MiWEndo is a spin-off from the University of Pompeu Fabra, the Hospital Clinic of Barcelona, the Polytechnic University of Catalonia and the Catalan Institute for Research and Advanced Studies, all in Barcelona, Spain.
Colonoscopy has long been considered the gold standard of screening for colorectal cancer. For this test, physicians use a flexible tube with a light and a camera at the end to see inside the lower bowel, viewing the image on a video screen. Any clumps of cells seen on the lining of the colon, known as polyps, can then be removed before they turn cancerous.
But colonoscopy is not infallible — nearly 10% of colorectal cancer cases are initially missed by colonoscopy. MiWEndo Solutions, a start-up in Barcelona, Spain, is trying to reduce that error rate. The company plans to augment colonoscopy with microwave imaging to “improve polyp detection and differentiate between malignant and benign tissues”, says Marta Guardiola, the company’s chief technology officer.
The system under development is based on Guardiola’s research at the University of Pompeu Fabra (UPF) in Barcelona. Guardiola, an electrical engineer, got her start in the field as a postdoc investigating how microwave technology might be used to detect breast cancers and cerebral haemorrhages. She had the idea of applying this approach to scanning the colon four years ago, during a conversation with Glòria Fernàndez Esparrach, a gastroenterologist and endoscopy specialist at the Hospital Clinic of Barcelona (HCP). “We realized that by combining our two disciplines, we could improve the prevention and diagnosis of colorectal cancer, which is one of the most prevalent cancers worldwide,” Guardiola says. Along with other colleagues, the pair co-founded MiWEndo in September 2019 as a spin-off of UPF, HCP, the Polytechnic University of Catalonia in Barcelona and the Catalan Institute for Research and Advanced Studies in Barcelona. Esparrach is the company’s chief medical officer.
A better view
Prateek Sharma, a gastroenterologist at the University of Kansas Medical Center in Kansas City, who is not involved with MiWEndo, says microwave technology could address two important unmet needs related to screening for colorectal cancer. One is the ability to characterize polyps in real time, “since some are benign heaps of normal tissues that can be safely left behind”, he says. The other is greater capacity to locate polyps smaller than one centimetre, which Sharma says are the ones most likely to be missed with standard colonoscopy. The camera of a colonoscope can provide images only in one direction: forward. And although physicians can turn the camera from side to side for a better view, polyps obscured behind flecks of faeces or folds in the colon can still be missed.
MiWEndo’s technology increases the field of vision to 360°. The device fits snugly around the end of a conventional colonoscope, like a ring, and contains miniaturized antennas that project microwaves and detect the reflected waves that bounce back (see ‘Microwave-assisted colonoscopy’). The resulting reflection patterns vary depending on the tissue’s dielectric properties — its capacity to impede electrical currents.
Tumours and pre-cancerous polyps contain more blood vessels and water — and therefore more electrically charged ions — than healthy tissue, so they can be readily distinguished by the company’s algorithms. According to Guardiola, the device can image even opaque tissues, and it sounds an alarm when suspicious lesions come into range, including the very small ones that are often missed with existing approaches. That alert, she says, removes some of the subjectivity from how physicians define what appear to be polyps during a colonoscopy. It also makes examination “less dependent on the experience of endoscopists”. A study of colon sections removed from 23 people showed that the device correctly ruled out the presence of cancer with 100% sensitivity — there were no false negatives1. It also correctly distinguished cancerous polyps from non-cancerous tissues with 95% specificity, which means that the system generated false-positive results only 5% of the time. In those instances, polyps were erroneously classified as cancer. Guardiola points out, however, that all the misclassified lesions were high-grade abnormal cells with pre-cancerous features. Avoiding false negatives, she emphasizes, is the greater priority. The company is planning to test the technology in pigs later this year.
Microwave imaging has yet to be approved for any clinical indication. And MiWEndo is moving into a crowded arena. The market for colonoscopy devices is projected to reach US$2.4 billion by 2025, driven in part by an ageing population. Numerous competing approaches are under development. Chromoendoscopy, for instance, involves spraying the colon with coloured stains that preferentially accumulate in cancerous tissues. And Raman spectroscopy can flag colorectal tumours on the basis of how they interact with light. But these approaches are not perfect. According to Sharma, chromoendoscopy has not been shown to consistently reduce the rate of missed cancers. And Raman spectroscopy, he says, is difficult to adapt to current endoscopy systems. The best improvements in colonoscopy so far, Sharma says, have come from increasing the resolution so that it is easier to spot flat polyps that spread across the colon surface. Researchers are also building artificial-intelligence systems that can improve polyp classification.
Aiden Flanagan, a research and development fellow with Boston Scientific Ireland in Galway, says that what makes MiWEndo’s approach unique is its capacity to locate and classify polyps simultaneously. Boston Scientific makes instruments for endoscopy, and Flanagan says that the company is collaborating with MiWEndo on efforts to bring its microwave imaging technology to the market. “We’re helping them make the business case by looking at market size, physician opinions, and assessments of the regulatory landscape in different countries,” he says.
According to Guardiola, around a dozen other endoscopy companies have also expressed interest. MiWEndo has current seed investments of €1.6 million (US$1.8 million), and is applying for European grants to fund clinical trials that it hopes to launch in Spain and the Netherlands in 2021. “MiWEndo still needs to show the device correctly identifies cancers and polyps on a human scale,” Flanagan says. “Clinical trials will be the ultimate test.”