Identification of the biomarker soluble TREM-1 as a candidate for a personalised medicine approach for nangibotide treatment of septic shock

Abstract Background:One of main new avenue of exploration in critical care medicine is the development of personalized medicine approaches to better determine which therapy provides the optimal risk/benefit ratio for subgroups of septic shock patients presenting with similar clinical syndromes. This paper describes the early clinical development of a personalized medicine approach for a novel therapeutic agent targeting the TREM-1 pathway, nangibotide, in the treatment of septic shock.Methods:Soluble TREM-1 (sTREM-1) is a mechanism-based candidate biomarker with the potential to select patients at high risk of death and who may respond best to an anti-TREM-1 strategy. Validation of a potential cut-off value and choice of surrogate endpoint for phase 2b clinical trial are built on data from a retrospective cohort study of 293 septic shock patients and data from a multicentre prospective phase 2a study of 49 septic shock patients treated either by placebo or nangibotide. Logistic regression modelling is combined with receiver operator characteristic analysis to validate the prognostic potential of sTREM-1. In addition, the potential predictive cut-off is defined by modelling the ability of different values to detect treatment effect.Results:sTREM-1 at baseline in the AdrenOSS-1 cohort is strongly associated with 28day mortality with a standardized odds ratio for mortality of 2.86 (95%CI 1.91-4.27, p<0.0001) and an optimum value for predicting mortality of 408pg/ml with an area under the receiver operator characteristic curve of 0.694. Based on the performance of different cut-off values as predictors of treatment response to nangibotide in the Phase IIa MOT-C-201 trial, a preliminary tentative value of 400pg/ml was selected. In patients with the high sTREM-1 group the mean change of SOFA from baseline to 48 hours is +1.5 points in the AdrenOSS-1 cohort. Conclusions: This study improves our understanding of the behaviour of sTREM-1 as a potential mechanism-based biomarker with the potential to offer both prognostic and predictive enrichment in clinical trials of nangibotide. These data inform the design of the ongoing ASTONISH phase 2b study which intends to validate the proposed cut-off value as a predictive biomarker in septic shock.Trial Registration: AdrenOSS-1 study: clinicaltrials.gov: NCT02393781, MOT-C-201: clinicaltrials.gov: NCT03158948