Tjitske van Engelen

265 Towards precision medicine in sepsis Biomarkers Septic patients can have different underlying pathophysiological mechanisms despite similar clinical presentations. Hence, biomarkers can be the key to personalized medicine [35]. Current biomarkers can be divided in diagnostic and prognostic markers: diagnostic biomarkers can discriminate sepsis from non-infectious critical illness, can differentiate between causative organisms, and may help assess severity of disease; prognostic biomarkers can assign risk profiles to patients and predict their outcomes. Key to precision medicine approach is the accuracy of biomarkers to stratify patients into subgroups based on specific pathophysiological features. A biomarker can aid in a theranostic approach to select a specific therapy for an individual patient and simultaneously be used to measure response to treatment [6]. When searching for novel biomarkers it is essential to assess their clinical utility: they should be measurable in easy obtainable samples, such as blood and urine, ideally as a rapid bedside test with limited hands-on time or need for special laboratories, facilitating rapid decision-making [11]. The complex pathophysiology of sepsis makes it unlikely that a single biomarker can provide sufficient information about the derailment of the host response. Instead, a panel of biomarkers may lead to sepsis-signatures needed to develop targeted therapies [36, 37]. Protein and cytokine biomarkers are regarded as ‘traditional’ biomarkers and over 178 have been evaluated in the context of sepsis [38]. Procalcitonin is by far the most studied biomarker, and is the only example that is often implemented as part of sepsis management, to guide duration and doses of antibiotics [39]. Other well studied biomarkers are C-reactive protein (CRP), lipopolysaccharide binding protein, IL-6, soluble triggering receptor expressed on myeloid cells (TREM)-1 and soluble urokinase plasminogen activator receptor (suPAR). None of these is specific enough to be used alone in the management of the critically ill patient. Omics techniques will enable us to stratify patients by pheno- and genotyping, rather than merely by routine parameters, such as age, sex, symptoms and signs of a disease, weight, height, concomitant diseases, medications, medical history, epidemiological and socio-economic context, and patient preference. Rapid development in this field has shifted the focus from traditional to omics biomarkers. Transcriptomics uses ribonucleic acid (RNA) molecules as biomarkers that can be incorporated into bedside tests. Diagnostic RNA biomarkers can already discriminate between infectious and non-infectious disease, such as in SeptiCyte™ LAB, the first rapid molecular assay approved by the US Food and Drug Administration to aid the Intensive Care physician as a diagnostic tool in critically ill patients [40]. The FAIM3:PLAC8 gene expression ratio was developed as a context-specific biomarker, to diagnose community-acquired pneumonia [41]. Transcriptomic biomarkers are also used to discriminate between causative pathogens, such as bacterial or viral, or non-infectious origin [42]; with regards to personalized medicine, these could also be used to stratify patients into subgroups based on their transcriptome of peripheral blood leukocytes. Genome-wide blood gene expression profiles have been used in prospective observational cohort studies aiming to identify biologically relevant molecular endotypes in patients with community-acquired pneumonia-associated sepsis. One study defined two sepsis response signatures (SRS), showing an immunosuppressive type (SRS1) when compared with SRS2 [43]. Another study provided a method for the molecular classification of patients with sepsis into four 11

RkJQdWJsaXNoZXIy MTk4NDMw