CFSRU, HMRU, and BPRP, "Preliminary Findings on the Evaluation of Hand-Held Immunoassays for Bacillus anthracis and Yersinia pestis"
January 2003 - Volume 5 - Number 1
Preliminary Findings on the Evaluation of Hand-Held Immunoassays for Bacillus anthracis and Yersinia pestis
Counterterrorism and Forensic Science Research Unit
Hazardous Materials Response Unit
Federal Bureau of Investigation
Bioterrorism Preparedness and Response Program
Bioterrorism Rapid Response and Advanced Technology Laboratory
National Center for Infectious Diseases
Centers for Disease Control and Prevention
Commercially available hand-held immunoassays (HHAs) for the detection of Bacillus anthracis and Yersinia pestis (the causative agents of anthrax and plague, respectively) were compared for sensitivity, specificity, repeatability, robustness, and stability. A total of six kits (four for anthrax and two for plague) were evaluated.
All of the HHAs detected the Pasteur and Sterne strains of B. anthracis, but an HHA from one manufacturer failed to detect a virulent strain of B. anthracis. The lower limits of detection for the HHAs varied from 105 to106 for unpurified anthrax spores and from 105 to108 for washed anthrax spores. The anthrax HHAs cross-reacted with nontarget bacteria, yielding positive results in the absence of B. anthracis from 3% to 83% of the time.
The ability of the plague HHAs to detect Y. pestis was dependent on the particular strain being analyzed as well as the temperature at which the strain was grown. Each strain was grown at either 25°C (77°F) or 37°C (98.6°F). No strain of Y. pestis that was grown at 25°C was detected by the HHAs. The lower limit of detection of the Y. pestis HHAs was between 104 and 105 cells. The plague HHAs cross-reacted with nontarget bacteria, yielding positive results in the absence of Y. pestis in 7% to 13% of the specimens tested. The results for sensitivity and specificity demonstrate the limitations of these assays as a reliable means of field detection.
In the interest of public safety, the following communication is being released before subsequent full publication.
During the emergency response phase of a suspected or potential exposure to a biological threat, first responders sometimes use commercially available hand-held immunoassays (HHAs) for the detection of biological threat agents. HHAs are small test strips which contain antibodies to a specific biological agent. A suspect sample is suspended in a liquid supplied with the test assay. The liquid suspension is then applied to the test strip and allowed to develop for approximately 15 minutes. If a biological threat agent is contained within the sample, a colored band will appear on the test strip. A quality control test is built into all the strips to indicate whether the assay materials are working properly. The Federal Bureau of Investigation and the Centers for Disease Control and Prevention tested six commercially available HHAs (four for B. anthracis and two for Y. pestis) in order to assess their capabilities and limitations under a variety of anticipated scenarios. Test strips were used according to the manufacturers’ instructions. Effectiveness of commercially available HHAs to detect B. anthracis or Y. pestis was evaluated by measuring sensitivity, specificity, repeatability, robustness, and stability. Preliminary results of this work are outlined in this short communication. Out of consideration to the HHA manufacturers, test strip names have been withheld. Names of test strips, complete procedures, and full discussion of the results will be provided in a detailed technical publication.
In order to establish the limit of sensitivity, the minimum number of microorganisms required to give a positive test result is referred to as the sensitivity of the assay. A stringent test was performed, in which ten samples out of ten tested had to be positive. If fewer than ten tests were positive at a certain concentration, then ten additional samples were tested with ten times the previous amount. This process was repeated until all ten samples tested gave positive results.
For anthrax sensitivity measurements, four different anthrax HHAs were tested against unpurified B. anthracis spores. In addition, a separate preparation of washed B. anthracis spores was tested. Overall, the sensitivity values of the assays ranged from 105 to 108 spores of B. anthracis.
Two plague HHAs were evaluated using 11 different strains grown at 25° and 37°C. The Y. pestis HHAs only detected organisms grown at 37°C because both HHAs are specific for a protein that is produced only when the organisms are grown at temperatures greater than 30°C. The sensitivity values of the assays ranged from 104 to 105 Y. pestis cells grown at 37°C. The lowest number of spores or cells required for visual detection of a positive result for each strip is provided in Table 1.
Note that the sensitivity values are much higher than those quantities reported to produce disease (8,000 - 50,000 spores for inhalation anthrax; 100 - 500 microorganisms for pneumonic plague).
See http:\\www.nbc-med.org\SiteContent\MedRef\OnlineRef\FieldManuals\ medman\Handbook.htm
Specificity refers to the ability of an assay to accurately discriminate between a target microorganism (either Bacillus anthracis or Yersinia pestis) and other bacterial species. In order to determine the specificity, each assay was tested for cross-reactivity with a variety of different bacterial species. Thirty non-B. anthracis species, strains, and preparations were tested using the anthrax HHAs, and 15 non-Y. pestis samples were tested using the plague HHAs. Under these conditions, any positive result observed would be considered a false-positive result. Table 1 provides the number of false-positive responses over the total number of different tests performed. The results indicate a false-positive rate between 3% and 83% for the anthrax assays and a false-positive rate between 7% and 13% for the plague assays.
In addition to these tests, both the anthrax and plague HHAs were tested with different strains of the respective pathogen. Two virulent strains and two vaccine strains of B. anthracis were tested with the anthrax HHAs from three suppliers (numbers 1-3 in Table 1). All the anthrax assays tested except one gave positive results with these samples. The HHAs of supplier 1 tested negative with one virulent (disease-causing) strain. Eleven virulent strains of Y. pestis grown at 37°C were tested. One HHA failed to detect two of these strains, and another HHA failed to detect one of these strains.
The repeatability of an assay reflects its ability to produce the same results under the same test conditions. For this analysis, all experiments were performed in duplicate, and the repeatability of the results is recorded in Table 1. The results are reported as the number of times the tests were repeatable relative to the total number of replicate tests performed. A high level of repeatability demonstrates optimal performance. The repeatability range for the HHAs tested was between 78% and 100%.
An assay that yields a very high number of correct responses when performed under a variety of different experimental conditions is regarded as robust. To determine the effect of interfering substances on the HHAs, tests were performed both with and without target pathogen in eight different solutions containing salt, sugar, milk, acid, and alkaline solutions, and with two extracts of either Georgia red clay or commercial potting soil. The ten tests were performed in duplicate with each assay. Success rates for each solution are shown in Table 1, listed as the number of times that an expected result was obtained relative to the total number of tests that were performed. There was no pronounced effect of interfering substances on the performance of the new generation HHAs. However, Georgia red clay gives a positive response with the anthrax strips of suppliers 2 and 3 and reduces the intensity of the reactions with the strip of supplier 1. High-salt solutions caused the strips of supplier 3 to react inconsistently.
HHAs were stored for 10 to 30 days at a field temperature of 30°C (86°F) without losing sensitivity. These results indicate that HHAs are generally stable. Therefore, special handling of test kits is not required.
This study was carried out to obtain information on the performance characteristics of hand-held immunological assays for two pathogens, Bacillus anthracis and Yersinia pestis. All HHAs tested gave false-positive results with at least one bacterial species that was not the target organism. These results indicate that further evaluation with environmental samples collected from a wide geographical region is warranted to better define the specificity of the HHAs.
The number of organisms reported to produce disease in humans is much fewer than the lower limit of consistent detection (105 to 108 spores of B. anthracis and 104 to 105 cells of Y. pestis) for both assays. Additionally, some HHAs were incapable of detecting infectious agents that cause anthrax and plague. For example, if highly infectious Y. pestis in captured fleas (the host that carries the pathogen) were tested, or organisms grown at room temperature (25°C), the results indicate the HHAs would not react. Therefore, the HHAs cannot be relied upon to always detect a pathogen when it is present, particularly the plague assays.
The HHAs generally performed well when substances known to interfere with immunological tests were added or conditions such as pH (acidity) and ionic strength (salt concentration) were changed. Results indicate that guidelines for the testing of environmental samples are needed. Variation in repeatability was observed with all anthrax HHAs, lowering the level of confidence that can be placed in these assay systems.
The low sensitivity, the high level of possible cross-reactivity with environmental bacteria, and the failure to detect dangerous species of B. anthracis and Y. pestis limit the usefulness of these hand-held assays.