July 10, 2025 | Efforts to find cancer early have been ongoing for more than a century now, but proven screening tests are available only for a handful of the most common types of malignancies and don’t always produce accurate results. Multi-cancer early detection (MCED) tests have emerged over the last decade that detect many different cancers in a single assay, but they excel only at the detection of cancer at stage three or four.
While this is better than finding cancer later when patients are more symptomatic, the holy grail for cancer detection is being able to consistently detect cancer at stage two or even one before it has spread significantly from its origin. As any oncologist will readily attest, this localized nature allows for more effective and less aggressive treatment options that often translate to higher survival rates and better long-term outcomes.
A growing number of promising young companies are now developing tests that are particularly good at picking up early cancer, initially for a single type of cancer but with the potential to eventually be extended to multiple cancer types. A handful of these startups will be spotlighted at the upcoming Next Generation Dx Summit, among them UK-based Xgenera.
The biotechnology company is laser-focused on miONCO, a blood-based MCED test that combines a predictive panel of microRNAs (miRNAs) with machine learning to detect 12 types of the most common and deadly cancers in a single test, according to CEO Andy Shapanis, Ph.D. Preliminary data from a study in a 21,000-patient cohort indicate miONCO can predict cancer with a sensitivity and specificity of 99%across all stages and up to 96% when predicting tumor site of origin.
Those 12 cancer types represent 68% of all new cancer cases in the UK, he reports. The analysis on the case-controlled population suggests that there may well be more than a dozen cancers detectable with miONCO; Xnenera just had too few patient samples to support that claim.
Since this is currently a microarray test and evaluated only for its ability to accurately identify individuals with and without cancer, the miONCO test is being ported into the gold standard PCR format, says Shapanis. A clinical trial evaluating the PCR version of the test, involving an analysis of blood samples from 8,000 patients enrolled in different long-term studies, is running now through mid-2027. It includes asymptomatic (or at least pre-diagnostic) patients, giving researchers their first indication of how well miONCO works in people who are seemingly fit and healthy.
The Xgenera team plans to publish detailed study findings once the external validation work is completed, and the intellectual property is fully protected, he says. The expectation is that they will have too few patients for the tool to predict with confidence more than five or six of the 12 cancers. But the most prevalent and lethal cancers—breast, prostate, pancreatic, lung, and colorectal—will be in the mix.
At that point, the company hopes to start commercializing miONCO in the U.S. as a laboratory developed test (LDT), says Shapanis. This will require either setting up a lab on its own and getting it certified and accredited or perhaps partnering with an existing high-volume facility.
Given the expansion plans, Shapanis says he is looking to expand his network mostly in the U.S. “I’m excited to explore potential future partnerships and collaborations with people across the U.S. in preparation for our 2027 launch... [and] always open for a general chat and eager to learn more from the people that know.”
Shapanis and Paul Skipp, professor of proteomics at the University of Southampton, developed the miONCO test and spun out the intellectual assets in 2023 to form Xgenera. Skipp serves as chief scientific officer for the six-person company. The miRNA biomarkers and quantification approaches could theoretically be applied to diseases beyond cancer, including Alzheimer’s, asthma, and autoimmune disorders, says Shapanis, but MCED is for now the sole focus and goal.
While working on his Ph.D. in proteomics, Shapanis homed in on skin cancer and found prognostic markers for squamous cell carcinoma, he says. But his dream was to find a pan-cancer diagnostic test, and to that end an extensive literature review and blood sample analyses turned up a solid panel of proteomic markers for specific cancers as well as tumor site-of-origin localization.
After expanding the search a bit, his interest turned to exosomes and, more specifically, what they carry—miRNA. Data from nine different peer-reviewed studies looking at miRNA in serum from 21,000 patients across a dozen different cancer types were used to train machine learning models to predict if cancer was present and where it was in the body, says Shapanis. These models were then applied to two new test sets to reveal the test’s impressive predictive power.
miONCO was originally built from microarray data, which serves as a useful discovery tool to identify important panels but would struggle to scale up to achieve the needs of a screening test, he says. The decision to turn this into a PCR test comes with multiple technical benefits, including “a higher dynamic range where you can go from 100 copies up to a billion copies” of miRNA molecules as well as better resolution for distinguishing closely related targets.” Additionally, porting to a quantitative PCR-based test has benefits for screening purposes as it is easier to adopt in current clinical diagnostic settings since it is “easier to automate and scale up and overall reduces the cost of the test, increasing accessibility.”
Better specificity relates to the fact that miRNA is very homogenous, with some molecules differing by a mere two or three bases, continues Shapanis. “Whilst that is still hard to do in qPCR, it’s easier to do than in microarray and so by tailor-making the qPCR process for our MCED, we can make it as accurate as possible for our specific targets.”
Admittedly, adoption of miONCO as a screening test in the UK may take a while. With the National Health Service (NHS), it’s a multi-step process requiring not just robust evidence of its effectiveness and a demonstration that the benefits outweigh the harms but consideration of ethical, social, and economic implications. Not even GRAIL’s MCED test, Galleri, is available in the UK outside of clinical trials.
That said, Xgenera has received £2.5 million in joint funding from the National Institute for Health and Care Research and the Office for Life Sciences, both closely connected to the NHS, to support development and clinical validation of its miONCO test, Shapanis says.
All told, Xgenera has raised about £4.5 million, over £1 million of that in equity financing. The next round of venture capital will need to be considerably larger to support a clinical trial with tens of thousands of participants in addition to a U.S. LDT launch.
While many of the healthcare companies in the MCED space are focused on the same biomarkers, Xgenera is one of a relatively small number focused on miRNA. This most notably includes the U.S. scientists who won a 2024 Nobel Prize for discovering the molecule governing how cells in the body function. “Now there is lots of hype in this field,” says Shapanis, “[but] luckily we got ahead of it and started this journey much earlier.”
Most of the big MCED developers rely on tests that sequence DNA and analyze methylation patterns in cell-free DNA, including GRAIL as well as Exact Sciences (Cologuard), Guardant Health (Shield), Freenome (LDT test), and Natera (under development). The main problem is that they all suffer from poor early-stage sensitivity, says Shapanis.
“That really boils down to one reason—they are measuring a direct signal from the cancer, so if the cancer gets bigger you get more of a signal, but in the early stages... it’s like trying to find a needle in the haystack,” he continues. “Because we’re looking at microRNA... [that] is circulating in our blood naturally... it is more about how that [miRNA] signature looks as to whether we predict someone has cancer or not.”
Most people working on tests for early cancer detection are unified by their motivation to improve patient outcomes and save lives, says Shapanis. “I’ve lost grandparents to cancer... everyone’s got a personal connection to cancer in one way, shape, or form.”
While Xgenera is in competition with some of these companies in a business sense, “I do still secretly hope that if it’s not us, it’s them,” he says. “I hope someone gets a product through because early diagnosis really is our best strategy currently.”