Advances in Identifying Urine/Serum Biomarkers in Alpha-1 Antitrypsin Deficiency for More Personalized Future Treatment Strategies

ABSTRACT

Alpha1-antitrypsin deficiency (AATD) is a genetic disorder characterized by reduced serum levels of alpha1-antitrypsin (AAT) and increased risk for developing both early-onset lung emphysema and chronic liver disease. Laboratory diagnosis of AATD is not just a matter of degree, although the AAT serum level is the most important determinant for risk of lung damage. While being a single-gene disease, the clinical phenotype of AATD is heterogeneous. The current standard of care for patients affected by AATD-associated pulmonary emphysema is replacement therapy with weekly i.v. infusions of pooled human purified plasma AAT. Although no treatment for liver disease caused by deposition of abnormal AAT in hepatocytes is available, innovative treatments for this condition are on the horizon.

This article aims to provide a critical review of the methodological steps that have marked progress in the detection of indicators described in the literature as being “clinically significant” biomarkers of the disease. The development and routine use of specific biomarkers would help both in identifying which patients and when they are eligible for treatment as well as providing additional parameters for monitoring the disease.

KEYWORDS:

• Alpha-1 antitrypsin
• desmosine
• fibrinogen degradation products
• MMP9
• orphan disease

Introduction

Alpha1-antitrypsin deficiency (AATD; OMIM +107400) is one of the most common inherited conditions mainly affecting the Caucasian population, 7.7% of the Northern European population and 25% of individuals in the Iberian Peninsula carry either the S- or Z-deficient alleles (1). Nevertheless, AAT deficiency affects individuals in all racial subgroups worldwide (2). This disorder is the underlying cause of approximately 1–2% of Chronic Obstructive Pulmonary Disease (COPD) cases (3,4). AATD predisposes individuals, especially smokers, to developing pulmonary emphysema in the fourth-fifth decade of adult life and predisposes children to childhood cirrhosis in about 10% of cases, with an initial presentation of prolonged neonatal jaundice. The detrimental effect of smoke exposure on the clinical phenotype of AATD has been recently reviewed (5). Although severe early-onset panacinar emphysema with a basilar predominance in adults is the classic pulmonary presentation of AATD, emphysema may also be diffusely distributed or predominant in the upper lobes (6). Bronchiectasis, with or without concomitant emphysema, is present, but its frequency is a matter of controversy (7). Dyspnea is generally the prominent symptom, although chronic cough or wheezing may also occur (6,8).

Pathophysiology of AATD

The pathophysiology of the disorder lies in variants of the SERPINA1 gene (also known as PI) encoding alpha1-antitrypsin (AAT), which results in severely reduced plasma and lung levels of AAT (0.15–0.25 g/L vs reference levels of 0.90–2.00 g/L) (9). It is well known that SERPINA1 variants dictate the plasma AAT concentration according to a co-dominant model, the most common SERPINA1 variants associated with AATD are the Z (rs28929474) and S variants (rs17580). The risk for COPD is well defined for subjects displaying absent (PI^*NullNull) or extremely low (PI^*ZZ; ∼15% of normal PI^*MM) AAT levels. Risk decreases progressively according to the following hierarchy: PI^*SZ (AAT level ∼40% of normal) > PI^*MZ (AAT level ∼60% of normal) > PI^*MS (AAT level ∼80% of normal) > PI^*MM. The clinical relevance of the problem is highlighted by evidence that approximately 45% of emphysema subjects referred for lung transplant in the United States are affected by AATD (10).

Diagnosis of AATD: Biochemical, genetic, and clinical tests

Laboratory diagnosis of AATD is currently performed in specialized centers since it requires a combination of different biochemical methods including nephelometric measurement of the AAT concentration; phenotyping by isoelectric focusing (IEF); genotyping (which mainly consists in the detection of S and Z alleles); and sequencing. Moreover, since systemic inflammatory status parallels increased levels of AAT and this increase might mask the presence of AATD variants, the simultaneous determination of C-reactive protein, and AAT in blood is useful for the correct diagnosis of heterozygotes carrying intermediate AATD genotypes (11). Nevertheless, the availability of matrices such as dried blood spots (DBS), which have facilitated the implementation of laboratory analysis for AATD and have also challenged laboratories to develop more reliable and reproducible techniques to evaluate dried blood (12).

The ranges of serum AAT in the general population, determined according to the main genotype classes, have been summarized in a recent report (13). Based on the AAT concentration, the predictive accuracy for classifying genotypes were also assessed including those not believed to represent a risk for developing emphysema (PI^*MM and PI^*MS) as well as those associated with intermediate AATD and arguably having a slightly increased risk for developing emphysema (PI^*SS and PI^*MZ). Accordingly, the optimal threshold provided a cutoff for AAT level at 1.00 g/L. For discriminating between PI^*MM and any other genotype carrying at least one S or Z allele, an optimal cut-off of 1.10 g/L was determined (13). Diagnosis of AATD requires biochemical and genetic tests, but clinical data may also support in addressing the laboratory analysis.

Physicians who are involved in the diagnosis of AAT deficiency should ask individuals about possible risk factors. These latter include: 1) smoking and exposure to dust, fumes, and other toxic substances; 2) medical history, i.e. the development of emphysema without any obvious cause at an unusually early age (45 years or younger); and 3) the family's medical history, i.e. if the subject has consanguineous relatives who have AATD, he is more likely to have the condition.

Variance with common COPD, in which airflow obstruction is believed to be largely secondary to airway damage and remodeling, in AAT deficiency emphysema is considered to play a major role (14). At physical examination, possible signs of the disease may be barrel-shaped chest and decreased breath sounds with or without wheezing and crackles. Chest x-ray may reveal hyperinflation of lungs, depression of diaphragm with its insertion of ribs evident, peripheral attenuation of pulmonary vessels, and heart shadow that is small relative to lungs. Lung function tests, including measurements reflecting airflow obstruction, lung hyperinflation, bronchial collapsibility, alveolar-to-capillary diffusion capacity, and bronchodilator response, are recommended to achieve a complete definition of the patient's status. While the former provides measurements of lung function associates with emphysema, high-resolution CT (HRCT) scanning is considered the most sensitive technique for directly detecting and quantifying pulmonary emphysema in vivo, but it may not be easy to obtain and it exposes the patient to radiation (15). When these features are present, doctors should suspect AAT deficiency and recommend tests to confirm or exclude its diagnosis. Moreover, among respiratory conditions, recommendation for diagnostic testing was made by the American Thoracic Society/European Respiratory Society also in the following settings: subjects with COPD, or asthma with airflow obstruction that is incompletely reversible after aggressive treatment with bronchodilators, and those with bronchiectasis without evident etiology (16).

Treatment strategies

Treatment of AATD is based on the individual's symptoms, the major goal of AATD management is prevention or slowdown of lung disease progression. Smoking cessation, if a person with AATD does smoke, is essential.

As in common COPD, treatments include the use of bronchodilators and prompt treatment with antibiotics for treating infectious exacerbations and other bacterial respiratory tract infections (17,18). Replacement (augmentation) therapy with weekly i.v. infusions of alpha1-antitrypsin purified from pooled human donor blood is available and represents the current standard of care for patients affected by AATD-associated pulmonary emphysema. Intravenous augmentation therapy in those with AATD is recommended for individuals with an FEV₁ less than or equal to 65% predicted and can be considered for those with an FEV₁ greater than 65% and rapid lung function decline (19). Lung transplantation may be an option for those who develop end-stage lung disease, according to guidelines (14).

Replacement therapy has been approved since 1989 in a number of countries, and currently several thousand AATD patients are on treatment. Not surprisingly, scientific literature in this field is rich with reports that show the efficiency of this therapy in reducing the rate of decline of lung function in AATD-affected individuals. Despite this evidence, due to the production of these data in non-controlled studies, their level of reliability is low. Measurement of emphysema progression, performed with computer-aided lung tomography in a multicenter randomized study on replacement therapy, showed that the trend offered by this treatment, although not significant, is indeed beneficial for patients (20). The recently published RAPID trial showed that replacement therapy was effective in reducing the annual loss of lung parenchyma, being demonstrated by a statistically significant reduction of the loss of lung density measured at total lung capacity of 34% (p = 0.03) (21). However, the high cost of this therapy renders its utilization increasingly questionable even when therapy is available. Indeed some countries have withdrawn the use of augmentation therapy while others refuse to reimburse the cost of this treatment. What is unknown is how effective this therapy may be once disease has developed, or who will benefit the most. A recent article pointed out the need to identify patients particularly at risk and hence most likely to benefit from treatment by using outcomes specific to the disease process, and suggested to monitor their efficacy where these are changing the most (22).

Inhalation of AAT, which can be therefore delivered directly to the respiratory epithelia, is considered a promising alternative to the i.v. route. The efficacy and safety of this approach have been recently reviewed (23).

Outcomes for AAT pharmacological trials

Provided that it is possibly a reliable diagnosis and an effective therapy for AATD, the availability of indicators capable of producing information on the clinical course would indeed be helpful to assess or predict changes during interventional trials.

Generally, among outcomes for pharmacological trials lung function and imaging are included (24). In common COPD, FEV₁ remains the primary end-point that regulatory authorities regard as an acceptable measure of efficacy for pharmacological trials. Unfortunately, because AATD is a rare disease it is evident that no clinical trial can ever achieve a sufficient power for obtaining statistically significant results using lung function as the primary outcome. Moreover, in AATD the primary pathophysiological alteration is emphysema and not the airway obstruction. The presence of emphysema can be evaluated with more specific tests, such as the diffusing capacity of the lung for CO (Dlco) and HRCT. However, it is not easy to determine whether a Dlco-measured change reflects a true change in pulmonary status or is only a result of test variability (25). HRCT imaging provides a means of accurately characterizing lung parenchymal changes and the nature of the image data facilitates quantitative assessment. The 15th percentile lung density (PD15) is now considered the most sensitive index of emphysema progression and of treatment modification in patients with AATD (26). Although densitometric evaluation allows the measurement of the progression of emphysema, a number of concerns have been raised regarding its use in COPD clinical trials. These mainly relate to the unresolved issues of the repeated exposure of patients to ionizing radiation and the high costs involved in its frequent use. Another issue is that this methodology has not been fully validated yet (27).

In spite of the increasing number of biomarkers (e.g. molecules, materials, cells, tissue) reflecting the disease process proposed so far in common COPD, useful biomarkers for AATD are still lacking. The next sections of this report will focus on the indicators most commonly designed as “clinically significant” biochemical markers of the disease, in order to provide a support in their use as an alternative or an integration to lung function or imaging in driving the therapy regimens. These biomarkers have been related to disease pathophysiology and the inflammatory and destructive process in the lung. In particular, the literature dealing with the identification of these molecules and the technological advances for their detection/quantitation from early years until present will be reviewed.

A few comments will also be dedicated to other indicators which can be found in the literature.

Biochemical markers

A brief “history” of desmosine detection

It has been speculated that the lung destruction observed in AATD-related COPD subjects may correlate with an imbalance in protease and AAT levels. Degradation of mature elastin results in the production of a variety of crosslinked elastin peptides containing desmosines (desmosine, DES and isodesmosine, IDES, hereafter referred as DESs), two rare tetrafunctional amino acid isoforms which, in humans, are unique to mature elastin in humans (28,29).

Under the assumption that fragments containing DESs deriving from elastin breakdown are quantitatively excreted in urine, an interesting debate on whether these amino acids should actually be considered surrogate end-points of extracellular matrix degradation started a few years ago (16). While initial studies focused on limited, specific aspects of DESs, an appreciation for the important role that these cross-linkages play in elastin fibers progressively increased the interest on their quantification. Thus, measurements in different groups of subjects with and without destructive lung disorders were initially performed to check for variations in pulmonary elastin fiber turnover. On the basis of observed significant differences in the amount of excreted DESs, healthy subjects and patients could easily be discriminated. This led to speculating that these amino acids were adequate indicators of elastin degradation (30,31). However, due to the low sensitivity and precision of methods available, solid evidence that DESs were clinically significant “biomarkers” of COPD was lacking. Advances in technological platforms, aimed at circumventing technical limitations and at enhancing instrumental sensitivity, promoted the development of new techniques for DESs quantification. The rationale for these improvements was that detection/quantification of analytes would be expected to become more accurate and reliable as the analytical sensitivity and specificity of the method improve, despite the complexity of biological matrices. Conventional methods, spanning from radioimmunoassay (RIA) (32), to high-performance liquid chromatography (HPLC) (33,34), capillary electrophoresis in the Capillary Zone Electrophoresis (CZE) (30), and Micellar Electrokinetic Chromatography (MEKC) (35) modes, allowed the accumulation of evidence that patients with destructive lung diseases excrete more urinary DESs than healthy subjects. However, it was the advent of MEKC coupled to laser-induced fluorescence detection (MEKC-LIF) that allowed the detection of (for the first time with high sensitivity and precision) these crosslinks in a variety of fluids other than urine from large cohorts of subjects suffering from different pathologies (36). A protocol was developed in our own laboratory in 2004, at the Department of Biochemistry, University of Pavia. Even though after derivatization with fluorescein isothiocyanate (FITC) the isomers co-migrated (thus the sum of the two was quantified), it worked well. The progressive replacement of the above-mentioned methods with this increasingly sophisticated technique and the use of well-standardized protocols resulted in much higher quality data on DESs concentration in urine, plasma, and/or induced sputum (37). DES levels measured in urine and plasma were in total agreement with the patient's clinical status and could easily be associated with type Z AAT-deficient patients with clinically significant emphysema or with individuals affected by Pseudoxanthoma elasticum (PXE) (38) or severe COPD (39). In the latter case, for the first time Capillary Electrophoresis coupled to laser-induced fluorescence detection (CE-LIF) allowed the analysis of sputum specimens producing data that correlated well with those from urine and plasma of the same patients and with their clinical parameters (forced respiratory volume, FEV₁). The application of CE-LIF to biological fluids from individuals with diverse connective tissue disorders facilitated the largest DES screening ever performed. Adoption of this approach made it quite clear that the greater the sensitivity and precision of the method employed, the more reliable was the correlation of disease severity with the levels of excreted crosslinks. Further technological advances observed with the advent of mass spectrometry (MS), confirmed this assumption. TMS-based monitoring procedures only select ions and their fragments, thus they circumvent the limitations associated with MEKC-LIF, and achieve greater specificity and sensitivity. Not surprisingly, this enhanced instrumental sensitivity has dramatically increased the popularity of this approach, allowing liquid chromatography–tandem mass spectrometry (LC–MS/MS) to emerge as the most widely utilized protocol in the area of DESs analysis. Among the variety of LC–MS/MS applications in DESs detection, the most representative examples are as follows. Ma et al. (40,41) performed accurate measurements (in urine, plasma, and sputum) of DESs as markers of elastin degradation in both AATD patients and non-AATD-related COPD subjects. In two different studies, Boutin et al. (42,43) measured urinary DESs in COPD patients and urinary hydroxylysylpyridinoline and lysylpiridinoline as biomarkers for Chronic Graft-versus-Host Disease. The ultrahigh performance liquid chromatographic (UPLC) technique, combined with ion mobility mass spectrometry (IM-MS), was applied by Davenport et al. (44) for the quantitation of free DES in urine and by Shiraishi et al. (45) to determine the total amount of urinary DESs in COPD-affected patients. Total urinary and plasma DESs correlated with lung function of subjects analyzed in another study by Lindberg et al. (46). To demonstrate that the amount of DESs is increased in urine or blood of patients with COPD compared to healthy non-smokers, Albarbarawi et al. (47,48) and Huang et al. (49) recently proposed the use of a validated isotope-dilution LC–MS/MS method.

Method development for DESs determination over the last fifteen years is summarized in Figure 1. It shows the DESs peak(s) that can be observed in a model electropherogram obtained by applying CE-UV (bottom); MEKC-LIF (middle); and LC–MS (top) to the analysis of a urine sample from a generic AATD patient.

Figure 1. Determination of desmosines in urine samples from AATD patients, obtained by applying CE-UV; CE-LIF; and LC–MS (bottom to top, respectively). Electropherograms/chromatograms reported are representative of those generated with the mentioned approaches, in the years indicated in the figure. Bottom: CE-UV profile. Peaks 1 and 2 represent IDES and DES, respectively. Experimental conditions: approximately 10 nL was injected into an uncoated fused-silica capillary (57 cm × 50 µm i.d.). Electrolyte background: 35 mM sodium tetraborate pH 9.3 containing 65 mM SDS. Applied voltage: 10 kV. Temperature: 25°C. Middle: CE-LIF electropherogram obtained from a diluted urine sample (approximately 10 nL) derivatized with FITC. Peak 1 represents endogenous desmosines (IDES plus DES) and peak IS corresponds to the internal standard (FITC-Asn). The peak indicated by an arrow represents the excess FITC. Inset: expansion of the square region containing peak 1. Experimental conditions: uncoated capillary (57 cm × 50 µm i.d). Electrolyte background: 20 mM sodium tetraborate pH 9.0 containing 60 mM SDS and 15% v/v methanol. Applied voltage: 30 kV. Temperature: 25°C. The laser module consisted of a 3 mW and a 488 nm air-cooled argon-ion laser with an emission band pass filter of 520 ± 2 nm. Top: LC-MS profile (panel A) obtained from a UPLC column (2.1 × 100 mm, 1.8 μm pore size) using a 7-min gradient from 99.5% of solvent A (5 mM ammonium formate containing 0.1% formic acid) to 90% solvent B (methanol containing 0.1% formic acid and 5 mM ammonium formate) at a flow rate of 0.5 mL/min. The sample volume injected was 10 μL. Panel B: multiple reaction monitoring (MRM) transition 526 > 397 of the desmosine peak (indicated by an arrow in panel A) was selected for quantitative and qualitative analyses, respectively.

Clinical validity of desmosine as biomarkers for AATD-related emphysema

The measurement of DESs will likely become more common practice; however, the question arises as to whether these crosslinks are ready to be introduced into the biomarker “hall of fame.” Given the number of critical questions that still need to be addressed, it seems plausible that the answer is: not yet, since the clinical validity and utility of this assay remain controversial. First of all, an important criticism to the concept that an increase in elastin turnover is primarily related to patients with pulmonary emphysema is that elastin is not unique to lung interstitium. This implies that increased DESs levels might be associated, for example, with accelerated elastin turn-over in the skin or major vessels. However, the well-documented presence of DESs in sputum (41) and their correlation with the emphysema phenotype, strongly suggest that lung would be a major source of elastin crosslinks in body fluids. Other important aspects are the lack of consensus about which parameters need to be analyzed (DES and IDES separately; DES + IDES; free or total) as well as the lack of fully validated methods for their quantification. Analytical validation of the method is a mandatory step to guarantee the reliability of results since data of inadequate quality may lead to over- or underestimation of new drug effects, inaccurate patient monitoring, or incorrect conclusions in clinical studies. As indicated above, a fast and reliable LC–MS method that uses D(5)-DES as an internal standard was developed and validated by Albarbarawi et al. (48) to measure total urinary DESs. More recently, Ongay et al. (50) were the first to develop and fully validate two other methods: one for the quantification of free DES and IDES, and the second for the quantification of total DESs. These two main approaches used for the absolute quantification of endogenous compounds in biological samples (i.e. authentic analyte in surrogate matrix and surrogate analyte in authentic matrix) were shown to produce the same results. When these approaches were applied to the analysis of urine from stable COPD patients and current- and never-smokers, statistically relevant differences in DESs levels were observed among groups. These results support the use of urinary DESs as promising biomarkers for COPD. However, the analysis of longitudinal samples from the same patients/controls was not performed and would have been very useful as confirmation of these findings. The longitudinal behavior and the relationship with progression and severity of the disease is, in fact, another important aspect of DESs validation. Large longitudinal studies are necessary to confirm their predictive power for patients’ clinical outcome. Indeed, these studies would add to the understanding of whether, besides their association with COPD in cross-sectional studies, DESs could be related to FEV₁ decline and to the worsening of diffusing capacity in longitudinal cases, and perhaps to changes in lung CT scan densitometry. This would certainly confirm their capacity for monitoring progression of disease severity and response in effective interventional trials. Unfortunately, convincing evidence that urinary and blood DESs are a useful biomarker of COPD status has only been provided by cross-sectional studies. A good example is the study by Lindberg et al. (46) (previously mentioned), who showed that several lung function parameters (FEV₁, FVC, RV, RV/TLC and DL, CO) had a significant association with urinary DESs after adjustment for age, sex, height, body mass index, and smoking status. By contrast, plasma DESs were significantly associated with FEV₁, DL, and CO only. These correlations were much more pronounced in COPD subjects than in individuals without COPD. Of great interest was the finding that DESs can be independently influenced by a number of factors after adequately correcting for risk factors to avoid confounding results. Huang et al. (49) also investigated the relationships of DESs levels with smoking status, disease entity, severity, exacerbations, and lung function. Taking advantage of the analytical validity of assays and the study size, they demonstrated that the elevation of DESs levels was only significant in patients with COPD, and not in those with asthma or in healthy smokers. In particular, they observed the elevation of urinary DES levels only during an exacerbation of COPD, and not in stable COPD. Blood DES levels were also found to be elevated in a subgroup of approximately 40% of patients with stable COPD.

It should be underlined that, although unsuccessful, attempts to use urinary DESs in the determination of therapeutic response (or efficacy) following drug treatments in patients with COPD (51) and AATD (52,15) have been previously performed. Interest in using these crosslinks as a surrogate endpoint in clinical trials for COPD has been revived mainly due to the technical feasibility of measuring plasma and sputum DES/IDS levels, which requires sensitive and specific assays.

Thus, what has emerged from the scientific literature over the course of these years is that: 1) the accuracy of DESs assessment parallels our understanding of elastin degradation in pulmonary disorders; and 2) the methodologies described have the potential to be used in clinical validation studies to establish a link with clinical outcomes. These results indicate that we are on the right path to utilizing these crosslinks as valid tools or “biomarkers” in the differential diagnosis and clinical management of these diseases.

It remains to be seen whether DES measurement could have an evidence-based role in stratifying patients for specific treatment or prognosis.

Fibrinogen degradation products

Fibrinogen is the most evolved blood biomarker in COPD, it has a well-established relationship with fatal outcome and its capacity for predicting exacerbations is promising (54). In particular, fibrinogen degradation products caused by neutrophil elastase could be considered valid biomarkers of AATD. In this context, with the aim of measuring human neutrophil elastase (HNE) activity, an assay for detecting of fibrinopeptide Aα(1–21), which is a cleavage product of fibrinogen, and its degradation products was developed several years ago (55).

Plasma levels of large fibrin(ogen) fragments formed by neutrophil elastase-mediated degradation (PMN-FDP) have been measured by Stolk et al. (56). Specific monoclonal antibodies were generated and used for the production of commercially available assays (57). The amount of these fragments is significantly greater in the plasma of subjects with AAT deficiency than in healthy controls, indicating an imbalance in the protease–antiprotease ratio in vivo at sites of inflammation where fibrin(ogen) is deposited (58). Although not disease-specific, fibrinogen is present at sites of inflammation and, as such, is relevant for patients with AAT deficiency who have increased inflammation in their lungs, even in the absence of a smoking habit.

The usefulness of applying an elastase-specific fibrinogen biomarker was shown by Stolk et al. (56) in a clinical study, in which AAT was administered intravenously. The amount of PMN-FDP fragments decreased in patients given the currently used dose, but did not reach levels seen in healthy individuals. In contrast, doubling the dose of AAT resulted in normal levels of fragments, which were maintained for ten days. These results suggest that fibrinogen fragments may serve as a marker for inflammation-induced proteolysis in the lung in vivo and that their formation can be inhibited with AAT doses higher than those currently recommended for augmentation therapy. To justify the cost of this treatment, assessment of the efficacy on the basis of biochemical markers of neutrophil-mediated alveolar destruction in these patients is feasible with these kinds of assays (59).

More recently, Carter and colleagues developed an assay for Aα(Val360), as a specific marker of neutrophil elastase activity (60). A cross-sectional study demonstrated that, while Aα(Val360) correlates with physiological, radiological, and symptomatic markers of disease severity in PiZZ AATD subjects, as assessed by transfer factor of the lung for carbon monoxide, this is only valid for disease progression in subjects with spirometry values in the normal range (61). Therefore, Aα(Val360) may be considered a tool for identifying subjects with progressive disease at an early stage and, in particular, as a measurement of the emphysematous component.

MMP-9

Matrix metalloprotease-9 (MMP-9) could be a biomarker that predicts disease progression without necessarily playing an etiologic role, since it is likely to be complex and time-dependent. It is not yet clear if MMP-9, by potentiating Interleukin-8, promotes an inflammatory pathway for lung damage and COPD exacerbations, or if its increase correlates with other factors that contribute to emphysema progression. In the past, the MMP9 polymorphism was demonstrated to act as a genetic factor for the development of smoking-induced pulmonary emphysema (62). In the 126 subjects with AATD-associated emphysema of the REPAIR trial (63), higher plasma levels of MMP-9 were associated with poorer pulmonary status at baseline and MMP-9 was longitudinally associated with certain decrements in pulmonary status and worse health outcomes. More recently, MMP-2 and MMP-9 have been used as biomarkers in BALF of mice treated with augmentation therapy and high-density lipoproteins (64). Moreover, investigations aimed at observing the effect of weekly intravenous AAT therapy (Prolastin®) on plasma MMP-9 and myeloperoxidase (MPO) levels showed that this therapy could significantly lower the levels of these two enzymes (65). Although only a few longitudinal studies on MMP-9 have been performed so far, this matrix metallopeptidase may be particularly important among the various biomarkers potentially associated with lung disease.

Biomarkers for liver damage and fibrosis

Serum gamma glutamyl transferase (GGT), a recognized marker of liver dysfunction, is commonly used as an indicator of biliary and liver disease. Nevertheless, the serum GGT level may also be an early marker of oxidative stress (66). Based on measurements and analysis of serum GGT in 334 PiZZ subjects from the UK AATD registry, Holme et al. (67) concluded that serum GGT, while reflecting the presence of liver disease, is independently associated with airflow obstruction and mortality. Whether serum GGT levels in AATD patients reflect oxidant stress associated with hepatic inflammation or a direct cytopathological effect of Z polymers is speculative. The role of this marker needs further investigation, ideally in prospective studies. Thus, even though GGT was prognostic for liver disease in children (68), in the Swedish alpha-1-antitrypsin screening study (69), it resulted in significantly higher clinically healthy PiZZ compared to PiMM subjects. The measurement of components with the extended liver fibrosis (ELF) test (70) in the same cohorts revealed its usefulness in identifying ZZ cases at an increased risk of developing liver disease later in life (71). In particular, with the ELF test, the tissue inhibitor of metalloprotease-1 (TIMP-1) was significantly higher in ZZ individuals than in matched MM controls. TIMP-1 is an early marker of extracellular matrix remodeling and fibrogenesis and its role as a prognostic biomarker of liver disease in PiZZ subjects deserves further investigation (71).

Recently, the presence of circulating polymers which arise from AAT, produced within the liver in all individuals with PiZZ AATD has been demonstrated (72). The authors of this study also demonstrated a relationship between the polymer level and self-reported abnormal liver function and liver disease in a cohort of 518 PiZZ individuals. Despite being a pilot study, these data allowed the correlation between circulating polymers and liver disease; moreover, the hepatic secretion of circulating polymers of Z-AAT has been clearly demonstrated in a recent article by Fra et al. (73). Extracellular polymers may contribute to inflammatory neutrophil infiltration, since they are chemotactic and stimulating for human neutrophils (74). Circulating polymers could also be biomarkers of AATD-related diseases such as panniculitis and vasculitis, mainly due to the deposition of AAT proinflammatory polymers in different body regions other than the liver. Other studies should be set up to explore these aspects of the disorder.

Angiopoietin-like protein 4

Alpha-1-antitrypsin is an acute-phase protein and in the past five years some non-protease inhibitory functions have been identified in vitro. One of the most interesting discoveries relates to the property of AAT to bind fatty acids and its possible association in vivo with dyslipidemia-related pathologies (75). In a small cohort of PiZZ emphysema patients receiving therapy with AAT (Prolastin®), it was recently found that plasma angiopoietin-like protein 4 (Angptl4) levels correlate with AAT levels and are significantly higher relative to non-treated PiZZ patients. The authors reported on the role of fatty-acid-bound AAT as a regulator of Angptl4 transcription and secretion (76). Angptl4 is a major inhibitor of lipoprotein lipase and is outlined in Figure 2;^{Table 1} the fatty acid-bound form of AAT might indirectly contribute to the inhibition of lipoprotein lipase and hence diminish the local uptake of triglyceride-derived fatty acids, which has a role in cardiovascular diseases. Thus, plasma Angplt4 levels could perhaps serve as novel biomarker to monitor the efficacy of augmentation therapy.

Figure 2. Role of plasma AAT as a fatty acid (FA) binding and transporting protein. AAT binds unsaturated FAs and can deliver FA into tissues and cells. Within the cell, FA rapidly binds intracellular fatty acid binding proteins (FABPs) in the cytosol, enters the nucleus, and induces expression of Angptl4 via activation of HIF-1 and the peroxisome proliferator-activated receptors (PPARs) pathways. Angptl4 is an inhibitor of lipoprotein lipase (LPL). Therefore, the FA-bound form of AAT might indirectly contribute to LPL inhibition (black line) leading to reduced triglyceride (TG) hydrolysis by circulating chylomicrons (CM) and very-low-density lipoproteins (VLDL, and to diminished uptake of TG-derived FAs. AAT-FA-mediated upregulation of CD36 and FABPs expression indicates the Angptl4-related switch in fuel utilization toward the use of AAT bound FAs (blue arrows).

A list of potential biomarkers for AATD is reported in Table 1.

Table 1. Characteristics of different biomarkers.

	Biomarkers
	Desmosine	Fibrinogen	AaVal-360	MMP-9	GGT	Circulating polymers of AAT
Clinical utility	Lung diseases (emphysema COPD- PXE)	Lung disease (exacerbation)	Lung disease (early emphysema)	Lung disease	Liver disease Lung disease (airflow obstruction)	Liver disease Lung disease
Specificity	High	Low	Low	Low	Low
Sensitivity	High	High for inflammation	High	High	Under investigation	Under investigation
Feasibility	Complexed	Feasible	High	High	High	Feasible
Biologic credibility	High	Good	Under investigation	Low	Low	Under investigation

GGT: gamma glutamyl transferase; MMP-9: matrix metalloprotease-9.

Genetic markers

Because of their high stability in tissue and fluids and their rapid release from tissue in to the bloodstream, the potentiality of miRNA as biomarkers is now evident. The ability of miRNA expression profiles to classify discrete tissue types and disease states has positioned their quantification as a promising new tool for a wide range of diagnostic applications. As far as COPD is concerned, Leidinger et al. (77) showed that lung cancer could be discriminated from COPD patients with 90.4% accuracy by comparing the expression of 863 human miRNAs in blood cells. After this interesting study, significant progress has been made toward the contribution of miRNAs in our understanding of COPD development.

The most recent study in this area identified seven miRNAs as promising biomarkers for COPD (78). The authors inferred that a combination of three of these (miR-145-5p, miR-338-3p, and miR-3620-3p) together with the clinical symptoms, laboratory examinations, and imaging should further enhance the early diagnosis of COPD (78).

In AATD, the study of miRNAs as biomarkers of the disease is still in its infancy. MiR-199a-5p was found to decrease in symptomatic MM and ZZ monocytes compared with asymptomatic counterparts (79). The role of this miRNA, as well as others that should be linked to AATD in the future could be prognostic. However, while being a key regulator of the unfolded protein response in AAT deficient monocytes, this miRNA cannot be considered a biomarker yet.

Future perspectives

Preferably, preservation of lung function and reduction in symptom score should be the goal of any new treatment strategy for this condition. These goals require longitudinal studies with a large number of patients and a long-term clinical trial in this study population. Therefore, surrogate markers have been developed and currently clinical researchers in the field are demonstrating that these new surrogates can be validated (80). These biomarkers are being evaluated in newly designed studies that look for the most effective dose of intravenous AAT to reduce biomarker levels in ZZ AATD patients similar to values found in healthy controls. While the focus of AAT augmentation treatment in the past has been on inhibition of alveolar tissue destruction by proteases, new hypotheses have been tested to study the non-protease inhibitor effects of AAT. The clinical importance of such effects on the course of both lung and liver disease in AATD patients remains to be determined.

In the past ten years, clinical scientists have collaborated to evaluate the performance of various biomarkers in plasma and urine from subjects with PiZZ alpha-1-antitrypsin deficiency. In the near future, hopefully, some of these biomarkers should be approved for research to evaluate why the expression of lung disease is so variable; rigorous prospective studies should elucidate the course of this lung disease. Addition of lung densitometry (22,81) and symptom scoring (82) data to these studies will give us more valuable clinical information about the expression variability in lung disease. In these studies, spouses may participate as study controls, an approach frequently used in other rare conditions. Furthermore, the aforementioned biomarkers may also be used in dose-finding studies to identify the optimal dose of alpha-1-antitrypsin treatment, be it intravenous or inhaled. This approach goes beyond the convention that the inherited reduced plasma level in genotype ZZ should be restored to the normal level seen in genotype MM alpha-1-antitrypsin.

There is considerable heterogeneity in the phenotypic expression of lung and liver disease in AATD, both in the incidence of liver tissue damage among individuals with PiZZ deficiency and in the extent of destructive lung disease. In fact, a number of PiZZ individuals who smoke do not develop clinical lung involvement associated with a reduced pulmonary function. All these issues suggest that genetic or external factors contribute to this variability, and these remain largely unidentified. Transcriptomic analysis by RNA sequencing may become a research tool to elucidate this clinical variability, but transcripts with significant changes in expression need to be validated and their role as biomarkers should be further investigated with the support of bioinformatic analyses. Epigenetic regulation of miRNA expression deserves more attention and could offer unexpected potentialities for the development of future personalized therapies in AATD-related COPD.

The emerging tools of network medicine offer a platform to systematically explore the molecular complexity of a particular disease, which is essential to identify and validate new drug targets and biomarkers. Despite its low prevalence, alpha-1-antitrypsin deficiency is characterized by different clinical multi-organ phenotypes. Innovative treatments for these conditions such as siRNA interference to knock down type ZZ AAT synthesis in the liver (23) and repair of damaged liver or lung tissue with bi-potent stem cells (83) are on the horizon, and Systems Medicine/Biology approaches are required to identify which patients are eligible for an early-stage medical intervention with these innovative treatments. This may lead to a reduction in disease severity and may guide clinicians in the application of expensive new treatments to better manage health resources.

Based on the above speculations, we cannot persuade the reader that actually biomarkers can be used in driving the therapy regimens but the following proposals are suggested:

–	The current available biomarkers (DESs, Aα(Val360),) should be evaluated in rigorous prospective studies to further validate them in comparison with HRCT PD15 in AATD patients and spouse controls.
–	The effects of AAT augmentation therapy in different clinical phenotypes of lung disease in the AATD population should be identified and the effects of different dose levels should be analyzed.