Analysis of Immune Checkpoint Inhibitor–Associated Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis Using the FAERS and OpenVigil Databases

Analysis of Immune Checkpoint Inhibitor–Associated Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis Using the FAERS and OpenVigil Databases | 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 Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 12 December 2025 V1 Latest version Share on Analysis of Immune Checkpoint Inhibitor–Associated Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis Using the FAERS and OpenVigil Databases Authors : Ziliang Zheng 0009-0004-2911-7074 and Zhu Shen [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.176554466.68585314/v1 220 views 136 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Background Immune checkpoint inhibitors (ICIs) have transformed cancer therapy but may induce severe immune-related adverse events, including Stevens–Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN). These reactions are rare yet life-threatening, and real-world evidence on ICI-associated SJS/TEN remains limited. Objective This study analyzed reporting characteristics and disproportionality signals of ICI-associated SJS/TEN using the FDA Adverse Event Reporting System (FAERS) and the OpenVigil platform to support clinical risk assessment. Methods FAERS reports from January 2011 to October 2025 were screened for SJS, TEN, and SJS/TEN overlap. Cases involving ICIs (nivolumab, pembrolizumab, atezolizumab, durvalumab, avelumab, ipilimumab, and tremelimumab) were identified, and demographics, outcomes, and drug distributions were summarized. Disproportionality analyses for PD-1, PD-L1, CTLA-4, and combination regimens were performed using ROR, PRR, EBGM, and EBGM05 based on OpenVigil 2.1. Results A total of 829 ICI-associated SJS/TEN cases were identified, predominantly in older adults. Reports increased after 2016. The mortality was 31.2%, highest in TEN (43.0%). Pembrolizumab and nivolumab accounted for most reports and deaths. Strong pharmacovigilance signals were observed for PD-1 inhibitors (ROR 3.821), CTLA-4 inhibitors (ROR 3.677), and PD-1+CTLA-4 therapy (ROR 4.784), while PD-L1 inhibitors showed no significant signal. Several concomitant medications, including sulfamethoxazole/trimethoprim, amoxicillin, and enfortumab vedotin, demonstrated strong additional signals. Conclusions Although rare, ICI-associated SJS/TEN carries high mortality. PD-1 inhibitors, CTLA-4 inhibitors, and their combination regimens show significant safety signals. Concomitant medications may further increase risk. Early recognition and careful evaluation of co-administered drugs are essential to improving clinical safety. Analysis of Immune Checkpoint Inhibitor–Associated Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis Using the FAERS and OpenVigil Databases Running title: Pharmacovigilance of ICI-Related SJS/TEN Ziliang Zheng 1,2 , and Zhu Shen 1,2,* , 1 Guangdong Cardiovascular Institute, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Guangzhou 510080, China, 2 Department of Dermatology, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou 510080, China *Corresponding author: Zhu Shen, MD, PhD, Department of Dermatology, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, No.106, Zhongshan 2nd Road, Guangzhou 510080, China ( [email protected] ). Prior postings and presentations The content of this manuscript has not been previously published or presented. Funding This work was supported in part by National Natural Science Foundation of China (No. 82273537). Abstract Background Immune checkpoint inhibitors (ICIs) have transformed cancer therapy but may induce severe immune-related adverse events, including Stevens–Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN). These reactions are rare yet life-threatening, and real-world evidence on ICI-associated SJS/TEN remains limited. Objective This study analyzed reporting characteristics and disproportionality signals of ICI-associated SJS/TEN using the FDA Adverse Event Reporting System (FAERS) and the OpenVigil platform to support clinical risk assessment. Methods FAERS reports from January 2011 to October 2025 were screened for SJS, TEN, and SJS/TEN overlap. Cases involving ICIs (nivolumab, pembrolizumab, atezolizumab, durvalumab, avelumab, ipilimumab, and tremelimumab) were identified, and demographics, outcomes, and drug distributions were summarized. Disproportionality analyses for PD-1, PD-L1, CTLA-4, and combination regimens were performed using ROR, PRR, EBGM, and EBGM05 based on OpenVigil 2.1. Results A total of 829 ICI-associated SJS/TEN cases were identified, predominantly in older adults. Reports increased after 2016. The mortality was 31.2%, highest in TEN (43.0%). Pembrolizumab and nivolumab accounted for most reports and deaths. Strong pharmacovigilance signals were observed for PD-1 inhibitors (ROR 3.821), CTLA-4 inhibitors (ROR 3.677), and PD-1+CTLA-4 therapy (ROR 4.784), while PD-L1 inhibitors showed no significant signal. Several concomitant medications, including sulfamethoxazole/trimethoprim, amoxicillin, and enfortumab vedotin, demonstrated strong additional signals. Conclusions Although rare, ICI-associated SJS/TEN carries high mortality. PD-1 inhibitors, CTLA-4 inhibitors, and their combination regimens show significant safety signals. Concomitant medications may further increase risk. Early recognition and careful evaluation of co-administered drugs are essential to improving clinical safety. Keywords Immune checkpoint inhibitors; Stevens–Johnson syndrome; Toxic epidermal necrolysis; Pharmacovigilance; FAERS; Disproportionality analysis; Cutaneous toxicity Key Points (1) Analysis of over 800 FAERS reports identified a clear and increasing real-world burden of ICI-associated SJS/TEN, with substantial fatality, particularly among TEN cases. (2) PD-1 inhibitors, CTLA-4 inhibitors, and PD-1+CTLA-4 combination therapy demonstrated strong disproportionality signals, whereas PD-L1 inhibitors showed no significant signal. (3) Pembrolizumab accounted for the largest proportion of both reported and fatal SJS/TEN cases, highlighting notable drug-level differences. (4) Several concomitant medications—especially SMX/TMP, amoxicillin, and enfortumab vedotin—exhibited strong safety signals and may potentiate risk. (5) These findings provide clinically actionable evidence to support early recognition, risk stratification, and safer immunotherapy decision-making. Plain Language Summary Immune checkpoint inhibitors (ICIs) are important cancer treatments, but in rare cases they can cause very serious skin reactions called Stevens–Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN). These conditions can be life-threatening. To better understand how often they occur and which drugs may carry higher risks, we examined reports from a large U.S. safety database containing real-world information submitted by doctors, patients, and drug manufacturers. We identified 829 reports of SJS or TEN linked to ICI treatment. Most patients were older adults, and the number of reports increased as ICIs became widely used. The overall death rate was high, especially among patients with TEN. Two commonly used drugs—pembrolizumab and nivolumab—appeared most frequently in these reports. We also found that ICIs targeting the PD-1 and CTLA-4 pathways, especially when used together, were more strongly linked to these severe reactions than drugs targeting PD-L1. Some antibiotics and chemotherapy drugs given at the same time may further raise risk. Although these reactions are rare, our findings highlight the need for careful monitoring, early recognition of warning signs, and attention to other medicines patients receive during cancer treatment. Introduction SJS/TEN are rare but lethal severe cutaneous adverse reactions (SCARs) characterized by acute onset, extensive epidermal and mucosal detachment, and potential multiorgan involvement. Although uncommon, they progress rapidly, with reported mortality of 1–5% for SJS and 25–35% for TEN 1 , and U.S. inpatient data showing mortality rates of ~4.8%, ~19.4%, and respectively 2 . As ICIs have become standard treatments for multiple malignancies, their ability to enhance antitumor immunity through T-cell activation has also introduced risks of immune-related adverse events, including severe cutaneous reactions such as SJS/TEN 3 . Case reports and small retrospective studies have implicated PD-1/PD-L1 inhibitors and CTLA-4 inhibitors in SJS/TEN, but real-world risk signals and outcomes remain insufficiently defined. Because SJS/TEN are extremely rare, clinical trials are underpowered to detect these events, making spontaneous reporting systems essential complementary tools for identifying potential drug-related risks. FAERS is one of the largest global spontaneous reporting databases and is widely used for pharmacovigilance signal detection. OpenVigil, an analytical platform built on FAERS, improves data quality through deduplication, terminology harmonization, and removal of redundant entries 4,5 . Integrating both systems enables more accurate assessment of ICI–SJS/TEN associations and supports evaluation of patient characteristics, outcomes, and concomitant medication patterns. Accordingly, this study used FAERS and OpenVigil to assess the strength of association between different ICIs and SJS/TEN, compare risks across agents, and characterize clinical features and severity profiles. By generating real-world pharmacovigilance evidence, the study aims to support early recognition of ICI-associated SJS/TEN and promote improved monitoring and intervention to enhance patient safety. Data Sources and Study Design This study employed a retrospective real-world pharmacovigilance design. Data were obtained from the FAERS database (https://www.fda.gov/drugs/surveillance/fdas-adverse-event-reporting-system-faers) and supplemented with signal detection using the OpenVigil 2.1 platform. FAERS is a spontaneous reporting system used for drug safety assessment and signal identification. FAERS Data Retrieval and Extraction In the FAERS database, the search period spanned from the first quarter of 2011 to October 20, 2025. Using the Medical Dictionary for Regulatory Activities (MedDRA) Preferred Terms (PT), we identified reports of “Stevens–Johnson Syndrome,” “Toxic Epidermal Necrolysis,” and “SJS–TEN Overlap,” yielding 23,773 records. After data extraction, standardized preprocessing was conducted, including duplicate removal based on FDA rules, drug-name harmonization, and correction of reporting formats. Identification of ICI-Related Cases Cases involving ICIs were identified from all SJS/TEN reports. The included agents were PD-1 inhibitors (nivolumab, pembrolizumab), PD-L1 inhibitors (atezolizumab, durvalumab, avelumab), and CTLA-4 inhibitors (ipilimumab, tremelimumab). In total, 829 ICI-associated SJS/TEN cases were retrieved for analysis. Description of Adverse Reaction Characteristics Basic information was extracted from all 829 cases. Reports listing both “Stevens–Johnson syndrome” and “toxic epidermal necrolysis” were excluded from phenotype classification. Ultimately, 430 cases were categorized as SJS, 12 as overlap, and 328 as TEN. Variables such as age, sex, report source, indication, concomitant medications, severity outcomes (e.g., hospitalization, life-threatening events, death), and annual reporting trends were then statistically analyzed. Signal Detection Analysis Using OpenVigil To evaluate the strength of association between ICIs and SJS/TEN, signal mining was conducted using OpenVigil 2.1. Based on the standardized, deduplicated FAERS dataset, pharmacovigilance indicators were calculated for five exposure groups: PD-1 inhibitors, PD-L1 inhibitors, CTLA-4 inhibitors, PD-1 plus CTLA-4 combination therapy, and PD-L1 plus CTLA-4 combination therapy. Signal detection employed ROR, PRR, EBGM, and EBGM05. A signal was considered statistically significant if the ROR 95% CI lower limit exceeded 1, PRR ≥ 2 with χ² ≥ 4, or EBGM05 > 1. Results Patient Demographics (Figure 1) Age Distribution (Figure 1A) SJS cases were primarily concentrated in older adults, particularly those aged 60–69 years (31.4%) and 70–79 years (31.7%), with fewer cases in individuals <50 years (6.7%) or ≥80 years (10.9%). TEN showed a similar age distribution, with most cases in the 60–69-year (39.0%) and 70–79-year (36.5%) groups, and only small proportions in those <50 years (4.7%) or ≥80 years (7.6%). Although overlap cases were few, most occurred in patients aged 60–69 years (50.0%) and 50–59 years (41.7%). Overall, ICI-associated SJS/TEN predominantly affected older populations. Country/Region Distribution (Figure 1B) Geographic patterns varied between phenotypes. Most SJS reports originated from Japan (n=224) and the United States (n=84), followed by France (n=29), Germany (n=14), and Taiwan (n=9). TEN reports were most frequently submitted from France (n=111), Japan (n=89), and the United States (n=25), with additional contributions from Germany (n=17) and Canada (n=15). Annual Trends (Figure 1C) Temporal trends showed minimal reporting from 2011–2015, followed by a marked increase after 2016. SJS cases grew from 10 in 2016 to 71 in 2024, while TEN increased from 4 to 70 in the same period. Overlap cases appeared consistently only after 2021 and displayed a gradual upward trend. These patterns parallel the rapid expansion in global ICI utilization. Weight Distribution (Figure 1D) Regarding body weight, SJS and TEN were most common in the 50–79 kg range. For SJS, 21.7%, 26.3%, and 16.4% of cases fell into the 50–59 kg, 60–69 kg, and 70–79 kg categories, respectively; corresponding proportions for TEN were 27.7%, 31.4%, and 20.8%. Although limited in number, 75% of overlap cases occurred in individuals weighing 90–99 kg. Gender Distribution (Figure 1E) Sex distribution was similar between SJS and TEN, with males slightly predominating (59% and 58%, respectively). In contrast, overlap cases were largely female (83.3%), though the small sample size (n=12) limits interpretation. Mortality and Distribution Characteristics of Individual ICI Agents (Figure 2) Mortality The mortality of ICI-related SJS, overlap, and TEN showed a clear severity-dependent gradient. SJS had a mortality rate of 19.5% (84/430), overlap 33.3% (4/12), and TEN the highest at 43.0% (141/328). Overall, 31.2% of all reported cases (259/829) were fatal, underscoring the substantial lethality of ICI-associated severe cutaneous adverse reactions. Mortality Risk by Individual ICI Agents (Figure 2A) Among fatal cases, the contribution of individual ICIs varied substantially. Pembrolizumab showed the highest number and proportion of fatal SJS/TEN reports, with a mortality rate of 38.0% (183/482), followed by nivolumab at 20.7% (57/275). Ipilimumab, atezolizumab, and durvalumab demonstrated intermediate mortality rates of 25.4%, 24.4%, and 22.7%, respectively. Tremelimumab and avelumab each had only one reported case, both fatal, yielding nominal 100% mortality; however, these were excluded from Figure 2 due to the extremely small sample size. Distribution of Individual ICIs (Figure 2B) Distribution across the three clinical phenotypes showed that SJS was most often linked to pembrolizumab (44.4%) and nivolumab (31.1%), followed by ipilimumab (14.1%), atezolizumab (7.1%), durvalumab (2.6%), and tremelimumab (0.8%). TEN displayed a similar pattern, dominated by pembrolizumab (55.9%) and nivolumab (25.9%), with lower proportions for ipilimumab (14.0%) and other PD-L1 inhibitors. Although overlap cases were few, pembrolizumab (53.3%) and nivolumab (26.7%) remained most frequent, with ipilimumab accounting for 20.0%. Overall, pembrolizumab and nivolumab constituted the majority of SJS, TEN, and overlap reports. This predominance most likely reflects their broad indications and extensive clinical utilization rather than inherently greater SJS/TEN-inducing potential, although pembrolizumab consistently showed the highest proportions across fatal and phenotype-specific cases. Pharmacovigilance Signal Analysis Based on OpenVigil (Figure 3) The association between immune checkpoint inhibitors and SJS/TEN varied across drug classes. PD-1 inhibitors showed strong and consistent signals (ROR 3.821; PRR 3.806; EBGM05 3.4769), and CTLA-4 inhibitors demonstrated comparable signal strength (ROR 3.677; PRR 3.663). Combination therapy further amplified risk, with CTLA-4 plus PD-1 inhibitors presenting the highest signal intensity (ROR 4.784; PRR 4.759; EBGM05 4.0710), supported by a substantial number of reports (n=119), indicating an increased likelihood of severe cutaneous toxicity under dual blockade. In contrast, PD-L1 inhibitors showed no significant signal (ROR 1.193; PRR 1.192; EBGM05 <1). Reports for CTLA-4 plus PD-L1 therapy were extremely limited (n=1), resulting in unstable estimates and precluding meaningful interpretation. Overall, PD-1 inhibitors, CTLA-4 inhibitors, and their combination displayed strong pharmacovigilance signals for SJS/TEN, whereas PD-L1 inhibitors did not. These results indicate potential mechanistic or risk differences among ICI classes. Characteristics of Concomitant Suspected Non-ICI Culprit Medications in Nivolumab- and Pembrolizumab-Related Reports (Table 1) Among nivolumab-associated SJS/TEN cases (n=275), sulfamethoxazole/trimethoprim was the most frequent suspected concomitant culprit drug (15 cases, 5.45%) and showed a very strong signal (PRR 10.777; ROR 10.923). Carboplatin (13 cases, 4.73%) and paclitaxel (11 cases, 4.00%) demonstrated moderate associations, while oxaliplatin (10 cases, 3.64%) and fluorouracil (9 cases, 3.27%) showed weaker or borderline signals. Although reported in only 9 cases (3.27%), amoxicillin displayed a markedly high signal (PRR 9.192; ROR 9.297), indicating a potentially substantial added risk when used with nivolumab. In pembrolizumab-associated cases (n=482), carboplatin was most commonly co-reported (143 cases, 29.67%) with a moderate signal. Pemetrexed (86 cases, 17.84%) showed a strong association, and paclitaxel (83 cases, 17.22%) demonstrated a stable moderate signal. Enfortumab vedotin-ejfv (29 cases, 6.02%) exhibited the strongest signal (PRR 17.178; ROR 17.585), representing the most prominent potential trigger. In contrast, lenvatinib (26 cases, 5.39%) showed a low association, likely reflecting indication-related co-use rather than true interaction risk. Overall, the patterns and signal intensities of concomitant non-ICI drugs varied between nivolumab and pembrolizumab. Certain antibiotics (e.g., SMX–TMP, amoxicillin) and antibody–drug conjugates (e.g., enfortumab vedotin) showed strong signals, suggesting they may act as additional triggering factors in ICI-related SJS/TEN. Indications for ICI Use and Common Concomitant Medications (Supplementary Material 1 and 2) Analysis of ICI indications showed that lung cancer was the most common underlying condition in both SJS (136 cases) and TEN (142 cases), followed by renal cancer, gastric cancer, and, to a lesser extent, hepatocellular carcinoma (SJS) and bladder cancer (TEN). Overall, ICI-associated severe cutaneous reactions occurred mainly in patients with lung and renal cancers, likely reflecting underlying disease prevalence and frequent ICI use in these malignancies. Concomitant medication patterns also differed between phenotypes. Among SJS patients (n=430), carboplatin, paclitaxel, folic acid, magnesium oxide, and lansoprazole were the most frequent co-administered drugs. In TEN patients (n=328), acetaminophen, ondansetron, omeprazole, folic acid, and atorvastatin were most common. These differences suggest that SJS cases more often involved chemotherapy and acid-suppressive agents, whereas TEN cases were more frequently associated with analgesics and antiemetics. Such patterns may offer clinically relevant clues for risk assessment and management of ICI-associated SJS/TEN. Discussion This study systematically evaluated the reporting characteristics of ICI-associated SJS/TEN using FAERS and OpenVigil data. The results revealed distinct demographic patterns, showing that SJS, TEN, and SJS/TEN overlap differ in age, sex, and body-weight distributions. These findings offer clinically relevant insights that may support risk identification and early warning in practice. Demographic Characteristics Overall, this study noted differences in age, sex, geography, and body weight among cases. Older adults, males, reports from Japan and France, and individuals weighing 50–79 kg were more common. However, these patterns likely reflect the characteristics of populations receiving ICIs, regional prescribing and reporting practices, and general reporting biases, rather than true demographic risk. Accordingly, these findings should be interpreted with caution. By contrast, the temporal patterns carry greater clinical relevance. Since 2016, global ICI use has expanded rapidly 6 , and FAERS reports of ICI-associated SJS/TEN have increased in parallel, reaching their highest levels in 2023–2024. This rise is more likely driven by the sharp growth in ICI prescribing volume rather than an intrinsic increase in SJS/TEN risk. Nonetheless, the absolute growth in reported cases presents a meaningful challenge for pharmacovigilance. Therefore, the rising number of SJS/TEN reports likely reflects the rapid expansion of ICI use and growing patient exposure rather than an intrinsic increase in risk, underscoring the need for strengthened surveillance of severe immune-related adverse events. Mortality and ICI-Related Risk Analysis This study identified an overall mortality rate of 31.2% (259/829), indicating that ICI-associated SJS/TEN carries a far higher fatality risk than most general irAEs. A clear severity-dependent gradient was observed: TEN had the highest mortality at 43.0%, followed by SJS/TEN overlap at 33.3% and SJS at 19.5%. Published data report mortality of 1–5% for SJS and 25–35% for TEN 1 , and U.S. inpatient cohorts show rates of 4.8%, 19.4%, and 14.8%, respectively 2 . Compared with these historical figures, ICI-related cases appear substantially more lethal, consistent with the profound epidermal necrosis, systemic inflammation, and multiorgan complications characteristic of TEN. The one-third mortality of overlap syndrome also indicates that it is not a mild intermediate phenotype. Mortality varied across ICI agents. Pembrolizumab accounted for the largest number of deaths, with a mortality rate of 38.0% and the highest proportions in both SJS (44.4%) and TEN (55.9%). Nivolumab showed a mortality rate of 20.7%, while ipilimumab, atezolizumab, and durvalumab demonstrated intermediate rates of 25.4%, 24.4%, and 22.7%. Tremelimumab and avelumab each had a single, fatal report. Overall, pembrolizumab was the predominant contributor to ICI-related mortality. These findings highlight the need for heightened clinical vigilance. Early manifestations—such as rash, mucosal involvement, or systemic symptoms—should prompt urgent evaluation for SJS/TEN. When suspected, immediate discontinuation of the ICI and timely initiation of corticosteroids, IVIG, or other interventions are critical to preventing progression and reducing the high mortality risk. Signal Strength and Mechanistic Interpretation This study showed prominent disproportionality signals for both PD-1 and CTLA-4 inhibitors (ROR ~3.8 and ~3.7), suggesting that these agents may heighten keratinocyte apoptosis and cutaneous immune injury through enhanced T-cell activation 7 . This mechanism aligns with the established pathophysiology of SJS/TEN and prior evidence that PD-1/CTLA-4 blockade is more likely to induce cutaneous irAEs 8 . In contrast, PD-L1 inhibitors showed no significant signal (ROR 1.193, 95% CI crossing 1), likely reflecting their more localized immunomodulatory effects 9 and a lower propensity for severe skin toxicities. Although the CTLA-4 + PD-L1 combination did not generate a meaningful signal, the very small number of reports limits interpretation and requires validation in larger datasets. Notably, the CTLA-4 + PD-1 combination showed the strongest signal (ROR 4.784; EBGM05 4.0709), clearly exceeding that of monotherapies. Although dual-checkpoint blockade enhances antitumor efficacy, its cumulative immune activation may substantially elevate the risk of severe irAEs. The strong signal observed here suggests that this regimen may represent a high-risk trigger for ICI-associated SJS/TEN, highlighting the need for closer dermatologic monitoring in patients receiving combination therapy. Overall, these findings demonstrate marked drug-class differences in SJS/TEN risk. PD-1 inhibitors, CTLA-4 inhibitors, and their combination exhibited the strongest associations, whereas PD-L1 inhibitors showed much weaker signals. Clinically, structured assessment of adverse drug reactions remains essential for improving immunotherapy safety and supporting precision risk prediction. Impact of Suspected Concomitant Medications In SJS/TEN reports involving nivolumab (n=275) and pembrolizumab (n=482), several non-ICI concomitant medications showed notable risk signals, suggesting potential additional triggering effects. Highly sensitizing antibiotics (e.g., SMX–TMP, amoxicillin) and chemotherapeutic agents (e.g., paclitaxel, carboplatin) were frequently reported with elevated PRR/ROR values for both PD-1 inhibitors. In the pembrolizumab group, enfortumab vedotin-ejfv demonstrated strong signals, consistent with prior reports linking it to SJS/TEN 10 . Paclitaxel also appeared frequently with PRR/ROR >1, supporting a possible contributory role of chemotherapy. In contrast, lenvatinib showed a reduced signal (PRR 0.429), suggesting that its co-administration may not increase SJS/TEN risk, aligning with clinical data showing manageable toxicity for the lenvatinib–pembrolizumab regimen 11 . Clinically, physicians should remain vigilant regarding these high-risk concomitant medications during ICI therapy. When drug reactions are suspected, structured causality tools—such as the Naranjo scale 12 and the ALDEN algorithm 13 —should be applied to clarify each agent’s contribution and support more refined risk management. Overall, this study highlights several high-risk medications—particularly SMX–TMP, amoxicillin, and enfortumab vedotin—as potential co-factors in ICI-related SJS/TEN. Possible mechanisms include overlap between antibiotic-induced hypersensitivity and ICI-driven T-cell activation 14 , chemotherapy-related impairment of skin integrity 15 , and immune dysregulation from multi-drug interactions. However, limitations of spontaneous reporting systems prevent distinction between true interactions and co-prescription patterns. Future pharmacokinetic, immunologic, and large-scale cohort studies are needed to better define these inter-drug relationships. Concomitant Medication Analysis The analysis of concomitant medications showed distinct patterns between phenotypes. Patients with SJS more frequently received chemotherapeutic agents such as carboplatin and paclitaxel, along with supportive treatments, whereas TEN patients were more often exposed to non-oncologic drugs including acetaminophen, ondansetron, and proton pump inhibitors. These differences suggest that specific medication combinations may influence both the risk and clinical presentation of severe cutaneous reactions. Notably, several agents—such as acetaminophen and proton pump inhibitors—have been previously linked to SJS/TEN 16,17 , providing additional insight for risk assessment and clinical management of ICI-related cases. Limitation This study has several limitations that require cautious interpretation. First, reliance on FAERS and OpenVigil spontaneous reports introduces issues such as underreporting, inconsistent quality, missing data, and potential duplication, which may affect signal estimates. These databases cannot establish causality and serve only to identify potential safety signals. Second, incomplete baseline information—including dermatologic history, genetic predisposition, detailed chemotherapy regimens, and dosing—limits adjustment for confounders. Disease severity was often based on subjective descriptions without standardized SJS/TEN grading, raising the risk of misclassification. Third, analyses of concomitant medications and treatment indications depend on categorical labels such as “suspect” and “concomitant,” which do not reliably differentiate true causal drugs from background therapies and may introduce bias. Variability in international reporting practices further affects comparability. Finally, signal detection using OpenVigil is influenced by report counts; thus, estimates for rare drugs or uncommon combinations may be unstable, and positive signals do not confirm true risk. Additional pharmacovigilance studies, real-world evidence, and mechanistic research are needed to validate these findings. In summary, the identified signals and epidemiologic patterns should be viewed as hypothesis-generating rather than causal, requiring confirmation through prospective and mechanistic studies. Conclusion Based on extensive FAERS and OpenVigil data, this study characterized the epidemiology, mortality, drug associations, and pharmacovigilance signals of ICI-related SJS/TEN. Both PD-1 and CTLA-4 inhibitors showed significant signals, with combination regimens presenting even higher risk. Lung cancer was the most common underlying malignancy, and mortality varied across ICI agents, with TEN exhibiting substantially higher fatality than SJS. Several chemotherapeutic, anti-infective, and supportive medications also appeared frequently, indicating that concomitant drug exposure may further contribute to severe cutaneous reactions during ICI therapy. Overall, although rare, ICI-associated SJS/TEN carries high mortality and potential interaction risks, underscoring the need for vigilant monitoring. Early recognition of high-risk patients and careful evaluation of co-administered drugs may help reduce severe cutaneous toxicity. Additional real-world evidence and mechanistic research are required to further elucidate risk factors and preventive strategies. Ethics Statement This study used publicly available, de-identified data from the FDA Adverse Event Reporting System (FAERS) and the OpenVigil platform. These databases contain no personal identifiers and do not involve interaction with human participants. Therefore, this research was exempt from institutional review board (IRB) approval and did not require informed consent in accordance with relevant regulations and ethical guidelines. Acknowledgments We would like to express our sincere gratitude for the financial support. Reference 1 Harr, T. & French, L. E. Toxic epidermal necrolysis and Stevens-Johnson syndrome. Orphanet journal of rare diseases 5 , 39, doi:10.1186/1750-1172-5-39 (2010).2 Hsu, D. Y., Brieva, J., Silverberg, N. B. & Silverberg, J. I. Morbidity and Mortality of Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis in United States Adults. The Journal of investigative dermatology 136 , 1387-1397, doi:10.1016/j.jid.2016.03.023 (2016).3 Zhu, J. et al. Stevens-Johnson syndrome/toxic epidermal necrolysis in patients treated with immune checkpoint inhibitors: A safety analysis of clinical trials and FDA pharmacovigilance database. EClinicalMedicine 37 , 100951, doi:10.1016/j.eclinm.2021.100951 (2021).4 Fang, H. et al. Exploring the FDA adverse event reporting system to generate hypotheses for monitoring of disease characteristics. Clinical pharmacology and therapeutics 95 , 496-498, doi:10.1038/clpt.2014.17 (2014).5 Böhm, R., Höcker, J., Cascorbi, I. & Herdegen, T. OpenVigil–free eyeballs on AERS pharmacovigilance data. Nature biotechnology 30 , 137-138, doi:10.1038/nbt.2113 (2012).6 Gougis, P. et al. Clinical spectrum and evolution of immune-checkpoint inhibitors toxicities over a decade-a worldwide perspective. EClinicalMedicine 70 , 102536, doi:10.1016/j.eclinm.2024.102536 (2024).7 Chen, C. B. et al. Immune checkpoint inhibitor-induced severe epidermal necrolysis mediated by macrophage-derived CXCL10 and abated by TNF blockade. Nature communications 15 , 10733, doi:10.1038/s41467-024-54180-7 (2024).8 Muntyanu, A., Netchiporouk, E., Gerstein, W., Gniadecki, R. & Litvinov, I. V. Cutaneous Immune-Related Adverse Events (irAEs) to Immune Checkpoint Inhibitors: A Dermatology Perspective on Management. Journal of cutaneous medicine and surgery 25 , 59-76, doi:10.1177/1203475420943260 (2021).9 Zhao, Y., Liu, L. & Weng, L. Comparisons of Underlying Mechanisms, Clinical Efficacy and Safety Between Anti-PD-1 and Anti-PD-L1 Immunotherapy: The State-of-the-Art Review and Future Perspectives. Frontiers in pharmacology 12 , 714483, doi:10.3389/fphar.2021.714483 (2021).10 Nguyen, M. N., Reyes, M. & Jones, S. C. Postmarketing Cases of Enfortumab Vedotin-Associated Skin Reactions Reported as Stevens-Johnson Syndrome or Toxic Epidermal Necrolysis. JAMA dermatology 157 , 1237-1239, doi:10.1001/jamadermatol.2021.3450 (2021).11 Mo, D. C., Luo, P. H., Huang, S. X., Wang, H. L. & Huang, J. F. Safety and efficacy of pembrolizumab plus lenvatinib versus pembrolizumab and lenvatinib monotherapies in cancers: A systematic review. International immunopharmacology 91 , 107281, doi:10.1016/j.intimp.2020.107281 (2021).12 Naranjo, C. A. et al. A method for estimating the probability of adverse drug reactions. Clinical pharmacology and therapeutics 30 , 239-245, doi:10.1038/clpt.1981.154 (1981).13 Sassolas, B. et al. ALDEN, an algorithm for assessment of drug causality in Stevens-Johnson Syndrome and toxic epidermal necrolysis: comparison with case-control analysis. Clinical pharmacology and therapeutics 88 , 60-68, doi:10.1038/clpt.2009.252 (2010).14 Teng, Y. S. & Yu, S. Molecular Mechanisms of Cutaneous Immune-Related Adverse Events (irAEs) Induced by Immune Checkpoint Inhibitors. Current oncology (Toronto, Ont.) 30 , 6805-6819, doi:10.3390/curroncol30070498 (2023).15 X, R. et al. Chemotherapy-induced toxic epidermal necrolysis in a patient with multiple myeloma, a case report and literature review. Frontiers in oncology 13 , 1227448, doi:10.3389/fonc.2023.1227448 (2023).16 Milosavljević, M. N., Pejčić, A. V. & Milosavljević, J. Z. A review of published cases of Stevens-Johnson syndrome and toxic epidermal necrolysis associated with the use of acetaminophen. Cutaneous and ocular toxicology 40 , 280-292, doi:10.1080/15569527.2021.1942896 (2021).17 Salloum, A., Nasr, D. & Maalouf, D. Dermatologic adverse reactions to proton-pump inhibitors: A synthetized review. Journal of cosmetic dermatology 20 , 1073-1079, doi:10.1111/jocd.13763 (2021). Table 1. Top Concomitant Non-ICI Medications Associated with Nivolumab- and Pembrolizumab-Related SJS/TEN Cases in the FAERS Database. Nivolumab(n=275) Sulfamethoxazole\Trimethoprim (29.67%, n=143) 10.777 10.923 Carboplatin (4.73%, n=13) 2.084 2.087 Paclitaxel (4.00%, n=11) 1.885 1.887 Oxaliplatin (3.64%, n=10) 1.195 1.196 Fluorouracil (3.27%, n=9) 1.271 1.271 Amoxicillin (3.27%, n=9) 9.192 9.297 Pembrolizumab(n=482) Carboplatin (29.67%, n=143) 2.084 2.087 Pemetrexed (17.84%, n=86) 3.705 3.719 Paclitaxel (17.22%, n=83) 1.885 1.887 Enfortumab Vedotin-Ejfv (6.02%, n=29) 17.178 17.585 Lenvatinib (5.39%, n=26) 0.429 0.429 Figure Legends Figure 1. Demographic Characteristics of ICI-Associated SJS, SJS/TEN Overlap, and TEN Cases : (A) Age Distribution; (B) Country/Region Distribution; (C) Trend from 2011–2024; (D) Weight Distribution; (E) Gender Distribution. Figure 2. Mortality and Drug-Specific Distributions of ICI-Related SJS, SJS/TEN Overlap, and TEN Cases: (A) Proportion of ICI in Death Cases; (B) Number and Proportion of ICI in Different Disease. Figure 3. Disproportionality Signal Analysis for Different Classes of ICI: ROR/PRR/EBGM(EBGM05). Information & Authors Information Version history V1 Version 1 12 December 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords cutaneous toxicity disproportionality analysis faers immune checkpoint inhibitors pharmacovigilance stevens–johnson syndrome toxic epidermal necrolysis Authors Affiliations Ziliang Zheng 0009-0004-2911-7074 Guangdong Provincial People's Hospital View all articles by this author Zhu Shen [email protected] Guangdong Provincial People's Hospital View all articles by this author Metrics & Citations Metrics Article Usage 220 views 136 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Ziliang Zheng, Zhu Shen. Analysis of Immune Checkpoint Inhibitor–Associated Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis Using the FAERS and OpenVigil Databases. Authorea . 12 December 2025. DOI: https://doi.org/10.22541/au.176554466.68585314/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.176554466.68585314/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:'a008abf25c4f1640',t:'MTc3OTU4NzY1OA=='};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())}}}})();