Skip to main content
Premium Trial:

Request an Annual Quote

Dxcover Advances Multicancer Detection Platform; Plans 2024 Launch for Brain Cancer Dx


NEW YORK – 2022 is turning out to be a transitional year for UK diagnostics firm Dxcover. 

The Glasgow-headquartered company recently presented data on the use of its spectroscopic liquid biopsy test for the early-stage detection of multiple cancers at the American Society of Clinical Oncology (ASCO) meeting in Chicago. Chief Technology Officer Matthew Baker said the dataset is "hugely important" for Dxcover, which to date has been working on early-stage brain cancer detection. 

"We have always known we have a platform and that our capability is not just in brain tumors but across all cancers," Baker said in an interview this week. "This shows its utility and proves to the community that we can detect those early-stage tumors," he said of the ASCO presentation.

Dxcover has other plans in play. According to CEO Mark Hegarty, it is working to raise a Series A round worth about $35 million by the end of the year. And plans for international expansion are also afoot, with the company mulling setting up a CLIA-compliant laboratory in California, as well as a business office, potentially in the Boston area. It is also hiring software developers and looking to bring on board a chief commercial officer and chief medical officer to oversee its US commercial endeavors. At the moment, Dxcover has 15 employees, Baker said.

It's a scale-up from where the firm stood three years ago, when it was a spinout of the University of Strathclyde called ClinSpec Diagnostics (it rebranded in 2021) with a handful of full-timers. 

Dxcover uses attenuated total reflection (ATR)-Fourier transform infrared (FTIR) spectroscopy to survey the chemical makeup of a patient's blood sample. Once spectra are obtained, the datasets are analyzed using machine-learning tools trained to recognize brain cancer signals. Detection is performed on silicon internal reflection element (Si IRE) slides, which permit the analysis of multiple samples at a time. 

In 2019, the company and partners at the University of Edinburgh trialed the test in 400 brain cancer patients. The work was published in Nature Communications in October 2019.

According to Baker, the development of its brain cancer test, branded as the Dxcover Brain Cancer Liquid Biopsy, continues, with "one pivotal trial" necessary before the company can launch it, first as a CE-IVD marked kit in Europe by the end of 2024 or in early 2025. Dxcover will also seek a UK Conformity Assessed certificate for the test. The UK intends to transition to its new, post-Brexit regulatory regime by July 2023.

In the US, Dxcover will likely offer the Dxcover Brain Cancer Liquid Biopsy as a laboratory-developed test first via a CLIA-compliant lab, as it embarks on an American validation study, and, eventually, a submission to the US Food and Drug Administration. Dxcover also has an eye on the Chinese market too, and Baker said the company will eventually seek Chinese FDA clearance for its test.

Baker said Dxcover could submit its test to both agencies in 2026.

At ASCO, Dxcover gave the market a look at what's to come beyond early-stage brain cancer detection. For its multicancer detection study, it recruited more than 2,000 patients, using the same spectrometry and AI technology to distinguish those with cancer from those without cancer, as well as to differentiate the organ of origin for patients' cancers, including brain, breast, colorectal, kidney, lung, ovary, pancreas, and prostate. 

According to the company, its model yielded 64 percent sensitivity for stage I cancers and 90 percent sensitivity for all cancers. Detection rates were 93 percent for stage I, 84 percent for stage II, 92 percent for stage III, and 95 percent for stage IV cancers. 

Separately at ASCO, Dxcover also presented data from a proof-of-concept study that explored the use of its approach to detect pancreatic cancer at an early stage. It aimed to discriminate between patients with cancer and controls, as well as those with cancer and symptomatic, nonmalignant controls. They reported that they could distinguish between both, with a sensitivity of 91 percent and a specificity of 88 percent.

Baker said that the data from both studies will be published soon. He said that reception to the firm's ASCO presentations had been "tremendous" and generated more interest in its platform. The data also aligns with Dxcover's goal of becoming an early-stage cancer detection company.

"Our mission is to save lives, and the best way to do that is to detect tumors at stage I and stage II, when they can be surgically treated rather than at stage III and stage IV, when it needs chemotherapeutic treatment," he said. 

The company's next step with its multicancer detection work is to run a larger study on roughly 12,000 samples that cover all of the cancers in its initial dataset, as well as some symptomatic controls, and then move forward with clinical trials on various cancers it selects, depending on the market need. Such trials could take place around the time that Dxcover gains a CE-IVD mark for its brain cancer test in the middle of the decade.

Dxcover's platform is already designed, and according to Baker, its hardware is "frozen," so that the platform that is used in clinical trials will be used clinically in the end. Its website displays a system consisting of sample slides, an autosampler to enable infrared analysis, and PerkinElmer's Spectrum Two instrument. 

According to Hegarty, Dxcover is in discussions with partners related to the spectrometer it will use in its final platform, and it continues to manufacture its slides in house. "We basically have control of all the manufacturing that is required at this stage," Hegarty said.

There is also the matter of competitors, of which there are many, from Grail to Freenome, working on cancer liquid biopsy tests for early-stage disease. But Baker did not rule out working with other companies, given Dxcover's capabilities.

"We can bring high sensitivity to highly specific genetic approaches," remarked Baker. "And we have a low barrier for integration into other workflows," he said.