Friday, December 23, 2016 Written by Ellen Jo Baron, Ph.D., D(ABMM), Prof. Emerita, Stanford University Director of Medical Affairs, Cepheid

Wake-Up Call: Proficiency Testing for Today's Technology

How Cepheid's GBS Assay Stimulated New Awareness and How Your Laboratory Should Respond to Avoid Similar Problems
Last year, a number of Cepheid's Xpert® GBS assay users "failed" some of the College of American Pathologists' group B streptococcal proficiency testing (P.T.) samples distributed in shipment D8. The users reported "positive for group B Streptococcus (GBS)" when the CAP had intended those samples to be negative and the provider of the samples had certified that they did not contain GBS. Laboratories using other nucleic acid amplification methods, such as BD GeneOhm, and those using the Cepheid SmartCycler® GBS product did not experience any positive results, and CAP's initial assessment was that the Xpert GBS assay was producing false positive results. CAP initially suspected that there was a cross-reaction with other organisms present in the challenge specimens. Although the early discussions focused on whether CAP should notify the FDA about the faulty performance of the assay, a conference call with several representatives from Cepheid and CAP resulted in an agreement to investigate the situation further. CAP's decision may have been influenced partially by a letter received from one participating laboratory documenting that the laboratory had recovered viable GBS from at least one of the supposedly negative P.T. samples after extended broth incubation.

At CAP's direction, additional P.T. samples from the contested lots were sent to a consultant on the CAP Microbiology Resource Committee as well as to Cepheid. Numerous samples were placed into enrichment broth for extended culture analysis, and others were retested in additional molecular platforms, as well as being run on GeneXpert® Systems at Cepheid. Again, a few of the "negative" samples yielded positive results for GBS in the GeneXpert, but not in the other assays. This time, the amplified products from the GeneXpert assay were sent to an independent laboratory for DNA sequence analysis, and sure enough, contained amplified sequences of GBS DNA. Scientists familiar with the issue had experienced a previous incident in which P.T. materials that should have been negative were inadvertently contaminated during specimen preparation with genetic material from the putative agent, and a number of corrective action interventions that had been successful in the past were suggested.

CAP then convened a conference call with the P.T. vendor, who acknowledged that the samples had been prepared primarily for culture analysis (for which the presence of dead organisms or genomic DNA posed no problem) and that not all processes were in place to avoid genetic, i.e., nucleic acid, contamination. Through this process, it became clear that because of their high levels of sensitivity (for MRSA as well as GBS) the GeneXpert assays were more prone to genomic DNA carryover than other methods used by participants (whether culture or nucleic acid amplification) and that a new standard for P.T. samples must be implemented. Fortunately the vendor had already identified potential sources of carryover and had begun to alter its production methods to accommodate the required stringency for nucleic acid amplification test methods. CAP issued a letter to its participants, noting that the results from Cepheid GeneXpert users would not be graded. In fact, the "false positives" were actually "true positives."

This incident should also serve as a "wake-up call" to GeneXpert users. In the PCR world, genetic targets can be inadvertently carried over from positive patient samples as well as from organisms, alive or dead, that may contaminate the workspace. The high sensitivity of PCR protocols can readily lead to problems if stringent specimen handling protocols are not followed. Batched testing is especially prone to target contamination, and in the age of closed system real-time PCR, target contamination probably accounts for the majority of contamination events leading to false positive results.1 For laboratorians, this is a modern version of the same age-old problem that causes occasional false-positive mycobacterial cultures and viral cultures. Sample contamination appears to be resurfacing in the age of molecular methods. C. difficile spores are also notorious for their ability to contaminate work environments so the potential for sample contamination is high.2 Good sample handling technique is critical to getting good results.

The GeneXpert cartridge is a closed system, which is an important attribute for controlling cross-contamination during the sample processing and amplification steps.

However, paying careful attention to the first steps of the process, that of adding the specimen carefully to the cartridge without carryover of material from previously processed specimens, is still important. This is especially true when batch testing is performed; tests performed in random order will not be as subject to this problem since the target-specific detection reagents are unique to each cartridge. Although some testing of small batches may be unavoidable, processing of large batches of identical tests does not take full advantage of the diagnostic horsepower under the hood of your GeneXpert, since batches take time to accumulate and this occurs at the expense of turnaround time. Moreover, it puts your laboratory at increased risk of experiencing a false-positive result due to inadvertent introduction of nucleic acid targets into a negative patient's sample or test cartridge. Remember to follow recommended procedures no matter how testing is being done; wash hands and change gloves before and after each new sample, clean the bench and surrounding workspace frequently with 10% bleach or DNA-destroying cleaner, and discard the used cartridges in a hard-sided container to avoid leakage of contents into the environment.

REFERENCES

1. Sloan, L. M. 2007. Real-Time PCR in Clinical Microbiology: Verification, Validation, and Contamination Control. Clinical Microbiology Newsletter. 29: 87-95.
2. Dumford, D. M., 3rd, et al. 2009. What is on that keyboard? Detecting hidden environmental reservoirs of Clostridium difficile during an outbreak associated with North American pulsed-field gel electrophoresis type 1 strains. Am J Infect Control. 37: 15-19.