Biomarkers to predict outcomes in diabetic foot ulcers

Biomarkers to predict outcomes in diabetic foot ulcers | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL View This is a preprint and has not been peer reviewed. Data may be preliminary. 29 January 2026 V1 Latest version Share on Biomarkers to predict outcomes in diabetic foot ulcers Authors : Julie Okiro 0009-0003-4269-3191 , Kellie Fortune , Eimear Daly , Luca McCann , Merhan Soltan , Seamus Sreenan , and Fabio Quondamatteo [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.176970084.40139491/v1 Published VIEW Version of record Peer review timeline 268 views 77 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Diabetic Foot Ulcers (DFUs) are among the most feared complications of Diabetes Mellitus (DM). The management of DFUs emphasises limb salvage, and to achieve this, clinical tools are utilised to identify patients who may require a more aggressive initial approach. Current clinical prediction models fail to account for variability in ulcer characteristics and patient-specific comorbidities, limiting their precision in individualising outcome prediction. This review explores the emerging role of molecular biomarkers in personalising DFU outcome prediction. The pathophysiology of DFUs is examined with an emphasis on disruptions in wound healing specific to DM, focusing on biomarkers involved at different stages of wound healing. This review highlights studies that have shown predictive potential of several biomarkers in a variety of biological samples from patients with DFUs. Despite promising findings, challenges remain in their clinical adoption. Larger studies and the development of accessible, biomarker-based diagnostics are essential to translate this approach into clinical settings and ultimately reduce the global burden of DFUs through personalised therapy, which would considerably increase the quality of life of people with DM. Introduction Diabetic foot ulcers (DFUs) are the result of a complex interplay of diabetes mellitus (DM) related complications: peripheral neuropathy 1-3 ; peripheral vascular disease (PVD) 1,4,5 , and immune dysfunction 6-9 . Peripheral neuropathy includes sensory neuropathy, which causes a loss of protective sensation, and motor neuropathy, which creates foot deformities, increasing the risk of injuries at high-pressure points. Meanwhile, PVD causes tissue hypoxia and ischaemic injury, creating an ideal environment for bacterial infections. These conditions, exasperated by chronic hyperglycaemia, disrupt cellular processes 10-15 that are crucial for the phases of wound healing. The challenges in fully understanding the trajectory of DFUs are further heightened by their extremely individual nature. Some ulcers are predominantly ischemic, or neuropathic, and others are mixed, neuro-ischemic 16-18 , each with distinct prognostic challenges. Comorbidities like cardiovascular disease 17,19,20 , renal dysfunction 19-21 , and obesity 22 further influence healing potential, as do behavioural factors, such as adherence to wound care and offloading protocols 23 . Psychosocial barriers, including depression or limited healthcare access 23 , add another layer of complexity. Even the physical characteristics of DFUs vary widely. Differences in size, depth, infection, or location profoundly impact outcomes. For instance, larger, deeper ulcers with bone involvement or infection and ulcers located on pressure points often carry poorer prognoses 16,17,20,24,25 . In essence, it is extremely frustrating for both patients and clinicians to predict the outcome of an individual’s foot ulcer. The complicated phenotype of DFUs emphasizes the need for individualized DFU prediction models. In this era of precision medicine, researchers are exploring the potential of biomarkers to predict a wound’s healing potential and offer a personalized approach to treatment. Such markers, when applied broadly in clinical practice, could revolutionize DFU care, enabling more targeted therapies 26,27 and significantly increasing the likelihood of successful outcomes. The present review highlights the molecular factors and cellular biological processes that regulate the healing of diabetic wounds. It examines the complex mechanisms that can either facilitate recovery or conversely, hinder progress, potentially leading wounds to a chronic and debilitating state. Beyond elucidating these mechanisms, this narrative explores their potential as predictive factors, offering insights that could significantly impact early stage wound management. To provide context, this paper briefly addresses key supplementary aspects of DFUs, including pathophysiology, epidemiology, and current treatment strategies, as well as the complex dynamics of the wound healing process. For a more comprehensive discussion of these aspects, readers are directed to several excellent reviews which are referenced in the next sections. General considerations on Diabetic Foot Ulcers DFUs are among the most feared complications of DM 28,29 , affecting an estimated 18.6 million people globally 30 , and are associated with significant morbidity, mortality, and socioeconomic burden. A recent meta-analysis revealed that the global prevalence of DFU amongst patients with diabetes mellitus stood at 6.3% in 2021 31 , emphasizing the scale of the problem. As the prevalence of diabetes continues to increase, with projections indicating 783 million affected individuals by 2045 worldwide 32 , the occurrence of diabetic foot ulcers (DFU) is also expected to increase. Moreover, the global health and financial burden of diabetic foot ulcers (DFUs) is substantial 33-43 . In 2014 it was estimated that the annual cost per ulcer episode may range from approximately 370 USD to 7,500 USD in low- and high-income countries respectively 44 . In addition to the financial burden, DFUs also pose significant morbidity and mortality. DFUs account for a significant number of Emergency Department visits and showed an increased trend of 28.2% between 2006 and 2010 in the US 45 . The high amputation 5,43 and mortality rates 46-49 of DFUs highlight the critical burden of the disease. It has been shown that 6.8 million people worldwide with DFU have undergone amputation and the years lived with disability (YLDs) due to amputation without prosthesis have increased by 31.5% since 1990 30 . In the U.S., 34,708 major and 75,932 minor amputations were reported in a 4-year period 45 further reflecting the growing impact of these complications of chronic non-healing ulcers. The five-year pooled mortality rate for DFUs is 24.6–30.5%, climbing to 45.4% at ten years 47,48 . These rates are comparable to those of many cancers 48 (Figure 1), emphasising the urgent need for enhanced prevention and management strategies. Among over 350 significant health conditions, diabetic foot disease (DFD) ranks 13 th in global disease burden, 20 th in global disability, and 21 st in global mortality burden, surpassing other diabetes complications such as ischemic heart disease (except in mortality ranks), chronic kidney disease, and diabetic retinopathy. 35 Considering all the evidence, it is undeniable that DFU represents one of the most devastating complications of diabetes. While they are curable, wounds that do eventually close can recur. Recurrence rates of 40% within a year and up to 65% within 5 years have been reported 50 . This cycle ensures that DFUs remain a persistent burden. Thus, there is a push to reduce the global incidence of DFD and subsequently chronic DFUs 35 . To achieve this, it is vital to understand the molecular mechanisms driving foot ulcer chronicity in diabetes mellitus. Chronic DFUs arise from disruptions at every phase of the normal wound healing process. In the next section, we will briefly highlight the specific molecular mechanisms underlying these disruptions, investigating how diabetes mellitus alters a physiological repair process into a chronic, persistent wound state. For a more comprehensive analysis of wound healing in general the reader can refer to several excellent review papers 51,52 . Impaired wound healing in diabetes mellitus Wound healing is a complex, dynamic process that typically progresses through four seamless phases: haemostasis, inflammation, proliferation, and remodelling 51,52 . In individuals with diabetes, this complex process can be disrupted at any of these stages, frequently leading to chronic wounds that are resistant to healing via the usual therapeutic interventions. Haemostasis phase of wound healing The process begins with the haemostasis phase which is initiated within minutes of injury. This phase leads to cessation of bleeding, establishes a barrier which protects the wound against pathogen invasion, preserves blood vessel integrity and prepares the wound site for subsequent repair. It is characterized by a series of cellular and molecular interactions. Endothelial cells initiate vasoconstriction by secreting vasoactive substances such as endothelin-1 53 to minimize blood loss. Concurrently, platelets adhere to exposed subendothelial structures, including collagen and von Willebrand factor (vWF) 54 . These platelets aggregate to form a temporary plug and secrete growth factors, including platelet-derived growth factor (PDGF) and transforming growth factor-beta (TGF-β), which recruit immune cells and trigger the inflammatory phase 51 . The formation of a fibrin clot stabilizes the wound and provides a scaffold for the migration of immune and repair cells 55 , initiating the subsequent phases of healing. In the context of diabetes, this critical phase is significantly disrupted. Chronic hyperglycaemia promotes the formation of advanced glycation end-products (AGEs), which generate oxidative stress and compromise both platelet and endothelial cell functions, 56-59 impairing the release of essential growth factors. Furthermore, cardiovascular comorbidities common in individuals with diabetes often necessitate the use of antiplatelet medications, which, to a small extent, could further compromise platelet function during this early stage 60 . Together, these factors undermine the foundation upon which the wound healing process depends. Inflammatory phase of wound healing Following haemostasis, the inflammatory phase commences 51 . The body’s innate immune system plays a central role in this phase, eliminating pathogens, debris, and necrotic tissue 52 to prepare the wound for repair. The process is initiated by vascular injury, whereby damaged blood vessels dilate and become more permeable. This process is mediated by histamine released from mast cells 61 and nitric oxide (NO) from endothelial cells 62 . The resulting vasodilation facilitates the infiltration of immune cells, with neutrophils acting as the most prominent first responders 61 . These cells phagocytose bacteria and debris, releasing antimicrobial agents such as reactive oxygen species (ROS) and proteases 63 . The inflammatory phase is characterized by a milieu of pro-inflammatory cytokines such as IL-1β, IL-6, and TNF-α, that amplify the immune response and recruit additional cells, including monocytes 61 . Upon arrival, monocytes differentiate into macrophages which ultimately become the predominant immune cell type. Macrophages eliminate apoptotic neutrophils through efferocytosis 64 and secrete anti-inflammatory cytokines such as interleukin-10 to transition the wound toward repair 65 . They also release growth factors, including vascular endothelial growth factor (VEGF) and PDGF, initiating the next step of healing 51 . Throughout this phase, pro-inflammatory cytokines create a feedback loop, ensuring a sustained and robust inflammatory response until the wound is sufficiently debrided and ready for the subsequent phases of wound healing 61 . Eventually, macrophages transition from a pro-inflammatory (M1) to an anti-inflammatory (M2) phenotype 66,67 which will coordinate efforts towards formation of new extracellular matrix (ECM). Diabetes significantly impairs the inflammatory phase through a combination of dysfunctional immune cell activity, prolonged inflammation, and disrupted signalling pathways 14 . Firstly, endothelial cells adversely affected by chronic hyperglycaemia exhibit a marked reduction in the secretion of nitric oxide 68 , impairing the early infiltration of immune cells into the wound. Secondly, in diabetic wounds, there is a persistent pro-inflammatory state characterized by elevated levels of inflammatory cytokines and chemokines. These molecules ensue that there is a continuous recruitment of neutrophils to the wound site, preventing their natural resolution and impeding the essential transition of macrophages from their M1-pro-inflammatory state to M2-anti-inflammatory state required for wound resolution 61,69 . Compounding this, diabetic macrophages exhibit impaired phagocytosis of apoptotic neutrophils, 69 resulting in the continued release of inflammatory cytokines by neutrophils and perpetuating a vicious inflammatory cycle. The consequences are severe and include repeated tissue damage, impaired angiogenesis, and delayed recruitment of fibroblasts and keratinocytes, which is essential for the later phases of healing 61 . Furthermore, diabetes-related microvascular complications impair oxygen delivery and promote ischemic injury. The net result is a self-sustaining loop of prolonged inflammation, tissue destruction and delayed repair, rendering DFUs especially susceptible to chronicity and infection. Proliferative phase of wound healing This phase of wound healing involves processes such as angiogenesis, re-epithelialisation and ECM deposition 51 . Fibroblasts recruited by growth factors such as PDGF and TGF-β, are important drivers of this phase of healing. These cells migrate into the wound bed and become the predominant cell type. Fibroblasts secrete ECM components, including collagens, fibronectin, and glycosaminoglycans, forming a scaffold for repair. Angiogenesis is a key feature of this phase, as new blood vessels form to supply oxygen to the active cells within the wound microenvironment 61 . This process is mediated by VEGF and fibroblast growth factor (FGF) and is vital for sustaining the repair process. At the wound surface and margins, keratinocytes proliferate and migrate, initiating re-epithelialization and restoring the epidermis. Guided by adhesion molecules such as fibronectin and integrins (α2β1, α3β1, α6β4, and αvβ5), and stimulated by growth factors including epidermal growth factor (EGF) and TGF-β, these cells work to cover the wound 51,61 . Fibroblasts, keratinocytes, and macrophages fill the wound bed with granulation tissue, a highly vascular and cellular matrix, preparing the wound for the final stage of wound healing, the remodelling phase. The proliferative phase is significantly disrupted by DM. This phase can be disturbed by a series of complex mechanisms including, cellular senescence, impaired fibroblast activity 10 , defective angiogenesis, delayed keratinocyte migration, and dysfunctional ECM remodelling 11 , all of which contribute to the development of chronic DFUs. Cellular senescence, a state in which cells permanently cease replication but remain metabolically active, has been linked to impaired healing in DFUs 70-72 . The diabetic microenvironment promotes excessive mitochondrial ROS production, causing DNA damage 70 and triggering cellular senescence. Senescent cells produce pro-inflammatory proteins, such as the senescence-associated secretory phenotype (SASP), which perpetuate local inflammation 72 . Key healing processes, such as angiogenesis, ECM remodelling, and re-epithelialization are disrupted by senescent endothelial cells, fibroblasts, and keratinocytes 70,71 . Additionally, diabetic fibroblasts exhibit reduced functionality, with impaired migration into the wound bed and diminished proliferative capacity 10,12,13,73,74 . This hinders the synthesis of critical ECM components, including collagen and fibronectin 75 , resulting in compromised granulation tissue and ECM structure. Concurrently, AGEs modify collagen 76 , making it resistant to degradation and remodelling, further impeding proper tissue repair. Hyperglycaemia also adversely affects endothelial cells by reducing their proliferation and migration 77,78 . Through the downregulation of VEGF receptors on endothelial cells, hyperglycaemia diminishes their responsiveness to VEGF 77,78 . This blunted response of endothelial cells to VEGF in diabetes results in defective angiogenesis 77 , a hallmark of chronic DFUs. Furthermore, the diabetic wound milieu suppresses keratinocyte proliferation and migration 79,80 , while AGEs disrupt integrins and other adhesion molecules 80 , impeding keratinocyte attachment to the ECM and their ability to re-epithelialize the wound surface. The cumulative effect of these disruptions is a wound that cannot progress beyond the proliferative phase, remaining in a persistent state of chronicity with indefinitely delayed healing. Remodelling phase of wound healing The remodelling phase, is the final and most prolonged stage of wound healing 51 , ensuring that the structural integrity of the tissue is restored. The hallmark of this stage is collagen remodelling and ECM maturation 61 . Matrix metalloproteinases (MMPs) are collagenases and gelatinases that degrade excess ECM proteins, while tissue inhibitors of metalloproteinases (TIMPs) act to counterbalance this by inhibiting MMP activity. An equilibrium between these two groups of molecules is necessary for maintaining tissue stability 52 . As the remodelling phase progresses and the metabolic demands of the wound decreases, many of the newly formed capillaries from the earlier proliferative phase undergo regression. The remaining vessels consolidate into a robust vascular network that sustains the newly repaired tissue 51 . Concurrently, macrophages transition to their M2 phenotype, which are mainly associated with healing 81 , and secrete growth factors such as TGF-β to mediate the differentiation of fibroblasts into myofibroblasts 61 . These specialized cells possessing smooth muscle-like properties contract the wound edges via actin-myosin cross-linking, thereby reducing the wound dimensions 52 . The result of this process is the formation of a collagen-rich scar that replaces the granulation tissue. While this scar tissue exhibits reduced vascularity and possesses diminished strength and functionality compared to the original tissue, it fulfils its primary purpose of sealing the wound and providing protection against further injury. The remodelling phase thus plays a critical role in the healing process, not only for completing tissue repair but also for ensuring its longevity. Through precise coordination of cellular mechanisms and biochemical signals, it establishes the foundation for durable healing, mitigating the risk of wound recurrence. In diabetes, this complex process is significantly impaired, rendering wounds susceptible to chronicity and recurrence. At the core of this dysfunction lies impaired fibroblast activity and abnormal collagen synthesis 10,13,75 . In place of the robust, organised collagen necessary for structural integrity, the diabetic wound environment produces disorganized and weak collagen. AGEs, also implicated in the earlier phases of wound healing, further exacerbate the situation by cross-linking with collagen fibres, rendering them brittle and resistant to proper remodelling 76 . As a result, chronic diabetic foot ulcers persist. The dysfunction extends beyond these factors. DFU fibroblasts exhibit increased senescence 10 and myofibroblasts fail in their duties to effectively achieve wound contraction 75 , contributing to prolonged wound duration. Concurrently, the delicate balance between matrix MMPs and their inhibitors (TIMPs) is disrupted 14,78,82-87 . Excessive MMP activity relative to TIMPs leads to unrestrained ECM degradation, hindering the formation of stable, resilient scars. Together, these disruptions result in wounds that fail to heal, scars that lack tensile strength, and a local environment conducive to wound recurrence and infections. Despite considerable advances in our understanding of the complex pathophysiological mechanisms underlying disturbed wound healing in diabetes mellitus, the current management remains challenging, with outcomes often inconsistent and unpredictable. The next section will briefly outline the current state-of-the-art treatment of DFUs. Current treatment modalities: Diabetic foot ulcer management is complex, involving local wound care, infection control, treatment of peripheral vascular disease, and blood glucose optimization, all with one primary goal: limb preservation. The most effective pathway to this goal lies in a multidisciplinary team (MDT) approach. The risk of amputation can be reduced by up to 51% when an MDT approach is employed 88 . The multidisciplinary foot team (MDFT) includes podiatrists, orthotists, diabetologists and diabetes specialist nurses, vascular and orthopaedic surgeons, infectious disease specialists, and dieticians, working in tandem to address the multifaceted challenges of DFU management. In addition to reducing major amputations, this approach has demonstrated improvements in healing times, minor amputation risks and all-cause-mortality rates 89 . This current section summarises the key components in the DFU management. Apart from glycaemic control, the cornerstone of DFU treatment is local wound management, which begins with thorough wound debridement. This involves carefully removing non-viable tissue until healthy, bleeding tissue is exposed. The debridement material is then discarded. The underlying principle of debridement is to clean the wound bed and eliminate necrotic tissue susceptible to infection, thereby allowing wound healing to proceed 90,91 . Following debridement, the wound should be examined for signs of infection. If present, systemic antimicrobials may be required. For localized infections, topical agents, such as silver or iodine-based dressings may be appropriate 90 . A critical step in wound care is selecting the most appropriate dressing. An effective dressing must efficiently manage exudate, protect the surrounding tissue, and prevent maceration. However, comorbidities, patient-specific factors, and the underlying pathophysiology of individual ulcers can complicate dressing selection. In general, dressings that promote moist wound healing are preferred as it optimizes healing 92 . The final consideration is the need for offloading strategies, where appropriate, such as total contact casting (TCC) or therapeutic footwear to mitigate stress on the ulcer 90 . During follow-up appointments, clinicians assess wound response to treatment. The management plan may be re-evaluated and modified if required, and adherence to prescribed footwear should be encouraged. In certain circumstances, transitioning to alternative options such as removable walkers or custom orthotics may enhance compliance with the treatment regimen. A critical juncture in the management of DFUs that are neither infected nor ischaemic is at the 4-week mark. Wounds with a 50% reduction in size on the fourth week of treatment are more likely to heal completely by 12 weeks 93 . However, for wounds that do not show adequate progress, alternative interventions such as maggot debridement 94 , as well as the use of negative pressure wound therapy (NPWT) 95 or hyperbaric oxygen therapy (HBOT) 96 should be considered without delay. In more complex cases, surgical interventions such as wound excision and skin grafting, or the utilization of cellular tissue products, autografts, or free flaps may be employed. As part of the initial assessment of DFU, it is paramount to exclude significant PVD. If vascular insufficiency is identified, timely revascularization procedures 90 , such as angioplasty or bypass surgery, can restore perfusion and facilitate healing. Finally, where all interventions prove ineffective, amputation, either minor or major, may become necessary, particularly in the presence of systemic signs of infection or critical limb ischemia 97 . Despite implementation of this structured, evidence-based approach to management, many DFUs fail to heal. Therefore, identifying the healing potential of DFUs at the earliest stages and determining which patients require an aggressive intervention sooner rather than later could revolutionize outcomes. Unfortunately, accurately predicting the healing outcomes of diabetic foot ulcers (DFUs) remains a significant challenge. Various clinical models are available to assist in the clinical assessment and prediction of DFU outcomes. Current clinical tools used in predicting DFU outcomes include: Ulcer Site, Ischaemia, Neuropathy, Bacteria Infection, Area and Depth (SINBAD), Diabetic Ulcer Severity Score (DUSS), Wound, Ischaemia and Foot Infection (WIFI), the University of Texas Staging System (UT), Wagner Diabetic Foot Ulcer Classification, and Perfusion, Extent, Depth, Infection and Sensation (PEDIS) 20,24,98-101 . These clinical assessment measures use patient specific wound characteristics to provide valuable insights into predicting the outcomes of DFUs. However, they are not robust enough to encompass multiple individual and comorbid issues such as lower limb oedema for instance. When applied across diverse populations, they are subject to confounding factors such as patient variables 98 , observer bias, and subjectivity 102 . In the last half decade, multiple studies have investigated the role of machine learning tools in predicting the outcome of DFUs. This approach shows promise. However, a standardised and validated algorithm and machine model is needed for this approach to be integrated into clinical practice. 103 Clearly, the prediction and treatment of DFUs remains a challenge (Figure 2). The future of DFU care lies in personalized prediction models with subsequent treatment targeting the specific pathogenic mechanisms underlying each patient’s chronic wound. Analysing cellular and molecular biomarkers directly obtained from the patient’s ulcer could pave the way for tailored and more effective management strategies. In this era of precision medicine, histopathological scoring systems utilizing patient-specific histological and molecular parameters may provide a more objective and individualized means of predicting outcomes. This paper synthesizes data from the rapidly advancing field of DFU management, examining emerging predictive models based on molecular and cellular biomarkers. We also explore biomarkers that influence the healing process, either by promoting or hindering repair, and their potential roles in shaping future predictive models and therapeutics. Although recent biotherapies targeting biomarkers such as MMPs, TIMPs, growth factors, and cellular components show significant promise, these approaches remain broad, failing to address the unique pathophysiological causes of individual ulcers. Looking ahead, the promise of personalized care guided by precise, patient-specific molecular data, offers the opportunity to refine therapeutic approaches for optimal outcomes. Biomarkers with Potential to Predict the Outcome of Diabetic Foot Ulcers There has been growing interest over the last two decades in identifying molecules that may help predict the trajectory of ulcers in individuals with DFU. Various biomaterials have been analysed for these markers, including serum, non-ulcerated skin biopsy, wound fluid, and wound biopsy samples. (Table. 1). A recent systematic review suggested that wound exudates may contain a vast amount of information that could help determine the healing potential of an ulcer 104 . In this section, we summarize studies that have examined biomarkers in various biological material, such as serum, wound exudate, wound biopsies, and debridement tissue and their predictive power or lack thereof (Figure 3). The majority of studies have been primarily focussed on matrix metalloproteases and their tissue inhibitors, growth factors, as well as cyto- and chemokines. In the next sections, we will highlight them in more detail. The Role of Matrix Metalloproteinases (MMPs) and Tissue Inhibitors of Metalloproteinases (TIMPs) in Predicting the Outcome of Diabetic Foot Ulcers MMPs and TIMPs are critical in ECM breakdown and remodelling during wound healing. The roles of individual MMPs in DFU healing have been comprehensively reviewed 105 . The key MMPs of interest in DFUs are MMP-1, -2, -8, and -9. MMP-2 and -9, secreted by neutrophils, fibroblasts, macrophages, keratinocytes, and endothelial cells, are essential during the inflammatory phase of wound healing, where they degrade ECM proteins to facilitate cell migration, tissue growth, and angiogenesis. MMP-1, secreted by fibroblasts and keratinocytes, and MMP-8, produced by neutrophils, are crucial in the proliferative phase, breaking down ECM proteins to enhance cell migration 105 . Imbalance in these proteases and their inhibitors 85,106,107 is associated with an excessive and prolonged proteolytic environment in the healing DFU leading to degradation of growth factors and the ECM and slowing wound healing. There is a growing body of evidence to support the utilization of these proteins in predictive models for DFU outcomes. The predictive values of MMPs and TIMPs examined in the serum, wound exudate, and DFU tissue samples at baseline and following treatment have been explored in recent studies. Elevated baseline levels of MMP-8 in wound exudate 84 and MMP-9 in both wound exudate 86,108 and serum 85 have been documented in poorly healing DFUs. Conversely, one study demonstrated no significant difference in MMP-9 levels in DFU exudates among subjects with varying ulcer outcomes 84 . Moreover, levels of MMP-8 and -9 have been observed to decrease over time in healing ulcers while remaining elevated in non-healing ulcers 84,85 . In fact, declining levels of MMP-9 in wound tissue have been positively associated with DFU healing in patients treated with total contact cast 87 . In contrast, elevated baseline MMP-1 levels correlate with better DFU healing in most studies 84,86 . The ratio of MMP-1/TIMP-1 obtained from wound exudates has been shown to predict ulcer outcomes, with a higher ratio at the initial visit predicting ulcer healing in 12 weeks 84-86 . Using a cutoff value of 0.39, Muller et al., found that this ratio in wound exudate predicted outcomes with a sensitivity of 71% and specificity of 87.5% 84 . This suggests that the MMP-1/TIMP-1 ratio at the initial visit can reliably and objectively predict DFU healing, regardless of patient and wound characteristics. Additionally, Luanraksa et al. emphasized the importance of the MMP/TIMP ratio in wound fluid, showing that individuals with favourable healing outcomes had higher MMP-1 levels and an MMP-1/TIMP-1 ratio at baseline that was 25 times higher than in those that had less favourable outcomes. In contrast, those with poor healing outcomes had higher baseline MMP-9 levels and an elevated MMP-9/TIMP-1 ratio, with the baseline ratio being 117 times higher than in those with favourable outcomes. Applying cutoff levels of 0.38 pg/µg protein for MMP-9, 0.056 for the MMP-1/TIMP-1 ratio, and 9.06 for the MMP-9/TIMP-1 ratio predicted wound healing with sensitivities of 81.8%, 81.8%, and 90.9%, and specificities of 64.6%, 55%, and 64.6%, respectively 86 . Other studies have looked at MMPs and TIMPs in serum of patients with DFUs and found elevated MMP-9 and MMP-9/TIMP-1 ratio in patients with ulcers that subsequently failed to heal. Serum MMP-9/TIMP-1 was able to predict better healing trajectory with a sensitivity and specificity of 63.6% and 58.6%, respectively, using a cutoff of <0.395 for the MMP-9/TIMP-1 ratio 85 . Additionally, studies have demonstrated the utility of pro-MMPs, the inactive form of MMPs, in predicting ulcer outcomes 107 . Liu et al., found higher pro-MMP-9/TIMP-1 and pro-MMP-2/TIMP-1 ratio in wound fluid correlated with poor ulcer outcomes and were able to predict healing with an 87% sensitivity and 91% specificity, accurately predicting healing outcomes in 94% of cases 107 . In addition to matrix metalloproteases, their inhibitors have also demonstrated predictive potential. The two protease inhibitors commonly investigated in DFU pathogenesis are TIMP-1 and TIMP-2. The majority of studies have not observed a significant difference in baseline serum or wound fluid levels of TIMP-1 and -2 between different ulcer outcomes. 84-86 . However, Liu et al. reported lower baseline wound fluid TIMP-1 levels in subjects with poor DFU outcomes 107 . Beyond baseline levels, the longitudinal evolution of MMPs and TIMPs over a follow-up period has been investigated to predict healing trajectories. Jindatanmanusan et al. monitored MMP-9 in wound fluid of DFUs over a 12-week period and observed that in patients with favourable healing outcomes, MMP-9 remained at low levels throughout the follow-up period compared to those with poor healing outcomes, in whom it remained persistently elevated 108 . Similarly, temporal monitoring of MMP-9 levels in wound fluid revealed a peak at week 8 in non-healing ulcers. In contrast, in healing ulcers, MMP-9 levels declined towards week 8 84,86 . Additionally, a reduction in MMP-1 was observed at week 8 86 . Comparable trends of declining MMP-9 levels and MMP-9/TIMP-1 ratio at the 4-week mark were also noted in the serum of subjects with better healing 85 . Thus, elevated MMP-9 levels measured in serum or wound fluid at later time points in the natural history of DFUs retains relevance as a predictor of the healing potential of DFU. This is particularly relevant in clinical practice, where subjects presenting with DFUs commonly do so later in the disease course. Several groups have analysed the dynamic changes in MMP and TIMP in the wound tissue following conventional DFU therapies such as total contact cast (TCC) 87 and negative pressure wound therapy (NPWT) 83 . Reduced MMP-1 83 , MMP-2 87 , and MMP-9 levels 83,87 , and increased TIMP-1 83,87 and TIMP-2 87 have been associated with good healing following treatment with TCC and NPWT. These studies support the idea that TIMP-1 upregulation and MMP-9, -2 downregulation are beneficial for DFU healing and these proteins may serve as markers to identify ulcers unlikely to heal and need a more aggressive treatment approach earlier. Furthermore, experimental treatment protocols have shown dynamic changes in MMPs that correlate with DFU healing outcomes. Ulrich et al. examined MMP-2 levels in wound fluid following treatment with an oxidized regenerated cellulose/collagen matrix and found that MMP-2 levels significantly declined over time, correlating with wound size reduction during the follow up period 109 . Additionally, Yao et al. investigated the impact of non-contact low-frequency ultrasound on chronic, non-healing DFU and found significant reduction in MMP-9 and pro-inflammatory cytokines in wound fluid, which positively correlated with wound area reduction 110 . These findings suggest that measuring these protein levels at different stages during DFU treatment may provide valuable insight into the ulcer’s trajectory, further supporting their role as potential predictive biomarkers. To summarise, MMPs and TIMPs can be utilised to personalise DFU outcome predictions. However, most of the available studies are limited by small sample sizes and population selection, often excluding subjects with soft tissue infections, osteomyelitis or peripheral vascular disease. Moreover, differences in how DFU healing is defined, as well as variations in ulcer duration prior to MMP/TIMP level analysis between studies, may account for discrepancies in the results of these studies. Nonetheless, most studies are in general agreement that MMP-1, MMP-9 and their ratios with TIMP-1 hold high predictive value. To encourage the clinical utility of these biomarkers, future research should involve larger sample sizes, more standardized ulcer outcome measures, and broader inclusion criteria to better reflect the diverse DFU patient population. The Role of Growth Factors in Predicting the Outcome of Diabetic Foot Ulcers Various growth factors have been investigated for their role in the wound microenvironment and in the pathophysiology of DFUs. Growth factors can be thought of as messengers, released by key cells that infiltrate the skin after wounding, including degranulated platelets and infiltrating immune cells. Growth factors are crucial in the proliferative phase of wound healing. Their receptors are expressed on fibroblasts, endothelial cells and keratinocytes, allowing them to play critical roles in ECM deposition, angiogenesis and re-epithelialization. Growth factors are mitogenic, triggering proliferation and also migration of these key cells involved in the proliferative phase of wound healing 111 . Therapeutic interventions that modulate growth factor levels further support their predictive value. Various studies have demonstrated improved wound healing outcomes following the introduction of topical or systemic growth factors 112-114 . This suggests that the presence or absence of growth factors in the DFU microenvironment may be a valuable prognostic tool. To support this theory, several research groups have explored the levels or expression patterns of growth factors such as VEGF, PDGF, TGF-β, epidermal growth factor (EGF), and granulocyte macrophage colony stimulating factor (GM-CSF) in diabetic wounds or the serums of subjects with DFUs and correlated their findings with healing outcomes. By assessing their baseline levels, dynamic expression, and pre- and post-treatment changes, studies have investigated their potential utility as predictive biomarkers. VEGF is one of the most extensively studied growth factors in DFU outcome prediction models. Elevated VEGF levels in the serum of patients with DFUs has been associated with favourable healing outcomes in one study 115 . Conversely, a large study by Xu et al., involving 502 subjects with DFUs found that high baseline levels of serum VEGF correlated negatively with wound healing, with a sensitivity of 89.5% and a specificity of 57.3% 116 . However, this discrepancy may be explained by population differences: the former study excluded subjects with PVD, 115 whereas the latter included this cohort and further demonstrated a statistically significant lower ABI in non-healers 116 . High circulating VEGF is positively correlated with PVD, 117 which itself is a poor prognostic indicator in DFUs 100,118 . This may account for the conflicting results between these two studies 115,116 . In wound tissue, high baseline VEGF level and VEGF expression has been correlated positively with DFU healing with a sensitivity and specificity of 93% and 65%, respectively for VEGF level and 89.1% and 74.3%, respectively for VEGF expression 116 . Changes in VEGF level or expression have been demonstrated following various treatment modalities. Increased VEGF in wound fluid 119 , wound biopsy samples 83 and both 120 has been shown to correlate positively with wound healing. Though there are conflicting findings regarding serum levels of VEGF and wound outcomes, most studies agree that high VEGF in the DFU wound material is associated with improved healing outcomes. Therefore, we propose that VEGF, when analysed in wound material, hold significant promise as a clinical biomarker for predicting DFU healing. Other growth factors, including PDGF, have also been investigated for their role in the pathophysiology of DFUs. The expression of PDGF and its receptors is significantly downregulated in keratinocytes and endothelial cells at the edge of human diabetic wounds 77 . Similarly, animal models of wound healing support this finding, demonstrating reduced expression of PDGF-A and its receptor in diabetic mouse skin and wound tissue 121 . Although these two studies did not compare molecule levels against ulcer outcomes, poor expression of these growth factors or their receptors in DFU tissue could potentially be used to predict DFU, as suggested by a few studies. Using serum analysis of biomarkers in 30 patients with DFUs, Dinh et al, noted raised serum levels of PDGF-AA, an isoform of PDGF, in subjects who failed to heal compared to those that did 122 . Additionally, the evaluation of PDGF following treatment has shown promise in predicting outcomes. In a randomized controlled trial of experimental treatment with Oxygen-Ozone therapy, Zhang et. al demonstrated a dynamic increase in PDGF levels in wound fluid and biopsy samples of DFUs that healed compared to those that did not. 120 We extrapolate from these findings that PDGF may serve as a potential biomarker for predicting if an ulcer is likely to heal. A potential explanation for this apparent discrepancy between these studies is that the study by Dinh et. al 122 focussed on a single PDGF isoform (PDGF-AA) whereas the study by Zhang et. Al 120 assessed the entire PDGF family. This underscores the possibility that individual isoforms of PDGF may have distinct roles in the context of the wounding healing process. Furthermore, an in-silico analysis model of wound healing has been used to explore biomarkers such as PDGF and TGF-β1, revealing that reduced TGF- β1 expression in DFUs is associated with poor healing. Interestingly, in scenarios of low TGF- β1 or high TNF (a pro-inflammatory cytokine) levels in the wound model, simulated wound debridement did not significantly improve healing outcomes 123 . This is an important finding, given that debridement is an essential component of conventional DFU treatment 90,91 . This finding suggests that variations in growth factor levels in the wound milieu may help explain why some DFUs fail to heal even following high quality debridement. It is therefore plausible that these subjects may have low expression of growth factors or high expression of proinflammatory cytokines in their ulcers. Clinical studies are needed to corroborate this finding. Similarly, clinical studies have demonstrated that reduced TGF-β1 in wound fluid is associated with poor DFU healing, compared to ulcers that successfully healed 107 . Additionally, in randomized controlled trials, increased TGF-β1 in wound fluids 120 and biopsy 83,120 following treatment have been correlated with healing ulcers. When TGF- β1 in wound fluid is used in combination with the pro-MMP-9/TIMP-1 ratio, ulcer outcomes could be reliably predicted in 94% of cases 107 . Although it appears unlikely that TGF- β1 can been used in isolation as a DFU prediction model, there is enough evidence that suggests that TGF-β1 in wounds, when combined with other molecular markers, may contribute significantly to a predictive model for DFU healing. Less studied but relevant biomarkers, including EGF and GM-CSF, have been shown to have increased expression in keratinocytes and endothelial cells at the wound margins of DFUs, though their receptors were not upregulated 77 . This suggests a possible signalling pathway impairment, which may be a contributory factor to poor ulcer outcomes. In interventional studies using animal models, GM-CSF has been linked to enhanced wound healing through the regulation of monocyte chemotactic factor 1 (MCP-1), facilitating angiogenesis and the recruitment of neutrophils and macrophages to the wound site. 124 . While evidence remains limited regarding the use of these biomarkers in predicting DFU outcomes, they show promise for inclusion in future predictive models. In summary, there is a growing interest and expanding knowledge surrounding the use of growth factors to aid prediction of DFU healing outcomes. However, further research is necessary to support the use of these growth factors in isolation for this purpose. Future studies should focus on integrating multiple biomarkers into predictive models to enhance their accuracy and clinical utility. The Role of Chemokines and Cytokines in Predicting the Outcome of Diabetic Foot Ulcers Chemokines and cytokines are proteins essential in all the phases of wound healing. They are key in coordinating immune responses, modulating inflammation, and facilitating tissue repair. These proteins can be broadly classified as either pro-inflammatory or anti-inflammatory. As such, their dysregulation in the DFU microenvironment could cause either an impaired immune response or a state of sustained inflammation, both of which can contribute to a non-healing DFU. Several studies have investigated the levels of chemokines and cytokines across different biological samples, including serum, wound exudate, and DFU tissue, to evaluate their potential as accurate predictive biomarkers for the clinical course of DFUs. A few studies have demonstrated the downregulation of key chemotactic factors in non-healing ulcers. ENA-78 (epithelial neutrophil activator-78) 125 , MCP-2 (macrophage chemotactic protein-2) 125 , and CXCL-6 (granulocyte chemoattractant protein-2, GCP-2) 126 may be able to predict DFU outcomes. Li et al, analysed ENA-78, a potent neutrophil chemotactic factor, in the wound fluid and plasma of subjects with DFUs and found lower levels in those with non-healing ulcers compared with those whose ulcers healed. ENA-78 in wound fluid, but not plasma, predicted wound outcome, with a sensitivity and specificity of 45.9% and 89.58%, respectively 125 . In the same study, MCP-2, a chemotactic protein that recruits cells of the innate immune system, was lower in the plasma and wound fluid of non-healing ulcers. However, it did not reach statistical significance in accurately differentiating ulcer outcomes in this study 125 . CXCL-6, a potent chemoattractant for neutrophils, was investigated in wound fluids as a potential predictive biomarker. Higher levels were observed in healing ulcers. CXCL-6 as a biomarker predicted DFU outcomes with a sensitivity and specificity of 87.27% and 95.56%, respectively 126 . Nonetheless, due to its association with other chronic inflammatory conditions, there is a potential risk for false positives. To mitigate this pitfall, a model that combines CXCL-6 with other prospective biomarkers should be evaluated. Despite this limitation, a recent scoping review examining biomarkers across 14 studies (10 human and 4 animal), including CXCL-6, ENA-78, SERPINB3, neutrophil elastase, citrulline histone H3 (CiTH3), soluble intercellular adhesion molecule-1 (sICAM-1), endothelin-1 (ET-1) and MMP-9/TIMP-1 ratio, identified CXCL-6 as the biomarker with the most predictive accuracy 127 . Increased levels of pro-inflammatory cytokines in serum or wound fluid have consistently been demonstrated to predict poor healing ulcers. Levels of tumour necrosis factor-alpha (TNF-α), Interferon-gamma (IFN-γ), interleukin-1 beta (IL-1β), interleukin-6 (IL-6) and interleukin-8 (IL-8) have been studied widely as potential predictive markers of DFU healing. In-silico analysis have implicated raised TNF-α levels in delayed DFU healing 123 . Some clinical studies have corroborated this finding, showing that baseline TNF-α levels are elevated in the serum of subjects with DFU that failed to heal 115,122 . Moreover, when levels are tracked during a follow up period of 12 weeks 115 or post-therapeutic interventions 83,110 , TNF-α levels followed a downward trajectory in ulcers that healed. Similarly, declining levels of IL-1β in wound biopsy 83 and fluid 110 , as well as reductions in IL-6 and IL-8 levels in wound fluid 110 , have been noted in DFUs with favourable outcomes. Similarly, significantly higher serum levels of IL-6 have been noted in DFUs with poor prognosis with a modest predictive performance of 0.669 AUC, and a sensitivity and specificity of 56% and 76% respectively 128 . However, the criteria used to define healing in this study is not comparable with most other studies 128 . Good prognosis was defined as ulcers with complete or near complete re-epithelialisation whilst most other studies defined healed as complete re-epithelialisation. Nonetheless, IL-6 remains a marker of poor DFU prognosis across the studies that investigated this cytokine. Less commonly studied cytokines show promise. For instance, IL-34, a pro-inflammatory cytokine involved in the differentiation of macrophages, has been found to be downregulated in non-healing ulcers based on both serum and wound biopsy data 129 . Likewise, SERPINB3, a serine protease inhibitor studied in DFU biopsies, has a sensitivity of 75% and a specificity of 62.5% in predicting ulcers likely to heal 130 . Other Biomarkers that Predict the Outcome of Diabetic Foot Ulcers More recently, serum level of Gremlin-1, a proangiogenic protein and a BMP (bone morphogenic protein) antagonist, have been demonstrated to be low in correlation with worsening Wagner DFU grade and in subjects who underwent lower extremity amputation, serving as a marker of poor prognosis in DFUs. A low Gremlin-1, with a cutoff of 2.47 ng/mL, was able to predict amputation with a sensitivity of 67% and specificity 46% 131 . However, this study did not stipulate a defined follow-up period. Besides, Gremlin-1 levels tended to be lower in subjects with PVD 131 , a cohort that are predictably more likely to have a negative outcome of their DFU. The role of Gremlin-1 is particularly interesting as it poses a conundrum in the pathogenesis of DFU. Gremlin-1 inhibits BMP-mediated repair 132 while also stimulating angiogenesis through VEGFR-2 133 creating a complex dynamic in DFU pathogenesis. Other biomarkers that have been studied in the outcome prediction of DFU include: transcription factors such as c-Myc, a transcription factor that controls cell growth and proliferation which, in a dysregulated state is negatively associated with wound healing 134 ; phosphorylated glucocorticoid receptor (p-GR) the activated form of the glucocorticoid receptor, which is associated with impaired keratinocyte migration and subsequent impaired wound closure 135 ; intracellular adhesion molecule-1 (ICAM-1), a pro-inflammatory molecule expressed on endothelial and immune cells; soluble suppression of tumorigenicity 2 (sST2), an immune receptor that reflects inflammation. A recent study investigated the performance of c-Myc and p-GR when used as biomarkers in combination with wound duration and size to predict healing outcomes of DFU in a 12-week follow up period. 136 This research analysed the nuclear presence of c-Myc and p-GR in a mixture of biopsy and debrided tissue in 107 subjects and discovered that independently, there were no statistically significant differences in baseline or week 4 levels of these markers in healed and non-healed wounds. However, when these markers were individually assessed in combination with a base model of wound duration and size, they had a modest predictive power with an AUC of 0.637 and 0.643 for c-Myc/base model and p-GR/base model respectively 136 . Similarly, in an observational study of 210 subjects with DFU and a follow up period of 6 months, higher ICAM-1 and sST2 levels were demonstrated in the serum of subjects with DFU showed good promise in their ability to predict DFU prognosis with a diagnostic performance of 0.69 and 0.682 AUC respectively, and sensitivities of 68.25% and 36.51% respectively and specificities of 67% and 91% respectively 128 , showing their ability to classify wounds as non-healing in majority of cases. When these biomarkers are combined with IL-6 in a nomogram their collective predictive performance improves with an AUC of 0.786, and a sensitivity and specificity of 68% and 83% respectively. Though the outcome measure of good and poor prognosis was ambiguous in this study with wounds classified as good prognosis if they completely or incompletely re-epithelialise, these markers show potential and should be studied further with clearer outcome measures. Whereas most of the work has focussed on serum and wound fluid or biopsies, very few studies have utilised wound debridement alone or biological material from wound dressings. Two studies by the same group investigated debrided material for gene expression of key proteins including growth factors, chemokines, cytokines, and tissue inhibitor of metalloproteinases, including M1-proinflammatory macrophages (CCR7, VEGF, CD80 and IL1-B) and M2-anti-inflammatory macrophages (PDGFB, TIMP-3 and MRC1) 137 138 . Individually, the expression levels of these genes did not differ between healing and non-healing ulcers over a follow up period. Thus, they developed a scoring system, namely, the M1/M2 score, based on a ratio between these M1 and M2 associated genes, representing the transition from M1 to M2 macrophage in the wound microenvironment and signifying progression towards healing. The group observed that this score increased over a follow up period in DFUs that did not heal and decreased in those that healed. Similarly, wound dressings have been studied in their ability to preserve molecules that are able to predict DFU outcome. In a study of 16 patients with mixed DFU and venous leg ulcers (VLU), wound dressings were analysed weekly for a 4-week follow up period for their potential to store biological material, and if these materials can be used to derive biomarkers that predict healing trajectory defined as greater or equal to 50% wound size reduction in four weeks. Using proteomics, the study identified diverse proteins. Increased levels of S100A7 and S100A11 were associated with non-healing wounds, whereas, healing wounds were associated with increasing levels of KRT6A (keratin-6A, a protein expressed in activated keratinocytes and involved in re-epithelialisation), SERPING1 (a protease inhibitor that regulates excessive tissue damage), haptoglobin, CATG (Cathepsin G, a proinflammatory protease) and ALDOA (aldolase A, a glycolytic enzyme involved in energy production) 139 . Though this study was a significantly small sample size (8 DFU) and a mixture of both DFU and VLU, it remains a pivotal study in research into biomarker discovery with emphasis on harvesting biomarkers from clinical waste products. These studies add to the growing body of knowledge on how biomarkers can be easily derived from routinely discarded materials. Additionally, the studies also highlight that patient specific biological materials can be easily accessible for use in further research into DFU outcome predictions. Eventually, should biomarkers be incorporated into clinical practice, clinicians may be able to avoid rigorous sampling as these discarded materials improve accessibility to patient wound biodata. Mechanistic molecules that are candidate biomarkers for future studies on DFU outcome prediction Whilst most described studies have focussed on identifying biomarkers that are able to discriminate between healed and non-healed ulcers, several studies have looked at dynamic changes in biomarkers following conventional or experimental treatment or the relationship of molecules with DFU diagnosis or severity. In a prospective study of 195 subjects, serum levels of NRF2, an antioxidant transcription factor with a positive role in wound healing 140 , was measured using ELISA. Ulcer severity classification (mild vs severe DFU) was the basis of comparison in this study. Patients with severe DFU had significantly lower serum NRF2 levels on initial presentation. In its utility as a biomarker for identifying severe ulcers, NRF2 performed well statistically, with an AUC of 0.746, a cutoff value of 185.65 pg/ml, sensitivity of 71.3%, and specificity of 63.0% 141 . Though this study did not necessarily compare healed vs non-healed wounds, NRF2 could be studied further in its ability to predict healing outcomes. 141 Similarly, in a randomised controlled trial using human mesenchymal stem cell gel in the treatment arm vs placebo, DFUs categorised as healing based on wound mass reduction in 7 days showed a statistically significant dynamic increase in CD163 mRNA (an M2 macrophage marker), and decrease in NF-κB p50 mRNA (a proinflammatory transcription factor), compared to placebo in wound biopsy specimens on day 7 of treatment 142 . Although this study does not directly evaluate the predictive power of CD163 and NF-κB p50, these molecules are mechanistically linked to healing since they are modulated by treatment only in patients who showed signs of healing. Thus, they are considered candidate biomarkers of interest for future predictive studies. . In summary, several notable biomarkers have the potential to be incorporated into a predictive model at a patient’s initial clinic visit, facilitating a personalised approach to predicting DFU outcomes. However, variability in inclusion criteria, such as some studies excluding infected or vascular ulcers, along with differences in ulcer duration prior to patient recruitment, follow-up periods, and criteria for wound healing makes it challenging to determine which biomarker is most likely to produce replicable results across different populations. Thus, further studies are required before definitive conclusions can be drawn regarding the predictive power of these biomarkers and their utility in clinical practice. Conclusion and outlook Based on the considerations discussed in the previous sections, it is evident that a significant body of research is attempting to predict the fate of DFU using a personalised patient-centred approach. Numerous biomarkers have been studied in different biological samples including DFU wound biopsies, debridement material from DFUs, wound exudates, biopsies taken from non-ulcerated skin, and serum samples. Despite the promising findings from research, the clinical adoption of biomarkers as a tool to aid prediction of the outcome of DFUs remains limited. At present, no biomarker or panel of biomarkers has been universally accepted or integrated into routine clinical practice for the management of DFUs. While the use of biomarkers as a predictive tool is not yet an established or standard practice, the direction of current research is encouraging. This opens avenues for future research exploring additional markers involved in the complex physiology of wound healing. These novel markers should not only have a high sensitivity and specificity in predicting ulcer outcomes but should be easily and safely obtained from patients as well as accessible to clinicians to encourage adoptability in clinical practice. Such discoveries, when implemented in diabetes podiatry clinics, would considerably improve the treatment of DFU, restructuring care towards a more personalised patient-centred approach with the ultimate goal of improving healing rates and patient outcome. Acknowledgement Julie Okiro is supported by the StAR MD Programme of the Royal College of Surgeons in Ireland in collaboration with the Hermitage Hospital, Dublin. Conflict of Interest The authors have no conflict of interest. References 1 Abbott, C. A. et al. The North-West Diabetes Foot Care Study: incidence of, and risk factors for, new diabetic foot ulceration in a community-based patient cohort. Diabet Med 19 , 377-384, doi:10.1046/j.1464-5491.2002.00698.x (2002).2 Pop-Busui, R. et al. Diabetic Neuropathy: A Position Statement by the American Diabetes Association. Diabetes Care 40 , 136-154, doi:10.2337/dc16-2042 (2017).3 Boulton, A. J. The pathway to foot ulceration in diabetes. Med Clin North Am 97 , 775-790, doi:10.1016/j.mcna.2013.03.007 (2013).4 Monteiro-Soares, M., Boyko, E. J., Ribeiro, J., Ribeiro, I. & Dinis-Ribeiro, M. Predictive factors for diabetic foot ulceration: a systematic review. Diabetes Metab Res Rev 28 , 574-600, doi:10.1002/dmrr.2319 (2012).5 Peters, E. J., Armstrong, D. G. & Lavery, L. A. Risk factors for recurrent diabetic foot ulcers: site matters. Diabetes Care 30 , 2077-2079, doi:10.2337/dc07-0445 (2007).6 Weigelt, C. et al. Immune mediators in patients with acute diabetic foot syndrome. Diabetes Care 32 , 1491-1496, doi:10.2337/dc08-2318 (2009).7 Tuttolomondo, A. et al. Adiponectin, resistin and IL-6 plasma levels in subjects with diabetic foot and possible correlations with clinical variables and cardiovascular co-morbidity. Cardiovasc Diabetol 9 , 50, doi:10.1186/1475-2840-9-50 (2010).8 Yu, X. T. et al. Tandem mass tag-based serum proteomic profiling revealed diabetic foot ulcer pathogenesis and potential therapeutic targets. Bioengineered 13 , 3171-3182, doi:10.1080/21655979.2022.2027173 (2022).9 Esposito, K. et al. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation 106 , 2067-2072, doi:10.1161/01.cir.0000034509.14906.ae (2002).10 Liang, L. et al. Integrative analysis of miRNA and mRNA paired expression profiling of primary fibroblast derived from diabetic foot ulcers reveals multiple impaired cellular functions. Wound Repair Regen 24 , 943-953, doi:10.1111/wrr.12470 (2016).11 Maione, A. G. et al. Three-dimensional human tissue models that incorporate diabetic foot ulcer-derived fibroblasts mimic in vivo features of chronic wounds. Tissue Eng Part C Methods 21 , 499-508, doi:10.1089/ten.TEC.2014.0414 (2015).12 Kashpur, O. et al. Differentiation of diabetic foot ulcer-derived induced pluripotent stem cells reveals distinct cellular and tissue phenotypes. Faseb j 33 , 1262-1277, doi:10.1096/fj.201801059 (2019).13 Lerman, O. Z., Galiano, R. D., Armour, M., Levine, J. P. & Gurtner, G. C. Cellular dysfunction in the diabetic fibroblast: impairment in migration, vascular endothelial growth factor production, and response to hypoxia. Am J Pathol 162 , 303-312, doi:10.1016/s0002-9440(10)63821-7 (2003).14 Chen, R. & Zou, L. Combined analysis of single-cell sequencing and bulk transcriptome sequencing reveals new mechanisms for non-healing diabetic foot ulcers. PLoS One 19 , e0306248, doi:10.1371/journal.pone.0306248 (2024).15 Theocharidis, G. et al. Single cell transcriptomic landscape of diabetic foot ulcers. Nat Commun 13 , 181, doi:10.1038/s41467-021-27801-8 (2022).16 Oyibo, S. O. et al. A comparison of two diabetic foot ulcer classification systems: the Wagner and the University of Texas wound classification systems. Diabetes Care 24 , 84-88, doi:10.2337/diacare.24.1.84 (2001).17 Gershater, M. A. et al. Complexity of factors related to outcome of neuropathic and neuroischaemic/ischaemic diabetic foot ulcers: a cohort study. Diabetologia 52 , 398-407, doi:10.1007/s00125-008-1226-2 (2009).18 Hurley, L. et al. A prospective study of risk factors for foot ulceration: the West of Ireland Diabetes Foot Study. Qjm 106 , 1103-1110, doi:10.1093/qjmed/hct182 (2013).19 Meloni, M. et al. Impact of heart failure and dialysis in the prognosis of diabetic patients with ischemic foot ulcers. J Clin Transl Endocrinol 11 , 31-35, doi:10.1016/j.jcte.2018.01.002 (2018).20 Prompers, L. et al. Prediction of outcome in individuals with diabetic foot ulcers: focus on the differences between individuals with and without peripheral arterial disease. The EURODIALE Study. Diabetologia 51 , 747-755, doi:10.1007/s00125-008-0940-0 (2008).21 Bonnet, J. B., Szwarc, I., Avignon, A., Jugant, S. & Sultan, A. Renal function is highly associated with podiatric risk in diabetic patients. Clin Kidney J 16 , 2156-2163, doi:10.1093/ckj/sfad106 (2023).22 Jalilian, M., Ahmadi Sarbarzeh, P. & Oubari, S. Factors Related to Severity of Diabetic Foot Ulcer: A Systematic Review. Diabetes Metab Syndr Obes 13 , 1835-1842, doi:10.2147/dmso.S256243 (2020).23 Coffey, L., Mahon, C. & Gallagher, P. Perceptions and experiences of diabetic foot ulceration and foot care in people with diabetes: A qualitative meta-synthesis. Int Wound J 16 , 183-210, doi:10.1111/iwj.13010 (2019).24 Beckert, S., Witte, M., Wicke, C., Königsrainer, A. & Coerper, S. A new wound-based severity score for diabetic foot ulcers: A prospective analysis of 1,000 patients. Diabetes Care 29 , 988-992, doi:10.2337/diacare.295988 (2006).25 Shah, P. et al. Wagner’s Classification as a Tool for Treating Diabetic Foot Ulcers: Our Observations at a Suburban Teaching Hospital. Cureus 14 , e21501, doi:10.7759/cureus.21501 (2022).26 Browne, S., Petit, N. & Quondamatteo, F. Functionalised biomaterials as synthetic extracellular matrices to promote vascularisation and healing of diabetic wounds. Cell Tissue Res 395 , 133-145, doi:10.1007/s00441-023-03849-4 (2024).27 Hosty, L., Heatherington, T., Quondamatteo, F. & Browne, S. Extracellular matrix-inspired biomaterials for wound healing. Mol Biol Rep 51 , 830, doi:10.1007/s11033-024-09750-9 (2024).28 Hendricks, L. E. & Hendricks, R. T. Greatest fears of type 1 and type 2 patients about having diabetes: implications for diabetes educators. Diabetes Educ 24 , 168-173, doi:10.1177/014572179802400206 (1998).29 Wukich, D. K. et al. Amputation and infection are the greatest fears in patients with diabetes foot complications. J Diabetes Complications 36 , 108222, doi:10.1016/j.jdiacomp.2022.108222 (2022).30 Zhang, Y. et al. Global Disability Burdens of Diabetes-Related Lower-Extremity Complications in 1990 and 2016. Diabetes Care 43 , 964-974, doi:10.2337/dc19-1614 (2020).31 Zhang, P. et al. Global epidemiology of diabetic foot ulceration: a systematic review and meta-analysis (†). Ann Med 49 , 106-116, doi:10.1080/07853890.2016.1231932 (2017).32 Magliano, D. J., Boyko, E. J. & committee, I. D. F. D. A. t. e. s. in Idf diabetes atlas (International Diabetes Federation © International Diabetes Federation, 2021., 2021).33 Jodheea-Jutton, A., Hindocha, S. & Bhaw-Luximon, A. Health economics of diabetic foot ulcer and recent trends to accelerate treatment. Foot (Edinb) 52 , 101909, doi:10.1016/j.foot.2022.101909 (2022).34 Waibel, F. W. A., Uçkay, I., Soldevila-Boixader, L., Sydler, C. & Gariani, K. Current knowledge of morbidities and direct costs related to diabetic foot disorders: a literature review. Front Endocrinol (Lausanne) 14 , 1323315, doi:10.3389/fendo.2023.1323315 (2023).35 Lazzarini, P. A., Raspovic, K. M., Meloni, M. & van Netten, J. J. A new declaration for feet’s sake: Halving the global diabetic foot disease burden from 2% to 1% with next generation care. Diabetes Metab Res Rev 40 , e3747, doi:10.1002/dmrr.3747 (2024).36 Hashempour, R. et al. Economic burden of diabetic foot ulcer: a case of Iran. BMC Health Serv Res 24 , 363, doi:10.1186/s12913-024-10873-9 (2024).37 Zhang, Y. et al. Cost-effectiveness of guideline-based care provision for patients with diabetes-related foot ulcers: A modelled analysis using discrete event simulation. Diabet Med 40 , e14961, doi:10.1111/dme.14961 (2023).38 Lo, Z. J., Surendra, N. K., Saxena, A. & Car, J. Clinical and economic burden of diabetic foot ulcers: A 5-year longitudinal multi-ethnic cohort study from the tropics. Int Wound J 18 , 375-386, doi:10.1111/iwj.13540 (2021).39 Syed, M. H. et al. The economic burden of inpatient diabetic foot ulcers in Toronto, Canada. Vascular 28 , 520-529, doi:10.1177/1708538120923420 (2020).40 Hicks, C. W. et al. Contribution of 30-day readmissions to the increasing costs of care for the diabetic foot. J Vasc Surg 70 , 1263-1270, doi:10.1016/j.jvs.2018.12.028 (2019).41 Toscano, C. M. et al. Annual Direct Medical Costs of Diabetic Foot Disease in Brazil: A Cost of Illness Study. Int J Environ Res Public Health 15 , doi:10.3390/ijerph15010089 (2018).42 Rice, J. B. et al. Burden of diabetic foot ulcers for medicare and private insurers. Diabetes Care 37 , 651-658, doi:10.2337/dc13-2176 (2014).43 Hicks, C. W. et al. Burden of Infected Diabetic Foot Ulcers on Hospital Admissions and Costs. Ann Vasc Surg 33 , 149-158, doi:10.1016/j.avsg.2015.11.025 (2016).44 Kerr, M., Rayman, G. & Jeffcoate, W. J. Cost of diabetic foot disease to the National Health Service in England. Diabet Med 31 , 1498-1504, doi:10.1111/dme.12545 (2014).45 Skrepnek, G. H., Mills, J. L., Sr. & Armstrong, D. G. A Diabetic Emergency One Million Feet Long: Disparities and Burdens of Illness among Diabetic Foot Ulcer Cases within Emergency Departments in the United States, 2006-2010. PLoS One 10 , e0134914, doi:10.1371/journal.pone.0134914 (2015).46 Mokoala, T. C., Sididzha, V., Molefe, E. D. & Luvhengo, T. E. Life expectancy of patients with diabetic foot sepsis post-lower extremity amputation at a regional hospital in a South African setting. A retrospective cohort study. Surgeon 22 , e109-e114, doi:10.1016/j.surge.2023.11.009 (2024).47 Jeyaraman, K., Berhane, T., Hamilton, M., Chandra, A. P. & Falhammar, H. Mortality in patients with diabetic foot ulcer: a retrospective study of 513 cases from a single Centre in the Northern Territory of Australia. BMC Endocr Disord 19 , 1, doi:10.1186/s12902-018-0327-2 (2019).48 Armstrong, D. G. et al. Five year mortality and direct costs of care for people with diabetic foot complications are comparable to cancer. J Foot Ankle Res 13 , 16, doi:10.1186/s13047-020-00383-2 (2020).49 Chen, L., Sun, S., Gao, Y. & Ran, X. Global mortality of diabetic foot ulcer: A systematic review and meta-analysis of observational studies. Diabetes Obes Metab 25 , 36-45, doi:10.1111/dom.14840 (2023).50 Armstrong, D. G., Boulton, A. J. M. & Bus, S. A. Diabetic Foot Ulcers and Their Recurrence. N Engl J Med 376 , 2367-2375, doi:10.1056/NEJMra1615439 (2017).51 Schultz, G. S., Chin, G. A., Moldawer, L. & Diegelmann, R. F. in Mechanisms of Vascular Disease: A Reference Book for Vascular Specialists (eds R. Fitridge & M. Thompson) (University of Adelaide Press© The Contributors 2011., 2011).52 Broughton, G., 2nd, Janis, J. E. & Attinger, C. E. The basic science of wound healing. Plast Reconstr Surg 117 , 12s-34s, doi:10.1097/01.prs.0000225430.42531.c2 (2006).53 Yanagisawa, M. et al. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 332 , 411-415, doi:10.1038/332411a0 (1988).54 Farndale, R. W., Sixma, J. J., Barnes, M. J. & de Groot, P. G. The role of collagen in thrombosis and hemostasis. J Thromb Haemost 2 , 561-573, doi:10.1111/j.1538-7836.2004.00665.x (2004).55 Clark, R. A. Fibrin is a many splendored thing. J Invest Dermatol 121 , xxi-xxii, doi:10.1046/j.1523-1747.2003.12575.x (2003).56 Kosmopoulos, M., Drekolias, D., Zavras, P. D., Piperi, C. & Papavassiliou, A. G. Impact of advanced glycation end products (AGEs) signaling in coronary artery disease. Biochim Biophys Acta Mol Basis Dis 1865 , 611-619, doi:10.1016/j.bbadis.2019.01.006 (2019).57 Soaita, I., Yin, W. & Rubenstein, D. A. Glycated albumin modifies platelet adhesion and aggregation responses. Platelets 28 , 682-690, doi:10.1080/09537104.2016.1260703 (2017).58 Rubenstein, D. A., Maria, Z. & Yin, W. Combined incubation of platelets and endothelial cells with glycated albumin: Altered thrombogenic and inflammatory responses. Diab Vasc Dis Res 11 , 235-242, doi:10.1177/1479164114531298 (2014).59 Esposito, C., Gerlach, H., Brett, J., Stern, D. & Vlassara, H. Endothelial receptor-mediated binding of glucose-modified albumin is associated with increased monolayer permeability and modulation of cell surface coagulant properties. J Exp Med 170 , 1387-1407, doi:10.1084/jem.170.4.1387 (1989).60 Undas, A., Brummel-Ziedins, K. E. & Mann, K. G. Antithrombotic properties of aspirin and resistance to aspirin: beyond strictly antiplatelet actions. Blood 109 , 2285-2292, doi:10.1182/blood-2006-01-010645 (2007).61 Rodrigues, M., Kosaric, N., Bonham, C. A. & Gurtner, G. C. Wound Healing: A Cellular Perspective. Physiol Rev 99 , 665-706, doi:10.1152/physrev.00067.2017 (2019).62 Murohara, T. et al. Vascular endothelial growth factor/vascular permeability factor enhances vascular permeability via nitric oxide and prostacyclin. Circulation 97 , 99-107, doi:10.1161/01.cir.97.1.99 (1998).63 Yager, D. R. & Nwomeh, B. C. The proteolytic environment of chronic wounds. Wound Repair Regen 7 , 433-441, doi:10.1046/j.1524-475x.1999.00433.x (1999).64 Bratton, D. L. & Henson, P. M. Neutrophil clearance: when the party is over, clean-up begins. Trends Immunol 32 , 350-357, doi:10.1016/j.it.2011.04.009 (2011).65 Zhang, Y., Kim, H. J., Yamamoto, S., Kang, X. & Ma, X. Regulation of interleukin-10 gene expression in macrophages engulfing apoptotic cells. J Interferon Cytokine Res 30 , 113-122, doi:10.1089/jir.2010.0004 (2010).66 Galli, S. J., Borregaard, N. & Wynn, T. A. Phenotypic and functional plasticity of cells of innate immunity: macrophages, mast cells and neutrophils. Nat Immunol 12 , 1035-1044, doi:10.1038/ni.2109 (2011).67 Sindrilaru, A. et al. An unrestrained proinflammatory M1 macrophage population induced by iron impairs wound healing in humans and mice. J Clin Invest 121 , 985-997, doi:10.1172/jci44490 (2011).68 Soro-Paavonen, A. et al. Advanced glycation end-products induce vascular dysfunction via resistance to nitric oxide and suppression of endothelial nitric oxide synthase. J Hypertens 28 , 780-788, doi:10.1097/HJH.0b013e328335043e (2010).69 Loots, M. A. et al. Differences in cellular infiltrate and extracellular matrix of chronic diabetic and venous ulcers versus acute wounds. J Invest Dermatol 111 , 850-857, doi:10.1046/j.1523-1747.1998.00381.x (1998).70 Berlanga-Acosta, J. A. et al. Cellular Senescence as the Pathogenic Hub of Diabetes-Related Wound Chronicity. Front Endocrinol (Lausanne) 11 , 573032, doi:10.3389/fendo.2020.573032 (2020).71 Wei, X. et al. Defective NCOA4-dependent ferroptosis in senescent fibroblasts retards diabetic wound healing. Cell Death Discov 9 , 138, doi:10.1038/s41420-023-01437-7 (2023).72 Samarawickrama, P. N. et al. Clearance of senescent cells enhances skin wound healing in type 2 diabetic mice. Theranostics 14 , 5429-5442, doi:10.7150/thno.100991 (2024).73 Loots, M. A., Lamme, E. N., Mekkes, J. R., Bos, J. D. & Middelkoop, E. Cultured fibroblasts from chronic diabetic wounds on the lower extremity (non-insulin-dependent diabetes mellitus) show disturbed proliferation. Arch Dermatol Res 291 , 93-99, doi:10.1007/s004030050389 (1999).74 Mendoza-Marí, Y. et al. Epidermal growth factor effect on lipopolysaccharide-induced inflammation in fibroblasts derived from diabetic foot ulcer. Scars Burn Heal 8 , 20595131211067380, doi:10.1177/20595131211067380 (2022).75 Maione, A. G. et al. Altered ECM deposition by diabetic foot ulcer-derived fibroblasts implicates fibronectin in chronic wound repair. Wound Repair Regen 24 , 630-643, doi:10.1111/wrr.12437 (2016).76 Monnier, V. M. et al. Skin collagen glycation, glycoxidation, and crosslinking are lower in subjects with long-term intensive versus conventional therapy of type 1 diabetes: relevance of glycated collagen products versus HbA1c as markers of diabetic complications. DCCT Skin Collagen Ancillary Study Group. Diabetes Control and Complications Trial. Diabetes 48 , 870-880, doi:10.2337/diabetes.48.4.870 (1999).77 Galkowska, H., Wojewodzka, U. & Olszewski, W. L. Chemokines, cytokines, and growth factors in keratinocytes and dermal endothelial cells in the margin of chronic diabetic foot ulcers. Wound Repair Regen 14 , 558-565, doi:10.1111/j.1743-6109.2006.00155.x (2006).78 Lu, Y. et al. Single-cell profiling reveals transcriptomic signatures of vascular endothelial cells in non-healing diabetic foot ulcers. Front Endocrinol (Lausanne) 14 , 1275612, doi:10.3389/fendo.2023.1275612 (2023).79 Huang, S. M. et al. High glucose environment induces M1 macrophage polarization that impairs keratinocyte migration via TNF-α: An important mechanism to delay the diabetic wound healing. J Dermatol Sci 96 , 159-167, doi:10.1016/j.jdermsci.2019.11.004 (2019).80 Zhu, P. et al. Impairment of human keratinocyte mobility and proliferation by advanced glycation end products-modified BSA. Arch Dermatol Res 303 , 339-350, doi:10.1007/s00403-010-1102-z (2011).81 Ploeger, D. T. et al. Cell plasticity in wound healing: paracrine factors of M1/ M2 polarized macrophages influence the phenotypical state of dermal fibroblasts. Cell Commun Signal 11 , 29, doi:10.1186/1478-811x-11-29 (2013).82 Izzo, V. et al. High matrix metalloproteinase levels are associated with dermal graft failure in diabetic foot ulcers. Int J Low Extrem Wounds 13 , 191-196, doi:10.1177/1534734614544959 (2014).83 Karam, R. A., Rezk, N. A., Abdel Rahman, T. M. & Al Saeed, M. Effect of negative pressure wound therapy on molecular markers in diabetic foot ulcers. Gene 667 , 56-61, doi:10.1016/j.gene.2018.05.032 (2018).84 Muller, M. et al. Matrix metalloproteinases and diabetic foot ulcers: the ratio of MMP-1 to TIMP-1 is a predictor of wound healing. Diabet Med 25 , 419-426, doi:10.1111/j.1464-5491.2008.02414.x (2008).85 Li, Z., Guo, S., Yao, F., Zhang, Y. & Li, T. Increased ratio of serum matrix metalloproteinase-9 against TIMP-1 predicts poor wound healing in diabetic foot ulcers. J Diabetes Complications 27 , 380-382, doi:10.1016/j.jdiacomp.2012.12.007 (2013).86 Luanraksa, S. et al. An MMP/TIMP ratio scoring system as a potential predictive marker of diabetic foot ulcer healing. J Wound Care 27 , 849-855, doi:10.12968/jowc.2018.27.12.849 (2018).87 Alvarez, O. M., Markowitz, L., Onumah, N. & Wendelken, M. Swift Downregulation of Gelatinases (MMP-2, MMP-9) in Neuropathic Diabetic Foot Ulcers Treated With Total Contact Cast. Wounds 31 , E39-e41 (2019).88 Musuuza, J. et al. A systematic review of multidisciplinary teams to reduce major amputations for patients with diabetic foot ulcers. J Vasc Surg 71 , 1433-1446.e1433, doi:10.1016/j.jvs.2019.08.244 (2020).89 Meloni, M. et al. Effect of a multidisciplinary team approach in patients with diabetic foot ulcers on major adverse limb events (MALEs): systematic review and meta-analysis for the development of the Italian guidelines for the treatment of diabetic foot syndrome. Acta Diabetol 61 , 543-553, doi:10.1007/s00592-024-02246-9 (2024).90 Schaper, N. C. et al. Practical Guidelines on the prevention and management of diabetic foot disease (IWGDF 2019 update). Diabetes Metab Res Rev 36 Suppl 1 , e3266, doi:10.1002/dmrr.3266 (2020).91 Armstrong, D. G., Tan, T. W., Boulton, A. J. M. & Bus, S. A. Diabetic Foot Ulcers: A Review. Jama 330 , 62-75, doi:10.1001/jama.2023.10578 (2023).92 Priyadarsini, S. L., Suresh, M. & Nikhila, G. Assessment framework for the selection of a potential interactive dressing material for diabetic foot ulcer. Heliyon 9 , e16476, doi:10.1016/j.heliyon.2023.e16476 (2023).93 Serena, T., Yaakov, S., Yaakov, R., King, E. & Driver, V. R. Percentage area reduction at week 4 as a prognostic indicator of complete healing in patients treated with standard of care: a post hoc analysis. J Wound Care 33 , S36-s42, doi:10.12968/jowc.2024.0141 (2024).94 Armstrong, D. G. et al. Maggot therapy in ”lower-extremity hospice” wound care: fewer amputations and more antibiotic-free days. J Am Podiatr Med Assoc 95 , 254-257, doi:10.7547/0950254 (2005).95 Widigdo, D. A. M., Sofro, Z. M., Pangastuti, H. S. & Dachlan, I. The Efficacy of Negative Pressure Wound Therapy (NPWT) on Healing of Diabetic Foot Ulcers: A Literature Review. Curr Diabetes Rev 20 , 1-11, doi:10.2174/0115733998229877230926073555 (2024).96 Zhang, Z., Zhang, W., Xu, Y. & Liu, D. Efficacy of hyperbaric oxygen therapy for diabetic foot ulcers: An updated systematic review and meta-analysis. Asian J Surg 45 , 68-78, doi:10.1016/j.asjsur.2021.07.047 (2022).97 Lin, C., Liu, J. & Sun, H. Risk factors for lower extremity amputation in patients with diabetic foot ulcers: A meta-analysis. PLoS One 15 , e0239236, doi:10.1371/journal.pone.0239236 (2020).98 Ince, P. et al. Use of the SINBAD classification system and score in comparing outcome of foot ulcer management on three continents. Diabetes Care 31 , 964-967, doi:10.2337/dc07-2367 (2008).99 Treece, K. A., Macfarlane, R. M., Pound, N., Game, F. L. & Jeffcoate, W. J. Validation of a system of foot ulcer classification in diabetes mellitus. Diabet Med 21 , 987-991, doi:10.1111/j.1464-5491.2004.01275.x (2004).100 Hicks, C. W. et al. The Society for Vascular Surgery Wound, Ischemia, and foot Infection (WIfI) classification system correlates with cost of care for diabetic foot ulcers treated in a multidisciplinary setting. J Vasc Surg 67 , 1455-1462, doi:10.1016/j.jvs.2017.08.090 (2018).101 Jeon, B. J., Choi, H. J., Kang, J. S., Tak, M. S. & Park, E. S. Comparison of five systems of classification of diabetic foot ulcers and predictive factors for amputation. Int Wound J 14 , 537-545, doi:10.1111/iwj.12642 (2017).102 McDermott, K., Fang, M., Boulton, A. J. M., Selvin, E. & Hicks, C. W. Etiology, Epidemiology, and Disparities in the Burden of Diabetic Foot Ulcers. Diabetes Care 46 , 209-221, doi:10.2337/dci22-0043 (2023).103 Weatherall, T. et al. The impact of machine learning on the prediction of diabetic foot ulcers - A systematic review. J Tissue Viability 33 , 853-863, doi:10.1016/j.jtv.2024.07.004 (2024).104 Longfield, M. S., Lourdesamy, J., Min, D. & Twigg, S. M. Do diabetes-related foot ulcer wound fluid measures have clinical utility as biomarkers for healing? A systematic review. J Wound Care 32 , xlvii-lxii, doi:10.12968/jowc.2023.32.Sup4a.xlvii (2023).105 Fu, K. et al. Role of matrix metalloproteinases in diabetic foot ulcers: Potential therapeutic targets. Front Pharmacol 13 , 1050630, doi:10.3389/fphar.2022.1050630 (2022).106 Lobmann, R. et al. Expression of matrix-metalloproteinases and their inhibitors in the wounds of diabetic and non-diabetic patients. Diabetologia 45 , 1011-1016, doi:10.1007/s00125-002-0868-8 (2002).107 Liu, Y. et al. Increased matrix metalloproteinase-9 predicts poor wound healing in diabetic foot ulcers. Diabetes Care 32 , 117-119, doi:10.2337/dc08-0763 (2009).108 Jindatanmanusan, P., Luanraksa, S., Boonsiri, T., Nimmanon, T. & Arnutti, P. Wound Fluid Matrix Metalloproteinase-9 as a Potential Predictive Marker for the Poor Healing Outcome in Diabetic Foot Ulcers. Patholog Res Int 2018 , 1631325, doi:10.1155/2018/1631325 (2018).109 Ulrich, D., Smeets, R., Unglaub, F., Wöltje, M. & Pallua, N. Effect of oxidized regenerated cellulose/collagen matrix on proteases in wound exudate of patients with diabetic foot ulcers. J Wound Ostomy Continence Nurs 38 , 522-528, doi:10.1097/WON.0b013e31822ad290 (2011).110 Yao, M. et al. A pilot study evaluating non-contact low-frequency ultrasound and underlying molecular mechanism on diabetic foot ulcers. Int Wound J 11 , 586-593, doi:10.1111/iwj.12005 (2014).111 Werner, S. & Grose, R. Regulation of wound healing by growth factors and cytokines. Physiol Rev 83 , 835-870, doi:10.1152/physrev.2003.83.3.835 (2003).112 Ohura, N. et al. Efficacy of autologous platelet-rich plasma gel in patients with hard-to-heal diabetic foot ulcers: a multicentre study in Japan. J Wound Care 33 , 484-494, doi:10.12968/jowc.2023.0088 (2024).113 Azam, M. S. et al. Efficacy of Platelet-Rich Plasma in the Treatment of Diabetic Foot Ulcer. Cureus 16 , e60934, doi:10.7759/cureus.60934 (2024).114 Bang, F. S., Leeberg, V., Ding, M. & Dreyer, C. H. The effect of VEGF stimulation in diabetic foot ulcers: A systematic review. Wound Repair Regen 32 , 384-392, doi:10.1111/wrr.13171 (2024).115 Theocharidis, G. et al. Integrated Skin Transcriptomics and Serum Multiplex Assays Reveal Novel Mechanisms of Wound Healing in Diabetic Foot Ulcers. Diabetes 69 , 2157-2169, doi:10.2337/db20-0188 (2020).116 Xu, J. et al. The risk factors in diabetic foot ulcers and predictive value of prognosis of wound tissue vascular endothelium growth factor. Sci Rep 14 , 14120, doi:10.1038/s41598-024-64009-4 (2024).117 Stehr, A. et al. VEGF: a surrogate marker for peripheral vascular disease. Eur J Vasc Endovasc Surg 39 , 330-332, doi:10.1016/j.ejvs.2009.09.025 (2010).118 Che, D. et al. Predictors of amputation in patients with diabetic foot ulcers: a multi-centre retrospective cohort study. Endocrine 85 , 181-189, doi:10.1007/s12020-024-03704-8 (2024).119 Asadi, M. R. et al. Angiogenic effects of low-intensity cathodal direct current on ischemic diabetic foot ulcers: A randomized controlled trial. Diabetes Res Clin Pract 127 , 147-155, doi:10.1016/j.diabres.2017.03.012 (2017).120 Zhang, J. et al. Increased growth factors play a role in wound healing promoted by noninvasive oxygen-ozone therapy in diabetic patients with foot ulcers. Oxid Med Cell Longev 2014 , 273475, doi:10.1155/2014/273475 (2014).121 Beer, H. D., Longaker, M. T. & Werner, S. Reduced expression of PDGF and PDGF receptors during impaired wound healing. J Invest Dermatol 109 , 132-138, doi:10.1111/1523-1747.ep12319188 (1997).122 Dinh, T. et al. Mechanisms involved in the development and healing of diabetic foot ulceration. Diabetes 61 , 2937-2947, doi:10.2337/db12-0227 (2012).123 Mi, Q., Rivière, B., Clermont, G., Steed, D. L. & Vodovotz, Y. Agent-based model of inflammation and wound healing: insights into diabetic foot ulcer pathology and the role of transforming growth factor-beta1. Wound Repair Regen 15 , 671-682, doi:10.1111/j.1524-475X.2007.00271.x (2007).124 Fang, Y. et al. Granulocyte-macrophage colony-stimulating factor enhances wound healing in diabetes via upregulation of proinflammatory cytokines. Br J Dermatol 162 , 478-486, doi:10.1111/j.1365-2133.2009.09528.x (2010).125 Li, J. Y., Wang, Z. J., Deng, A. P. & Li, Y. M. ENA-78 Is a Novel Predictor of Wound Healing in Patients with Diabetic Foot Ulcers. J Diabetes Res 2019 , 2695436, doi:10.1155/2019/2695436 (2019).126 Wang, X., Li, J., Wang, Z. & Deng, A. Wound exudate CXCL6: a potential biomarker for wound healing of diabetic foot ulcers. Biomark Med 13 , 167-174, doi:10.2217/bmm-2018-0339 (2019).127 Qin, Q., Haba, D. & Nakagami, G. Which biomarkers predict hard-to-heal diabetic foot ulcers? A scoping review. Drug Discov Ther 17 , 368-377, doi:10.5582/ddt.2023.01086 (2024).128 Zhou, Y., Zhou, W., Guo, Y. & Hu, C. Correlation of Wound Prognosis with serum IL-6, ICAM-1 and sST2 in Patients with Diabetic Foot and Construction of a Nomogram Model. Int J Low Extrem Wounds , 15347346251345262, doi:10.1177/15347346251345262 (2025).129 Zheng, A., Xu, Y., Cen, N. & Wu, B. A Lower IL-34 Expression Is Associated with Non-Healing Diabetic Foot Ulcers. Comb Chem High Throughput Screen 27 , 1533-1543, doi:10.2174/0113862073273222231005065757 (2024).130 Fadini, G. P. et al. The molecular signature of impaired diabetic wound healing identifies serpinB3 as a healing biomarker. Diabetologia 57 , 1947-1956, doi:10.1007/s00125-014-3300-2 (2014).131 Açık, Ş., Çetinkalp, Ş., Aslan Ç, G. & Akçay, Y. Gremlin-1 and Wagner Classification: Potential Biomarker for Amputation in Diabetic Foot Patients. Int J Low Extrem Wounds , 15347346251337157, doi:10.1177/15347346251337157 (2025).132 Walsh, D. W., Godson, C., Brazil, D. P. & Martin, F. Extracellular BMP-antagonist regulation in development and disease: tied up in knots. Trends Cell Biol 20 , 244-256, doi:10.1016/j.tcb.2010.01.008 (2010).133 Mitola, S. et al. Gremlin is a novel agonist of the major proangiogenic receptor VEGFR2. Blood 116 , 3677-3680, doi:10.1182/blood-2010-06-291930 (2010).134 Waikel, R. L., Kawachi, Y., Waikel, P. A., Wang, X. J. & Roop, D. R. Deregulated expression of c-Myc depletes epidermal stem cells. Nat Genet 28 , 165-168, doi:10.1038/88889 (2001).135 Jozic, I. et al. Stress Signals, Mediated by Membranous Glucocorticoid Receptor, Activate PLC/PKC/GSK-3β/β-catenin Pathway to Inhibit Wound Closure. J Invest Dermatol 137 , 1144-1154, doi:10.1016/j.jid.2016.11.036 (2017).136 Kirsner, R. S. et al. Evaluation of c-Myc and Phosphorylated Glucocorticoid Receptor (p-GR) for Predicting Diabetic Foot Ulcer Healing-A Diabetic Foot Consortium Study. Wound Repair Regen 33 , e70044, doi:10.1111/wrr.70044 (2025).137 Bajpai, A. et al. Effects of Non-thermal, Non-cavitational Ultrasound Exposure on Human Diabetic Ulcer Healing and Inflammatory Gene Expression in a Pilot Study. Ultrasound Med Biol 44 , 2043-2049, doi:10.1016/j.ultrasmedbio.2018.05.011 (2018).138 Nassiri, S., Zakeri, I., Weingarten, M. S. & Spiller, K. L. Relative Expression of Proinflammatory and Antiinflammatory Genes Reveals Differences between Healing and Nonhealing Human Chronic Diabetic Foot Ulcers. J Invest Dermatol 135 , 1700-1703, doi:10.1038/jid.2015.30 (2015).139 Soto, V. et al. Turning Waste Into Insight: A Novel Proteomic Approach to Non-Invasive Wound Biomarker Discovery. Mol Cell Proteomics 24 , 101058, doi:10.1016/j.mcpro.2025.101058 (2025).140 Han, Q., Gu, Y. & Qian, Y. Study on the mechanism of activating SIRT1/Nrf2/p62 pathway to mediate autophagy-dependent ferroptosis to promote healing of diabetic foot ulcers. Naunyn Schmiedebergs Arch Pharmacol 398 , 3015-3025, doi:10.1007/s00210-024-03400-4 (2025).141 Yang, W., Cao, H., Lin, S. & Tian, F. Investigating serum NRF2 as a screening biomarker and potential therapeutic target in diabetic foot ulcers: a prospective observational study. Eur J Med Res 30 , 841, doi:10.1186/s40001-025-03086-1 (2025).142 Sadyah, N. A. C., Heri, N., Putra, A. & Riwanto, I. Secretome of Human MSC Gel Improves DFU Healing through NF-?B p50 and CD163 mRNA Expression. Pak J Biol Sci 28 , 151-161, doi:10.3923/pjbs.2025.151.161 (2025).143 2024, C. F. F. (2024). 116 VEGF ELISA 502 52 Higher plasma VEGF levels in non-healers AUC: 0.728 Cut off - 1.77 μg/L Sensitivity - 89.50% Specificity - 57.30% 85 MMP-9/TIMP-1 ELISA 93 12 Higher MMP-9/TIMP-1 ratio in non-healers AUC: 0.658 Cut off - <0.395 Sensitivity - 63.6% Specificity - 58.6% 131 Gremlin-1 ELISA 62 No defined follow up period (2019 to 2020) Lower Gremlin-1 levels in subjects with an endpoint of amputated AUC: 0.649 ± 0.064 Cut off - 2.47 ng/mL Sensitivity - 67% Specificity - 46% IL-6 ELISA 210 26 Higher serum levels in poor prognosis AUC: 0.669 Cut off - 107 pg/mL Sensitivity - 56% Specificity - 76% ICAM-1 ELISA 210 26 Higher serum levels in poor prognosis AUC: 0.69 Cut off - 374.294 ng/mL Sensitivity - 68% Specificity - 67% sST2 ELISA 210 26 Higher serum levels in poor prognosis AUC: 0.682 Cut off - 44.737 ng/mL Sensitivity - 37% Specificity - 91% IL-6/ICAM-1/sST2 ELISA 210 26 Higher serum levels in poor prognosis AUC: 0.786 Sensitivity – 68% Specificity – 83% Biomarkers in wound tissue 116 VEGF WB 502 52 Higher VEGF levels in healing ulcers AUC: 0.790 Cut off - 1.24 μg/L Sensitivity - 93.00%, Specificity - 65.10%. 116 VEGF expression IHC 502 52 Higher VEGF expression in healing ulcers AUC: 0.759 Sensitivity - 89.10% Specificity - 74.30% 130 SERPINB3/total protein ELISA and Gene expression analysis 47 26 Higher SERPINB3/total protein in healing ulcers AUC: 0.665 Cut off - 1.13 ng ml −1 μg −1 μl −1 Sensitivity of 75% Specificity of 62.5% 136 c-Myc + wound size and duration IHC 107 12 No statistically significant difference in the nuclear presence of c-Myc between healed and non-healed wounds AUC 0.637 136 p-GR + wound size and duration IHC 107 12 No statistically significant difference in the nuclear presence of p-GR between healed and non-healed wounds AUC 0.643 Biomarkers in wound fluid 125 ENA-78 ELISA 84 24 Lower levels of ENA-78 in patients with non-healing ulcers AUC: 0.705 Cut off - 1792.00 ng/ml Sensitivity - 45.90 % Specificity - 89.58 % 84 MMP-1/TIMP-1 ELISA 16 12 Higher baseline MMP-1/TIMP-1 ratio was associated with >82% wound surface area at week 4 AUC: 0.821 Cut off – 0.39 Sensitivity – 71 % Specificity - 87.5 % 86 MMP-1/TIMP-1* ELISA 22 12 Baseline MMP-1/TIMP-1 ratio was higher in healing ulcers AUC: 0.802 Cut off - 0.38 pg/µg protein Sensitivity – 81.8 % Specificity – 64.6 % 86 MMP-9/TIMP-1* ELISA 22 12 Baseline MMP-9/TIMP-1 ratio was higher in non-healing ulcers AUC: 0.777 Cut off - >9.06 Sensitivity – 90.9 % Specificity – 64.6 % 107 Pro-MMP-9/TIMP-1 Zymography was used to detect MMP-9 ELISA was used to detect TIMP-1 62 12 Higher baseline pro-MMP-9 and pro-MMP-9/TIMP-1 ratio was associated with non-healing ulcers AUC: 0.94 (No cut off for Pro-MMP-9/TIMP-1 ratio was provided) Sensitivity – 87 % Specificity – 91 % 126 CXCL-6 ELISA 100 24 Higher baseline CXCL-6 was associated with healing ulcers AUC: 0.965 Cut off: 846.90 ng/ml Sensitivity - 87.27 % Specificity - 95.56 % Table 1: Biomarkers (measured at baseline) that have been shown to have predictive value in DFU outcomes. Biomarkers appear more than once if they were evaluated in multiple studies and have been denoted by an asterisk (*). MMP, matrix metalloproteinase; TIMP, tissue inhibitor of matrix metalloproteinase; VEGF, vascular endothelial growth factor; SERPINB3, serpin peptidase inhibitor, clade B (ovalbumin), member 3; ENA-78—epithelial neutrophil-activating peptide 78; TGF-β1, transforming growth factor beta 1; CXCL6—chemokine (C-X-C motif) ligand 6; IHC, Immunohistochemistry; WB, Western blot; ICAM-1, intracellular adhesion molecule-1; sST2, soluble suppression of tumorigenicity 2; IL-6, interleukin-6; p-GR, phosphorylated glucocorticoid receptor Figure 1: Five-year mortality rates of nine of the most prevalent cancers according to the American Cancer Society, Cancer Facts & Figures 2024 143 compared to the five-year mortality of DFU 48 . Figure 2: Two matched subjects with DFUs and a similar Texas score on initial presentation with images showing varied outcomes. Patient 1 A). Initial presentation Texas A1; B). 6 months after initial presentation, outcome of minor limb amputation. Patient 2 C). Initial presentation Texas A1; D). 6 months after initial presentation, outcome of complete healing. Figure 3: Comparing the performance of biomarkers that have been studied in predicting the outcome of diabetic foot ulcers (DFU) using their Area Under the Curve (AUC), where a higher AUC suggests better performance of the biomarker in distinguishing the outcome of DFU in individual studies. Orange represents biomarkers in serum; blue represents biomarkers in wound fluid; green represents biomarkers in wound tissue. Biomarkers appear more than once if they were evaluated in multiple studies and have been denoted by an asterisk (*). MMP-9/TIMP-1 85 , serum VEGF 116 , CXCL-6 126 , ENA-78 125 , MMP-1/TIMP-1* 86 , MMP-1/TIMP-1 84 , MMP-9 86,108 , MMP-9/TIMP-1* 86 , Pro-MMP-9/TIMP-1 107 , SERPINB3 130 , Tissue VEGF 116 , VEGF expression IHC 116 , ICAM-1 128 , sST2 128 , IL-6 128 , p-GR 136 , c-Myc 136 , combined IL-6/ ICAM-1/ sST2 128 MMP, matrix metalloproteinase; TIMP, tissue inhibitor of matrix metalloproteinase; VEGF, vascular endothelial growth factor; SERPINB3, serpin peptidase inhibitor, clade B member 3; ENA-78—epithelial neutrophil-activating peptide 78; TGF-β1, transforming growth factor beta 1; CXCL6—chemokine (C-X-C motif) ligand; IHC, Immunohistochemistry; ICAM-1, intracellular adhesion molecule-1; sST2, soluble suppression of tumorigenicity 2; IL-6, interleukin-6; p-GR, phosphorylated glucocorticoid receptor Supplementary Material File (okiro et al biomarkers for dfu_ table 1.docx) Download 34.77 KB File (okiro et al biomarkers for dfu_ fig 1.docx) Download 112.18 KB File (okiro et al biomarkers for dfu_ fig 2.docx) Download 1.83 MB File (okiro et al biomarkers for dfu_ fig 3.docx) Download 67.91 KB Information & Authors Information Version history V1 Version 1 29 January 2026 Peer review timeline Published VIEW Version of Record 28 Apr 2026 Published Copyright This work is licensed under a Non Exclusive No Reuse License. Collection View Keywords diabetic foot diabetic foot ulcer biomarkers diabetic foot ulcer healing diabetic foot ulcer outcome prediction diabetic foot ulcers limb salvage Authors Affiliations Julie Okiro 0009-0003-4269-3191 RCSI Department of Anatomy and Regenerative Medicine View all articles by this author Kellie Fortune Connolly Hospital Blanchardstown View all articles by this author Eimear Daly Connolly Hospital Blanchardstown View all articles by this author Luca McCann Royal College of Surgeons in Ireland View all articles by this author Merhan Soltan Royal College of Surgeons in Ireland View all articles by this author Seamus Sreenan Connolly Hospital Blanchardstown View all articles by this author Fabio Quondamatteo [email protected] RCSI Department of Anatomy and Regenerative Medicine View all articles by this author Metrics & Citations Metrics Article Usage 268 views 77 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Julie Okiro, Kellie Fortune, Eimear Daly, et al. Biomarkers to predict outcomes in diabetic foot ulcers. Authorea . 29 January 2026. DOI: https://doi.org/10.22541/au.176970084.40139491/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176970084.40139491/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe4aa268c9e4193',t:'MTc3OTIxMDA5Nw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();