Population modeling of the relationship between TACI Ig exposure and biomarker response in patients with rheumatoid arthritis (RA)
Nestorov, I.1, A. Munafo2, O. Papasouliotis2, J. Visich1 and M. Rogge1
1 ZymoGenetics, Seattle, WA, USA; 2 Serono International, Geneva, Switzerland
Background: TACI‑Ig, a recombinant fusion protein containing the extracellular, ligand-binding portion of the receptor TACI, acts as an inhibitor of BLyS and APRIL, two potent stimulators of normal and autoimmune B-cell maturation, proliferation and survival. The molecule is currently being developed for several indications, including treatment of patients with RA.
Objectives: To develop population PK/PD models of the relationships between circulating levels of IgG, IgM and IgA (markers of biological activity) and the systemic exposure with TACI‑Ig after repeated subcutaneous (SC) doses in patients with RA.
Methods: Data were generated from a Phase 1 placebo-controlled study with six cohorts of RA subjects (total n=73) suffering from active moderate to severe RA who received TACI-Ig either as a single SC dose of 70 (n=6), 210 (n=6) or 630 (n=6) mg TACI-Ig or as a repeated dose two weeks apart of 3x70 (n=9), 3x210 (n=9), or 7x420 (n=19) mg. Exposure was determined by measuring total serum concentrations of TACI‑Ig. As markers of biological activity, circulating levels of immunoglobulins (Ig) A, G and M were measured. Exposure was modeled using a two compartment, first order absorption model with nonlinearities (dose dependence) in the parameters. Biomarker responses (defined as percentage of measured baseline values) were characterized by indirect PD models with TACI-Ig dependent inhibition of immunoglobulin production.
Results: The models fit the clinical data well and the parameter values estimated characterize the inhibitory potential of TACI-Ig for the three Ig types. The modeling results are in line with the observations from the trial indicating that circulating IgM levels were the most reactive to TACI-Ig exposure, followed by IgA and IgG. IgM depletion is characterized by maximum inhibition (Imax) of at least 75-90% of baseline, IgA has an Imax of 50-65%, and IgG – 18-28% of baseline. The inhibitory concentration of total TACI-Ig that achieves 50% of the maximum inhibition (IC50) is similar for all three biomarkers: in the interval between 1.39 mg/L (for IgG), and 2.00 mg/L (for IgA). Based on the percentage of baseline profiles, the correlations between the three biomarkers are high, with correlation coefficients in excess of 0.7.
Conclusion: Owing to the informative design of the study, population PK/PD models on all three biomarkers were developed successfully. The exposure-dependent biological activity of TACI-Ig in RA patients, in line with the assumed mechanism of action, was clearly demonstrated by the quantitative characterization of the existing relationship between TACI-Ig exposure and IgM, IgA and IgG responses provided by the population PK/PD models developed. The latter can be used in exploring various scenarios for the design of dosing regimens to support future studies.