Lumping or splitting autoimmune rheumatic disorders? Lessons from Sjögren's syndrome.
British Journal of Rheumatology 1998;37:1263–1264
LUMPING OR SPLITTING AUTOIMMUNE RHEUMATIC DISORDERS?
LESSONS FROM SJO¨GREN’S SYNDROME
A rheumatic disorders constitute a diverse
group of illnesses with significant clinical and immuno-
logical variability. Their clinical presentation and
course vary, ranging from subclinical and mild to
severe and life-threatening, and from slow and indolent
to acute and catastrophic. Some of these disorders can
be manifested simultaneously in the same individual
or, alternatively, one nosological entity may evolve
into another. For instance, Sjo ¨gren’s syndrome (SS)
frequently co-exists with various other rheumatic con-
ditions, whereas a patient with an organ-specific disease
(e.g. myasthenia gravis or thyroiditis) can progress to
a systemic autoimmune disorder, such as systemic
lupus erythematosus (SLE) or SS. A multi-faceted
activation of the immune system occurs in these dis-
orders, as illustrated by the presence of autoantibodies
directed against one or multiple autoantigens and/or
mononuclear inflammatory infiltrates in the affected
tissues. The interplay of genetic factor(s) and environ-
mental agent(s) is thought to account for the develop-
ment and clinical course of these diseases. The nature
of such factors, however, remains obscure, whereas the
implication of microorganisms has been speculative
for the last three decades .
In an effort to establish an avenue of communication
among clinical investigators, diagnostic/classification
criteria have been established for a number of auto-
immune rheumatic disorders [2, 3]. Nevertheless, group-
ing individuals with diverse clinical and immunological
manifestations into adisorder carries the risk of masking
certain important or differentiating features.
Recent advances in the clinical, immunological and
immunogenetic profile of one of these disorders, namely
SS, can serve as a paradigm to illustrate the problem
and offer some solutions. This autoimmune disorder is
common and affects primarily females at the fourth to
fifth decade of life. It presents with a diverse clinical
spectrum ranging from disease confined in the exocrine
glands to systemic disorder, whereas in a considerable
number of patients lymphoid neoplasia ensues. The
lungs, kidneys, liver and thyroid gland are frequently
affected, and several lines of evidence suggest that
epithelial cells are the main targets of destruction .
SS can be found alone or in association with almost all
autoimmune rheumatic disorders.The diagnosis is based
on the criteria put forward by a European multicentre
study . Accordingly, all conditions (metabolic, infec-
tious, degenerative, inflammatory, neoplastic and drug
side-effects) that mimic the clinical and histopathological
features of SS should be excluded.
In the late 1970s, comparison of the clinical, sero-
logical and immunogenetic profiles of patients with SS
without evidence for another autoimmune disorder and
those of patients with rheumatoid arthritis (RA) and
SS has shown similarities and differences. Consequently,
the terms ‘primary’ and ‘secondary’ SS were proposed
for the former and the latter groups of patients.
Compared to patients with secondary SS, patients with
primary SS present significantly more often parotid
gland enlargement and systemic manifestations, such as
Raynaud’s phenomenon, lymphadenopathy, spleno-
megaly, purpura and renal tubular acidosis. In addition,
primary SS, but not secondary SS, is characterized
by the occurrence of serum autoantibodies against
Ro(SSA) (in 60%) and La(SSB) (in 40%) ribonucleo-
proteins, and is associated with distinct HLA haplotypes
[5, 6]. Although the incidence, clinical expression and
serology of SS in various autoimmune diseases have
been evaluated , comparative studies of well-defined
groups of patients with SS co-expressed with auto-
immune rheumatic diseases other than RA have never
been performed. Despite this fact, during the present
decade, the term ‘secondary SS’ has been arbitrarily
granted to SS occurring in patients with any auto-
immune disorder .
Today, 20 yr after the original classification, clinical,
histopathological and serological evidence implies the
existence of at least two additional SS groups.
(1) Primary SS overlapping with other autoimmune
disorders. This group includes individuals who clinically
express SS and suffer from another defined rheumatic
disease. These patients express autoantibodies to
Ro(SSA) andLa(SSB)proteins as frequentlyas patients
with the ‘primary’ disorder. Immunogenetic studies have
revealed that the immune response to Ro(SSA) and
La(SSB) autoantigens is associated with distinct major
histocompatibility complex (MHC) class II alleles,
regardless of the clinical entity where they occur [8, 9].
Patients with autoimmune disorders who present fea-
tures of SS and express anti-Ro(SSA) and anti-La(SSB)
responses could qualify for this group . Careful
evaluation of the clinical and immunogenetic profile of
this group, and in comparison to that of primary or
secondary SS groups, should be performed.
(2) SS with other autoantibody specificities. The
patients of this group are seemingly indistinguishable
from primary SS. On closer scrutiny, however, these
patients reveal characteristic clinical and serological
features. Their autoantibody profile consists of reactiv-
ities compatible with other autoimmune disorders
(namely, autoantibodies to mitochondria, thyroid per-
oxidase or centromere) [11, 12]. Most often, these
individuals do not express the overt clinical picture of
the correspondingdisorder (i.e.primarybiliary cirrhosis,
thyroiditis or limited scleroderma). However, the histo-
pathological evaluation of liver or thyroid tissues in SS
© 1998 British Society for Rheumatology
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 37 NO. 121264
patients displaying anti-mitochondrial or anti-thyroid
peroxidase antibodies reveals an early lesion of biliary
cirrhosis or thyroiditis, respectively. Nail-fold capillaro-
scopy of SS patients with anti-centromere antibodies has
revealed vascular lesions comparable to those encoun-
tered in scleroderma (paper submitted for publication).
On the basis of their immune response, SS patients
can be subdivided into three major subgroups: those
with anti-Ro(SSA) and anti-La(SSB) response; those
without specific autoantibody response; and those with
autoantibodies against certain autoantigens, such as
thyroid peroxidase, mitochondria and centromere.
Furthermore, the detailed analysis of the initial patho-
logical lesions in the exocrine glands may be also
instructive for the subclassification of patients with SS.
Studies have suggested that in patients with ‘primary’
SS and in those with ‘secondary’ SS associated with
RA or primary biliary cirrhosis, the lesion starts in
close proximity to exocrine glandular ducts, whereas
in patients with sicca manifestations occurring in the
context of SLE or infection by the human immuno-
deficiency virus (HIV) the initial lesion is perivascular
[13, 14]. In contrast, in patients with chronic graft-
versus-host disease who develop an SS-like syndrome,
Histopathological studies involving multicentre collab-
oration could also be extended to comparisons with
the SS-like tissue-infiltrating lesions found during infec-
tion by HIV or by the hepatitis C virus. Finally, SS
patients can also be subdivided into groups according
to their MHC class II associations. It will be very
interesting to evaluate the clinical similarities and
differences of these groups, as well as their clinical
course and disease evolution.
Splitting patients into subgroups on the basis of more
global assessment may provide more homogeneous
populations of patients for study and allow more precise
evaluation of disease parameters. This is likely to
become particularly productive if careful studies of
groups are designed and performed. Are these studies
feasible? In order to collect a reasonable number of
individuals with SS from the various patient subgroups,
multicentre/multinational studies involving several dis-
ciplines (i.e. rheumatology, ophthalmology, dentistry,
pathology and immunology) should be initiated. The
acquisition of appropriate funding for the elaboration
of these studies appears a daring task. Such an orches-
trated effort is mandatory and granting agents should
It can be anticipated that the SS subgroups described
above represent a common spectrum of the disorder.
Ultimately, splitting SS patients in subgroups may lead
to such a realization. However, this may well pre-date
identification of the factor(s) that are responsible for
the disease, whose discovery remains imperative.
H. M. M and M. N. M
Department of Pathophysiology, School of Medicine,
National University of Athens, Greece.
Correspondence to: H. M. Moutsopoulos, Department of
Pathophysiology, School of Medicine, National University
of Athens, Mikras Asias 75, Athens 115 27, Greece.
1. Talal N. Molecular autoimmunity. London: Academic
2. Schumacher HR Jr. Classification of rheumatic diseases.
In: Klippel JH, Dieppe PA, eds. Rheumatology. London:
Balestrieri G, Bencivelli W, Bernstein RM et al.
Preliminary criteria for the classification of Sjo ¨gren’s
syndrome. Results of the EEC prospective concerted
action. Arthritis Rheum 1993;36:340–7.
4. Moutsopoulos HM. Sjo ¨gren’s syndrome: Autoimmune
epithelitis. Clin Immunol Immunopathol 1994;72:162–5.
5. Moutsopoulos HM, Chused TM, Mann DL, Klippel
JH, Fauci AS, Frank MM et al. Sjo ¨gren’s syndrome
(sicca syndrome): current issues. Ann Intern Med
6. Anonymous. Primary and secondary Sjogren’s syn-
drome. [Editorial] Lancet 1984;ii:730–1.
7. Talal N, Moutsopoulos HM, Kassan SS (eds) Sjogren’s
syndrome. Heidelberg: Springer-Verlag, 1987:165–92.
8. Raveille JD, MacLeod MJ, Whirrington K, Arnett FC.
Specific amino acid residues in the second hypervariable
region of HLA-DQA1 and DQB1 chain genes pro-
mote the Ro(SSA)/La(SSB) autoantibody responses.
J Immunol 1991;146:3871–7.
9. Tambur AR, Friedman A, Safirman C, Markiziu A,
Ben-Chetrit E, Rubinow A et al. Molecular analysis of
HLA-class II genes in primary Sjo ¨gren’s syndrome
patients: a study of Israeli Jewish and Greek non-Jewish
patients. Hum Immunol 1993;36:235–42.
10. Skopouli FN, Andonopoulos AP, Moutsopoulos HM.
Clinical implications of anti-Ro(SSA) antibodies in
11. Skopouli FN, Barbatis C, Moutsopoulos HM. Liver
involvement in primary Sjo ¨gren’s syndrome. Br J
Moutsopoulos HM. Patients with anti-centromere anti-
bodies (ACA): Clinical features, diagnosis and evolution.
Br J Rheumatol 1993;32:297–301.
Galanopoulou D, Papadimitriou CS, Moutsopoulos
HM. Histologic lesion in the labial salivary glands in
systemic lupus erythematosus patients. Oral Surg Oral
Pathol Oral Med 1991;72:208–12.
14. Itescu S, Brancato LJ, Buxbaum J, Gregersen PK, Rizk
CC, Croxson TS et al. A diffuse infiltrative lymphocytosis
syndrome in human immunodeficiency virus (HIV)
infection: A host immune response associated with HLA-
DR5. Ann Intern Med 1990;112:3–10.
15. Gratwhol AA, Moutsopoulos HM, Chused TM et al.
Sjo ¨gren-type syndrome after allogeneic bone-marrow
transplantation. Ann Intern Med 1977;87:703–6.