320 Background: MEDI-565, a bispecific single-chain antibody, targets human CEA on tumor cells and the CD3 epsilon subunit of the human T-cell receptor complex. In murine models, MEDI-565 showed antitumor activity in CEA-expressing tumors (J Immunother 2009;34:341-52). Methods: This phase 1, multicenter, open-label, dose-escalation study enrolled adults with GI adenocarcinomas (including esophageal, gastric, small intestine, colorectal, biliary tract, and pancreatic). MEDI-565 was given intravenously over 3 h on days 1–5 in 28-day cycles, with 4 single-patient (pt) (0.75–20 μg) and 5 standard 3+3 escalation (60 μg–3 mg; 1.5–7.5 mg with dexamethasone [dex]) cohorts. Primary objective was to determine the maximum tolerated dose (MTD); secondary objectives were to evaluate pharmacokinetics (PK), antidrug antibody (ADA), and antitumor activity. Results: Study enrolled39 pts: mean age 59 y; 56% male; 28 (72%) colorectal, 6 (15%) pancreatic, 5 (13%) other. Dose-limiting toxicities (grade ≥ 3 nonhematologic) were seen in 4 pts (2 at 3-mg; 2 at 7.5-mg + dex): hypoxia (n = 2), diarrhea, and cytokine release syndrome (CRS). Grade 3 treatment-related adverse events (AEs) seen in 5 pts: diarrhea, CRS, increased alanine aminotransferase, hypertension (all n = 1), and hypoxia (n = 2). Treatment-related serious AEs seen in 6 pts: diarrhea, vomiting, pyrexia, CRS (all n = 1), and hypoxia (n = 2). Five pts discontinued treatment due to AEs: diarrhea, CRS, central nervous system metastases, and hypoxia (n = 2). MEDI-565 exposures increased in approximately dose-proportional manner, with clearance (35–77 L/d) and half-life (2–7 h) typical of drug class. ADA had minor impact; 19 pts (48.7%) had ADAs, 5/39 (12.8%) with high titer, with decreased MEDI-565 concentrations in 2 pts. Plasma inflammatory cytokines were elevated posttreatment in several pts at 1.5- and 3-mg (no dex) dose levels. No objective responses were observed; 11 (28%) pts had stable disease as best response. Conclusions: The MTD of MEDI-565 in pts with GI adenocarcinomas was 5 mg with dex. PK was linear, with fast clearance and short half-life. No objective responses were observed. Clinical trial information: NCT01284231.
Development of a recombinant human IL-15·sIL-15Rα/Fc superagonist with improved half-life and its antitumor activity alone or in combination with PD-1 blockade in mouse model
