Abstract 3855
Background
CEA-CD3 TCB (RG7802, RO6958688) is a novel T-cell bispecific antibody targeting CEA on tumor cells and CD3e on T cells. In mouse models, CEA-CD3 TCB displays potent anti-tumor activity, leads to increased intratumoral T cell infiltration and activation and up-regulates the PD-L1/PD-1 pathway.
Methods
Biodistribution was assessed in mice using SPECT/CT. Patient (pt) samples were from 2 dose-escalation studies in CEA-positive solid tumors. Study 1 (S1): single agent weekly (qw) (0.052-600 mg IV; n = 80), and Study 2 (S2): CEA-CD3 TCB qw (5-160 mg IV) plus atezolizumab 1200 mg q3w (n = 46). Analytics: [CEA-CD3 TCB]—bifunctional PK assay; antidrug antibodies—ELISA; immunophenotyping in peripheral blood (PB)—flow cytometry (FCM), in baseline (BSL) and on-treatment (OT; week 7) biopsies by immunohistochemistry and FCM; plasma cytokines—multiplex assay; PD-L1—SP142 assay.
Results
In mice, CEA-CD3 TCB preferentially accumulated in CEA-positive tumors. CEA-CD3 TCB showed near-linear PK in both studies (S1: 35; S2: 28). In pts with matched BSL and OT biopsies, 7/10 CRC pts treated with ≥ 60 mg of CEA-CD3 TCB in S1 had > 2.4-fold increase in CD8 T cells, which did not correlate with RECIST response. In S2, 2/2 CRC pts receiving ≥ 80 mg of CEA-CD3 TCB (with RECIST reduction ≥ 25%), showed > 8-fold increase in CD8/Ki67 T cells. SUVmax decrease (FDG-PET) correlated with BSL levels of CD4-OX40 and CD8-PD1 in S1 and CD8-OX40 in S2. In PB at week 4, a > 4-fold expansion of activated CD8 T cells (HLA-DR/Ki67) but not CD4, was detected in most pts at doses ≥ 60 mg (S1: 24; S2: 9). In most pts, increases in IL-6 were seen after the first TCB infusion and in fewer cases after the second/third infusion in both studies (S1: 62; S2: 33).
Conclusions
On-treatment increases in intratumoral CD8 T cells consistent with the mechanism of action and support that CEA-CD3 TCB is the first tumor-targeted T cell bispecific showing biological activity. The activation level of intratumoral T cells at BSL could be a predictive biomarker of response. In preclinical models, tumor targeting has been demonstrated. Updated data will be presented. Clinical data are reported separately.
Clinical trial identification
NCT02324257
Legal entity responsible for the study
F Hoffmann-La Roche Ltd.
Funding
F Hoffmann-La Roche Ltd.
Disclosure
I. Melero: Advisory board: Bristo-Myers Squibb, Roche-Genentech, AstraZeneca, Lilly, Merck Serono, Bayer, Genmab, Alligator, Bioncotech, Tusk Grants from: Roche-Genetech, Bristo-Myers Squibb, Bioncotech. N.H. Segal: Consulting and research funds from Genentech/Roche. J. Saro: Employee of Roche and stock holder of Roche. A. Marabelle: PI: Roche, Bristo-Myers Squibb, Merck, Pfizer, Lytix pharma, Eisai, AstraZeneca/Medimmune Scientific Consulting: Roche, Pierre Fabre, Onxeo, EISAI, Bayer, Genticel, Rigontec, Daichii Sankyo, Imaxio, Sanofi, BioNTech. J.M. Cleary: Research funding to the institution from Merrimack Pharmaceuticals, Taiho Oncology, Merck, Roche, Abbvie, Precision Biologics, and Bristo-Myers Squibb. H.I. Hurwitz: Honoraria: Roche and Lilly. Consultant: Roche, Bristo-Myers Squibb, Lilly, Novartis, Incyte, TRACON Pharma, Acceleron Pharma, GlaxoSmithKline, OncoMed. Institutional Funding: Roche, GlaxoSmithKline, Novartis, TRACON Pharma, Bristo-Myers Squibb, Regeneron, Lilly, Macrogeneics, NCI. C. Jamois, S. Bouseida, F. Sandoval, V. Karanikas: Roche employee. E. Andersson: RICM participation in the Roche Connect program. M. Bacac: Employed by Roche and own stock options. T. Nayak: Roche stocks. All other authors have declared no conflicts of interest.