Y Pang’s scientific contributions

Background Risankizumab (RZB) has been evaluated in subjects with moderately to severely active ulcerative colitis (UC) in the INSPIRE and COMMAND trials. Population pharmacokinetic (PPK) and exposure-response (ER) analyses were conducted at the end of Phase 2b and following Phase 3 to characterize RZB PK and its relationship with clinical efficacy and safety to support Phase 3 induction dose selection, and the final dose recommendations for induction and maintenance treatment. Methods PPK analyses in UC used a previously developed PPK model for Crohn’s disease with additional evaluation of covariates relevant to UC. ER relationships between RZB exposures and efficacy endpoints at the end of induction (Week 12), re-induction (Week 24; exploratory only) and maintenance (Week 52) periods were established using exploratory graphical analyses and logistic regression modeling. ER analyses for safety were performed for key variables through Weeks 12, 24 and 52 using graphical analyses. Results A two-compartment model with first-order absorption for SC administration and first-order elimination adequately described RZB PK. Among the various covariates assessed, body weight, baseline (BL) serum albumin, BL high-sensitivity C-reactive protein, BL fecal calprotectin, sex, advanced therapy IR status, BL pancolitis and corticosteroid use were statistically correlated with RZB clearance but their impact on exposure was not clinically relevant for efficacy. ER analyses for efficacy using Phase 2b induction data showed that while incremental improvement in efficacy was predicted with doses increasing from 1200 to 1800 mg, the magnitude of difference in efficacy was modest especially for endoscopy-driven endpoints. Phase 3 ER analyses for efficacy showed statistically significant trends of exposure-dependent increase in efficacy following induction treatment with 1200 mg IV at Week 0, 4 and 8, with greater response rates observed with higher exposures across all evaluated endpoints at Week 12, in line with the results from Phase 2b. ER analyses for maintenance demonstrated an exposure-dependent increase in efficacy for Week 52 endpoints, with modest improvement in efficacy when increasing the dose from 180 mg to 360 mg, and largely overlapping confidence intervals. ER analyses for safety indicated no apparent exposure-dependent safety events over the induction or maintenance treatment. Conclusion The PPK and ER analyses of RZB in subjects with UC characterized the disposition of RZB and its relationship to efficacy and safety, as well as the lack of impact by intrinsic and extrinsic factors on PK and efficacy. Results supported the final dosing regimen recommendations for the treatment of moderately to severely active UC.

Background In order to support risankizumab dose recommendation for the treatment of Crohn’s disease (CD), population pharmacokinetic (PPK) and exposure-response (ER) analyses were conducted for risankizumab using data from Phases 2 & 3 studies in moderate to severely active CD patients to characterise PK and its relationship with clinical efficacy and safety. Methods PPK analyses used non-linear mixed-effects modelling, leveraging a previously developed PPK model for plaque psoriasis & CD, with additional evaluation of covariates relevant to CD. ER relationships were evaluated for efficacy endpoints (clinical remission/response based on CDAI or stool frequency/abdominal pain, endoscopic endpoints, etc.) at Week 12 (induction), Week 24 (re-induction; exploratory only), and Week 52 (maintenance), using graphical analyses at Week 12, 24 and 52 and logistic regression analyses at Week 12 and 52. ER analyses for safety evaluated key safety variables through Week 12, 24 and 52 using graphical analyses. Results Observed risankizumab concentrations in CD patients were adequately described by the risankizumab PK model. PK in subjects with CD was linear and time-independent, similar to healthy subjects and plaque psoriasis. The majority of the evaluated intrinsic and extrinsic factors showed no statistically significant effect on risankizumab PK, including age, race, country/region, liver function markers, creatinine, baseline CDAI, baseline SES-CD, duration of disease, prior biological therapies, immunosuppressant use, and immunogenicity. Sex, baseline fecal calprotectin, corticosteroid use, baseline creatinine clearance, body weight, and baseline albumin were statistically correlated with risankizumab clearance but their impact on exposure was not clinically relevant for efficacy or safety. ER analyses for induction demonstrated that exposures associated with 600 mg IV at Week 0, 4 and 8 achieved a near maximal response for all evaluated efficacy endpoints at Week 12, with negligible added benefit from 1200 mg IV. ER analyses for maintenance showed higher efficacy response at higher exposures for most of the Week 52 endpoints, particularly the more stringent endoscopic endpoints. ER analyses for safety indicated no apparent relationship between exposure and incidence of any AE, SAE, infection or serious infection over the 12 or 24 weeks of induction or 52 weeks of maintenance treatment. Conclusion PPK and ER analyses of risankizumab in subjects with CD supported the final dose recommendation of 600 mg IV at Week 0, 4, 8 followed by 360 mg SC at Week 12 and Q8W thereafter for the treatment of moderate to severely active CD.

Background Glucocorticoids (GC) are potent drugs used for treating many inflammatory diseases. While GCs are effective in many immune diseases, dose and duration of administration is limited due to significant side effects. Resting immune cells have very little TNF expression on the cell surface and it is only in an activated state that TNF expression is upregulated. Upon immune cell stimulation, TNF is upregulated and although a significant amount of TNF is cleaved from an activated cell, a portion remains on the cell surface. We have observed that anti-TNF antibodies bind to transmembrane TNF (tmTNF) and undergo endocytosis to the lysosome (1). We have developed a stable antibody drug conjugate (ADC), ABBV-3373, that has a proprietary, highly potent, glucocorticoid receptor modulator (GRM) payload linked to an anti-TNF monoclonal antibody (mAb) that is able to deliver the GC payload to activated immune cells. Objectives We hypothesized that a TNF ADC with a GRM payload would be able to deliver increased efficacy through both TNF inhibition and targeted GRM payload delivery to activated immune cells while sparing systemic glucocorticoid side effects. Methods A mouse surrogate TNF GRM ADC was characterized in an acute in vivo contact hypersensitivity model of inflammation (CHS) and in a mouse model of collagen induced arthritis (mCIA). Additionally, the human anti-TNF GRM ADC, ABBV-3373 has been characterized in healthy volunteers. Results In the CHS model the anti-TNF GRM ADC significantly inhibited the inflammatory response with minimal effect on systemic GC biomarkers. In mCIA a single dose of an anti-TNF GRM ADC, administered at disease onset, was able to completely inhibit arthritis for greater than 30 days while an anti-TNF mAb only partially inhibited disease. ABBV-3373, a human anti-TNF GRM ADC with a GC payload, was evaluated in a Phase 1 study in healthy volunteers. ABBV-3373 demonstrated antibody-like PK profile and ABBV-3373 did not impact cortisol levels at predicted efficacious doses while control subjects that received a single oral dose of 10 mg prednisone demonstrated expected decreases in cortisol levels. Conclusion These data suggest that an anti-TNF ADC delivering a GRM payload into activated immune cells may provide improved efficacy in immune mediated diseases, while minimizing systemic side effects associated with standard GC treatment. References [1]Deora, A. et al. MAB s. 2017;9(4):680-695. Disclosure of Interests Bob Stoffel Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Michael McPherson Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Axel Hernandez Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Christian Goess Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Suzanne Mathieu Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Wendy Waegell Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Shaughn Bryant Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Adrian Hobson Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Melanie Ruzek Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Yinuo Pang Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Hartmut Kupper Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Ronilda D’Cunha Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Julie Parmentier Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie, Timothy Radstake Shareholder of: AbbVie, Grant/research support from: AbbVie, Employee of: AbbVie