Article Text

Download PDFPDF

Concise report
Anti-carbamylated protein (anti-CarP) antibodies precede the onset of rheumatoid arthritis
  1. Jing Shi1,
  2. Lotte A van de Stadt2,3,
  3. E W Nivine Levarht1,
  4. Tom W J Huizinga1,
  5. Dörte Hamann4,
  6. Dirkjan van Schaardenburg2,5,
  7. René E M Toes1,
  8. Leendert A Trouw1
  1. 1Department of Rheumatology, Leiden University Medical Center, Leiden, The Netherlands
  2. 2Division of Rheumatology, Jan van Breemen Research Institute | Reade, Amsterdam, The Netherlands
  3. 3Sanquin Research and Landsteiner Laboratory, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
  4. 4Division Diagnostic Services, Sanquin Blood Supply, Amsterdam, The Netherlands
  5. 5Department of Rheumatology, VU University Medical Center, Amsterdam, The Netherlands
  1. Correspondence to Dr L A Trouw, Department of Rheumatology, Leiden University Medical Center, C1-R, LUMC, PO Box 9600, Leiden 2300 RC, The Netherlands; l.a.trouw{at}


Objective The presence of anti-citrullinated protein antibodies (ACPA) and IgM-rheumatoid factor (IgM-RF) years before the clinical diagnosis of rheumatoid arthritis (RA) suggests they are possibly involved in the pathogenic process underlying RA. In this study, we analysed whether anti-carbamylated protein (anti-CarP) antibodies, a novel autoantibody system against carbamylated proteins, can also be detected in healthy individuals before they developed RA.

Methods Multiple sera from asymptomatic blood donors prior to the onset of their RA symptoms and sera from age-matched and sex-matched controls were tested for the presence of antibodies directed against carbamylated-fetal calf serum (Ca-FCS), carbamylated-fibrinogen (Ca-Fib), cyclic citrullinated-peptide 2 and IgM-RF.

Results Anti-Ca-FCS and anti-Ca-Fib antibodies were each present in 27% and 38% of the last serum samples of blood donors prior to the diagnosis of RA. Both anti-Ca-FCS and anti-Ca-Fib antibodies could be detected many years before the onset of RA. Anti-CarP antibodies as well as ACPA are, on average, detected earlier than IgM-RF.

Conclusions In addition to ACPA and IgM-RF, also the newly identified anti-CarP antibodies appear many years before the diagnosis of RA.

  • Early Rheumatoid Arthritis
  • Autoantibodies
  • Ant-CCP
  • Rheumatoid Factor

Statistics from


Rheumatoid arthritis (RA) is a systemic autoimmune disease primarily affecting the synovial joints.1 Early and aggressive intervention in individuals who are developing RA may prevent irreversible bone loss and induce early remission.2 Identification of individuals who could benefit most from such an intervention is a challenge, but the presence of autoantibodies could be a useful marker. Besides IgM-rheumatoid factor (IgM-RF) also anti-citrullinated protein antibodies (ACPA) are implicated in the disease progression of RA3 and are now included in the 2010 American College of Rheumatology (ACR)/The European League Against Rheumatism criteria for RA.4 Both ACPA and RF can be detected in serum many years prior to the onset of symptoms.57 Recently, we identified a novel autoantibody family consisting of anti-carbamylated protein (anti-CarP) antibodies.8 Carbamylation is a post-translational modification in which lysines are chemically converted to homocitrullines.9 The presence of anti-CarP antibodies is associated with more radiological progression over time in ACPA-negative RA patients8 and with the conversion from arthralgia to RA.10

In this study, we have analysed whether anti-CarP antibodies are already present in healthy individuals before the diagnosis of RA and how the appearance of anti-CarP antibodies relates to the appearance of ACPA and IgM-RF.


Study population

Sera from 79 asymptomatic blood donors (48 women), with a mean age at diagnosis of 51 years, collected before they developed RA as well as from 141 age-matched and sex-matched controls were obtained as described before.6 ,11

In brief, sera of patients who fulfilled the ACR 1987 criteria for RA12 obtained before the diagnosis of RA were collected from the Sanquin Blood Bank. A median of 5 (IQR 4–6) sequential sera from blood donors obtained at 1–6-year intervals were available for analysis. The clinical information of patients was retrieved from medical records. Sera of 141 sex-matched and age-matched healthy blood donors collected in the same period as the samples of the individuals who developed RA were also collected from the Sanquin Blood Bank. All patients and controls are Caucasian. Informed consent was obtained from the participating RA patients. The study was approved by the Slotervaartziekenhuis and Reade ethical review board, Amsterdam, The Netherlands, with the condition, however, that we were not allowed to contact the control donors for reasons of privacy.

Antibody detection

Sera collected on the last visit prior to the diagnosis of RA (median time before diagnosis: 1.4 years, IQR: 0.8–1.8 years) were measured for the presence of anti-carbamylated-fetal calf serum (anti-Ca-FCS), anti-carbamylated-fibrinogen (anti-Ca-Fib), anti-cyclic citrullinated-peptide 2 (anti-CCP2) antibodies and IgM-RF. Next, of individuals positive for at least one autoantibody, all consecutive samples were analysed.

Anti-Ca-FCS and anti-Ca-Fib antibodies were detected by ELISA as described before.8 The anti-CCP2 ELISA (Euro-Diagnostica) was performed following the manufacturer's instructions. The IgM-RF ELISA was performed as previously described.6 ,11 The cut-off for the anti-Ca-FCS and anti-Ca-Fib ELISA was set as the mean plus two times the SD of healthy controls.


Mann–Whitney U test (one-tailed) was performed to compare the time points before the diagnosis of RA when the different autoantibodies were detectable for the first time.


The characteristics of the 79 asymptomatic blood donors who developed RA later used in this study have been described previously.6 ,11 Sera collected on the last visit prior to the diagnosis of RA were analysed for the presence of anti-Ca-FCS, anti-Ca-Fib, anti-CCP2 antibodies and IgM-RF. At the time of diagnosis, the patients reported a 0.9-year median duration of joint symptoms. Twenty-one (26.6%) donors were positive for anti-Ca-FCS antibodies, 30 (38.0%) for anti-Ca-Fib antibodies, 33 (41.8%) for anti-CCP2 antibodies and 19 (24.0%) for IgM-RF (figure 1A). The numbers of single-positive, double–positive and triple-positive donors are listed in figure 1B.

Figure 1

Anti-carbamylated protein (Anti-CarP) antibodies are present in asymptomatic blood donors before they developed rheumatoid arthritis (RA). (A) Anti-carbamylated-fetal calf serum (anti-Ca-FCS), anti-carbamylated-fibrinogen (anti-Ca-Fib), anti-cyclic citrullinated-peptide 2 (anti-CCP2) antibodies and IgM-rheumatoid factor (IgM-RF) are each present in 27%, 38%, 42% and 24% of the serum samples taken from donors most recently before their diagnosis of RA. (B) The absolute numbers of single-positive, double-positive, triple-positive and autoantibody-negative donors are listed. ACPA, anti-citrullinated protein antibodies.

Analysis of the longitudinal samples from the donors who were positive for at least one autoantibody at the last visit prior to the diagnosis of RA revealed that similar to ACPA and IgM-RF also anti-Ca-FCS and anti-Ca-Fib antibodies can be detected years before the diagnosis of RA (figure 2A). There is a steady increase in the percentage of individuals with a positive antibody status for each of the autoantibodies studied over time. The moment when IgM-RF, anti-Ca-FCS, anti-Ca-Fib and anti-CCP2 antibodies were detectable for the first time were 10, 14, 14 and 14 years before diagnosis, respectively (figure 2A,B). These are the time points when the first samples of these donors were collected. The median (IQR) of time points when anti-CCP2, anti-Ca-FCS, anti-Ca-Fib antibodies and IgM-RF were detectable were 6 (3–10), 5 (3–7), 7 (4–10) and 2 (1–5) years before the diagnosis (figure 2B). Comparing the time points when each of these three autoantibody families could be detected for the first time in this cohort revealed that anti-CarP and anti-CCP2 antibodies appeared around the same time, while IgM-RF appeared significantly later (figure 2B). We observed an increase in the number of autoantibody reactivities present in the samples over time. Six years before the diagnosis, more than 50% of the autoantibody-positive donors harboured only one autoantibody family, whereas at least 30% of the autoantibody-positive donors displayed all three autoantibody families within 4 years before diagnosis (figure 2C).

Figure 2

The development of autoantibodies over time. (A) The percentage of autoantibody-positive donors increases over time. (B) The first time point that IgM-rheumatoid factor (IgM-RF) appeared is significantly later than anti-cyclic citrullinated-peptide 2 (anti-CCP2) and anti-CarP antibodies reacting with Ca-FCS and with Ca-Fib. (C) The percentages of blood bank donors that harbour 0, 1, 2 or 3 autoantibody families are calculated. The number of autoantibody-positive donors and the number of recognised autoantibodies of autoantibody-positive donors increase over time.

Next to positivity, the levels of autoantibodies increased within the majority of followed blood donors. The levels of anti-CCP2, anti-Ca-FCS, anti-Ca-Fib antibodies and IgM-RF levels each increased in 83% (33/40), 78% (31/40), 53% (21/40) and 85% (34/40) of donors. The median levels of all autoantibodies start to increase around 5–7 years before the diagnosis or RA (figure 3).

Figure 3

The median levels of autoantibodies increase over time. Median level of each autoantibody was calculated for every year. The median levels at time point 0 of all autoantibodies were set at 100%. The median levels of other time points were presented as the percentage of the median level of time point 0. The number below the x-axis is the number of all available samples at the given time points. Time points before 10 years were not shown since the total sample number is too low. (A) Anti-Ca-FCS, (B) Anti-Ca-Fib, (C) anti-cyclic citrullinated-peptide 2 (anti-CCP2) and (D) IgM-rheumatoid factor (IgM-RF).


ACPA and IgM-RF have been reported to be present many years before the onset of the clinical symptoms of RA.57 Here, we observed that also anti-CarP antibodies are present before the onset of the clinical symptoms of RA. The moment of first appearance of anti-CarP antibodies is comparable with ACPA and earlier than IgM-RF. Levels of each of these autoantibodies appear to increase with similar kinetics. From the current data, it is not possible to conclude on the relative contribution of each of these antibodies on the pathogenesis of RA.

Incidences and levels of anti-CCP2 antibodies and IgM-RF in this report are slightly different from previous reports6 ,11 on this cohort, because of a different approach in patient follow-up. In the present report, only donors who had a positive test in the last sample before the diagnosis of RA were followed, while previous reports mentioned cumulative percentages of positivity in all samples.

Homocitrulline is rather similar to citrulline but is one methylene group longer. Nonetheless, anti-CarP antibodies and ACPA appear by and large as two different autoantibody systems as anti-CarP antibodies are present in both ACPA-positive and ACPA-negative RA patients. In addition, inhibition studies, ACPA depletion studies and western blot analyses have revealed that next to a ‘cross-reactive’ component both antibody families have components recognising only citrulline-containing or homocitrulline-containing proteins/peptides.8 ,13 Compared with anti-Ca-FCS antibodies, anti-Ca-Fib antibodies may have more cross-reactivity to citrullinated proteins.13 In this study, we observed five blood bank donors who were ACPA negative but positive for either anti-Ca-FCS (1) or anti-Ca-Fib antibodies (4). We also observed six donors who were only positive for ACPA. Furthermore, ACPA and anti-CarP antibodies do not appear at the same moment in individuals as in 26 double-positive donors, 13 donors developed anti-CarP antibodies first, 16 donors developed both at the same time and 6 donors developed ACPA first. Collectively these observations strengthen the concept that next to a cross-reactive component also non-cross-reactive antibodies exist. Because of the limited sample size of 79 individuals, replication studies in comparable cohorts are warranted.

Since autoantibodies can be detected many years before the onset of clinical symptoms, it is tempting to speculate that these antibodies may contribute to the pathogenesis of RA, for instance, by inducing osteoclastogenesis, tumour necrosis factor (TNF)-α secretion by macrophages or complement activation.1416 Our finding that anti-CarP antibodies can be present years before diagnosis adds another source of antibodies, which can potentially contribute to RA pathogenesis. Several genetic and environmental factors, such as shared epitope, PTPN22 and smoking, have been reported to associate with ACPA-positive RA as recently reviewed.17 It would be interesting to know whether similar risk factors also predispose to anti-CarP antibodies.

The assays used to detect ACPA, anti-CarP antibodies and IgM-RF may differ in definition of cut-off and sensitivity, potentially limiting the comparability. Interestingly, IgM-RF seems to occur pathogenically subsequent to ACPA or anti-CarP antibodies. Observations made in American-Indian RA patients and their first-degree relatives may support this hypothesis.18 These studies revealed that although either anti-CCP antibodies or IgM-RF can be found in both patients and their healthy first-degree relatives, the combination of both anti-CCP and RF is predominantly found in the RA patients. This supports the notion that ACPA and anti-CarP antibodies may initiate the primary target recognition but that amplification of IgM-RF is important for progression towards clinical RA, for example, by enhancing complement activation.16 ,19 ,20

In conclusion, we discovered that anti-CarP antibodies can be present in healthy blood donors many years before the development of clinical symptoms of RA.


We acknowledge the excellent support of Margret de Koning regarding sample storage and handling.


View Abstract


  • Handling editor Tore K Kvien

  • Contributors JS has contributed to study conception and design, acquisition, analysis and interpretation of data, drafting the article and revising it. LAvdS and LAT have contributed to study conception and design, analysis and interpretation of data, drafting the article and revising it. EWNL has contributed to study conception and design, acquisition of data and revising the article. TWJH, DvS and REMT have contributed to study conception and design, analysis and interpretation of data and revising the article. DH has contributed to study conception and design, acquisition, analysis and interpretation of data and revising the article.

  • Funding The financial support was from the Dutch Arthritis Foundation, The Netherlands Organisation for Scientific Research, Masterswitch project FP7, the IMI JU funded project BeTheCure, contract no 115142-2, Pfizer, The Netherlands Proteomics Center and the Center for Medical Systems Biology as part of The Netherlands Genomics Initiative. LAT is supported by a ZON-MW Vidi grant and by a fellowship from Janssen Biologics.

  • Competing interests LAT is supported by a fellowship from Janssen Biologics.

  • Ethics approval Slotervaartziekenhuis and Reade ethical review board.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.