Background For anti-TNFα agents, infections&malignancies are key safety concerns of interest. It is unknown whether safety events are driven by peak (C_max), trough (C_min), average (C_avg) conc or by total systemic exposure (AUC). This PK/safety analysis was performed from pooled Ph3 intravenous(IV)/subcutaneous(SC) GLM studies to investigate if there is any relationship of PK exposure to selected safety events.

Objectives To explore the correlation between GLM exposure & occurrence of infections, serious infections(SIs), serious adverse events(SAEs) & malignancies after administration of IV or SC GLM in RA, PsA or AS pts.

Methods 7 Ph3 studies(2 IV&5 SC) in GLM rheum program were incl to leverage data from a large number of pts over a wide range of dosing regimens. Dosing regimens ranged from 2mg/kg GLM Q8W to 4 mg/kg GLM Q12W in IV studies to 50mg GLM Q4W to 100mg GLM Q4W in SC studies, & incl pts receiving GLM as monotherapy or on background of MTX/other non-bio DMARDs. Established pop PK models after SC&IV admin of GLM were used to generate empirical Bayesian estimates of steady-state GLM PK exposure metrics (C_max, C_avg, C_min & Cumulative AUC) for individual pts using their actual dosing records. Infections, SIs&SAEs were evaluated thru approx 1yr after initial GLM exposure. Malignancies were evaluated thru the latest data cut for the IV program(Aug 15, 2012&thru wk160 for SC studies to leverage as much data available as possible due to lower incidence of malignancies vs infections&SAEs. A total of 2486 pts were incl in the PK/safety dataset. For infections, SIs& SAEs, 2 sets of analyses were performed incl a quartile analysis where the prop of pts who experienced infections, SIs or SAEs were assessed by 4PK quartile subgrps. The 2nd analysis incl the distribution of PK exposure vs. number of occurrences. For malignancies, distribution of PK exposure vs occurrence of malignancies were plotted.

Results From the quartile analyses of pooled SC&IV data, as systemic exposures to GLM increased there was no trend of increasing infections, SIs or SAEs regardless of distribution of the number of safety events for all PK exposure metrics. When comparing SC to IV data for SIs, there was no trend of increasing SIs with higher exposures. From distribution plots of combined IV&SC data, it was observed that for Cavg, Cmax, &AUC, ranges of exposures across categories of having 0,1, or >1 infections were similar, while for SAEs&SIs, higher rates of infections trended to occur in pts having lower exposures. A trend toward observation of higher systemic exposure to GLM correlating with more safety events was not observed, contrary to what might be expected. From the distribution plots for malignancies, it was observed that there is no difference in any of the PK exposure metrics for pts who had no malignancies vs.pts who had malignancies.

Conclusions No correlation of the occurrence of safety events with GLM Cmax, Cavg, Cmin & Cumulative AUC was observed with regard to infections, SIs& SAEs over 1yr with SC or IV GLM. There was no correlation of PK GLM exposure with malignancy occurrences up to 3yrs of treatment. The data suggests that IV&SC dosing regimens of GLM evaluated in these Ph3 studies have similar safety profiles.

Disclosure of Interest J. H. Leu Employee of: Janssen R&D, LLC, A. Beutler Employee of: Janssen R&D, LLC, A. Mendelsohn Employee of: Janssen R&D, LLC, S. Liao Employee of: Janssen R&D, LLC, H. Davis Employee of: Janssen R&D, LLC, H. Zhou Employee of: Janssen R&D, LLC, Z. Xu Employee of: Janssen R&D, LLC