A simple ten-minute universal cancer test that can be detected by the human eye or an electronic device - published in Nature Communications (Dec 2018) by the Trau lab at the University of Queensland. Red indicates the presence of cancerous cells and blue doesn't.
Matt Trau, a professor of chemistry at the University of Queensland, stunned the science world back in December when the prestigious journal Nature Communications published his lab's discovery about a unique property of cancer DNA that could lead to a simple, cheap, and accurate test to detect any type of cancer in under 10 minutes.
No one believed it. I didn't believe it. I thought, "Gosh, okay, maybe it's a fluke."
Trau granted very few interviews in the wake of the news, but he recently opened up to leapsmag about the significance of this promising early research. Here is his story in his own words, as told to Editor-in-Chief Kira Peikoff.
There's been an incredible explosion of knowledge over the past 20 years, particularly since the genome was sequenced. The area of diagnostics has a tremendous amount of promise and has caught our lab's interest. If you catch cancer early, you can improve survival rates to as high as 98 percent, sometimes even now surpassing that.
My lab is interested in devices to improve the trajectory of cancer patients. So, once people get diagnosed, can we get really sophisticated information about the molecular origins of the disease, and can we measure it in real time? And then can we match that with the best treatment and monitor it in real time, too?
I think those approaches, also coupled with immunotherapy, where one dreams of monitoring the immune system simultaneously with the disease progress, will be the future.
But currently, the methodologies for cancer are still pretty old. So, for example, let's talk about biopsies in general. Liquid biopsy just means using a blood test or a urine test, rather than extracting out a piece of solid tissue. Now consider breast cancer. Still, the cutting-edge screening method is mammography or the physical interrogation for lumps. This has had a big impact in terms of early detection and awareness, but it's still primitive compared to interrogating, forensically, blood samples to look at traces of DNA.
Large machines like CAT scans, PET scans, MRIs, are very expensive and very subjective in terms of the operator. They don't look at the root causes of the cancer. Cancer is caused by changes in DNA. These can be changes in the hard drive of the DNA (the genomic changes) or changes in the apps that the DNA are running (the epigenetics and the transcriptomics).
We don't look at that now, even though we have, emerging, all of these technologies to do it, and those technologies are getting so much cheaper. I saw some statistics at a conference just a few months ago that, in the United States, less than 1 percent of cancer patients have their DNA interrogated. That's the current state-of-the-art in the modern medical system.
Professor Matt Trau, a cancer researcher at the University of Queensland in Australia.
(Courtesy)
Blood, as the highway of the body, is carrying all of this information. Cancer cells, if they are present in the body, are constantly getting turned over. When they die, they release their contents into the blood. Many of these cells end up in the urine and saliva. Having technologies that can forensically scan the highways looking for evidence of cancer is little bit like looking for explosives at the airport. That's very valuable as a security tool.
The trouble is that there are thousands of different types of cancer. Going back to breast cancer, there's at least a dozen different types, probably more, and each of them change the DNA (the hard drive of the disease) and the epigenetics (or the RAM memory). So one of the problems for diagnostics in cancer is to find something that is a signature of all cancers. That's been a really, really, really difficult problem.
Ours was a completely serendipitous discovery. What we found in the lab was this one marker that just kept coming up in all of the types of breast cancers we were studying.
No one believed it. I didn't believe it. I thought, "Gosh, okay, maybe it's a fluke, maybe it works just for breast cancer." So we went on to test it in prostate cancer, which is also many different types of diseases, and it seemed to be working in all of those. We then tested it further in lymphoma. Again, many different types of lymphoma. It worked across all of those. We tested it in gastrointestinal cancer. Again, many different types, and still, it worked, but we were skeptical.
Then we looked at cell lines, which are cells that have come from previous cancer patients, that we grow in the lab, but are used as model experimental systems. We have many of those cell lines, both ones that are cancerous, and ones that are healthy. It was quite remarkable that the marker worked in all of the cancer cell lines and didn't work in the healthy cell lines.
What could possibly be going on?
Well, imagine DNA as a piece of string, that's your hard drive. Epigenetics is like the beads that you put on that string. Those beads you can take on and off as you wish and they control which apps are run, meaning which genetic programs the cell runs. We hypothesized that for cancer, those beads cluster together, rather than being randomly distributed across the string.
Ultimately, I see this as something that would be like a pregnancy test you could take at your doctor's office.
The implications of this are profound. It means that DNA from cancer folds in water into three-dimensional structures that are very different from healthy cells' DNA. It's quite literally the needle in a haystack. Because when you do a liquid biopsy for early detection of cancer, most of the DNA from blood contains a vast abundance of healthy DNA. And that's not of interest. What's of interest is to find the cancerous DNA. That's there only in trace.
Once we figured out what was going on, we could easily set up a system to detect the trace cancerous DNA. It binds to gold nanoparticles in water and changes color. The test takes 10 minutes, and you can detect it by eye. Red indicates cancer and blue doesn't.
We're very, very excited about where we go from here. We're starting to test the test on a greater number of cancers, in thousands of patient samples. We're looking to the scientific community to engage with us, and we're getting a really good response from groups around the world who are supplying more samples to us so we can test this more broadly.
We also are very interested in testing how early can we go with this test. Can we detect cancer through a simple blood test even before there are any symptoms whatsoever? If so, we might be able to convert a cancer diagnosis to something almost as good as a vaccine.
Of course, we have to watch what are called false positives. We don't want to be detecting people as positives when they don't have cancer, and so the technology needs to improve there. We see this version as the iPhone 1. We're interested in the iPhone 2, 3, 4, getting better and better.
Ultimately, I see this as something that would be like a pregnancy test you could take at your doctor's office. If it came back positive, your doctor could say, "Look, there's some news here, but actually, it's not bad news, it's good news. We've caught this so early that we will be able to manage this, and this won't be a problem for you."
If this were to be in routine use in the medical system, countless lives could be saved. Cancer is now becoming one of the biggest killers in the world. We're talking millions upon millions upon millions of people who are affected. This really motivates our work. We might make a difference there.
AI was integral to creating Moderna's mRNA vaccine against COVID.
But the truth is that before ChatGPT nabbed the public’s attention, AI was already here, and it was doing more important things than writing essays for lazy college students. Case in point: It was key to saving the lives of tens of millions of people.
AI-designed mRNA vaccines
As Dave Johnson, chief data and AI officer at Moderna, told MIT Technology Review‘s In Machines We Trust podcast in 2022, AI was integral to creating the company’s highly effective mRNA vaccine against COVID. Moderna and Pfizer/BioNTech’s mRNA vaccines collectively saved between 15 and 20 million lives, according to one estimate from 2022.
Johnson described how AI was hard at work at Moderna, well before COVID arose to infect billions. The pharmaceutical company focuses on finding mRNA therapies to fight off infectious disease, treat cancer, or thwart genetic illness, among other medical applications. Messenger RNA molecules are essentially molecular instructions for cells that tell them how to create specific proteins, which do everything from fighting infection, to catalyzing reactions, to relaying cellular messages.
Johnson and his team put AI and automated robots to work making lots of different mRNAs for scientists to experiment with. Moderna quickly went from making about 30 per month to more than one thousand. They then created AI algorithms to optimize mRNA to maximize protein production in the body — more bang for the biological buck.
For Johnson and his team’s next trick, they used AI to automate science, itself. Once Moderna’s scientists have an mRNA to experiment with, they do pre-clinical tests in the lab. They then pore over reams of data to see which mRNAs could progress to the next stage: animal trials. This process is long, repetitive, and soul-sucking — ill-suited to a creative scientist but great for a mindless AI algorithm. With scientists’ input, models were made to automate this tedious process.
“We don’t think about AI in the context of replacing humans,” says Dave Johnson, chief data and AI officer at Moderna. “We always think about it in terms of this human-machine collaboration, because they’re good at different things. Humans are really good at creativity and flexibility and insight, whereas machines are really good at precision and giving the exact same result every single time and doing it at scale and speed.”
All these AI systems were in put in place over the past decade. Then COVID showed up. So when the genome sequence of the coronavirus was made public in January 2020, Moderna was off to the races pumping out and testing mRNAs that would tell cells how to manufacture the coronavirus’s spike protein so that the body’s immune system would recognize and destroy it. Within 42 days, the company had an mRNA vaccine ready to be tested in humans. It eventually went into hundreds of millions of arms.
Biotech harnesses the power of AI
Moderna is now turning its attention to other ailments that could be solved with mRNA, and the company is continuing to lean on AI. Scientists are still coming to Johnson with automation requests, which he happily obliges.
“We don’t think about AI in the context of replacing humans,” he told the Me, Myself, and AI podcast. “We always think about it in terms of this human-machine collaboration, because they’re good at different things. Humans are really good at creativity and flexibility and insight, whereas machines are really good at precision and giving the exact same result every single time and doing it at scale and speed.”
Moderna, which was founded as a “digital biotech,” is undoubtedly the poster child of AI use in mRNA vaccines. Moderna recently signed a deal with IBM to use the company’s quantum computers as well as its proprietary generative AI, MoLFormer.
Moderna’s success is encouraging other companies to follow its example. In January, BioNTech, which partnered with Pfizer to make the other highly effective mRNA vaccine against COVID, acquired the company InstaDeep for $440 million to implement its machine learning AI across its mRNA medicine platform. And in May, Chinese technology giant Baidu announced an AI tool that designs super-optimized mRNA sequences in minutes. A nearly countless number of mRNA molecules can code for the same protein, but some are more stable and result in the production of more proteins. Baidu’s AI, called “LinearDesign,” finds these mRNAs. The company licensed the tool to French pharmaceutical company Sanofi.
Writing in the journal Accounts of Chemical Research in late 2021, Sebastian M. Castillo-Hair and Georg Seelig, computer engineers who focus on synthetic biology at the University of Washington, forecast that AI machine learning models will further accelerate the biotechnology research process, putting mRNA medicine into overdrive to the benefit of all.
This article originally appeared on Big Think, home of the brightest minds and biggest ideas of all time.
