Guides › Nomenclature

Why HLA-C antigens are written Cw but HLA-C alleles are written C

Updated 2026-10-30 · allele examples checked against IPD-IMGT/HLA 3.66.0 on 2026-10-30

HLA-C alleles are written with C, as in C*07:02, and HLA-C serological antigens are written with Cw, as in Cw7. The w was removed from allele names in the April 2010 nomenclature update but deliberately kept in antigen names.

So Cw*07:02 is neither form. It mixes the antigen prefix with allele syntax, and in current nomenclature it should be read as C*07:02. The rest of this guide explains where the w came from, why it survives for antigens, and how to handle it in data.

What the 2010 update changed

The April 2010 nomenclature update introduced colons between fields and, at the same time, removed the w from HLA-C allele names. Its published examples:

The same page says the w was kept in HLA-C antigen names, to avoid confusion with components of the complement system and with the C1 and C2 epitopes that HLA-C carries as KIR ligands. A bare "C1" in a serology report could otherwise mean an HLA-C antigen, a complement component or a KIR-ligand group.

The antigen names still use Cw

The WHO nomenclature lists HLA-C antigens with the w. The HLA antigens page at hla.alleles.org includes Cw7, for example.

The machine-readable file agrees. hla_nom.txt in the WMDA folder of IPD-IMGT/HLA release 3.66.0 lists HLA-C antigens under the locus label Cw:

• Cw1 to Cw10, dated between 1975 and 1987;
• Cw11, deleted on 14 November 1991 with the reason "Sequence error";
• entries dated 28 January 2026, added with the 2026 nomenclature report, including Cw12, Cw14 to Cw18 and four-digit designations such as Cw0304.

The WMDA README explains that the 2026 report added new antigen specificities and associated antigens, and that three-digit associated antigen codes are being extended to four digits. In rel_dna_ser.txt, the file that maps alleles to serological equivalents, C*07:02:01:01 now carries the value 0702 and C*01:02:01:01 the value 1.

The allele names do not

In release 3.66.0 every HLA-C allele in Allelelist.txt is written with C*. No current allele name begins Cw*.

The first-field groups present are C*01 to C*08, C*12 and C*14 to C*18. There is no C*09, C*10, C*11 or C*13 group. That is not a gap to fill: two early names in those groups, Cw*1101 and Cw*1301, were deleted as sequence errors in 1991 and 2002, and the deletion file records both.

The relationship between a C* allele group and a Cw antigen is the usual one in HLA nomenclature: the first field often corresponds to the serological antigen, but not always. Check rel_dna_ser.txt for a stated equivalent rather than reading one off the allele name.

The forms you will find in old data

Data written over several decades can contain HLA-C in four shapes:

The name history in IPD-IMGT/HLA traces the pre-2010 forms. Cw*0702 appears in releases 1.05.0 to 1.08.0 and Cw*07020101 in 1.16.0 to 2.28.0, both under the same HLA ID as today's C*07:02:01:01 (HLA00434).

The hybrid form is the one to watch. It looks modern because of the colon, so a validator that only checks for colons will pass it, while one that checks names against the allele list will reject it. Neither is right. It is a recognisable spelling of C*07:02 and should be converted, with a note that the spelling is deprecated.

Some old names do not convert to anything current. Cw*0402 was deleted in June 1999 as "shown to be identical to C*04:01:01:01", so it maps to that allele through the deletion record, not by changing the prefix.

Handling Cw in a pipeline

1. Separate antigens from alleles first. A value with an asterisk is an allele name; one without, such as Cw7, is serology and needs a different path.
2. Map Cw* to C* for alleles, and add colons if they are missing. Cw*0702 and Cw*07:02 both stand for C*07:02.
3. Check the result against the allele list for the release. Mechanical conversion can produce a name that was deleted or never assigned; deleted names need their successor.
4. Keep the original string. Store what the lab reported next to the normalised name and the release used.
5. Do not add a w to allele names when exporting, even for systems that display antigens as Cw.

The guide on converting pre-2010 names covers the colon conversion and the rollover cases in more detail.

How HLA-Verify reads Cw

POST /v1/normalize accepts the legacy forms. Sent Cw*0702 or Cw*07:02, it returns C*07:02 with the flag deprecated_name. POST /v1/verify reports both as valid, with current_2field set to C*07:02 and the same flag, so a report written in the old format is not treated as fabricated, while an invented name with a Cw prefix is still reported as hallucinated.

In match calculations the names are normalised before comparison. Checked against release 3.66.0, a donor written with Cw*0702 and a recipient written with C*07:02 match at that position; the spelling is not counted as a mismatch.

Frequently asked questions

Why is it HLA-Cw7 but HLA-C*07:02?

Cw7 is a serological antigen and C*07:02 is an allele. The 2010 update removed the w from allele names but kept it in antigen names to avoid confusion with complement components and the C1/C2 KIR-ligand epitopes.

Is Cw*07:02 a valid HLA allele name?

No. It combines the antigen prefix with current allele syntax. Read it as C*07:02 and record that the spelling is deprecated.

When did HLA-C allele names lose the w?

In April 2010, with the same nomenclature update that introduced colons between fields.

Are there HLA-C*13 alleles?

Not in IPD-IMGT/HLA release 3.66.0. The HLA-C groups present are C*01 to C*08, C*12 and C*14 to C*18; an early Cw*1301 name was deleted in 2002 as a sequence error.