalence answers a separate compatibility question. K is absent from most donor units, so K-negative units are usually obtainable. High-prevalence k is present on nearly all units, making anti-k support difficult despite its lower immunogenicity. The same principle applies to anti-U, anti-Ku, anti-Jk3, anti-Lu3, anti-Vel, and antibodies against other high-prevalence antigens. An antibody's clinical significance determines the required compatibility; prevalence determines how far the search may need to extend. Weigh both as soon as the antibody is identified, so the clinical team hears early when a search will take time.1,2
Practice
References
- Bloch EM, Campbell-Lee S, McKenna DH Jr, Montemayor-Garcia C, Schwartz J, Shaz B, Storry J, eds. Technical Manual. 22nd ed. AABB; 2026. Back to text
- International Society of Blood Transfusion. ISBT Blood Group Database, September 2026 release, version 19. Updated September 1, 2026. Accessed September 7, 2026. Back to text
- International Society of Blood Transfusion Working Party on Red Cell Immunogenetics and Blood Group Terminology. RHD blood group alleles. Updated September 30, 2022. Accessed August 27, 2026. Back to text
- Determination of the Rh factors, 21 CFR § 640.5(c) (2026). Accessed August 27, 2026. https://www.ecfr.gov/current/title-21/chapter-I/subchapter-F/part-640/subpart-A/section-640.5 Back to text
- AABB, College of American Pathologists, American College of Obstetricians and Gynecologists, et al. Joint Statement on Phasing-In RHD Genotyping for Pregnant Women and Other Females of Childbearing Potential With a Serologic Weak D Phenotype. 2015. Accessed August 27, 2026. Back to text
- AABB. Association Bulletin #24-02: Use of Rh Immune Globulin and Considerations in the Setting of Supply Shortages and Limited Availability. July 15, 2024. Accessed August 29, 2026. Back to text
- Jung HH, Danek A, Walker RH, Frey BM, Peikert K. McLeod neuroacanthocytosis syndrome. In: GeneReviews. University of Washington, Seattle. Updated September 16, 2021. Accessed September 7, 2026. Back to text
- Bradley L, Yewhalaw D, Hemming-Schroeder E, et al. Epidemiology of Plasmodium vivax in Duffy negatives and Duffy positives from community and health centre collections in Ethiopia. Malar J. 2024;23:76. doi:10.1186/s12936-024-04895-1 Back to text
- IPD-IMGT/HLA Database. Release 3.65 version report. July 2026. Accessed August 27, 2026. Back to text
- Versiti. Human Platelet Antigen Gene Database. Accessed August 27, 2026. Back to text
- Giblett ER. A critique of the theoretical hazard of inter vs. intra-racial transfusion. Transfusion. 1961;1(4):233-238. doi:10.1111/j.1537-2995.1961.tb00048.x Back to text
- Moiz B, Hasan M, Salman M, Kumari B, Black E, Liew YW. Anti-LW masquerading as anti-D in a D+ patient. Immunohematology. 2025;41(1):11-16. Back to text
- Chen Q, Xiao J, Zhang M, Huang C, Li M, Flegel WA. A null allele caused by a four-base-pair duplication within the RHCE gene encoding a D−− phenotype. Transfusion. 2021;61(3):E23-E26. doi:10.1111/trf.16211 Accessed September 11, 2026. https://pmc.ncbi.nlm.nih.gov/articles/PMC9066653/ D−− carriers lose CcEe expression and can form anti-Rh17; the Rhnull amorph phenotype and anti-Rh29 are covered in the AABB Technical Manual. Back to text