Safety of immune checkpoint inhibitors in patients with preexisting autoimmune disorders.

Abstract

e14568 Background: In recent years, immune checkpoint inhibitors (ICIs) have been approved for a growing number of cancer types. Approximately 10-30% of cancer patients have a concomitant diagnosis of autoimmune disorders (AID). Activation of the immune system with ICIs poses a potential risk of AID worsening. The majority of the ICI clinical trials excluded these patients from the study. There is a paucity of data regarding the benefits and risks of ICIs in cancer patients with AIDs. The primary objectives of this study were to determine the incidence of immunotherapy related AID worsening and all immunotherapy related adverse events (irAEs). Secondary outcomes were time to AID worsening and survival difference. Methods: All adult patients (age ≥18 years) with solid malignancy who received ICIs between Jan 2016 to June 2019 were identified using the University of Louisville pharmacy database. Medical records were reviewed to include all the patients with preexisting AIDs. Descriptive statistics were used to determine the incidence of AID worsening and all irAEs. Baseline characteristics were compared between cancer patients with versus without AID worsening using Pearson chi2 and Student’s t-test, where appropriate. Multivariate Cox regression analysis was used to compared survival between the 2 groups. Results: A total of 40 patients with AIDs were identified during the study period. The cancer types were melanoma (57.5%), lung (15%), breast (5%), and others (22.5%). AIDs were rheumatological (52.5%), gastroenterological (12.5%), neurological (12.5%), dermatological (20%) and hematological (2.5%). The incidence of all irAEs was 60% (grade ≥3 in 20%) and AID worsening was 40% (grade ≥3 in 15%). Baseline characteristics of the 2 comparison groups are shown in table. The median time from ICI initiation to AID worsening was 94.5 (range 21-431) days. In multivariate Cox regression analysis, adjusted for demographics, cancer type, and stage, survival was similar for patients who had AID worsening versus those who did not (HR 0.30 (95%CI 0.06-1.40, p=0.128). Conclusions: In our single-institution study, cancer patients with preexisting AID do have an increased risk of irAEs with high-grade toxicities in 20% compared to historically reported 5% in the general cancer population; about 15% of patients reported grade ≥3 worsening of their AIDs. However, though the risk of irAEs is numerically higher in patients with AID, it may be acceptable to the patients if the potential benefits of ICIs outweigh the risks. Baseline demographics and clinical characteristics.[Table: see text]