Roche’s Deutschmann: When Alternative is not Alternative
The veteran microbiologist discussed experimentation with a rapid microbiological CFU counting technique, and maintained that validation should not be significantly different than with traditional methods.
By Paul Thomas, Senior Editor
At the PDA 2011 Annual Meeting in San Antonio, Sven Deutschmann, microbiologist with Roche Diagnostics GmbH, spoke about his company’s experimentation with a rapid microbiological method based upon an imaging technology that captures the fluorescence emitted by cells. (The system was Rapid Micro Biosystems’ Growth Direct, though Deutschmann chose not to name the vendor.) The talk was titled, “Qualification and Validation of an Automated Rapid Growth Based System for In-Process-Control and Water Testing.”
Deutschmann noted that implementing an electronic, automated colony-forming unit (CFU) counting system was “a piece of the puzzle in our paperless lab solution.” Especially in bioprocessing, he noted, it is valuable to have quick results (typically 24 hours) in order to inform colleagues to proceed, or not, with a production run.
First, Deutschmann provided an overview of the machine itself, which includes an automated robotic sampling arm, sample cassettes, an incubator, and camera, plus electronics and chillers. The Growth Direct, he continued, “detects microcolonies by illuminating them with blue light and directing the resulting cellular yellow-green autofluorescence onto a CCD chip.”
Deutschmann took pains to emphasize that the use of the system was not wholly different, from a regulatory standpoint, from more traditional testing methods, and thus he did not view this approach as an “alternative” method. Though the system is not a traditional Petri-dish-based method, “This is not an alternative technology,” he said, “and this is very important for validation.” Roche uses its standard membrane filtration method, and depends upon compendial microbial enumeration tests. “Thus, the technology retains the advantages of traditional culture methods while reducing long test turnaround/incubation times”—those times are typically 24 to 36 hours for most typical microorganisms, he noted.
To justify his position, Deutschmann turned to EP Chapter 5.1.6, section 3.1, “Types of Microbiological Tests,” which states: “It is critical to the validation to identify the portion of the test addressed by the alternative method.” To Deutschmann, the incubation period was the notable difference. After further reviewing the regulatory literature, Deutschmann determined that only the accuracy and precision of the recovery system need to be validated, not the entire test. He then detailed the work his team did to qualify the system for accuracy and precision.
After consulting with European reviewers and regulators, Deutschmann set up experiments to test for stressed microorganisms, using temperature and pH stresses. Using S. aureus, P. aeruginosa and C. albicans, his team conducted an incubation procedure comparing traditional methods and Growth Direct, finding that it is possible to detect stressed organisms in 24 to 36 hours. Tests upon pH-stressed microbes (at pH 3.0 for 0 to 60 minutes) showed similar time results.
In summary, Deutschmann said, “Growth Direct is a very accurate, precise counting machine. . . . We have in our facility several other alternative methods, but this machine generates a final result in CFU per milliliter, which can be used with other methods.” He also noted that his lab has reduced incubation times to 24 hours, and uses the software function of the machine to email results as they are obtained to various parties, and that it has been integrated with their LIMS system, fulfills 21 CFR Part 11, and thus eliminates issues associate with paper-based systems.