Form, Function, and Dysfunction: Airway Diseases Are Associated with Increased Risk for Rheumatoid Arthritis

Whether it be the sweeping eagle in his flight, or the open apple-blossom, the toiling work-horse, the blithe swan, the branching oak, the winding stream at its base, the drifting clouds, over all the coursing sun, form ever follows function, and this is the law.

Louis Sullivan

Rheumatoid arthritis (RA) risk is strongly influenced by environmental exposures and the organ with the largest interface with our environment is the lung. To effectively exchange oxygen for carbon dioxide, the lung must support a delicately thin mucosal border (2 μm) spread over 140 m² (1). While ideal for gas exchange, the immensity of the lung’s surface area makes it challenging to mitigate threats from inhaled pathogens and toxins while simultaneously promoting immunologic tolerance to harmless epitopes. A failure to maintain homeostatic host-environment interactions along this mucosal border is implicated in the pathogenesis of many lung diseases and growing evidence suggests it is also implicated in the pathogenesis of RA.

Pulmonary diseases occur commonly in RA. Clinically significant interstitial lung disease occurs in 5–10% of patients with RA and subclinical interstitial lung disease can be detected in 30–50% of patients with RA (2). Other pulmonary manifestations of RA include pleural effusions, bronchiectasis, pulmonary hypertension, pulmonary nodules, and both follicular and constrictive bronchiolitis. Recently, RA was also identified as a risk factor for chronic obstructive pulmonary disease (COPD) (3). The biologic mechanisms underlying RA-associated lung disease are poorly understood but likely involve systemic autoimmunity and inflammation. Overexpression of TNF- α in mice causes both joint inflammation and lung disease, while in a SKG murine model of RA, the development of pulmonary disease requires a “second-hit” lung injury (e.g. cigarette smoke or bleomycin) (4, 5). Additionally, autoimmunity is now implicated as a cause of lung destruction in COPD (6).

However, lung disease may not simply be a consequence of RA but also a cause of RA pathogenesis. Epidemiologic studies have identified cigarette smoking as a major risk factor for RA and the relative risk for RA conveyed by certain HLA-DRB1 polymorphisms is synergistic with a history of smoking (7). Smokers have increased expression of peptidylarginine deiminase (PAD) enzymes which citrullinate proteins through the conversion of arginine to citrulline (8). In turn, these citrullinated proteins can trigger formation of anti-citrullinated protein autoantibodies (ACPA) that are highly specific for RA (9). Citrullinated proteins and ACPA can be found at higher concentrations in bronchoalveolar fluid than blood from individuals with early RA (8, 10), and RA is also associated with the presence inducible bronchus-associated lymphoid tissue (iBALT) which contain ACPA producing B-cells (11). Airway inflammation can also be identified in early ACPA positive RA as well as individuals who have serum elevations of ACPA and/or rheumatoid factor but do not have inflammatory arthritis (10, 12, 13). In addition, ACPA and rheumatoid factor can be found in sputum from individuals at-risk for future RA (14, 15). The risk of RA conveyed by cigarette smoke is likely a consequence of pulmonary injury and inflammation rather than a direct effect of cigarette smoke. Inflammation is a common cause for citrullination in many tissues, and many volatile environmental exposures increase RA risk including mineral and textile dust, not just cigarette smoke (9, 16). In some studies, ACPA titers were higher in smokers with COPD than those without COPD (17). Additionally, individuals with COPD from alpha-1 antitrypsin deficiency have an increased risk for elevated ACPA titers, but amongst these individuals, there is no association between ACPA titers and smoking (18).

If indeed pulmonary injury and inflammation are key steps in RA pathogenesis, then one might hypothesize that chronic inflammatory airway diseases like COPD and asthma should also increase RA risk. Yet there have been no studies demonstrating COPD is a RA risk factor and the evidence for asthma as a RA risk factor has been limited until now. However, in this issue of Arthritis and Rheumatology, Ford and colleagues provide compelling evidence that both asthma and COPD increase the risk of RA, independent of smoking history. They used Cox proportional hazard models for incident RA in a prospective cohort study of 205,153 women in the Nurses’ Health Study. Exposures were asthma or COPD as determined by self-report of physician diagnosis of disease in combination with validated questionnaires. The outcome variable was RA as determined by medical record review by two rheumatologists. Seropositive RA was determined by the presence of elevated rheumatoid factor or ACPA. Importantly, their multivariate model adjusted for potential confounders including smoking status (current/former/never), duration and intensity of smoking (pack-years), and passive smoke exposure. Amongst women with COPD, the multivariable-adjusted hazard ratio (HR) for developing RA was 1.89 (95%CI 1.31,2.75). Interestingly, there was an increased risk for seropositive RA (HR=2.07, 95%CI 1.31,3.25) but not seronegative RA (HR=1.59, 95%CI 0.83,3.05). Because COPD commonly occurs amongst older individuals, the authors evaluated a subgroup of ever-smokers aged >55 years and identified a stronger association between COPD and seropositive RA (HR=2.85, 95%CI 1.63,4.99) in this subgroup. Amongst women with a diagnosis of asthma, the multivariable-adjusted HR for developing RA was 1.53 (95%CI 1.24,1.88). Asthma was associated with both seropositive and seronegative RA, but interestingly, the presence of asthma in women who never smoked was associated with seronegative RA (HR=1.90, 95%CI 1.22, 2.96) but not with seropositive RA (HR=1.32, 95%CI 0.88,1.96).

The study by Ford and colleagues was well-designed and their use of a large and thoroughly characterized cohort of subjects provided them the opportunity to demonstrate that asthma and COPD are risk factors for RA. Naturally, this study also raises important questions. Only women were included in the study and therefore findings may not be generalizable to men. There was also no assessment of medication use for asthma or COPD (e.g. inhaled corticosteroids), and it would be important to understand the role medications have in decreasing or increasing subsequent RA risk. Additionally, while validated tools were used to identify the presence of asthma or COPD, both these diseases are remarkably complex and heterogenous. COPD involves multiple pathologic manifestations, including emphysema, small airway disease, mucous metaplasia, and vascular remodeling. These pathologic features occur to variable degrees amongst affected individuals and therefore current pre-clinical and clinical assessments of COPD include measurements of airflow obstruction, percent radiographic emphysema, respiratory diffusion capacity, patient reported symptoms, and emerging biomarkers (19). Similarly, asthma is a complex disease involving various subtypes (i.e. endotypes). While some individuals with asthma have classic features of atopy and TH₂ inflammation, other subtypes of asthma are more closely associated with the presence of obesity, neutrophilic inflammation, and/or exposure to volatile pathogens and toxins (e.g. cigarette smoke) (20). There is also an increasing appreciation of a subgroup of patients with asthma-COPD overlap syndrome. While poorly defined, patients with this condition have features of both diseases. Future studies will require thorough phenotyping of COPD and asthma to more completely understand their relationships to RA.

The need for more extensive phenotyping may be particularly important for asthma because of the authors’ intriguing findings that asthma and COPD have different associations with RA seropositivity. The authors identified COPD as a risk factor for seropositive RA but not seronegative RA, a finding rooted in mechanistic evidence as described above. In contrast, the authors identified asthma as a risk factor for seronegative RA but not seropositive RA amongst never smokers, and one can only speculate about the cause for increased seronegative RA in this population. What are the inflammatory profiles and mechanisms underlying this relationship? Is the relationship strongest amongst those with atopic asthma? Are specific exposures or genetic associations involved? Understanding how asthma subtypes relate to seronegative RA may lead to the identification of yet to be discovered mechanisms and treatments for RA.

Despite the limitations of this study, these findings have the potential to immediately impact care of patients with asthma and COPD. By informing us of the heightened risk for RA amongst individuals with asthma and COPD, we can decrease time to diagnosis and initiate treatment earlier. In nature, form follows function. Ford and colleagues remind us that the lung’s form also exposes it to environmental factors that in turn contribute to diseases both inside and outside the lung.