Pyoderma Gangrenosum: Treatment Options

Pyoderma gangrenosum (PG) is an ulcerative inflammatory skin disease classified within the neutrophilic dermatoses group. It is a rare disease with a prevalence of 5.8 cases per 100,000 adults [1, 2]. PG negatively impacts quality of life and carries a three times higher mortality risk than that of the general population [3]. The classic clinical presentation are very painful skin ulcers with a predilection for the lower extremities in approximately 80% of cases [2, 4, 5] (Fig. 1). When the lower extremities are involved, it is particularly important to fully evaluate for other causes of skin ulceration. For example, venous and arterial status must be assessed, and frequent evaluation for infection is necessary. In addition, skin biopsy can rule out many other differential diagnoses. However, any hair-bearing skin on the body has the potential to develop PG. Less common presentations including bullous, verrucous and pustular variants have been described [6]. Other ulcerative variants include peristomal PG. In the initially acute inflammatory stage of classic PG, patients often have an erythematous papule or pustule that rapidly develops into a painful skin ulceration or multiple ulcerations with gunmetal grey borders and violaceous erythema. Once the acute inflammation has subsided, large wounds may also present without classic or distinctive features, making the diagnosis even more challenging [6]. The pathergy phenomenon has been reported in up to 30% of patients with PG and describes the formation of new ulcers after minor trauma (e.g. needle injury). Sometimes this is a clinical clue for the diagnosis of PG; however, it is not specific to PG and has also been described in other patients with auto-inflammatory or neutrophilic conditions [7].

PG still represents a diagnostic challenge, as it is a rare disease with several clinical mimickers. With a misdiagnosis rate of up to 39%, it is considered the prototype of misdiagnosis among other skin ulcerations where the initial diagnosis was PG [8]. The most common clinical mimickers are venous leg ulcers, vasculitis, vasculopathies and factitial ulcers (Table 1). Recent efforts from clinical researchers, including experts worldwide, have supported the implementation of diagnostic frameworks to improve this critical issue in clinical practice. Three diagnostic criteria have been proposed: the Su and Delphi consensus, and the PARACELSUS score. All these diagnostic criteria seem to be useful in clinical practice, but the PARACELSUS score has shown the highest sensitivity among comparative studies [9, 10]. However, the specificity of all of them remains relatively low. The classic histological findings in PG have been well described. In early pustules or ulcerations, diffuse neutrophilic infiltrates are found around hair follicles [11]. However, this inflammatory infiltrate evolves into a non-specific inflammatory infiltrate (e.g. lymphohistiocytic infiltrate) once the ulceration becomes chronic. In two retrospective chart reviews of patients with PG, characteristic biopsy findings of PG were reported in only 7% (8 of 103) and 11% (5 of 47), respectively [4, 9]. PG is often associated with other inflammatory disorders such as inflammatory bowel disease (IBD 13.0–22.7%) and rheumatoid arthritis (RA 9.2–16.9%) [12‐14]. It is also important to note that PG does not reflect the clinical course of the underlying disease, and it is not uncommon for patients with PG to have well-controlled IBD or vice versa [14]. Some patients also have underlying haematologic disorders (monoclonal gammopathy of unknown significance) or malignancies (leukaemias or other neoplasia) [15, 16]. Despite these comorbidities being clinical clues in the diagnosis and workup for PG, they seem to have a more preponderant role in the prognosis and/or selection an appropriate therapeutic approach [17] (Fig. 2). Besides medical management to control the associated comorbidity and the inflammation inherent to PG with either local pharmacologic therapy and/or systemic therapy, wound care management and pain control also needs to be addressed throughout the treatment course.

The occurrence of PG after taking drugs has been repeatedly reported in the literature (Table 2). Although the underlying pathomechanisms are not clearly known, it must be assumed that a dysregulated inflammatory reaction with disturbed migration and activation of neutrophil granulocytes also occurs as a result of the drugs in genetically predisposed individuals. Keratinocyte apoptosis and alteration of epigenetic mechanisms might play a role as well [18]. A direct assessment of the causal relationship of the intake of a drug and the induction of a PG is not always easy and reports with drug rechallenge are rare. Mostly, these are descriptions of single patients. Many of those patients had an underlying disease such as (haematologic) neoplasia or a chronic inflammatory disease [19]. Thus, it cannot always be said with certainty whether the drug was the trigger for the PG or whether the disease would have occurred even without the drug. Another important aspect is the temporal relationship between the start of drug therapy and the occurrence of PG. Very different time periods have been described [20, 21]. For example, if a large proportion of the world's population received appropriate vaccinations during the Covid-19 pandemic, the temporal context would include patients with PG [22]. It is then only possible to determine whether clustering has occurred by analysing large patient collectives. In the context of rarely occurring PG, however, these are exclusively individual case descriptions, which in principle, could also be random. However, other immune-mediated clinical diseases were also found significantly more frequently in association with the Covid-19 vaccinations, so a connection could certainly exist here [23].

In the first study summarising drug-induced PG, colony-stimulating factors (CSFs) and tyrosine kinase inhibitors (TKIs) were the most common culprits [18]. CSFs are haematopoietic growth factors used for chemotherapy-induced bone marrow suppression and have been reported as possible culprits since the 1990s [24, 25]. TKIs are small molecules used as targeted therapy in several cancers; the most common culprit in drug-induced PG within this group has been sunitinib [26]. Subsequently, in an analysis of 44 publications with 52 cases of drug-induced PG based on the Naranjo criteria [19], the most common probable culprits were cocaine and levamisole. The anthelmintic agent levamisole is often added to cocaine because it is an inexpensive additive that prolongs the effects of cocaine. Levamisole is known to alter the chemotaxis response of neutrophils. In addition, levamisole and cocaine can increase inflammation and autoimmunity [27]. Reviewing the list of drugs that have been implicated in causing PG, it is striking that these are often medicaments that have otherwise been successfully used for PG therapy. These paradoxical reactions have already been described for various other autoimmunological disease patterns, such as psoriasis, hidradenitis suppurativa or inflammatory bowel diseases [28, 29]. They begin a few weeks to months after the start of biologic therapy and resolve after discontinuation of the drugs without specific therapy [30]. In an analysis of all case reports in which PG was thought to have been triggered by the administration of biological therapy, 57 patients were identified. Here, 41 cases involved rituximab, 12 cases involved TNF-α inhibitors, 3 cases involved interleukin 17A and one case involved cytotoxic T-lymphocyte-associated protein 4 antibodies [31]. It is hypothesised that in these paradoxical reactions, there is a shift of the cytokine balance towards a molecular pattern typical of the respective induced disease. Looking more closely at the case reports, almost all published patients had systemic inflammatory diseases. Therefore, it is important to consider that the PG was possibly not induced by the biologic therapy but rather the biologic did not sufficiently suppress the induction of PG. After discontinuation of the drugs, other immunosuppressive or immunomodulatory therapies were almost always used, so the improvement of the PG could also be attributed to the newly initiated therapies.

Interestingly, some of the drugs are administered by injection. Here, it must be clarified whether the drug or a pathergy phenomenon related to the physical stimulus was causally responsible. If, for example, PG occurs at the injection site of enoxaparin [32] or erythropoietin [33], it must be assumed that the pathergy phenomenon was more likely the contributing factor. Interestingly, for azacitidine, the occurrence of a PG has been described both at the injection site [34] and independently [35]. Drug-induced PG is diagnostically challenging and often seen in the presence of other predisposing causes (e.g. underlying inflammatory conditions). Meaningful testing or other direct detection methods have not yet been developed. It is therefore very important to recognise the potential association and avoid the triggering drugs if possible.

Once the diagnosis of PG has been established, the objective clinical features of severity for PG such as ulcer size, depth, number and location can guide the next steps. Patients with PG often present to clinicians at a late stage with ulceration, and the focus is placed on systemic therapies to gain rapid control of the inflammation. However, a subset of patients might present with early and small ulcerations, e.g. patients who are having a recurrence, and be in a position to start local treatment early in the course of the disease. Based on our experience, small PG ulcers (4 cm² or less) without compromise of the deep structures (e.g. muscle/tendon) might be ideal for trying local or intra-lesional immunosuppressants. Since local therapy can help limit the dose and duration of systemic therapeutics, they can be useful throughout the entire treatment period.

While the main therapeutic options for classic PG remain those listed by Maronese et al. [49], evidence from a large, multi-centre, retrospective cohort study as well as an expert survey study shows that PG patients receive an average of two different systemic agents. This underscores the importance of combination treatments [50, 51] to achieve ulcer healing [2]. Combination treatment has not been defined but based on our experience, overlapping classic immunosuppressants and corticosteroid sparing agents for at least 4 weeks throughout the course of the disease is becoming standard of care in clinical practice when patient require systemic treatments (Table 3).

From our experience, local pharmacologic therapy only can be attempted when there are single small (≤ 4 cm²) PG skin lesions with or without ulceration. The most common approach is to use a high potency steroid (e.g. clobetasol) or a calcineurin inhibitor (e.g. tacrolimus). In a comparative prospective study of these two alternatives, ulcer size was an independent predictor of healing [38]. When using topical therapy for PG, it is important to closely monitor progress with objective measurements and photography, in addition to establishing a strict timeline for re-assessment of therapy. The addition of systemic treatment should be considered if progress is lacking after 2–4 weeks of therapy. In addition, given the prolonged duration needed to control the inflammation, treating PG with local therapy can contribute to bridging or tapering the systemic immunosuppression and decrease the possibility of side effects. In general, topical therapy for PG is used as an adjuvant to systemic therapy and is usually applied to the ulcer itself and/or ulcer edges [41].

Systemic therapy should be considered when an ulcer is large (> 4 cm²) or if there are numerous or PG ulcers. CS and/or cyclosporine have been used for many years due to their rapid onset of action in stopping and/or decreasing the progression of skin ulcerations. However, the healing rate is less than 50% at 6 months and classical side effects associated with each type of medication determine which patient might benefit from one or the other. Sparing agents are often added to first-line immunosuppressive medications to maintain the inhibition of the immune system. Currently, the most commonly used sparing agents are biologics. Most the data, including one randomised control trial (RCT), favours the use of TNF-α inhibitors, particularly infliximab. However, growing evidence has shown that inhibitors of IL-1, IL-12/23, IL-17 and C5a are another alternative to be considered (Fig. 3).

Biologics can already be used as first choice drugs in patients with inflammatory comorbidities such as IBD and inflammatory arthritis. Here, approval exists and several disease aspects can be treated simultaneously with one therapeutic agent.

In paraneoplastic PG, there is stronger evidence that treatment of the underlying malignancy may be helpful in treating PG. This does not always preclude the initiation of systemic immunosuppression as initial treatment when it is first required. Nevertheless, immunomodulatory therapy with options such as dapsone or IVIG can be considered initially.

It is always difficult to determine when systemic therapy should be stopped. It is important to remember that systemic therapy is intended to control inflammation. If no more inflammation can be detected, then systemic therapy can be successively reduced and does not necessarily have to be continued until the wounds are completely healed. However, this decision must always be made on an individual basis.

The clinical picture of PG was first described by Brocq over a hundred years ago [102]. It is still an enigmatic disease for healthcare providers and even more so for patients who often endure consultation with many physicians prior to finding one who might be familiar with the diagnosis and treatment of PG and able to provide much needed care and comfort. Sustaining progress in the management of PG relies on understanding its pathogenesis. It is still not known what stimulates neutrophil trafficking to cause skin ulcerations. It seems that abnormal activation of the innate immune system in patients with genetic predisposition provide an ideal state for PG to develop [103]. However, neutrophil-T cell crosstalk causing the activation of Th17 downstream cytokines is now supported by gene expression studies [6, 104].

Systemic therapies with CS and cyclosporine have remained the first-line options for most PG patients for several years. Despite the increased use of biologic therapy and the association of PG with inflammatory comorbidities, there is still a lack of strong evidence in support of their acceptance by regulatory agencies such as the Food and Drug Administration (FDA) and European Medicines Agency (EMA) [49]. If these patients do not have an associated comorbidity, the approval of biologics in clinical practice is even more challenging. The current evidence for the use of other, non-biologics favours MMF and dapsone (Table 3). These can be used as sparing agents, but their onset of action seems to be slower. However, the use of dapsone might confer Pneumocystis jirovecii prophylaxis (PJP) in patients treated with immunosuppressive for longer periods of time [105]. In addition, the appearance of PG ulcers on more sensitive areas (face and neck, genitalia) or the association with extracutaneous manifestations (e.g. lung involvement) might prompt the consideration of more aggressive initial treatment which could include pulses of CS or combination of CS with cyclosporine.

With all forms of therapy, it is very important to clarify in advance whether these treatments are at all feasible for patients who often have underlying illnesses. This involves clarifying side effects, but also the type of therapy to be carried out and whether the implicit costs will be covered by the insurer. Especially in the inflammatory phase, patients must be monitored closely. With control of the inflammation, the pain should decrease and the livid border around the wound should recede. If clinical symptoms progress despite 2–4 weeks of therapy, therapy should be extended or changed. Moreover, in the event of unsuccessful therapy, the PG diagnosis should also be critically questioned and, if necessary, re-examined. Misdiagnosis is still the main issue with PG [8]. Despite the proposal of new diagnostic frameworks, the impact on the misdiagnosis rate is yet to be seen.

As clinical guidelines are in development, healthcare providers rely on their knowledge of PG pathophysiology combined with their experience in wound care principles and training in analgaesic strategies to determine the most appropriate approach to treat these patients. Although more and more drugs are available for the medical treatment of PG, most are still considered off label and access is an issue for patients worldwide. When patient does not respond appropriately to immunosuppressive therapy, misdiagnosis is a consideration. However superimposed cellulitis or other concomitant ulcer aetiologies (e.g. chronic venous insufficiency) have to be addressed to assure treatment response. Conversely, high doses or levels of immunosuppressants can also negatively affect healing throughout the course of the disease.

In the future, we need to deepen our understanding of the pathophysiology of this disease, search for specific markers to aid in the enrolment of patients in clinical studies and possibly to distinguish PG from other ulcerative skin diseases, and ultimately develop targeted therapies that can feasible be studied in RCTs.

PG is a rare inflammatory dermatological disease that leads to very painful ulcerations and causes a significantly reduced quality of life. Once the diagnosis is confirmed with validated scores, numerous immunosuppressive and immunomodulatory therapies are now available. In particular, biologics will play an increasingly important role in the treatment regimens of patients with PG.