LAUSR

To the Editor, We read with great interest the article ‘Toxic epidermal necrolysis during camrelizumab treatment for oesophageal squamous cell carcinoma’ by Peng et al., presenting a case of a man who developed toxic epidermal necrolysis (TEN) during treatment with camrelizumab and in whom antinuclear antibodies were detected. The article highlighted the potential relationship between SSA/Ro autoimmune antibodies and TEN after anti-PD1 immunotherapy. We reviewed patients with TEN related to anti-PD1 therapy (two patients treated with pembrolizumab and one patient with camrelizumab) who were hospitalized in the Department of Dermatology, Shanghai Ruijin Hospital between January 2020 and January 2022. Clinical data was summarized in Table 1. In our cohort, all three patients had no history of autoimmune diseases but the serum samples taken at the onset of TEN were positive for anti-SSA antibody. Furthermore, we reviewed another seven cases of TEN triggered by other medications such as non-steroidal anti-inflammatory drugs, allopurinol, oseltamivir and antibiotics in our hospital. Two of them were also positive for anti-SSA antibody. The serum samples were also taken at the onset of TEN and detected by immunoblotting in the same lab. One of the patients had a medical history of rheumatoid arthritis with positive anti-SSA antibody. Anti-SSA/Ro antibody is the most prevalent antibody to extractable nuclear antigen, often identified in patients with Sjögren's syndrome, systemic lupus erythematosus (SLE) and other autoimmune diseases. SSA/Ro autoantigen could be subdivided into two types, Ro-52 antigen (52 kDa) and Ro-60 antigen (60 kDa). Autoantigen Ro-52 is normally located in the cytoplasm in keratinocytes, and moves to the surface of keratinocytes after exposure to UV lights and other stress factors. We hypothesize that tissue damage caused by TEN could lead to the release of large amounts of intracellular Ro-52 and induce strong autoimmune reactions in predisposed individuals. According to the previous literatures, only two reports mentioned the relationship between TEN and antiSSA antibody. In one report from a large cohort study including 123 patients with Steven–Johnsons syndrome (SJS), SJS/TEN overlap or TEN, five cases were positive for anti-SSA antibody during the surveillance after TEN and developed Sjögren's syndrome or SLE as long term sequelae. The other report presented one case with asymptomatically anti-SSA positivity at the onset of TEN, who developed severe pulmonary complication 1 month after TEN onset and died 1 year later. In addition, anti-PD1 therapy may disrupt the balance of the immune system and induce autoimmunity. Le Burel et al. reported that anti-PD1 therapy triggered Sjögren's Syndrome in two patients with anti-SSA positivity at baseline. Thus, the relationship among anti-SSA antibody TEN and anti-PD1 therapy needs further exploration. However, different with what Peng et al. reported, the intervals between drug exposure and the onset of cutaneous eruptions were longer and the outcomes were better in our retrospective observation. In TEN, the typical latency before eruption is 4 days to 4 weeks of first use of drug. However, in the initial reports of SJS or TEN related to anti-PD1 therapy, the exposure time seemed to be longer (6 weeks, 77 days and 140 days). Recently in one multi-centre case