EDITORIAL - Can we rely on anti-CCP antibody determination for the diagnosis of early RA?

[Guest guest]

Journal of Rheumatology

December 2006

Editorial

--------------------------------------------------------------------------------

Can We Rely on Anti-Citrulline Antibody Determination for the Diagnosis of

Early Rheumatoid Arthritis?

DAVID A. BELL, MD, FRCPC,

Division of Rheumatology,

St. ph's Health Care London,

London, Ontario, Canada.

--------------------------------------------------------------------------------

While intuitively the early diagnosis of any disease is important, it has

now become dogma that the early diagnosis and treatment of rheumatoid

arthritis (RA) is critical to prevent crippling. Accordingly, if one accepts

this view, there is an increasing need to have readily available tools to

identify early RA. The challenge in very early RA is to distinguish this

disease from other disorders that mimic it or are transient (e.g., viral

arthritis) or are not anticipated to lead to destructive arthritis [e.g.,

systemic lupus erythematosus (SLE)]. Perhaps a more important objective is

to also identify which patients with RA are likely to have an aggressive

course of their disease. The serologic measurement of anti-citrulline

antibodies has been proposed as a means to achieve these objectives, since

the test has an apparent high specificity for established RA and is at least

as sensitive for the diagnosis as rheumatoid factor (RF). While

anti-citrulline antibodies have been known to be associated with RA for over

40 years1, it is only recently that their measurement has been facilitated

by a relatively simple ELISA employing a proprietary cyclic polypeptide,

CCP2 2.

In this issue of The Journal, Matsui and coworkers3 examine the performance

of a series of RA-associated markers, including anti-CCP2 antibodies, in a

cohort of patients with early RA (very early or < 6 months and relatively

early or < 2 years) compared to patients with other disorders that could

mimic RA. These included SLE, mixed connective tissue disease, and Sjögren's

syndrome (SS). The authors found that anti-CCP2 antibodies had a sensitivity

greater than 67% for the diagnosis of very early RA, and 82% for the overall

diagnosis of RA, among a group of Japanese patients seen in one academic

medical center. Other diagnostic markers were also prevalent in this

population group, including RF (84%) and anti-galactosyl IgG antibodies

(approximately 90%). However, anti-CCP2 antibodies were more specific for RA

than any combination of other markers (anti-agalactosyl IgG, RF, matrix

metalloproteinase-3). Further, anti-CCP2 antibodies had a positive

predictive value of approximately 91% for RA. Importantly, RF was positive

in 61% of anti-CCP2-negative patients with very early RA, while anti-CCP2

was positive in 22% of RF-negative subjects. The authors concluded that

anti-CCP2 antibodies are the single most useful test for the overall

diagnosis of RA, but that a combination of anti-CCP2 and RF is more useful

than either test alone for the diagnosis of very early RA.

How do these results compare to the reports from other population groups

with RA? The sensitivity of anti-CCP2 antibodies for the diagnosis of RA in

most other reported series, particularly in Caucasians, is generally lower

(60%-70%) and the specificity for RA is generally higher (95%-98%)4. In the

series of patients reported here by Matsui, et al, a relatively high

frequency of anti-CCP2 antibodies was also seen in patients with non-RA

connective tissue disorders such as SLE (15%), SS (14%),

polymyositis/dermatomyositis (23%), and scleroderma (16%). The median level

of anti-CCP2 antibodies was, however, generally lower in these non-RA

patients (7-35 units per ml) compared to RA (100 units per ml or more).

These are surprising results, and they urge some caution in generalizing

from the conclusions of the studies in this somewhat selected group of

Japanese patients with RA, since some of the patients with anti-CCP2

antibodies may have arthritis that can clinically mimic early RA but that is

not generally recognized to lead to joint destruction. This may account for

the fact that among non-Asian population groups previously studied, the

specificity of anti-CCP2 antibodies for RA has been reported to be higher

than in the studies by Matsui, et al. These authors also suggested the

interesting possibility that the common denominator predicting the

likelihood of positive anti-CCP2 antibodies was the occurrence of

inflammatory arthritis, regardless of its association with disease. While

there is no evidence yet supporting this hypothesis, there are a growing

number of reports of anti-citrulline antibody in non-RA subjects.

Symmons, et al5 have recently stated that to date, " no set of predictive

criteria has been able to discriminate between individuals destined to

develop ultimate RA (those with persistent destructive disease) and those

not developing RA. " These authors assert that the majority of patients with

early inflammatory polyarthritis have undifferentiated disease, and that the

only justification for using American College of Rheumatology (ACR) criteria

or any criteria to define RA in patients with early inflammatory

polyarthritis is to satisfy the insistence of manuscript reviewers! They

have argued that the label " rheumatoid arthritis " should be reserved for

those with " established disease " and that " its use in the first few months

of early disease should be dropped. "

There are a few reports of anti-citrulline antibodies (measured by CCP2) in

psoriatic arthritis6,7 and even in psoriasis without arthritis8. We have

identified patients who appear to have primary SS with an inflammatory

polyarthritis who have anti-citrulline antibodies (by CCP2 assay)9. Many of

these patients had clinical evidence of erosive or destructive arthritis

similar to RA. The occurrence of anti-citrulline antibody in human RA is

closely linked to the expression of the RA shared epitope (SE). HLA typing

among these patients with anti-Ro/SSA-positive SS with anti-citrulline

antibody revealed that many expressed one copy of the SE but only on one

parental allele and DR3 (often linked to anti-Ro/SSA antibody expression) on

the other allele. It has been pointed out that citrullination (due to

deimination by the enzyme peptidyl arginine deiminase of arginine residues

in certain polypeptides) may be a nonspecific event linked to inflammation

in general10. The ability to produce anti-citrulline antibodies

experimentally, however, is linked to the expression of the SE, as we have

recently shown in DR4 transgenic mice11. Therefore, the presence of synovial

inflammation in individuals with the appropriate genetic background (the SE)

may be necessary to generate an immune response to citrullinated

polypeptides. We did not observe anti-citrulline antibodies among a

carefully selected group of patients with psoriatic arthritis who lacked the

SE12.

Consistent with these latter experimental observations11 are numerous

reports linking the presence of anti-citrulline antibodies in human RA to

the expression of MHC class II molecules with the SE13-15. While this

combination of circumstances is more likely to occur in what we understand

as RA, it is entirely conceivable that patients with other disorders not

classically recognized as fulfilling ACR criteria for this diagnosis may

have anti-citrulline antibodies. Van der Helm-van Mil and coworkers16

recently provided evidence that SE alleles did not independently contribute

to RA among a group of patients with undifferentiated arthritis, but rather,

influenced the development of anti-citrulline antibodies. The level of

anti-citrulline antibodies was higher among patients with the SE than in

those without this genetic marker. Similarly, Inigoyen and coworkers15

recently reported that the SE is strongly associated with the production of

anti-CCP antibodies, but not RF, and that the odds ratio of having

anti-citrulline antibodies is doubled when 2 copies of the SE are expressed.

These observations, taken together, are consistent with the hypothesis that

SE expression is linked directly to the immune response to citrulline,

confirming the experimental observations11. Given these findings, the

expression of anti-citrulline antibody may be more closely linked to the SE

than to RA itself, and could account for the growing number of observations

that anti-citrulline antibodies may be present in patients with other

diseases who express the SE. It appears, therefore, that the dust has not

yet settled on the issue of whether anti-citrulline antibodies, usually

tested by reactivity to CCP2, perform as an ideal marker for the diagnosis

of early RA.

A second issue is whether or not anti-citrulline antibodies are useful in

predicting severity of RA. This had been alluded to earlier. There are a

number of studies among Caucasians that suggest this association17 and with

the SE. It is possible, as stated, that the linkage here is with

anti-citrulline antibodies. That is, given the studies described above, it

is possible that the association of the SE with more severe disease is a

reflection of its influence on the production of anti-citrulline antibodies.

Moreover, direct experimental evidence exists that anti-citrulline

antibodies are pathogenic. First, the studies by Kuhn and collaborators,

showing that in mice, citrullinated collagen II enhances its

arthritogenicity compared to uncitrullinated collagen II, support this

possibility18. Second, we have shown that mice transgenic for the SE (but

not non-SE transgenic mice) can be induced to develop inflammatory arthritis

following immunization with citrullinated but not non-citrullinated

fibrinogen19. However, the thesis that anti-citrulline antibodies are

pathogenic in RA, and predict severity, is difficult to reconcile with their

observed appearance years before the onset of obvious RA20. This suggests

that additional factors influence the expression of synovitis and its

severity.

The rediscovery of anti-citrulline antibodies and their relationship to RA

has opened a new chapter in our understanding of this disease. Like the

earlier discovery of RF, the strong association of anti-citrulline

antibodies in RA challenges us to examine how the immune response to

citrullined polypeptides is involved in the pathogenesis of the disease, and

how reliable anti-citrulline antibodies are in its early diagnosis and in

predicting its progression. More work is needed.

http://www.jrheum.com/subscribers/06/12/2369.html

Not an MD

I'll tell you where to go!

Mayo Clinic in Rochester

http://www.mayoclinic.org/rochester

s Hopkins Medicine

http://www.hopkinsmedicine.org