Circadian biology of the systemic inflammatory response during critical illness caused by acute pancreatitis

Abstract

Acute pancreatitis is localised inflammation of the pancreas characterised by a sudden onset. It is associated with a wide range of complications, the most severe being premature death. Currently, there are no specific treatments, and the standard of care is supportive management; however, this is limited and nuanced due to its diverse nature. Given the paucity of treatments, alternatives must be pursued to expand the knowledge base, generate new treatment strategies, and ultimately mitigate adverse outcomes.

Recently, the importance of circadian rhythms in the development and progression of disease has come to the forefront of medical research, although this has largely focused on their relationship with mental health conditions and metabolic diseases. This has not been extended to acute pancreatitis. Therefore, this thesis seeks to bridge the respective domains by analysing the molecular profile of two cohorts utilising a multiomics approach.

Through the collection of multiple blood samples, an extensive time series dataset was combined with clinical data to enable the investigation of numerous molecules, including the transcriptome, metabolome, and proteome. For each, harmonic regression was used to define the overall circadian rhythm signature of the two cohorts. Furthermore, models were adjusted for disease severity to determine if severe acute pancreatitis was associated with greater circadian dysregulation than a mild disease course. Circadian dysregulation was defined as a statistically significant loss of rhythmicity in -omics components.

Other statistical methods including clustering, linear mixed effects modelling, and generalized additive models were used to address secondary aims. The latter two were utilised to determine if there was an association between the time of day of acute pancreatitis onset and several adverse outcomes, including in-hospital mortality, admission to critical care, and a composite endpoint of the above termed ‘negative outcome’. The findings concluded that disease onset during the night or morning was significantly associated with poorer outcomes. In contrast, onset during the day was associated with the lowest risk.

This thesis presents novel evidence of the relationship between the circadian rhythms and acute pancreatitis across multiple molecular ‘-omes’. Its overall findings may facilitate further exploration of the unique molecular pathways that link the two domains, whilst encouraging the development of circadian-based therapies.