Clinical Features and Outcomes Based on Liver Injury Patterns in Liver Injury Caused by Immune Checkpoint Inhibitors [post]

Takanori Ito, Masatoshi Ishigami, Takafumi Yamamoto, Kazuyuki Mizuno, Kenta Yamamoto, Norihiro Imai, Yoji Ishizu, Takashi Honda, Hiroki Kawashima, Satoshi Yasuda, Hidenori Toyoda, Kenji Yokota (+12 others)
2021 unpublished
BackgroundThe clinical course of liver injury induced by immune checkpoint inhibitors (ICIs) varies among individuals, and there were few reports on the therapeutic effects of corticosteroids based on the patterns of liver injury.MethodsWe evaluated the characteristics and clinical course of immune-related liver injury in 1087 patients treated with ICIs for advanced malignancies between August 2014 and December 2020. ResultsDuring the follow-up period (median, 270 days), 56 patients (5.2%) had
more » ... atients (5.2%) had immune-related liver injury (≥Grade 3). The liver-injury patterns were hepatocellular (n = 25, 44.6%), mixed (n = 10, 17.9%), or cholestatic (n = 21, 37.5%), and the median time to onset of liver injury was 36, 85, and 53 days, respectively; the hepatocellular pattern occurred earlier than the other types (P = 0.036). Corticosteroids were administered to 29 (51.8%) patients. While liver injury was improved in almost all patients with the hepatocellular pattern (n = 13/14, 92.9%), that failed to show improvement in over half of the patients with the non-hepatocellular patterns (mixed, n = 8; cholestatic, n = 7), and three patients with mixed patterns needed secondary immunosuppression with mycophenolate mofetil. Liver biopsies performed in 13 patients mainly showed lobular injury, endothelialitis, and spotty necrosis with infiltration of T cells positive for CD3 and CD8, but not CD4 or CD20. ConclusionThe incidence pattern and therapeutic response to corticosteroids in immune-related liver injury differs according to the injury type. Although corticosteroids were effective for the hepatocellular pattern, an additional strategy for refractory non-hepatocellular patterns is needed.
doi:10.21203/rs.3.rs-483760/v1 fatcat:cbw74f3ml5fy5naalhddh5ikzy