Tumor flare reactions secondary to T-cell engaging immunotherapies: A study from the French REISAMIC registry.

Abstract

2645 Background: Tumour flare reactions (TFR) were recently reported in patients receiving T-cell engagers (TCE) and require further investigations. This study aims to investigate incidence, predictive factors, and outcomes of pts with TFRs related to TCE. Methods: This observational cohort study is nested in the French academic pharmacovigilance register, Registre des Effets Indésirables Sévères des Anticorps Monoclonaux Immunomodulateurs en Cancérologie (REISAMIC, CNIL number 2098694v0). All patients treated at Gustave Roussy (France) with TCE, for all tumor indications except acute leukemia, were included. The main objectives were to determine the incidence, predictive factors, associated clinical and biomarker profiles, as well as the outcomes of patients. TFRs were clinically defined with transient worsening of tumor symptoms including tumor pain, effusions, or compressive symptoms, mimicking tumor growth but without reflecting disease progression. Statistical analyses included log-rank tests and Cox regression models. Results: Overall, 222 TCE-treated patients were included, median [range] age: 53 [5 - 87] years; male-to-female ratio: 1.44, median prior lines of therapies was 3 [2-9]. Of them, 147 (66.2%) had solid tumors, including prostate cancer (n=51), high-grade serous ovarian carcinoma (n=28), and small-cell lung cancer (n=24), and 75 (33.8%) had hematologic malignancies, primarily multiple myeloma (MM, n=35) and diffuse large B-cell lymphoma (DLBCL, n=31). Common TCE targets included CD3/CD20 (n=40), CD3/KLK2 (n=36), CD3/BCMA (n=35), CD3/DLL3 (n=28), and CD3/B7-H4 (n=27). TFRs occurred in 54 pts (24.3%), with higher TFR frequency in solid tumors (34.0%) than lymphomas (12.1%) and none in MM. Median TFR onset was 1 day [1–8]. Symptoms of TFRs were pain (92.6%), compression syndrome (22.2%), and effusion (7.4%). Severity of TFRs was grade 3–4 in 68.5% of cases. No TFR-related deaths occurred. TFRs management included corticosteroids (35.2%), opioids (61.1%), paracentesis (7.4%), JJ stenting (3.7%) and tocilizumab (3.7%). TFRs correlated with transient increases of CRP (182 vs. 69 mg/L; p=0.0002) and LDH (316 vs. 225 U/L; p=0.04). Patients with TFRs were more frequently exposed to cytokine release syndrome (p=0.0001). Predictors included tumor serous localization (p=0.052) and a higher CD4+/CD8+ ratio in blood (p=0.048). In solid tumors pts; TFRs were associated with higher response rates (18.0% vs. 6.3%; p=0.013) and disease control rates (72.0% vs. 49.5%); PFS (2.83 vs. 2.76 months; p=0.865) and OS (9.92 vs. 9.72 months; p=0.539) were comparable regardless of TFRs. Conclusions: TFRs related to TCE are clinically significant adverse events, primarily observed in solid tumor pts. TFRs may indicate a unique pattern of antitumor response distinct from progression. Better recognition and management of TFRs should help to optimize tolerability of TCE therapies.