Body: To address the increasing complexity of HLA typing, P and G group classifications were developed to group alleles that share identical amino acid (P) or nucleotide (G) sequences within the antigen binding domain (ABD). These groupings help simplify donor-recipient matching. Most epitopes recognized by anti-HLA antibodies or T cell receptors fall within this region. However, antibodies targeting regions outside the ABD have also been reported. In such cases, patients may form antibodies against alleles that are otherwise considered a match under P or G group classifications.
This has direct implications for virtual crossmatch (vXM) in solid organ transplantation. If the presence of antibodies to non-ABD epitopes is overlooked, vXM results may be falsely interpreted as compatible. This risk is especially relevant in sensitized patients. High-resolution HLA typing for both living and deceased donors is essential to accurately assess compatibility and avoid missed donor-specific antibodies that can compromise graft outcomes.
Patient 1 is a 60-year-old male evaluated for a second kidney transplant. Class II antibody testing showed reactivity to all DQ beads except for the one carrying DQB1*02:02-DQA1*02:01 (Fig 1). Based on SSOP typing and his DR typing, he was presumed to be homozygous for DQB1*02:02. Although this allele belongs to the DQB1*02:01P and DQB1*02:01:01G groups, it differs by a single amino acid at position 135 in exon 3 (Glycine vs. Aspartic acid), which likely contributed to the specific antibody reactivity.
Patient 2 is a 33-year-old male awaiting a second transplant. He developed antibodies to epitopes on his previous donor’s DQ6 and DPA1*01. Unexpectedly, he also showed reactivity to DPB1*17:01, which is in the same P and G groups as his own DPB1*131:01:01 (Fig 2). These two alleles differ at position 96 in exon 3, suggesting that antibodies were formed against this non-ABD epitope.
Conclusion: These cases illustrate how differences outside the ABD can lead to clinically significant sensitization. Relying only on P and G group data may overlook meaningful mismatches. Accurate virtual crossmatching requires high-resolution typing that includes non-ABD regions to ensure informed and safe donor selection.
