Cetuximab-related hypersensitivity reactions associated with pre-existing cetuximab-specific IgE antibody

Abstract

9097 Background: Cetuximab is a chimeric (mouse/human) IgG1 monoclonal antibody against the epidermal growth factor receptor and is approved for use in patients (pts) with colorectal cancer and head and neck squamous cell carcinoma. Hypersensitivity reactions caused by cetuximab (C-HSR) have been reported; however, the mechanism underlying these reactions is unknown. We hypothesize that C-HSR are mediated by pre-existing cetuximab-specific IgE antibodies (C-IgE). Methods: A total of 140 serum samples were obtained under IRB approved protocols and retrospectively analyzed across 2 cohorts: 1) 71 pretreatment sera from cetuximab-treated pts collected from multiple centers (47 pts with no HSR and 24 pts with any grade HSR; samples were biased for HSR pts), and 2) 69 sera from healthy volunteers in the Nashville TN area. The samples were analyzed for total-, cetuximab-specific and mouse-specific IgE levels using a modified ImmunoCAP assay (Phadia US Inc.). Severe HSR was defined as grade 3/4 reactions during the first infusion by NCI CTC version 3.0 Allergic reaction/hypersensitivity criteria by a reviewer blinded to the immunoCAP assay results. Results: Of the 71 cetuximab-treated pts from cohort 1, 21 experienced severe HSR by retrospective evaluation. All C-IgE(+) pts (15/15) experienced severe HSR and were immediately discontinued from therapy. Of the remaining 6 severe HSR pts with C-IgE(-), 4 pts were re-challenged and completed their cetuximab infusion without any further reaction, suggestive of a non-IgE mediated mechanism, while 2 pts were not re-challenged. Also, the C- IgE(+) pts tended to have higher levels of total IgE compared to the C-IgE(-) pts. All 47 non-HSR pts were C-IgE(-). Mouse-specific IgE was not detected in any sera from the pts. Analysis of sera from healthy volunteers from the cohort 2 revealed that 15/69 (21.7%) were C-IgE(+), suggestive of pre-existing C-IgE; however, the association with C-HSR could not be made. Conclusion: Our data suggest that C-IgE antibodies are present prior to treatment and appears highly predictive of severe HSR during the first infusion; however, IgE-mediated reactions may not account for all cases of HSR. Prospective validation of the association between C-IgE and cetuximab-induced HSR is warranted. [Table: see text]