Rapid Next-Generation Sequencing–Based Diagnostics of Bacteremia in Septic Patients

Abstract

The increasing incidence of bloodstream infections including sepsis is a major challenge in intensive care units worldwide. However, current diagnostics for pathogen identification mainly depend on culture- and molecular-based approaches, which are not satisfactory regarding specificity, sensitivity, and time to diagnosis. Herein, we established a complete diagnostic workflow for real-time high-throughput sequencing of cell-free DNA from plasma based on nanopore sequencing for the detection of the causative agents, which was applied to the analyses of eight samples from four septic patients and three healthy controls, and subsequently validated against standard next-generation sequencing results. By optimization of library preparation protocols for short fragments with low input amounts, a 3.5-fold increase in sequencing throughput could be achieved. With tailored bioinformatics workflows, all eight septic patient samples were found to be positive for relevant pathogens. When considering time to diagnosis, pathogens were identified within minutes after start of sequencing. Moreover, an extrapolation of real-time sequencing performance on a cohort of 239 septic patient samples revealed that more than 90% of pathogen hits would have also been detected using the optimized MinION workflow. Reliable identification of pathogens based on circulating cell-free DNA sequencing using optimized workflows and real-time nanopore-based sequencing can be accomplished within 5 to 6 hours following blood draw. Therefore, this approach might provide therapy-relevant results in a clinically critical timeframe. The increasing incidence of bloodstream infections including sepsis is a major challenge in intensive care units worldwide. However, current diagnostics for pathogen identification mainly depend on culture- and molecular-based approaches, which are not satisfactory regarding specificity, sensitivity, and time to diagnosis. Herein, we established a complete diagnostic workflow for real-time high-throughput sequencing of cell-free DNA from plasma based on nanopore sequencing for the detection of the causative agents, which was applied to the analyses of eight samples from four septic patients and three healthy controls, and subsequently validated against standard next-generation sequencing results. By optimization of library preparation protocols for short fragments with low input amounts, a 3.5-fold increase in sequencing throughput could be achieved. With tailored bioinformatics workflows, all eight septic patient samples were found to be positive for relevant pathogens. When considering time to diagnosis, pathogens were identified within minutes after start of sequencing. Moreover, an extrapolation of real-time sequencing performance on a cohort of 239 septic patient samples revealed that more than 90% of pathogen hits would have also been detected using the optimized MinION workflow. Reliable identification of pathogens based on circulating cell-free DNA sequencing using optimized workflows and real-time nanopore-based sequencing can be accomplished within 5 to 6 hours following blood draw. Therefore, this approach might provide therapy-relevant results in a clinically critical timeframe. Bloodstream infections, in particular sepsis, represent one of the main causes of death, with a mortality rate of up to 30% to 50% in intensive care units worldwide—and this trend is rising.1Fleischmann C. Scherag A. Adhikari N.K.J. Hartog C.S. Tsaganos T. Schlattmann P. Angus D.C. Reinhart K. International Forum of Acute Care TrialistsAssessment of global incidence and mortality of hospital-treated sepsis. Current estimates and limitations.Am J Respir Crit Care Med. 2016; 193: 259-272Crossref PubMed Scopus (1292) Google Scholar, 2Stevenson E.K. Rubenstein A.R. Radin G.T. Wiener R.S. Walkey A.J. Two decades of mortality trends among patients with severe sepsis: a comparative meta-analysis*.Crit Care Med. 2014; 42: 625-631Crossref PubMed Scopus (427) Google Scholar, 3Vincent J.L. Marshall J.C. Namendys-Silva S.A. Francois B. Martin-Loeches I. Lipman J. Reinhart K. Antonelli M. Pickkers P. Njimi H. Jimenez E. Sakr Y. ICON InvestigatorsAssessment of the worldwide burden of critical illness: the intensive care over nations (ICON) audit.Lancet Respir Med. 2014; 2: 380-386Abstract Full Text Full Text PDF PubMed Scopus (513) Google Scholar, 4Alberto L. Marshall A.P. Walker R. Aitken L.M. Screening for sepsis in general hospitalized patients: a systematic review.J Hosp Infect. 2017; 96: 305-315Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar To provide an appropriate treatment for sepsis patients, the major challenge is not only the timely diagnosis of sepsis, but also the identification of the sepsis-causing agents.4Alberto L. Marshall A.P. Walker R. Aitken L.M. Screening for sepsis in general hospitalized patients: a systematic review.J Hosp Infect. 2017; 96: 305-315Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar A rapid detection of the causative pathogens enables early targeted antimicrobial therapy, which significantly increases survival rate, prevents subsequent complications, and reduces drug-related side effects as well as medical expenses.4Alberto L. Marshall A.P. Walker R. Aitken L.M. Screening for sepsis in general hospitalized patients: a systematic review.J Hosp Infect. 2017; 96: 305-315Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar, 5David V.L. Ercisli M.F. Rogobete A.F. Boia E.S. Horhat R. Nitu R. Diaconu M.M. Pirtea L. Ciuca I. Horhat Horhat M. C. of sepsis incidence in patients using 2017; PubMed Scopus (17) Google Scholar, J. and PubMed Scopus Google Scholar, M.M. A. H. R. Reinhart K. Angus D.C. C.S. S.A. J.L. R. including the for of severe sepsis and septic Care Med. PubMed Scopus Google Scholar, A. B. R. L. A. M. of of antimicrobial is the critical of survival in septic Care Med. PubMed Scopus Google Scholar medical in the of sepsis, are to satisfactory diagnostic for a and identification of pathogens in the K. C. C. R. A. A. pathogen PubMed Scopus Google Scholar, L. diagnosis of sepsis: and J 2014; PubMed Google Scholar, for the diagnosis of PubMed Scopus Google Scholar, A. E. of rapid diagnostic in patients with 2014; PubMed Scopus Google Scholar Therefore, and blood are the standard of care for sepsis to an and targeted for the diagnosis of PubMed Scopus Google H. and septic and treatment Crit 2017; PubMed Scopus Google Scholar a clinically validated workflow for identification and of sepsis-causing pathogens was established based on circulating cell-free DNA from plasma high-throughput P. C. T. A. K. sequencing diagnostics of in septic Med. 2016; PubMed Scopus Google Scholar, A. C. P. Y. K. T. and sequencing diagnostics for the detection of in patients with septic of a and J 2017; Scopus Google Scholar, C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar this and relevant for could be identified within to hours after blood draw. a established the sepsis clinically relevant pathogens could be from based on this the and with to and of and molecular-based L. diagnosis of sepsis: and J 2014; PubMed Google could be an workflow to be more and is an for up time to diagnosis. the sequencing and high-throughput of of next-generation sequencing 2016; PubMed Scopus Google Scholar of the sequencing is the MinION a that is based on sequencing using J. L. A. H. identification for nanopore DNA PubMed Scopus Google M.M. A the MinION 2014; PubMed Scopus Google Scholar throughput and sequencing to as of and the of in time this for A of with nanopore 2017; PubMed Scopus Google Scholar the of this sequencing with to been for from to diagnostics from samples and antimicrobial P. J. identification of pathogens in samples real-time nanopore sequencing Med. PubMed Scopus Google Scholar, C. M. L. P. MinION of the to in pathogen 2017; Scholar, J. E. L. sequencing for 2016; PubMed Scopus Google Scholar, J. A.J. A. of nanopore-based sequencing from 2017; Scholar, K. M. C. J. J. of pathogens and antimicrobial from nanopore-based 2017; PubMed Scopus Google Scholar, J. P. C. J. K. K. T. E. E. K. M. A. P. sequencing and real-time nanopore sequencing in a of PubMed Scopus Google Scholar, T. H. R. C. J. diagnosis of using nanopore-based Scholar However, for sepsis diagnostics of patient a rapid real-time workflow not been established the nanopore sequencing in time with the to time to diagnosis to a a workflow was established to for the diagnosis of sepsis in of sensitivity, specificity, and from a of a of patients in a A. C. P. Y. K. T. and sequencing diagnostics for the detection of in patients with septic of a and J 2017; Scopus Google C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar which was in the intensive care of and patients were the of the of from a of a of four septic patients and and three patients for and in a of which was in the intensive care of A. C. P. Y. K. T. and sequencing diagnostics for the detection of in patients with septic of a and J 2017; Scopus Google C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar the of and was only the of sepsis of was also time after the with a of samples to and in a of eight samples from four septic on the and clinically validated this of septic samples infections a of a and a with a and for this of three patients of were as and to a of and patients were the of the of A patient with a that a and a of the with were the patient from that an was with the for an as well as a the patient septic to an of the with subsequent that was in which the was and an was and to a positive with from an the treatment was to and after the of septic the treatment to be to a of the a with and was However, the patient after the of septic A patient with a of with the for a as well as a after the the patient septic to a with the for a as well as a treatment was with after of septic positive for However, this not in a targeted treatment for revealed a of the treatment with was to The patient after sepsis A patient with septic to a with subsequent a was treatment was with The patient well and was to be within the Two following a of the to a a patient with septic to a of the a was and an treatment with and was to septic was after the of septic after the of septic of revealed of was However, could not be the patient of in an treatment with was The patient but was to be after the of septic samples in this were within a C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar for plasma blood samples were and for minutes a of the the plasma and for plasma was was using the and the DNA from to the with the following plasma was for minutes and to The of the from was with of The and of were with the the and the to using a sequencing were using 5 of using the DNA to the including the following were with and was with the of a was with the DNA and the of the following for minutes of for for and for and for were with of and for the the of the was with of library samples to only a optimized was with of and and of of time was of with and time sequencing was to the using on a of from to to sequencing from to were on the all were on samples were for for were within a C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar preparation and sequencing were as P. C. T. A. K. sequencing diagnostics of in septic Med. 2016; PubMed Scopus Google Scholar from of using the library preparation with a of the was on a in on were and using to a for M. M. M. MinION and and PubMed Scopus Google H. nanopore MinION sequencing and 2016; PubMed Scopus Google Scholar MinION were from using based on K. M. H. R. J. C. J. E. The library for a for 2017; PubMed Scopus Google Scholar all than were and all than were to this using in a from the using all a of were The of and was using the M. C. and sequencing PubMed Scopus Google Scholar To for with was R. R. and of PubMed Scopus Google Scholar The approach is from the to A. R. A and to DNA PubMed Scopus Google Scholar was to a using with and H. R. and with PubMed Scopus Google was and with to to the H. A for and from sequencing PubMed Scopus Google H. B. A. T. J. R. and PubMed Scopus Google Scholar in P. C. T. A. K. sequencing diagnostics of in septic Med. 2016; PubMed Scopus Google Scholar for was to to using the for and were also with to a from that were identified in a and as in the C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar of to from the was using after with H. A for and from sequencing PubMed Scopus Google H. B. A. T. J. R. and PubMed Scopus Google Scholar with a were for only hits on the were and a to the The MinION were using the results C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar and to as the the The the of this are in the the sequencing from nanopore sequencing can be found the the can be and for pathogen identification from plasma is to low and of as well as for real-time sequencing throughput is and library preparation protocols are optimized for DNA input of fragments have an of to and of 5 and were in plasma of healthy and septic patients, C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar of a low input for the preparation of a is applied to of the this was optimized for three samples of one septic patient and to from as as 5 of were and for time and increasing library to as optimized could be within of that well with an of hours of sequencing optimization the and were optimized to and could increase the up to to as optimized for optimized optimized the standard to the only with an and a of to septic of and in a the standard to the only with an and a of to septic of and all of MinION which the to for MinION sequencing of and To with were and only from to with an were which that of were a of would only of to be and of and to to and are in the the within the hours of the the of were as for of all that were after hours a of was for the optimized samples the of that could be for pathogen detection were than with the optimized samples after hours which also for time after hours and hours samples from four septic and and three patients and were from a sepsis cohort the performance of the diagnostic A. C. P. Y. K. T. and sequencing diagnostics for the detection of in patients with septic of a and J 2017; Scopus Google Scholar for and only the samples sepsis were for a of time was from the to on the results from an workflow of a C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar this been to represent infections for a of a and a with all hits were validated a and as C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar a for of the workflow within this the following of more than could be to the with a of than of were for analyses of agents, with and were using and the in a to but with the MinION to for nanopore-based J. to for nanopore sequencing PubMed Scopus Google Scholar an appropriate was that of of a be to a of By the of to the MinION sequencing not a could be in a of were with and and of By the from to the of was from to with the of detected from to However, the of not the burden of of rate, of to of is as detected with as not to as in and of to of septic of and of to of is as detected with as not to as in and of to of in a septic of and relevant pathogens and were to be MinION and results and However, a of were also in which were not detected with with the an of hits was found from all with increasing to to only 5 hits this the of hits also but from to By the of pathogen hits and hits increasing the from to With the of only a hits were with and pathogen the one that the hits are sequencing not also to the low the the more and of the also with increasing and that are hits the mainly hits and also from the pathogen detection was up to a of only significantly a of the of was that the for pathogens with a of is the of with the which pathogens could be detected the MinION on the was Therefore, MinION were for all pathogens that were identified from the the of of pathogen hits were By increasing the to the detection rate to of and to a of of the of in to the was the were to for a the of MinION with was and was for all pathogen hits for MinION could be found the and MinION was found the with not hits to the throughput and of MinION with the and the for pathogen detection are the of a pathogen detection rate of were low to four and the of was which a relevant and more By increasing the could be but was To a of pathogens was from the sepsis cohort to the C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar were and all were to this the were all to be relevant pathogens for bloodstream infections from septic patients 239 samples to an patient cohort patients, as the to the of also which would not be with a low of healthy controls, a was to relevant pathogen and all hits within the were as clinically and from the sepsis patient samples as pathogen detection could be in the pathogen hits were but a of pathogen were also in all samples and and with using were of of MinION for to of sepsis in a sepsis using the this trend was and hits were the of pathogen with with the results from with the of with for a pathogen detection rate of including the pathogen C. in patient was and the to the based on was this workflow the one sepsis patients have to be with for diagnosis, which is to A. M.M. Antonelli M. R. for of sepsis and septic Care Med. 2017; PubMed Scopus Google Scholar of the septic are can a more targeted treatment be However, an an early targeted antimicrobial is for with the mortality rate is A. B. R. L. A. M. of of antimicrobial is the critical of survival in septic Care Med. PubMed Scopus Google H. C. T. M. Marshall J. P. M.M. M. A. H. T. C. C. C. R. B. K. A. Reinhart K. of a for on sepsis a Care Med. 2017; PubMed Scopus Google Scholar with a more and next-generation sequencing time to diagnosis is the of the MinION in of was to sequencing the is only after the sequencing sequencing for the of following start of the in time over the of sequencing this the time which the pathogen in eight septic patient samples were based on the workflow. on pathogen were within minutes after sequencing patient which a with a of E. the E. after of sequencing from a to with low to of DNA of up to for 6 of the were within after start of sequencing for of the eight pathogen hits were found within of sequencing. the with low of DNA from three was identified only one of which the and hours after sequencing start from the pathogen hits from eight hits were within hours after start of the sequencing time to diagnosis to 6 hours after blood the of over the performance for pathogen detection the of a is in this positive early time is more considering the pathogen hits in all samples and also in samples as which was positive for C. with only three all of detected within the hours of sequencing To the MinION would have on a sepsis the MinION was with of pathogen hits from 239 patient samples of the a MinION were based on the of pathogens for which hits were from results in of as well as A and The this the the that can be from the MinION workflow. that for the workflow with and a of for one would to more with MinION workflow as with are to but of in to library sequencing the only hits a of standard library of were for in of the C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar an for a MinION would have been However, a of a than with and an more of three a than with were the detection rate for septic samples a increase using for detection were using a of and for one and three and a using the of and for one and three 5 and When the a detection rate was more than 90% of the complete cohort with of pathogen hits using a of one using a of three a as with the pathogen hits including more than of the hits of the complete cohort would be considering the clinically more relevant early time within the hours following sepsis diagnosis to the detection as for the workflow with detection of and one three for were on the of for the identification of pathogens from not only based on P. C. T. A. K. sequencing diagnostics of in septic Med. 2016; PubMed Scopus Google R. P. E. of in an with next-generation PubMed Scopus Google E. M. H. R. E. R. J.L. diagnosis of next-generation J Med. 2014; PubMed Scopus Google Scholar but also nanopore sequencing using the P. J. identification of pathogens in samples real-time nanopore sequencing Med. PubMed Scopus Google Scholar, C. M. L. P. MinION of the to in pathogen 2017; Scholar, J. E. L. sequencing for 2016; PubMed Scopus Google Scholar, J. A.J. A. of nanopore-based sequencing from 2017; Scholar, K. M. C. J. J. of pathogens and antimicrobial from nanopore-based 2017; PubMed Scopus Google T. H. R. C. J. diagnosis of using nanopore-based Scholar for the time the of the MinION for pathogen detection in plasma samples from septic patients and performance with a that was validated an of intensive care C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar to diagnosis a critical in this but more than not only but also with next-generation diagnostic P. C. T. A. K. sequencing diagnostics of in septic Med. 2016; PubMed Scopus Google Scholar, A. C. P. Y. K. T. and sequencing diagnostics for the detection of in patients with septic of a and J 2017; Scopus Google Scholar, C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar However, nanopore sequencing real-time of relevant for diagnosis the to time to diagnosis. on of library preparation protocols and workflows, sequencing from septic patients using the MinION might provide pathogen within the 6 hours following blood as in this for of eight patient detection of the from the pathogens on was for patient with to E. hits in the a time after sepsis of the patient with a significantly burden revealed only with hits within the However, were not the time of hours but were the hours of a not only but also and of pathogen the of a is and with to a diagnostic as for patient with only three which were as C. within the a real-time approach to be established in the of sepsis diagnosis in the to that were for with P. J. identification of pathogens in samples real-time nanopore sequencing Med. PubMed Scopus Google C. M. L. P. MinION of the to in pathogen 2017; J. A.J. A. of nanopore-based sequencing from 2017; K. M. C. J. J. of pathogens and antimicrobial from nanopore-based 2017; PubMed Scopus Google Scholar throughput of a MinION with only and on is than for the established diagnostic workflow using of pathogen hits of can be with the using the for eight septic patient results are based on an low of a of MinION to in this is three more than with the workflow a of However, for of the pathogens identified in the could be for as for the three in for and the 5 in with the MinION would have been might only on that the DNA of is in blood and be MinION library which is a with the an extrapolation of the of pathogen detection in a more were for a complete sepsis cohort within 239 this we to also the of one as to not only on one for pathogen but also on a of three are more than in the one three with three and would to and to a standard library of with of one and three MinION However, based on the MinION using the detection of all more and using one and three MinION would be detection with one MinION as could have been which was mainly to a of the with a results a to the that the the and MinION workflows are The using the the might not all to be with sepsis and have to be also and M.F. P. J. Walker and can 2014; PubMed Scopus Google Scholar be as not to to for from the C. Y. P. K. T. K. performance of next-generation sequencing diagnostics with standard of care diagnostics in patients from septic Care Med. PubMed Scopus (17) Google Scholar To the pathogens and a cohort to be established with patients to for to the The only of using a is for the applied as as the of nanopore sequencing on is the of of and to the of J. A.J. A. of nanopore-based sequencing from 2017; Scholar and applied a of on hits the to hits a subsequent and to relevant pathogens from a be only pathogens with from the to also with all of the and However, samples from healthy and a is the critical in sequencing and be also that sequencing are of for are and only for the are MinION sequencing in the of this would including library and sequencing with with The main are the a and the of the one are from the to the from the sequencing which the and on the to significantly for with which be in a diagnostic a DNA DNA could also and also detection the is throughput and nanopore sequencing was to for pathogen detection from of septic patients within the as and rapid and more the performance of nanopore-based sequencing for in including sepsis results that sequencing for the detection of bloodstream pathogens is using the MinION sequencing within a clinically relevant of hours after blood this approach for sepsis diagnosis. Herein, we a tailored and bioinformatics on samples and to a clinically validated the and the of nanopore-based sequencing. been for a of pathogens bloodstream infections within an early and targeted of an diagnosis of the the MinION workflow a for the pathogens. we the a real-time to this MinION workflow as a approach in intensive care units with a real-time and the the of a rapid and targeted to patient care and

Topics & Concepts