Nassim Djebli (1), David Carlile (2), Felix Jaminion (1), Johann Laurent (1), Francois Mercier (1), Nicole A Kratochwil (1), Ann-Marie E Bröske (3), Natalie Dimier (2), Cristiano Ferlini (1), Tom Moore (1), Martin Weisser (3), Alexander Phipps (2), Nicolas Frey (1), Peter N Morcos (4)
(1) Roche Innovation Center Basel, Roche Pharma Research and Early Development, Basel, Switzerland, (2) Roche Products Ltd, Welwyn Garden City, United Kingdom, (3) Roche Innovation Center Munich, Roche Pharma Research and Early Development, Penzberg, Germany, (4) Roche Innovation Center New York, Roche Pharma Research and Early Development, New York, NY, USA
Objectives: CD20-TCB (RG6026) is a novel T-cell-engaging bispecific (TCB) antibody. It has a ‘2:1’ molecular format that comprises two fragment antigen binding (Fab) regions that bind CD20 (on B cells) and one Fab region that binds CD3 (on T cells). Versus other bispecific formats, CD20-TCB has the potential for increased tumor antigen avidity, rapid T-cell activation and enhanced tumor cell killing. In NP30179 (NCT03075696), an ongoing Phase I dose-escalation study in relapsed or refractory (R/R) non-Hodgkin lymphoma (NHL) patients, CD20-TCB has shown promising antitumor activity and manageable safety [1]. CD20-TCB population pharmacokinetics (PopPK) and exposure-response (E-R) relationships for efficacy and safety were investigated.
Methods: In NP30179, indolent (i) and aggressive (a) R/R NHL patients received 1000mg obinutuzumab (G) as pre-treatment on Cycle 1 Day −7 to mitigate for any potential cytokine release syndrome (CRS). Patients then received CD20-TCB doses of 0.005–25mg q2w or q3w. A PopPK model was developed in NONMEM v7.4 using spare PK data. The influence of physiologically relevant covariates on CD20-TCB PK variability was then investigated. In order to estimate CD20-TCB receptor occupancy (RO%) in the presence of G concentrations competing for CD20 receptors over time, full G concentration-time profiles were constructed using the G PopPK model [2]. Using logistic regression, E-R relationships between CD20-TCB time-averaged RO% (AvgRO%) up to Cycle 3 Day 1 and objective response rate (ORR) and complete response rate (CRR) were investigated in aNHL patients who reached Cycle 3 Day 1. E-R relationships between CD20-TCB AvgRO% over the first 24 hours and CRS (as defined by Lee et al. 2014 [3]) were investigated in iNHL and aNHL patients combined.
Results: PopPK were analyzed in 167 iNHL and aNHL patients with at least one PK sample. E-R relationships were analyzed in 87 aNHL patients with PK and efficacy data at Cycle 3 Day 1 and in 139 iNHL and aNHL patients with PK and safety data. CD20-TCB PK were best described using a two-compartment PK model with linear clearance, and were found to be influenced by body weight, an effect that was retained when theory-based allometric scaling was used. PK were comparable in patients with iNHL and aNHL. A significant positive E-R relationship between AvgRO% up to Cycle 3 Day 1 and efficacy was observed in aNHL patients (ORR and CRR, p<0.001). ORR and CRR in the highest tertile of AvgRO% (≥0.593%) was 79.3% and 58.6%, respectively, versus 62.1% and 37.9% (0.095–0.593%) and 34.5% and 20.7% (<0.095%) in the lower tertiles. Consistent with the clinical safety profile showing first-dose-dependent CRS, with the majority of events occurring within the first 24 hours [1], a significant positive E-R relationship between CD20-TCB AvgRO% over the first 24 hours and Grade (Gr) ≥2 CRS (p<0.001) was identified. With a 10/16mg (10mg in Cycle 1 and 16mg thereafter) q3w dosing regimen, AvgRO% (median [P10–P90]) up to Cycle 3 was 0.42% (0.24–1.04%); based on the exposure-efficacy model, a CRR at Cycle 3 of 45.1% (38.0–57.0%) was predicted, consistent with the clinical CRR observed in the 10 and 16mg q3w cohorts [1]. With an initial 10mg dose in Cycle 1, AvgRO% over the first 24 hours was 0.61% (0.37–1.17%); based on the exposure-safety model, a Gr ≥2 CRS rate of 29.6% (25.1–36.1%) was anticipated, consistent with the clinical CRS incidence observed in the 10mg q3w cohort [1].
Conclusions: In the current study, we characterized CD20-TCB PopPK and E-R relationships for efficacy and safety. These models are being used to identify CD20-TCB dosing regimens (including fixed and step-up) that maximize efficacy and minimize CRS risk and to inform optimal biological dose selection for use in combination regimens.
Acknowledgments: NP30179 is sponsored by F. Hoffmann-La Roche Ltd. Third party medical writing assistance, under the direction of Nassim Djebli, was provided by Stephanie Lacey of Gardiner-Caldwell Communications and was funded by F. Hoffmann-La Roche Ltd.
References:
[1] Dickinson et al. Hematol Oncol 2019;37(S2):92–93.
[2] Gibiansky et al. CPT Pharmacometrics Syst Pharmacol 2014;3:e144.
[3] Lee et al. Blood. 2014;124(S2):188–195.
Reference: PAGE () Abstr 9392 [www.page-meeting.org/?abstract=9392]
Poster: Oral: Drug/Disease Modelling