The HLA System and Its Role in Forensic Relationship Testing

The MHC on chromosome 6 is conventionally divided into three regions. The class I region encodes HLA-A, HLA-B, and HLA-C, which are expressed on virtually all nucleated cells in the body. The class II region encodes HLA-DR, HLA-DQ, and HLA-DP, which are expressed constitutively only on professional antigen-presenting cells such as dendritic cells, macrophages, and B lymphocytes. Class III encodes complement proteins and other immune mediators and is not directly relevant to HLA typing for relationship testing.

In the serological era, forensic typing focused on class I antigens (HLA-A and HLA-B) because antibody panels were better characterised for class I and class I antigens are present on lymphocytes used in CDC assays. Class II typing, particularly HLA-DR, became practical with molecular methods and was added to paternity panels in the late 1980s and early 1990s, substantially increasing discrimination power. The combined polymorphism across HLA-A, HLA-B, and HLA-DR means that the probability of two unrelated individuals sharing all three typings is very low in most population groups.

The CDC assay was developed by Paul Terasaki and John McClelland in the 1960s and became the standard method for HLA typing in both transplant medicine and forensic serology. It detects HLA antigens on the surface of living lymphocytes by exploiting the fact that antibody binding activates complement, which punches pores in the cell membrane and kills the cell.

The procedure runs as follows. Lymphocytes are isolated from peripheral blood by density gradient centrifugation. A small volume of the lymphocyte suspension is transferred into each well of a Terasaki microtray, which contains predispensed HLA-specific antisera. Each well tests for one HLA specificity. After incubation at room temperature to allow antibody binding, rabbit complement is added to every well. A second incubation allows complement-mediated lysis to occur. A vital dye such as eosin-Y is then added; dead cells take up the dye and appear dark under a phase-contrast microscope, while live cells exclude it and remain refractile. Wells with more than approximately 20 percent dead cells are scored as positive, meaning the cells carry the antigen detected by that well's antiserum.

Despite its limitations, CDC typing was forensically admissible in many jurisdictions from the 1970s onward. German courts were among the earliest to use HLA serology in paternity proceedings. In the United States, the American Association of Blood Banks established proficiency standards for HLA typing laboratories from 1978. In India, HLA typing evidence was admitted in paternity suits under the Indian Evidence Act 1872 (now the Bharatiya Sakshya Adhiniyam 2023) with courts assessing the weight of the evidence through expert testimony about antigen frequencies and population databases. The UK courts similarly admitted HLA serology under the Family Law Reform Act 1969, which governed blood tests for paternity. European civil law systems generally required court-appointed experts to perform the testing.

Molecular HLA typing began to enter clinical and forensic practice in the early 1990s. The fundamental shift is that PCR-based methods interrogate the DNA sequence directly rather than detecting surface antigen with antibodies. This eliminates the cross-reactivity problem and extends typing to samples that cannot provide viable lymphocytes: blood stains, buccal swabs, and, with some caveats, skeletal material.

Three main molecular strategies are in use. Sequence-specific primer PCR (SSP, also called PCR-SSP or ARMS-PCR) uses primer pairs whose 3-prime ends are matched to specific allele sequences. Amplification occurs only when the target allele is present. A panel of reactions covering relevant alleles is run simultaneously, and the resulting gel pattern identifies the alleles present. SSP gives low to intermediate resolution, sufficient for forensic paternity and kinship work, and is rapid enough for single-sample urgent cases.

Sequence-specific oligonucleotide probe typing (PCR-SSOP) amplifies exons 2 and 3 of class I or exon 2 of class II genes with generic primers, then hybridises the product with an array of labelled probes each specific for a known allele sequence. The pattern of positive and negative hybridisations identifies the allele. SSOP offers higher throughput than SSP and is suitable for population studies and large-scale kinship cases. Sequence-based typing (SBT) directly sequences the polymorphic exons and compares them with reference databases such as the IMGT/HLA database maintained by the European Bioinformatics Institute. SBT gives the highest resolution and is now standard in transplant matching; in forensic contexts it is used when maximum discrimination is needed, as in immigration kinship cases where the question is whether a claimed parent-child relationship is biological.

Because HLA loci are tightly clustered on chromosome 6, alleles at HLA-A, -B, -C, -DR, and -DQ are usually inherited together as a haplotype. A child receives one haplotype from each parent, so the child's two haplotypes can in principle be traced back to the parental contributions. In a paternity case, the haplotype the child did not receive from the mother is the obligate paternal haplotype. If the alleged father does not carry any HLA haplotype consistent with the obligate paternal haplotype, he is excluded as the biological father.

When the alleged father is not excluded, a paternity index is calculated. For a single locus, the PI is the probability that the alleged father transmitted the obligate allele divided by the probability that a random man from the relevant population transmitted it. The denominator is the allele frequency in the population database. Multiple loci contribute independent PI values that are multiplied together. However, because HLA alleles at different loci are not always in linkage equilibrium, using individual allele frequencies and multiplying them can overestimate or underestimate the true haplotype frequency. Proper forensic calculation should use observed haplotype frequencies rather than multiplied single-locus allele frequencies when population data permit.

Linkage disequilibrium also affects the interpretation of apparent exclusions. The classical example is the HLA-A1, B8, DR3 haplotype, which is found at elevated frequency in northern European populations due to positive selection. A paternity case involving an alleged father from that background will show a higher denominator than expected from multiplied individual allele frequencies, correctly reducing the paternity index. Laboratories using HLA typing for forensic paternity were required to maintain population-specific allele and haplotype frequency tables, usually stratified by self-reported ethnic origin, a requirement that maps directly onto the current practice for STR profiling.

STR profiling entered forensic practice in the early 1990s and gradually displaced HLA serology from paternity testing because STR analysis is faster, requires less starting material, is not affected by antibody cross-reactivity, and provides a standardised statistical framework that courts find easier to present. By the late 1990s, most forensic paternity laboratories in Europe and North America had moved to STR-only panels. The transition was complete in most jurisdictions by the early 2000s.

During the transition period, some laboratories used HLA typing as a supplementary tool when STR evidence was inconclusive. A case where the STR paternity index was borderline could be strengthened by adding HLA data. The combined evidence from both types of marker was presented to the court as a single combined likelihood ratio, converting both STR and HLA results into a unified probability framework. This practice required careful attention to the independence assumption: HLA loci must be shown to be in linkage equilibrium with the STR loci used, which they are, given that STR markers used in forensic panels are distributed across all chromosomes and the MHC STR-equivalent markers are not part of standard commercial panels.

In some regions, HLA typing remained in use for kinship testing longer than in paternity testing. Immigration cases involving claimed parent-child or sibling relationships, where DNA sampling of the alleged relative in the home country was logistically difficult, sometimes relied on HLA typing of the claimant and the established family members in the host country. This was particularly common in UK immigration casework in the 1980s and 1990s. The Joint Entry Clearance Unit cases documented by the UK Home Office represent the best-recorded body of HLA kinship evidence in an immigration forensic context.

HLA typing retains forensic relevance in three distinct contemporary contexts. The first is transplant-related investigations. Organ trafficking is a serious crime in multiple jurisdictions: the WHO estimates that illicit transplants constitute a substantial fraction of global organ transactions, though precise figures are contested. When investigators seize medical records or interview alleged victims, HLA typing data from pre-transplant matching may be available in clinical files. Forensic analysis of those records can establish whether the donor and recipient HLA types were genuinely matched, helping to determine whether a legitimate clinical pathway was followed.

The second context is post-transplant forensic casework. After a bone marrow or stem cell transplant, the recipient's haematopoietic system becomes progressively chimaeric, containing cells from both donor and recipient. STR profiling of blood from a transplant recipient may show a mixed or completely donor-replaced profile. If that individual becomes involved in a criminal case, such as leaving blood at a crime scene, standard STR profiling of blood will yield the donor's profile, not the recipient's. The DNA Identification Act in the United States and equivalent legislation in the UK and Germany contain provisions or have been judicially interpreted to address this problem, but the burden falls on the forensic analyst to identify potential post-transplant chimerism. HLA typing of tissues other than blood, such as buccal mucosa or hair follicles, can confirm the individual's constitutional HLA type distinct from the donor profile.

The third context is mass disaster victim identification when bone marrow material is available. Bone marrow is a source of nucleated cells and thus of both HLA antigens and amplifiable DNA. In disasters where bodies are severely decomposed, haematopoietic cells from bone marrow may be better preserved than soft tissue, and HLA typing from marrow aspirates has been used in some mass-fatality investigations to assist in identification when ante-mortem HLA typing records from transplant workups exist for the victim. This application is narrow but forensically sound. The broader forensic biotechnology context for such identifications is covered in the forensic biotechnology subject index.