Comparison of multiplex PCR hybridization-based and singleplex real-time PCR-based assays for detection of low prevalence pathogens in spiked samples

https://doi.org/10.1016/j.mimet.2016.11.005Get rights and content

Highlights

  • Standardized methods need to be established for evaluating low-prevalence pathogen diagnostic devices.

  • Biothreat and emerging agents such as Francisella tularensis and Babesia microti are examples of low-prevalence pathogens.

  • The effectiveness of standardized methods was demonstrated on two testing platforms, real-time PCR and TEM-PCR.

  • Compared results from two types of sample matrix, healthy donor blood and blood from patients with signs of infection.

  • Reproducibility of comparisons demonstrated the standardized methods are effective for evaluating diagnostic devices.

Abstract

Molecular diagnostic devices are increasingly finding utility in clinical laboratories. Demonstration of the effectiveness of these devices is dependent upon comparing results from clinical samples tested with the new device to an alternative testing method. The preparation of mock clinical specimens will be necessary for the validation of molecular diagnostic devices when a sufficient number of clinical specimens is unobtainable. Examples include rare pathogens, some of which are pathogens posing a biological weapon threat. Here we describe standardized steps for developers to follow for the culture and quantification of three organisms used to spike human whole blood to create mock specimens. The three organisms chosen for this study were the Live Vaccine Strain (LVS) of Francisella tularensis, surrogate for a potential biothreat pathogen, Escherichia coli, a representative Gram-negative bacterium and Babesia microti (Franca) Reichenow Peabody strain, representing a protozoan parasite. Mock specimens were prepared with blood from both healthy donors and donors with nonspecific symptoms including fever, malaise, and flu-like symptoms. There was no significant difference in detection results between the two groups for any pathogen. Testing of the mock samples was compared on two platforms, Target Enriched Multiplex-PCR (TEM-PCR™) and singleplex real-time PCR (RT-PCR). Results were reproducible on both platforms. The reproducibility demonstrated by obtaining the same results between two testing methods and between healthy and symptomatic mock specimens, indicates the standardized methods described for creating the mock specimens are valid and effective for evaluating diagnostic devices.

Introduction

It has been more than two decades since the introduction of PCR (Mullis et al., 1986; Shampo and Kyle, 2002). Although this technology was implemented almost immediately in research laboratories, it has only been within the past decade that new diagnostic PCR-based technologies are increasingly being integrated into clinical microbiology laboratory practice (Sobel et al., 2008). These technologies enable the detection and quantification of pathogens with increased sensitivity and speed. The implementation of PCR-based nucleic acid tests in blood donor screening and for diagnosing infectious diseases has substantially improved the ability to obtain rapid, actionable information (Brittain-Long et al., 2011; Roth et al., 2012). There are numerous competitive nucleic acid based technologies that have been utilized for the detection and characterization of microorganisms. These technologies can be separated into several broad categories such as direct hybridization, nucleic acid amplification, and a variety of methods for post-amplification analysis such as sequencing, melt-curve analysis and others. Several methods combine the sensitivity of end-point PCR amplification with hybridization-based methodology, allowing the detection of multiple pathogens in one reaction (multiplex) which increases the throughput of the testing platform without compromising assay performance. This work is aimed at comparing the performance of multiplex end-point TEM-PCR coupled with hybridization-based detection to singleplex RT-PCR, for detection of low prevalence pathogens spiked into human matrix such as whole blood.

The development of novel molecular devices for detection of emerging low prevalence pathogens such as Francisella tularensis and Babesia microti is hindered by the lack of clinical samples necessary to conduct clinical sensitivity studies required for U.S. Food and Drug Administration (FDA) clearance or approval (FDA/CDRH, 2014). The use of appropriate specimen matrix for assessment of analytical sensitivity is part of the regulatory requirements for laboratory developed tests or in vitro diagnostics (FDA/CDRH, 2014). Although sample transport medium or another simulated matrix can be used for demonstration of analytical sensitivity, analytical performance of molecular assays should be tested using pathogen-free clinical matrices (CLSI, 2008, FDA/CDRH, 2014). In addition to comparison of analytical sensitivity for multiplex and singleplex PCR tests, we compared human blood matrix collected from patients with fever or other signs of disease to blood collected from healthy donors because the composition of symptomatic blood can have an impact on analytical performance of both molecular methods that is different from the performance when testing healthy blood. The utilization of both testing platforms was recently demonstrated in work to standardize procedures for spiking low prevalence pathogens in human matrices (Dong et al., 2016). As in the first study, the crucial importance of this work is not so much the novelty of the methods described, but the standardization and validation of the methods for use by other investigators. By utilizing these standardized methods evaluation of new technology can be effectively compared to previously established platforms. This study expands upon previously described standardized procedures by developing methods for producing E. coli, F. tularensis, and B. microti mock specimens.

Section snippets

Materials and methods

The production of mock specimens for all three organisms shared a common work flow (Fig. 1). The organisms were cultured, aliquoted and cryopreserved. DNA extracted from the aliquots was serially diluted and tested with RT-PCR to determine the limiting dilution at which the sample could be detected, which was the basis for quantification of the cryopreserved aliquots used for spiking into blood.

Testing negative control blood specimens

The performance of two amplification methods, singleplex real time PCR (RT-PCR) and Target Enriched Multiplex PCR (TEM-PCR) was evaluated based on percentage of positive samples detected (a measure of sensitivity) and percentage of true negative samples classified correctly in these assays. Coded samples were tested and results were compared after completion of testing. The quantitative output of each assay that indicates positive detection of the pathogen (the cut off) was set in both

Discussion

RT-PCR and multiplex PCR are commonly used methods for detecting low abundance pathogens. Singleplex RT-PCR pathogen detection is considered to be a gold standard methodology for high sensitivity, reproducibility and ease of use. To reduce the burden of clinical validation for low abundance pathogens, mock specimens can be created from human matrices and cultured pathogens. However, spiking should be done by standardized methods to obtain results that can be compared appropriately (Dong et al.,

Conclusion

The standardized methods of preparing mock (spiked) clinical specimens for evaluating diagnostic devices intended to test for low prevalence pathogens have effectively demonstrated the equivalence of a proprietary multiplex platform to individual real-time PCR. The equivalence between TEM-PCR, optimized to screen for the presence of Babesia microti, Escherichia coli and Francisella tularensis, and individual real-time PCR assays for each pathogen over a three-log range of pathogen

Acknowledgements

The Francisella tularensis culture was initiated from a colony gratefully obtained from Dr. Karen Elkin's laboratory (FDA/CBER). We wish to acknowledge that this work was supported by NIAID through an Interagency Agreement AAI13005-002 awarded to CBER/FDA. Thanks to Karen Elkins, PhD and Rene Reese, PhD for reading and commenting on the manuscript.

References (20)

There are more references available in the full text version of this article.

Cited by (0)

Our contributions are an informal communication and represent our own best judgment. These comments do not bind or obligate FDA.

View full text