RHEUMATOLOGY GRAND ROUNDS AT RUSH
EDITORS:ROBERT S. KATZ, MD, AND JOEL A. BLOCK, MD
HLA-B27 and Its Pathogenic Role
Muhammad Asim Khan, MD, MACP, FRCP
T
he year 1973 is a very memorable year for me because it
was the year that the remarkable association was first
reported between HLA-B27 and ankylosing spondylitis (AS),
as I had suffered since 1956 from this illness. Moreover, in
1973, I passed on my HLA-B27 gene to my first offspring,
finished my fellowship training in rheumatology, and started
my academic career.
DISCOVERY OF HLA-B27
It was Erik Thorsby from Norway who first described
HLA-B27, originally named TH-FJH.
1
It was detected by
using an antibody that was produced by “planned immuni-
zation,” a method that was being used by several investi-
gators. He immunized a colleague, FJH, with a skin trans-
plant and cells from a healthy donor who was HLA
identical with the recipient because Eric assumed that the
donor might possess a hitherto undetected HLA antigen.
The recipient produced an antibody that detected a “new”
antigen, which he named TH-FJH, and this antigen was
later given the name HLA-B27.
Erik told me that later when he tested his own family,
he noticed that one family member who carried this new
antigen had AS, whereas another had questionable AS (per-
sonal communication). However, since most of his family
members who carried the new antigen were completely
healthy, he discarded this finding as being caused by chance.
He has ever since regretted that he did not pursue this
possible association by testing some other patients who suf-
fered from AS. Had he done so he would have been the very
first to detect an association between an HLA antigen and a
disease!
HLA-B27 PREVALENCE AND DISEASE
ASSOCIATION
The prevalence of HLA-B27 varies markedly in the
various population groups worldwide (see Table 1).
2
Its
association with AS is among the strongest of an MHC
antigen and any disease. HLA-B27 also shows a strong
association with the related spondyloarthropathies (SpA),
although not as strong as with AS (Table 2).
2
Moreover, the
strength of these disease associations also varies among the
various ethnic and racial groups worldwide.
2–5
Speaking in
general terms, the prevalence of AS roughly correlates with
the prevalence of HLA-B27 but with certain glaring excep-
tions. For example, HLA-B27 is present in West African
countries, such as Senegal and Gambia, but it is very difficult
to find any cases of AS in these countries.
CLINICAL UTILITY OF HLA-B27 TEST AS AN
AID TO DIAGNOSIS
To use of B27 typing as an aid to diagnosis in clinical
medicine, we must know how to use laboratory tests that are
imperfect; i.e., they are neither 100% sensitive nor 100%
specific and cannot be used as a diagnostic or screening test.
The result of the test has to be interpreted in light of the a
priori clinical likelihood that the person had that disease
before the test was ordered. Those interested in this subject
are referred to my prior publications on this subject.
3–6
HLA-B27 POSITIVE VERSUS HLA-B27 NEGATIVE
DISEASE
There are phenotypic differences between B27 positive
and B27 negative patients with AS; for example the B27-
positive patients are significantly more likely to get acute
anterior uveitis and are more likely to have a family history
of AS and related spondyloarthropathies than a B27 negative
patient.
7
The disease can occur in the absence of B27 because
there are additional yet undiscovered disease-predisposing
genes in the rest of the genome that are also playing a role.
One such gene lies on chromosome 9 that is strongly asso-
ciated with acute anterior uveitis in patients with AS.
8
HLA-B27 HOMOZYGOSITY AND DISEASE
ASSOCIATION
Many years ago, we first reported that B27 homozy-
gosity increases the likelihood of disease by a factor of more
than 3.
9
This aspect was not thoroughly pursued by other
investigators since then, but finally this observation has
recently been confirmed.
10
Why is it that homozygous indi-
viduals with B27 are at least 3 times more likely to get the
disease as opposed to heterozygous individuals? If some of
the B27-extended haplotypes carry additional genetic risk
factors for AS, both inside and outside the MHC region on
chromosome 6, it would not be surprising that an individual
From Rheumatology Grand Rounds at Rush University Medical Center,
Chicago, IL USA. Editors: Robert S. Katz, MD, and Joel A. Block, MD.
From the Case Western Reserve University, MetroHealth Medical Center,
Division of Rheumatology, Cleveland, OH.
Reprints: Muhammad Asim Khan, MD, MACP, FRCP, Case Western
Reserve University, MetroHealth Medical Center, Division of Rheuma-
tology, 2500 MetroHealth Drive, Cleveland, OH 44109-1998. E-mail:
mkhan@metrohealth.org.
Copyright © 2008 by Lippincott Williams & Wilkins
ISSN: 1076-1608/08/1401-0050
DOI: 10.1097/RHU.0b013e3181637a38
JCR: Journal of Clinical Rheumatology • Volume 14, Number 1, February 2008 50