Introduction
The ultimate treatment goal in ulcerative colitis (UC) is comprehensive disease control (CDC), a composite of clinical, laboratory, endoscopic, and histologic remission. Previously, we identified subpopulations of responders using individual patient-reported outcomes (PRO)—rectal bleeding (RB) and stool frequency (SF)—in the VARSITY trial1. Symptom response trajectories and clinical, endoscopic, and histologic endpoints were mapped to transcriptomics to investigate the biologic basis of different treatment responses.
Aims & Methods
A multivariate trajectory model2 was used to characterise longitudinal patient trajectories by response.3 Trajectory response groups, as well as response groups defined by clinical remission, endoscopic improvement, and histologic remission, were then integrated with mucosal tissue transcriptomics. Analysis of differential gene expression (DGE) and pathways was conducted at baseline, week 14 and week 52, to distinguish between response groups in patients treated with adalimumab (ADA) and vedolizumab (VDZ). Targeted analysis of 42 molecular signatures derived from prior studies and associated with remission, disease activity, or prognosis was performed.
Results
The analysis included 214 VDZ and 187 ADA patients who had RNA-seq samples. The symptom-based trajectory modelling resulted in super/fast responders (SR), responders (R), partial responders (PR), and incomplete or non-responders (NR). Faster responses were linked to higher rates of CDC at week 52. Transcriptomes were evaluated at baseline and post-treatment time points to identify molecular predictors of distinct PRO response trajectories and response groups based on week 52 study endpoints. DGE results in the NR of VDZ based on study endpoints were linked to fibroinflammatory pathways and cellular types (such as angiogenesis, interferon signalling, ECM interactions, neutrophil degranulation, inflammatory fibroblasts, B cells and myeloid subtypes), whereas in responders were associated with epithelial barrier and metabolic processes at both baseline and post-treatment. Findings were more convergent across response comparisons (i.e. PRO trajectories and study endpoints) for VDZ and ADA treated patients respectively at week 14. Targeted analysis of the molecular signatures revealed a subset linked to TNF and Tl1a resistance, and characterised by fibroblast, myeloid, and lymphoid cell activity, elevated in non-responders to VDZ and ADA.
Conclusion
This analysis extends on prior work showing that endpoint response and symptom response trajectories define biologically different UC patient subpopulations with different prospects to achieve CDC at week 52. These findings add clarity to the transcriptomics analysis by identifying discreet and shared pathways linked to NR to VDZ or ADA.
References
1. Sands BE, et al. N Engl J Med. 2019;381(13):1215-1226.
2. Nagin DS, et al. Stat Methods Med Res. 2018;27(7):2015-2023.
3. Schreiber S, et al. United European Gastroenterol J. 2023;11(suppl 8).
Disclosure
Funding: This study was sponsored by Takeda Pharmaceutical Company Limited.
Maria D. Paraskevopoulou, Saman Farahmand, Ruixue Hou, Jie Cheng, Pratik Panchal, Christian Agboton, Ninfa Candela, Vijay Yajnik, and Paresh Thakker are employees of Takeda Pharmaceuticals and hold stock/options in Takeda Pharmaceutical Company Limited. Parambir Singh Dulai reports consulting fees from Abbvie, Abivax, Adiso, Alimentiv, Bristol Meyer Squibb, Boehringer Ingelheim, Campfield, Celltrion, Genentech, Geneoscopy, Janssen, Lilly, Merck, Pfizer, Sanofi, Shattuck, Takeda and Xencor. Stefan Schreiber reports consultancy and personal fees from AbbVie, Arena Pharmaceuticals, Biogen, Bristol Myers Squibb, Celltrion, Dr Falk Pharma, Eli Lilly, Fresenius Kabi, Galapagos NV, Gilead, Hikma, I-MAB Biopharma, Janssen, Merck Sharp & Dohme, Pfizer, Sanofi, Takeda, and Teva.
