Project aims to ward off illness before it happens

Study looks for markers in healthy people that could help predict cancer, other diseases before symptoms occur

One of the sobering facts about cancer treatment is that it often begins when it is already too late: Studies show that an alarming number of treatable cancers are diagnosed in the advanced stages of disease.

That has long bothered Dr. Sam Gambhir, a top cancer researcher at Stanford University who lost his teenage son to brain cancer in 2015. Gambhir wondered if there was some surefire way to detect cancer long before people got sick.

“In the cancer field we often find problems long after people have symptoms,” he said. “We rarely find things early.”

Now Gambhir is leading a large study that seeks to better understand the transition from normal health to disease. The study, called Project Baseline, could lead to the identification of new markers in the blood, stool or urine of healthy people that help predict cancer, cardiovascular disease and other leading killers of Americans.

It is a joint effort between Stanford and Duke universities and Verily, a life sciences company owned by Alphabet, the parent company of Google. Researchers are recruiting 10,000 adults across the United States who will be examined in extreme detail and followed intensively for at least four years.

Many of the people joining the study are healthy adults, which differs from traditional medical trials that focus largely on people who are already ill. Another key difference is that the researchers are collecting a staggering amount of medical data on their subjects: analyzing their microbiomes, sequencing their genomes, subjecting them to a variety of scans and assessing their cognitive health. They are also equipping volunteers with new wearable technology from Verily that records their nightly sleep patterns and tracks their heart rhythms and physical activity.

In another unusual move, the Project Baseline investigators are sharing the research results with their subjects, everything from how much plaque or calcium they find in their arteries to which bacterial strains inhabit their guts.

Some experts worry, however, that providing such detailed medical data to healthy adults could lead to new problems. Dr. Eric Topol, a professor of genomics at the Scripps Research Institute in California, cautioned that the sheer amount of testing, scans and other “deep interrogations” could produce incidental findings that cause unnecessary anxiety. “Sometimes it leads to unwarranted further testing that could even be harmful,” he said.

Topol is involved in a similar study, called All of Us and financed by the National Institutes of Health, which is building a “biobank” of health information collected from 1 million Americans. The researchers intend to return genetic data and some other results to participants but are figuring out the best way to do that.

“This is the new challenge in a democratized world of medical research,” Topol said. “I’m really in favor of it, but it sets up this new issue of dealing with unexpected results that are difficult to interpret.”

The Project Baseline researchers are learning this firsthand. They say they have discovered and promptly alerted participants to potentially lethal conditions that might otherwise have gone unnoticed, like cancer and aortic aneurysms, so they can seek appropriate medical care. But some of the participants have also been frightened by fairly innocuous findings, like chest X-rays that reveal small, usually benign nodules in their lungs that they may look up on the internet and think are cancerous, said Charlene Wong, a Project Baseline investigator.

“For most of our participants, it will not be cancer. But we’re still in the process of working with participants to find out if we can return that data in the right way so that we minimize the anxiety it can cause,” she said.

While there have been plenty of longitudinal studies in the past, many of the largest and most important were not very diverse. The landmark Framingham heart study that began in 1948, for example, focused mostly on white adults. Dr. Svati Shah, an associate professor of medicine at Duke, said Project Baseline is recruiting many people who are black, Hispanic, Asian and other ethnicities so the study can shed light on differences in disease risk factors among people of different backgrounds.

That includes people like Rosa Gonzalez, 57, a nurse who lives in Concord, N.C. Gonzalez, who is Mexican-American, joined the study earlier this year and has encouraged at least a dozen Latino friends and acquaintances to join it as well.

“Other studies present data and talk about Latinos, but they don’t have Latinos in the study,” she said. “I’m trying to set an example so other Latinos see that it’s good to take part so that we can have data that shows how we’re the same or different.”

Gambhir said the idea for Project Baseline was hatched in 2013, when Google executives approached him and said they wanted to do a landmark study on human health. Gambhir proposed a study to find early markers of cancer in people who are otherwise healthy.

“We have always thought that if we learn more about what your body is doing before you become ill, then we would have a much better chance of ideally preventing or at least detecting things early,” he said. Google liked the idea but suggested broadening the scope to include other diseases.