Laboratory tools for the direct detection of bacterial respiratory infections and antimicrobial resistance: a scoping review

Rapid laboratory tests are urgently required to inform antimicrobial use in food animals. Our objective was to synthesize knowledge on the direct application of long-read metagenomic sequencing to respiratory samples to detect bacterial pathogens and antimicrobial resistance genes (ARGs) compared to PCR, loop-mediated isothermal amplification, and recombinase polymerase amplification. Our scoping review protocol followed the Joanna Briggs Institute and PRISMA Scoping Review reporting guidelines. Included studies reported on the direct application of these methods to respiratory samples from animals or humans to detect bacterial pathogens ±ARGs and included turnaround time (TAT) and analytical sensitivity. We excluded studies not reporting these or that were focused exclusively on bioinformatics. We identified 5,636 unique articles from 5 databases. Two-reviewer screening excluded 3,964, 788, and 784 articles at 3 levels, leaving 100 articles (19 animal and 81 human), of which only 7 studied long-read sequencing (only 1 in animals). Thirty-two studies investigated ARGs (only one in animals). Reported TATs ranged from minutes to 2 d; steps did not always include sample collection to results, and analytical sensitivity varied by study. Our review reveals a knowledge gap in research for the direct detection of bacterial respiratory pathogens and ARGs in animals using long-read metagenomic sequencing. There is an opportunity to harness the rapid development in this space to detect multiple pathogens and ARGs on a single sequencing run. Long-read metagenomic sequencing tools show potential to address the urgent need for research into rapid tests to support antimicrobial stewardship in food animal production.


Rationale 3
Describe the rationale for the review in the context of what is already known.Explain why the review questions/objectives lend themselves to a scoping review approach.
Pages 4-6 (Lines 42-121) Objectives 4 Provide an explicit statement of the questions and objectives being addressed with reference to their key elements (e.g., population or participants, concepts, and context) or other relevant key elements used to conceptualize the review questions and/or objectives.

Protocol and registration 5
Indicate whether a review protocol exists; state if and where it can be accessed (e.g., a Web address); and if available, provide registration information, including the registration number.

Eligibility criteria 6
Specify characteristics of the sources of evidence used as eligibility criteria (e.g., years considered, language, and publication status), and provide a rationale.Describe all information sources in the search (e.g., databases with dates of coverage and contact with authors to identify additional sources), as well as the date the most recent search was executed.
Page 7 (Lines 137-143) Search 8 Present the full electronic search strategy for at least 1 database, including any limits used, such that it could be repeated.Describe the methods of charting data from the included sources of evidence (e.g., calibrated forms or forms that have been tested by the team before their use, and whether data charting was done independently or in duplicate) and any processes for obtaining and confirming data from investigators.

Data items 11
List and define all variables for which data were sought and any assumptions and simplifications made. Pages

Selection of sources of evidence 14
Give numbers of sources of evidence screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally using a flow diagram.

Characteristics of sources of evidence 15
For each source of evidence, present characteristics for which data were charted and provide the citations.Pages 9-10, Lines 202-215, Table 2, Supplementary Table S2 Critical appraisal within sources of evidence 16 If done, present data on critical appraisal of included sources of evidence (see item 12).N/A

Results of individual sources of evidence 17
For each included source of evidence, present the relevant data that were charted that relate to the review questions and objectives.
Pages 10-15, Lines 202-320, Tables 2-7, and Supplementary Table S2 Synthesis of results 18 Summarize and/or present the charting results as they relate to the review questions and objectives.

Summary of evidence 19
Summarize the main results (including an overview of concepts, themes, and types of evidence available), link to the review questions and objectives, and consider the relevance to key groups.† A more inclusive/heterogeneous term used to account for the different types of evidence or data sources (e.g., quantitative and/or qualitative research, expert opinion, and policy documents) that may be eligible in a scoping review as opposed to only studies.This is not to be confused with information sources (see first footnote).‡ The frameworks by Arksey and O'Malley (6) and Levac and colleagues (7) and the JBI guidance (4,5) refer to the process of data extraction in a scoping review as data charting.§ The process of systematically examining research evidence to assess its validity, results, and relevance before using it to inform a decision.This term is used for items 12 and 19 instead of "risk of bias" (which is more applicable to systematic reviews of interventions) to include and acknowledge the various sources of evidence that may be used in a scoping review (e.g., quantitative and/or qualitative research, expert opinion, and policy document).
From:  Organization developed a Global Action Plan for AMR urging every member state to adopt and adapt five objectives to their national context [1].The WHO, FAO and OIE made joint commitments to improve antimicrobial stewardship in the human and animal sectors, including investment in new diagnostic tools [2].Evidence-based veterinary medicine based on diagnostic testing promotes antimicrobial stewardship in livestock management [2].
In Canada, bovine respiratory disease (BRD) management is the most common reason for parenteral antimicrobial use in feedlot cattle [3].Antimicrobial use (AMU) protocols for B [4]RD are typically guided by clinical risk assessment using demographic information (source, age, weight class, breed), temperature measurements, and behaviour monitoring to make metaphylaxis and treatment decisions [5].The BRD complex includes a variety of bacterial and viral pathogens -Mannheimia haemolytica, Mycoplasma bovis, Pasteurella multocida, Histophilus somni, bovine viral diarrhea virus (BVDV), bovine respiratory syncytial virus (BRSV), bovine herpesvirus 1 (BoHV-1), and parainfluenza three virus (PI3V) [6].Laboratory-based diagnostic testing for bacterial BRD pathogens in live animals currently relies on culture-based diagnostic methods for bacteria and antibiotic susceptibility testing [7].
These phenotypic tests take 5-7 days to receive results [8] and may be missing some bacteria due to differential growth rates and demanding requirements for laboratory cultivation [8].As a result, improved rapid diagnostic tests are needed to provide more timely and accurate information for veterinarians to manage BRD in feedlot cattle.The changing international environment may require diagnostic information to justify AMU in livestock in the near future [4].
Rapid and immediate pen side identification of pathogens and AMR determinants to guide antimicrobial selection in individual animals is not yet an option.However, methods to provide diagnostic information within 1-2 days would be a vast improvement to inform evidence-based AMU at pen or herd management group [9].Laboratory diagnostic methods that may provide rapid sample-to-result are longread sequencing technology, polymerase chain reaction (PCR) [10], recombinase polymerase amplification (RPA), and loop-mediated isothermal amplification (LAMP) of target/specific gene sequences.
The PCR, LAMP and RPA methods are limited to known target sequences [11].The polymerase chain reaction is a popular technology for detecting low-abundance nucleic acids to identify pathogens and AMR genes [12].However, pathogen identification requires a known, specific gene target, as does AMR gene detection.The process requires multiple cycles of heating for DNA denaturing, specific probedirected amplification, and synthesis, followed by detection of the target using a thermocycler [13], electrophoresis, or in real-time using either fluorogenic probes [14] or intercalating agents [15].If multiplexed, it can identify more than one target sequence in a single assay but still has limits to the number of target sequences.In contrast, isothermal amplification techniques such as LAMP and RPA occur at one reaction temperature under simple conditions (e.g., water bath), eliminating the limitation of thermocycling in PCR, making point-of-care applications more realistic [16].The LAMP method uses a set of four target-specific primers to recognize particular sites flanking the target DNA sequence, and amplification occurs at a constant temperature of 60 -65 o C [16].Depending on detection chemistry and instrumentation, LAMP products can be detected as early as one hour.While RPA takes a bit longer (30-90 minutes), it can be run at a lower temperature (37-42 o C).Importantly, PCR, LAMP and RPA methods can all be applied directly to samples, precluding the need for isolation of organisms.This greatly reduces the time to receive diagnostic results, which is crucial for a diagnostic strategy for BRD management.
Second and third generation sequencing methods can also be applied directly to metagenomic samples [17], [18].Short-read methods are more efficient for sequencing genomes of isolated organisms, which is not timely in rapid testing.Conversely, long-read sequencing techniques, such as The Oxford Nanopore platform (MinION, GridION, and PromethION), provide great promise for rapid diagnostic testing through direct metagenomic application to respiratory samples [19].The development of VolTRAX library preparation offers the potential to deploy mobile long-read sequencing in nonlaboratory environments [20].In addition, long-read techniques for metagenomic sequencing can discover novel sequences since the method does not rely on the amplification of target sequences.This opportunity for rapid sequencing of known and unknown genes will be helpful to inform treatment and management protocols for infectious diseases.Therefore, it is essential to summarize what is currently known about these rapid diagnostic technologies to inform the future development of a BRD laboratory diagnostic strategy.
In this scoping review, we will summarize the current state of knowledge regarding the direct application of long-read metagenomic sequencing technology to respiratory samples to diagnose respiratory infections in animals and humans.We will collect emerging insights concerning the methods, opportunities, and barriers for long-read sequencing of bacteria, viruses, virulence genes, and AMR genes.Further, we will compare long-read sequencing with other direct rapid molecular technologies for gene identification and characterization, specifically PCR, LAMP and RPA.
During a preliminary search of Ovid Medline/Embase, Cochrane Library and Joanna Briggs Institute Systematic Review Register on December 8, 2020 (see Appendix for preliminary search string), we found no existing scoping review or systematic reviews on this topic.The search indicated that relevant research exists for review.

Rationale
Rapid and efficient diagnostic testing strategies are required to meet future requirements to support AMU in livestock.Current methods take up to 5-7 days and rely on traditional culture and antimicrobial susceptibility testing.Rapid methods such as long-read metagenomics sequencing, PCR, RPA, or LAMP will support antimicrobial stewardship by providing information about known bacterial and viral pathogens, virulence genes, and AMR genes [16].Exploring these new opportunities for rapid diagnostic testing will support the development of new diagnostic methods for BRD management and antimicrobial stewardship efforts.[21].Short-read sequencing techniques are not well suited for direct application to respiratory samples and thus excluded from this review.However, PCR, LAMP and RPA can be valuable tools for the rapid identification and characterization of known target genes (specific for pathogen identification, antimicrobial resistance or virulence).

Research question
What is the current state of knowledge regarding rapid diagnostic tools for respiratory infections and related antimicrobial resistance in humans and animals? Objectives: 1. To synthesize available knowledge in peer-reviewed literature and relevant grey literature (e.g., white papers) about the direct sample application of long-read sequencing for rapid diagnosis of respiratory infections and related antimicrobial resistance.
2. To compare long-read, direct sequencing methods to other rapid diagnostic techniques for nucleic acid detection.

Types of participants
Any study that includes animals and humans of any age with respiratory infections diagnosed using long-read sequencing, RPA, LAMP or PCR will be included.Any study that uses long-read sequencing, PCR, LAMP, or RPA to detect, identify, diagnose or confirm the presence of respiratory infections in humans or animals will be included.The comparator groups are other rapid diagnostic technologies such as PCR, LAMP and RPA.

Concept
Rapid refers to a real-time, same-day or otherwise fast method that provides the required detection, identification, diagnostic or confirmatory information within 48 hours.Our outcome of interest is the diagnosis, detection, identification or confirmation of bacterial or viral respiratory infections, AMR genes and related, integrative and conjugative elements (ICEs) and plasmids, and/or virulence factors in samples from the respiratory tract.Also, we will allow the term 'rapid' to evolve and be defined according to the context of the final data set.

Context
There were no limits applied to language, geographical location, long-read sequencing types and date published.

Types of studies
The review will include published peer-reviewed articles.We will also hand-search the websites of technology companies to capture relevant proprietary primary research (e.g., White Papers).We would also hand-search references and citations of most relevant articles (that is, articles that make it through second-level screening).

Exclusion Criteria
• Preprints, books, book chapters, theses, dissertations, commentary, opinion pages, editorials, newspaper articles • Describes research that does not investigate the use of long-read sequencing or other pertinent methods (e.g., PCR, LAMP or RPA).
• Describes research that does not investigate rapid methods (i.e., results within 48 hours).
• Describes research that is not for specific detection of nucleic acids (i.e., it is research on antigen detection).
• Describes research that does not investigate the diagnosis, detection, confirmation, or identification of respiratory infections, pathogens or AMR genes and their related ICEs or plasmids, and/or virulence factors for respiratory pathogens.
• Research focusing on bioinformatics tools and protocols/pipelines.

Approach
The scoping review will follow the framework outlined in the Joanna Briggs Institute Reviewer's Manual, which includes identifying the research question, finding relevant studies, selecting studies, charting the data, and reporting the results [22].The reporting guidelines developed by Tricco et al. in their extension of the PRISMA checklist for scoping reviews will be followed [23].The process of selecting relevant inclusion or exclusion criteria variables required frequent team discussions.The iterative process will continue as we design relevant data extraction templates and variables of interest.

Search strategy
We will develop search strategies in consultation with a librarian and execute the search in the following databases: MEDLINE®, AGRICOLA™, BIOSIS Previews ® , CABI and EMBASE ® .We will hand-search the websites of technology companies to capture relevant proprietary primary research.We will review reference lists of included articles and future articles citing articles to identify any articles missed by the search.Outlined below are the main/target content areas and the proposed search terms in each domain.An initial search was completed in Ovid (Medline and Embase) using a preliminary search string; informal analysis of the terms used in the titles, abstracts and indexes of relevant articles identified other terms pertinent to the search string.Synonyms and variations on the main terms are included in the more comprehensive search string.
The final search string (see Appendix) will be adapted and applied across the above databases.The reference lists from the included studies will be reviewed to identify additional articles of relevance using the snowball technique.

Study selection and Screening Process
Two independent reviewers will screen articles at all levels of screening per the descriptions below for each level.When disagreements arise concerning the inclusion of a particular reference, the reviewers will discuss until they reach a consensus.In a situation where they cannot reach an agreement, then a third researcher will arbitrate.A third level of screening was added on Nov 2, 2021 to focus the review on questions of specific interest to the project team after obtaining a large number (>500) of articles for data extraction after level 2 screening (see Appendix 2).Furthermore to the third level screening, we made a decision to add a level 3b screening starting on January 24, 2022 to focus the review on one question (rather than multiple) of special interest to the project team after obtaining 271 articles for data extraction following level 3a screening (see Appendix 3).

Level 1 Screening
• Two independent reviewers will screen the full text of each article..They will review disagreements to determine if the screening questions are working as intended.If required, screening questions will be modified and a protocol amendment dated.Once this is complete, each reviewer will review half of the remaining articles independently for efficiency.
• We will use a stacked questionnaire to screen the titles and abstracts captured by the initial search by two independent reviewers.
• If the article fully or partially meets the inclusion criteria (i.e., all screening questions are answered either 'yes' or 'unsure'), the paper will proceed to Level 2 screening.
• If the article meets any one of the exclusion criteria, it will not proceed to Level 2 screening, and the reason(s) for exclusion will be indicated.
Level 2 Screening: • Two independent reviewers will screen the full text of each article.
• In Level 2 screening, only articles that meet all the inclusion criteria will be included in the review.'Unsure' is not an option at this stage.
• One or more answers of "No" or "Unsure" to the Level 2 screening questions below lead to the exclusion of the article; information regarding the reason(s) for exclusion will be recorded.
• Google Translate™ will be used to screen non-English full article text.

Level 3 screening for the 594 full articles initially at the data extraction stage (appendix 2)
 Include only papers that focus on the diagnosis or investigation of bacterial diseases.
 Include all papers that report laboratory diagnostic qualitative and quantitative metrics of interest

Level 3b screening for 271 full articles initially at data extraction stage
 Include only papers that report limit of detection (aka detection limit, analytical sensitivity.)  Include all papers that report assay turnaround time (aka run-time, hands-on-time)

Screening Questions
Level 1: the following questions will be answered using a stacked form to screen each title and abstract: If the answer to any of the above screening questions is "no," the article will not move forward to secondary screening.An answer of "unsure" warrants additional investigation and the article will move forward to secondary screening.

For screening articles retrieved in 2021
For screening articles retrieved in 2022 We will assess the agreement between the inclusion decisions of the two reviewers using the kappa statistic measure of agreement [24].

Data extraction
EndNote X9 and Distiller will be used to manage the citations.All eligible articles will be uploaded to EndNote X9 for automatic and manual removal of duplicates.The remaining articles will be uploaded to DistillerSR® and checked again for duplicates.Eligibility for inclusion will be determined with the software-created screening forms.
A data extraction form will be created in DistillerSR®.Data extracted will summarize the metadata, population, intervention, comparators and outcome of the articles.• Run-time (hands-on time, turn-around time) • Measures of quality, as adapted from the minimum requirement for (meta)genome and qPCR publication (annex 2) • Measures of precision and accuracy -Sensitivity and specificity (diagnostic, clinical, epidemiologic)

Presentation of Results
We will present the results in a narrative summary format with the aid of tables and figures as required.Each important result of interest will have a narrative summary.Tables will include characteristics of the studies, comparisons based on the factors investigated.ti(rapid* OR immediat* OR "pen side" OR "bedside" OR "bed side" OR "point of care" OR "same day") AND ab(rapid* OR immediat* OR "pen side" OR "bedside" OR "bed side" OR "point of care" OR "same day") AND noft(genom* OR metagenom* OR "high throughput" OR "amino acid" OR "polymerase chain reaction*" OR "base sequence" OR "base sequences" OR "hybridization*" OR "nanopore technology* interpretation of the results with respect to the review questions and objectives, as well as potential implications and/or next steps.funding for the included sources of evidence, as well as sources of funding for the scoping review.Describe the role of the funders of the scoping review.Page 28 (Lines 617-621) JBI = Joanna Briggs Institute; PRISMA-ScR = Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews.* Where sources of evidence (see second footnote) are compiled from, such as bibliographic databases, social media platforms, and Web sites.
Studies must describe the use of rapid technology and diagnosis of respiratory infection as primary points of interest.The study must include at least one of the techniques of interest: long-read sequencing, RPA, LAMP or PCR.There are currently only two methods available for long-read sequencing, also known as third-generation sequencing: nanopore methods from Oxford Nanopore Technology (ONT) and PacBio's single-molecule real-time sequencing (SMRT).
Characteristics of the study, including: • Year of publication • Type of document • Country: the setting of study participants, as reported by the author (e.g., animal, human, clinic, farm, laboratory-based) • Year(s) of data collection • Host species • Sample type: nasal swab, nasopharyngeal swab, bronchoalveolar lavage, post-mortem respiratory tissue samples • Sample host and characteristics o Human (Age, sex, health or risk status etc.) o Animals (healthy vs sick; exposure status etc.) o Laboratory-stored samples (frozen, fixed, etc.) • Synthetic/contrived/mock samples Type of laboratory diagnostic tool used • Long-read sequencing, LAMP, RPA, PCR Description of and results for factor(s) investigated, including:

Table 1 :
Search term blocks for based on research question

For screening articles retrieved in 2021 For screening articles retrieved in 2022
The following questions guide the screening of each full article: