Interleukin-32 in systemic sclerosis, a potential new biomarker for pulmonary arterial hypertension

Background Pulmonary arterial hypertension (PAH) is a severe complication of systemic sclerosis (SSc), associated with a progressive elevation in pulmonary vascular resistance and subsequent right heart failure and death. Due to unspecific symptoms, the diagnosis of PAH is often delayed. On this basis, it is of great value to improve current diagnostic methods and develop new strategies for evaluating patients with suspected PAH. Interleukin-32 (IL-32) is a proinflammatory cytokine expressed in damaged vascular cells, and the present study aimed to assess if this cytokine could be a new biomarker of PAH during SSc. Methods The IL-32 expression was evaluated in the sera and skin samples of 18 SSc-PAH patients, 21 SSc patients without PAH, 15 patients with idiopathic PAH (iPAH) and 14 healthy controls (HCs), by enzyme-linked immunosorbent assay (ELISA) and immunohistochemistry (IHC). Receiver-operating characteristic (ROC) curves were performed to evaluate the cut-off of IL-32 in identifying patients with PAH. Furthermore, in SSc patients, correlation analyses were performed between IL-32 sera levels and mean pulmonary artery pressure (mPAP) evaluated by right heart catheterization (RHC) and systolic pulmonary artery pressure (sPAP), obtained by echocardiography. Additionally, the number of skin IL-32+ cells was correlated with modified Rodnan skin score (mRSS). Results In SSc-PAH patients, IL-32 sera levels were significantly higher when compared with SSc patients without PAH and patients affected by iPAH. The analysis of ROC curve showed that IL-32 sera levels above 11.12 pg/ml were able to predict patients with PAH (sensitivity = 90%, specificity = 100%). Furthermore, the IL-32 sera levels of patients with SSc correlated with both mPAP and sPAP. In the skin derived from SSc-PAH patients, the number of IL-32+ cells was significantly increased when compared with the skin derived from SSc patients without PAH, correlating with the mRSS. Conclusion Our study suggested that sera determination of IL-32 may be a promising approach to evaluate the presence of PAH in SSc patients and together with longitudinal future studies could help to increase the understanding how these biomarkers mirror the vascular changes and the inflammatory process during SSc.


Background
Systemic sclerosis (SSc) is a systemic autoimmune disease characterised by microvascular damage and progressive fibrotic manifestations of the skin and internal organs [1][2][3][4], classified in two different subsets, displaying different clinical patterns and outcomes: limited cutaneous (lcSSc) and diffuse cutaneous (dcSSc). Cardiopulmonary involvement is the most common cause of morbidity and mortality in those patients [5]. According to the last clinical classification of pulmonary arterial hypertension (PAH), pulmonary hypertension can occur due to pure vascular involvement (WHO Group 1), secondary to SSc-interstitial lung disease (WHO Group 2) and secondary to cardiac involvement (post-capillary pulmonary hypertension, WHO Group 3) in those patients [6]. Isolated PAH (in the absence of lung fibrosis) is one of the more severe complications of SSc, associated with a very high mortality rate, [7]. It has been reported that the 1-year survival rates for SSc-PAH patients range from 50 to 81% [8], having about a threefold increase in the mortality risk when compared with iPAH [9]. Thus, an early diagnosis is essential, because the late referral to cardiologists, when the symptoms of right heart dysfunction are already present, predicts a worse survival [10,11]. However, early and accurate diagnosis of PAH is a clinical challenge, needing right heart catheterization (RHC), an invasive test which have to be performed in those patients at increased risk of PAH, following echocardiographic results and additional criteria [12][13][14]. In this context, the availability of suitable diagnostic biomarkers for an early identification of SSc-PAH patients, still in asymptomatic phase, may significantly impact patient's survival. A large number of circulating biomarkers have been investigated in the last years, but so far, only B-type natriuretic peptide (BNP) and N-terminal proBNP (NT-proBNP) are routinely used in the clinical setting, since being associated with the severity of heart involvement [15,16].
It is well known that pulmonary arteries display complex structural and functional changes during PAH and pulmonary endothelial cells' (ECs) dysfunction plays a crucial role in disease progression, leading to vascular defects observed in SSc patients [17][18][19][20][21][22]. Hyperproliferative ECs may contribute to pulmonary arterial remodelling, leading to the obliteration of vascular lumen [23]. IL-32, a proinflammatory cytokine, originally described as a transcript termed NK4, is considered a key regulator molecule of EC activities [24,25]. Although 9 different isoforms are described according to varying mRNA splicing [26], IL-32α, IL32-β and IL-32γ are the most prominent isoforms [24]. Biologically, IL-32 resembles the activity of another cytokine of the same family, the IL-33 [27], largely activated in damaged ECs of SSc [28] and over-expressed in arteries of giant cell arteritis patients [29,30], acting as a "danger signal/alarmin" molecule . The EC production of IL-32 is enhanced  by various inflammatory stimuli, such as IL-1b, IL-1a,  IL-6, and IL-8, and among them IL-1 and IL-8 are potent promoters of angiogenesis [24,31]. Of interest, IL-32 has been identified in the abnormal ECs, populating the plexiform lesions in the lungs of patients with iPAH [25], and probably involved in both activation and proliferation of these ECs. Increased IL-32 levels have been found associated with [25,32,33] activity and/or severity of Behcet's disease [34], systemic lupus erythematosus and atopic dermatitis [35,36]. Presently, the role of IL-32 in SSc is still unknown; thus, we assessed the expression of IL-32 in the sera of SSc patients with PAH (WHO Group 1) and compared these values with those from SSc patients without PH and patients with iPAH. In addition, we correlated the IL-32 sera levels with the mean pulmonary artery pressure (mPAP) assessed by RHC and the values of systolic pulmonary artery pressure (sPAP), obtained by echocardiography, to understand if IL-32 may be considered an additional non-invasive screening tool for SSc patients with signs and symptoms of PAH.
The pathogenic mechanisms of experimental PAH largely mirror what described in SSc vascular defects [17,[19][20][21][22], and on these bases, we searched for IL-32 expression in the skin samples of our SSc patients. Finally, we correlated the number of IL-32+ cells in the affected skin with the modified Rodnan skin score (mRSS), a surrogate for disease activity, severity and mortality in patients with dcSSc [37].

Patients and controls
Eighteen SSc patients with WHO group 1 PAH [6], consecutively selected during the diagnostic RHC, and 21 patients with SSc [38] without PAH, selected among those patients consecutively admitted in our outpatient clinics, who underwent sPAP evaluation, showing a normal sPAP associated with no sign or symptom referred to heart involvement, in the context of their routinely clinical assessment, were enrolled for this cross-sectional study. All patients fulfilled the 2013 classification criteria for SSc [39]. The diagnosis of PAH was confirmed according to the European guidelines [14]. Fifteen patients with iPAH and 14 HCs were also evaluated as controls. For this study, sera of the SSc-PAH patients and sera of iPAH patients, collected at the time of diagnostic RHC, were matched with the sera of SSc patients without PAH, and with the sera of HCs. Patients with infections, malignant diseases or other diseases, which could be possibly associated with increases of IL-32 sera levels, were excluded. SSc patients without PAH, needing prostanoids, calcium-channel blockers, phosphodiestererase-5 inhibitors and endothelin-1 receptor antagonists, were excluded from this study. Samples from SSc-PAH patients were collected before starting vasoactive therapies. All patients were evaluated by echocardiography, the initial non-invasive test for the screening of pulmonary hypertension [40,41]. sPAP was estimated from peak tricuspid regurgitation jet velocities according to the equation: sPAP = 4 (V)2 + RA pressure, where V is the peak velocity (in metres per second) of TRV, as previously performed [42].

RHC
RHC was performed via femoral venous access, with zero reference levelled at mid chest in the supine position. Parameters concerning pulmonary circulation were measured as follows: mean right atrium pressure, right ventricle pressure, pulmonary artery pressure and wedge pressure obtained after catheter balloon inflation at the end of expiration. Cardiac output was measured using the thermodilution technique, through a thermistortipped Swan-Ganz catheter, or the Fick technique in patients with severe tricuspid regurgitation. Pulmonary vascular resistance was calculated as (mPAP-PAWP)/ CO.

Enzyme-linked immunosorbent assay
Sera levels of IL-32 were determined by commercial human ELISA using Human IL-32 ELISA (R&D, USA), according to the manufacturer's protocol. All experiments were performed in duplicate.

Skin biopsies
Full-thickness biopsy samples, 2 × 0.5 cm, isolated from excisional biopsy, were obtained from clinically involved skin of one third of the distal forearm of patients affected by SSc. Skin with a mRSS of ≥ 1 was considered to be clinically involved [43]. Skin samples obtained from donors, matched for age and gender, undergoing a surgical treatment for trauma of arms, were used as controls. Both skin and blood samples derived from patients undergoing RHC were collected at the time of catheterization.

Immunohistochemistry
Each biopsy sample was fixed in 10% buffered formalin, dehydrated in graded alcohol series, and embedded in paraffin. Skin sections (thickness 3 μm) were deparaffinised, treated with endogenous peroxidase blocking (Dako, USA) and then with Dako Protein block (Dako, USA) to block non-specific binding. After blocking, sections were incubated with anti-IL-32 antibody (AbCam, UK). Visualisation of the primary antibodies was performed using EnVision Flex/HRP and DAB (diaminobenzidine) (both Dako, USA). No immunohistochemical staining was noted in negative control samples where the primary antibody was omitted. Sections were examined and photographed under light microscope (Olympus BX53). The number of positive cells was counted by two pathologists, blinded to tissue source and expressed as the mean of two observations for each sample. Results were reported as the median (range) of number of positive cells per microscopic field, considering the nonparametric distribution.

Ethics committee approval
The local ethics committee approved the study protocol (ASL1 Avezzano Sulmona L'Aquila, L'Aquila, Italy, protocol number #1039, Università degli Studi di ROMA "La Sapienza" protocol number #1011), which has been performed according to the Good Clinical Practice guidelines and the Declaration of Helsinki.

Statistical analysis
According to data distribution, results were presented as mean and standard deviation (SD) or median and interquartile range (IQR) as appropriate, and consequently, parametric or non-parametric t tests were used to compare these variables. Spearman's correlation was used to correlate IL-32 with sPAP. Furthermore, the receiveroperating characteristic (ROC) curves were performed to evaluate the predictivity of IL-32 sera levels in identifying patients with PAH. The best cut-off for ROC curves was calculated by the Youden's index. Due to the relatively simple study design, few missing data were managed by exclusion of these from analyses. Statistical significance was expressed by a p value < 0.05. GraphPad Prism 5.0 software and Statistics Package for Social Sciences (SPSS version 17.0, SPSS Inc) were used for statistical analyses.

Baseline characteristics of the study population
Eighteen SSc-PAH patients, 21 SSc patients without PAH, 15 patients with iPAH and 14 HCs were analysed. Demographic and clinical characteristics of our cohorts are summarised in Table 1.
The mean age of SSc-PAH patients was 69.8 ± 8.0 years; 16 patients were Caucasian female. The mean age of SSc patients without PAH was 60.5 ± 9.3 years; 18 patients were Caucasian female. The mean value of systolic pulmonary artery pressure (sPAP) estimated by Doppler echocardiography was 73.1 ± 16.6 in SSc-PAH patients, 23.3 ± 4.4 in SSc patients without PAH, and 68.7 ± 20.5 in patients with iPAH. Among the SSc-PAH patients, 13 patients showed the dcSSc form and 5 showed the lcSSc form, and among the SSc patients without PAH, 14 patients were classified as dcSSc and 7 as lcSSc. The mean value of mRSS of dcSSc-PAH patients was 16.8 ± 2.4; the mean value of mRSS of dcSSc patients without PAH was 10.6 ± 7.1.

Sensitivity and specificity of IL-32 as biomarker to identify SSc-PAH patients
The area under the ROC curve resulted to be 0.950 (95% CI 0.889-1.000, p < 0.0001) for IL-32 sera levels. By analysis of the ROC curve, we assessed that the best cutoff for IL-32 sera levels was 11.12 pg/ml, to predict patients with PAH, providing a sensitivity of 90% and a specificity of 100% (Fig. 2).
Correlations between IL-32 sera levels and mPAP evaluated by RHC in SSc patients Correlations between IL-32 sera levels and sPAP evaluated by Doppler echocardiography in SSc patients

Discussion
Our study suggests that IL-32, already identified in the lung of iPAH patients [25], may be considered a potential biomarker to identify SSc patients with WHO group 1 PAH. We enrolled 21 SSc-PAH patients, 13 showing the diffuse form and 5 the limited form of the disease. Although, it has been generally accepted that PAH is more common in lcSSc [7,44,45], recent evidence failed to confirm these previous data, showing that PAH may also affect dcSSc patients at least as frequently as the limited form [46,47]. We observed that IL-32 sera levels were higher in patients with PAH; on the contrary, those levels were quite undetectable in SSc patients without PAH and HCs, confirming the relevance of this cytokine in highlighting the presence of WHO group 1 PAH. It has been reported that IL-32 may play a key role in the vascular alterations occurring during iPAH and showing a proinflammatory effect in several inflammatory and autoimmune diseases [25,[33][34][35]. In fact, IL-32 has been described in the abnormal ECs, populating the plexiform lesions in the lungs from iPAH patients [25], and its production seems to promote the leukocyte recruitment via the production of pro-inflammatory cytokines, such as tumour necrosis factor-α (TNF-α), IL-1β, IL-6 and IL-8 [27], all contributing to EC injury [24,25,31]. Furthermore, IL-32 may play an additional role, modulating activation and proliferation of ECs during the angiogenesis [25], although conflicting results are present in literature [48]. IL-32 may modulate the VEGF expression during cancer, promoting the pro-angiogenic program [25,49], while in human bronchial epithelial cells, IL-32 appears to suppress the pro-angiogenic signals [48]. In our setting, the increase of IL-32 levels, observed only in the patients with PAH, should be referred to the specific EC damage, associated with PAH, more than a disturbance of the immune-homeostasis, as reported for other autoimmune diseases [32,33]. Although it is still unknown what isoform is associated with the organisation of plexiform lesions, during iPAH, one study suggested that IL-32γ may be the main isoform, involved in the The area under the ROC curve resulted to be 0.950 (95% CI 0.889-1.000, p < 0.0001) for IL-32 sera levels. The analysis of ROC curve shows that the best cut-off for IL-32 sera levels is 11.12 pg/ml (sensitivity = 90%; specificity = 100%) angiogenic process [25]. Nine different IL-32 isoforms have been identified [26], and among them, IL-32α, IL-32β and IL-32γ are the most representative forms [26]. In this study, mirroring what was already done in others' papers measuring the levels of IL-32 as biomarker of diseases, we used ELISA kit to be able to recognise the three main isoforms (α, β and γ), as reported by others [34,50,51]. Considering that the main goal of our work was to assess the possible use of IL-32 as a new and relevant biomarker to discriminate those SSc patients at higher risk of PAH, we did not analysed the specific role of any isoform. Surely, further studies are needed to establish the pathogenic role of each isoform in SSc patients developing PAH.
At present, no specific biomarker for PAH has been identified so far, although a wide variety of molecules have been explored [14], all of them lacking a significant specificity to identify patients to be referred to cardiologists for performing RHC and/or to follow up patients with already confirmed PAH. In our study, we identified the specific cut-off (IL-32 sera levels of 11.12 pg/ml, sensitivity = 90%, specificity = 100%), which is able to discriminate SSc patients at higher risk of WHO group 1 PAH. Although this study retrospectively assessed a relatively low number of patients, the strong sensitivity and specificity observed in our work suggest that IL-32 sera levels may emerge as a new specific biomarker, helpful for an accurate screening of SSc patients with suspected WHO group 1 PAH, being our results a proof-ofconcept to design further and powered studies to confirm and validate this result.
Of note, the IL-32 sera levels significantly correlated with the mPAP observed during RHC, in both iPAH and PAH-SSc patients. These results confirmed on one the hand the possibility that IL-32 may be a helpful biomarker in PAH patients and, on the other hand, its possible pathogenic role [25]. Furthermore, our results showed that the IL-32 sera levels significantly correlated with the sPAP evaluated with echocardiography, which is a non-invasive and feasible assessment of suspected PAH according to current guidelines [14,52]. In fact, sPAP, evaluated by echocardiography, correlates with RHC. In this context, the need for biomarkers identifying the patients' profile at the highest risk for a poor outcome has become of crucial interest in autoimmune diseases [53,54], and the correlation between IL-32 and both sPAP and mPAP could offer a new promising screening tool to be readily integrated in clinical practice. Furthermore, a serum determination could overcome the limitations of echocardiography, such as operator expertise, machine quality and patient acoustic window [55], suggesting the possible usefulness of an integrated approach.
Additionally, we assessed the expression of IL-32 in the skin of SSc patients. Our results showed that the number of IL-32+ cells was significantly increased in the skin of SSc patients, when compared with HC skin. IL-32 was expressed in keratinocytes, vascular cells and fibroblast cells present in the derma. It has been reported that IL-32 may be produced by epithelial cells, fibroblasts keratinocytes and inflammatory cells, after stimulation with pro-inflammatory cytokines, including interferon-γ and TNF-α [56], and all these cells are have been frequently found to be involved in SSc pathogenesis [57][58][59][60]. Furthermore, higher expression of IL-32 has been reported in human epidermal keratinocytes of atopic dermatitis, correlating with the severity of the diseases [61]. Interestingly, different from the relatively poor IL-32+ cells observed in the skin of patients without PAH, the SSc skin of patients with PAH showed a very rich IL-32+ cells, strongly confirming the very close association between IL-32 production and vascular complications leading to PAH.
In this study, we showed a correlation between the number of IL-32+ cells and the mRSS of dcSSc patients. It is well known that the mRSS is the main measure of outcome in dcSSc with less than 5 years from the onset of the first symptom different from Raynaud's phenomenon. Thus, any biomarker, able to correlate with mRSS, may be considered as a surrogate outcome measure for severity and mortality in patients with dcSSc [37,62,63].
Although our study shows promising results, we are well aware about some limitations to generalise these results, including the single-centre retrospective design and the relatively small sample size, thus suggesting the need of further confirmatory studies.

Conclusions
In conclusion, our study suggests that higher IL-32 sera levels may be a useful biomarker for screening patients with WHO group 1 PAH, before invasive assessments. Despite future multi-centre, longitudinal studies are needed, on a larger cohort of patients, our preliminary results support the concept that IL-32 sera levels may contribute to an earlier detection of SSc patients to be referred to cardiologist for cardiac assessment, and ultimately improving the outcome, being the early diagnosis of PAH crucial for initiating the proper treatment.