NEW YORK (GenomeWeb) – A multi-center trial evaluating BioMérieux's Vitek MS clinical microbiology platform has found the MALDI mass spec-based system is able to identify molds commonly tested for in patients in clinical settings with high accuracy.
Using the instrument to test 1,601 mold isolates, researchers found that 98 percent of the identifications the system provided were correct, indicating the usefulness of the platform in clinical mycology.
The study, published last month in the Journal of Clinical Microbiology, was part of BioMérieux's submission of the system to the US Food and Drug Administration. The company received 510(k) clearance from FDA in July for the expanded identification of mycobacteria, Nocardia, and molds on the Vitek MS. The platform received its original 510(k) clearance in 2013, which covered use of the system for detection of Gram-negative bacteria, Gram-positive bacteria, yeast, and anaerobes.
Bruker, BioMérieux's main competitor in the MALDI clinical microbiology space received the first 510(k) clearance that same year for its MALDI Biotyper platform, and the two companies have since worked to expand their databases with molds being a prominent target.
With the recent 510(k) clearance, BioMérieux has stepped momentarily ahead of Bruker, at least from a regulatory perspective, as the latter company has yet to receive the FDA's nod for a MALDI Biotyper library targeting mold. The company developed research-use-only libraries for these organisms that contain 127 species of filamentous fungi and 164 species and 780 strains of mycobacteria.
Both platforms identify microbes by matching the protein profiles of sample organisms generated via MALDI mass spec to profiles contained in a proprietary database. According to Jenna Rychert, a researcher at reference laboratory ARUP and first author on the study, in the case of fungi, MALDI could enable facilities to make confident identifications without requiring the level of expertise traditionally needed to evaluate these organisms.
"In the clinical lab, when you go to identify a mold, commonly the first path is to literally look at it macroscopically. What is the plate [it grows on]? What does it look like growing on a plate?" she said. "Then look at it microscopically, and from those two things you can often get it to at least the genera and sometimes to the species. But that's obviously a very subjective thing, and you're going to have to have people who are very well educated on how to do those identifications."
With a MALDI system like the Vitek MS, on the other hand, "you are putting the organism in the instrument and it's giving you an answer," she said. "The level of education needed is a little bit less. The level of work involved is a little bit less. It's not as subjective. So in that sense, it's better."
While speed is commonly cited as a major advantage of MALDI methods in the bacterial ID space, this is less of a factor with molds, Rychert said, as clinicians still have to take the organism of interest through the culturing process. However, she noted, in cases where clinicians aren't able to make an identification based on their visual inspection of the organism's morphology, using MALDI could allow them to skip the various additional, and time-consuming, tests required to further evaluate the organism.
Sequencing approaches can serve a similar role, but Rychert noted that MALDI is significantly less expensive than sequencing as well as faster and simpler. That said, whether a facility uses sequencing or MALDI for their mold identification is likely a function of their larger clinical microbiology needs and processes, she said.
"It's going to differ with every laboratory you ask," she said. "You might have a clinical laboratory at a hospital that has a mass spec instrument that doesn't do sequencing. At a different hospital, you may have a situation where they have sequencing, but they don't have mass spec. It's a bigger picture question than just identifying mold."
In the JCM study, Rychert and collaborators at University of Utah, Memorial Sloan Kettering Cancer Center, Weill Medical College, and the University of Texas Health Science Center found the Vitek MS could identify isolates represented in its database to the species level 91 percent of the time, with an additional 2 percent identified to the genus level. The system was unable to identify 91, or 6 percent, of the isolates tested, and provided incorrect identifications of 15 isolates. Of those 15, in 13 cases the mistaken identification still placed the organism in the correct genus.
"I think if you compare [the performance of the Vitek MS] to someone who is a trained mycologist who has been looking at cultures for 20 years, it's probably not a whole lot different," Rychert said. "But if you compare it to a whole mycology laboratory that has everyone from someone who just got out of school to that 20-year veteran, it's going to give you a more accurate answer more of the time."
That suggests the system could find adopters among smaller hospital labs and other facilities without extensive mycology expertise, she said, particularly if they have already purchased the platform for bacterial work.
"You might have a hospital that brought on this instrument when it came out for bacteria, and they've been using it for bacteria and yeast, and now they have this opportunity to use it for their mycology lab," she said. "It would be an easy transition."
The platform as tested in the JCM paper is focused on a core set of commonly encountered organisms, Rychert said, noting that this would also dictate where it might prove most useful.
"If you're in a large, academic medical center that has, for instance, lots of transplant patients, the kinds of molds that you're going to see are different than if you're a community-level hospital, or hospital system that doesn't really have a transplant center, or doesn't really do oncology services," she said. "They all have those kind of same, base-level common organisms [covered by the Vitek MS database]. But then those more complicated, academic medical centers are going have all the extras. And this database isn't geared towards all those extras."
ARUP's mycology work tends to be focused on more esoteric organisms, Rychert noted. "We get all of the kind of unusual things that other people can't really identify themselves."
This means the lab's needs extend beyond the Vitek MS database evaluated in the JCM study, and, in fact, Rychert said, ARUP uses Bruker's Biotyper for its clinical microbiology work due in part to what she said is that platform's greater openness to customization.
While both the Vitek MS and MALDI Biotyper are clinical in vitro diagnostic instruments, Rychert said there are "subtle differences" she attributed to the Biotyper's research-based origins that make that instrument "a little easier to customize than the other."
"That's not bad, or good, or whatever," she added. "It's just a matter of what you need."