Host response signature trends in bacteraemia

To the Editor,

We read with interest the article published online on 13 December 2023 by Peri et al. [1], which evaluated the performance of a transcriptional host immune response signature (SeptiCyte RAPID, Immunexpress Inc.) in a time course study of patients with proven bacteraemia due to Staphylococcus aureus (S.a.) or Gram-Negative Bacilli (GNB) infection. The goal of the study was to determine whether SeptiCyte RAPID could differentiate complicated from uncomplicated bacteraemia and be a guide to patient management decisions related to source control and treatment duration.

The study included 86 patients with proven bacteraemia, 40 with S.a. and 46 with GNB and, at a median of 2.5 days following collection of the index blood culture, follow-up blood cultures and SeptiCyte RAPID tests were performed daily for four days (identified as Days 1–4 in the study). The authors observed that SeptiCyte RAPID test results (SeptiScores on a scale of 0–15) were higher in patients with GNB compared to S.a. bacteraemia, and that higher SeptiScores were associated with positive blood cultures at follow-up. Higher SeptiScores were also associated with a diagnosis of metastatic infection on Days 1 and 2 in the context of GNB bacteraemia but not for S.a. bacteraemia.

We commend the authors for a well written paper and for clearly acknowledging limitations of their study, including that SeptiCyte RAPID was used outside of its primary indication. However, we believe that several aspects of the study warrant further discussion, and that there are other limitations worth noting.

The intended use of SeptiCyte RAPID, as recently cleared by the Therapeutic Goods Administration of Australia, where the study was performed, states: ‘The SeptiCyte RAPID test is indicated for use at the time of admission or during a patient’s hospital stay when ≥2 signs of Systemic Inflammatory Response Syndrome (SIRS) are exhibited, combined with clinical signs of life-threatening illness due to suspicion of sepsis or incipient sepsis. Retest is warranted if deterioration of the patient’s clinical state is observed leading to increased suspicion of sepsis; or to obtain confirmatory evidence of improved clinical status’. With respect to this Australian intended use, we note that SeptiCyte RAPID provides a likelihood of sepsis (not bacteraemia) at the time sepsis is first suspected and should be used in combination with clinical signs. The initial SeptiCyte RAPID test result (at admission or upon first suspicion of sepsis or ‘Time 0’) is an important baseline value against which to compare subsequent results in light of patient clinical deterioration or improvement.

In the study by Peri et al. the median time between taking the index blood culture at Time 0 and the ‘study Day 1’ sample was 2.5 days, which is an adequate period for patients to have responded to appropriate antibiotics and source control. Importantly, the SeptiCyte RAPID test was not performed when the index blood culture was taken. We note that the authors do not provide information on whether patients were on antibiotics prior to the start of the study, or antibiotic use (e.g. type, dose, route of administration, duration, changes, sensitivity) during the study. We believe that a precise interpretation of the SeptiCyte RAPID test results in the study of Peri et al. is made difficult by the lack of an index (Time 0) SeptiCyte RAPID result, the relatively long duration between the index blood culture and study Day 1 sample, a lack of information about antibiotic use, and potential differences in source control.

As part of the development of SeptiCyte RAPID we have evaluated its performance using a number of different types of datasets, including those publicly available, microarray data from over 1000 patients and our own clinical trials [2,3]. We have not observed any statistically significant difference in SeptiCyte RAPID scores (SeptiScores) when comparing Gram positive versus Gram negative blood culture positive sepsis cases, Staphylococcus versus other genera of bacterial infections, or Staphylococcus aureus versus other genera of bacterial infections, when samples are taken at Time 0.

An important point is that SeptiCyte RAPID results are available within one hour which compares favourably with the median 11-h time to blood culture positivity in the study of Peri et al. That is, SeptiCyte RAPID results are available at least 10-h prior to gaining any blood culture results and within the 3-h period recommended by the Surviving Sepsis Campaign [4] for administration of antibiotics for patients suspected of sepsis (without shock). We therefore believe, when used as intended, SeptiCyte RAPID can provide clinical utility through rapid results, and guide management decisions in patients suspected of sepsis irrespective of the type of infection causing sepsis (bacterial (Gram positive or negative), viral, fungal, parasitic), including determining response to therapy, making discharge decisions, and enabling antibiotic stewardship requirements.

Disclosure statement

All authors of this Letter to the Editor are shareholders and/or employees of Immunexpress Inc. Peri et al. state that the study was supported by The University of Queensland and that Immunexpress did not provide any funding to the study, nor was involved in study design, samples’ testing or samples’ supply, data analysis or manuscript drafting. However, Immunexpress Inc. provided a Biocartis Idylla instrument to Pathology Queensland to enable SeptiCyte RAPID cartridges to be processed.