Three cases of persistent erythema multiforme, two of unknown aetiology and one precipitated by influenza are reported. Lesions were widespread, mostly atypical in appearance and regressed in response to immunosuppressants (systemic steroids and/or azathioprine) or, in one case, to dapsone. One patient developed erythroderma responding eventually to etretinate. Histology in all patients was consistent with the mixed, epidermodermal pattern of erythema multiforme. There were no significant laboratory abnormalities nor marked symptomatology apart from itching. The persistent form appears to belong to the spectrum of erythema multiforme being heterogeneous with respect to inducing stimuli, including viral antigens, neoplastic or inflammatory disease or unknown causes. Whenever it is possible, treatment should be adjusted depending on the causative agent.

Background Checkpoint inhibitors provide an effective approach for the melanoma treatment. They prolong lymphocyte effects, which explains the cytotoxicity underlying immune‐related adverse events (IrAEs). Cutaneous IrAEs affect nearly 40% of PD‐1i and 50% of CTLA4i‐treated patients. Severe cutaneous irAE do not often occur but could be life‐threatening and may persist despite treatment discontinuation. Methods We aimed to investigate cutaneous IrAEs in a cohort of patients treated with ICI across Europe in an effort to characterize the reactions in a real‐world, phase IV, post‐marketing study using a follow‐up questionnaire. Data since November 2016 until March 2021 were obtained from the Melskintox database, a European multicentric biobank dedicated to the follow‐up of melanoma and cutaneous adverse events, supported by EADO. The dermatoses reported were pooled into four categories: inflammatory dermatosis, bullous diseases, drug‐related eruptions and pigmentary diseases. Results Inflammatory benign dermatoses (n = 63) represented the most common group of reactions (52.5%), followed by drug‐related eruptions (n = 24, 20%), pigmentary diseases (n = 23, 19.2%) and bullous diseases (n = 10, 8.3%). Grade II (n = 41, 34.2%) are represented by bullous pemphigoid, eczema, hypodermitis, lichenoid eruption, maculopapular rash, pruritus, psoriasis‐like rash, urticarial eruption and vitiligo. Grade III (n = 18, 15.0%) are represented by bullous pemphigoid, lichenoid eruption and rashes. Grade IV (n = 2, 1.7%) is only represented by bullous disease. Most cutaneous IrAEs led to immunotherapy continuation (n = 95, 88.0%). CR is associated with more severe the cutaneous irAEs. We report an average time‐to‐onset of 208 days and some late‐onset events. Conclusion Our study has characterized the clinical spectrum of cutaneous irAEs, their timing and severity and their relationship with tumour response. Grade I–II cutaneous IrAE are easily managed allowing ongoing anticancer treatment. Severe late‐onset cutaneous irAE are not uncommon. A dermatological follow‐up helps mitigate the risk of life‐threatening adverse events. These findings highlight the importance of oncodermatological involvement in management of patients with melanoma receiving immunotherapy.

No abstract