Systemic biomarkers of inflammation and haemostasis in patients with chronic necrotizing pulmonary aspergillosis

The ubiquitous mould Aspergillus fumigatus causes a variety of clinical syndromes ranging from allergic disease in immunocompetent individuals to invasive pulmonary aspergillosis in severely immunocompromised patients [1, 2]. Chronic necrotizing pulmonary aspergillosis (CNPA) is a locally destructive process of the lung due to semi invasive infection by Aspergillus species often accompanied by development of a fungus ball (aspergilloma). CNPA which is a relatively uncommon form of Aspergillus infection, typically occurs as a complication of preexisting chronic lung diseases such as emphysema, sarcoidosis, tuberculosis sequela and other fibrotic lung diseases as well as in mildly immunocompromised patients e.g. diabetes mellitus, chronic liver disease and alcoholism [3, 4]. The most common symptoms are chronic, productive cough with or without hemoptysis and pleuritic chest pain, but constitutional symptoms (i.e. weight loss, fever and lack of energy) are also seen [3‐5]. Although a few reports suggest a dysregulated cytokine response and involvement of complement activation in CNPA [6], little is known about the immunopathogenic mechanisms of this disorder. Moreover, whereas thrombosis and vascular invasion are histopathological hallmarks of invasive aspergillosis in severely immunocompromised patients [7, 8], the role of platelet and endothelial cell activation as well as the levels of prothrombotic mediators in CNPA are largely unknown [4, 6]. To further elucidate these issues, we measured selected biomarkers involved in inflammation and thrombosis in 10 patients with CNPA and in 19 sex- and age-matched healthy individuals.

CNPA is characterized by persistent inflammation caused by an unresolved infection possibly reflecting an altered antifungal immune response. Several names have been proposed for different entities of this condition [3] as it is thought to represent a spectrum of diseases depending on the degree of invasion of pulmonary parenchyma, formation and expansion of multiple cavities over time and the development of fibrosis in surrounding tissue [13]. The term CNPA is often used to describe all entities in this spectrum of illness.

CD40-CD40L interaction regulates several aspects of cell-mediated immunity, including activation of antigen-presenting cells, CD4⁺ and CD8⁺ T cells, and stimulation of IL-12/IFN-γ production [14, 15]. The CD40-CD40L axis has been implicated in anti-fungal defense mechanisms [16‐20]. Notably, the patients with CNPA had decreased expression of the gene encoding CD40L in PBMC and decreased levels of sCD40L. sCD40L is primarily derived from platelets [21‐23]. Since A. fumigatus has been shown to induce platelet activation [24, 25], low serum sCD40L in CNPA patients could reflect degranulated platelets secondary to chronic activation. On the other hand, the reduced expression of CD40L in PBMC may suggest a more general impairment in the CD40-CD40L axis in patients with CNPA, possibly contributing to a decreased ability to clear the fungus in these patients as has been demonstrated in other mycoses [16‐18, 20].

A major observation in the present study was the markedly increased IL-10 levels in the CNPA group as seen both in serum and mRNA expression in PBMC. IL-10 may have suppressive effects on the antifungal activity of mononuclear cells against A. fumigatus in vitro [6], but the role of IL-10 in pulmonary Aspergillus manifestations is still debated. It is conceivable that high IL-10 levels could contribute to the resolution of the Aspergillus induced pulmonary inflammation, but high IL-10 concentrations have also been associated with development of invasive aspergillosis in non-neutropenic immunocompromised patients and with colonization with A. fumigatus and allergic bronchopulmonary aspergillosis in patients with cystic fibrosis [26]. It is therefore tempting to hypothesize that the net effect of the high IL-10 levels in CNPA patients is to contribute to the maintenance of persistent Aspergillus infection within the pulmonary tissue in these patients, inducing a state of nonresolving inflammation.

We further show that patients with CNPA are characterized by significantly increased serum and plasma levels of several inflammatory markers including inflammatory cytokines (IL-6 and TNF-α), chemokines (IL-8 and RANTES) and the adhesion molecule ICAM-1. The reduced levels, although not significant, of IL-8 in patients with a history of CNPA more than 2 years may reflect a down-regulation of the inflammatory response with sustained infection. Although CNPA is a localized disease with only involvement of pulmonary tissue, several of the patients have constitutional symptoms which may be associated with the signs of chronic inflammation reflected in our findings.

Complement defects, and in particular low levels of MBL, have also been related to the development of CNPA [27]. In the present study we did not find any differences in MBL levels between CNPA patients and controls. However, in our opinion, the present study is too small to make any firm conclusion concerning the potential role of MBL.

Aspergillus fumigatus is in general angioinvasive, causing intravascular thrombosis and dissemination of the fungus through the blood stream [7, 8]. Thrombosis and vascular events have also been documented in patients with CNPA [28]. Our population of CNPA patients had increased levels of the pro-thrombotic mediators TF and PAI-1, without any changes in the anti-thrombotic mediator thrombomodulin. In addition, the CNPA patients showed signs of enhanced endothelial cell activation as assessed by increased vWF levels. The normal concentration of E-selectin might seem in conflict with this finding, but it has previously been shown that the expression of E-selectin on the endothelium may be differently regulated than other markers of endothelial cell activation [29].

Formulations of amphotericin B are proinflammatory stimulants of innate immune cells and the echinocandins have been shown to enhance the antifungal properties of neutrophiles and macrophages [30]. Azoles are the least active to modulate the host`s antifungal defences [30], but voriconazole has nevertheless been shown to induce production of TNF-α in the human, monocytic cell line THP-1 in vitro [31]. In the present study, three out of ten patients used voriconazole at the time of sampling. However, we did not observe any differences in the levels of measured biomarkers, although limited by low numbers, when comparing patients using voriconazole with those who did not. However, when including voriconazole as an independent variable, the difference between patients and controls did not reach statistical significance for IL-8, suggesting some influence of this medication on the immune response in CNPA.

Our findings suggest a possible dysfunction of the CD40-CD40L axis, thus compromising the immune response against Aspergillus in patients with CNPA. Elevated levels of the anti-inflammatory cytokine IL-10 may well contribute to an impaired antifungal response. Both findings may predispose to persistence of the fungal infection and unresolved inflammation. CNPA patients have signs of endothelial activation and a pro-thrombotic phenotype, possibly contributing to the well known histopathology of infections with A. fumigatus.