Guidelines for the Management of Stevens–Johnson Syndrome and Toxic Epidermal Necrolysis 2025 Supplement

Stevens–Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are characterized by high fever, body malaise, and multiple erythema, erosions, and blisters on the skin and mucocutaneous junction. They are considered life-threatening and intractable diseases that leave sequelae, including blindness and respiratory disorders. The diagnostic criteria for SJS/TEN were developed in 2005 by the Japanese Ministry of Health, Labour and Welfare Research Group on Severe Erythema Multiforme in Japan. To further understand the actual clinical scenario, the first epidemiological survey targeting dermatologists' full-time facilities nationwide was conducted from 2005 to 2007 and subsequently published. In 2015, the Japanese guidelines for the management of SJS and TEN were established. The second national survey for SJS/TEN, conducted from 2016 to 2018, identified vascular disease, malignancy, and diabetes mellitus as risk factors for SJS/TEN development. Moreover, they also identified sepsis and diabetes as risk factors for death and sepsis, respectively. In this survey, the mortality rates of both SJS and TEN were higher than those recorded in the first national survey. Based on these results, we revised the existing guidelines by incorporating recent knowledge on the pathogenesis, medical complications, treatment, and prognosis prediction of SJS/TEN with the aim of saving the lives of patients with SJS/TEN in the future. This addendum provides standard guidelines at this point. However, several of these guidelines are based on the committee's opinions, because clinical studies with high-level evidence are limited due to the rarity of these diseases, and their pathogeneses are often unclear. In actual practice, the treatment approach must consider each patient's situation, and the guidelines are not intended to force treatment selection or limit a physician's discretion. Levels of Evidence Classification: I: Systematic review/meta-analysis. II: One or more randomized controlled trials. III: Nonrandomized controlled trials. IVa: Epidemiological studies, such as cohort studies. IVb: Epidemiological studies, such as case–control and cross-sectional studies. V: Descriptive studies, including case reports and case series studies. VI: Personal opinion. Recommendation Statement: Keratinocyte death leads to the formation of blisters and erosions. The death of these cells is mediated via different types of regulated cell death processes, including apoptosis and necroptosis. Explanation The culprit drug is recognized as a foreign antigen in individuals with specific human leukocyte antigen (HLA) alleles. Subsequently, activated T cells and natural killer (NK) cells release soluble and cytotoxic molecules, resulting in the death of epidermal keratinocytes. Death of epidermal keratinocytes is mediated by different types of regulated cell death processes, such as apoptosis and necroptosis. Apoptosis is induced by soluble Fas ligand, perforin/granzyme B, and granulysin [1, 2] (Figure 1). Necroptosis is caused by annexin A1 released by monocytes [3]. Monocyte-derived annexin A1 binds to formyl peptide receptor 1 (FPR1), an annexin A1 receptor specifically expressed on SJS/TEN keratinocytes. The de novo FPR1 expression on keratinocytes is induced by an antimicrobial peptide LL-37 derived from neutrophil extracellular traps (NETs) released by activated skin-infiltrating neutrophils [4] (Figure 1). Cytokines, including tumor necrosis factor-alpha (TNF-α), released by cytotoxic T cells and other immune cells also play an important role in inducing death of epidermal keratinocytes [5]. TNF-α not only induces apoptosis and necroptosis in epidermal keratinocytes but also upregulates granulysin mRNA expression in immune cells [6]. The other possible mechanisms underlying the death of epidermal keratinocytes include defective regulatory T-cell function due to concomitant infection and aberrant T-cell activation due to the enhanced production of pro-inflammatory cytokines. Recommendation Statement: For SJS and TEN, the most commonly suspected high-risk drugs include antibiotics and antipyretic analgesics. Explanation In the second national survey of SJS/TEN, we analyzed 315 SJS and 174 TEN cases [7]. In our results, the suspected SJS-associated drugs were antibiotics (19.2%), antipyretic, analgesic, and anti-inflammatory drugs (8.1%), antiepileptic drugs (7.2%), gastrointestinal ulcer treatment drugs (6.9%), and cardiovascular disease treatment drugs (6.6%). The suspected TEN-associated drugs included antibiotics (30%); antipyretic, analgesic, and anti-inflammatory agents (13.5%), cardiovascular drugs (11%); gastrointestinal ulcer treatment drugs (6.7%); and anticancer agents (4.7%). The top three suspected drugs in this survey were the same as in the 2005–2007 survey, with antibiotics and antipyretic, analgesics, and anti-inflammatory drugs accounting for more than one-third of the overall drug-related risk of SJS/TEN [7]. Recently, there have been scattered reports of association of SJS/TEN with novel anticancer agents, molecularly targeted drugs, immune checkpoint inhibitors, and prostate cancer hormone therapy drugs. Given that these new drugs are expected to be increasingly used in clinical practice, dermatologists will need to acquire sufficient knowledge about them. Moreover, dermatologists must actively participate in treatment decisions involving other specialties, such as considering the continuation of treatment in light of an underlying disease. Recommendation Statement: Among the existing diagnostic methods for SJS/TEN, patch tests, and drug-induced lymphocyte stimulation test (DLST, i.e., lymphocyte transformation test, LTT) are relatively useful in identifying the causative drug and are routinely performed in clinical practice. Explanation The second national survey on SJS/TEN [7] helped elucidate the current status of causative drug identification for SJS/TEN in Japan. The patch tests were performed in 13.3% and 13.8% of SJS and TEN cases, with positive rates of 28.6% and 20.8%, respectively. Further breakdown of representative positive cases revealed that four and three cases of SJS arose due to the use of antibiotics and antipyretic, analgesic, and anti-inflammatory agents. In contrast, the use of antibiotics; anticoagulants; antihypertensive agents; expectorants; and antipyretic, analgesic, and anti-inflammatory agents each led to one TEN case [7]. DLST was performed in 67.3% and 66.1% of SJS and TEN cases, with positive rates of 50% and 51.3%, respectively. The use of antipyretic, analgesic, and anti-inflammatory agents, antibiotics, and antiepileptic drugs led to the emergence of 48, 18, and 14 SJS cases, respectively. In contrast, the use of antipyretic, analgesic, and anti-inflammatory agents, antibiotics, and psychiatric drugs led to the emergence of 29, ten, and 6, TEN cases, respectively. These findings indicated that the use of antibiotics and antipyretic, analgesic, and anti-inflammatory agents was more commonly associated with positive cases. Given that DLST was covered by health insurance in 2008 and has become more widely adopted, the implementation rates for patch test and DLST observed in the second survey were lower [8] and higher [7] than the rates observed in the first survey, respectively. Recommendation Statement: Numerous HLA alleles have been reported to be associated with the drug-related risk of developing SJS/TEN. Explanation HLA alleles associated with the risk of developing SJS/TEN have been reported in diverse populations (Table 1) [9-25]. Genetic analyses of Japanese patients with SJS/TEN have identified several HLA alleles associated with the risk of developing SJS/TEN, including HLA-B*58:01 in the antigout drug allopurinol and HLA-B*15:11, HLA-B*51:01, and HLA-A*02:07 in the antiepileptic drugs carbamazepine, phenobarbital, and zonisamide, respectively [13, 14, 17, 19]. Another study reported an association between HLA-A*02:06 and the risk of developing SJS/TEN with ocular complications caused by cold medications [18]. Regarding genetic polymorphisms in drug-metabolizing enzymes that affect systemic drug concentrations, an international meta-analysis involving Taiwanese, Malaysian, and Japanese populations reported that the loss-of-function allele (CYP2C9*3) of the primary metabolizing enzyme CYP2C9 for the antiepileptic drug phenytoin is associated with the risk of developing phenytoin-induced SJS/TEN [21, 22]. Recommendation Statement: Cardiovascular disease, malignant tumors, diabetes, autoimmune diseases, as well as Mycoplasma and human immunodeficiency virus (HIV) infections are risk factors for developing SJS/TEN. Sepsis is a risk factor for mortality in patients with SJS/TEN. Explanation In the second survey, the underlying conditions and mortality risk factors of 315 SJS patients and 174 TEN patients were analyzed [7, 26]. The most prevalent underlying condition in SJS patients was cardiovascular disease, followed by malignant tumors, autoimmune diseases, and Mycoplasma infection [7]. The most common underlying condition in TEN patients was cardiovascular disease, followed by malignant tumors, diabetes, and Mycoplasma infection. Furthermore, the incidence of malignant tumors was higher among SJS/TEN patients than in the general Japanese population [26]. In Germany, the prevalence of systemic lupus erythematosus (SLE) is higher among SJS/TEN patients than in the general population [27]. However, data on the prevalence of SLE in Japanese patients with SJS/TEN remain limited. SLE patients might suffer from TEN-like acute cutaneous lupus erythematosus, which can clinically resemble SJS/TEN, underscoring the importance of careful differential diagnosis [27]. In the United States, the incidence rate of SJS/TEN is 100-fold higher among HIV-infected individuals than in noninfected individuals [28]. Factors suggested to contribute to the higher risk of onset of SJS/TEN in HIV-infected individuals included polypharmacy, immune dysregulation, and mixed infections. Recommendation Statement: Sepsis is a risk factor associated with mortality in patients with SJS/TEN. The main risk factors for sepsis onset include diabetes and systemic conditions requiring intensive care unit-level management. Explanation In the second national survey, patients with SJS and TEN exhibited mortality rates of 4.1% and 29.9%, respectively [7]. The key risk factor for mortality in SJS/TEN patients was sepsis [7, 26]. The mortality rate was higher among SJS/TEN patients who developed sepsis during the 2016–2018 period compared to those who did not develop sepsis [7, 26]. These results were consistent with the findings from previous domestic and international studies [29, 30]. The main risk factors for sepsis included diabetes and systemic conditions requiring intensive care unit-level management [26]. Although not statistically significant, patients receiving systemic corticosteroids before the onset of SJS/TEN tended to exhibit a higher incidence of sepsis [26]. Recommendation Statement: Flu-like symptoms suggestive of microbial infections are often observed in SJS/TEN cases with severe ocular complications, and the administration of medications to treat these symptoms frequently triggers SJS/TEN onset. Explanation Regarding the pathogenesis of SJS/TEN induced by cold medications [31-33], the suppression of prostaglandin E2 (PGE2) production has been suggested to inhibit the anti-inflammatory mechanism mediated by the PGE2–EP3 receptor [31]. In mouse models, PGE2 has been shown to suppress skin and mucosal inflammation, including that on the ocular surface, via the EP3 receptor [34, 35]. In the SJS/TEN patients with severe ocular complications, EP3 protein expression is markedly downregulated in ocular surface tissues [36]. Furthermore, increased levels of certain plasma microRNAs (miRNAs) have been implicated in the dysregulation of innate immune responses [37, 38]. These findings suggested that the abnormalities in the innate immune response to microbial agents might play a role in the development of SJS/TEN with severe ocular complications [39]. Recommendation Statement: Novel treatment options for chronic ocular sequelae in SJS/TEN include specially designed rigid contact lenses and cultured autologous oral mucosal epithelial transplantation. Both are currently covered by health insurance in Japan. Explanation In the acute phase of SJS/TEN, extensive sloughing of the ocular surface epithelium leads to the loss of corneal epithelium, eventually resulting in conjunctival epithelial overgrowth onto the cornea—a condition known as corneal epithelial stem cell deficiency. This condition causes visual impairment due to corneal opacity, irregular astigmatism, and ocular surface adhesions. To address these complications, a limbal rigid contact lens (Sun-con Kyoto-CS) was developed and is currently covered by health insurance in Japan for patients with ocular sequelae of SJS/TEN [40, 41]. This lens, available in 13.0- and 14.0-mm diameters, is worn during the day and removed at night. It fully covers the cornea, improving visual acuity in patients who do not achieve sufficient vision with conventional glasses or contact lenses. Additionally, it alleviates dry eye symptoms and gradually reduces ocular surface inflammation with continued use [42]. In severe cases with ocular sequelae, ocular surface adhesion might progress. In such cases, amniotic membrane transplantation has been attempted but with limited success. Recently, cultured autologous oral mucosal epithelial sheet transplantation (COMET) has shown promising potential in improving severe adhesions [43]. The COMET sheet (Sakuracy) was approved in 2022 and has been covered by health insurance in Japan since 2023. COMET, followed by the use of a limbal rigid contact lens, can improve visual acuity, even in end-stage cases [44, 45]. This contact lens also helps maintain long-term ocular surface stability. In cases where corneal opacity complicates cataract surgery, combining COMET with cataract surgery facilitates the procedure and enhances visual outcomes. Recommendation Statement: In patients who develop SJS/TEN at a young age after receiving non-steroidal anti-inflammatory drugs (NSAIDs), ocular mucosal epithelial damage might worsen even after the treatment has been initiated. Explanation The presence of epithelial defects or pseudomembranes on the ocular surface during the acute phase of SJS/TEN indicates severe ocular involvement and is associated with a high risk of long-term ocular sequelae, including visual impairment and dry eye [32]. Ocular lesions tend to worsen when SJS/TEN is triggered by cold medications or NSAIDs [32]. Additionally, SJS/TEN onset at a younger age is associated with the development of more severe ocular complications [32]. Even with prompt commencement of treatment after diagnosis, the ocular complications might progress [46]. in and young who develop SJS/TEN be of the high risk of ocular surface damage Recommendation Statement: In cases with severe ocular involvement during the acute phase of SJS/TEN, therapy be systemic with for potential infections. Explanation In cases of severe ocular involvement during the acute such as epithelial defects or intensive ocular surface inflammation on the ocular In such cases, be in with systemic with for the development of potential infections Ocular surface inflammation often than systemic anti-inflammatory therapy be more gradually than systemic In a case series involving patients with severe acute ocular involvement but long-term sequelae patients systemic followed by an of of which was based on their clinical corticosteroids were for an of after treatment included eye and eye corticosteroids were from the second were from the levels of in were even after systemic inflammation which was consistent with clinical These findings the need for the use of anti-inflammatory Recommendation Statement: Severe gastrointestinal and impairment Explanation In SJS/TEN, other than the skin and that are reported to be include the and including and have been reported to in of cases In severe cases, patients who develop acute respiratory syndrome often a In a previous study by involving SJS and TEN cases in Japan the prevalence rates of respiratory diseases, including were and in SJS and TEN cases, respectively. In a nationwide survey conducted by the Severe Erythema Multiforme Group of the Ministry of Health, and of cases analyzed developed respiratory therapy was in cases, with cases requiring a which is a of SJS/TEN, has been reported in cases, with respiratory in the disease and symptoms gradually several In severe cases, transplantation be with a is For on the diagnosis and treatment of to the for the of by the Japanese and have each been associated with the onset of new respiratory diseases and exhibit [7]. Among the patients in this study who during the of the disease, and [7]. In one case of drug-induced syndrome was is acute characterized by loss and necrosis of and chronic where leads to the loss of and results in of drug-induced syndrome associated with SJS/TEN are have been only such cases, with studies on cases. Among the cases, transplantation with a prognosis or Among the cases analyzed in the national survey of SJS/TEN damage at the of their of after One case transplantation due to acute and one case drug-induced to cases with levels of at the of the disease, cases with of a with and a mortality rate of at 1 an even The are by caused by a in due to epidermal cell Additionally, drug-induced be caused by or and other The conditions associated with include damage or and factors for the onset of acute include such as sepsis, medications antibiotics, and chronic disease, and is observed in of cases, with and of the cases requiring and long-term respectively Additionally, the of mortality include infections chronic diseases, and Recommendation Statement: administration of systemic corticosteroids and therapy is Explanation In the second survey the of the mortality rate by the for to the actual mortality rate was by each treatment The mortality rate was by therapy followed by therapy the of both were observed between the mortality rate by the and the actual mortality rate for the treatment methods [7]. studies have also indicated that skin compared to care and reduces the actual mortality rate compared to the mortality rate Additionally, reports have suggested that therapy corticosteroids and the prognosis of SJS/TEN patients and the of therapy is Recommendation Statement: In SJS/TEN cases where the disease therapy be considered with the aim of disease Explanation therapy is used in there is a general use in conditions requiring prompt such as respiratory or status administration for SJS/TEN is associated with an increased risk of study has reported a higher mortality rate in these cases compared to cases receiving care case series have reported that therapy is in ocular lesions in SJS In SJS/TEN patients with acute ocular therapy of onset the risk of severe ocular sequelae Furthermore, an epidemiological survey conducted in Japan from 2016 to that and the mortality rates [7]. Regarding systemic administration for TEN, reports have suggested that it the risk of infection and indicated that therapy did not the and was associated with mortality therapy be considered for severe cases, cases, or cases with but severe ocular However, in cases where epidermal has to extensive is as it the risk of infection and inhibit epidermal The treatment for SJS/TEN is shown in Recommendation Statement: the skin symptoms the diagnostic criteria for SJS or TEN and also the diagnostic criteria for it is to as an between SJS/TEN and Although an between the diseases Explanation is a drug characterized by onset and It is caused by a limited of causative drugs. This condition is via clinical and findings of a such as disease high fever, and of human SJS/TEN is via skin there are cases that the diagnostic criteria for also the criteria for such a is to as an In most cases with an and during the of the disease, such as blisters and symptoms of SJS/TEN in the of or develop as skin symptoms during a of The incidence of such is in a national epidemiological survey of conducted in of cases were to with SJS/TEN. The in as or and progress to is characterized by cell in the with but findings severe epidermal damage to SJS/TEN. Additionally, mucosal involvement might in it is as and of the oral and However, in cases, extensive and severe and when they are by of the and conjunctival the criteria for SJS are Ocular involvement is often limited to conjunctival but cases leave sequelae, such as dry eye or The mechanism underlying the the of progress to SJS/TEN, and NSAIDs during treatment or medications for infections the onset of SJS/TEN. Recommendation Statement: has been as a factor for SJS/TEN and be Explanation The most prediction for SJS/TEN that has been widely used to is This was in and several including epidermal concomitant malignant tumors, and levels of and The mortality rate can be based on the Although this is widely used it and is of several including rate and which can recent study suggested that or the mortality rates based on the data from cases identified in the second national survey of SJS/TEN in a new prediction was This factors with high for mortality and one to each The in clinical findings and medical at the including epidermal concomitant malignant tumors, diabetes, cardiovascular diseases, use of antibiotics as the causative mucosal damage in three and and of systemic therapy (Table The of in Japan is with an the of Furthermore, cases also an of that the of was to that of existing prediction is considered a and useful as of are based on the clinical findings and medical However, further of cases is for Recommendation Statement: are cases that are not Explanation Among the cases as SJS/TEN based on epidermal drug cases of epidermal are to due to than from a These cases not be as SJS/TEN. For is a of an and Given that is also expressed in epidermal a induces epidermal damage In severe cases, the clinical might resemble SJS/TEN but the underlying mechanisms from those of SJS/TEN, and it is important to it as SJS/TEN based on epidermal Recently, the of such case reports has been and the SJS/TEN diagnostic criteria have damage caused by the of cytotoxic anticancer agents, not as a condition to be from the findings (Table In cases of skin damage caused by the of cytotoxic agents, systemic administration of corticosteroids or not be therapy oral and be Recommendation Statement: In patients with to severe SJS/TEN or those with of with corticosteroids is as it to the of of the and a in use is of evidence Explanation Based on the evidence that TNF-α is in the pathogenesis of epidermal cell death in SJS/TEN, treatment with the TNF-α has been and reports have been published. randomized controlled in including patients compared the with the to the the a skin of and 14 in and severe cases, respectively Although the was not statistically significant, the mortality rate from to Furthermore, a study reported that the statistically in both mortality and to skin compared with the meta-analysis also suggested that TNF-α inhibitors, including contribute to and mortality in SJS/TEN patients Moreover, a specific clinical of regulated clinical conducted in Japan reported that for cases with response to the use of in a to of compared to previous reports of were only to with with these and other the of skin and the of the are multiple studies, the in to severe cases. Furthermore, a mortality was observed in in severe was there was a for to compared to the the other is currently not the primary approach due to limited data with the the current selection to severe cases with a of or cases that are or at high risk for Furthermore, have not been and the implementation of randomized is that is not approved for this in Japan of it is also not covered by health and are for This is the of the that was in of the Japanese of The have for from the of the Japanese of have their to the of this These guidelines are a of the Japanese for the of Stevens–Johnson and and This study was by the and Research from the Ministry of Health, and Welfare of Japan to an of the of and a of this To was from to the of this for The other of The data that the findings of this study are available from the