Nathan Constantine-Cooke

# Nathan Constantine-Cooke

Affiliations: MRC Human Genetics Unit & Centre for Genomic and Experimental Medicine

I am a Precision Medicine PhD student specialising in modelling biomarker trajectories and disease outcomes for inflammatory bowel disease. I am in the Vallejos group and collaborate closely with clinical colleagues in the Lees group.

My research interests include survival analysis, electronic health records, Bayesian statistics, and machine learning.

I am also an R package developer. My packages can be found on GitHub.

In my personal life, I enjoy swimming, the gym, playing bass guitar, and all things technology.

## News

Aug 17, 2022 Latent Crohn’s Disease Subgroups are Identified by Longitudinal Faecal Calprotectin Profiles is available on medRxiv! gameR has been released on CRAN! This website has gone live! datefixR has been released on CRAN!

## Selected publications

1. TAG
Real-world experience with tofacitinib in ulcerative colitis: a systematic review and meta-analysis
Therapeutic Advances in Gastroenterology Dec 2021

Background and aims: Tofacitinib is a Janus kinase inhibitor (JAKi) recently approved for the treatment of moderate to severe ulcerative colitis (UC) based on robust efficacy and safety data derived from OCTAVE clinical trials. Evidence on the outcomes of tofacitinib therapy in real-world UC patients is needed, as a number of these patients would be deemed ineligible for clinical trials. We have therefore summarised data derived from observational, real-world evidence (RWE) studies on the effectiveness and safety of tofacitinib in moderate to severe UC patients. Methods: We searched the PubMed, EMBASE, Scopus, Web of Science and Cochrane databases for observational studies on the use of tofacitinib in UC patients, published between 30 May 2018 and 24 January 2021. Pooled induction (8-14 weeks) and maintenance (16-26 weeks) clinical response and remission rates were calculated, as well as the proportion of reported adverse events using random effects models. Results: Nine studies were included, comprising 830 patients, of which 81% were previously treated with anti-tumour necrosis factor (TNF) and 57% with vedolizumab. Induction of clinical response and remission were achieved in 51% (95% confidence interval, 41-60%) and 37% (26-45%) of patients, after a median follow-up of 8 weeks. At the end of a median follow-up of 24 weeks, maintenance of clinical response and remission were met in 40% (31-50%) and 29% (23-36%) of patients, respectively. Thirty-two percent of the patients had at least one adverse event, the most commonly reported being mild infection (13%) and worsening of UC, requiring colectomy (13%). A third of the patients (35%) discontinued tofacitinib, most frequently due to primary non-response (51%). Conclusion: Tofacitinib is a safe and effective therapy in real-world UC patients, as previously reported by clinical trials.

2. medRxiv
Longitudinal Faecal Calprotectin Profiles Characterise Disease Course Heterogeneity in Crohn’s Disease
Oct 2022

Background and Aims: The progressive nature of Crohn’s disease is highly variable and hard to predict. In addition, symptoms correlate poorly with mucosal inflammation. There is therefore an urgent need to better characterise the heterogeneity of disease trajectories in Crohn’s disease by utilising objective markers of inflammation. We aimed to better understand this heterogeneity by clustering Crohn’s disease patients with similar longitudinal faecal calprotectin profiles. Methods Latent class mixed models were used to model faecal calprotectin trajectories within five years of diagnosis and to cluster subjects. Information criteria, alluvial plots, and cluster trajectories were used to decide the optimal number of clusters. Chi-squared, Fisher’s exact test, and ANOVA were used to test for associations with variables commonly assessed at diagnosis. Results: Our study cohort comprised of 365 patients with newly diagnosed Crohn’s disease and 2856 faecal calprotectin measurements taken within five years of diagnosis (median 7 per subject). Four distinct clusters were identified by characteristic calprotectin profiles: a cluster with consistently high faecal calprotectin and three clusters characterised by different downward longitudinal trends. Cluster membership was significantly associated with smoking (p = 0.015), upper gastrointestinal involvement (p < 0.001), and early biologic therapy (p < 0.001). Conclusions: Our analysis demonstrates a novel approach to characterising the heterogeneity of Crohn’s disease by using faecal calprotectin. The group profiles do not simply reflect different treatment regimes and do not mirror classical disease progression endpoints. We believe these profiles represent an entirely new way of classifying disease behaviour in Crohn’s disease.