As mentioned above, AATD is caused by the abnormal folding of the AAT Z variant, which promotes its polymerization and intracellular retention, constituting the initial event in the pathogenesis [1,11].

The accumulation of misfolded AAT in the ER causes cellular stress and activates the unfolded protein response (UPR), a protective mechanism aimed at restoring cellular balance. However, when this activation is chronic, it contributes to inflammation, cell death, and progression of liver damage. At the same time, reduced levels of functional AAT in the circulation disrupt the protease–antiprotease balance in the lung, allowing enzymes such as neutrophil elastase to damage lung tissue. Cells normally remove abnormal proteins through quality control systems, such as endoplasmic reticulum-associated degradation (ERAD), as well as proteasomal degradation and autophagy. When these mechanisms are overwhelmed or impaired, misfolded proteins accumulate further, worsening tissue injury [12–15].

These processes represent important therapeutic targets. Emerging strategies such as protein folding modulators, autophagy inducers, polymerization inhibitors, and gene therapies targeting SERPINA1 aim to directly intervene in the underlying molecular mechanisms, with the goal of correcting both hepatic toxicity and systemic deficiency which will be further discussed ahead.

SERPINA1 is a highly polymorphic gene with more than 500 variants described to date [16], exhibiting a broad spectrum of pathogenicity ranging from benign to disease-causing. An ever-increasing spectrum of SERPINA1 mutations has recently been identified, giving rise to AAT variants associated with deficiency and/or dysfunctional behavior [17,18].

Pathogenic AAT variants are classified as null or deficient. Null variants (Q0) are characterized by lack of AAT production, leading to undetectable serum levels and a high risk of early-onset emphysema, even in non-smokers [19]. Deficient variants are caused by point mutations causing intracellular retention, polymer formation, increased degradation, or reduced antiprotease activity. These mechanisms are not mutually exclusive producing variable degrees of quantitative and functional deficiency. For example, the Z allele, the most common severe variant, is highly polymerogenic and characterized by marked intracellular retention, although the small fraction that is secreted retains residual, but impaired, inhibitory activity against neutrophil elastase [20]. In contrast, the rare F variant is non-polymerogenic, functionally impaired, and only mildly reduced in circulating levels [21].

Genetic mutations alone, or in combination with environmental factors such as cigarette smoke, which is known to induce post-translational modifications including oxidation or polymerization [22], may disrupt AAT trafficking in pulmonary endothelial cells, including its cellular uptake, intracellular transport, and secretion, thereby affecting its functional bioavailability within specific lung compartments [23].

The diverse pathogenic mechanisms associated with AAT variants, which are expressed in a codominant fashion and can therefore result in a wide range of genotypic combinations [24], represent the phenotypic consequences of the heterogeneity in the clinical presentation of AATD patients.

Recent spatial and genomic studies have refined our understanding of AATD-associated liver disease. The spatial study recapitulates known markers of disease progression and identifies dysregulated proteins. A key finding is the early activation of the peroxisomal response in patients with low-grade liver fibrosis, whereas this response is delayed in advanced fibrosis. Because this association is not observed in alcohol-related liver disease, it may represent a mechanism specific to AATD [25]. Analysis of somatic mutations has shown that chronic Z-AAT proteotoxicity influences hepatocyte clonal evolution, favoring the expansion of clones with acquired alterations that reduce mutant AAT production or its intracellular burden [26].

AATD clinical diagnosis is challenging due to phenotypic heterogeneity and the lack of pathognomonic symptoms [27,28]. Contrary to the typical AATD patient profile of a young (under 50 years of age at the time of diagnosis), nonsmoker individual with basal panacinar emphysema, over 50% of cases are diagnosed after age 50 [29–31], around 36% of patients present apical emphysema [32], and approximately 80% are current smokers (8%) or former smokers (71%) [32–34].

The diagnostic algorithm must be initiated upon detection of specific pulmonary, hepatic, or systemic “red flags” (Table 1) [35]. According to the ATS/ERS statement, AATD testing is recommended in individuals with COPD or otherwise unexplained persistent airflow obstruction, unexplained chronic liver disease, necrotizing panniculitis, and unexplained bronchiectasis [6,28]. These documents also specifically mention granulomatosis with polyangiitis/PR3-ANCA-positive vasculitis. However, this indication should not be generalized to all ANCA-associated vasculitis [36]. Genetic counseling and targeted testing should be offered to first-degree relatives of individuals with a confirmed deficiency-associated SERPINA1 variant. A family history of premature severe COPD may increase clinical suspicion but is not, by itself, an indication for familial testing [4,6,37,38].

The most significant advance has been implementing active case-finding programs and diagnostic algorithms in primary care combining clinical criteria with initial laboratory tests to facilitate early detection [8,39].

Despite the success of active case-finding and primary care algorithms in improving early detection, clinical and radiological criteria alone cannot confirm the disease. A definitive diagnosis requires the use of comprehensive molecular strategies, including SERPINA1 genotyping and sequencing. Although there is no universally accepted diagnostic algorithm, a widely accepted one has been proposed [5,8,35]. The diagnostic approach includes: (i) measurement of plasma AAT levels, (ii) protein phenotyping, (iii) genotyping targeting the most common variants [28], and (iv) full sequencing in those cases with low AAT serum levels but no abnormal allele is found by PCR or when null or rare variants are suspected [40–42].

Overall, these molecular techniques are needed to identify rare alleles or null variants improving diagnostic accuracy where rapid methods fail [43–45]. In addition, the development of high-throughput platforms allows familial and targeted population detection in at-risk populations. Importantly, precision molecular biology-based approaches are being developed that correlate genotype with clinical risk, enabling better patient stratification [5].

Biomarkers proposed for use in AATD differ substantially in their intended purpose. Some serve a diagnostic role (confirming deficiency or genotype), others are pharmacodynamic (demonstrating biochemical exposure to treatment), and others are proposed as markers of ongoing disease activity or progression. However, currently, no single circulating biomarker is specific and validated enough to measure pulmonary disease activity or emphysema progression in AATD.

Identifying reliable biomarkers for monitoring the AATD progression is an area of intense research. From a clinical point of view, it is necessary to identify biomarkers that help identify patients with a poorer prognosis.

Preclinical models of AATD, including cell cultures, and animal and human models have been essential for understanding disease mechanisms and developing new therapies.

Common cellular models include HEK293, COS-7, and CHO cells engineered to express mutant AAT variants, especially the Z allele, and are used to study polymerization, ER retention, and degradation pathways such as ER-associated degradation and autophagy [80–82]. Hepatocyte-derived cell lines such as HepG2 and Huh7 hepatoma-derived cells provide a more physiological hepatic context and are widely used to analyze secretion defects, polymer accumulation, ER stress, and pharmacological responses [83,84]. These systems are valued for high transfection efficiency, strong protein expression, and suitability for proteostasis and therapeutic screening.

More advanced cellular models include patient-derived iPSCs that has been differentiated into hepatocyte-like cells [85], which preserve patient genetic background and reproduce intracellular polymer accumulation, making them highly valuable for precision medicine and gene-editing validation [86–88]. In AATD-patient-derived iPSCs, the correction of Z variant can restore the structure and function of AAT [86].

Additionally, three-dimensional liver organoids recapitulated the key features of Z-AAT deficiency [89,90]. Liver organoid models derived from iPSCs, primary cells and from liver biopsies of patients with specific genotypes (i.e. ZZ, MZ) are being very useful for investigating responses and pathogenic features enabling an improved tissue architecture, and long-term modelling [91,92].

Animal models are essential for studying systemic AATD. PiZ transgenic mice (C57BL/6-PiZ and NSG-PiZ), expressing human Z-AAT, are widely used for liver disease and show hepatocellular polymer accumulation, ER stress, and progressive liver injury [81,93,94]. However, PiZ mice do not develop spontaneous emphysema, since murine neutrophil elastase is not efficiently inhibited by human Z-AAT, and mice lack an endogenous protease–antiprotease imbalance comparable to that in humans [93,95]. This model is therefore useful for hepatic but not pulmonary disease. In addition, the Serpina1 a–e knockout mice lack endogenous AAT and model pulmonary disease with protease–antiprotease imbalance and emphysema-like damage. Limitations include the absence of the toxic gain-of-function polymer accumulation seen with the Z variant, meaning this model isolates loss-of-function effects but does not recapitulate the combined mechanisms operating in human PiZZ disease [96]. Recently, a new AAT-null PiZ strain combines SERPINA1 knockout with human PiZ expression [97].

More recent approaches have generated mice that both lack murine AAT and express human Z-AAT, aiming to combine loss-of-function and gain-of-function mechanisms in a single model. CRISPR/Cas9-based gene editing has also been used to generate more precise humanized models and to explore gene-correction strategies as potential therapeutic approaches. These models remain relatively new, and long-term data on their ability to reproduce progressive emphysema and their utility for therapeutic screening are still limited [98,99].

The evolution of gene therapy for AATD arises from already classic studies that established the proof of concept for hepatic gene transfer. Pioneering research by Aliño et al. first demonstrated the feasibility of using non-viral vectors and liposomes to deliver the SERPINA1 gene in animal models [100,101]. Further studies by this group, explored hydrodynamic injection to achieve sustained expression of human AAT protein within the liver parenchyma [102,103].

Currently, the field has advanced toward correcting the SERPINA1 locus using adenine base editors (ABEs). Preclinical studies using lipid nanoparticles achieved 35% editing efficiency in hepatocytes, raising serum AAT above the therapeutic threshold of 11μM and reducing the intrahepatic polymers [104]. Additionally, CRISPR/Cas9 via AAV8 vectors deleted the Z allele, reducing fibrosis and inflammation [98]. These approaches represent a shift from augmentation therapy to a potential one-time curative treatment by addressing the genetic cause in both the liver and the lungs. Recently, it has been shown that ABE-based correction of the Z variant in patient iPSCs reduces aberrant AAT accumulation and restores its secretion [91,105]. Finally, ABE editing has recently entered clinical development for AATD. BEAM-302 is a liver-targeted adenine base editor designed to correct the pathogenic SERPINA1 Z variant without generating double-strand DNA breaks and is currently being evaluated in a first-in-human phase 1/2 trial (NCT06389877). Preliminary findings have shown the appearance of corrected M-AAT, increased total and functional AAT levels, and reduced circulating Z-AAT, providing initial clinical proof of mechanism. Nevertheless, the BEAM-302 findings remain preliminary, and longer follow-up and larger cohorts are required to establish safety, editing durability, and effects on pulmonary and hepatic clinical outcomes.

Altogether, these results demonstrate that base editing is a feasible and effective tool for correcting the Z variant in AATD patient cells. However, despite this progress, critical translational challenges remain, including off-target genotoxicity and immunogenicity [106].

RNA editing using endogenous adenosine deaminases acting on RNA (ADAR) enzymes enables temporary A-to-I conversion, enabling the translation of functional M allele [107]. without altering the genome, diminishing the risk of permanent genotoxicity while achieving efficiencies above 50% and raising serum AAT levels to >30μM [108]. Clinically, the RestorAATion-2 trial (NCT06405633) has demonstrated the restoration of protective M-AAT levels in ZZ patients with a favorable safety profile [97].

Antisense and small RNA interference (siRNA) therapies significantly advance AATD-related liver disease treatment by blocking misfolded Z-AAT protein synthesis and reducing hepatic polymer accumulation. Fazirsiran, a hepatocyte-targeted siRNA, has demonstrated up to a 94% reduction in serum Z-AAT, 83–94% decrease in liver protein concentration and significant fibrosis regression in 58% of patients without affecting lung function [9,10]. Similarly, agents like belcesiran, an investigational RNA-interference therapy designed to reduce hepatic AAT synthesis, entered clinical development for AATD-associated liver disease. Initially the drug showed effective dose-dependent AAT reduction, but its development was subsequently discontinued [109].

Altogether, these results indicate that gene silencing is a potential therapeutic approach to stop the progression of liver damage. Phase 1 and 2 studies of gene-silencing therapies have provided evidence regarding their safety and potential efficacy. However, definitive evidence awaits the results of RCTs.

Neutrophil elastase (NE) and proteinase 3 (PR3) are the major proteases inhibited by AAT [114]. Both proteases are synthesized by neutrophils and stored inside the azurophilic granules of these cells [115].

NE is involved in processes such as disruption of the alveolar-capillary barrier, oxidative stress, inflammation, autophagy, and apoptosis. While it contributes to host defense at physiological levels, its dysregulation can cause emphysema and chronic inflammation, making its inhibition a promising therapeutic strategy [116,117].

Sivelestat is an irreversible NE inhibitor that reduces vascular permeability in acute lung injury and is approved for treating acute respiratory distress syndrome in some countries such as China and Japan [118].

POL6014 is a selective inhaled inhibitor that achieves high lung concentrations and significantly reduces NE activity in cystic fibrosis patients [119].

Similarly, Sirtinol has shown promising results in animal studies for treating inflammatory lung diseases [120].

Alvelestat is an orally administered, reversible NE inhibitor that is well tolerated, binds elastase similarly to AAT and penetrates lung tissue [121,122]. Clinical trials (ASTRAEUS and ATALANTa) showed that doses of 120mg and 240mg twice daily significantly reduced NE activity, with greater suppression at 240mg [123]. However, no improvements in lung function were observed over 12 weeks, likely due to the short follow-up period [124].

Epigenetic modulation of SERPINA1 is an emerging therapeutic strategy. DNA methylation, histone modifications, and non-coding RNAs regulate the SERPINA1 gene expression and may explain interindividual variability in disease severity [125].

DNA methylation has been shown to influence SERPINA1 transcriptional activity. Differential methylation of the SERPINA1 promoter has been observed in peripheral blood mononuclear cells and is associated with variations in circulating AAT levels, supporting its role as an epigenetic regulatory mechanism [126].

In addition to transcriptional control, post-transcriptional regulation plays a key role in determining AAT expression. Alternative splicing of non-coding exons and the generation of distinct mRNA isoforms contribute to tissue-specific expression patterns of SERPINA1[127]. Furthermore, RNA structural elements and upstream open reading frames in the 5′ untranslated region have been shown to modulate translation efficiency, highlighting an additional layer of regulatory complexity [128].

MicroRNAs also contribute to the epigenetic regulation of SERPINA. For example, miR-320c has been identified as a regulator of SERPINA1 expression and is associated with inflammatory lung disease, suggesting that microRNA-mediated mechanisms may influence disease variability and progression [129]. Increased expression of miR-199a-5p has been linked to the regulation of UPR components in the monocytes of asymptomatic PI*ZZ individuals, suggesting its potential as a biomarker for disease severity [125].

Importantly, these regulatory mechanisms are potentially reversible, making them attractive pharmacological targets.