Wounds are injuries to the skin that can be superficial, such as erosions affecting only the epidermis and healing without scarring. There are also deeper injuries, such as ulcers, encompassing both the epidermis and the underlying dermis, which heal with scars [1]. In both cases, acute wound healing is achieved by a tightly controlled process of sequential, overlapping phases of hemostasis, inflammation, proliferation, and remodeling. In contrast, chronic, non-healing wounds are those that do not heal in an orderly and timely fashion [2]. They have been referred to as a silent epidemic [3] and are a major clinical, social, and economic issue worldwide [4]. In developed countries, including the US and the EU, it has been estimated that 1 to 2% of the total population will experience a chronic wound at some point in life [5, 6]. The patient population comprises more than 10 million individuals in the US alone [3]. This number is expected to increase due to the aging population and growing incidence of metabolic diseases [7].

In addition to high morbidity, chronic wounds cause increased mortality. The 5-year mortality rate due to diabetic foot ulcers (DFU) is 46.2% and 56.6% after minor and major amputations, respectively. This is higher than the pooled mortality rate for all cancers (31%) [8]. While typical diabetes comorbidities such as cardiovascular and renal disease contribute to the mortality, DFU remains a risk factor for death after multivariate analysis [9]. Moreover, in non-diabetic patients with chronic lower extremity wounds, a high mortality rate (28% in 2 years) has been documented [10]. Hospital-acquired pressure ulcers alone are responsible for 60,000 deaths annually in the US [11].

Chronic wounds also cause a big economic burden, not only because the costs for treatment amount to 2–5% of total health care costs [6, 12, 13] but also due to loss of productivity of patients and caregivers.

Current treatment options are unsatisfactory, and no new drugs have been approved in this field for more than 20 years [14]. Chronic wounds have a multifactorial etiology and are dependent on different variables, e.g., the underlying disease (such as diabetes, arterial occlusions and venous insufficiency), age, nutritional status, the microbial environment [2, 15, 16]. This poses a challenge for the development of new therapies, especially since regular cell culture and animal models do not adequately represent chronicity [17–19].

In this review we will present the underlying biology, as well as current and future therapy options, focusing on human studies and translational models.

Cutaneous wound healing is a tightly regulated biological process requiring the coordinated actions of numerous cellular players to restore the integrity and function of damaged skin. It proceeds through four overlapping but distinct phases: hemostasis, inflammation, proliferation, and remodeling [20].

Hemostasis marks the immediate response to injury, wherein blood vessels constrict to limit blood loss, and activated platelets aggregate with fibrin to form a stable clot. Beyond hemostasis, platelets release key growth factors, such as PDGF (Platelet-Derived Growth Factor), TGF-β (Transforming Growth Factor β), and VEGF (Vascular Endothelial Growth Factor), as well as chemokines, initiating inflammatory and cellular recruitment cascades [21, 22].

Inflammation follows rapidly, serving to clear pathogens and debris while priming the wound bed for repair. This phase is driven by recognition of Damage-Associated Molecular Patterns (DAMPs) and Pathogen-Associated Molecular Patterns (PAMPs) via Pattern Recognition Receptors (PRRs) expressed by both immune (macrophages, neutrophils, dendritic cells) and non-immune (keratinocytes, fibroblasts) cells. DAMPs, including ATP, DNA and intracellular heat-shock proteins, elicit sterile inflammation, while PAMPs, as bacterial toxins or viral single-stranded RNA, stimulate immune responses to infection [23, 24].

These molecular cues initiate a coordinated response involving different cell types. Keratinocytes release cytokines and chemokines, recruit immune cells, and secrete antimicrobial peptides [25, 26]. Dermal adipocytes modulate macrophage responses via adipokines and lipids [27]. Fibroblasts adopt a pro-inflammatory phenotype, secreting cytokines and MMPs (Matrix Metalloproteinases) that shape the wound environment [28]. Neutrophils infiltrate the wound and eliminate microbes via ROS (Reactive Oxygen Species), proteases, and NETs (Neutrophil Extracellular Traps) [29]. Monocytes differentiate into pro-inflammatory macrophages that clear debris and pathogens, stimulate angiogenesis, and amplify the inflammatory milieu [20].

Proliferation begins as inflammation resolves, driven by a phenotypic shift in macrophages toward a pro-resolution state following efferocytosis of apoptotic neutrophils and a metabolic switch to oxidative phosphorylation [30]. Re-epithelialization involves keratinocytes forming a migratory front and a surrounding proliferative zone [31]. Cells at the wound edge undergo partial epithelial–mesenchymal transition, increasing expression of migration-associated integrins and MMPs to traverse the provisional matrix [32]. While pro-inflammatory macrophages support re-epithelialization early, fibroblasts become central in the proliferation phase by producing growth factors that boost keratinocyte migration [31].

Concurrently, fibroblasts proliferate, migrate into the wound bed, and differentiate into myofibroblasts. These cells deposit type III collagen, contract the wound, and release angiogenic factors to promote capillary sprouting [33, 34]. Macrophage-derived TGF-β supports myofibroblast differentiation, although excessive activity contributes to fibrosis and scarring. Notably, tissues that exhibit scarless healing, such as fetal tissues, oral mucosa, and reindeer velvet, are characterized by minimal inflammation and low macrophage recruitment [35, 36].

Remodeling involves the gradual replacement of type III collagen by cross-linked type I collagen, restoring tensile strength [37, 38]. Mechanical cues stimulate myofibroblasts to deposit ECM (Extracellular Matrix) and close the wound [39]. Apoptosis and senescence of myofibroblasts are essential for terminating the repair process and preventing fibrotic overgrowth [40, 41].

Chronic Wounds: Unlike acute wounds, chronic wounds fail to progress through the canonical healing phases. Despite diverse etiologies, they share core features such as persistent inflammation, impaired epithelial and fibroblast function, ischemia, and microbial dysbiosis (Fig. 1).

While elevated pro-inflammatory cytokines (e.g., IL-1β, IL-8) in chronic wound fluid historically supported the notion of excessive inflammation [43–45], recent -omics data suggest a more nuanced picture. Chronic wounds often exhibit deficient rather than exaggerated immune activation, with reduced numbers of pro-inflammatory macrophages in VLUs (Venous Leg Ulcers) and non-healing DFUs (Diabetic Foot Ulcers) [31, 46, 47]. In contrast, healing DFUs show fibroblast subtypes enriched in inflammatory mediators like CHI3L1 and IL-6 [48]. Thus, dysregulated—not merely prolonged—immunity appears to be central to chronicity.

Microbial factors further exacerbate dysfunction. While commensals may promote healing through non-inflammatory immune signaling [49, 50], biofilm-dominant pathogens in DFUs and other chronic wounds maintain a state of immune activation via continuous PAMP release [51–54]. These biofilms, however, evade immune clearance, perpetuating a cycle of impaired immunity and persistent infection.

At the edge of chronic wounds, keratinocyte hyperproliferation with impaired migration is common. In DFUs and VLUs, downregulation of FOSL1 and upregulation of miR-193b-3p have been linked to migration impairment [31, 55]. Meanwhile, β-catenin and c-MYC activation contribute to excessive proliferation [56, 57]. Yet, this phenotype is heterogeneous, with some DFU wounds showing reduced keratinocyte proliferation due to deficient EGF/HGF (Epidermal Growth Factor/Hepatocyte Growth Factor) signaling, while increased keratinocyte apoptosis is reported in some PU (Pressure Ulcer) wounds [31, 58].

Fibroblast dysfunction is another hallmark. Cells isolated from chronic wounds show impaired migration, reduced proliferation, and poor growth factor responsiveness [59–61]. Spatial profiling of VLU and DFU samples revealed an absence of proliferative fibroblasts, and single-cell transcriptomics showed lower expression of ECM/inflammation-related genes in non-healing wounds, underscoring the dual role of fibroblasts in ECM deposition and immune modulation to mediate healing [31, 46, 48].

Vascular insufficiency, whether macrovascular (e.g., arterial ulcers) or microvascular (e.g., diabetic wounds or VLUs), limits oxygen and nutrient supply and is another feature common to many chronic wounds. While transient hypoxia promotes healing in acute wounds by induction of myofibroblasts and ECM deposition, persistent hypoxia in chronic wounds is deleterious, promoting cell death and limiting angiogenesis [42]. Chronic wounds also display imbalanced angiogenic signaling and protease-mediated degradation of growth factors such as VEGF [62, 63], exacerbating the impaired angiogenesis in chronic wounds.

Chronic wounds are classified based on their etiology, and there are frequent overlaps. The most common wound types are venous leg ulcers, arterial ulcers or a combination of the two (mixed ulcers), diabetic foot ulcers, and pressure ulcers (Fig. 2). For more detailed information, the reader is referred to comprehensive review articles on classification and treatment [2, 64–70] and position papers of the European Wound Management Association (EWMA; https://ewma.org/).

Venous leg ulcers are caused by venous valve insufficiency and venous hypertension, frequently associated with varicose veins or thrombosis. Increased capillary pressure and edema can lead to skin breakdown, typically in the gaiter regions, especially the anterior to medial malleolus and the pretibial lower third of the leg. Ulcers are usually shallow with irregular edges and a granulating base, often with heavy exudate. The surrounding skin shows edema, varicosities, hyperpigmentation (hemosiderin), and lipodermatosclerosis [67].

Arterial ulcers arise from advanced peripheral arterial disease (PAD) causing limb ischemia that prevents wound healing. By definition, PAD begins at an ankle-brachial index (ABI) of < 0.9, but the ABI of patients with an arterial leg ulcer is typically below 0.6 or even lower when chronic critical limb ischemia (CLI) is already present [71]. Arterial leg ulcers are extremely painful and typically located on the pretibial area, the distal lateral lower leg, on the tips of toes, or on bony protrusions on the foot [72, 73]. The prognosis depends largely on the degree of PAD, extent and depth of the wound, and degree of tissue infection [71, 74]. Mixed arterial and venous ulcers combine features of the two [73].

Diabetic foot ulcers arise from a combination of diabetic polyneuropathy, repetitive mechanical stress, and peripheral artery disease, which typically affects the lower-leg arteries most severely in cases of long-standing, poorly controlled diabetes mellitus [75, 76]. Motor-, sensory-, and anatomic neuropathy contribute to foot deformity, loss of protective pain sensation, and dry skin – factors that increase the likelihood of developing a DFU [76]. Diabetic polyneuropathy leads to denervation of the short foot muscles, causing the foot skeleton to lose stability and collapse, resulting in deformations that are exposed to intense pressure and shear forces, particularly in the metatarsophalangeal joint of the big toe, the sole above the middle metatarsal heads, the backs of the toes, and the lateral edge of the foot. Moreover, the diminished sensation associated with diabetic neuropathy prevents patients from adequately perceiving pressure-induced tissue damage. Consequently, repetitive stress may progress unnoticed, leading to large defects in the skin and soft tissues of the diabetic foot [77]. In parallel, impaired perfusion and increased susceptibility to infection common in diabetes mellitus further exacerbate tissue vulnerability and increase the risk of severe complications [76, 78]. Prevention consists of optimal foot support, ideally by wearing orthopedic shoes, and revascularization to improve tissue perfusion. Given the high recurrence rate (40% within 1 year to 65% within 5 years [75]), combined with the fact that neuropathy renders the lesions painless, careful monitoring of the feet is necessary to avoid recurrence or complications like infection and osteomyelitis, which can further delay wound healing and frequently lead to amputation [79].

Pressure ulcers typically develop in bedridden or wheelchair-bound patients upon prolonged unrelieved pressure over bony prominences such as the sacrum, trochanters of the thighs, ischial tuberosity, heels, or over the spine or back of the head. In addition to immobility, other risk factors include poor nutritional status (malnutrition), friction, shear stress, or moisture. These wounds may be accompanied by undermining or tunneling. Many pressure ulcers can be prevented by regular repositioning and appropriate support surfaces [80].

Atypical ulcers are considered wounds that do not belong to the most typical wound categories, i.e., venous, arterial, mixed, pressure, or diabetic foot ulcers, and they account for approximately 20% of all chronic wounds. Differential diagnosis needs to be broad to include rare as well as more common entities [81]. Some of the more common atypical ulcers are ecthyma (chronic ulcers caused by bacterial infections), vasculitis (inflammatory condition that destroys the wall of blood vessels, leading to ischemic necrosis and chronic wounds), pyoderma gangraenosum (rare ulcer based on autoinflammatory dysregulation of neutrophil granulocytes), Martorell ulcers (caused by hypertensive ischemic atherosclerosis), and the related calciphylaxis, which occurs in patients with terminal renal failure, as well as drug-induced chronic wounds, such as hydroxyurea ulcers (see [81, 82] for comprehensive reviews of atypical ulcers). Also belonging to atypical wounds are those with underlying genetic diseases including epidermolysis bullosa, with gene mutations leading to loss of adhesion and affecting epidermal integrity [83], sickle cell anemia [84] or Klinefelter syndrome [85]. Some atypical wounds are also of neoplastic origin, e.g., Marjolin’s ulcer, a squamous cell carcinoma arising in a chronic wound or scar. Other skin tumors, e.g., basal cell carcinoma, malignant melanoma, cutaneous B cell lymphomas, and Kaposi sarcoma, present as non-healing wounds [67]. Diagnosis usually requires biopsy and specialized testing to identify the underlying cause [73].

For all chronic wounds, accurate diagnosis is crucial to guide therapy. A structured, practical approach such as the ABCDE rule is recommended [86]:

Often, chronic wound evaluation is multidisciplinary—involving dermatologists, wound care specialists, vascular surgeons, endocrinologists, podiatrists or orthopedic surgeons, and other specialists. Regular documentation of wound dimensions and characteristics is recommended to objectively assess the healing trajectory [65].

Recent advances in translational research in wound healing treatments reveal a fundamental shift from passive wound management to active, multi-modal, and increasingly personalized therapeutic interventions. Most of the trials in the clinicaltrials.gov database comprise medical devices and observational parameters. There is a paucity of clinical studies with novel pharmacological treatments for chronic skin ulcers, with the majority focusing on biologic therapies and diabetic foot ulcers as an indication (Fig. 3 and Suppl. material 1: table S2).

Acknowledging the failure of past single-agent approaches in the complex pathophysiology of chronic wounds, a new class of multimodal therapies addressing the multifactorial nature of chronic wounds is reaching clinical maturity. They encompass trials focusing on placental-based products and stem cell therapies. Placenta-based products deliver a supporting scaffold as well as a natural cocktail of growth factors and healing promoting substances [140]. Stem cell therapy holds promise to enhance tissue regeneration due to their ability to adaptively respond to tissue injury and inflammation by providing paracrine signals which alter the wound environment toward a pro-healing state or even directly participate in tissue regeneration when applied topically [141]. An important development was the identification of the ABCB5 surface marker on skin stem cells, which is used for isolating homogenous populations of stem cells from discarded skin tissues, thus decreasing the variability of the treatment and fulfilling the stringent standards of good manufacturing practice [142]. A special class of therapy in clinical development is AUP-16, a genetically engineered live bacterium delivering directly onto the wound a cocktail of three human proteins: Colony Stimulating Factor (CSF-1), Fibroblast Growth Factor 2 (FGF-2), and Interleukin 4 (IL-4). Through this mechanism, AUP-16 exerts a multi-target mode of action, simultaneously modulating inflammation, angiogenesis, and proliferation [143]. Furthermore, therapy using genetically corrected stem cells for the treatment of epidermolysis bullosa has progressed from preclinical ex vivo correction of patient keratinocytes to early clinical trials, demonstrating durable restoration of laminin 332 and sustained wound healing in junctional EB [144, 145]. Current efforts and clinical studies focus on the challenging delivery and sustained expression of corrected collagen 7 to patients with a more severe form of the disease, recessive dystrophic EB [146].

In contrast to the high number of trials focusing on biologicals, the investigation of drug substances for therapy of chronic wounds is still at early stages. Currently, only 12 pharmacological candidates are undergoing clinical investigation, covering various mechanisms of action (Suppl. material 1: table S2). While anti-inflammatory properties are a recurring theme, many compounds target distinct aspects of the wound-healing process. Among the substances focusing on the reduction of chronic inflammation are: i) the pan-JAK inhibitor MDI-1228 [147, 148]; ii) TCP-25, a thrombin-derived peptide with anti-inflammatory properties based on scavenging of pathogen-derived DAMPs [149]; iii) S42909, a NADPH oxidase inhibitor that dampens the production of reactive oxygen species (ROS), thereby alleviating chronic inflammation [150]; and iv) Melatonin, which exerts antioxidant and anti-inflammatory effects beyond its role in the circadian rhythm [151]. Substances with other diverse modes of action include: i) TR-987, a yeast glucan polymer which acts as an infection decoy and activates host immune cells to reboot the healing process [152]; ii) Esmolol and timolol, beta blockers with pleiotropic actions including promotion of keratinocyte-, fibroblast- and endothelial cell migration [153]; iii) Folinic acid, involved in biosynthetic pathways of amino acids and hence important for maintaining normal cell growth and replication [154]; iv) ENERGI-F703, an adenine-containing gel found to accelerate wound closure in an early clinical study with DFUs, presumably by increasing cellular energy [155]; v) Nitric oxide, an endogenously produced mediator that promotes vasodilation, angiogenesis and antimicrobial activity [156]; vi) Deferoxamine, an iron chelator that stabilizes HIF-1α and induces hypoxia-responsive, pro-angiogenic gene expression, promoting angiogenesis and tissue regeneration [157]; and vii) Cilostazol, a PDE3 inhibitor and antiplatelet agent with vasodilatory properties [158].

The advent of artificial intelligence is expected to improve the management of chronic wounds (reviewed in [159]) (Fig. 3). The application of AI is already leading to fast and efficient documentation of the wound with smartphone apps or dedicated 3D camera systems, encompassing many parameters of the WUHWS triangle of wound assessment, e.g., wound area, depth, and tissue composition [159–162]. By combining wound imaging documentation and information from electronic health records of >1.2 million wounds, wound healing trajectories can be predicted with high accuracy, aiding in early identification of patients at risk of developing a chronic wound [163]. With smart dressings containing electrochemical sensors, AI could also help monitor biological processes directly in the wound environment [159, 164] to inform the healthcare professional remotely when a change of bandage or therapy is necessary. However, these devices are still in the experimental stage. Future development is also focusing on engineering dressings that deliver medication upon specific triggers in the wound environment. Ideally, in the future, treatment could be continuously adapted to respond to actual changes in the wound environment without the stress of changing bandages [165].

However, currently such strategies are severely hampered by the limited number of therapeutics in clinical use and registered for chronic wounds. Considering the complexity of chronic wounds, it is unlikely that targeting a single biological effect/process is sufficient to correct the entire pathology of a chronic wound. AI and machine learning can integrate large -omics datasets to identify biomarkers, new drug targets, or target combinations, as recently shown [166, 167]. The major challenge remains the validation of these targets for human therapy, since the predictive value of presently used animal models is limited [18, 168]. Therefore, test systems using human wound material are desirable for the characterization of future therapeutics. This includes 3D models of human skin, either skin biopsies or in vitro models, such as skin-on-a-chip, organoids, and bioprinted skin equivalents [19], as well as personalized cellular assays of chronic wounds [44]. Doerfler et al. (2025) have recently shown that exudates from chronic wounds transfer their chronic phenotype to normal human fibroblasts grown in cell culture, leading to production of inflammatory mediators and impaired proliferation. The assay was used to test the rescue capacity of approved wound therapeutics in the context of individual wound exudates, paving the way for personalized pre-testing of drugs for chronic wound therapy, as well as its use as a companion diagnostic. This assay also has the potential to help identify and characterize new drugs or aid in the validation of targets identified by -omics approaches [44, 169]. The use of wound exudates as a source of biomarkers for personalized medicine is also being investigated by proteomics [45]. Clinical validation of wound exudate-based cellular and molecular biomarkers will be necessary to finally demonstrate their usefulness to improve translational medicine for chronic wounds.

A major challenge in translational science for chronic wounds is the regulatory framework. The current regulatory requirement of full and lasting wound closure for approval of a new drug or therapeutic device is unrealistic and represents a big hurdle. Wound healing societies such as the American WHS and the Wound Care Collaborative Community (WCCC) initiative (https://www.woundcarecc.org/wp-content/uploads/2025/05/2025-WCCC-Combined-Summit-Deck-.pdf, accessed 29.07.2025) are working together with the FDA on defining meaningful and patient-centric clinical endpoints. Suggestions include percent wound area reduction, reduced infection, reduced pain/analgesia use, increased function and ambulation, and improved quality of life. Hopefully, incorporating these measures will make it more attractive for pharma and device companies to conduct clinical studies that include patients with problematic wounds that are unlikely to close within the currently postulated time frame of 12 weeks.

Chronic wounds remain a major challenge in modern medicine, reflecting a convergence of complex pathophysiological mechanisms, comorbidities, and socioeconomic burdens. While significant progress has been made in understanding the cellular and molecular drivers of chronicity—such as persistent inflammation, ischemia, microbial dysbiosis, and cellular dysfunction [42, 170]—this knowledge has not yet translated into major treatment breakthroughs. This is due in part to the multifactorial nature of non-healing wounds and the marked heterogeneity of patients, who often present with multiple comorbidities and diverse medication regimens. The success of future therapeutic interventions will likely depend on the application of personalized medicine, using biomarkers to stratify patients during clinical development and/or companion diagnostics to predict the therapeutic success of specific treatments. Promising biomarker candidates [14, 45] and ex vivo models for new drug discovery [44] have been reported and await clinical validation. Another obstacle to clinical development is the limited predictive value of conventional preclinical animal models, which fail to fully replicate the chronic wound microenvironment [18].

Despite these challenges, the field is experiencing a surge of innovation. Multimodal therapies—such as placental-derived products, stem cell-based interventions, and engineered biologicals—are being developed to address the diverse cellular and molecular deficits in chronic wounds. Assuming clinical efficacy is demonstrated, these therapies still come with major challenges related to manufacturing and standardization, high costs, and low accessibility due to the need for specialized medical facilities. Therefore, there is a major need for the development of effective drugs to promote healing of chronic wounds. Drugs are manufactured with standardized and controlled processes, such as chemical synthesis and bioreactors, leading to low variability, which decreases costs and increases accessibility beyond specialized medical centers all the way to home-based care.

The integration of artificial intelligence offers powerful opportunities for real-time wound assessment, predictive modeling, and personalized therapy, particularly when combined with smart dressings and exudate-based biomarker assays. Moreover, the adoption of biologically relevant human preclinical models and more meaningful clinical endpoints as alternatives to complete wound closure promises to bridge the translational gap. Collectively, these developments mark a pivotal shift toward individualized, patient-centric care strategies that could redefine the therapeutic landscape for chronic wound management (Fig. 5).