Molecular Early Sepsis Identification Study

Background Sepsis has recently been redefined (Sepsis 3) as an infection leading to organ dysfunction (1). Although this definition is not generally accepted by all researchers (2,3) due to the potential for delayed diagnosis, mortality related to this entity could be significantly reduced by early adoption of diagnostic protocols and appropriate and early use of antimicrobial therapy (ATB).

A recent study in Catalonia (4) showed an incidence of 212 cases of sepsis per 100,000 inhabitants/year with a hospital mortality of 21%, which is significantly higher if consider ICU patients (40%) or those affected by septic shock (>50%) (5,6). These data highlight the current importance of this entity and highlight the need to seek new alternatives for early and appropriate diagnosis and treatment to have a favourable impact on the high mortality rate.

Ineffective treatment of sepsis is often due to late diagnosis of sepsis, due to the inherent complication of nonspecific clinical signs. Although procalcitonin (PCT) appears to be an adequate biomarker for the diagnosis of bacterial infection (7,8), different studies show a discrimination of only around 80% (7,8) and recent international guidelines do not recommend its use (9,10,11). These same guidelines recommend the use of scores, such as SOFA (sequential organ failure assessment) and qSOFA (a simplified version) based on the presence of altered consciousness, hypotension, and tachypnoea) for early risk stratification of patients suspected of infection.

However, sepsis and non-sepsis systemic inflammatory syndromes present as clinically similar entities, and both are frequently complicated by the appearance of organ dysfunction. Therefore, the identification of sepsis remains subjective and possibly leads to overdiagnosis of sepsis and consequently a high use of ATB in patients who do not have sepsis and who do not benefit from this treatment (12).

The availability of new molecular techniques and the automation of these laboratory analyses have stimulated the search for biomarkers related to the host response to sepsis. Analysis of host gene expression through RNA transcripts (transcriptomics) offers an opportunity in this complex setting of the critically ill patient. Host gene expression differs greatly between septic patients and healthy individuals (13). However, although gene expression between septic and non-septic inflamed patients overlaps to a large extent, several differences between transcripts observed in sepsis can help distinguish infection from other causes of inflammation [14]. The SeptiCyte™ LAB transcription assay was developed to diagnose infection in critically ill patients with suspected sepsis [15]. It is the first host response assay based on quantitative reverse transcriptase PCR (qRT-PCR) methods to be approved by the US Food and Drug Administration (FDA). SeptiCyte™ LAB allows the assessment of host response to infection by measuring the expression of specific genes involved in immune function and inflammatory signaling in whole blood. The first clinical performance study of SeptiCyte™ LAB analyzed 345 patients who were enrolled as part of the Molecular Diagnosis and Sepsis Risk Stratification (MARS) cohort in two ICUs in the Netherlands [15]. Patients were categorized as having sepsis based on a post-hoc medical assessment of the available clinical, radiological, and microbiological evidence [16] resulting in estimates for the AUC (0.77 to 0.99), sensitivity (79% to 100%) and specificity (33% to 91%) varied widely, depending on the prevalence of sepsis, the confidence level with respect to the reference diagnosis and whether patient registration was consecutive. As this analysis took about 6 hours, an early sepsis diagnostic tool (SeptiCyte RAPID (17) has been developed that is based on the measurement of two of the four gene expression biomarkers in peripheral blood (PLAC8 and PLA2G7) using the same quantitative reverse transcription polymerase chain reaction (RT-qPCR) and correlates closely with the clinical results of SeptiCyte LAB, but with a much simpler workflow. Based on the results, the device issues a report with a sepsis probability scale. (see attached algorithm) This new analysis methodology, together with the usual biomarkers, can improve the early diagnosis of sepsis, differentiate between inflamed and septic patients.

The FilmArray® blood culture identification (BCID®) assay is a molecular test approved for direct identification of BSI causing pathogens from positive BC (18). A recently updated version of the panel (BCID2®) comprises improved species identification characteristics and allows for the detection of one expanded-spectrum β-lactamase (ESBL)- and several carbapenemase-encoding genes. The BCID2® Panel tests for 43 targets associated with bloodstream infections, including gram-negative bacteria, gram-positive bacteria, yeast, and 10 antimicrobial resistance genes-all with one test and with results available in about an hour from positive blood culture. In addition, BCID2® Panel menu includes seven additional resistance genes, including carbapenemases, colistin resistance genes and a MREJ assay allows for more specific MRSA identification. A recent study (19) evaluates the clinical performance of the BCID2 assay for species identification in 180 positive blood cultures (BCs). BCID2 results were concordant with the standard of care (SOC) in 159/180 (88.3%) BCs; 68/74 (91.9%) and 71/74 (96.0%) of all samples growing monobacterial, Gram-positive or Gram-negative pathogens, respectively, were identified, in agreement with SOC results.

The usual implementation of the BCID2® is done once the blood culture is positive. However, some preliminary results (20) suggest that the performance of BCID2® in detecting microorganisms very early, may be adequate with only a few hours (6-8h) of blood culture incubation and before the blood culture becomes positive.

Since early diagnosis and early administration of effective antibiotic therapy is of crucial importance to improve patient outcome, the aim of the present study is to develop and evaluate a clinical protocol that combines molecular diagnosis of infection and early determination of microorganisms in blood cultures with only 6-8 hours of incubation before they are found to be positive.

Endpoints

The primary endpoints:

1. - To determine the ability of filmarrays BCID 2® for the early identification of microorganisms in BC with a few hours of incubation and before they have become positive.
2. - To determine the ability of SeptiCyte® for the molecular diagnosis of sepsis in critically ill patients with unclear infectious focus.

The secondary endpoints:

1. - To determine the possible impact of the results on the management of antimicrobial treatment.
2. - To determine if there is an association between common serum biomarkers (PCT, CRP) and positive SeptiCyte® and BCID 2® results.

Main inclusion/exclusion criteria

Inclusion Criteria:

1. - Adults (>16 years old)
2. - At least 2 SIRS criteria
3. - RCP (> 10 mg/dL) or PCT (> 0.5 ng/mL in community or > 1.0ng/mL in nosocomial infection) elevated

3.- Without a clear focus that justifies sepsis 4.- High suspicion of infection that requires starting antimicrobial treatment.

Exclusion criteria:

1. - Failure to obtain the samples within 12 hours after the infection is suspected and antibiotic administration
2. - Previous administration of antimicrobials in the last 24 hours
3. - Documented or evident sepsis focus on ICU admission

Procedures:

MOLECULAR DIAGNOSTIC PROTOCOL AND EARLY DIRECTED TREATMENT (See attached Schedule of Assessments) Once the usual blood samples have been obtained for the determination of PCR, PCT, white blood cell count (WBC) and blood cultures (4 bottles), as well as clinically significant cultures, and if the inclusion criteria are met, the blood cultures (4 bottles) will be sent together with a EDTA tube from those obtained for the usual analysis to the microbiology department where the determination will be carried out with SeptiCyte® and BCID 2® (see attached algorithm).

1. SeptiCyte RAPID® will be performed on all patients at the time of sample receipt in microbiology department. The test result will be stored in the study CRF. This result will not be known to the patient's treating physician.
2. BioFire® FilmArray® blood culture identification panel (BCID 2®) will be performing on the 4 blood culture bottles after 6-8 hours of incubation and especially before it becomes positive. The test result will be stored in the study CRF. This result will not be known to the patient's treating physician. A new BCID 2 determination shall be performed once the blood culture is positive, and the result shall be compared with the one obtained previously with 6-8 hours of incubation.

NOTE: In no case will the information received be used to define behaviours or treatments related to the status of the patient Follow-up Period Patients will be followed up until the event that originated the disorder is considered resolved. Only the first episode of possible sepsis will be considered for each patient.

Patients will be classified as confirmed, probable or possible sepsis according to the definitions from "The International Sepsis Forum Consensus Conference on Definitions of Infection in the Intensive Care Unit. (Calandra T et al. Crit Care Med 2005; 33:1538-1548).

Sample's size calculation According to what was published by Klein Klouwenberg (12), only 25% of patients with suspected of sepsis and who receive antibiotics upon admission to the ICU, are finally classified as having sepsis definitive. The study hopes to be able to identify at least 40% of sepsis in patients with suspected sepsis. Based on this difference and for a power of 90% with an error level of 5%, the number of patients needed to include is 60. Considering a possible loss of 20% of patients (17 patients) the number rises to 100 patients. To ensure the number of patients and the power of the study, it was decided to include a further 20%, making a total of 120 patients.

Data analysis plan:

Qualitative variables will be expressed as percentages, while quantitative variables will be expressed as mean and standard deviation (SD) or median and interquartile range 25-75%. To determine clinical differences between groups with and without BCID 2 and SeptiCyte positives, Chi-square and Fisher tests will be used for categorical variables, and Student's t-test or Mann-Withney U-test for quantitative variables. Multivariate analysis by binary logistic regression will be performed to determine the risk factors associated with the early presence of microorganisms in BCID 2 and for the diagnosis of high risk of sepsis with SeptiCyte.

Finally, an empirical antimicrobial treatment algorithm will be developed for critically ill patients related to the type of microorganisms isolated and the resistance genes evolved.

Limitations The main limitation of the study is the possibility of showing a smaller number of patients with sepsis among the included individuals than the one used to calculate the sample size. This situation could lead to a lack of statistical significance due to the lack of power of the study. To avoid this, it is planned to perform an interim analysis when 70% of the patients are reached. The second limitation is the inclusion of patients at a low rate. Although it has been considered that 6 months of fieldwork will be sufficient, the study could be extended by 4 more months to complete the number of patients.

Ethical considerations The study is being reviewed by the reference Clinical Research and Ethics Committee. Due to its retrospective design and to improve the quality of care, the committee has considered that it is not necessary to request informed consent.