People of certain ages or with certain risk factors know that they need regular cancer screening. Men over 50 often get prostate exams, the fair-skinned have moles checked annually and those with cancer in their family histories can opt to have genetic testing done. But what happens if a person has no apparent symptoms or risk factors? In many instances, cancer has an unfortunate way of sneaking up on both patients and doctors. In the worst cases, it goes undetected until it reaches later stages of metastasis (Stage 3 and Stage 4), when tumors spread to nearby lymph nodes, tissues and organs or other areas of the body.
Thankfully, detection could soon become so simple that it could be a part of an annual physical exam, along with screening for diabetes and high cholesterol.
At the recent TEDGlobal conference in Rio de Janeiro, Miroculus, a startup founded in part by microbiologists and data scientists, debuted Miriam. The device, which is currently a prototype, will screen for dozens of cancers by looking for biomarkers known as microRNAs in easy-to-draw blood samples.
Many researchers, including Miroculus co-founder Fay Christodoulou, have shown that microRNAs—small molecules that control how our genes are expressed and the action of proteins in the body—are biomarkers that might indicate certain types of cancer, including lymphoma and breast, prostate and lung cancer. Unlike other invasive and prolonged cancer screening methods, including mammograms and biopsies, doctors can acquire a microRNA sample from blood and use it to spot cancers in otherwise asymptomatic patients. Essentially, the presence of a single microRNA or a grouping of them (the human genome has over 2,000) acts as a fingerprint for disease.
If that sounds too good to be true, that’s because, up until now, it has been. Though microRNA tests do exist, they require large, expensive equipment, highly trained lab techs and patience for long turnaround times. One lung cancer test, for instance, costs nearly $6,400 and takes seven days to return results. When Miroculus launches Miriam, the device will cost about $510 and nurses will be able to operate it. Each test will cost as little as $150 and produce results within 90 minutes.
According to Chief Technology Officer Jorge Soto, Miriam represents an important inflection point in microRNA research. “There’s a lot of scientific validation [for the usefulness of microRNA],” he says, “but there is not a lot of clinical validation. We are at the beginning of a new wave that will study the clinical application.”
Soto and Christodoulou met in 2008, when she shared with him her idea for a device that could trap microRNA in a simple “point-of-care device.” Because microRNA research is already a 20-year-old field, the team was able to pull their concept together within a month and build a clinical-trial-ready prototype within 18 months.
The Miroculus works in three steps: sample, react and analyze. First, a tech or doctor uses an off-the-shelf kit to extract RNA from a one-milliliter blood sample. Then, he or she places drops of the extracted sample into 96 wells on a prepared plate, known as a bioassay. Each well on the plate is pre-treated with its own bio-chemical mix that’s designed to react with a specific type of microRNA. The tech then places the plate inside a freestanding device that seals the plate off from light and keeps it at the proper temperature to incite a reaction. If a well glows, the specified mircoRNA is present.
The tech places a smartphone on top of the device so that its camera can peer inside. (Later versions of Miriam will have their own onboard computer.) Over the course of 60 minutes, the camera takes a series of pictures of the wells and tracks changes—including which wells are glowing, how frequently and how intensely. Data is sent to Miroculus’s cloud server for analysis. It compares those results with existing data to determine if any patterns are present that might indicate a particular type of cancer.
So far, the team has been able to identify lung, breast and pancreatic cancer in mice. The researchers are also participating in a clinical trial in Germany that includes 200 human breast cancer patients.
But Miroculus will have competition, eventually. A similar research effort in Japan, involving the country’s National Cancer Center, aims to bring a similar product to market within five years. The group's work will analyze blood samples from some 6,500 patients in order to spot the microRNA signatures of 13 types of cancer.
While these methods show promise, experts do caution that we still need more data to make microRNA diagnosis as foolproof as Miroculus plans. Simple things like a common cold or taking an aspirin can affect the presence of microRNA in the blood, Muneesh Tewari of the Fred Hutchinson Cancer Research Center explained to Wired. Therefore, building out Miroculus’s database of reference points will be just as important as the accuracy of the bioassay itself.
Over the next several months, Miroculus will release Miriam to pharmaceutical manufacturers, who will use it to test the effectiveness of new medications. It’s a quid-pro-quo arrangement, in which Miroculus will also reap the benefits of that data in order to further build out its system’s understanding of microRNA patterns. “The populations for those trials are controlled,” says Soto, “so they will provide better data.” He estimates it will be three to five years before a product will clear the FDA for clinical diagnosis.
Since Soto’s presentation at TED, the company has found other avenues for data collection. “We’ve been approached by a lot of hospitals and sample banks, telling us they want to volunteer themselves and their data,” he explains. “We also have interest in trials from all over the world—from India to Japan to the Middle East to Central America and the U.S. That’s very powerful for us, because it will help us in reaching our goals.”