Rheumatoid arthritis (RA) is a chronic autoimmune inflammatory disease characterized by joint inflammation, which causes symmetric stiffness, pain, swelling, motion limitation, and progressive cartilage and bone damage. This occurs due to inflammatory cell infiltration into the synovium, joints, and various body organs.1

A common and serious extra-articular manifestation of RA is pulmonary involvement in any lung compartment affecting 60%–80% of patients. Interstitial lung disease (ILD) is a particular type of lung involvement that is the leading cause of morbidity and mortality in ∼40% of patients with RA.2

RA interstitial lung disease (RA-ILD) pathogenesis involves a complex interaction of risk factors, including smoking history, male gender, and advanced age, and genetic factors, including HLA-DRB1 shared epitope and high titer of anticitrullinated peptide antibodies (ACPA). Data shows that dysregulated inflammatory cascades in RA-ILD can produce cytokines, chemokines, and growth factors that induce epithelial and endothelial cellular damage, angiogenesis, fibroblast proliferation, and, ultimately, lung fibrosis. ILDs have recently been on the spotlight because of severe lung damage and patients’ life quality.3–5

Interferon-gamma (IFN-γ) is an immunomodulatory cytokine with a significant role in RA pathogenesis.6 The chemokines are proteins with a chemotactic activity that target cells with chemokine receptors. They are categorized as C, CC, CXC, CX3C, and IFN-γ-inducible protein 10 (CXCL10, also called IP-10) chemokines. Monocytes, endothelial cells, neutrophils, mesenchymal cells, fibroblasts, and dendritic cells are among the cells that secrete CXCL10.7,8

CXCL10 is an inflammatory chemokine that regulates different biological responses, including inflammation, chemotaxis, and angiogenesis.6 CXCL10 is considered an ILD biomarker because of its pathogenesis involvement. B cells, type 1 T helper (Th1) cells, mast cells, dendritic cells, and fibroblasts all express CXCR3 in the synovium. CXCL10 controls Th1 cell migration from the blood to the synovium. Th1 type cell recruitment is diminished in patients with RA by inhibiting the reaction between CXCR3 and CXCL10, thereby proving the important role of CXCL10 in ILD pathogenesis.1

The current study aimed to identify the role of CXCL10 as a peripheral RA-ILD blood marker, the correlation of CXCL10 with disease activity, and to identify its association to RA-ILD.

Among the 73 enrolled patients with RA, evidence of interstitial lung abnormalities was found in 33 patients detected by HRCT. Patients suffering from RA-ILD were older with a longer disease duration than patients with RA without ILD (Table 1; p=0.001, and 0.002 respectively). Other demographic variables or environmental factors were comparable between both groups without significant difference between them regarding sex, smoking status, positive family history, and body mass index (p=0.622, 1, 0.744, and 0.379 respectively).

Table 1 enumerated all the clinical features of the participants. General characteristics were comparable between both studied groups with no statistically significant difference (p>0.05) between them. Meanwhile, patients with RA-ILD suffer from more cough, pain, expectoration, and dyspnea compared to patients with RA without ILD (p<0.001).

Patients with ILD show higher disease activity as measured by the DAS28 score (p=0.007), more restricted PFT (p<0.001), more abnormal findings in HRCT (p<0.001), and lower life quality as measured by the MHAQ (p=0.033; Table 2).

An insignificant difference was found between both studied groups regarding the type and dose of administered treatment, except for the number of patients with ILD who have received steroid treatment, which was statistically higher than patients without ILD (37 [74%] versus 7 [4%], p<0.001).

Patients with RA-ILD show significantly higher CXCL10, and C-reactive protein (CRP) than those without ILD (p<0.001, and 0.004, respectively; Table 2).

A significant positive association existed between the serum level of CXCL10 and the age of the studied patients (p=0.024), disease duration (years) (p<0.001), erythrocyte sedimentation rate (ESR) (p=0.002), CRP (p=0.001), and rheumatoid factor (RF) (p<0.001), with a negative significant correlation with MHAQ (p=0.015; Table 3).

The means of CXCL10 were compared among different DAS categories using a one-way analysis of variance test, which showed a significant difference in the mean CXCL10 between groups with mild and high disease activities (p=0.046) but was insignificantly different between groups with mild and moderate disease activities (p=0.877). The mean CXCL10 in the moderate disease activity group was lower than in the high disease activity group (p=0.065; Table 4).

ROC curve for ILD detection in patients with RA was done. CXCL10 has sensitivity and specificity for RA-ILD of 88% and 70%, respectively, at a cutoff of 128.15, with an area under the curve of 0.807 (0.72–0.894, p<0.001; Table 5 and Fig. 2).

Fig. 2.

ROC curves for ILD detection in patients with RA. CXCL0 (blue) and reference line (green). Area under the curve=0.807 (0.72–0.894), p value=0.000.

(0.12MB).

Univariate logistic regression analysis showed that age, disease duration, CXCL10, and CRP variables were significantly associated with RA-ILD. This finding was confirmed on multivariate logistic regression analysis which shows that age, and CXCL10 variables are still significantly associated with RA-ILD. CXCL10 was the most associated factor among them; were patients with CXCL10 ≥128 were about 10 times more likely to developed RA-ILD as compared to patients with CXCL10 <128 (OR=9.566, 95% CI 2.492–36.716, p<0.001; Table 6).

One of the mortality causes in patients with RA is interstitial lung disease, highlighting the need for biomarkers to identify patients at risk of developing ILD.15 This study aimed to evaluate the serum CXCL10 level as a biomarker for RA-ILD by including 73 patients with RA, of whom 33 had evidence of lung abnormalities detected by HRCT. In RA-ILD, dysregulated inflammatory cascades can produce cytokines, chemokines, and growth factors that cause epithelial and endothelial cellular damage, angiogenesis, fibroblast proliferation, and, eventually, lung fibrosis. CXCL10 is an inflammatory chemokine that controls a variety of biological responses such as inflammation, chemotaxis, and angiogenesis. CXCR3 is expressed in the synovium by B cells, type 1 T helper (Th1) cells, mast cells, dendritic cells, and fibroblasts. CXCL10 regulates the migration of Th1 cells from the blood to the synovium. Th1 cell recruitment is reduced in RA patients by inhibiting the reaction between CXCR3 and CXCL10, demonstrating the importance of CXCL10 in ILD pathogenesis. Individuals with RA-ILD were found to be older and had longer disease duration than patients with RA without ILD due to lung fibrosis by comparing the baseline demographic data among both studied groups. In line with the findings of the current study, Chen et al., Restrepo et al., and Salaffi et al.5,16,17 discovered that patients with RA-ILD were older and had a longer disease duration than those with RA without ILD.

The provided information is not exhaustive regarding gender and ILD; some researchers discovered a positive association between male gender and RA-ILD16,18,19; however, others did not find such a correlation as the current study found.20,21 This could be due to the higher prevalence of RA among female patients, which will also explain the absence of the difference between smoking status and ILD development among the patients in the current study with RA due to the lower smoking rate among Egyptian females.

Regarding the clinical presentation of the studied participants, the current study revealed that patients with RA-ILD were suffering from higher cough, pain, expectoration, dyspnea, and more restricted PFT (p<0.001) as compared to those with RA without ILD (p<0.001 for all), which was in line with Chen et al. and Spagnolo et al.2,5

The present study used DAS28-ESR and revealed that patients with RA-ILD have higher significant disease activity than those without ILD (p=0.007). Congruent with the current study, Chen et al. and Restrepo et al.5,16 observed that patients with RA-ILD have higher DAS28 scores than those without ILD. However, Kass et al. and Salaffi et al.4,17 discovered insignificant differences in the measurement of DAS28-ESR between both groups because the authors reported differences in the methotrexate usage and tumor necrosis factor inhibitors between those groups, with both drugs usage being higher in the RA without ILD subgroups than with ILD (p=0.05).

The current research regarding MHAQ revealed that patients with RA-ILD suffered from lower life quality as compared to those without ILD (p=0.033). This is mainly due to cough, pain, expectoration, and dyspnea, which were more common among patients with RA-ILD. The current study supported the studies of Natalini et al. and Fadda et al.22,23 who reported that patients with RA-ILD suffered from lower life quality compared to those without ILD as assessed by MHAQ (p=0.009).

Patients with RA-ILD were found to show significantly higher CRP than those without ILD (p=0.004). Several studies were in line with the current study in comparing different laboratory findings between both studied groups.16,19,24–27 In contrast to the current study, Fadda et al.23 and Salaffi et al.17 found no statistically significant differences in CRP levels between both studied groups. This conflict between the previously mentioned studies could be explained by the degree of RA activity during the assessment of this inflammatory serological marker.

According to the current findings, patients with RA-ILD have higher significant CXCL10 levels than those without ILD (p<0.001). In line with the findings of the current study, Chen et al.5 discovered that serum CXCL10 levels were significantly higher in patients with RA-ILD. Kameda et al.28 recently reported that serum CXCL10 levels and bronchoalveolar lavage fluid were significantly higher in patients suffering from interstitial pneumonia with autoimmune features or collagen vascular diseases associated with ILD than in patients with IPF.

The mean CXCL10 was discovered to be highly significant among patients with RA suffering from high disease activity when compared with the mean CXCL10 across different DAS28-ESR categories. Additionally, a positive significant correlation was discovered between CXCL10 and age, disease duration, CRP, ESR, and RF, as well as a negative significant correlation with MHAQ. Furthermore, the CXCL10 and disease activity correlation provides stronger proof that this serum protein can be utilized as an efficient biomarker for the pathology of specific organs in systemic inflammatory disorder. Whereas the comparatively high CXCL10 level in patients with RA-ILD with higher disease activity reinforces that extra-articular complication is caused by a conjunction of immunity dysregulation and excessive tissue remodeling. Supporting the current findings, Chen et al. and Kuan et al.5,29 stated that serum and CXCL10 concentrations may be utilized as a disease activity biomarker in patients with RA.

The role of the CXCL10 biomarker was previously examined in RA pathogenesis; however, its level has never been studied to reflect disease activity in patients with RA, except in two previously mentioned studies by Kuan et al. and Chen et al.5,29 Hence, the findings of the current study cannot be compared, and larger sampled studies are required to confirm the findings.

Multivariate logistic regression analysis which shows that age, and CXCL10 and CRP variables are still significant predictors for RA-ILD. The most significant predictor among them was CXCL10 (p=0.001). Mori et al.20 stated that logistic regression analysis indicated a strong association of ILD with age, smoking and titers of RF, also Salaffi et al.17 when studied multivariate regression analysis, they highlighted the correlation between age, ACPA titer, age at RA onset, smoking, and ILD. Additionally, Muhsin et al.30 cleared that on regression CXCL10 was significant predictor for RA. Kotrych et al.31 stated that regression analysis with multiple variables taking into account patient gender and age, disease duration, and the CXCL10 GG genotype, this genotype was found to be the independent factor associated with an increased risk of developing extra-articular manifestations.

Similar studies need to be conducted on a larger number of patients with RA-ILD, and further studies should estimate ACPA and its correlation with CXCL10 in RA-ILD.

CXCL10 is a good biomarker for RA disease activity and is a sensitive biomarker for RA-ILD. Additionally, CXCL10 is a significant predictor for development of RA-ILD.

The authors declare they have no conflict of interest.