A One Health perspective on the use of genotypic methods for antimicrobial resistance prediction

Kelli J. Maddock Veterinary Diagnostic Laboratory, North Dakota State University, Fargo, ND

Search for other papers by Kelli J. Maddock in
Current site
Google Scholar
PubMed
Close
 MS
,
Claire R. Burbick Washington Animal Disease Diagnostic Laboratory, Department of Veterinary Microbiology and Pathology, College of Veterinary Medicine, Washington State University, Pullman, WA

Search for other papers by Claire R. Burbick in
Current site
Google Scholar
PubMed
Close
 DVM, PhD, DACVM
,
Stephen D. Cole School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA

Search for other papers by Stephen D. Cole in
Current site
Google Scholar
PubMed
Close
 VMD, MS, DACVM
,
Joshua B. Daniels College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO

Search for other papers by Joshua B. Daniels in
Current site
Google Scholar
PubMed
Close
 DVM, PhD, DACVM
,
Tessa E. LeCuyer School of Veterinary Medicine, University of California-Davis, Davis, CA

Search for other papers by Tessa E. LeCuyer in
Current site
Google Scholar
PubMed
Close
 DVM, PhD, DACVM
,
Xian-Zhi Li Veterinary Drugs Directorate, Health Canada, Ottawa, ON, Canada

Search for other papers by Xian-Zhi Li in
Current site
Google Scholar
PubMed
Close
 MD, PhD
,
John Dustin Loy Nebraska Veterinary Diagnostic Center, School of Veterinary Medicine and Biomedical Sciences, University of Nebraska-Lincoln, Lincoln, NE

Search for other papers by John Dustin Loy in
Current site
Google Scholar
PubMed
Close
 DVM, PhD, DACVM
,
Susan Sanchez Department of Infectious Diseases, College of Veterinary Medicine, University of Georgia, Athens, GA

Search for other papers by Susan Sanchez in
Current site
Google Scholar
PubMed
Close
 PhD
,
Brianna L. S. Stenger Veterinary Diagnostic Laboratory, North Dakota State University, Fargo, ND

Search for other papers by Brianna L. S. Stenger in
Current site
Google Scholar
PubMed
Close
 PhD
, and
Dubraska Diaz-Campos College of Veterinary Medicine, The Ohio State University, Columbus, OH

Search for other papers by Dubraska Diaz-Campos in
Current site
Google Scholar
PubMed
Close
 DVM, PhD

Click on author name to view affiliation information

Abstract

Antimicrobial resistance is a global One Health concern with critical implications for the health of humans, animals, and the environment. Phenotypic methods of bacterial culture and antimicrobial susceptibility testing remain the gold standards for the detection of antimicrobial resistance and appropriate patient care; however, genotypic-based methods, such as PCR, whole genome sequencing, and metagenomic sequencing, for detection of genes conferring antimicrobial resistance are increasingly available without inclusion of appropriate standards for quality or interpretation. Misleading test results may lead to inappropriate antimicrobial treatment and, in turn, poor patient outcomes and the potential for increased incidence of antimicrobial resistance. This article explores the current landscape of clinical and methodological aspects of antimicrobial susceptibility testing and genotypic antimicrobial resistance test methods. Additionally, it describes the limitations associated with employing genotypic-based test methods in the management of veterinary patients from a One Health perspective. The companion Currents in One Health by Maddock et al, AJVR, March 2024, addresses current and future needs for veterinary antimicrobial resistance research.

Abstract

Antimicrobial resistance is a global One Health concern with critical implications for the health of humans, animals, and the environment. Phenotypic methods of bacterial culture and antimicrobial susceptibility testing remain the gold standards for the detection of antimicrobial resistance and appropriate patient care; however, genotypic-based methods, such as PCR, whole genome sequencing, and metagenomic sequencing, for detection of genes conferring antimicrobial resistance are increasingly available without inclusion of appropriate standards for quality or interpretation. Misleading test results may lead to inappropriate antimicrobial treatment and, in turn, poor patient outcomes and the potential for increased incidence of antimicrobial resistance. This article explores the current landscape of clinical and methodological aspects of antimicrobial susceptibility testing and genotypic antimicrobial resistance test methods. Additionally, it describes the limitations associated with employing genotypic-based test methods in the management of veterinary patients from a One Health perspective. The companion Currents in One Health by Maddock et al, AJVR, March 2024, addresses current and future needs for veterinary antimicrobial resistance research.

Introduction

Antimicrobial resistance (AMR) is a critical threat to public health worldwide. A One Health concern, AMR has animal, human, and environmental health implications. The use of some molecular test methods targeting specific genes for the detection of AMR has become common in human medicine, but their utility in veterinary medicine is still unclear.1,2 With the rapid development of molecular-based technologies, sequence-based or genotypic antimicrobial resistance testing (ART) methods are gaining use in clinical practice; however, these tools have important limitations that may not be obvious to the clinician. Genotypic ART must be used in the appropriate situations and with the understanding that phenotypic antimicrobial susceptibility testing (AST) methods remain the gold standard. In short, genotypic ART tells you which antimicrobials you cannot use, while AST tells you which are appropriate for use. The purpose of this article is to review genotypic ART through a One Health lens, provide guidance for use, and describe current limitations for the diagnostic utility of molecular methods of ART in veterinary medicine.

In optimal clinical practice, AST is performed when it is challenging to predict susceptibility to a specific antimicrobial due to the potential for acquired resistance mechanisms. Antimicrobial susceptibility testing is also performed to assess the appropriateness of empirically chosen antimicrobials, to investigate cases in which patients exhibit inadequate responses to the initially selected therapy, or for surveillance purposes.3 Typically, first-line treatments are used for noninvasive infections until AST results are available.46 In the case of a severe or invasive infection, broad-spectrum antimicrobials are often chosen for empiric treatment, with de-escalation and optimized treatment choices recommended once laboratory testing is complete.5,7 Ultimately, treatment with an antimicrobial of the appropriate spectrum combined with AST results confirming susceptibility maximizes the likelihood of treatment success. This information is critical for patient care and a key component of antimicrobial stewardship.7

Conventional (Phenotypic) Culture and AST

Phenotypic culture and AST are the current gold-standard test methods for identification of clinically significant bacterial pathogens and are critical management tools that aid clinical decision-making.3 Bacterial culture is necessarily slow, requiring on average 24 hours for most bacteria, due to the time required for visible growth. Culture plates are evaluated, and bacteria, if present, are identified using various methods such as biochemical tests, carbohydrate utilization, matrix-assisted laser desorption ionization time-of-flight mass spectrometry, PCR, or DNA sequencing. Some bacteria may be primary pathogens in a particular body site, such as a pure culture of Escherichia coli in a sterile body site of an animal with compatible clinical signs, in contrast to E coli in a fecal culture, which is normal and expected in a healthy individual. Microbiologists evaluate growth critically to ensure a correct interpretation of significance in the clinical context. Often, they must sort through contaminating commensal microbiota from mucosa, skin surfaces, or other tissues to identify primary pathogens or to determine whether there is an apparent overgrowth of commensal microbiota that may represent opportunistic infection, such as Staphylococcus pseudintermedius from a skin source in a dog (Figure 1).8 In general, few commensals are associated with significant infections and reporting all identified bacteria from such cultures should be avoided. If a bacterium is identified in a clinical context that supports likely involvement in the disease process, it is then further subjected to AST depending on its likelihood to develop resistance.

Figure 1
Figure 1

Examples of bacterial cultures from animals. These cultures are assessed by trained microbiologists to determine whether a potential pathogen is present in a clinical sample. The microbiologist assesses whether the microbiota present are appropriate for a particular body site or whether contamination occurred during collection and any potential pathogen is present. A—This is an example of a sample that was likely contaminated during collection and was reported as “mixed bacterial growth.” Different colony types (species) of bacteria are present at each arrow. Reporting all bacteria present and performing antimicrobial susceptibility testing (AST) on bacteria from this sample would be poor antimicrobial stewardship and would lead to overtreatment. B—This is a pure culture of Serratia marcescens from a wound. This is a clinically significant bacterium, and AST is appropriate.

Citation: Journal of the American Veterinary Medical Association 262, 3; 10.2460/javma.23.12.0687

Antimicrobial susceptibility testing methods determine whether a bacterium’s growth is inhibited in the presence of an antimicrobial agent and are considered a phenotypic test. There are multiple methods, including dilution-based tests and diffusion-based tests resulting in MIC or zone diameter (ZD) data. In vitro susceptibility data are then compared to published interpretive guidelines, which rely on testing to be performed using a standard method and appropriate quality control, such as those developed by the Clinical and Laboratory Standards Institute (CLSI) or European Committee on Antimicrobial Susceptibility Testing.6,810 Deviations from these standardized protocols may result in a bacterium appearing more or less resistant to an antimicrobial, leading to less accurate prediction of treatment response.9,10

Phenotypic AST interpretive categories, sometimes referred to as clinical breakpoints, are the result of complex assessments of animal species-specific pharmacokinetic data; concentrations of the antimicrobial agent needed to inhibit the growth of a specific family, genera, or species of bacteria; properties of the antimicrobial agent (time vs concentration-dependent activity); and population modeling. Ideally, interpretive criteria are validated with clinical outcomes data; however, these studies are infrequent in veterinary medicine. The CLSI is the leading standards organization in the US for phenotypic AST methods and clinical breakpoints.9 Although not currently included in the veterinary AST standards, CLSI VET01S, the analogous human AST document, CLSI M100, includes interpretations for molecular ART targets for specific clinical utility and troubleshooting procedures.11,12 Neither CLSI VET01S nor CLSI M100 currently has approved guidelines for the use of sequence-based test methods for prediction of AMR from clinical samples.

Molecular Approaches

Polymerase chain reaction

Because this report focuses on the potential limitations of genotypic ART, first it is important to discuss genotypic test methods to provide context and acknowledge strengths of the methodology (Figure 2; Table 1). Polymerase chain reaction is a widely used diagnostic technique for detection of infectious pathogens, AMR genes that confer antimicrobial resistance, and oncology markers. Polymerase chain reaction has become an accepted replacement for some conventional laboratory techniques such as virus isolation, identification of fastidious organisms that will not grow by conventional means, or to differentiate between field and vaccine strains of various pathogens.13,14 It is also important for toxin typing or identification of virulence-associated genes for a variety of veterinary bacterial pathogens, including Clostridium perfringens and E coli, in which there are strong associations between gene presence and pathogenicity.15,16

Figure 2
Figure 2

Comparison of the use of PCR, whole genome sequencing, and metagenomic sequencing on a surgically collected tissue. *Some laboratories refer to 16s rRNA sequencing as metagenomic sequencing. This type of sequencing requires no knowledge of the target and yields a biased assessment of bacterial DNA content in a sample. Created with BioRender.com.

Citation: Journal of the American Veterinary Medical Association 262, 3; 10.2460/javma.23.12.0687

Table 1

Summary of molecular-based antimicrobial resistance test methods.

Properties PCR WGS Metagenomic sequencing
Target specific Yes Noa No
A priori knowledge required Yes No No
Rapid (< 12 h for results) Yes No No
Cost < $75 (USD) Yes No No
Diagnostic standards for quality Yes No No
Can detect DNA of fastidious or unculturable bacteria Yes Yes Yes
Database required for result interpretation No Yes Yes
Should be performed in conjunction with culture and phenotypic AST Yes Yes Yes

AST = Antimicrobial susceptibility testing. USD = US dollars. WGS = Whole genome sequencing.

aTargeted approaches may be used.

In general, PCR is designed to be very specific and requires large collections of strains and/or clinical samples to validate, ensuring there is well-documented analytical and diagnostic sensitivity and specificity. Errors can also occur due to sequence variability in the nucleic acids of the target organism and/or reaction inhibition due to substances present in the clinical sample, leading to false-negative results. Detection is limited to the pathogens or targets the assay is designed to detect, limiting test utility when an alternative pathogen is not suspected. Polymerase chain reaction assays may be designed to detect multiple targets, referred to as a multiplex PCR, which can reduce the chance of missing important disease-causing agents and manage costs, especially when designed around detection of common agents associated with a disease syndrome. In veterinary medicine, PCR is used to detect primary pathogens, such as Tritrichomonas foetus, Salmonella, bovine respiratory syncytial virus, or Cryptosporidium parvum in clinically ill animals. Alternatively, PCR is also useful for detecting pathogens that require long incubation times or are difficult to grow by conventional methods, such as many fungal or mycobacterial species.

Sequencing

Under the blanket term “DNA sequencing” there are multiple approaches available, including whole genome sequencing (WGS), targeted sequencing, and metagenomic (shotgun) sequencing approaches. Like PCR, sequencing is also useful for detecting or identifying organisms with long incubation times or unusual growth requirements or that lack traditional identification tools.17 Whole genome sequencing is often used for sequencing a single bacterial species to minimize genome variability in the sample. Genes present from WGS of a pure isolate can be assessed without reliance on a priori knowledge of targets, but skilled analysis is time-consuming. Beyond AMR genes, WGS data can be used to assess virulence and housekeeping genes and to compare bacterial strains from outbreaks at the level of single nucleotide polymorphisms or sequence types.

Differing from WGS, targeted sequencing utilizes specific primers in the test workflow to identify particular genes of interest. Similar to multiplex PCR, targeted sequencing can be offered as part of a panel to detect specific regions of a particular virus, intracellular bacterium, or oncology marker in a single reaction.1820 However, when using this approach the user is limited to examining the genes of interest rather than the entire genome of interest.

Alternatively, metagenomic approaches sequence all genomes present in a sample, including host genomic material. Because AMR genes may be present in the commensal microbiota of healthy and sick individuals (humans and animals) or the environment, metagenomic methods can be used to detect numerous AMR and virulence genes that enhance understanding of complex AMR epidemiology across One Health surveillance sectors. However, metagenomic sequencing methods may face challenges of low sensitivity and detection bias. They require a robust analysis pipeline or global standardization methods capable of differentiating commensal and pathogenic microbial genes and that exclude host genomes to ensure accurate results.14,21 Unusual animal species will make this analysis more difficult because reference genomes that aid in excluding the host genome rely on well-curated databases; these species are less represented than domestic animal species even in the largest databases.

A major advantage of WGS is that it is a powerful surveillance tool for the detection of AMR genes and allows for epidemiologic evaluation of strains, adding richness to phenotypic AST data.22 Whole genome sequencing for surveillance of nosocomial pathogens and high-level resistance genes has been employed with success in human medicine and increasingly in veterinary medicine, but it is one of many tools to support antimicrobial stewardship and thus cannot replace other approaches and phenotypic methodologies.

Clinical Applications of Genotypic ART in Human Medicine

Targeted molecular assays, such as PCR, are used routinely in human medicine as a screening tool to detect colonization with common nosocomial pathogens that harbor resistance mechanisms of concern, such as methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, or carbapenemase-producing carbapenem-resistant Enterobacterales, particularly before hospitalization for elective procedures or in long-term care facilities.1 These tools help prevent the spread of difficult-to-treat and potentially deadly nosocomial infections among hospitalized patients; detection of these organisms triggers use of infection-control practices like contact precautions and isolation procedures. Targeted PCR tests are also used to identify bacteria and critical AMR determinants from blood culture–collection systems before conventional culture results are available.1 Additionally, syndromic PCR panels are used to identify various pathogens and AMR genes (BIOFIRE, BioFire Diagnostics; and Verigene, Nanosphere), increasing the likelihood of identifying a significant pathogen or resistance gene faster than with phenotypic methods and reducing the need for PCR testing to detect a single pathogen. This reduces the impact of the specificity of individual PCR testing. The accuracy and rapid turnaround time of PCR provides a powerful tool that a clinician or other public health professional (in the case of disease surveillance) can act upon quickly to improve patient outcomes, reduce the spread of resistant bacteria, and minimize inappropriate treatment with critically important antimicrobials.23

Although widely used for surveillance and research, sequencing is not currently a common tool for routine clinical diagnostic use. This is primarily attributed not only to the cost of testing but mainly to challenges in data interpretation.1,24 Sequencing in human clinical diagnostics is primarily conducted to analyze bacterial housekeeping genes for retrospective surveillance, in conjunction with phenotypic culture and AST.24,25 Furthermore, clinical microbiology experts with a deep understanding of sequencing must be available to aid clinicians with result interpretation, appropriate use, and limitations of such tests.1

Increasingly, sequencing is used to detect novel or rare infections; these methods have shown utility in identifying pathogens in culture-negative samples in cases for which infection is highly suspected.26 However, interpretation of sequencing results from clinical cases with negative culture requires significant expertise to ensure the presence of novel DNA signals are not misinterpreted or overinterpreted due to potential assay or bioinformatic bias and that the same challenges with interpretation of results is given appropriate consideration.

Limitations

Molecular methods are highly specific for detecting AMR genes; however, the relationship between genetic detection and phenotypic expression of AMR is not validated in many cases. There is an incredible diversity of resistance mechanisms that bacteria employ, and not all have a known genetic predictor or are a mixture of mechanisms creating an additive effect, such as non–carbapenemase-producing carbapenem-resistant gram-negative bacteria.27 In addition, complex interactions between genes or epistatic effects are only partially known.28 Antimicrobial-resistance detected by sequence-based methods essentially differentiate wild-type (lack of acquired resistance genes) and non–wild-type (AMR gene–possessing) bacterial populations.29 This interpretation is similar to the MIC or ZD distribution data that can segregate populations into a present/absent category on the basis of an epidemiologic cutoff MIC/ZD evaluation but contains no clinically predictive information. This approach does not necessarily help the selection of an appropriate antimicrobial treatment regimen, highlighting a key limiting factor for genotypic approaches.

The detection of a resistance gene does not guarantee clinically relevant expression because the mere presence of a resistance gene does not indicate gene functionality. Additional challenges arise when pharmacokinetic-pharmacodynamic relationships for certain antimicrobial agent classes do not correlate with genetic resistance mechanisms. An example of this lies in the differences between human MIC breakpoints in CLSI M100 and canine MIC breakpoints for ampicillin for Enterobacterales in CLSI VET01S.11,12 The canine resistance breakpoint is 32-fold lower (more conservative) than the human resistance breakpoint. The canine breakpoint is lower because oral dosing regimens do not reach plasma concentrations high enough to reach wild-type E coli MIC ranges, which are unrelated to bacterial AMR gene presence and expression. There is an additional ampicillin susceptible breakpoint for Enterobacterales isolated from canine urine as the drug concentrates in urine to levels reaching realistic MICs for E coli, which is in accordance with the human susceptible breakpoint and the epidemiologic cutoff value for wild-type E coli. The canine urine and human breakpoints are more useful for predicting acquired resistance in the bacterial isolate but this information is not widely understood, and if the systemic canine breakpoint was used for correlation with genotypic data, it would be grossly misleading. Interestingly, breakpoints that call all bacteria resistant regardless of the presence of acquired resistance, as in the case of the lower ampicillin canine breakpoint, are unique to veterinary medicine and complicate efforts in surveillance and understanding of AMR. Currently, there are no clinical breakpoint approaches that support the use of genotypic ART methods without confirmation by culture and phenotypic AST.

Molecular tests detect the nucleic acids of all organisms present in a sample, both viable and not. This means that without confirmation by phenotypic bacterial culture and AST, treatment could be prolonged even if no live bacteria are present or initiated when not clinically indicated. For example, a strategy used in bovine respiratory disease (BRD) involves PCR-based detection of both potential pathogens and genes that are highly correlated with the isolation of a pathogen that is resistant to macrolides and/or tetracycline class antimicrobials.30,31 Outbreaks of BRD typically involve large populations of animals, so rapid information on the presence of AMR gene determinants may be helpful for therapy initiation; however, confirmation with pathologic examination, culture of appropriate lower airway specimens, and subsequent AST and interpretation using BRD-specific breakpoints in CLSI VET01S are required to confirm the significance of the molecular testing and avoid the challenges described above with gene detection alone.

Resistance genes can be detected in a polymicrobial patient sample but may not be associated with a clinically relevant pathogen. For example, direct sequencing or PCR of an animal wound sample may detect S pseudintermedius, a mecA gene (which confers methicillin-resistance in Staphylococcus species), and a commensal Staphylococcus species. Without the phenotypic AST of this sample, there is no way of determining whether the mecA gene is associated with the S pseudintermedius or the commensal Staphylococcus species (Figure 3). This could prompt the clinician to treat inappropriately for methicillin-resistant S pseudintermedius when the more probable pathogen may in fact be susceptible to first-line treatments.1

Figure 3
Figure 3

An example of genotypic and phenotypic test results that may be obtained from a dog wound. Treatment decisions should not be based on the genotypic antimicrobial resistance testing (ART) results alone. In this case, ART results suggest that a first-generation cephalosporin would not be appropriate for treating this patient’s infection due to the presence of the mecA gene, which confers resistance to β-lactam/β-lactam combination agents and first-, second-, and third-generation cephalosporins. However, with ART we cannot differentiate which organism carries the mecA gene. On the basis of the phenotypic AST results, we observe that the primary pathogen, Staphylococcus pseudintermedius, is actually susceptible to cephalosporins based on the oxacillin test result. Use of ART results alone increases the risk of overtreatment with more reserved classes of antimicrobials that are critically important for human and veterinary health. Note that “with mixed bacterial growth” in the phenotypic culture example includes contaminant and commensal organisms with low potential for pathogenicity such as Staphylococcus epidermidis, which was also detected by genotypic testing and may harbor the mecA gene. Created with BioRender.com.

Citation: Journal of the American Veterinary Medical Association 262, 3; 10.2460/javma.23.12.0687

Currently, sequencing is not a cost-effective replacement for conventional diagnostics, such as culture, serology, or PCR. There is a high cost associated with the purchase of the instrumentation, reagents, consumables, data storage, and personnel time required for hands-on work, data analysis, and interpretation. In addition, the training and specialization required to perform and interpret these types of tests are not well developed in veterinary medicine and are quite costly for most diagnostic laboratories. Sequencing may be attractive because some bacterial or viral pathogens could be identified in less than an 8-hour work shift with a well-developed assay and analysis pipeline; however, that is not routinely feasible or practical outside of an outbreak response or research program.3235 The hands-on personnel and turnaround times for appropriately analyzed samples would rarely shorten the time to test results compared to phenotypic culture and AST.

In addition, bacteria detected using a metagenomic approach may be present on the skin of a normal animal as commensal or contaminating microbiota. In general, the microbiologist assesses the quality of the specimen, probability of contaminating microbiota, supportive clinical data, and relative number of bacteria grown in phenotypic culture methods to determine probable significance in an infection. Recent papers have suggested that results can be interpreted on the basis of relative abundance, but there are issues with these approaches, including false overrepresentation of certain bacterial species and resistance genes due to sequencing bias effects.21,36 Currently, there are no standards for determining amounts consistent with commensal microbiota versus the relative abundance that constitutes a potential pathogen for most organisms. Interpretation of sequence-based results of exotic animal species is further complicated by the limited literature available regarding standard pathogens or commensals of these species. In contrast, semiquantitative evaluation of commensal versus pathogen is well established for conventional culture methods.37

Databases used for analyzing sequencing test results are only as good as the quality of information that has been included in the tool.3841 Bioinformatics tools require coding knowledge and the development of data analysis pipelines that are validated to produce the correct results when a gene is present (or not) in a sample. The selection of bioinformatics pipelines for data analysis can impact which resistance markers are detected in a sample.42 In addition, the quality of coverage will affect which resistance markers are detected.42,43 Although robust and improving daily, most bioinformatics tools are not developed enough for routine “point and click” use without a specially trained workforce.

Genotypic-based test methods are prone to inhibition by substances present in a clinical specimen, and the test may cross-react with commensal microbiota present in a sample.13,44 Complete test method validation, which ensures that a test is performing and detecting what it is designed to detect, requires testing on all sample types and all animal species for which the test is intended. Some body fluids contain more inhibitors than others, and commensal microbiota vary depending on the body site; a properly validated test ensures that the test is not inhibited by normal substances at that body site and does not cross-react with commensal microbiota present in that site.13,44 Again, standards for validation and verification of sequence-based test methods have not been developed, so the quality of test results cannot be ensured.

Laboratory Quality Standards

Veterinary diagnostic laboratories should adhere to quality control (QC) standards appropriate for the type of testing offered as a diagnostic service to veterinary clients. Veterinary diagnostic laboratories may adhere to standards set forth by a third-party accreditation body such as the American Association of Veterinary Laboratory Diagnosticians or a standards-setting organization such as the International Organization for Standardization. However, unlike laboratories performing testing on human samples, laboratory accreditation is not compulsory to offer testing on animal samples. This means that test offerings from unaccredited laboratories may not guarantee quality, stringent method validation procedures, accuracy, or reproducibility of test results. Laboratory accreditation is a reasonable assurance to a client that appropriate validation procedures, quality checks, and technical competence of personnel are achieved and maintained.45 Whether affiliated with a hospital, university, or privately owned, veterinary diagnostic laboratories are responsible for ensuring QC of all tests offered.

Genotypic tests performed in veterinary diagnostic laboratories are largely laboratory developed; those offered by unaccredited laboratories are not evaluated by an outside agency, and claims of utility by the testing laboratory or manufacturer are not necessarily validated.13 Several publications suggest QC parameters, but no international consensus standards have been set for sequence-based pathogen detection and ART.42,43 If sequence-based ART is being provided as a service, veterinarians should understand that interpretation of results is not based on any internationally accepted standards for quality; consequently, clinical utility is unknown. Quality-control parameters and acceptance criteria are necessary for all properly performed laboratory procedures; therefore, in the absence of such standards, those offering this service are doing so with in-house laboratory-developed test methods and unregulated QC standards. Such testing requires reasonable scrutinization of test result accuracy.

Diagnostic stewardship is a form of QC in laboratory medicine; it is the principle of judicious use of laboratory tests, appropriate test interpretation, and communication of limitations to users of their diagnostic test methods to improve patient outcomes and minimize cost (Figure 4). The choice of diagnostic tools requires careful consideration of test quality and clinical utility of the results. Any diagnostic test method should yield a result that is actionable and may be predictive of treatment success or failure. Currently, human and veterinary medical literature does not support using genotypic ART as a suitable replacement for phenotypic AST methods. If a laboratory offers genotypic ART, a disclaimer and description of potential limitations must be included for the client to interpret the report properly.1

Figure 4
Figure 4

Standard diagnostic stewardship workflow promoting appropriate timing, sample collection, and test selection. Created with BioRender.com.

Citation: Journal of the American Veterinary Medical Association 262, 3; 10.2460/javma.23.12.0687

As mentioned previously, no international consensus standards exist for genotypic ART. However, CLSI M100, a standard document for the performance of AST on bacteria isolated from human patients, suggests appropriate molecular targets for confirming specific resistance genes. This appendix also suggests applications of molecular methods from direct patient samples; however, as of the 33rd edition, no standards for sequence-inferred ART have been recommended. The CLSI working group responsible for future revision of CLSI VET01S is planning to draft similar guidance suggesting appropriate molecular targets to confirm specific resistance genes and applications of molecular methods from direct patient samples for veterinary use.

Proposed Clinical Veterinary Applications of Genotypic ART in Support of One Health

A primary benefit of genotypic ART methods in a clinical setting is the rapidity of test results and the immediate action taken, such as therapy de-escalation, empiric therapy revision when resistant organisms are detected, and isolation precautions. Genotypic ART methods have not been widely adopted for routine use in veterinary medicine, likely due to the high cost of these test methods and the technical expertise needed. Unfortunately, nosocomial infections in veterinary hospitals are expected to rise as more patients require intensive care.46,47 As in human medicine, using genotypic ART for blood cultures and other body fluids in tandem with phenotypic culture and AST may benefit patient care in critical cases.1

Until there is standardization for genotypic ART that includes defined sample types, library preparation method/kit, sequencing platform, and analysis pipeline, laboratories are free to use whatever library prep kits and sequencer and any number of bioinformatics tools strung together in a pipeline.42,4852 Not all workflows are created equal, and even similar methods may not produce the same results.42,53 Even if a suitable method is developed, clinical utility and patient outcome studies are scarce, thus warranting further investigation.

Genotypic ART, in conjunction with phenotypic AST data, is valuable to collect and use for surveillance and research purposes because it contributes to a better understanding of AMR genotype-phenotype correlation for relevant veterinary pathogens. In North America, the FDA’s Veterinary Laboratory Investigation and Response Network and the USDA’s National Animal Health Laboratory Network have established AMR surveillance programs to collect phenotypic and sequence-based AMR gene detection on veterinary pathogens from clinically ill animals.54 Such surveillance networks help further knowledge of resistance phenotypes and genotypes present in bacteria collected from animal hosts.22,47,55,56

Adoption of genotypic ART directly from the patient sample as it currently stands will likely drive AMR due to inappropriate treatment after nonspecific resistance gene detection. Detection of a gene does not mean that the primary cause of infection possesses functional copies of that resistance gene, and it may in fact belong to a commensal bacterium that would not require treatment. With current sample preparation steps and data analysis pipelines for sequence-based AST, it is also likely that turnaround time will not be faster than conventional culture and phenotypic AST.

Finally, the current literature does not support genotypic ART results to guide clinical decision-making. Further clinical studies are necessary to determine the utility and whether these test methods can improve the quality of care. Because current usage is limited, clinicians may need training and substantial assistance determining the significance of results reported on a genotypic ART report.1,57 Reports should include interpretations of the results and evidence-based treatment guidance, if known.1 Continued resources for veterinary-specific national surveillance programs and AMR research should be top priorities for funding agencies.

Acknowledgments

The authors are grateful to Ms. Lori Selden for her support.

Disclosures

Many of the authors are voluntary (unpaid) members of the Clinical and Laboratory Standards Institute Veterinary Antimicrobial Susceptibility Testing subcommittee. The views expressed in this article do not necessarily reflect those of Dr. Li’s affiliation with Health Canada.

No AI-assisted technologies were used in the generation of this manuscript.

Funding

The authors have nothing to disclose.

References

  • 1.

    Banerjee R, Patel R. Molecular diagnostics for genotypic detection of antibiotic resistance: current landscape and future directions. JAC Antimicrob Resist. 2023;5(1):dlad018. doi:10.1093/jacamr/dlad018

    • Search Google Scholar
    • Export Citation
  • 2.

    Humphries RM. Update on susceptibility testing: genotypic and phenotypic methods. Clin Lab Med. 2020;40(4):433-446. doi:10.1016/j.cll.2020.08.002

    • Search Google Scholar
    • Export Citation
  • 3.

    Clinical and Laboratory Standards Institute. Understanding Susceptibility Test Data as a Component of Antimicrobial Stewardship in Veterinary Settings. CLSI VET09. Clinical and Laboratory Standards Institute; 2019.

    • Search Google Scholar
    • Export Citation
  • 4.

    Lappin MR, Blondeau J, Boothe D, et al. Antimicrobial use guidelines for treatment of respiratory tract disease in dogs and cats: antimicrobial guidelines working group of the International Society for Companion Animal Infectious Diseases. J Vet Intern Med. 2017;31(2):279-294. doi:10.1111/jvim.14627

    • Search Google Scholar
    • Export Citation
  • 5.

    Weese JS, Blondeau J, Boothe D, et al. International Society for Companion Animal Infectious Diseases (ISCAID) guidelines for the diagnosis and management of bacterial urinary tract infections in dogs and cats. Vet J. 2019;247:8-25. doi:10.1016/j.tvjl.2019.02.008

    • Search Google Scholar
    • Export Citation
  • 6.

    Hillier A, Lloyd DH, Weese JS, et al. Guidelines for the diagnosis and antimicrobial therapy of canine superficial bacterial folliculitis (Antimicrobial Guidelines Working Group of the International Society for Companion Animal Infectious Diseases). Vet Dermatol. 2014;25(3):163-e43. doi:10.1111/vde.12118

    • Search Google Scholar
    • Export Citation
  • 7.

    Allerton F, Prior C, Bagcigil AF, et al. Overview and evaluation of existing guidelines for rational antimicrobial use in small-animal veterinary practice in Europe. Antibiotics (Basel). 2021;10(4):409. doi:10.3390/antibiotics10040409

    • Search Google Scholar
    • Export Citation
  • 8.

    Timofte D, Broens EM, Guardabassi L, et al. European Network for Optimization of Veterinary Antimicrobial Treatment (ENOVAT); ESCMID Study Group for Veterinary Microbiology (ESGVM); European College of Veterinary Microbiology (ECVM); European Association of Veterinary Laboratory Diagnosticians (EAVLD). Driving laboratory standardization of bacterial culture and antimicrobial susceptibility testing in veterinary clinical microbiology in Europe and beyond. J Clin Microbiol. 2021;59(6):e02572-e20. doi:10.1128/JCM.02572-20

    • Search Google Scholar
    • Export Citation
  • 9.

    Feßler AT, Wang Y, Burbick C, et al. Antimicrobial susceptibility testing in veterinary medicine: performance, interpretation of results, best practices and pitfalls. One Health Adv. 2023;1:26. doi:10.1186/s44280-023-00024-w

    • Search Google Scholar
    • Export Citation
  • 10.

    Schwarz S, Silley P, Simjee S, et al. Editorial: assessing the antimicrobial susceptibility of bacteria obtained from animals. J Antimicrob Chemother. 2010;65(4):601-604. doi:10.1093/jac/dkq037

    • Search Google Scholar
    • Export Citation
  • 11.

    Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals. 6th ed. CLSI Supplement VET01S. Clinical and Laboratory Standards Institute; 2023.

    • Search Google Scholar
    • Export Citation
  • 12.

    Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing. 33rd Edition. CLSI Supplement M100. Clinical and Laboratory Standards Institute; 2023.

    • Search Google Scholar
    • Export Citation
  • 13.

    Daniels JB. Molecular diagnostics for infectious disease in small animal medicine: an overview from the laboratory. Vet Clin North Am Small Anim Pract. 2013;43(6):1373-1384, vii. doi:10.1016/j.cvsm.2013.07.006

    • Search Google Scholar
    • Export Citation
  • 14.

    Compton SR. PCR and RT-PCR in the diagnosis of laboratory animal infections and in health monitoring. J Am Assoc Lab Anim Sci. 2020;59(5):458-468. doi:10.30802/AALAS-JAALAS-20-000008

    • Search Google Scholar
    • Export Citation
  • 15.

    Rood JI, Adams V, Lacey J, et al. Expansion of the Clostridium perfringens toxin-based typing scheme. Anaerobe. 2018;53:5-10. doi:10.1016/j.anaerobe.2018.04.011

    • Search Google Scholar
    • Export Citation
  • 16.

    Casey TA, Bosworth BT. Design and evaluation of a multiplex polymerase chain reaction assay for the simultaneous identification of genes for nine different virulence factors associated with Escherichia coli that cause diarrhea and edema disease in swine. J Vet Diagn Invest. 2009;21(1):25-30. doi:10.1177/104063870902100104

    • Search Google Scholar
    • Export Citation
  • 17.

    Votintseva AA, Bradley P, Pankhurst L, et al. Same-day diagnostic and surveillance data for tuberculosis via whole-genome sequencing of direct respiratory samples. J Clin Microbiol. 2017;55(5):1285-1298. doi:10.1128/JCM.02483-16

    • Search Google Scholar
    • Export Citation
  • 18.

    Kattoor JJ, Nikolai E, Qurollo B, Wilkes RP. Targeted next-generation sequencing for comprehensive testing for selected vector-borne pathogens in canines. Pathogens. 2022;11(9):964. doi:10.3390/pathogens11090964

    • Search Google Scholar
    • Export Citation
  • 19.

    Anis E, Ilha MRS, Engiles JB, Wilkes RP. Evaluation of targeted next-generation sequencing for detection of equine pathogens in clinical samples. J Vet Diagn Invest. 2021;33(2):227-234. doi:10.1177/1040638720978381

    • Search Google Scholar
    • Export Citation
  • 20.

    Harrison BM, Loukopoulos P. Genomics and transcriptomics in veterinary oncology. Oncol Lett. 2021;21(4):336. doi:10.3892/ol.2021.12597

  • 21.

    Dulanto Chiang A, Dekker JP. From the pipeline to the bedside: advances and challenges in clinical metagenomics. J Infect Dis. 2020;221(suppl 3):S331-S340. doi:10.1093/infdis/jiz151

    • Search Google Scholar
    • Export Citation
  • 22.

    Hendriksen RS, Bortolaia V, Tate H, Tyson GH, Aarestrup FM, McDermott PF. Using genomics to track global antimicrobial resistance. Front Public Health. 2019;7:242. doi:10.3389/fpubh.2019.00242

    • Search Google Scholar
    • Export Citation
  • 23.

    WHO Advisory Group on Integrated Surveillance of Antimicrobial Resistance. Critically Important Antimicrobials for Human Medicine. 6th revision. WHO; 2019. Accessed October 28, 2023. https://iris.who.int/bitstream/handle/10665/312266/9789241515528-eng.pdf?sequence=1

    • Search Google Scholar
    • Export Citation
  • 24.

    Avershina E, Khezri A, Ahmad R. Clinical diagnostics of bacterial infections and their resistance to antibiotics-current state and whole genome sequencing implementation perspectives. Antibiotics (Basel). 2023;12(4):781. doi:10.3390/antibiotics12040781

    • Search Google Scholar
    • Export Citation
  • 25.

    Wilson MR, Sample HA, Zorn KC, et al. Clinical metagenomic sequencing for diagnosis of meningitis and encephalitis. N Engl J Med. 2019;380(24):2327-2340. doi:10.1056/NEJMoa1803396

    • Search Google Scholar
    • Export Citation
  • 26.

    Hong HL, Flurin L, Thoendel MJ, et al. Targeted versus shotgun metagenomic sequencing-based detection of microorganisms in sonicate fluid for periprosthetic joint infection diagnosis. Clin Infect Dis. 2023;76(3):e1456-e1462. doi:10.1093/cid/ciac646

    • Search Google Scholar
    • Export Citation
  • 27.

    CRE technical information. CDC. Accessed October 28, 2023. https://www.cdc.gov/hai/organisms/cre/technical-info.html

  • 28.

    Fulton RW, Confer AW. Laboratory test descriptions for bovine respiratory disease diagnosis and their strengths and weaknesses: gold standards for diagnosis, do they exist? Can Vet J. 2012;53(7):754-761.

    • Search Google Scholar
    • Export Citation
  • 29.

    Tyson GH, Zhao S, Li C, et al. Establishing genotypic cutoff values to measure antimicrobial resistance in Salmonella. Antimicrob Agents Chemother. 2017;61(3):e02140-16. doi:10.1128/AAC.02140-16

    • Search Google Scholar
    • Export Citation
  • 30.

    Dutta E, Loy JD, Deal CA, et al. Development of a multiplex real-time PCR assay for predicting macrolide and tetracycline resistance associated with bacterial pathogens of bovine respiratory disease. Pathogens. 2021;10(1):64. doi:10.3390/pathogens10010064

    • Search Google Scholar
    • Export Citation
  • 31.

    Loy JD, Leger L, Workman AM, Clawson ML, Bulut E, Wang B. Development of a multiplex real-time PCR assay using two thermocycling platforms for detection of major bacterial pathogens associated with bovine respiratory disease complex from clinical samples. J Vet Diagn Invest. 2018;30(6):837-847. doi:10.1177/1040638718800170

    • Search Google Scholar
    • Export Citation
  • 32.

    O’Donnell VK, Grau FR, Mayr GA, Sturgill Samayoa TL, Dodd KA, Barrette RW. Rapid sequence-based characterization of African swine fever virus by use of the Oxford Nanopore MinION sequence sensing device and a companion analysis software tool. J Clin Microbiol. 2019;58(1):e01104-19. doi:10.1128/JCM.01104-19

    • Search Google Scholar
    • Export Citation
  • 33.

    Bold D, Souza-Neto JA, Gombo-Ochir D, et al. Rapid identification of ASFV, CSFV and FMDV from Mongolian outbreaks with MinION short amplicon sequencing. Pathogens. 2023;12(4):533. doi:10.3390/pathogens12040533

    • Search Google Scholar
    • Export Citation
  • 34.

    Taxt AM, Avershina E, Frye SA, Naseer U, Ahmad R. Rapid identification of pathogens, antibiotic resistance genes and plasmids in blood cultures by nanopore sequencing. Sci Rep. 2020/05/06 2020;10:7622. doi:10.1038/s41598-020-64616-x

    • Search Google Scholar
    • Export Citation
  • 35.

    McLaughlin HP, Bugrysheva JV, Conley AB, et al. Rapid nanopore whole-genome sequencing for anthrax emergency preparedness. Emerg Infect Dis. 2020;26(2):358-361. doi:10.3201/eid2602.191351

    • Search Google Scholar
    • Export Citation
  • 36.

    Damerum A, Malka S, Lofgren N, Vecere G, Krumbeck JA. Next-generation DNA sequencing offers diagnostic advantages over traditional culture testing. Am J Vet Res. 2023;84(8):ajvr.23.03.0054. doi:10.2460/ajvr.23.03.0054

    • Search Google Scholar
    • Export Citation
  • 37.

    Leber AL, ed. Clinical Microbiology Procedures Handbook. 4th ed. John Wiley & Sons Inc; 2016.

  • 38.

    Lupo V, Van Vlierberghe M, Vanderschuren H, Kerff F, Baurain D, Cornet L. Contamination in reference sequence databases: time for divide-and-rule tactics. Perspective. Front Microbiol. 2021;12:755101. doi:10.3389/fmicb.2021.7551017

    • Search Google Scholar
    • Export Citation
  • 39.

    Wang C, Xia Y, Liu Y, et al. CleanSeq: a pipeline for contamination detection, cleanup, and mutation verifications from microbial genome sequencing data. Appl Sci (Basel). 2022;12(12):6209. doi:10.3390/app12126209

    • Search Google Scholar
    • Export Citation
  • 40.

    Goig GA, Blanco S, Garcia-Basteiro AL, Comas I. Contaminant DNA in bacterial sequencing experiments is a major source of false genetic variability. BMC Biol. 2020;18(1):24. doi:10.1186/s12915-020-0748-z

    • Search Google Scholar
    • Export Citation
  • 41.

    Wally N, Schneider M, Thannesberger J, et al. Plasmid DNA contaminant in molecular reagents. Sci Rep. 2019;9(1):1652. doi:10.1038/s41598-019-38733-1

    • Search Google Scholar
    • Export Citation
  • 42.

    Alvarez Narvaez S, Shen Z, Yan L, et al. Optimized conditions for Listeria, Salmonella and Escherichia whole genome sequencing using the Illumina iSeq100 platform with point-and-click bioinformatic analysis. PLoS One. 2022;17(11):e0277659. doi:10.1371/journal.pone.0277659

    • Search Google Scholar
    • Export Citation
  • 43.

    Ellington MJ, Ekelund O, Aarestrup FM, et al. The role of whole genome sequencing in antimicrobial susceptibility testing of bacteria: report from the EUCAST Subcommittee. Clin Microbiol Infect. 2017;23(1):2-22. doi:10.1016/j.cmi.2016.11.012

    • Search Google Scholar
    • Export Citation
  • 44.

    Toohey-Kurth K, Reising MM, Tallmadge RL, et al. Suggested guidelines for validation of real-time PCR assays in veterinary diagnostic laboratories. J Vet Diagn Invest. 2020;32(6):802-814. doi:10.1177/1040638720960829

    • Search Google Scholar
    • Export Citation
  • 45.

    Toohey-Kurth KL, Mulrooney DM, Hinkley S, et al. Best practices for performance of real-time PCR assays in veterinary diagnostic laboratories. J Vet Diagn Invest. 2020;32(6):815-825. doi:10.1177/1040638720962076

    • Search Google Scholar
    • Export Citation
  • 46.

    Stull JW, Weese JS. Hospital-associated infections in small animal practice. Vet Clin North Am Small Anim Pract. 2015;45(2):217-233, v. doi:10.1016/j.cvsm.2014.11.009

    • Search Google Scholar
    • Export Citation
  • 47.

    Cole SD, Peak L, Tyson GH, Reimschuessel R, Ceric O, Rankin SC. New Delhi metallo-β-lactamase-5-producing Escherichia coli in companion animals, United States. Emerg Infect Dis. 2020;26(2):381-383. doi:10.3201/eid2602.191221

    • Search Google Scholar
    • Export Citation
  • 48.

    Feldgarden M, Brover V, Gonzalez-Escalona N, et al. AMRFinderPlus and the Reference Gene Catalog facilitate examination of the genomic links among antimicrobial resistance, stress response, and virulence. Sci Rep. 2021;11(1):12728. doi:10.1038/s41598-021-91456-0

    • Search Google Scholar
    • Export Citation
  • 49.

    Sayers EW, Beck J, Bolton EE, et al. Database resources of the National Center for Biotechnology Information. Nucleic Acids Res. 2021;49(D1):D10-D17. doi:10.1093/nar/gkaa892

    • Search Google Scholar
    • Export Citation
  • 50.

    Feldgarden M, Brover V, Haft DH, et al. Validating the AMRFinder tool and resistance gene database by using antimicrobial resistance genotype-phenotype correlations in a collection of isolates. Antimicrob Agents Chemother. 2019;63(11):00483-19. doi:10.1128/AAC.00483-19

    • Search Google Scholar
    • Export Citation
  • 51.

    Alcock BP, Huynh W, Chalil R, et al. CARD 2023: expanded curation, support for machine learning, and resistome prediction at the Comprehensive Antibiotic Resistance Database. Nucleic Acids Res. 2023;51(D1):D690-D699. doi:10.1093/nar/gkac920

    • Search Google Scholar
    • Export Citation
  • 52.

    Florensa AF, Kaas RS, Clausen PTLC, Aytan-Aktug D, Aarestrup FM. ResFinder - an open online resource for identification of antimicrobial resistance genes in next-generation sequencing data and prediction of phenotypes from genotypes. Microb Genom. 2022;8(1):000748. doi:10.1099/mgen.0.000748

    • Search Google Scholar
    • Export Citation
  • 53.

    Forry S, Servetas S, Kralj J, et al. Variability and bias in microbiome metagenomic sequencing: an interlaboratory study comparing experimental protocols. medRxiv. Forthcoming.

    • Search Google Scholar
    • Export Citation
  • 54.

    NARMS 2019 animal pathogen AMR data. FDA. Accessed October 28, 2023. https://www.fda.gov/animal-veterinary/national-antimicrobial-resistance-monitoring-system/2019-animal-pathogen-amr-data

    • Search Google Scholar
    • Export Citation
  • 55.

    Ceric O, Tyson GH, Goodman LB, et al. Enhancing the one health initiative by using whole genome sequencing to monitor antimicrobial resistance of animal pathogens: Vet-LIRN collaborative project with veterinary diagnostic laboratories in United States and Canada. BMC Vet Res. 2019;15(1):130. doi:10.1186/s12917-019-1864-2

    • Search Google Scholar
    • Export Citation
  • 56.

    Tyson GH, Ceric O, Guag J, et al. Genomics accurately predicts antimicrobial resistance in Staphylococcus pseudintermedius collected as part of Vet-LIRN resistance monitoring. Vet Microbiol. 2021;254:109006. doi:10.1016/j.vetmic.2021.109006

    • Search Google Scholar
    • Export Citation
  • 57.

    Simner PJ, Dien Bard J, Doern C, et al.; Antibacterial Resistance Leadership Group. Reporting of antimicrobial resistance from blood cultures, an antibacterial resistance leadership group survey summary: resistance marker reporting practices from positive blood cultures. Clin Infect Dis. 2023;76(9):1550-1558. doi:10.1093/cid/ciac952

    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 3307 3307 312
PDF Downloads 1323 1323 129
Advertisement