If you are among the third of the population who will someday develop cancer, your body will contain warning signs well before your doctor is able to diagnose the disease. If these subtle signals in your cells and your bloodstream could only be detected sooner, you’d have a far greater chance of surviving. The problem is that the changes that mark the early stages of cancer are remarkably complex – and often slight, even on a molecular level.
But James Heath, a physical chemist at the California Institute of Technology, believes that nanotechnology could finally provide the solution to this molecular riddle. Heath is betting that banks of ultrasmall silicon wires, each made to detect a specific cancer-related protein, could pick up even the most subtle changes in our body chemistry. The nanosensors that Heath and his Caltech coworkers are developing will simultaneously look for hundreds or even thousands of different biomolecules in, say, a drop of blood. If they work, these nanosensors could be the basis for cancer tests that are not only more accurate but, because they don’t involve tissue sampling and lab analysis, cheaper and more convenient than those now available.
That’s not saying much, of course. Screening for most cancers remains primitive, often involving simple physical exams to find evidence of tumor growth, or crude imaging methods such as mammography and x-rays. Blood tests exist for a few cancers, such as prostate and ovarian cancers, but their performance is woeful; not only are they slow and costly, but they’re notoriously unreliable. To diagnose prostate cancer, for example, doctors look for a protein called PSA (prostate-specific antigen) in the blood. But only 25 to 30 percent of men who go through the immensely stressful process of having tissue biopsies because of high PSA levels in their blood actually have prostate cancer. “PSA is always in the prostate,” points out Heath, “and is leaked out into the blood in small quantities all the time. When there is some sort of trauma to the prostate – which could be cancer or something else – it leaks out in greater quantities. But it is a very poor marker for early-stage prostate cancer, since there really isn’t too much trauma to the prostate at that stage.”
A more accurate cancer test would better reflect the complexity of biomolecular events. Heath’s ambition is to construct devices that can not only make multiple measurements at once, from a drop of blood or a few cells taken from a particular tissue, but also detect extremely small quantities of biomolecules. “We are trying to develop a finger prick–based test,” he explains. “We would like this test to eventually be something analogous to what is used for diabetics. Diabetics can now monitor their glucose levels, and because they can do that on a regular basis, they take control of the disease. We would like to develop a similarly enabling platform for cancer.”
Piecing Together the Puzzle
Cancer research might seem an unlikely place for James Heath to have ended up. As a graduate student at Rice University in Houston during the early 1980s, he began studying the properties of tiny chunks of materials. He was part of the team that, in 1985, discovered the soccer ball–shaped carbon molecule C60; the discovery won Heath’s professor, Richard Smalley, a Nobel Prize 11 years later and helped launch today’s interest in nanotech. But Heath later shifted his focus to semiconductors, such as silicon, used by the microelectronics industry, looking for ways to fashion them into ever smaller devices. Recently, he and collaborators at the University of California, Santa Barbara, devised a method for making silicon wires just a few nanometers wide, about ten times smaller than the smallest features in today’s integrated circuits.
The advance was a milestone in the continued miniaturization of electronics. And, says Heath, “We hoped that by solving such a difficult problem, other opportunities would present themselves.” They did: Heath realized these nanowires could also serve as ultrasensitive biosensors.
He also realized, however, that incorporating nanowires into an effective diagnostic tool would not be easy. Changes in a person’s state of health are reflected in wild swings in concentrations of biomolecules as different genes switch on and off. But over the past several years, geneticists and molecular biologists have come to realize that genes don’t generally act independently. They tend to operate in groups and networks, and they can regulate each other’s expression. So making sense of the molecular “fingerprints” of disease requires a systems-level understanding of how genes and proteins work together.
That’s where Heath’s collaborator, Leroy Hood, founder of the Institute for Systems Biology in Seattle, comes in. Systems biologists look at the cell much as an electrical engineer looks at a complex circuit: as a highly interconnected system of components that switch each other on and off and relay signals. Heath’s sensors might provide thousands of clues to a person’s state of health, but Hood’s systems-biology approach is needed to piece all those bits of information together into a coherent picture.
Hood and his team have, for example, looked at how genes are expressed to produce proteins in cells and tissues affected by prostate cancer. “Our idea,” says Hood, “is that the difference between normal and diseased cells is that the protein and gene regulatory networks in diseased cells have been perturbed, and these disease perturbations are reflected in altered patterns of protein expression controlled by the networks. A fraction of these perturbed proteins will find their way into the blood and constitute molecular fingerprints that are diagnostic not only of health and disease but of what disease and what type of a particular disease.” (There are at least three different types of prostate cancer, for example.)
“We have identified 300 [cancer marker] genes that are uniquely expressed in the prostate,” says Hood, “and we predict that about 62 of these may be secreted into the blood. We tested one of these by making antibodies against it and demonstrated that it was only present in the blood of patients with prostate cancer.” Hood’s team is now testing five more prostate cancer–secreted proteins. It has also found a similar array of genes that should be diagnostic for ovarian cancer.
A Fluid Situation
What exactly would a nanosensor to detect such proteins look like? To turn a nanowire into a transistor, the researchers bring each of its ends into contact with metal wires so that a current can be passed through it. They then position an electrode close to the nanowire. Charging this electrode alters the conductivity of the nanowire, turning it “on” and “off” – all familiar stuff to any electrical engineer.
Heath then transforms his nanowire transistors into tiny biosensors. Say, for instance, that one nanowire is to act as a sensor for a particular protein. The researchers coat the surface of the wire with antibodies that will stick to the target protein but not to other molecules. When proteins bind to the antibodies, they interact with the electrons traveling in the nanowire’s surface layer, altering its conductivity. If the wire is only a few nanometers thick, there is a significant – and measurable – change in its overall conductivity. “If the wire is really, really small,” says Heath, “instead of putting a voltage on it, we can put molecules on it, and a chemical event is what causes the transistor to switch.”
Their small size also makes the devices very sensitive. Ultimately, the number of molecules required to produce a reading will depend on how tightly they bind to the receptor groups on the sensor surface; but it might be possible to detect individual molecules. Heath says that, although his group has not yet reached that level of sensitivity, it has succeeded in detecting just a few molecules. (Charles Lieber of Harvard University, meanwhile, has demonstrated nanosensors that can detect a single viral particle*).
But it’s not just high sensitivity that Heath is relying on for easy and early detection of disease. “We can make thousands of these sensors in a very small area,” he says. This means the ability to screen the varied molecular contents of individual cells. Heath is collaborating with Stanford University microfluidics expert Stephen Quake to fabricate chips in which fluids pumped down microscopic channels shuttle single cells into position over a nanosensor array, where they can be studied one at a time.
In the end, all this technology has to be integrated in a device that can be used in the clinic, which means solving yet more technical and practical problems. In 2003, the Institute for Systems Biology, Caltech, and the University of California, Los Angeles, established the NanoSystems Biology Alliance to ensure that the new tools reflect the latest advances in cancer biology and immunology. The diagnosis of cancer and other diseases, says Quake, will be “carried out automatically, in a few seconds or minutes, on just a handful of cells or their contents.” And that conjecture, he predicts, “will be turned into a reality within this decade.”
Philip Ball’s latest book is called Critical Mass: How One Thing Leads to Another.
DeepMind’s cofounder: Generative AI is just a phase. What’s next is interactive AI.
“This is a profound moment in the history of technology,” says Mustafa Suleyman.
What to know about this autumn’s covid vaccines
New variants will pose a challenge, but early signs suggest the shots will still boost antibody responses.
Human-plus-AI solutions mitigate security threats
With the right human oversight, emerging technologies like artificial intelligence can help keep business and customer data secure
Next slide, please: A brief history of the corporate presentation
From million-dollar slide shows to Steve Jobs’s introduction of the iPhone, a bit of show business never hurt plain old business.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.