Discussion
This is the first study in the literature investigating on a retrospective basis the immunologic profile of a relatively large cohort of patients with CPFE and comparing it to that observed in patients with IPF without emphysema, using a complete panel of clinical, serum and histopathologic markers. Our main findings are the following: 1) An increased number of CPFE patients with positive ANA profile (17/40, 42.5%) compared to patients with IPF without emphysema (16/60, 26.6%) (p < 0.05). 2) An increased number of CPFE patients with elevated serum p-ANCA titers (7/40, 17.5%) compared to none with IPF without emphysema (p < 0.05). The latter finding was accompanied by an increased prevalence of MPA (3 out of 7 patients with positive p-ANCA profile) based on renal biopsy showing pauci-immune necrotizing glomerulonephritis coupled with diffuse alveolar hemorrhage while in the remaining 4 patients immunological (positive p-ANCA titers) clinical (mild dyspnea on exertion) and serological (microscopic hematuria with normal renal function and 24-hour urine protein concentrations) findings were suggestive of an occult autoimmune disorder compatible with systemic vasculitis namely MPA. 3) A statistically significant improved survival of patients with CPFE and positive autoimmune profile compared to those with negative one suggesting that presence of autoimmunity may be associated with favorable prognosis. 4) Massive infiltration of clusters of CD20+ cells forming lymphoid follicles immediately adjacent to areas of fibroblastic foci in lung biopsy samples from CPFE patients with positive serum immunologic profile (ANCA + and/or ANA+) compared to patients with negative profile, suggesting the presence of antibody producing B cells within the injured lung. 5) A linear positive correlation of CD20+ cells with median survival in patients with CPFE indicating this histopathologic marker as a potentially reliable disease prognosticator.
CPFE represents a distinct underecognised entity seminally defined by Cottin et al. as a syndrome characterized by frequent paraseptal emphysema with upper zone predominance imaging changes consistent with lower lobe fibrosis (UIP pattern in most cases), relatively preserved lung volumes contrasting with disproportionally impaired gas exchange, severe exercise hypoxemia and poor prognosis especially when pulmonary hypertension is present. Despite a characteristic functional and radiological profile it is still debatable whether CPFE represents a distinct syndrome or a coincidence of pulmonary fibrosis with emphysema since no common pathogenetic mechanisms have been drawn to encompass both disease entities.
Seminal observations regarding common pathogenetic pathways between emphysema and lung fibrosis identified overexpression of key molecules including tumor necrosis factor alpha (TNFa), platelet derived growth factor (PDGF) and metalloproteinases (MMPs) 2, 7, 9 as critical events leading to both alveolar and endothelial cell destruction as well as thickening of the alveolar septa depending on the lung zone. Moreover, the past few years, several studies implicated autoimmune derangement and loss of immune tolerance in the pathogenesis of both COPD and lung fibrosis, as indicated by a global numerical and functional impairment of T regulatory cells and increased titers of circulating anti-nuclear antibodies in patients with COPD and pulmonary fibrosis. Further extending the latter observations a close pathogenetic association between lung fibrosis and MPA, a type of small vessel vasculitis with renal and pulmonary involvement, has been recently demonstrated both in vivo and in vitro. More specifically, Tzelepis et al. reported an increased incidence of pulmonary fibrosis in patients with MPA while anti-MPO antibodies have been demonstrated to induce oxidative burst and fibroblast proliferation, thus, directly contributing to lung fibrosis. Further extending these evidence, our group reported a case of a patient initially diagnosed with CPFE that suddenly developed respiratory and renal failure finally attributed to MPA with lung and renal involvement.
On the basis of this finding,we conducted the first study in the literature investigating autoimmunity profile in a relatively large cohort of patients with CPFE. To this end, we used a complete panel comprising of clinical, serologic and histopathologic biomarkers and we compared it to that observed in patients with IPF without emphysema. Intriguingly both disease entities presented with a distinct immunologic profile with patients with CPFE characterized by an increased prevalence of positive ANA and p-ANCA serum titers. Additionally, elevated circulating anti-MPO concentrations were accompanied by a relevant clinical phenotype compatible with either a typical diagnosis of MPA with pulmonary involvement as assessed by positive renal biopsy and diffuse alveolar alveolar hemorrhage leading to respiratory failure or a more occult clinical phenotype encompassing asymptomatic microscopic hematuria with normal renal function and mild hypoxemia. Additional subgroup analysis of CPFE patients with positive immunologic profile revealed that there was no difference in terms of respiratory functional status as well as radiological and histopathological pattern between subjects with ANA+, ANCA + or ANA-ANCA- profile (Table 3). However, Kaplan-Meier survival analysis revealed that the presence of autoimmunity may be linked with a more favorable prognosis since patients with CPFE and positive ANA profile exhibited a statistically significant improved survival compared to those with negative profile (median survival of 51 vs. 38 months, p = 0.052, respectively). The latter observation is line with previous reports showing that signs and serological evidence of autoimmune disease may identify a group of patients with NSIP exhibiting improved survival.
Investigating beyond this observation, representative lung biopsy specimens from CPFE patients, with and without positive serum immunologic profile, were immunostained with CD20, a marker of B cell presence. Intriguingly, CD20+ B cells forming lymphoid follicles were found within the fibrotic interstitium in areas adjacent to fibroblastic foci in CPFE patients with positive serum ANA and p-ANCA profile whereas CD20+ cells were almost absent in biopsy samples from patients with CPFE and negative ANA profile as well as in control lung samples. In addition, an almost linear positive correlation between the number of CD20+ cells/mm of lung tissue and median survival in patients with CPFE was also demonstrated (Figure 1b) evidence that further supports the premise that serological and histopathologocal autoimmune markers may be used as reliable disease prognosticators. Nonetheless, it is worth noting that this is a retrospective study enrolling patients under different treatment arms and therefore a close association between positive autoimmune markers and disease survival cannot be safely drawn. Further, prospective larger and well designed studies using highly defined patients are sorely needed to prove such hypothesis.
The presence of B cells, immediately adjacent to fibroblastic foci in CPFE patients with positive immunologic profile, gives credence to the view that in certain cases development of both fibrosis and emphysema may be associated with massive B cell infiltration that migrate into the lung and are directed at infectious or apoptotic material derived from repetitive injurious stimuli including smoking and/or viruses. B cell follicular structures then arise and give birth to germinal centers enabling B cell isotype switching and antibody production. While on-going smoking exposure, chronic lung inflammation and lung injury proceed circulating neutrophils via endogenous signals such as, IL-1 and other inflammatory cytokines become primed and express intracellular antigens, such as MPO on their surfaces. This process may last for prolonged periods of time even after smoking cessation and may ultimately lead to the unintended recognition of self-antigens by somatically hypermutated antibodies including ANA and/or p-ANCA that were originally targeted to non self-antigens. Circulating MPO-ANCA antibodies then attach to the primed neutrophil membranes promoting their destruction and the release of reactive oxygen species located within the neutrophilic cytoplasmic domain. The latter results in an uncontrolled oxidative burst within the pulmonary interstitium that could promote alveolar epithelial and endothelial cell apoptosis and trigger fibroblast proliferation, as it has been clearly demonstrated with MPO-induced advanced oxidation protein products from patients with MPA. Additionally direct cytotoxicity of MPO-ANCA antibodies in endothelial and epithelial cells as well as fibrogenicity in resident lung fibroblasts has also been proposed. Moreover, the aforementioned secondary protease-antiprotease imbalance derived from neutrophil degranulation may explain focal alveolar epithelial and endothelial cell destruction leading to emphysema.
Finally, the presence of a distinct histopathologic feature (CD20+ lymphoid follicles) that closely correlates with serum immunologic profile and clearly differentiates patients with CPFE into different groups (Table 3), with distinct prognostic patterns (Figure 1a), is of major importance since it may identify a specific subgroup of CPFE patients with an asymptomatic ongoing autoimmune process that may benefit from targeted immunomodulatory or immunosuppressive therapeutic approaches. To this end, CPFE patients exhibiting elevated titers of p-ANCA without positive renal biopsy (n = 4) but accompanied by an insidious clinical phenotype resembling to systemic vasculitis, with mild hypoxemia and asymptomatic microscopic hematuria and normal renal function and 24-hour urine protein levels were switched from pirfenidone treatment to low doses of corticosteroids, azathioprine and NAC. Due to the presence of relatively mild renal and pulmonary involvement, we decided to apply a more conservative therapeutic regimen and closely monitor renal and lung function rather than implementing an aggressive treatment approach with pulses of cyclophosphamide and methylprednisolone or anti-CD20 biological agents. Given their significant side-effects the application of the aforementioned therapeutic agents could be limited for more aggressive types of disease or otherwise in the context of large randomized control trials in selective groups of CPFE patients. Preliminary follow-up data seems promising since all patients (n = 4) exhibited satisfactory treatment response as assessed by both clinical and functional stabilization (Table 4) as well as improved survival.
Despite relative enthusiasm arising from the above data our study exhibited a number of limitations that should be addressed cautiously before rigid conclusions can be drawn. Firstly and most importantly, this is a retrospective study presenting with its original caveats and therefore it is hard to delineate a temporal relationship between presence of auto-antibodies and specific clinical, radiological and histopathological phenotypes. Secondly, based on the study design it is impossible to prove a causal-effect relationship between circulating auto-antibodies and B cells infiltration with alveolar epithelial and endothelial cell destruction leading to both lung fibrosis and emphysema. Our results are more suggestive of the incidence of autoimmune markers in patients with CPFE rather than supportive of a specific pathogenetic linkage. Whether CD20+ lymphoid follicular areas within the fibrotic lung represent a causal event or a compensatory protective response against fibrogenesis constitutes the subject of ongoing in vitro and experimental studies. Thirdly and most importantly patients with CPFE exhibiting only ANA positivity could simply reflect false-positivity – and not convey any autoimmune etiology. Finally, it is important to underline the fact that there is a significant background prevalence of ANA positivity in similar groups of otherwise healthy individuals and therefore the presence or absence of autoimmunity should not be based solely on non-specific serologic markers. In line with the aforementioned observations ANA positivity may simply reflect loss of immune tolerance and not a flare of autoimmunity. A complete panel of clinical and serological markers should be administered in order to rigidly define the diagnosis of autoimmune disease.