Icon Legend

This session is not in your schedule.

This session is in your schedule. Click again to remove it.

Event Icons

Assessor Sessions

Committee Meetings

Foundation Events

Networking Events

Oral Abstracts

CABP Eligible

Video Recording

Oral Abstract

Immunohematology and Genetic Testing (red cells, leukocytes and platelets)

Oral Abstract Session - Novel RHD Variants and Approaches to Overcoming Them

#OA1-AM25-MN-36 - Partial RHD Genotypes in Genetic-Ancestry Genome Databases

Monday, October 27, 2025

5:00 PM - 6:00 PM PT

Room: SDCC 30CDE

CE: N/A

Presenting Author(s)

Glenn Ramsey, MD (he/him/his)

Feinberg School of Medicine, Northwestern University
Chicago, Illinois

Disclosure information not submitted.

Handouts

Background/Case Studies: In patients, partial RHD genotypes pose risk for anti-D alloimmunization, cause some weak or discrepant RhD phenotypes, and influence RhD typing methods (DVI-negative reagents). RHD gene surveys are published mainly in Black donors or sickle cell disease (SCD) patients. Large genome databases could provide broader RHD information for other genetic ancestries, irrespective of phenotype. Variant co-occurrence is a database tool for genetic disease to predict whether two intragenic variants occur in cis or trans.

Study

Design/Methods: We examined all alleles designated as partial RHD in International Society of Blood Transfusion RHD allele table v6.4. Single-nucleotide variant (SNV) allele frequencies (AF) were obtained from Genome Aggregation Database (gnomAD) v4.1.0 (global) and All of US public genomic database v8 (USA) (GRCh38 assemblies). We converted AF to hemi- or homozygous (HHZ) genotype frequencies (GF) with Hardy-Weinberg formulas using combined D-negative database AF for RHD*01N.01 plus markers for RHD*03N.01 and RHD*08N.01. Multinucleotide-variant (MNV) GF including RHCE gene conversions were estimated from informative HHZ cis-variant co-occurrences in gnomAD (available only in v2.1.1; no Mideast). MNV allele designations included small groups of similar alleles. We limited DAU estimates to Africans and Admixed Americans (formerly Hispanic/Latino) because c.1136T (DAU0) is in reference genome assemblies and unavailable for co-occurrence analysis. African & European GF were compared to those derived from published USA RHD AF surveys.

Results/Findings: The Table shows the top 6 partial RHD genotypes predicted in 6 genetic ancestries using gnomAD. All of US SNV results were similar. USA RHD gene surveys generally corroborated database-derived African & European GF. Variants c.455C, c.667G and c.733G occur as SNVs (DNT, DFV, DUC2) or in other MNV alleles not shown. Three RHD-CE-D hybrid genotypes were found by HHZ cis-variant co-occurrence: DIV type 4 in 2/4580 Ashkenazi Jews (AJ) (0.04%), and weak D type 14 (not considered partial) and DFR-1, each in 1/47,707 Europeans (0.002%). No DVI patterns were seen (< 1:12,400 Europeans, 95% confidence interval). Predicted RhD-negative rates in Amish and Mideast groups were 22-23%.

Conclusions: Genome databases can provide useful insights into partial RHD GF. We successfully adapted variant co-occurrence analysis to examine the most common MNV partial RHD genotypes, although gnomAD v2.1.1 is older (GRCh37) and relatively small. DVII should be included in RHD genotyping assays. Partial RHD alleles in Admixed Americans reflected their proportion of African ancestry. gnomAD provided novel access to Amish and AJ RHD data, notably for DII and DIV type 4 markers in AJ. DAU analysis was challenging because of DAU0 in reference genomes. Future genome database enhancements will enable more robust analysis of RHD variants in diverse genetic ancestries.