Accelerating early disease detection with nanobiotechnology

Share this post:

Imagine this scenario:  Annual physical examinations are supplemented by an affordable home diagnostic chip, allowing you to regularly monitor your baseline health with just a simple urine sample. Though outwardly you appear to be in good health, the device reveals a fluctuation in your biomarker profile, indicating the possible emergence of early stage cancer development or presence of a virus.

Diagnostic devices like a home pregnancy test have been around since the 1970s. It revolutionized a woman’s ability to find out if she was pregnant without having to wait for a doctor’s appointment to confirm her suspicions. The test relies on detecting a hormone, human chorionic gonadotropin, present in urine. But could detecting cancer, or a deadly virus, from a similar kind of sample and device be as simple and non-invasive?

nanoDLD chip

IBM Research’s 2cmx2cm nanoDLD chip mounted in a microfluidic jig

Over the last couple of years, a cross-disciplinary team at IBM Research led by Dr. Joshua Smith, research scientist and master inventor, and Dr. Gustavo Stolovitzky, program director of IBM Translational Systems Biology and Nanobiotechnology, have been exploring this idea by retooling silicon technology and using it to separate nanoscale-sized particles like the elements of viruses and cancers.

The answer lies at the intersection of human biology and nanotechnology. First, you need a “liquid biopsy,” which can be obtained from urine or saliva samples. Then the nanometer-tiny biomolecules need to be separated for downstream detection of a disease, even while its presence is physically undetectable. Today’s separation techniques require expensive equipment, a biochemistry lab, and trained technicians, making regular screenings impractical. But a chip-based nanotech approach could offer a simple and affordable option to enable physicians to potentially detect and monitor a disease, even at its earliest stages, long before physical symptoms manifest and when the prognosis for treatment options is most positive. It’s the holy grail for medical professionals: treat people before they get chronically sick.

The team’s big advantage is combining state-of-the-art nanotechnology and fabrication techniques with computational biology expertise. This potent synergy is difficult to come by in other research laboratories.

Know your bioparticles

Separating bioparticles is by no means a new concept, but in their latest experiments, Smith and Stolovitzky’s team is the first to have successfully separated nanoscale biomolecules – specifically exosomes (cell-derived vesicles found in bodily fluids containing genetic cargo released from the mother cell) down to just 20nm in diameter, roughly 1/5,000th of a human hair. At this scale, a host of important bioparticles can be separated, including the aforementioned exosomes, DNA, viruses, and protein complexes, which can potentially signal the early onset of a disease state, existence of a virus soon after exposure, or be used to monitor disease progression.

Using a technology called nanoscale deterministic lateral displacement, or nanoDLD, a liquid sample can be passed, in a continuous flow, through a silicon chip especially designed with an asymmetric pillar array. This array allows the system to sort a microscopic waterfall of particles, separating particles by size down to tens of nanometers resolution. Separation occurs as smaller particles move in a zig-zag motion in the direction of the fluid, while larger particles bump through the array along the direction of pillar asymmetry, like truck drivers forced into the truck lane on a freeway, allowing materials of different sizes to be isolated for downstream detection or analysis.

“The societal impact of this research is that it could enable physicians to detect cancer early…when there are more possibilities of being cured,” Stolovitzky said. “We wanted this research to be in the area of cancer and also the area of detecting DNA and viruses like Zika. Everything reduces down to the same thing: being able to have a small and affordable diagnostic tool that can detect minute quantities of biomarker particles that tell physicians something about a person’s health.”

scale of biology

Scale of biology from human to atoms and the new nanoscale ranges that IBM Research’s nanoDLD device is able to reach [full infographic]

Smith added: “The ability for physicians to regularly screen an individual’s biological profile in an affordable and noninvasive manner has the potential to usher in a new era of preventative healthcare and further our understanding of the factors that contribute to disease progression.”

Using exosomes to pioneer preventative medicine

As part of its on-going development for this technology, IBM Research is collaborating with a team from Mount Sinai Health System, led by Carlos Cordon-Cardo, MD, PhD, Chair of the Department of Pathology at the Mount Sinai Health System and Professor of Pathology, Genetics and Genomic Sciences, and Oncological Sciences at the Icahn School of Medicine. The collaboration brings together Mount Sinai’s domain expertise in cancer and pathology while IBM Research brings its bioinformatics experience and its latest nanoscale separation technology.

In the war on cancer, exosomes are increasingly being viewed as useful biomarkers for the diagnosis and prognosis of malignant tumors. Exosomes are released in easily accessible body fluids such as blood and urine, and represent a precious biomedical tool, as they can reveal the origin of a cancer.

“We have realized that cells talk to themselves by sending messengers.  Some of these messengers are stored and transferred inside various small vesicles or exosomes. We have learned that inside of these envelopes a lot of critical information can be found, including those encoded by proteins, or even our own cell’s messengers in the form of DNA-gene fragments and RNA molecules,” Cordon-Cardo said.

Cordon-Cardo is motivated by the opportunity to use the information carried within exosomes to get ahead of diseases.

“Today we do not pursue ‘healthcare’ in the context of its thorough definition. Today, we conduct essentially what we can define as ‘sick care.’ Healthy individuals, medical professionals and society in general need to be educated in order to accept and implement ‘healthcare.’ We find the time to take our car for its annual inspection, but not ourselves. Often time it’s because we are afraid something could be wrong.  Exosomes could assist in early detection of disease. Questioning and interrogating which messages are starting to be sent between organs to predict what may happen, it would be one of the most interesting new ways to look at healthcare.”

Cordon-Cardo believes that with IBM nanoDLD technology, experts can more effectively eavesdrop on the messages being sent between cells and organs in a way that is noninvasive. Monitoring this intercellular conversation more regularly would allow experts to act as a virtual chaperone, carefully observing until the time comes when either the condition has reversed and the patient has returned to a healthy state, or allowing doctors to decide that something more invasive is justifiably needed.

Image of exosomes

Electron micrograph of exosomes on a surface.

“When we are ahead of the disease we usually can address it well; but if the disease is ahead of us, the journey is usually much more difficult. One of the important developments that we are attempting in this collaboration is to have the basic grounds to identify molecular signatures that can be there very early on,” he said.

Testing with prostate cancer

IBM Research and Mount Sinai plan to test the nanoDLD technology on prostate cancer, the most common cancer in men in the US. In prostate cancer, doctors recommend that men older than 50 have a yearly test for a prostate specific antigen (PSA) taken from a blood sample. However, having an elevated PSA level doesn’t necessarily mean a person has or will get prostate cancer. Conversely, not testing positive for PSA doesn’t preclude an individual from forming prostate cancer either.

The hope is to look for specific, sensitive biomarkers, and exosomes today represent a new frontier offering clues that might hold the answer to whether a person has cancer.

“With exosomes you get a glimpse of what a cancer cell is processing. Cells shed exosomes all the time. If we detect them we can see what is inside without touching the cancer,” Stolovitzky said.

Right now, the research is still in its earliest stages. The IBM Research team’s latest scientific results were published today in Nature Nanotechnology*. As part of next steps, the team hopes to increase the volume and throughput of what the device can handle and to engineer the device so it can separate bioparticles smaller than 20 nanometers, thus taking them into the realm of individual proteins.

They also plan to confirm their device is able to pick up the kinds of biomarkers they’d expect to see in the exosomes of prostate cancer patients, and will continue to work on building a clinical-grade prototype device. Clinical trials would be required before a device could reach the market. But just like the pregnancy test changed the game for expectant women, some day in our future we might be able to use a simple point-of-care device that gives us tell-tale signs of our health before we even notice any symptoms.


*Nature Nanotechnology: Nanoscale Lateral Displacement Arrays for Separation of Exosomes and Colloids Down to 20n, DOI: 10.1038/NNANO.2016.134

Benjamin H. Wunsch (IBM Research), Joshua T. Smith (IBM Research), Stacey M. Gifford (IBM Research), Chao Wang (current affiliation: Arizona State University), Markus Brink (IBM Research), Robert Bruce (IBM Research), Robert H. Austin (Princeton University), Gustavo Stolovitzky (IBM Research), and Yann Astier (current affiliation: Roche Molecular Systems)

More Nanotechnology stories

Text2Scene: Generating Compositional Scenes from Textual Descriptions

At CVPR 2019, IBM researchers introduce techniques to interpret visually descriptive language to generate compositional scene representations from textual descriptions.

Continue reading

Overcoming Challenges In Automated Image Captioning

At CVPR 2019, IBM researchers introduce an improved method to bridge the semantic gap between visual scenes and language to produce diverse, creative and human-like captions.

Continue reading

AI Models Predict Breast Cancer with Radiologist-Level Accuracy

Our team of IBM researchers published research in Radiology around a new AI model that can predict the development of malignant breast cancer in patients within the year, at rates comparable to human radiologists.

Continue reading