Skip to main content

GeneMe Commercializing Rapid COVID-19 Tests With Extremely Low Limits of Detection

NEW YORK – Genomic testing startup GeneMe is channeling enhanced isothermal molecular diagnostic technologies into point-of-care tests for COVID-19. The Gdansk, Poland-based company claims its extraction-free loop-mediated isothermal amplification (LAMP) test has a limit of detection of 20 viral copies per milliliter, placing it among the most sensitive molecular assays in the world.

The firm is currently partnering with Virgin Atlantic to rapidly test flight crews and also plans to adapt a rapid PCR-based saliva test to develop automated self-testing kiosks. This week, GeneMe secured €5.2 million ($7.0 million) in seed funding from angel investors. It also recently partnered with lyophilization firm Biolyph to support further scaling of its tests.

GeneMe's currently commercialized test is called FRANKD, an acronym for fast, reliable, accurate, nucleic acid-based kit for COVID-19 detection. The test uses an oropharyngeal swab sample incubated on standard thermal cyclers, and is paired with a digital health app from UK-based firm Yoti to deliver results to a phone in approximately 30 minutes.

The core technology in FRANKD is standard RT-LAMP chemistries that are enhanced through modifications of the enzymatic reaction. Specifically, the firm modifies the polymerase with a protein to improve the speed and accuracy of the reaction.

In an interview, Colin Brown, founder of an advisory firm called NextD that is coordinating GeneMe's commercialization, noted that an analytical evaluation of the FRANKD assay by the National Health Service of Scotland revealed the test had a limit of detection of 20 viral copies per milliliter.

The firm is also developing an assay it calls salvia amplification viral detection, or SAVD, a saliva-based test that uses rapid RT-PCR. Brown noted that an evaluation of this test showed it had an LoD of 2 copies per milliliter.

GeneMe also has a test system in development called influenza COVID effective detection, or ICED, for multiplexed SARS-CoV-2 and influenza assessment, Brown said.

For the FRANKD RT-LAMP test, GeneMe recently joined up with a global collaborative group called gLAMP that is troubleshooting LAMP-based SARS-CoV-2 testing.

Chris Mason, a genomics researcher at Weill Cornell Medicine and one of the founders of gLAMP, confirmed in an email that GeneMe's reported LoD's would be on par with laboratory-based PCR tests.

"That LoD, if confirmed, is quite impressive," Mason said. "If they really have 0.002 copies per microliter, or 2 per milliliter, then this represents one of the best LoDs on the market," he also said.

However, Mason also noted that a Nature Biotechnology study he co-authored demonstrated differences between the reported LoD and the actual detection limits when tests are used by others. The variation can be one to two orders of magnitude, he said, and overall the LoDs of current tests on the market range five- to six-fold in logarithmic differences.

GeneMe's Brown is aware of these issues, and also cited the recent US Food and Drug Administration panel evaluations, which are reported at nucleic acid amplification test (NAAT) detectable units, or NDUs.

"There can be considerable variation in the manufacturer's published LoD in copies per mL and the FDA's derived value of NDU per mL LOD for the same test," Brown said.

For example, "Abbott's ID NOW LoD published by the manufacturer is 125 copies per milliliter, but when the FDA tested it with their own reagents they came up with a LoD of 300,000 NDU per milliliter," while another independent evaluation of the same test has shown it to have a LoD of 20,000 copies per milliliter, he said.

So, across the industry it seems that in the absence of calibrated reference standards, "Where the 'true' LOD value lies is open to debate," Brown said. "The FDA's 'NDU per mL' LoD unit does not answer that question particularly, but just allows a more reliably fair comparison between tests," he said, as does Mason and colleagues' Nature Biotech study.

That said, there have been other reports of surprising limits of detection with LAMP-based tests for SARS-CoV-2. For example, a team at Harvard Medical School utilized RT-LAMP and an extraction chemistry known as "glass milk," recently reporting in PNAS a LoD as low as 1 viral RNA copy per microliter, which translates to 1,000 copies per milliliter.

A recent report in medRxiv described modifications to the Bst enzyme meant to reduce the LoD of LAMP-based SARS-CoV-2 tests. That team, based at the University of Texas, Austin, reported that adding fusion domains could enhance the LoD to 50 copies per 3 microliters. And, as described in bioRxiv, the UT team has developed a multiplex test for detecting SARS-CoV-2 in saliva that could reliably signal as few as 10 genomic RNA copies in a sample using endpoint fluorescence.

But, the ways in which LoD interacts with pre-analytical issues to impact sensitivity and specificity is a bit of a black box in real-world settings. A test that can detect few viral particles could be highly sensitive, but sampling and workflow also matter.

For example, GeneMe notes that the FRANKD test is 97 percent sensitive and 100 percent specific although it purportedly has a lower LoD than virtually all gold-standard PCR tests.

Core tech, and future directions

Kasjan Szemiako, chief technology officer at GeneMe, said the firm's core technology can also be applied to other types of genetic analysis. "We have developed our own polymerases, and these enzymes are used in all genetic diagnostics for bacteria, viruses, or in our own DNA," he said.

The polymerase "provides very high specificity and sensitivity" to detect single mutations, Szemiako said, and is combined with a protein called NeqSSB from Nanoarchaeum equitans archaeon bacteria that can bind to double- and single-stranded DNA and keep the polymerase attached to the target nucleic acids. "It upgrades the polymerase with specificity and processivity, which means we get the results very fast," he said.

GeneMe has applied for an international patent for the method.

Szemiako said GeneMe quickly adapted its technology in March to address COVID-19 testing needs by adding the reverse transcriptase and adapting it for RNA. The firm is sourcing any LAMP reagents not made in house from various suppliers, although it is not using a common source of LAMP reagents, New England Biolabs. And, Brown said that so far GeneMe has not negotiated licensing with Eiken Chemical, the developer of the LAMP technology.

The firm submitted its FRANKD test to the FDA for Emergency Use Authorization in July, Brown said, but has not made much progress to date. Although the FDA has called for more point-of-care tests, Brown attributes the delay to the fact that GeneMe is "an underdog" and not an established "big pharma" player.

"We are a little company in Poland, but we've produced two of the most accurate COVID-19 tests in the world," he said.

The firm's FRANKD test is currently being used in a program to test Virgin Atlantic flight crews at London's Heathrow Airport.

Specifically, it is used for routes to certain countries in Asia where crews are tested on arrival, and if they test positive they have to quarantine for two weeks. Earlier in the pandemic, "They were running out of flight crews," Brown said, because the airline essentially couldn't get their crews back for weeks.

The UK is also now rolling out a "test and release" program run by collaborators Collinson Group and Swissport, Brown said, expanding airport testing to travelers in parking garages and to additional airports.

The FRANKD test kit can be run on a variety of instruments, but GeneMe has also created a small isothermal testing instrument that is the size of a lunchbox that costs $250 to manufacture, Brown said.

For SAVD, the firm is developing a mass-testing instrument akin to an ATM it expects to launch at the end of March.

The machine will contain devices to perform 19 tests in parallel. This testing kiosk can theoretically be used in airports or hospitals to process 5,000 saliva samples in 18 hours, he said, whereby travelers or visitors would download an app, drop a sample in the test machine, and get a test result on their smartphone 30 minutes later.

Post-COVID, "Our vision in the future is for it to be outside the hospital, but filled with a range of different tests," Brown said.

For its saliva test, the team is adapting the SalivaDirect test protocol developed at Yale University for COVID-19 testing.

Anne Wyllie, a researcher at Yale who has worked on SalivaDirect, commented that "LAMP-based tests are a fantastic approach to improve on PCR testing seeing that it can be faster and can in some cases use more simple machines "

Wyllie said that GeneMe provided some of its FRANKD tests for the Yale team to try out, although there is no formal collaboration in place and it has not yet had the bandwidth to assess them. The Yale team has authorized 56 labs in 26 states to run its SalivaDirect test and is currently working to bridge to additional PCR instrumentation, functionally equivalent steps, automation, more saliva collection devices, at-home collection, and trained observation, Wyllie also said.

In the future, GeneMe plans an additional 240 assays using its technology, of which it has developed 40 so far, Brown said. These include more infectious disease tests, as well as ones for oncology and wellness testing.

Overall, the SARS-CoV-2 pandemic seems likely to change the landscape of diagnostic testing for some time. Yale's Wyllie said, "It has been incredible to see the rapid development of different technologies which are likely to forever change the future of public health for the better."

Brown agreed. In terms of diagnostics, "COVID has completely changed societal and scientific norms," he said. "Where we found our sweet spot is being highly accurate, but fast."