Plasma cytokine profiles in very preterm infants with late-onset sepsis.

INTRODUCTION: Deficiencies in innate immune responses may contribute to the increased susceptibility to infection in preterm infants. In vivo cytokine profiles in response to sepsis in very preterm infants are not fully understood. AIMS: To characterise plasma pro- and anti-inflammatory cytokine concentrations and pre-defined ratios in very preterm infants with late-onset sepsis (LOS).
METHODS: In this observational study, peripheral blood samples were collected at the time of evaluation for suspected LOS from 31 preterm infants (<30 weeks gestational age). Plasma cytokine concentrations were determined by 12-plex immunoassay.
RESULTS: IL-10, IFN-γ, IL-12p70, IP-10, IL-6 and CCL2 were elevated in the majority infants with LOS (n = 12) compared to those without LOS (n = 19). There was no difference in TNF-α, IL-1β, IL-17AF, IL-8 and IL-15 concentrations between groups. IL-10/TNF-α ratios were increased, while CCL2/IL-10 and IL-12p70/IL-10 ratios were decreased in infants with LOS compared to those without.
CONCLUSION: Very preterm infants have a marked innate inflammatory response at the time of LOS. The increase in IL-10/TNF-α ratio may indicate early immune hypo-responsiveness. Longitudinal studies with a larger number of participants are required to understand immune responses and clinical outcomes following LOS in preterm infants.

Introduction

Globally, approximately 15 million infants are born preterm (<37 weeks gestational age, GA) each year, of which ~15% are born less than 32 weeks GA. [1] Preterm infants are susceptible to invasive bacterial infections and the risk is associated with lower GA and birthweight. [2] Despite advances in neonatal care, late-onset sepsis (LOS; occurring >72 hours after birth) is a frequent complication in preterm infants, affecting up to 40% of infants born <28 weeks GA. [2] Infection-related inflammation is associated with increased mortality and morbidity, particularly long-term adverse neurodevelopmental outcomes. [3]

The mechanisms underpinning the distinct neonatal immune responses to bacterial infections are incompletely understood. Pro- and anti-inflammatory cytokines are critical key initiators and regulators of inflammation and host response to bacterial infection. [4] In adults, both pro- and anti-inflammatory cytokines are simultaneously produced during the early stages of sepsis and correlate with disease progression and mortality. [5] Elevated ratios of anti- and pro-inflammatory cytokines (e.g. IL-10/TNF-α and IL-6/IL-10) are associated with multiple organ failure and are proposed markers of sepsis-induced immunosuppression in adult patients following sepsis. [6, 7] Recovery from sepsis may depend on the balance of pro- and anti-inflammatory immune responses to achieve homeostasis. [8]

It is unclear whether the immediate response of preterm infants with LOS is hyper- or hypo-inflammatory. [9] Cytokine concentrations at the time of neonatal sepsis are inconsistent, partly as studies often include both LOS and early-onset sepsis (EOS; occurring <72 hours after birth), [10-12] and a wide range of GA, [11, 13–18] making interpretation of cytokine responses challenging. We therefore aimed to characterise key cytokines [5] in plasma at the time of evaluation for suspected LOS in a homogeneous cohort of very preterm infants (born <30 weeks GA) enrolled in an observational cohort study of innate immune ontogeny. We also investigated whether infants with LOS have increased pro- and anti-inflammatory cytokine responses and elevated pre-defined cytokine ratios, as observed in adults, compared to neonates without LOS.

Materials and methods

Study participants

This study was approved by the Women and Newborn Health Services Human Research Ethics Committee, Perth, Australia (1627/EW). Written, informed consent was obtained by the Principal Investigator or delegate from the parent(s)/legal guardian prior to study participation. Infants born less than 30 weeks GA without birth defects or genetic abnormalities who were admitted to the Neonatal Intensive Care Unit (NICU) at King Edward Memorial Hospital, Perth, Australia were eligible to participate in this prospective observational cohort study of innate immune system ontogeny, with recruitment from July 2009 to October 2011. Samples taken at the time of evaluation for suspected LOS were collected opportunistically during normal working hours from enrolled infants. Blood culture was performed on infants if they had clinical signs suggestive of LOS, including increased ventilation or oxygen requirements, lethargy, decreased perfusion, temperature instability, vomiting, feed intolerance, increased apnoea and/or bradycardia. For ethical reasons only one blood sample was collected at the time of sepsis evaluation, when the clinical decision was made to investigate sepsis and commence antibiotic therapy. Infants who had episodes of EOS (n = 2) or necrotising enterocolitis (NEC) prior to the episode of suspected LOS (n = 1) were excluded. Placental histology were analyzed using an adaptation of a widely accepted semi-quantitative scoring system, as previously described. [19] Haematological parameters were recorded close to the time of blood culture sampling (mean ± SD, 124 ± 177 min).

Definition of late-onset sepsis

‘Confirmed LOS’ was retrospectively defined as a positive blood culture with a single organism and a C-reactive protein (CRP) of >15mg/L within 72hrs of blood culture sampling and antibiotic therapy for ≥ 5 days, as per the local NICU guidelines during the study period. Episodes without evidence for LOS, referred to as ‘no LOS’, were defined as a negative blood culture. Four infants were considered to have contaminating blood cultures (based on the presence of Gram-positive commensal bacteria, lack of an inflammatory response with two or more CRP measurements <15mg/L and absence of clinical features of sepsis) were excluded from analysis.

Blood sampling and processing for cytokine analyses

Approximately 0.5ml of peripheral blood was collected by venepuncture into Lithium-heparin tubes (BD Biosciences, North Ryde, Australia) at the time of blood culture (68% of samples) or close to the time of blood culture sampling (mean ± SD, 1.2 ± 2.6 hr). Sample processing and analysis was performed at the Children’s Clinical Research Facility in Perth, Australia. Samples were centrifuged at 6,000 x g for 2 minutes, and plasma extracted and stored at -80°C until cytokine measurement.

Multiplex immunoassay

Plasma cytokine concentrations were quantified using a custom 12-plex magnetic bead-based immunoassay kit (ProcartaPlex Biosystems eBioscience, San Diego, California) following the manufacturer’s instructions. The kit provided a mixture of magnetic beads conjugated with primary antibodies against IL-1β, IL-6, IL-12p70, IL-15, IL-17AF, IL-8, IL-13, IL-10, TNF-α, interferon (IFN)-γ, inducible protein (IP)-10 and chemokine (C-C motif) ligand-2 (CCL2). Fluorescence for each cytokine bead region was acquired electronically in real time on the BioPlex 200 System (Bio-Rad, Gladesville, Australia) and analysed using BioPlex Manager 5.0 Software. Cytokine concentrations, in pg/mL, were generated from a seven-point, five-parameter (four-parameter for IL-13) logistic standard curve. All samples analysed were run as a single sample on one plate. Concentrations above the highest standard were repeated at a dilution that fell within the standard curve. Concentrations below the lowest standard were assigned a value of half the lowest standard for statistical analysis.

Statistical analysis

Continuous data were summarised with median and interquartile ranges (IQR, 25th-75th) and categorical data with frequency distributions. All cytokine data were tested for normality using the Shapiro-Wilk normality test and presented using nonparametric data summaries, medians and IQR (25th-75th percentiles). Comparisons between the ‘confirmed LOS’ and ‘No LOS’ groups were by the Mann-Whitney test for continuous data and Chi-square tests for categorical data. All statistical analysis was performed using Prism 8 software (Graphpad Software, San Diego, California). P values <0.05 were considered significant.

Results

Study subject characteristics

There were no differences in the basic demographic and clinical details between the infants with confirmed LOS (n = 12) and no LOS (n = 19), details are shown in Table 1. One infant from the confirmed LOS group (sampled at 16 days postpartum) died aged 44 days of respiratory failure, and two infants in the no LOS group (both sampled at day 7 postpartum) died aged 19 and 56 days of NEC and respiratory failure, respectively. No other infants in the no LOS group developed infections or NEC prior to discharge from the NICU.

Table 1

Basic demographic and clinical parameters of study cohort.

	Confirmed LOS (n = 12)	No LOS (n = 19)	P value
Gestational age (weeks)	26.7 (24.5–28.4)	25.6 (24.6–26.6)	0.383
Birthweight (grams)	835 (669–1101)	705 (555–895)	0.194
Male	4 (33.3)	12 (63.2)	0.149
Caesarean section	5 (41.7)	11 (57.9)	0.473
Membranes rupture >24h before delivery	4 (33.3)	9 (47.4)	0.484
Antenatal steroidsb	11/12 (91.7)	18/18 (100)	0.400
Histological chorioamnionitisb	5/11 (45.5)	9/16 (56.3)	0.704
Mechanical ventilation	12 (100)	19 (100)	>0.999
Duration (hours)	241 (26.3–842)	394 (115–1091)	0.417
CPAP	12 (100)	18 (94.7)	>0.999
Duration (hours)	1003 (771–1284)	1057 (563–1298)	0.723
IVH (grade III/IV)	0 (0)	2 (10.5)	0.510
ROP (stage III/IV)	0 (0)	1 (5.3)	>0.999
Length of NICU stay (days)	83 (60–109)	109 (94–149)	0.150

aData are expressed as median (IQR) or n (%), as appropriate.

bFrom available reports. CPAP, continuous positive airway pressure; IVH, intraventricular haemorrhage; ROP, retinopathy of prematurity

The clinical parameters at the time of sepsis evaluation are shown in Table 2. For the 12 episodes of confirmed LOS, coagulase-negative staphylococci (CoNS) were the most common infectious agents isolated (10 cases, 83%), consistent with local epidemiology. [20] All infants recovered from the LOS episode.

Table 2

Clinical parameters at the time of sepsis evaluation .

	Confirmed LOS (n = 12)	No LOS (n = 19)	P value
Age (days)	12 (9–14)	10 (7–15)	0.404
Gram-positive organisms
CoNS	10 (83.4)	-	-
Bacillus sphaericus	1 (8.3)	-	-
Enterococcus faecalis	1 (8.3)	-	-
CRP (mg/L)
Highest within 72hr blood culture	59 (26–91)	17 (5–30)	0.0008
At blood culture samplingb	20 (7–36)	12 (5–24)	0.106
White blood cell count (x109/L)c	14.5 (9.4–18.1)	24.8 (15.0–31.5)	0.095
Neutrophil count (x109/L)c	11.1 (5.0–15.4)	16.7 (9.6–20.0)	0.244
Platelet count (x109/L)c	137 (102–219)	215 (118–332)	0.253
Mechanical ventilation commenced	4 (33.3)	5 (26.3)	0.704
Inotrope use	0 (0)	0 (0)	>0.999

aData are expressed as median (IQR) or n (%), as appropriate.

bAt or before (mean ± SD, 2.5 ± 2.5 hr) blood culture sampling.

cData available for n = 11 confirmed LOS and n = 18 no LOS. CoNS, coagulase-negative staphylococci; CRP, C-reactive protein.

Circulating cytokine concentrations at the time of late-onset sepsis evaluation

Overall, infants with confirmed LOS had markedly higher concentrations of IL-10, CCL2, IFN-γ, IL-12p70, IL-13, IL-6 and IP-10 compared to no LOS, whereas concentrations of TNF-α, IL-15, IL-1β, IL-17AF and IL-8 were similar in the two groups (Table 3). For IL-1β, IL-17AF and IL-13, fewer than 25%, 15% and 10%, respectively, of concentrations were above the lowest limit of detection, and results should be interpreted accordingly.

Table 3

Circulating cytokine concentrations at the time of late-onset sepsis evaluation.

	Multiplex pg/mL range		Confirmed LOS (n = 12 episodes)			No LOS (n = 19 episodes)			P value
	Minimum	Maximum	Median	IQR	Range	Median	IQR	Range
IL-6	9.2	37,800	231	62.8–922	4.6–4309	4.6	4.6–149	4.6–2881	0.017
IFN-γ	12.4	50,700	347	65.7–712	6.2–1512	6.2	6.2–6.2	6.2–949	0.001
IL-12p70	6.9	28,100	32.8	3.4–72.7	3.4–85.7	3.4	3.4–24.4	3.4–39.3	0.049
TNF-α	7.2	29,500	9.1	6.1–32.2	3.6–49	6.1	3.6–14.0	3.6–23.2	0.168
IL-15	3.1	12,500	43.8	1.5–383	1.5–576	218	1.5–276	1.5–670	0.676
IL-1βb	2.0	8,250	1.0	1.0–9.7	1.0–164	1.0	1.0–3.8	1.0–33.1	0.861
IL-17AFb	6.1	25,000	3.1	3.1–332	3.1–960	3.1	3.1–3.1	3.1–347	0.172
IL-8	2.2	8,900	234	137–1103	51.1–5138	174	84.2–707	33.7–6628	0.164
IP-10	2.2	8,800	13079	8348–16026	2661–18013	1066	602–1498	57.6–16033	<0.0001
CCL2	4.8	19,500	3407	2428–7627	1780–15105	1350	473–2588	184–7701	0.002
IL-10	2.3	9,250	194	46.8–255	22.9–487	6.5	1.1–22.9	1.1–529	<0.0001
IL-13b	3.3	13,400	1.6	1.6–3.8	1.6–16.5	1.6	1.6–1.6	1.6–1.6	0.049

aData are expressed as median, IQR (25th-75th) and range in pg/mL.

bFewer than 25% of data points in both groups are above the lowest detectable standard. LOS, late-onset sepsis; IL, interleukin; IFN, interferon; IP, inducible protein, CCL2, chemokine (C-C motif) ligand-2; TNF, tumour necrosis factor; ns, not statistically significant. Level of significance measured by Mann-Whitney test.

Analysis of anti- and pro-inflammatory ratios in infants with suspected late-onset sepsis

We next evaluated whether the ratios of anti- and pro-inflammatory cytokine concentrations, namely IL-10/TNF-α, IL-6/IL-10 and IP-10/IL-10, differed between groups, as previously reported in adults [6] and infants. [12-14] Infants with confirmed LOS had a markedly elevated IL-10/TNF-α and IP-10/IL-10 ratio compared to infants with no LOS (median (IQR) 10 (5–38) vs 0.9 (0.3–3), and 72 (38–144) vs 188 (82–533), respectively; Fig 1A and 1B), yet IL-6/IL-10 was similar across groups (Fig 1C).

Fig 1

Cytokine ratio at the time of late-onset sepsis evaluation.

Comparison of a) IL-10/TNF-α, b) IP-10/IL-10, c) IL-6/IL-10, d) CCL2/IL-10, e) IL-12p70/IL-10 and f) IFN-γ/IL-10 from infants with confirmed LOS (n = 12 episodes, closed circles) and no LOS (n = 19 episodes, closed triangles). Cytokines were measured by bead-based multiplex assay. All data shown on a log scale with median with 95% confidence intervals. Symbols depict level of significance between confirmed LOS and no LOS groups (*P = 0.048; **P = 0.003; ***P = 0.0002) by Mann-Whitney test. LOS, late-onset sepsis; IL, interleukin; IFN, interferon; IP, inducible protein; CCL2, chemokine (C-C motif) ligand-2.

As concentrations of the pro-inflammatory cytokine IFN-γ, IL-12p70 and chemokine CCL2 were associated with confirmed LOS, we evaluated the ratios of CCL2 to IL-10, IFN-γ to IL-10 and IL-12p70 to IL-10 to determine if an anti- or pro-inflammatory response predominated. A low ratio of CCL2/IL-10 and IL-12p70/IL-10 were strongly associated with confirmed LOS compared to the no LOS group (median (IQR) 38 (13–78) vs 286 (52–563) and 0.2 (0.03–0.9) vs 1.9 (0.5–3.0), respectively. Fig 1D and 1E), but there was no difference in the IFN-γ/IL-10 ratio between the patient groups (Fig 1F).

Discussion

In this study, we report that very preterm infants with microbially confirmed LOS had elevated concentrations of both pro-inflammatory cytokines, IFN-γ, IP-10, IL-12p70, IL-6, CCL2 and the anti-inflammatory cytokine IL-10, compared to those without LOS. TNF-α levels were relatively reduced in infants with LOS. Furthermore, we observed an elevated IL-10/TNF-α ratio and decreased CCL2/IL-10 and IL-12p70 ratios in infants with confirmed LOS, compared to the no LOS group.

Elevated circulating cytokine concentrations are associated with late-onset sepsis

Elevated plasma concentrations of IL-10, IFN-γ, IP-10, IL-12p70, IL-6 and CCL2, were strongly associated with confirmed LOS compared to no LOS; consistent with other preterm infant sepsis studies. [10, 13, 14, 16, 18]. Concentrations of TNF-α, IL-1β, IL-17AF and IL-8 were similar between the two groups. Published concentrations of TNF-α, IL-1β, IL-17AF and IL-8 in infants with sepsis are inconsistent, with some studies reporting results similar to ours, [10–12, 15, 17] and others reporting higher cytokine concentrations in septic infants. [14-17] The reasons underlying these discrepant results are unclear. Several factors may affect cytokine concentrations, including GA, [10] EOS vs LOS (which generally have different aetiologies), [12] sepsis definition (culture proven vs clinical sepsis) [15] and inclusion of Gram-negative, fungal and NEC-related infections (which may elicit different cytokine responses). [13] To our knowledge, this is the first study to measure IL-15 in neonates with and without sepsis. IL-15 has anti-apoptotic properties that may prevent sepsis-induced apoptosis of immune cells in septic patients, [21]. Similar to adult sepsis studies, [22] we did not observe a difference in IL-15 between preterm infants with and without sepsis.

Cytokine ratios as potential indicators of neonatal late-onset sepsis

The elevated ratio of IL-10/TNF-α in infants with confirmed LOS may indicate early immune hypo-responsiveness. Elevated ratio of IL-10/TNF-α has been observed in adult sepsis data and suggested to indicate sepsis-induced immunosuppression, [6]; it may therefore indicate relative immunosuppression in preterm infants with sepsis. The balance between these cytokines may be important in sepsis pathogenesis and susceptibility to infections in preterm infants, as described in other children and adults with sepsis. [6, 23] Other studies report markedly lower levels of LPS-, R848-, and LPS/ATP-induced TNF-α/IL-10 ratios in neonatal cord blood compared to adult peripheral blood, and an increased LPS-stimulated IL-10/TNF-α ratio in cord blood from infants born very preterm compared to moderate and late preterm. [24, 25] Further, peripheral blood from infants with LOS had a markedly higher IL-10/TNF-α ratio following stimulation with live Staphylococcus epidermidis compared to uninfected infants, indicating infected infants have an impaired immune response to bacteria. [26] This suggests that a marked anti-inflammatory response may be characteristic of infants and particularly those who are very preterm. Our results of elevated IL-10/TNF-α and decreased IP-10/IL-10 ratios are similar to other studies looking at septic neonates with disseminated intravascular coagulation (DIC). [13, 14]

We did not observe a difference in the IL-6/IL-10 ratio between groups. Previous studies report an elevated IL-6/IL-10 ratio in septic infants, [12] and specifically in those with disseminated intravascular coagulation, which is often associated with fungal and Gram-negative bacterial infections that are known to elicit different cytokine responses. [13, 14] The infants with LOS in our study had Gram-positive bacterial LOS and did not have DIC, which may explain why an elevated ratio was not observed in our cohort. Further studies with well-defined sepsis definitions are needed to resolve these discrepant findings.

The elevated CCL2 and IL-12p70 observed in LOS infants is consistent with data on neonatal [11, 14, 16] and adult sepsis. [5] To our knowledge, this is the first study to report lower ratios of CCL2/IL-10 and IL-12p70/IL-10 in infants with LOS compared to infants without LOS. The role of CCL2, a member of the chemokine CC subfamily with potent monocyte chemotactic activity, in sepsis is unclear. In a murine model of caecal ligature and puncture-induced infection, higher levels of CCL2 modulates peritoneal bacterial clearance, suggesting CCL2 is important in antimicrobial defence. [27] In contrast, adult studies report that sepsis mortality is associated with increased CCL2, [5] potentially due to chemotaxis of innate leukocytes and marked production of pro-inflammatory cytokines that contribute to organ failure, septic shock and death. [28] Elevated levels of IL-12p70, an IFN-γ inducing pro-inflammatory cytokine, have been reported in other preterm infant sepsis studies. [18] Further studies investigating the role of CCL2 and IL-12p70 in LOS in preterm infants are warranted.

Study strengths and limitations

This is the first study to show a skewed anti-inflammatory response in very preterm infants with only Gram-positive LOS. Our study focused on a well-defined population of very preterm infants with LOS, which may reduce some of the inter-individual variability seen in other neonatal studies. We acknowledge some limitations, including modest sample size, which precluded investigation of rarer Gram-negative and fungal associated LOS and infrequent outcomes, such as septic shock and death. We were unable to collect serial samples over the course of the episode and therefore lack important kinetic data. Lastly, it was not considered ethical to take blood samples from very preterm infants without clinical indication, so the no LOS group had clinical signs that led to investigations for sepsis at the time of sample collection. This group, therefore, may not be representative of responses in otherwise healthy very preterm infants.

Conclusions

Very preterm infants with LOS have elevated concentrations of both pro- and anti-inflammatory cytokines with a prominent shift to an anti-inflammatory immune response. These results suggest that neonatal LOS maybe associated with a hypo-responsiveness, similar to sepsis-induced immunosuppression observed in adults. Consistently reported patterns of cytokines production associated with sepsis may lead new diagnostic methods, prognostic tools and highlight possible therapeutic targets. Future studies should further explore the kinetic profile of circulating cytokines in preterm infants with sepsis by serial sampling over the course of the septic episode and ideally would be powered to evaluate associations between cytokine profile and disease severity and outcomes.

31 Dec 2019

PONE-D-19-33625

Plasma cytokine profiles in very preterm infants with late-onset sepsis

PLOS ONE

Dear Ms Hibbert,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Feb 14 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Philip Alexander Efron, MD, FACS, FCCM

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Methods section, please provide additional information about the participant recruitment method and the demographic details of your participants. Please ensure you have provided sufficient details to replicate the analyses such as: a) the recruitment date range (month and year), b) a description of any inclusion/exclusion criteria that were applied to participant recruitment, c) a description of how participants were recruited, and d) descriptions of where participants were recruited and where the research took place.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Title: Plasma cytokine profiles in very preterm infants with late-onset sepsis

Journal: PLOS ONE

Summary: Thank you for the opportunity to review this manuscript by Hibbert et al. The authors present an interesting study in which blood samples were obtained from 31 preterm neonates at the time of evaluation for late-onset sepsis. Plasma cytokines concentrations were determined, and inferences were made about the status of neonatal immune responses. The authors should be applauded for performing the study given the difficulties in blood collection in this vulnerable population, and the need for improved understanding of the neonatal immune response. Overall the study is very well presented, but could benefit from some clarification and potentially a multivariate logistic regression.

Major Points:

1. Why were patients with necrotizing enterocolitis excluded from the study? This is an important source of gram-negative, late-onset sepsis. Is this disease process considered fundamentally different from the LOS cases studied?

2. In Methods, Blood sampling: blood was collected close to the time of blood culture sampling. How was blood collected and how much was drawn? Was the blood draw performed separately from the blood culture collection? Was it a capillary heelstick, venipuncture, or drawn from an existing line?

3. Could the composite cytokine score ratio of anti- to pro-inflammatory cytokines be biased by the fact that there are only two anti-inflammatory cytokines (IL-13 and IL-10) included compared to many more (10) pro-inflammatory cytokines in the ratio? Since sums of the normalized scores are used for the ratio, it seems that the ratio will always be small given the large sums in the denominator for 10 different inflammatory cytokines.

4. Did the authors consider performing multivariate logistic regression to predict sepsis vs. non-sepsis? This is a well-performed prospective cohort study with fairly well-balanced sepsis vs. non-sepsis groups. It would be interesting to evaluate which clinical and cytokine factors emerge as the most significant predictors of sepsis, as this could help determine which cytokines should be measured to assist with the diagnosis of neaonatal sepsis. If possible, I would suggest having sepsis vs. non-sepsis be the outcome of the regression, and include multiple explanatory varaibles (such as IL-10, IFN-gamma, gestational age at birth, day of life, CRP, platelet count). This would aid in creating a predictive model with an area-under-the-curve to predict late-onset sepsis based on parameters that can be obtained from POC tests/clinical information. This would also lend support to statements in the conclusion about the association of cytokine levels with sepsis vs. non-sepsis.

Minor Points:

1. Table 1 has a line describing “mortality during episode”. What does this mean, i.e. how is the episode defined? I would consider deleting this line given that there is already a descriptive statement in the results stating that only one infant died 12 days after the initial sample collection.

2. Table 2: Please include the calculated p-value rather than “ns” as this offers a more complete picture of the analysis. For example, median IL-15 levels were 43.8 in the sepsis group and 218 in the non-sepsis group. A p-value rather than “ns” here would help with interpretation of that result.

Recommendation: Major Revision

Reviewer #2: The manuscript, “Plasma cytokine profiles in very preterm infants with late-onset sepsis” evaluates various cytokine profiles in the plasma of very preterm infants. They found that there are marked innate inflammatory responses, with a relatively high anti-inflammatory to pro-inflammatory ratio. While the understanding of immune responses in preterm infants is important, a number of issues need to be addressed prior to publication:

1. Can the authors state very clear inclusion and exclusion criteria in the methods section? There is a mention in the limitations that the “no LOS” group were patients that had concerns for sepsis, but ultimately did not have a positive blood culture. Without further description, this is not necessarily an adequate comparison group. Was there any possibility to collecting blood from healthy pre-term infants or did the authors feel this was not clinically ethical? In addition, the authors state that one infant from the “no LOS” group had NEC. Did any of the other “no LOS” patients have an infection? This would be extremely important to know, as not all septic patients will have positive blood cultures and/or bacteremia may be transient enough to miss.

2. Can the authors state why they chose a single early time point and what was the goal for sample collection? In the methods section, they state that, “hematological parameters were recorded close to the time of blood culture sampling” of about 3h (124 ±177min). Was the time of blood culture sampling presumably time zero of diagnosis of LOS?

3. Although the authors state that 83% had gram positive cultures, can the remainder of the organism types be listed (number and %). Were there any causes for the bacteremia (ie what was the infectious source? Was this primary bacteremia or from other sources)?

4. While the LOS group had significantly higher levels of both pro- and anti-inflammatory cytokines, this does not necessarily predict immune cell function and therefore one cannot conclude that the patients are “immunosuppressed”. Do the authors have any insight into actual immune cell function?

5. In the “multiplex immunoassay” section in the Methods, the authors state that cytokine concentrations were generated from a seven-point, five parameter logistic standard curve, except for IL-13. Is there a particular reason for this or was it simply the assay standard?

6. A minor detail: the legend for Figure 2 labels d) as IL-12p70/IL-10 and e) as CCL2/IL-10, however these are reversed in the actual figure.

7. Since the authors set alpha at p<0.05, this is what should be considered statistically significant and there is no need for further statistical indices. In other words, p<0.004 is not “more significant” that p<0.05 as indicated in Figure 2.

8. Did any patients in either group require surgery or any other major invasive procedure? This would obviously skew the data on inflammation if so. Please address.

9. Were there any other outcomes assessed, such as 30d mortality, ventilator days, ICU length of stay, or hospital length of stay? This would make the data more robust.

Reviewer #3: PONE-D-19-33625

In this study, the investigators utilized multiplex cytokine kits to evaluate the cytokine signature of premature infants with late onset sepsis. The manuscript is well written without egregious spelling or syntax errors. I have some comments for the authors.

1. This study is not hypothesis driven and the cohort is extremely small making it difficult to draw any concrete conclusions. How will the results from the current study assist the providers that treat preterm neonates with sepsis?

2. The “no LOS” group was not clearly defined. Were these infants that displayed the signs and symptoms of sepsis ie increased respiratory support, temp instability, feeding intolerance or A's and B's but just did not have positive blood cultures? So by definition of the LOS group having positive blood culture and CRP>15 and >5 days of antibiotic therapy did that imply that you could have one or two of these criteria but not all three and be included in the “no LOS” group? Also, are the criteria utilized to define LOS utilized in the majority of neonatal literature examining sepsis?

3. Under the definition of LOS, I don't understand why four positive blood cultures were classified as no LOS (line 98). Please explain the rationale for this statement and criteria.

4. In the methods section for the description of normal values for the cytokine multiplex, please provide those data in a table (line 117-122).

5. How do the authors explain/justify that the IL-6/IL-10 ratio was not different between the two groups? How did they choose the ratios to examine; isn't there a nearly infinite number of combinations? In holding with this question, why did they choose IL-10 and not another anti-inflammatory cytokine for the examination described in Fig 2 (lines 198-204)?

6. In the first line of the discussion the authors use the term "very preterm". This term should be defined and is this term the nomenclature routinely utilized in the neonatal literature?

7. Since the authors did not see a difference in IL-6/IL-10 ratio and it is reportedly related to DIC in sepsis, they should investigate whether any of their LOS patients had DIC.

8. If the authors would have utilized the data from this study to try to predict which preterm infants at risk for LOS in a separate cohort, it would make this manuscript much more relevant.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

27 Mar 2020

Journal requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

Answer: Formatting of headings and sub-headings have been corrected.

2. In your Methods section, please provide additional information about the participant recruitment method and the demographic details of your participants. Please ensure you have provided sufficient details to replicate the analyses such as: a) the recruitment date range (month and year), b) a description of any inclusion/exclusion criteria that were applied to participant recruitment, c) a description of how participants were recruited, and d) descriptions of where participants were recruited and where the research took place.

Answer: Additional information about the participant recruitment method have been updated in the "Methods" section on page 4/5.

Reviewer #1: Title: Plasma cytokine profiles in very preterm infants with late-onset sepsis. Summary: Thank you for the opportunity to review this manuscript by Hibbert et al. The authors present an interesting study in which blood samples were obtained from 31 preterm neonates at the time of evaluation for late-onset sepsis. Plasma cytokines concentrations were determined, and inferences were made about the status of neonatal immune responses. The authors should be applauded for performing the study given the difficulties in blood collection in this vulnerable population, and the need for improved understanding of the neonatal immune response. Overall the study is very well presented, but could benefit from some clarification and potentially a multivariate logistic regression.

Major Points:

1. Why were patients with necrotizing enterocolitis excluded from the study? This is an important source of gram-negative, late-onset sepsis. Is this disease process considered fundamentally different from the LOS cases studied?

Answer: We agree that necrotising enterocolitis (NEC) is a very important potential source of infection and inflammation that requires further understanding, however, the pathogenesis of NEC and the relationship between NEC and LOS are unclear (reviewed in PMID: 25171544). Further, blood culture-positive NEC is most often associated with Gram-negative bacteria, for which infections are more severe and elicit a stronger inflammatory response (PMID: 22633519). There were two infants in our cohort who developed blood culture-positive NEC, one prior to the episode of LOS and one 12 days following a ‘no LOS’ episode. It is not within the scope (or power) of this study to evaluate the relationship or impact NEC has on subsequent LOS episodes, therefore the infant with the blood culture-positive NEC episode prior to the suspected LOS episode was excluded from our analysis. Exclusion of this infant from analysis has been clarified on page 4 of the revised manuscript.

2. In Methods, Blood sampling: blood was collected close to the time of blood culture sampling. How was blood collected and how much was drawn? Was the blood draw performed separately from the blood culture collection? Was it a capillary heelstick, venipuncture, or drawn from an existing line?

Answer: Thank you for bringing this to our attention. We collected 0.5ml of peripheral blood by venepuncture, with the majority of samples collected either at the time of sampling for blood culture (n=21) or 2-3 hours prior to sampling for culture (mean � SD, 124 � 177 min). The process for blood collection has been updated on page 5 of the revised manuscript.

3. Could the composite cytokine score ratio of anti- to pro-inflammatory cytokines be biased by the fact that there are only two anti-inflammatory cytokines (IL-13 and IL-10) included compared to many more (10) pro-inflammatory cytokines in the ratio? Since sums of the normalized scores are used for the ratio, it seems that the ratio will always be small given the large sums in the denominator for 10 different inflammatory cytokines.

Answer: We agree with the reviewer and have removed the composite cytokines scores and ratios from the manuscript.

4. Did the authors consider performing multivariate logistic regression to predict sepsis vs. non-sepsis? This is a well-performed prospective cohort study with fairly well-balanced sepsis vs. non-sepsis groups. It would be interesting to evaluate which clinical and cytokine factors emerge as the most significant predictors of sepsis, as this could help determine which cytokines should be measured to assist with the diagnosis of neonatal sepsis. If possible, I would suggest having sepsis vs. non-sepsis be the outcome of the regression, and include multiple explanatory variables (such as IL-10, IFN-gamma, gestational age at birth, day of life, CRP, platelet count). This would aid in creating a predictive model with an area-under-the-curve to predict late-onset sepsis based on parameters that can be obtained from POC tests/clinical information. This would also lend support to statements in the conclusion about the association of cytokine levels with sepsis vs. non-sepsis.

Answer: We agree with performing multivariate logistic regression in principle, however, following consultation with our biostatistician colleague, we were advised that this would require a larger sample size. With the current sample size, we believe that regression models would be unstable and findings would not necessarily be robust, and therefore, is beyond the scope of this project.

Minor Points:

1. Table 1 has a line describing “mortality during episode”. What does this mean, i.e. how is the episode defined? I would consider deleting this line given that there is already a descriptive statement in the results stating that only one infant died 12 days after the initial sample collection.

Answer: Thank you, we agree with the reviewer and have removed this variable from Table 1.

2. Table 2: Please include the calculated p-value rather than “ns” as this offers a more complete picture of the analysis. For example, median IL-15 levels were 43.8 in the sepsis group and 218 in the non-sepsis group. A p-value rather than “ns” here would help with interpretation of that result.

Answer: Thank you, the P values in Table 2 have been updated.

Reviewer #2: The manuscript, “Plasma cytokine profiles in very preterm infants with late-onset sepsis” evaluates various cytokine profiles in the plasma of very preterm infants. They found that there are marked innate inflammatory responses, with a relatively high anti-inflammatory to pro-inflammatory ratio. While the understanding of immune responses in preterm infants is important, a number of issues need to be addressed prior to publication:

1. Can the authors state very clear inclusion and exclusion criteria in the methods section? There is a mention in the limitations that the “no LOS” group were patients that had concerns for sepsis, but ultimately did not have a positive blood culture. Without further description, this is not necessarily an adequate comparison group. Was there any possibility to collecting blood from healthy pre-term infants or did the authors feel this was not clinically ethical? In addition, the authors state that one infant from the “no LOS” group had NEC. Did any of the other “no LOS” patients have an infection? This would be extremely important to know, as not all septic patients will have positive blood cultures and/or bacteremia may be transient enough to miss.

Answer: Thank you for raising these points. This was a prospective observational study of early life innate immune system ontogeny and all infants born less than 30 weeks’ gestational age without birth defects or genetic abnormalities and admitted to the NICU at King Edward Memorial Hospital were eligible for enrolment (updated on page 4 of the revised manuscript). The scope of this project, characterisation of cytokines at the time of sepsis, was not the primary aim of the main study, therefore collection of blood samples at the time of LOS evaluation were collected opportunistically. We could not ethically justify the collection of blood from infants who did not have a septic workup. The majority of very preterm have a multitude of co-morbidities, including chronic lung disease, intraventricular haemorrhage, retinopathy of prematurity, making it challenging to identify a cohort of infants that are ‘healthy’ for comparison. The ‘no-LOS’ group of infants represent the closest matched group to our sepsis group at the time of suspected sepsis and differ only for characterisation in the presence of detectable infection at that time. However, we can also confirm that no other infants, aside from the one infant who developed NEC 12 days following a ‘no LOS’ episode, in the no LOS group developed an infection during their admission to the NICU, this has been updated on page 7 of the revised manuscript.

2. Can the authors state why they chose a single early time point and what was the goal for sample collection? In the methods section, they state that, “hematological parameters were recorded close to the time of blood culture sampling” of about 3h (124 ±177min). Was the time of blood culture sampling presumably time zero of diagnosis of LOS?

Answer: In our study, and the majority of other neonatal sepsis studies, blood sample collection at the time of blood culture is common practice and is considered the onset of infection. For ethical reasons one blood sample at the time of blood culture was permitted, and has been clarified on page 5 of the revised manuscript. Abnormal WBC, neutrophil and platelet counts have been shown to be associated with the onset of sepsis and were therefore reported at the time of onset of LOS.

3. Although the authors state that 83% had gram positive cultures, can the remainder of the organism types be listed (number and %). Were there any causes for the bacteremia (ie what was the infectious source? Was this primary bacteremia or from other sources)?

Answer: Thank you, we have clarified the number and percent of the other causative organisms in Table 2. The infants in our cohort were born <30 weeks gestational age with a median birthweight of <1000g. Very preterm birth and very low birthweight are the leading factors contributing to the increased risk of LOS, while the infectious source is often unknown for neonatal late-onset sepsis (PMID: 20876594).

4. While the LOS group had significantly higher levels of both pro- and anti-inflammatory cytokines, this does not necessarily predict immune cell function and therefore one cannot conclude that the patients are “immunosuppressed”. Do the authors have any insight into actual immune cell function?

Answer: We agree with the reviewer, and as part of the original study characterising innate immune system ontogeny in preterm infants, we have found that the infants with LOS have impaired cytokine responses when challenged with live bacteria under ex vivo conditions. The manuscript describing these results has recently been published with Clinical Infectious Diseases (PMID: 31960030) and reference has been made in this manuscript on page 12 of the revised manuscript.

5. In the “multiplex immunoassay” section in the Methods, the authors state that cytokine concentrations were generated from a seven-point, five parameter logistic standard curve, except for IL-13. Is there a particular reason for this or was it simply the assay standard?

Answer: Five-parameter (5PL) and four-parameter (4PL) logistic modelling are the most widely accepted curve fitting methods for immunoassays (PMID: 24918306; PMID: 29536273). For S-shaped curves 4PL fits symmetrical data whereas 5PL has an added parameter that allows a better fit for asymmetrical data. Our IL-13 data was symmetrical and the 4PL model was used. The remaining cytokines/chemokines data were asymmetrical and fit using the 5PL model.

6. A minor detail: the legend for Figure 2 labels d) as IL-12p70/IL-10 and e) as CCL2/IL-10, however these are reversed in the actual figure.

Answer: Thank you for pointing this out, the figure legend has been corrected.

7. Since the authors set alpha at p<0.05, this is what should be considered statistically significant and there is no need for further statistical indices. In other words, p<0.004 is not “more significant” that p<0.05 as indicated in Figure 2.

Answer: Thank you, we understand the concern raised by the reviewer, but think it is important to provide the reader with an indication of the P value, as such we have updated the legend with the exact P value.

8. Did any patients in either group require surgery or any other major invasive procedure? This would obviously skew the data on inflammation if so. Please address.

Answer: We can confirm that during the first 4 weeks of life no infants in the cohort required surgery. Due to the prematurity of infants born <30 weeks gestational age a large proportion of the infants will experience some level of invasive intervention (e.g. mechanical ventilation, blood infusion) or medical condition (e.g. intraventricular haemorrhage) that may have an impact on inflammation, however, given the sample size of this study, only adjustment for a limited number of variables is feasible.

9. Were there any other outcomes assessed, such as 30d mortality, ventilator days, ICU length of stay, or hospital length of stay? This would make the data more robust.

Answer: We agree with your suggestion and have added several outcomes to Table 1, including duration of mechanical ventilation, duration of CPAP, intraventricular haemorrhage, ROP and length of NICU stay.

Reviewer #3: In this study, the investigators utilized multiplex cytokine kits to evaluate the cytokine signature of premature infants with late onset sepsis. The manuscript is well written without egregious spelling or syntax errors. I have some comments for the authors.

1. This study is not hypothesis driven and the cohort is extremely small making it difficult to draw any concrete conclusions. How will the results from the current study assist the providers that treat preterm neonates with sepsis?

Answer: We acknowledge that the sample size is modest, reflecting the realities of a real-world study in this vulnerable population. However, but are confident that the findings from this study, especially the IL-10 and CCL2, results will prompt further investigation using larger studies to examine the utility of these cytokines as adjunct sepsis diagnostic makers to blood culture. The results from this study also add to and support the existing literature in this field, despite the acknowledged limitations.

2. The “no LOS” group was not clearly defined. Were these infants that displayed the signs and symptoms of sepsis ie increased respiratory support, temp instability, feeding intolerance or A's and B's but just did not have positive blood cultures? So by definition of the LOS group having positive blood culture and CRP>15 and >5 days of antibiotic therapy did that imply that you could have one or two of these criteria but not all three and be included in the “no LOS” group? Also, are the criteria utilized to define LOS utilized in the majority of neonatal literature examining sepsis?

Answer: Thank you, that is correct, a ‘no-LOS’ infant could have an elevated CRP, but without a positive blood culture. Currently, there is no consensus definition for neonatal sepsis (PMID: 26766602; PMID: 24751791), and positive blood culture is the accepted standard gold standard for diagnosing neonatal sepsis (PMID: 24751791). However, we also acknowledge that blood culture has limited sensitivity and specificity, the latter potentially leading to false-positive results (PMID: 24751791). For this reason, we included an additional objective marker, elevated CRP, as an indicator of inflammation and the clinical indicator of antibiotic therapy for �5 days as additional inclusion criteria, an approach previously used in analogous studies (PMID: 28367457, PMID: 16613997). Until there is a consensus definition for neonatal sepsis the classification will vary in the neonatal literature. Together with an international neonatal sepsis group we have highlighted the need for an improved and unified neonatal sepsis diagnosis (Pediatric Research, in press).

3. Under the definition of LOS, I don't understand why four positive blood cultures were classified as no LOS (line 98). Please explain the rationale for this statement and criteria.

Answer: Coagulase-negative Staphylococci (CoNS), a group of skin commensals, are the most commonly isolated group of organisms in very preterm infant late-onset sepsis. However, given these organisms are present on the skin, contamination of blood cultures is a frequent occurrence in the neonatal population leading to false-positive identifications of sepsis (3%-18%; PMID: 23331501, PMID: 11136522). These four blood cultures in our study were classified as contaminants based on the lack of an inflammatory response (no CRP rise within 48h of blood culture), negativity of a subsequent blood culture, and absence of clinical features of sepsis.

4. In the methods section for the description of normal values for the cytokine multiplex, please provide those data in a table (line 117-122).

Answer: Thank you for suggesting this, the multiplex minimum and maximum range has been added to Table 2 and removed from the methods section.

5. How do the authors explain/justify that the IL-6/IL-10 ratio was not different between the two groups? How did they choose the ratios to examine; isn't there a nearly infinite number of combinations? In holding with this question, why did they choose IL-10 and not another anti-inflammatory cytokine for the examination described in Fig 2 (lines 198-204)?

Answer: Elevated IL-6/IL-10 ratios have been shown in septic neonates with disseminated intravascular coagulation, which is commonly associated with Gram-negative infections. Our cohort of infants with Gram-positive LOS did not develop DIC and we feel this may contribute to the discrepancy in results – this has been clarified on page 13 of the revised manuscript.

We chose IL-10/TNF�, IL-6/IL-10 and IP-10/IL-10 ratios because they have been previously reported in adults and infants with infection as possible diagnostic markers, as referenced on page 10 of the revised manuscript (PMID: 10608764, PMID:27997530, PMID: 12719394). We observed a significant increase in CCL2, IFN� and IL-12p70 in infants with LOS and explored the ratio of these cytokines with IL-10. IL-13 was the only other anti-inflammatory cytokine in our panel, however the majority of levels were below the detection rate (common among neonatal sepsis PMID: 24013483; PMID: 29562764), therefore it was not a good candidate for exploring ratios.

6. In the first line of the discussion the authors use the term "very preterm". This term should be defined and is this term the nomenclature routinely utilized in the neonatal literature?

Answer: Thank you, we have now defined this term on page 3 of the revised manuscript. The World Health Organisation sub-categorises preterm infants based on gestational age: extremely preterm (<28 weeks), very preterm (28-32 weeks) and moderate to late preterm (32-37 weeks) and this nomenclature is standard in the neonatal literature.

7. Since the authors did not see a difference in IL-6/IL-10 ratio and it is reportedly related to DIC in sepsis, they should investigate whether any of their LOS patients had DIC.

Answer: None of the infants with LOS in our cohort developed DIC, most likely related to the Gram-positive aetiology that predominated in our cohort. We have clarified this on page 13 of the revised manuscript.

8. If the authors would have utilized the data from this study to try to predict which preterm infants at risk for LOS in a separate cohort, it would make this manuscript much more relevant.

Answer: This is a valid point, and we agree in principle, however, this requires a larger sample size (as advised by a biostatistician colleague) and a separate independent validation cohort. This is beyond the scope of this project. We are currently conducting a larger neonatal study with a primary focus on sepsis, and anticipate completion in the next 2-3 years. We hope that other groups will also aim to validate our findings in other preterm populations.

27 Apr 2020

Plasma cytokine profiles in very preterm infants with late-onset sepsis

PONE-D-19-33625R1

Dear Dr. Hibbert,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Philip Alexander Efron, MD, FACS, FCCM

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: (No Response)

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Reviewer #3: No

4 May 2020

PONE-D-19-33625R1

Plasma cytokine profiles in very preterm infants with late-onset sepsis

Dear Dr. Hibbert:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Philip Alexander Efron

Academic Editor

PLOS ONE