NEW YORK – A Wales-based diagnostic company is drawing on biomarker research from Aberystwyth University to create a point-of-care test to diagnose lung cancer from a patient's urine sample.
The lateral flow test, which is currently in development, from iOmics detects five proprietary biomarkers licensed from Aberystwyth that could indicate the presence of non-small cell lung cancer and potentially other types of lung cancers.
Discovering the five biomarkers was a process that required taking sputum, blood, and urine samples from 127 patients and 46 controls across the UK and first looking for differences in metabolites across the samples, said Keir Lewis, iOmics' clinical lead and a professor of respiratory medicine at Swansea University.
The samples were run through high resolution-mass spectrometers to generate measures of more than 5,000 chemicals in the samples that were then mapped to clinical characteristics, such as the presence of early-stage cancer or lung cancer type. Once the chemicals were mapped, the field was narrowed down to five biomarkers that, combined, seem "to accurately predict the presence of lung cancer," Lewis said. Those five markers include chemicals that are involved in cellular turnover and cell membrane synthesis "so you'd expect them to be raised or lowered in cancer if you've got increased cellular turnover."
The team used advanced statistical approaches to determine which compounds were raised or lowered in patients with lung cancer compared to patients without and then applied machine learning tools to see what combinations of those compounds produced the best results. "If you can see something in the [cluster analysis] that differentiates the two groups, you definitely can say 'Well, actually it doesn't happen often, or not at all, in healthy people, but it's always raised in lung cancer,' [and] that's your best biomarker."
The levels of five of those cell combinations were raised in most of the 97 patients with non-small cell lung cancer in the sample cohort, Lewis said. Although many groups focus on blood, the company decided to go with urine samples because collection is "easy to do" and more pleasant for patients while presumably detecting the same chemicals.
"The same blood that goes through your lungs where the cancer is also goes through the kidneys," he noted.
The overall diagnostic accuracy of the biomarkers was between 87 percent and 92 percent with an area under the curve of .91, results Lewis said the team was "really happy with." Those results outperform the accuracy rates of chest X-rays and are close to the accuracy rates of CT scans, with the benefit of being "a lot cheaper" to detect, not requiring a specialist to interpret, and not involving radiation.
Two patents for the combination of metabolites in urine that will be used in the test have been accepted by the US Patent and Trademark Office, with iOmics expecting the patents to go through in the next couple of months, Lewis said. Until then, the names of the biomarkers are under a nondisclosure agreement and further data on the markers won't be published. Lewis did share that some of the biomarkers are connected to altered steroid levels.
Lewis was careful to note that the diagnostic capabilities of the biomarkers need to be tested on a broader cohort of patients, which the iOmics team is working on now. A larger study of 1,600 patients to validate the markers is already funded, with some help from the university and a significant portion from the UK government, he said. The company is also seeking funding for the development of a point-of-care device with a manufacturing partner to ensure the company can hit the ground running if the biomarkers are further validated.
"We think [the biomarkers are] there. They make biological sense. We just want to be sure that the results are as good as they look," Lewis said.
That larger study will also help iOmics determine if the biomarkers can be used to differentiate between stages of lung cancer and detect early lung cancer, he said. The study will look at patients with small cell lung cancer to determine if the biomarker levels are different depending on the type of cancer. The team has also seen different signals in patients with primary lung cancer and those with secondary lung cancer — i.e., cancer that originated elsewhere but has spread to the lungs — that could prompt it to look elsewhere in the body.
The firm and its partners at Aberystwyth University are doing further work on urine biomarkers for prostate and gynecological cancers, as well.
As the lung cancer test is further developed, iOmics is looking to partner with other diagnostic companies that can guide the company in bringing the test to market, since it doesn't have the commercialization experience. Moving forward with the test would then be "a question … of labeling antibodies and developing a lateral flow test band that's small enough and cheap enough to be mass produced," Lewis said.
While the use case of the test depends on the results of the larger study, Lewis said he sees it being effective in primary care settings, and since there are only five biomarkers, it has "the potential [to be] a handheld or point-of-care test."
It could also be built into national cancer screening programs. Sinan Eccles, a consultant respiratory physician at Royal Glamorgan Hospital who is not involved with the company, said that a biomarker test could be used as a first step in a screening program to pick out high-risk patients who should have a CT scan.
Eccles cautioned that implementing a biomarker test as part of a national program would require "pretty strong evidence," not just of the accuracy of detecting more cancers but that it reduces mortality as well. "Just detecting more cancers is probably not enough," he said.
The test could also be used "opportunistically" beyond a large-scale screening program, such as in settings where patients are worried about lung cancer or are symptomatic but don't meet the criteria to participate in a screening program, he said.
Although early cancer detection and diagnostics is a hot field, with more companies entering the market each year, Lewis noted that the vast majority of these firms are looking at blood — specifically circulating tumor DNA in the blood, rather than metabolites. He emphasized that iOmics isn't looking to compete with ctDNA blood tests, but rather to provide another method of detection.