Background/Purpose:

Periodontitis (PD) has been implicated as an etiologic risk
factor in rheumatoid arthritis (RA), an association that is speculated to be
related to the oral pathogen Porphyromonas gingivalis (Pg).  Among prokaryotes,
Pg has the unique ability to citrullinate
antigens, and anti-Pg immunity in established RA has
been linked to the expression of anti-citrullinated
protein antibody (ACPA).  To date, no studies have investigated whether an
association exists between anti-Pg antibody and
pre-clinical ACPA expression or the future development of RA.  The
objective of this study was to evaluate the association between anti–Pg antibodies and development of RA as well as RF and
anti-CCP2 antibodies in preclinical RA.

Methods:

A longitudinal case-control design was used to study 73
U.S. military personnel who eventually developed seropositive RA and 73
controls, matched on age at diagnosis, sex, race, number of samples available,
duration of sample storage and enlistment region. Stored serum samples taken
prior to RA diagnosis were tested for RF (isotypes
and nephelometry), anti-CCP2 antibody, and antibody
to Pg outer membrane antigen. Generalized estimating
equations (GEE) were used to determine the association of anti-Pg antibody with RA case status. In secondary analysis,
mixed effects linear regression was used to determine the association between
anti-Pg antibody and autoantibody
concentrations in preclinical RA. All antibody levels were log-transformed for
analysis.

Results:

At the time of diagnosis, RA cases had a mean (±SD) age of
40 (10) years, 59% were men, and a majority were Caucasian (67%).  Cases
and controls had a mean of 3.5 (1.3) samples available for analysis (2.9 ± 1.3
pre-diagnosis among cases) with a median of 1.4 (IQR 0.8-2.7) years from the
last sample prior to diagnosis to diagnosis.  As shown in the Table, there
was no association of pre-clinical anti-Pg antibody
and RA case status (p=0.32).  Among cases, there was no association
between anti-Pg and anti-CCP2 antibody concentrations
(p=0.63). However, anti-Pg antibody levels were
significantly associated with RF measured by nephelometry
(p=0.009).  Although not reaching statistical significance, isotype analysis demonstrated the strongest association
between anti-Pg antibody and RF IgM
concentration. 

Conclusion:

There was no evidence in this study to support an
association of anti-Pg antibody with the risk of
developing seropositive RA nor was anti-Pg antibody
associated with circulating ACPA during the pre-clinical period.  The
association of anti-Pg antibody with RF warrants
further investigation, particularly in light of recent work demonstrating that
RF acts synergistically with ACPA in promoting inflammation in RA.