Precision medicine aims to carefully target cancer and other diseases with the right drug at the right time for the right patient (a mantra that is getting a bit old). It’s a worthwhile goal, but it spends a lot of health care time, energy, and money on the tail end of disease.

I believe we should be more aggressively pursuing precision health: ways to prevent disease and, when that isn’t possible, intercept and treat it earlier.

My colleagues and I at Stanford have been working on novel technologies that, although they are still in development stages, point the way to how precision health could work.


Precision health begins with customized risk monitoring based on models drawn from every individual’s age, family history, genetic sequencing, and exposome (the collection of nongenetic exposures that affect human health and disease). Each of us would be monitored in ways that are unique to our risks, which would likely be different from our co-workers or neighbors, who have different disease risks.

This kind of monitoring would occur in several ways. Devices like the Fitbit and Apple Watch, already used by millions of people, provide ways to measure physiology. Integrating sensors into clothing will extend their reach. But we must also monitor biomarkers in the blood, in exhaled air, in saliva, and in urine and stool. This would ideally be done by home devices that don’t require you to wear something or change your behavior.

The smart toilet represents one such device. In developed countries, every home has a toilet, which the inhabitants use every day. Adapting this tool to detect markers of disease such as blood in the stool, sugar in the urine, or circulating tumor cells and send warning signals to the user — and his or her physician — could detect disease while it is still preventable, or at least treatable.

Early cancer detection presents a challenge: the amount of a biomarker or the number of circulating cancer cells in the blood is extremely low. A tube of blood or two, or six, might have too little to measure. Our team decided to flip around the problem. Instead of bringing blood to the sensor, we brought the sensor to the blood. We developed the MagWIRE, a magnetic wire that attracts circulating cancer cells labeled with magnetic nanoparticles. Inserted into a vein for 10 to 20 minutes, it would be able to “scan” almost the entire blood volume for cancer cells and capture them, allowing for accurate measurement.

But what about cancers that aren’t showing their hand, or aren’t making the biomarkers we’ve learned to look for? It’s possible to force them to reveal themselves using tiny rings of single-stranded DNA called minicircles. We initially used minicircles that code for a protein called secreted embryonic alkaline phosphatase, which isn’t present in adult cells. The minicircles also contain a promoter that regulates production of a protein called survivin. The promoter is “on” only in cancer cells. When we injected these minicircles into mice, all cells took them up. Two days later, embryonic alkaline phosphatase began to appear in the blood of mice with cancer, but not in those without.

If this technology pans out in humans, you could pop an early cancer detection pill once a year and then have your blood or urine tested for the molecule wired into the minicircle. This technology would also let doctors see where those active cancer cells are by forcing them to make a signal that shows up on imaging.

To make early cancer detection happen in real time, rather than on artificial monthly or yearly schedules, we have been looking at ways to re-engineer immune cells to constantly patrol the body for signs of cancer. When such genetically modified immune cells encounter cancer cells, they send a signal by secreting a compound of our choosing into the blood that could be measured in the urine by a smart toilet.

This represents a powerful approach because those cells would potentially live with you forever, serving as 24/7 cancer detectors.

To make these kinds of advances work, we must develop a better understanding of the biology of cancer and other diseases. Today, the technology is way ahead of the biology, especially regarding the transition from health to illness. Much more work is needed to identify the earliest tell-tale signs of illness and their molecular signs.

To do that, my colleagues and I have been working with Google and Verily on Project Baseline. Designed from the ground up to study in a deep way what’s happening when you are healthy, it will offer clues as to what happens when people transition to illness.

The reality of precision health is still a long way off. But the goal of preventing disease, and identifying it early when it occurs, is a payoff worth working for.

Sanjiv Sam Gambhir, M.D., is chair of radiology at Stanford University School of Medicine and director of the Canary Center at Stanford for Cancer Early Detection.

Leave a Comment

Please enter your name.
Please enter a comment.

Sign up for our Daily Recap newsletter

A roundup of STAT’s top stories of the day in science and medicine

Privacy Policy