Article Text

Anti-carbamylated proteins antibody repertoire in rheumatoid arthritis: evidence of a new autoantibody linked to interstitial lung disease
  1. Raul Castellanos-Moreira1,
  2. Sebastian Cruz Rodríguez-García1,
  3. Maria Jose Gomara2,
  4. Virginia Ruiz-Esquide1,
  5. Andrea Cuervo1,
  6. Ivette Casafont-Solé3,
  7. Julio Ramírez1,
  8. Susana Holgado3,
  9. Jose A Gómez-Puerta1,
  10. Juan D Cañete1,
  11. Isabel Haro2,
  12. Raimon Sanmarti1
  1. 1 Rheumatology Department, Arthritis Unit, Hospital Clinic de Barcelona, Barcelona, Spain
  2. 2 Consejo Superior de Investigaciones Científicas, Unit of Synthesis and Biomedical Applications of Peptides, CSIC-IQAC, Barcelona, Spain
  3. 3 Rheumatology Department, Hospital Germans Trias i Pujol, Badalona, Spain
  1. Correspondence to Dr Raimon Sanmarti, Arthritis Unit, Rheumatology Department, Hospital Clinic de Barcelona, Barcelona 08036, Spain; sanmarti{at}


Objective To analyse the association between anti-carbamylated protein antibodies (Anti-CarP) and interstitial lung disease (ILD) in rheumatoid arthritis (RA) patients.

Methods Cross-sectional study including RA patients fulfilling the 2010 ACR/EULAR criteria. The main population comprised two groups: (1) RA patients diagnosed with RA-ILD (RA-ILD group); (2) RA patients without ILD (non-ILD RA group). Non-ILD RA patients in whom ILD was suspected underwent a diagnostic work-up and, if ILD was diagnosed, were switched to the RA-ILD group. ILD was diagnosed by high-resolution computed tomography and confirmed by a multidisciplinary committee. An independent replication sample was also obtained. Three Anti-CarP IgG autoantibodies against fetal calf serum (Anti-FCS), fibrinogen (Anti-Fib) and chimeric fibrine/filagrine homocitrullinated peptide (Anti-CFFHP) and one Anti-CarP IgA against FCS (Anti-FCS-IgA) were determined by home-made ELISA. Associations between Anti-CarP and ILD were analysed using multivariable logistic regression adjusted by smoking, sex, age, RA disease duration, rheumatoid factor and anticitrullinated protein antibodies.

Results We enrolled 179 patients: 37 (21%) were finally diagnosed with RA-ILD. Anti-CarP specificities were more frequent in RA-ILD patients (Anti-FCS 70% vs 43%; Anti-Fib 73% vs 51%; Anti-CFFHP 38% vs 19%; Anti-CarP-IgA 51% vs 20%, p<0.05 for all comparisons). Serum titers of Anti-CarP were significantly higher in RA-ILD patients. Anti-CarP specificities showed a robust effect towards increasing the odds of ILD in the multivariate analysis (Anti-FCS (OR: 3.42; 95% CI: 1.13 to 10.40), Anti-Fib (OR: 2.85; 95% CI: 0.83 to 9.70), Anti-CFFHP (OR: 3.11; 95% CI: 1.06 to 9.14) and Anti-FCS-IgA (OR: 4.30; 95% CI: 1.41 to 13.04)). Similar findings were observed in the replication sample.

Conclusions Anti-CarP were strongly associated with ILD. The role of homocitrullination in RA-ILD merits further investigation.

  • rheumatoid arthritis
  • smoking
  • autoantibodies
  • pulmonary fibrosis
  • anti-CCP

Statistics from

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.

Key messages

What is already known about this subject?

  • Rheumatoid arthritis-associated interstitial lung disease (RA-ILD) entails a high mortality. Therefore, an early detection is crucial in establishing an individualised treatment strategy.

  • Anti-carbamylated protein antibodies (Anti-CarP) are associated with poor disease outcomes in RA patients and have been detected in various chronic lung diseases regardless of the RA history.

What does this study add?

  • For the first time an association between RA-ILD and Anti-CarP has been found, which was independent of smoking, sex, age, RA disease duration, anticitrullinated protein antibody and rheumatoid factor.

How might this impact on clinical practice or future developments?

  • These findings suggest a possible link between homocitrullination and the development of ILD in RA patients exists.


Interstitial lung disease (ILD) is a severe extra-articular manifestation of rheumatoid arthritis (RA), that affects 4%–50% of RA patients depending on the screening method and the population examined.1 RA-associated ILD (RA-ILD) entails a shortened survival with a mortality rate up to 10 times higher than those without ILD,2 prompting a great effort to achieve an earlier diagnosis.

Various risk factors have been identified, including smoking, male sex, higher disease activity, longer disease duration, older age, positive rheumatoid factor (RF) and anticitrullinated protein antibodies (ACPAs).1 3 In recent years, several biomarkers have been proposed for RA-ILD screening,4 5 although there is no universally accepted screening test.

Anti-carbamylated protein antibodies (Anti-CarP) recognise homocitrullinated peptides, which are generated by a post-translational modification (PTM) of lysine residues.6 First described in 2011, Anti-CarP have been associated with poor disease outcomes in RA patients, including higher disease activity and radiographic progression.7–9 A recent study found higher mortality in Anti-CarP positive RA patients, which was particularly attributed to respiratory diseases.10 Interestingly, Anti-CarP have been found in patients with various non-ILD chronic lung diseases irrespective of smoking or RA history.11 12

The objective of this study was to analyse the association between Anti-CarP and ILD in a population of RA patients.


Study design

We performed a cross-sectional study including RA patients diagnosed according to the 2010 ACR/EULAR criteria assessed in a rheumatology department outpatient clinic of a tertiary university hospital. Individuals fulfilling other inflammatory arthritis or connective tissue disease diagnostic criteria were excluded.

Study population

The main population comprised two groups; (1) patients diagnosed with RA-ILD before study inclusion (RA-ILD group); (2) consecutive patients with RA with no diagnosis of ILD, assessed by a single rheumatologist (RC-M) between July 2017 and July 2018 (non-ILD RA group).

All non-ILD RA patients in whom ILD was suspected, underwent an appropriate work-up (see RA-ILD diagnosis section). Subjects resulting in a new diagnosis were switched to the RA-ILD group. Criteria for suspicion were based on the presence of RA-ILD risk factors or persistent symptoms such as dyspnoea or cough. An independent replication sample with similar characteristics was obtained from another hospital.

RA-ILD diagnosis

RA-ILD was diagnosed using high-resolution computed tomography and classified according to the American Thoracic Society/European Respiratory Society 2013 multidisciplinary classification criteria of idiopathic interstitial pneumonias.13 Diagnoses were confirmed by a multidisciplinary committee including pneumologists, radiologists, pathologists, clinical immunologists, internists and rheumatologists.

Demographic, clinical, therapeutic and serological data

In the main population, the variables analysed included demographics (age, sex and ethnicity), disease duration at study inclusion, smoking status and smoking cumulative dose (number of packs per day×number of years of smoking). Disease activity and disability were assessed using the Disease Activity Score in 28 joints (DAS28) and the Health Assessment Questionnaire Disability Index (HAQ-DI), respectively. Current conventional synthetic and biological disease-modifying antirheumatic drugs (csDMARDs and bDMARDs, respectively) and glucocorticoid use were evaluated. Hand and foot X-rays were obtained at study entry for analysis of radiological damage.

Antibodies measurement

Autoantibody status was measured in sera collected at study enrolment. Three carbamylated antigens, namely fetal calf serum (FCS), fibrinogen (Fib) and chimeric fibrin/filaggrin homocitrullinated peptide (CFFHP) were obtained. CFFHP and its non-homocitrullinated version were synthesised by solid-phase peptide synthesis following a 9-fluorenylmethoxycarbonyl/t-butyl strategy with subsequent cyclisation in solution by forming a disulfide bridge.14 15 Four Anti-CarP autoantibodies were determined by home-made ELISA tests with either carbamylated proteins or CFFHP as antigens, respectively; three IgG specificities namely Anti-FCS, Anti-Fib and Anti-CFFHP and one IgA specificity against carbamylated FCS (Anti-FCS-IgA). Receiver operating characteristic (ROC) curve analysis and regression analysis were conducted using the GraphPad Prism5 programme and the cut-off values were determined with a specificity of 95% compared with a healthy population of blood donors (n=179). A series of successive dilutions of a pool of sera from four positive patients was used as a reference standard in all plates and to convert optical density (OD) values to arbitrary units (AU).

A positive cut-off value was defined as ≥173.5 AU/mL, ≥166.9 AU/mL, ≥146.5 AU/mL and ≥257.0 AU/mL for Anti-FCS, Anti-Fib, Anti-CFFHP and Anti-FCS-IgA, respectively. A test was only considered positive and specific for homocitrulline when the AU/mL values were higher than the respective cut-off and the OD value difference between carbamylated (homocitrullinated) and native (non-homocitrullinated) antigens was at least 0.1 in agreement with the methodology previously reported.16 Further details on the ELISA techniques are provided in the online supplementary text.

RF was measured by nephelometry and ACPA was assessed by anti-CCP3 chemiluminescent immunoassay (QUANTA Flash CCP3, Inova Diagnostics). The cut-offs for positive values were 25 international units (IU) and 20 chemiluminescence units (CU), respectively.

Statistical analysis

Between-group differences were analysed using descriptive statistics as appropriate. Proportions were compared using the χ² or Fisher’s exact test. Continuous variables were analysed using the Student’s t-test or the Mann–Whitney U test and presented as means with SD (±) or medians with IQR.

The association between Anti-CarP and ILD was evaluated using logistic regression including a set of variables known to be related to the development of ILD such as smoking, sex, age, RA disease duration, ACPA and RF and others taken from the univariate analysis (p<0.2). Potential confounders such as treatments or disease activity were not included in the models because of concerns regarding timing, dosing and study design. Different models were built using the all possible equations approach and the final models were chosen according to the Akaike information criterion and the area under the ROC curve after assessing possible first order interactions.

Exploratory analysis, including the association between Anti-FCS-IgA and RA-ILD, and the unadjusted assessment of the overlap between antibodies and the association between Anti-CarP and smoking were performed in the main population.

There was only one observation missing for HAQ in the main population which was imputed using the average HAQ values in the respective group.

Two-tailed p values <0.05 were established to be statistical significance in all analyses, which were performed using both IBM SPSS for Windows V.23.0 and STATA V.15.


Signed informed consent was obtained from all patients before study enrolment. The Strengthening the Reporting of Observational Studies in Epidemiology guidelines were followed.


Study population baseline features

One hundred and seventy-nine patients were enrolled in the main population: 31 in the RA-ILD group and 148 in the non-ILD group. The diagnostic work-up resulted in six new ILD diagnoses: these patients were relocated to the RA-ILD (see figure 1). Thus, 37 patients (21%) were finally included in the RA-ILD group and 142 in the non-ILD RA group. The replication sample was composed of 25 patients with RA-ILD and 50 RA patients without ILD. The baseline features of the main population and the replication sample are summarised in table 1 and online supplementary table S1, respectively.

Figure 1

Main population flow chart. ILD, interstitial lung disease; RA, rheumatoid arthritis.

Table 1

Main population demographic, clinical and therapeutic features

In the main population, RA-ILD patients had a mean age at ILD diagnosis of 64.2 (±9.7) years and a mean time from RA diagnosis to ILD diagnosis of 8.5 (±7.4) years. Usual interstitial pneumonia (38%) and non-specific pneumonia (38%) were the most frequent ILD patterns, followed by respiratory bronchiolitis-associated ILD (11%), cryptogenic organised pneumonia (8%) and desquamative interstitial pneumonia (5%).

Subjects within the RA-ILD group were older and had a longer disease duration than their counterparts at inclusion (p<0.005) as well as greater disease activity (DAS28 3.71 (±1.35) vs 2.74 (±1.05), p<0.005), a higher proportion of erosive disease (70% vs 44%, p<0.005) and functional disability (HAQ-DI ≥1; 27% vs 12%, p value: 0.02). No differences in current (19% vs 16%, p value: 0.69) and ever-smoking (57% vs 44%, p value: 0.15) were observed, although patients with RA-ILD had a higher smoking cumulative dose (30.7 (±11.1) vs 21.8 (±12.0) packs/year, p<0.005). No between-group differences in the proportions of current csDMARDs (89% vs 86%, p value: 0.60), bDMARDs (30% vs 25%, p value: 0.59) or glucocorticoids (70% vs 58%, p value: 0.17) were observed.

Anti-CarP and RA-ILD

The proportions of positive Anti-CarP in the main population were 49%, 56%, 23% and 27% for Anti-FCS, Anti-Fib, Anti-CFFHP and Anti-FCS-IgA, respectively. All Anti-CarP fine specificities were more frequent in the RA-ILD group (Anti-FCS: 70% vs 43%; Anti-Fib: 73% vs 51%; Anti-CFFHP: 38% vs 19%; Anti-FCS-IgA: 51% vs 20%, p<0.05 for all comparisons) (table 2). Similar results were observed in the replication sample (Anti-FCS: 92% vs 48%, p<0.005; Anti-Fib: 76% vs 58%, p value: 0.12; Anti-CFFHP: 36% vs 18%, p value: 0.08).

Table 2

Main population autoantibody status in patients with and without ILD

In addition, the mean titers were higher in the RA-ILD group, ranging from approximately 1.3 for Anti-Fib to nearly fourfold for Anti-CFFHP; the differences were statistically significant for Anti-FCS, Anti-CFFHP and Anti-FCS-IgA (figure 2A–C,F).

Figure 2

Boxplots of autoantibody titers in patients with and without ILD. ACPA, anticitrullinated protein antibody; Anti-CFFHP, IgG antibodies against chimeric fibrine/filagrine homocitrullinated peptide; Anti-FCS, IgG antibodies against carbamylated fetal calf serum; Anti-Fib, IgG antibodies against carbamylated fibrinogen; Anti-FCS-IgA, IgA antibodies against carbamylated fetal calf serum immunoglobulin A; AU, arbitrary units; CU, chemiluminescence units; ILD, interstitial lung disease; IU, international units; RA, rheumatoid arthritis; RF, rheumatoid factor.

A logistic regression model adjusted for age, RA disease duration, ACPA, RF, sex and smoking cumulative dose showed that Anti-FCS (OR: 3.42; 95% CI: 1.13 to 10.40), Anti-CFFHP (OR: 3.12; 95% CI: 1.06 to 9.14) and Anti-FCS-IgA (OR: 4.30; 95% CI: 1.41 to 13.04) were independently associated with ILD. The results of the analysis for Anti-Fib (OR: 2.85; 95% CI: 0.83 to 9.71) followed a similar trend, although not significant. This tendency was confirmed in the replication sample; Anti-FCS (OR: 10. 42; 95% CI: 1.68 to 64.46); Anti-Fib (OR: 1. 65; 95% CI: 0.40 to 6.86) and Anti-CFFHP (OR: 1. 49; 95% CI: 0.42 to 5.29) (table 3).

Table 3

Logistic regression models for the association between Anti-CarP specificities with RA-ILD


No differences were found in the proportion of patients who tested positive for ACPA (78% vs 70%, p value: 0.29) or RF (76% vs 59%, p value: 0.06) according to ILD status. Numerically higher titers of both antibodies (median (IQR), 674 (2215) vs 143 (1132) CU, p value: 0.07) and (105 (298) vs 34 (110) IU, p value: 0.06) ACPA and RF, respectively) were observed in the RA-ILD group (see table 2 and figure 2D,E), although statistical significance was not reached.

Overlap between Anti-CarP with ACPA and RF

The overlap between Anti-CarP specificities in ACPA and RF positive patients is shown in table 4. Among those patients negative to both RF and ACPA, a small subset of positive Anti-FCS (9%), Anti-Fib (14%), Anti-CFFHP (3%) and Anti-FCS-IgA (6%) was observed. ACPA and RF titers were significantly higher in patients seropositive for Anti-FCS, Anti-Fib, Anti-CFFHP and Anti-FCS-IgA (p<0.005 for all comparisons).

Table 4

Overlap between Anti-CarP specificities with ACPA and RF in the main population

Anti-CarP and smoking

Numerically higher proportions of all Anti-CarP were observed in ever and current smokers, although the difference was only significant for Anti-FCS (58% vs 41% and 67% vs 45%, respectively), Anti-Fib (65% vs 48% and 77% vs 52%, respectively) and Anti-FCS-IgA (37% vs 18% and 50% vs 22%), p<0.05 for all comparisons. No differences in the smoking cumulative dose according to the Anti-CarP status were observed. Details in the unadjusted association between Anti-CarP and smoking are provided in online supplementary table S2 and S3.


This is the first study to analyse the association between Anti-CarP and RA-ILD. We found higher proportions and titers up to nearly fourfold greater of different Anti-CarP specificities in the RA-ILD group when compared with patients without this lung involvement. All Anti-CarP showed a consistent effect towards increasing the odds of ILD, even after adjustment for pertinent confounders using logistic regression modelling.

Compared with the general population, RA entails a 50% higher risk of mortality.17 Over the past two decades, the mortality rate has declined, mostly due to a reduction in deaths related to cardiovascular disease and malignancies,18 19 possibly attributed to an earlier diagnosis and treat-to-target strategies. Conversely, incident deaths from pulmonary disease have remained stable19 and may become the leading cause of death in RA patients. RA-ILD plays a major role in the burden of pulmonary-related deaths and constitutes an unmet need, even with available treatment options. RA-ILD patients usually have greater comorbidity and a higher incidence of infection, reflected by a higher mortality2 20 21 and a greater disability. Therefore, early detection is crucial in establishing an individualised treatment strategy.

Various novel candidates have been proposed for RA-ILD screening, including imaging techniques such as positron emission tomography and pulmonary ultrasound,22 23 or serum biomarkers such as the matrix metalloproteinases 7, surfactant protein D and Krebsvon del Lungen-6.4 5 RA-ILD has been associated with different antimodified protein antibodies (AMPAs) based on citrulline24 25 and more recently, malondialdehyde–acetaldehyde adducts.26 Screening techniques can increase RA-ILD diagnostic accuracy, although a model based on features used in clinical practice such as age, sex, smoking, RF and ACPA has a potent screening capacity for symptomatic and subclinical RA-ILD.4 Based on these findings, we decided to include these features in addition to RA disease duration as adjusting variables in our models.

The relationship between ACPA and ILD has been addressed in different studies with most showing a higher seroprevalence and titers in RA-ILD cohorts.1 27–29 However, no association between ACPA and RA-ILD has been confirmed in other cohorts30 31 and up to 17% of patients with RA-ILD are reported to be ACPA-negative.20 In a recent meta-analysis, ACPA positivity was associated with a higher risk for ILD (OR: 4.7, 95% CI: 2.1 to 10.6) although a single study contributed almost 75% to the weighted average and was not specifically concerned with RA-ILD.29 We found that ACPA and RF were numerically higher in patients with ILD, but the differences were not significant, unlike Anti-CarP.

The association between Anti-CarP and RA-ILD observed in the main population and the replication sample opens a debate on the potential role of homocitrullination (carbamylation) and its antibody response on its development. So far, no studies have evaluated homocitrulline in lung tissue of RA patients as opposed to citrulline.32 However, carbamylated peptides have been detected in the synovial of RA patients33 and lung tissue from smokers and asthma patients.34 35 Interestingly, Anti-CarP have been found in subjects with bronchiectasis, cystic fibrosis and chronic obstructive pulmonary disease with no smoking or RA history.11 12

Carbamylation can change the structure, function and antibody antigen binding avidity of a wide range of proteins.36 Thus, Anti-CarP may target homocitrullinated peptides found in the lung and promote a subsequent inflammatory response, which is essential in the genesis and progression of ILD.37 38 An alternative explanation is a direct pulmonary insult by the inflammatory cascade precipitated by an increased sensitivity of proteinases and the presence of free radicals and other inflammatory mediators enhanced by carbamylated free amino acids found in the lung.36 Notably, the association between ILD and Anti-CarP-IgA presented the greatest effect size, fuelling the hypothesis that RA-related antibodies originate in the respiratory mucosa.39

Although ACPA and Anti-CarP present important differences in their genetic40 and environmental41 backgrounds, both antibody systems form part of the AMPA family and have great structural similarity.36 As in previous studies,8 9 a substantial overlap between autoantibodies was observed in our cohort. However, inhibition studies and the fact that a subset of ACPA negative patients can harbour Anti-CarP (up to 29% in our population) support the hypothesis that the overlap is only partial.7 42 Recent evidence indicates that exposure to a purified specific PTM such as homocitrulline, acetaldehyde or citrulline can induce AMPAs against antigens containing other PTMs.43 Hence, a single PTM can generate a breaking of tolerance against multiple modified antigens.

Smoking, the leading environmental risk factor for RA and RA-ILD is also a known risk factor for breaking tolerance against multiple modified antigens.44 Smoking can induce carbamylation directly by inhalation of thiocyanate or via myeloperoxidase release by neutrophils at inflammation sites.8 34 36 Like a previous study,44 we found a higher seroprevalence of Anti-CarP among smokers. There were no between-group differences in ever or current smoking according to the presence of ILD, yet a higher smoking cumulative dose was observed. No interaction between the smoking cumulative dose and Anti-CarP specificities was observed in the regression models, suggesting that the association between Anti-CarP and ILD is not a simple surrogate marker for smoking.

Our study has some limitations. First, a relatively small proportion of patients with RA-ILD was included, so we had to be conservative while building our models because of the risk of overfitting, so residual confounding cannot be ruled-out. Treatment and disease activity were not included; the former due to concerns about the accuracy of timing and dosing and current literature’s inability to validate a true relationship between RA-ILD and methotrexate45 or bDMARDs.46–48 The latter, because of the design of the study with only a single measurement, which has not shown to be relevant in previous reports,49 50 as opposed to its longitudinal evaluation.51 Likewise, we cannot rule-out subclinical RA-ILD in the control group, mainly due to the lack of a universally accepted evidence-based screening approach. Finally, most of our patients are of Mediterranean Caucasian origin, which has been associated with a less aggressive RA course and a lower prevalence of extra-articular disease.52

In conclusion, a robust association between different Anti-CarP and RA-ILD was found after adjusting for multiple confounders including ACPA and RF. These findings pose the debate whether a link between homocitrullination and the development of this devastating extra-articular manifestation exists. However, our results should be interpreted with caution and further studies validating our findings are needed.


The authors wish to thank Loreto Carmona and Miguel Angel Descalzo for their assistance in the methodology and the Interstitial Lung Disease Committee staff at the Hospital Clinic of Barcelona for their collaboration.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Handling editor Josef S Smolen

  • Twitter @raul_cast_morei, @sdlcrodriguez

  • RC-M and SCR-G contributed equally.

  • Correction notice This article has been corrected since it published Online First. The second affiliation has been updated.

  • Contributors RC-M, SCR-G, IH and RS contributed to the conception and study design. RC-M, JR, JG-P, VR-E, IC-S, SH and JDC contributed to data collection. RC-M, SCR-G and MJG analysed the data. RC-M, SCR-G, VR-E and IH contributed to interpretation of the data. RC-M, SCR-G and RS wrote the first version of the manuscript and AC, JR, JG-P, JDC, VR-E, IC-S, SH and IH revised it critically. All authors read and approved the final manuscript.

  • Funding Financial support from the Hospital Clinic of Barcelona, Research, Innovation and Education Department (Grant # 37 933 to RC-M and the Spanish Ministry of Economy, Industry and Competitiveness and the European Regional Development Fund (Grant # RTI2018-094120-B-I00 to IH).

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Patient consent for publication Not required.

  • Ethics approval The study was conducted in accordance with the Declaration of Helsinki and was approved by the Hospital Clinic of Barcelona Ethics Committee (approval number 2017/0679).

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

  • Data availability statement Data are available upon reasonable request. Data is available upon reasonable request, all data relevant to the study are included in the article.