stanford medicine


Test questions

Special Report

Test questions

Consumer genomics enters the classroom

Stanford graduate student Steve Harris stared at the unopened file on his computer screen containing his risk for Parkinson’s disease.

Did he want to open it? Six weeks earlier, he had spit into a test tube, sealed it up, then shipped it off via FedEx to a genetics-testing lab. Now he had the results before him — his risk for more than 80 major health conditions, the Parkinson’s result in a separate file.

He recalled his father’s torturous seven-year struggle with the disease and felt sadness wash over him. “I sat there for 10 minutes, but I didn’t open it,” said Harris, a sixth-year MD/PhD student, who asked that his real name not be used. When he later returned to the file, he still couldn’t bring himself to learn the result.

Better, he decided, at age 30 not to live with the anxiety of knowing he had a dramatically increased risk of a terrible neurological condition with no known cure.

“This information is coming. We need to learn how to apply it for the benefit of our population.”

Harris is a pioneer of sorts, one of 54 students in a School of Medicine class called Genomics and Personalized Medicine. The historic class, the first in the country to offer students the option of commercial genotyping, was approved after highly charged debates that Dean Philip Pizzo, MD, said were “among the most important that our faculty and students have engaged in.”

The eight-week course aims to teach medical and graduate students how the explosion of knowledge about genetics could play out in medical practice. As part of the class, students had the opportunity to get their own genotyping results to help them understand, through personal experience, the myriad ethical, clinical, social, legal and other challenges that patients across the country will face in the near future.

“This information is coming,” cardiologist Euan Ashley, MD, PhD, said in a July presentation to the students. “We need to learn how to apply it for the benefit of our population.”

The class comes at a time of extremely rapid progress in genetics research. Every week brings to light some new discovery of a gene or snippet of DNA related to a particular disease or trait. Meanwhile, the cost of genomic testing is plummeting. In 2003, it took more than $400 million and a gargantuan effort to sequence the first complete human genome, the molecular blueprint of life. Just six years later, in 2009, Stanford bioengineering professor Stephen Quake, DPhil, managed to sequence his own genome for just $48,000, using a technology he invented. Within the next year or two, any individual will be able to secure his or her own genome for just a few thousand dollars, predicted Michael Snyder, MD, PhD, professor and chair of genetics.

“The day is not far away when, at the time you’re born or when you turn 18, it will be standard to have your genome sequenced,” said Keyan Salari, the sixth-year MD/PhD student who proposed and directed the course.

Already, consumer genomics companies charge less than $1,000 for the type of report Harris received. Instead of providing the full genome sequence, which contains billions of nucleotide subunits, these tests report on hundreds of thousands of SNPs, or single nucleotide polymorphisms — single changes in the DNA sequence that can have a meaningful impact on a person’s disease risk or trait.

Though this information thus far has had little impact on clinical practice, ultimately its effects could be far-reaching. Many predict that it will enable doctors to make the best choice of drugs and drug doses. It will also help determine what diagnostic tests patients should get and what lifestyle changes they could make to minimize their disease risks, said Ashley, who directs the Stanford Center for Inherited Cardiovascular Diseases. For instance, a person found to be at greater risk of diabetes could go for blood sugar screening, obtain drug therapy, and change his or her eating and exercise habits to reduce the impact of the disease, he said.

In addition, whole-genome sequencing will offer much greater power to detect the cause of genetic syndromes in families, said Ashley.

“The technology could revolutionize our approach to family screening in these cases,” he said.

Leslie Williamson
Medical student Keyan Salari, pictured here among genome sequencing machines, led a historic class on genomics in medicine.
Medical student Keyan Salari, pictured here among genome sequencing machines, led a historic class on genomics in medicine.

But the availability of genetic information also has the potential to generate enormous anxiety, as people might learn they are at higher risk for some devastating disorders, such as Alzheimer’s disease, over which they have little or no control.

While this avalanche of data is due to arrive shortly, few practitioners are equipped to help patients interpret the massively complex results and cope with what could be disquieting news. Currently, there are only about 3,000 genetic counselors in the United States and 1,100 medical geneticists, far too few to accommodate the enormous demand, said Kelly Ormond, director of the genetic counseling training program at Stanford and associate professor of genetics.

“Everyone, no matter what area you work in, is going to need a higher level of knowledge,” Ormond told students in explaining the need for the course.

While no one disputed the value of the class, some faculty members were troubled by the prospect of offering students personal genotyping. Critics said it could be viewed as coercion on the part of the school, as students would feel pressured to submit to testing to please the course instructors. And some questioned how the school could subject students to a test that might deliver devastating results.

“I was alarmed by this possibility, frankly,” Michael Greicius, MD, a professor of neurology, told the students.

When several faculty leaders first planned in 2009 to offer the genotyping to incoming medical students, as well as some graduate students and internal medicine residents, Pizzo put it on hold until the school could assess the risks and liabilities of the genetic testing.

“I fully recognize that genotyping will be an important part of our future, but also that the current technologies are limited and potentially problematic for those who are less than fully informed,” Pizzo said.


A 27-member task force — including clinicians, basic scientists, ethicists, genetic counselors, lawyers and students — debated the issue for a year, ultimately approving the course proposed by Salari, with some modifications. For instance, the course was offered as an elective in which student genotyping was optional. Students could choose one of two commercially available tests offered by 23andMe or Navigenics, which report on SNPs. Students would pay $99 for the testing — enough to prove their commitment but not enough to break their budgets, the task force decided. Stanford would pay the $300 balance for each student.

“The day is not far away when, at the time you’re born or when you turn 18, it will be standard to have your genome sequenced.”

In classroom exercises, students could use their own data or publicly available genome data, and instructors would not know which set they had chosen. The school also offered free counseling to students by the companies’ genetic counselors or members of the psychiatry faculty who volunteered their services. Test results were kept confidential, and instructors would never know which students came forward for testing.

Pizzo said the debate over the class reflects a field in its infancy. “There will be lots of opportunities for continued dialogue as the sophistication of the technology improves and the costs fall. It is our obligation to educate and train students to be future leaders. It is gratifying to see students behaving as leaders in the present — which certainly speaks well for the future.”


The disagreements about the course were freely discussed in classroom sessions. On the day before students were to decide whether to be tested, Greicius and biomedical ethicist Hank Greely, JD, presented some powerful arguments for why they thought students should opt out.

Greicius, who studies the genetics of neurological disease, warned the students against getting tested for the ApoE4 gene variant, which puts carriers at much greater risk of early Alzheimer’s and is among the variants tested by some personal genotyping companies.

“If you have a healthy 20-year-old without a clinical problem, I don’t think (testing for the gene) makes sense,” he said. “In fact, it does more harm than good, especially when you have 40 years to worry about it. Every time you get a B+ on a test, you will be thinking, ‘It’s early Alzheimer’s.’”

Greicius also pointed out that while this is among the most informative Alzheimer’s test offered, it is still fraught with uncertainty: ApoE4 carriers can live long lives and never get the disease, while a large proportion of Alzheimer’s disease patients do not have the ApoE4 variant at all.

Greely, director of the Center for the Law and Biosciences, assumed a lawyerly stance and with his booming voice presented his arguments point by point as if he were in a court of law. He said he never objected to the class, just the genotyping of students, who could obsess about the results or be misled by them. For instance, a woman who tests negative for the BRCA1 or BRCA2 gene mutations, which carry a greatly increased risk of breast cancer, might forgo cancer screening in the mistaken belief that she had no risk of the disease.

He said he also worried about the lack of government regulation over these direct-to-consumer tests and the fact that Stanford’s use of the tests could amount to a tacit endorsement of them (though class organizers issued a disclaimer to the contrary).

“I think that is a bad message to send because, from the medical side, I don’t think there’s anything valuable here,” Greely told the class. He also argued that the technology employed by the current genotyping tests soon will be replaced by whole-genome sequencing, making them obsolete by the time students begin their medical practices.

Stuart Kim, PhD, professor of developmental biology and of genetics and the course’s faculty sponsor, was visibly tense during the presentation and said later that he sharply disagreed with those views. Similar debates in the task force had left faculty members at odds. Kim said he believes that while the current tests don’t supply the range of data available in a fully sequenced genome, they are still quite useful.

“You don’t wait until it’s perfect before you learn it,” he said. “You learn what you can. You learn the principles of how human genetics works, so that you can best use new genetic discoveries as soon as they are made.”

Kim argued on the side of knowing — that knowledge can be power. For instance, a woman who tests positive for a mutation in the BRCA1 gene could have more frequent screenings and catch the cancer early, when it can be effectively removed before it spreads to the lymph nodes, he said.

“I think it’s worth it for women to have that information,” said Kim, who had his own genotyping done two years ago. “I would take life over ignorance.”

He said in some instances, there is a clear value in having genotype data in hand, the best example being the case of the frequently prescribed drug warfarin, a blood thinner that helps prevent clots. Without genetic information, the drug can be hard to calibrate. Too much can cause hemorrhaging and death, explained Russ Altman, MD, PhD, a professor of bioengineering, medicine and genetics, who serves on a national advisory panel that is developing physician guidelines for the drug.

“You can’t tell the dose by looking at a patient because genetics plays a huge part,” said Altman, who disclosed to the students that he serves on the scientific advisory board of 23andMe.

Altman led a class exercise using genotype data to get each student’s dose right.

Students engaged in a number of other in-class exercises, including one in which they used genetic data to calculate their chances of living to age 100. Sam Pearlman, a bioinformatics PhD student, and two others in the class created a software tool for the exercise, based on a study published in the July 2 issue of Science Express. The controversial study by Boston scientists identified 150 SNPs associated with longevity and used those genes to build a model for predicting a person’s “aging signature.”

The class winner of the longevity lottery was Kathy Wei, a PhD student in bioengineering who was found to have a 99.8 percent chance of reaching 100. Wei said she was excited though surprised by the result, as nothing in her family history suggests she has such hardy genes.

Kim, who gave her a prize of chocolates and clam chowder (clams are among the longest-living animals in the world), cautioned the students not to put too much stock in the result, as many questions remain about the validity of the study, which used data from Caucasians only. Nonetheless, this is where genetics is headed, he said.

“Whether it’s aging, diabetes or heart disease, this kind of genetic signature is part of the future,” he said.

Ultimately, 33 students in the class opted for the genotyping test, with varying levels of angst. Alexander Morgan, a PhD student in bioinformatics and pediatrics, said he didn’t hesitate to get tested as he thought he could cope with any results that might ensue. “If I had some horrible news, I think I would not consider it a death sentence,” he said. “People do recover emotionally from bad news.”

Pearlman, on the other hand, said he agonized over the decision because of a family history of Parkinson’s.

“Finding out I have this variant would not change anything — it will just make me worry,” said Pearlman, who is considering a career in genetic counseling. On the other hand, “Being a scientist, I probably would want to know because I feel more information is good information, even though at this point I can’t do much about it.” He is still waiting for his results.

Krystal St. Julien, a PhD student in biochemistry, said she decided not to go for the test because she was afraid she would obsess over the data.

“I tend to freak out about the most random things, and I don’t know what I’d be OK with or not OK with,” she said. However, the class has taught her what genotyping has to offer, so she has become more interested in having it done, she said.

Students say the class provoked them to think about many heady issues affecting themselves, their families and the future of medical practice.

Pearlman said the class just reinforced the pitfalls of making genetic data available, as it is poorly regulated, open to widespread confusion and misinterpretation, and constantly changing.

“Scientists like to view information as pure, but when it comes to health, it gets to be a quagmire of ethics, politics and regulation,” he said. “Will it be helpful or destructive?”

St. Julien noted that the class only scratched the surface of what’s to come, as human genome sequencing will open a Pandora’s box of potential problems.

“When you do sequencing, you can pinpoint exactly where something is going wrong,” she said. “I think it’s going to be very hard for physicians and scientists to explain everything to everybody.”

“It does more harm than good, especially when you have 40 years to worry about it. Every time you get a B+ on a test, you will be thinking, ‘It’s early Alzheimer’s.’”

But the class is still a good beginning, she said. “I think it’s a good start to figure out the logistics of how people should study personalized medicine and the medical implications. It’s good to build a community of people who can express to patients what it all means.”

As for Harris, the class gave him a whole new perspective on the field and the limitations of genetic testing. “Having undergone the kind of testing and knowing the anxiety it can provoke and the meaning of the results and the limitations — that is an experience that is useful for physicians,” he said. “Now you’ll understand what kinds of stresses patients will have. I think it’s a unique way for physicians to empathize with patients because you went through it firsthand.” His Parkinson’s file meanwhile remains unopened.


E-mail Ruthann Richter





©2010 Stanford University  |  Terms of Use  |  About Us