Genetic variants associated with sepsis.

BACKGROUND: The variable presentations and different phenotypes of sepsis suggest that risk of sepsis comes from many genes each having a small effect. The cumulative effect can be used to create individual risk profile. The purpose of this study was to create a polygenic risk score and determine the genetic variants associated with sepsis.
METHODS: We sequenced ~14 million single nucleotide polymorphisms with a minimac imputation quality R2>0.3 and minor allele frequency >10-6 in patients with Sepsis-2 or Sepsis-3. Genome-wide association was performed using Firth bias-corrected logistic regression. Semi-parsimonious logistic regression was used to create polygenic risk scores and reduced regression to determine the genetic variants independently associated with sepsis.
FINDINGS: 2261 patients had sepsis and 13,068 control patients did not. The polygenic risk scores had good discrimination: c-statistic = 0.752 ± 0.005 for Sepsis-2 and 0.752 ± 0.007 for Sepsis-3. We found 772 genetic variants associated with Sepsis-2 and 442 with Sepsis-3, p<0.01. After multivariate adjustment, 100 variants on 85 genes were associated with Sepsis-2 and 69 variants in 54 genes with Sepsis-3. Twenty-five variants were present in both the Sepsis-2 and Sepsis-3 groups out of 32 genes that were present in both groups. The other 7 genes had different variants present. Most variants had small effect sizes.
CONCLUSIONS: Sepsis-2 and Sepsis-3 have both separate and shared genetic variants. Most genetic variants have small effects sizes, but cumulatively, the polygenic risk scores have good discrimination.

Background

Studies show that sepsis is a leading cause of mortality, the incidence is increasing, and survivors frequently have long-term physical, psychological, and cognitive impairments [1-4]. Sepsis is a syndrome, previously described by the presence of markers of systemic inflammation–abnormal vital signs and leukocyte counts—associated with suspected infection. Recently, the definition has changed to one that relies on organ system dysfunction [5-8]. Sepsis is treated with antibiotics, intravenous fluids, vasopressors and supportive care. Attempts to decrease mortality in sepsis via drugs that target specific enzymes, mediators, or proteins associated with sepsis have invariably failed, likely because sepsis has multiple pathways involving many enzymes, mediators, and proteins coded by a plethora of genes, making sepsis a polygenic disease, like hypertension [9, 10].

Sepsis and the immune response to infection have a genetic component–about 10% of human genes codes for immune mediators [11]–and adult adoptees have nearly a five-fold higher risk of infection-related mortality if a biologic parent died of infection [12]. This is a higher heritability than cardiovascular disease or cancer [12]. Studies have implicated many genes across a large spectrum of immune and coagulation proteins including interleukins, receptors, and fibrinogen [13]. However, many of these studies were limited by relatively small patient populations, which may limit the reproducibility of the findings. More recently, genome-wide association studies (GWAS) have been performed on patients with pneumonia or with severe sepsis [14, 15]. These studies found a few associated genes, in particular FER (which regulates cell-cell adhesion and mediates signaling from the cell surface to the cytoskeleton via growth factor receptors), DRD1 (which is a G protein coupled receptor that stimulates adenylate cyclase and helps mediate vascular tone), and IGF-1 (a potent activator of the AKT signaling pathway and a potent inhibitor of apoptosis) [15-17].

Complex human diseases may be explored using genetic approaches to gain insight into the complex functional pathways that characterize disease [18]. Given the polygenic nature of many diseases, in particular, sepsis where ~10% of genes code for immune mediators [11], creating a polygenic risk score to summarize the estimated effects of an individual’s genetic variants on the sepsis phenotype might be clinically useful and furthermore identifying genetic variants that are associated with sepsis may provide important biological insights. GWAS, by analyzing a multitude of genetic variants in large patient populations, has the potential to assess many candidate genes for sepsis, determine their relative contributions, and be used to create polygenic risk scores. The primary purpose of this study is to create a polygenic risk score for sepsis and to assess its discrimination. The secondary purpose it to identify those genetic variants most associated with the polygenic risk score.

Methods

Study design and setting

This study was approved by the Institutional Review Board (HUM00168165). We used the STREGA checklist when writing our report. All participants had given written informed consent. The Michigan Genomics Initiative (MGI) is a biobank that collects blood samples on adult perioperative patients for later genetic analysis. The MGI cohort and data have been previously described [19]. Briefly, subjects were recruited primarily during surgical encounters at Michigan Medicine and provided consent for linking of their electronic health records and genetic data for research purposes. Samples were genotyped on customized Illumina HumanCoreExome v12.1 bead arrays (Illumina, Inc. San Diego, CA) and subsequently imputed to the Haplotype Reference Consortium using the Michigan Imputation Server [20], providing ~14 million DNA single nucleotide polymorphisms (SNP) with a minimac imputation quality R2>0.3 and minor allele frequency >10−6. European ancestry was inferred using principal component analysis with samples from the Human Genome Diversity Panel as known references.

Outcome measure and data collection

Adult (> 18 years) subjects who were participating in the MGI had their electronic medical record searched for suspected sepsis. Using previously created and validated software [21], sepsis was separately defined using both the Sepsis-2 and Sepsis-3 definitions [5, 6]. Briefly, patients had to have had body fluid (blood, sputum, urine, wound, other) cultures obtained between Jan 28, 2008 and June 26, 2016, received antibiotics started within the 24 hours before or the 72 hours after obtaining cultures, and had ≥ 2 Systemic Inflammatory Response Syndrome (SIRS) or new Sequential Organ Failure Assessment (SOFA) points. Controls were MGI participants who did not meet sepsis criteria and their medical records showed no body fluid cultures with organism growth. Additionally, controls could not have International Classification of Diseases codes for sepsis, SIRS or infection (S1 Table, Fig 1) [22].

Fig 1

Patient flowchart.

Data analysis

GWAS was performed using Firth bias-corrected logistic regression implemented in the epacts software (https://genome.sph.umich.edu/wiki/EPACTS). Univariate logistic regressions were run with each sepsis definition as the dependent variable and each SNP allele dosage as the independent variable. SNPs were assessed for correlations, and for groups of SNPs with correlations greater than 0.5, only one SNP was retained per group. Additionally, one SNP was removed due to lack of variation in predicted genotype, with greater than 99.9% of patients having the same predicted genotype. SNPs with p ≤ 0.01 in the univariate regressions for a sepsis outcome definition were included as independent variables in the respective multivariable logistic regression along with inferred sex, and principal components 1 through 4. Non-parsimonious models were first created to produce the polygenic risk score and the discriminations of the models were assessed as the areas under the receiver operating characteristic curve (c-statistics). Then, forward selections by Akaike Information Criteria, with inferred sex and principal components 1–4 forced to remain in the models, were performed on each multivariable model to find a parsimonious model and the c-statistics were calculated. Genetic variants with final p<0.05 and 95% confidence intervals that excluded 1 were deemed significant [23-25]. Post hoc, we conducted a similar regression analysis to determine the variants associated with 90-day mortality in patients with sepsis. Regression and correlation analyses were run using RStudio version 1.4.1103.

Results

We had 2261 patients with sepsis: 2040 (90%) had Sepsis-2, 1295 (57%) had Sepsis-3 (1074 (47%) had both Sepsis-2 and Sepsis-3), (Fig 2), and 13,068 control patients without sepsis.

Fig 2

Venn Diagram showing the number of patients with sepsis by Sepsis-2 (N = 2040) or Sepsis-3 (N = 1295) criteria or both (N = 1074) and the numbers of genes and genetic variants associated with each sepsis phenotype.

Q-Q plots confirmed that the analyses were well-controlled, without an elevated Type I error (S1 and S2 Figs). The Manhattan plots (Figs 3 and 4) showed that we had no variants at univariate p-values < 5 x 10−8. We did, however, find 772 genetic variants univariately associated with Sepsis-2 and 442 associated with Sepsis-3 at the p < 0.01 level. After removal of collinear genetic variants, we created a polygenic risk score from the remaining 320 variants associated with Sepsis-2. This score had good discrimination (c-statistic = 0.752 ± 0.005) and the 218 variants associated with Sepsis-3 had similar discrimination (c-statistic = 0.752 ± 0.007). Both scores also had good calibration (Fig 5).

Fig 3

Manhattan plot showing the single nucleotide polymorphisms and their univariate p-values for Sepsis-2.

Fig 4

Manhattan plot showing the single nucleotide polymorphisms and their univariate p-values for Sepsis-3.

Fig 5

Calibration plot showing the fraction of patients with sepsis for each decile of risk calculated from the polygenic risk score.

Sepsis-2 (left), Sepsis-3 (right).

After multivariate adjustment to create a parsimonious model, we found 95 variants on 72 genes were independently associated with Sepsis-2 sepsis (S2 Table). Most variates were associated with small increases in the odds ratio of developing sepsis, only five were associated with a doubling or more and five others with at least a halving of the odds ratio compared to the reference gene. A variant in IL12RB1, 19:18184213:T:A, had the largest increase in odds ratio (95% confidence interval) = 2.63 (1.57, 4.42), p = 0.0003 and a variant in TJP1, 15:30181207:C:T, had the greatest decrease in odds ratio (95% confidence interval) = 0.29 (0.14, 0.56), p = 0.0003.

In Sepsis-3 patients, after adjustment and reduction, we found 68 variants in 54 genes to be associated with sepsis (S3 Table). Six variants were associated with a more than doubling and four with at least a halving of the odds ratio of sepsis. PPARA, 22:46579337:A:T, was associated with the largest increase in odds ratio (95% confidence interval) = 2.30 (1.50, 3.52), p = 0.0001 while CHRNA7, 15:32413847:G:A had the greatest decrease in odds ratio (95% confidence interval) = 0.12 (0.02, 0.56), p = 0.008. There were 25 variants that were identically present in both the Sepsis-2 and Sepsis-3 groups out of 32 genes that were present in both groups.

In the mortality analysis, 116 (5%) patients died by 90 days. After adjustment, we found 11 variants on 10 genes to be associated with mortality (S4 Table). The 7:105901555:C:T variant of NAMPT was associated with a 10-fold higher risk of death—odds ratio (95% confidence interval) = 10.61 (2.47, 45.64), p = 0.002.

Discussion

We found that the polygenic risk score identifies patients with sepsis with good discrimination. We also found a collection of genetic variants associated with sepsis, either Sepsis-2 or Sepsis-3 and 25 genetic variants associated with both. While Sepsis-2 and Sepsis-3 have distinct definitions, in practice, some patients have both and mortality is higher in patients with both phenotypes than with either one or the other alone [21].

As genotyping becomes increasingly common, the automatic calculation of polygenic risk scores for various diseases by identifying patients at high risk for sepsis, particularly as a complication to other treatment such as chemotherapy, immunotherapy, or surgery, might help physicians better protect these patients or to institute anti-sepsis therapy sooner. For example, a patient at high risk of sepsis undergoing surgery might receive a longer perioperative course of prophylactic antibiotics. Or, a patient receiving immunotherapy for an autoimmune disorder would have a personalized risk of sepsis determined to help understand the risks (infections and sepsis) and benefits of immunotherapy (resolution or amelioration of the disease). This would need to be tested in prospective studies.

We found genes involved in a variety of functions including immune, metabolic, vascular, and signaling. Immune related genetic variants may contribute to the success or lack thereof of clearing the infection, while non-immune genetic variants may contribute to organ damage in sepsis. As many genes have pleomorphic effects, some of these genes are better known for other diseases and have available therapy for them. For example, we found that a PCKS9 variant was associated with a near doubling of the odds of developing either Sepsis-2 or Sepsis-3. PCSK9 inhibitors lower LDL cholesterol and decrease incidence of strokes and myocardial infarction [26]. Two approved inhibitors, alirocumab and evolocumab, are in trials to reduce sepsis-related mortality (NCT03634293 and NCT03869073), respectively. Activation of PPARA, a variant we found to be associated with Sepsis-3, inhibits the production of proinflammatory molecules by interfering with macrophages [27]. In a murine model of sepsis, fibrates preserved neutrophil chemotaxis by blocking the effects of lipopolysaccharide [28]. PPARA can be stimulated by fibrates, such as gemfibrozil and fenofibrate, to metabolize lipids. Fibrates lose their ability to inhibit inflammation when PPARA gene is knocked out in mice [29]. We found ANGPT2, a potent disrupter of microvascular integrity and contributor to vascular leakage, to be associated with both Sepsis-2 and -3. Levels are higher in septic patients and progressively rise in non-survivors. Monoclonal antibodies to ANGPT2 restore vascular integrity in mice [30].

Previous prospective trials of inhibitors or antibodies to gene products linked to sepsis, such as IL1RN, CD14, and tissue factor, failed to have benefit, suggesting that these genes are only a small part of a polygenic disease; blocking only this small part would have little overall effect [31-33]. While we didn’t find any variants of IL1RN or CD14 to be associated with sepsis, we found two variants of tissue factor pathway inhibitor (TFPI), an anticoagulant protein, to be associated with Sepsis-2, but not Sepsis-3.

We also found genes associated with immune function or energy regulation to be associated with mortality. Intracellular NAMPT converts nicotinamide to nicotinamide mononucleotide and is responsible for most of the NAD+ formation in mammals. Administration of NAD+ prevents murine septic shock [34]. Extracellular NAMPT activates TLR4, promotes B cell maturation, and inhibits neutrophil apoptosis [35]. Notably, we also found a mutation in CHRNA7 associated with a doubling of the odds of death. Stimulation of this receptor has been associated with inhibited release of tumor necrosis factor and high mobility group box 1, reduced nuclear factor-kappa B activation, and improved survival in a murine model of sepsis [36].

Sepsis presents with various phenotypes [37]. Even for patients who meet Sepsis-2 or Sepsis-3 criteria, patients may have different physiologic or different organs dysfunctions. E.g., one patient may meet Sepsis-2 criteria by having an abnormal heartrate and respiratory rate, while another may have hypothermia and leukopenia. In Sepsis-3, one may have severe isolated renal dysfunction, while another may have hypotension or less severe hypotension plus thrombocytopenia. While some of these differences may be caused by the infectious agent, other differences may be from individual genetic variability. That is, out of the many genetic variants associated with sepsis, each individual will have a subset of these variants, leading to individuals having different sepsis phenotypes. Unfortunately, our study does not have enough patients to investigate this hypothesis.

GWAS in sepsis have potential to improve therapy and outcomes by two methods. First, individualized or precision medicine may be used to tailor therapy based on the particular genetic variants that each individual has [38]. Second, similar to checkpoint inhibitors in cancer chemotherapy that are effective in treating patients with a plethora of diverse mutations, GWAS may help find sepsis checkpoints–particular gene products on which a variety of pathways converge [39, 40].

GWAS analyses typically use very restrictive p values, usually p< 5x10-8, to limit the false discovery rate. However, such restrictive p-values only find variants that are relatively common or have relatively large effect sizes [41, 42]. At this threshold, studies of breast cancer using GWAS have missed the association of BRCA1 and BRCA2 (prevalences <<1%), even though they are associated with a lifetime risk of breast cancer in women of 65% and 45%, respectively [43, 44]. GWAS to create polygenic risk scores use a much more liberal p threshold and then use a variety of statistical methods to reduce the number of independent variants. A study on multiple sclerosis suggested p < .2 and one on schizophrenia and bipolar disorder used a p < 0.5 [23, 24]. We chose to use an intermediate threshold, p<0.05, after adjustment. While this produces a large number of variants, mostly with small effect sizes, Boyle et al. have argued that this is explained by genetic networks where the variants are transcribed or activated in relevant tissues [25]. Further studies are needed to determine if, when, where, and how these variants are activated in sepsis.

There are few previous GWAS studies of sepsis. Our study differs from two other studies which assessed the contributions of gene variations and mortality in sepsis [14, 45]. Our findings of different genetic variants may be related to different types of sepsis or patient populations. Our study also differs from a third study, which evaluated prognosis and response to therapy, whereas, our study also assessed genetic variants associated with sepsis occurrence [15]. Our study is more similar to another study, which searched for associations between alleles and sepsis occurrence in extremely premature infants [46]. In their small study with 351 septic infants, they found possible associations with two genes: FOXC2 and FOXL1. We found neither gene to be associated with sepsis in our adult population. There are several differences between their and our study that may explain our failure to reproduce their gene candidates. They studied a multiracial population while we studied a population from European ancestry. While all their cases had positive cultures or meningitis, the physiologic derangements or criteria of sepsis were not described. Neither Sepsis-2 nor Sepsis-3 criteria require positive cultures and most cases of adult sepsis are culture negative [47, 48]. Srinivisan et al. also found separate genes associated with gram positive, gram negative, and fungal organisms [46]. Further study is needed to determine if there are differences in genes associated with culture positive and culture negative sepsis.

In a study of Han Chinese trauma patients, Lu et al. studied 64 genetic variants and found 4 to be associated with sepsis-2: NOS2, PPARG, HSPA12A, and TLR1 [49]. We found PPARG to be associated with only Sepsis-2, similar to their study, giving credence to its association. We did not find the other 3 genes. This may be related to using different gene candidates in the studies or to differences in ethnicity. Further study is needed.

There are several limitations of this study. First, the analysis was conducted on an inferred mostly European ancestry population. GWAS done on other populations may find other genetic variants related to sepsis. Second, our control population was created on lack of sepsis or infection as documented in the University electronic medical record. As we don’t have access to medical records for treatment received at other hospitals, we may have missed episodes of sepsis or infection treated elsewhere. Similarly, control patients may develop sepsis after the analysis. This may bias our results in unknown ways. Further study is needed using patients with more comprehensive records. Third, our GWAS population was selected from patients presenting for surgery with anesthesia at an academic referral medical center. They may not be representative of non-surgical patients or the population at large. Given the constructs of our study, we don’t know if patients donated DNA for analysis before or after sepsis. If donated after sepsis, there may be a survival bias in our results. Finally, our study should be validated in other GWAS studies.

The main strength of this study is that the sepsis patients were obtained from careful review of the medical record to find patients who met either Sepsis-2 or Sepsis-3 criteria. Studies of sepsis that use administrative records, such as ICD codes, miss many cases of sepsis, which would introduce biases in GWAS [50, 51].

In conclusion, we found we found that polygenic risk scores predict sepsis with good discrimination. We found 95 variants on 72 genes to be independently associated with Sepsis-2 and 68 variants in 54 genes to be associated with Sepsis-3 sepsis.

Q-Q plots of the single nucleotide polymorphisms for Sepsis-2.

MAF—minor allele frequency.

(PDF)

Click here for additional data file.

Q-Q plots of the single nucleotide polymorphisms for Sepsis-3.

MAF—minor allele frequency.

(PDF)

Click here for additional data file.

ICD9 codes suggestive of infection that were used to remove possible control patients.

(DOCX)

Click here for additional data file.

Variants associated with Sepsis-2.

C-statistic = 0.684 (95%confidence interval = 0.672–0.696) identify genetic variants associated with both Sepsis-2 and Sepsis-3 sepsis. Factor shows the odds ratios for each of the other components of the analysis. PC–principal component.

(DOCX)

Click here for additional data file.

Variants associated with Sepsis-3.

C-statistic = 0.693 (95%confidence interval = 0.678–0.707) identify genetic variants associated with both Sepsis-2 and Sepsis-3 sepsis. Factor shows the odds ratios for each of the other components of the analysis. PC–principal component.

(DOCX)

Click here for additional data file.

Variants associated with mortality.

C-statistic = 0.721 (95% confidence interval = 0.671–0.768). PC–principal component.

(DOCX)

Click here for additional data file.

14 Dec 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Yonglan Zheng, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

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

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

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2.Thank you for stating the following in the Competing Interests section:

(Nicholas Douville was supported by a Foundation for Anesthesia Education and Research (FAER) Mentored Research Training Grant

Michael Maile was supported by the American Diabetes Association.

Milo Engoren has received consulting fee from Aerogen and Masimo.

The other authors have nothing to declare.)

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

"Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[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: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: 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: No

Reviewer #2: 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

**********

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: Engoren et al study here the genetic variants associated with two international consensus definitions of sepsis (Sepsis 2 and Sepsis 3).To that end they analyze by GAWS multiple genetic variants associated with sepsis from stored DNA of > 18 years old adult patients undergoing surgery at Michigan University Medical Center. Electronic medical records were searched for sepsis using the sepsis-2 and sepsis-3 definitions. In addition microbiological cultures, antibiotics use and �  2 SIRS or SOFA points were also recorded. Controls did not fulfill the International Classification of Diseases Code for sepsis, SIRS or infection. The authors found 25 genetic variants present in both the sepsis-2 and sepsis-3 groups. They consider that most genetic variants have small effect sizes but cumulatively the polygenic risk scores have good discrimination.

The manuscript is enticing and the underlying idea of considering sepsis as a polygenic disease like hypertension is very attractive, correct in my opinion and worth exploring further.

Major points

1.My main concern regards the selection of the controls as individuals �  18 years that did not develop sepsis. I understand as stated in Material and Methods (Page 6, lines 125) that patients and controls were followed from January 28, 2008 through June 26, 2016. Is that correct? Eight years of follow-up is a period long-enough to detect sepsis predisposition in old individuals (for instance �  75 years) but too short for 30 years old individuals that underwent a surgery (perhaps a minor orthopedic surgery) at Michigan University Medical Center. Younger controls might develop sepsis many years after finishing the 8 years follow-up period. Could you clarify this point?

2.Could you compare sepsis and control individuals regarding age, sex and main comorbidities linked to sepsis (diabetes, cancer, HIV, other immunodepressions) in one initial table?. This way the reader could see that both arms, septic and control , are well balanced.

3.Could you analyze the effect of these genetic variants studied on sepsis mortality as others did ?(References 14 and 42)

Reviewer #2: The manuscript is about an interesting and significant topic, the genetic predisposition to sepsis. Presented original research is within scope of the journal. Clear, professional English language is used in the text.

There is appropriate Introduction to show context and own results. Literature is well referenced and relevant. Structure of this manuscript is in line with the PLOS ONE standards; supplementary material containing tables and figures is relevant, high quality, well labeled and described.

Research question is well and precisely defined. The primary purpose of this study was to create a polygenic risk score for sepsis and to assess its discrimination; the secondary purpose it to identify those genetic variants most associated with the polygenic risk score.

Investigation has been performed in agreement to high technical and ethical standards, and methods are described well, with sufficient information to replicate.

In this GWAS study authors analyzed about 14 million SNPs in a group of 2261 patients with sepsis and more than 13.000 controls founding 772 genetic variants associated with Sepsis-2 and 442 with Sepsis-3 criteria. After multivariate adjustment, 100 variants on 85 genes were associated with Sepsis-2 and 69 variants in 54 genes with Sepsis-3; twenty-five variants were present in both sepsis groups. Most variants showed small effect sizes but cumulatively, the polygenic risk scores had good discrimination.

The results of this study could have practical implementation; based on obtained results, polygenic risk scores could be further explored for diagnostic and prognostic purposes. This is important because sepsis is frequent cause of mortality and morbidity with still unexplained determinants.

The authors critically discussed some limitations of the study linked mainly to the selection criteria for studied population (ethnical origin of patients with sepsis, medical history of both patients and controls). The specificity of this study is that the sepsis patients were obtained from careful review of the medical record to find patients who met either Sepsis-2 or Sepsis-3 criteria. That is important because studies of sepsis that use administrative records, such as ICD codes, miss many cases of sepsis, leading to the biases in GWAS.

All underlying data have been provided; they are sufficient, statistically correct and controlled. Contemporary bioinformatics tools have been used adequately and expertly.

Conclusions are well stated, linked to original research question and to supporting results. The main conclusion is that that established polygenic risk scores predict sepsis with good discrimination.

Authors found 95 variants on 72 genes to be independently associated with Sepsis-2 and 68 variants in 54 genes to be associated with Sepsis-3 criteria.

I suggest accepting this manuscript in present form; no modifications are required.

**********

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: Yes: Full Professor Maja Surbatovic, MD, PhD

[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.]

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 PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

22 Jan 2022

Journal Requirements:

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

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

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Response: We have made the style changes.

2.Thank you for stating the following in the Competing Interests section:

(Nicholas Douville was supported by a Foundation for Anesthesia Education and Research (FAER) Mentored Research Training Grant

Michael Maile was supported by the American Diabetes Association.

Milo Engoren has received consulting fee from Aerogen and Masimo.

The other authors have nothing to declare.)

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Response: We have added that statement to the competing interest section of the cover letter and have added a statement on requesting the data.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Response: We have done so.

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

"Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

Response: We obtained the data, with permission, from the Michigan Genomics Initiative. We have contact information for the Michigan Genomics Initiative so that others may similarly apply for the data. Data Availability: Data are available from the Michigan Genomics Initiative https://precisionhealth.umich.edu/our-research/michigangenomics/#request.

[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: Yes

________________________________________

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

Reviewer #1: Yes

Reviewer #2: 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: No

Reviewer #2: Yes

Response: We have added this Data Availability: Data are available from the Michigan Genomics Initiative https://precisionhealth.umich.edu/our-research/michigangenomics/#request.

________________________________________

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

________________________________________

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: Engoren et al study here the genetic variants associated with two international consensus definitions of sepsis (Sepsis 2 and Sepsis 3).To that end they analyze by GAWS multiple genetic variants associated with sepsis from stored DNA of > 18 years old adult patients undergoing surgery at Michigan University Medical Center. Electronic medical records were searched for sepsis using the sepsis-2 and sepsis-3 definitions. In addition microbiological cultures, antibiotics use and �  2 SIRS or SOFA points were also recorded. Controls did not fulfill the International Classification of Diseases Code for sepsis, SIRS or infection. The authors found 25 genetic variants present in both the sepsis-2 and sepsis-3 groups. They consider that most genetic variants have small effect sizes but cumulatively the polygenic risk scores have good discrimination.

The manuscript is enticing and the underlying idea of considering sepsis as a polygenic disease like hypertension is very attractive, correct in my opinion and worth exploring further.

Major points

1.My main concern regards the selection of the controls as individuals �  18 years that did not develop sepsis. I understand as stated in Material and Methods (Page 6, lines 125) that patients and controls were followed from January 28, 2008 through June 26, 2016. Is that correct? Eight years of follow-up is a period long-enough to detect sepsis predisposition in old individuals (for instance �  75 years) but too short for 30 years old individuals that underwent a surgery (perhaps a minor orthopedic surgery) at Michigan University Medical Center. Younger controls might develop sepsis many years after finishing the 8 years follow-up period. Could you clarify this point?

Response: We agree that controls can later develop sepsis. We have expanded our Limitations to discuss this. It now reads “As we don’t have access to medical records for treatment received at other hospitals, we may have missed episodes of sepsis or infection treated elsewhere. Similarly, control patients may develop sepsis after the analysis. This may bias our results in unknown ways.”

2.Could you compare sepsis and control individuals regarding age, sex and main comorbidities linked to sepsis (diabetes, cancer, HIV, other immunodepressions) in one initial table?. This way the reader could see that both arms, septic and control , are well balanced.

Response: Unfortunately, the data we received are deidentified and without age, sex, and comorbidities and we are unable to obtain that data.

3.Could you analyze the effect of these genetic variants studied on sepsis mortality as others did ?(References 14 and 42)

Response: We have added this as Supplemental Table 3. We describe this in the Methods, Results, and Discussion sections.

Reviewer #2: The manuscript is about an interesting and significant topic, the genetic predisposition to sepsis. Presented original research is within scope of the journal. Clear, professional English language is used in the text.

There is appropriate Introduction to show context and own results. Literature is well referenced and relevant. Structure of this manuscript is in line with the PLOS ONE standards; supplementary material containing tables and figures is relevant, high quality, well labeled and described.

Research question is well and precisely defined. The primary purpose of this study was to create a polygenic risk score for sepsis and to assess its discrimination; the secondary purpose it to identify those genetic variants most associated with the polygenic risk score.

Investigation has been performed in agreement to high technical and ethical standards, and methods are described well, with sufficient information to replicate.

In this GWAS study authors analyzed about 14 million SNPs in a group of 2261 patients with sepsis and more than 13.000 controls founding 772 genetic variants associated with Sepsis-2 and 442 with Sepsis-3 criteria. After multivariate adjustment, 100 variants on 85 genes were associated with Sepsis-2 and 69 variants in 54 genes with Sepsis-3; twenty-five variants were present in both sepsis groups. Most variants showed small effect sizes but cumulatively, the polygenic risk scores had good discrimination.

The results of this study could have practical implementation; based on obtained results, polygenic risk scores could be further explored for diagnostic and prognostic purposes. This is important because sepsis is frequent cause of mortality and morbidity with still unexplained determinants.

The authors critically discussed some limitations of the study linked mainly to the selection criteria for studied population (ethnical origin of patients with sepsis, medical history of both patients and controls). The specificity of this study is that the sepsis patients were obtained from careful review of the medical record to find patients who met either Sepsis-2 or Sepsis-3 criteria. That is important because studies of sepsis that use administrative records, such as ICD codes, miss many cases of sepsis, leading to the biases in GWAS.

All underlying data have been provided; they are sufficient, statistically correct and controlled. Contemporary bioinformatics tools have been used adequately and expertly.

Conclusions are well stated, linked to original research question and to supporting results. The main conclusion is that that established polygenic risk scores predict sepsis with good discrimination.

Authors found 95 variants on 72 genes to be independently associated with Sepsis-2 and 68 variants in 54 genes to be associated with Sepsis-3 criteria.

I suggest accepting this manuscript in present form; no modifications are required.

Response: Thank you for your kind comments.

________________________________________

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: Yes: Full Professor Maja Surbatovic, MD, PhD

[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.]

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 PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Response.docx

Click here for additional data file.

23 Feb 2022

Genetic Variants Associated with Sepsis

PONE-D-21-35066R1

Dear Dr. Engoren,

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

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. 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 help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- 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.

Kind regards,

Yonglan Zheng, Ph.D.

Academic Editor

PLOS ONE

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 #1: 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 #1: Yes

**********

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

Reviewer #1: 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 #1: 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 #1: 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 #1: I am pleased with the changes introduced in the revised version of the manuscript and now in my opinion n it might be accepted in PLOSONE

**********

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 #1: No

3 Mar 2022

PONE-D-21-35066R1

Genetic Variants Associated with Sepsis

Dear Dr. Engoren:

I'm 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 let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, 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.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Yonglan Zheng

Academic Editor

PLOS ONE