Tag Archives: research

Newborn screening collides with privacy fears

The wrinkled heel of nearly every baby in the United States is pricked at birth, and a few drops of blood are dabbed on filter paper and shipped off for analysis. Started in the 1960s, this newborn screening program tests for more than 30 rare and serious diseases that are treatable if caught early in life. Now, many public health experts who help run or advise the program are worried what the future holds. A new law shaped by a coalition of privacy advocates and conservative politicians requires consent for federally funded research on newborn blood spots, which include DNA but no names. Seeking consent sounds innocuous, even welcome. But experts are concerned that the law, which took effect in March, could hamstring not just fundamental research but also the kind of studies that routinely improve screening. Efforts to improve newborn testing often require studies on hundreds of thousands of stored blood samples; seeking consent for each one would be prohibitive and impractical. When California researchers sought informed consent to test a cuttingedge screening technology on blood spots from 400,000 newborns, for example, overworked hospital staff did not contact nearly half of eligible families, hampering the study. “Do you want genetic privacy at the expense of everything else?” asks David Orren, chief legal counsel of the Minnesota Department of Health in St. Paul.

When it began lumbering through Congress, the Newborn Screening Saves Lives Reauthorization Act of 2014 was unremarkable; it simply updated an expiring 2007 law that provided federal support for state-run newborn screening programs. In early 2014, the bill passed in the Senate—unanimously, and “in about 30 seconds,” says Cynthia Pellegrini of the March of Dimes in Washington, D.C., who advocated for the bill. The controversy began a few days before the House of Representatives voted on the bill last June, when a nurse named Twila Brase, who runs the Citizens’ Council for Health Freedom, a nonprofit in St. Paul that presses for medical privacy, reached out to the office of Michele Bachmann, a tea party icon whose district included the northern suburbs of Minneapolis-St. Paul until she retired from Congress earlier this year. Brase, who also opposes the federal mandate for electronic health records and the Affordable Care Act, had been fighting storage and research on newborn blood spots for years. Brase’s contact had its desired effect: When the bill reached the House floor, Bachmann delivered an emotional speech. “This legislation presumes that every parent of every newborn in the United States of America pre-agrees that the government can have their baby’s blood sample, which contains their DNA code,” she said. “Americans should not see the death of privacy, especially of the most sensitive private information that every American can have.”

Bachmann’s speech came too late to affect the House vote. The bill passed. But because legislators had added some minor tweaks to the language before voting, the bill had to return to the Senate, so that the two chambers were passing identical text. That gave time for Bachmann’s qualms to catch the attention of members of the conservative Senate Steering Committee, including Rand Paul (R–KY) and Patrick Toomey (R–PA). They sought input from her, as well as from officials from the March of Dimes, the National Institutes of Health (NIH), and other research and advocacy groups. After much discussion, the senators settled on the clause mandating informed consent when newborn blood spots were used in federally funded research. It passed both chambers and was signed into law by President Barack Obama a week before Christmas. At the crux of scientists’ and public health advocates’ concerns is what fits under the umbrella of “research,” which federal regulations define as investigations that “develop or contribute to generalizable knowledge.” Does testing a new screening technology qualify as research? What about studies of a test for a disease not currently on a screening panel, to determine whether it should be added? “There are public health functions that are mixed up with” what might be considered “pure” research, says Logan Spector, an epidemiologist at the University of Minnesota, Twin Cities. And some research that seems unrelated to newborn screening might not be: Probing leukemia’s origins, as researchers studying blood spots have done, could also represent nascent steps toward a test for leukemia risk.

Jeffrey Botkin, a pediatrician and bioethicist at the University of Utah in Salt Lake City, who is part of a federal advisory panel on newborn screening, worries about the impact of mandating informed consent. But he’s sympathetic to its appeal. “It’s good to be the subject of much more public dialogue and scrutiny,” Botkin says. Many acknowledge that screening programs could do a far better job of educating parents and doctors, ideally before a baby’s birth rather than in the distracted hours afterward. The Office for Human Research Protections is drafting guidelines on the law and plans to define what qualifies as research. In the meantime, scientists and state health departments are trying to anticipate the law’s effects. “We’ve essentially frozen” our repository, says Michael Watson, the executive director of the American College of Medical Genetics and Genomics in Bethesda, Maryland, which runs a virtual bank of dried blood spots. None of the four participating states plans to provide information from blood spots collected after March, when the law took effect. A pilot study to develop a test for detecting Duchenne muscular dystrophy “has been slowed down tremendously,” Watson says. There’s also a big question about whether the law is an early jolt of a larger seismic shift in how deidentified samples are handled. Until now, studying such samples, which carry no names or addresses and are not linked to an individual’s health records, hasn’t required informed consent. But in January, NIH began expecting grantees on genomic research to seek consent before using deidentified samples. The newborn screening law is turning that recommendation into a national requirement, at least for blood spots. (A handful of states already mandate consent.) Other samples, like tumor tissue or deidentified blood samples from adults, could be next. The Department of Health and Human Services is rewriting its “Common Rule” governing human subject research. An upcoming draft will reveal whether it wants consent for all deidentified samples. Once those regulations are finalized, perhaps within a couple of years, the newborn screening requirement for consent will be subsumed by the Common Rule. The Common Rule is Brase’s next frontier. She plans to comment on the proposed draft rules when they’re released, to urge that all deidentified samples be subject to informed consent before scientists can access them. “When researchers decide we’re theirs, that sets people up to oppose what’s happening in research,” she says.

Jennifer Couzin-Frankel, Science Magazine

Ancestry.Com Is Quietly Transforming Itself Into A Medical Research Juggernaut

In 1984, a genealogy geek named John Sittner published The Source, a book meant to unearth and analyze never-before-seen records that genealogists could use to put together family histories with unprecedented detail. Several years later, he founded Ancestry magazine to teach people how they could use public archives and technology — which, back then, meant CD-ROMS and primitive websites and search engines — to build out their family trees.

Sittner sold the company long ago, but three decades after it began, Ancestry.com — the $1.6 billion Internet company that his magazine evolved into — is poised to become one of the most unlikely, yet powerful, scientific tools in the world. For about three years, it’s been collecting and analyzing genetic information through a service called AncestryDNA, and in the process, quietly asking consumers if they’d be willing to share their data with Ancestry for research. To date, it’s banked more than 800,000 samples from customers all over the world, rivaling the database of Google-backed genetics-analysis company 23andMe, which boasts about 900,000 samples. And now, armed with mountains of health data, Ancestry.com is slowly transforming itself from a retiree’s hobby into a medical research juggernaut.

“We actually do think that health is a pretty natural extension of the core mission to help everyone discover, preserve and share their family history,” Ancestry.com CEO Tim Sullivan told me earlier this week, during a visit to the company’s San Francisco offices. “We’re exploring ways that we could participate in health and provide our users with health insights, for sure….ways that we could leverage the data we’ve aggregated to support research efforts, similar to what 23andMe has done with Genentech and others.”

Long before Ancestry.com got into the DNA game, it had ties to the Mormon church. Its owners were two Brigham Young University grads who had made their fortune selling Latter-day Saints publications on floppy disks. Access to Ancestry.com was free at LDS Family History Centers, and recently the company signed a deal with the church’s genealogy non-profit, FamilySearch.org.

Ancestry.com’s huge advantage over services like 23andMe is its age; since it has been collecting ancestral data about its users for decades, it knows health information not just about its users, but about their great-grandparents and great-great-grandparents. That information, coupled with surveys and modern-day genetic testing, can inform users of any hereditary conditions that run in their family, and help them project health problems in their future. Recently, the company has been testing a “family health history experience,” which will eventually help people use their family trees to aggregate family health history from their living family members.

“Our records give us a lot of family health history. It’s super interesting. Your family health history is what your doctor always asks about. It’s extremely informative for your future health,” said Kenny Freestone, the product director for AncestryDNA. “The actual genetic markers and data also are informative, but they’re one piece of the data. We are really interested in a holistic approach.”

As Ancestry.com pivots into medical research, it would be wise to learn from the example of 23andMe, which has spent much of its life tangled up with federal regulators. 23andMe is Silicon Valley’s biotech darling — a sexy, headline-grabbing company that was co-founded by Anne Wojcicki, a biologist who married Google co-founder Sergey Brin—but its reception by the government has been less glowing. After being told that it wasn’t allowed to market its spit-in-a-vial genetic test as a medical diagnostic, 23andMe went ahead and did it anyway. Last year, the U.S. Food and Drug Administration sent the company a cease and desist letter, essentially shutting down its direct-to-consumer genetics arm domestically. In February, the company got the FDA’s green light to sell consumers its genetic test for Bloom Syndrome, a rare genetic condition.

Outside the consumer realm, though, 23andMe has had some victories. The company’s massive database landed it some megadeals with pharmaceutical giants Genentech and Pfizer earlier this year, and last month, it launched its own drug-discovery lab, 23andMe Therapeutics.

Ancestry.com has a chance to succeed on the same scale, but first, it will need to navigate some of the same pitfalls 23andMe did.

“We want to be the largest personal genomics company on the planet,” Sullivan said. “But we want to get it right.”

Already, Ancestry.com is marketing a spit-box test similar to 23andMe’s that analyzes a user’s DNA and spits out a detailed breakdown of ethnic heritage, including where the user’s her ancestors came from, and which other Ancestry.com users he or she might be related to. Since it doesn’t have approval to conduct medical diagnostic tests, Ancestry.com can’t tell users if their DNA indicates a higher risk for ovarian cancer or Alzheimer’s disease. But the company is getting ready to talk to the FDA about that, and hopes to win the approval that 23andMe didn’t.

As it aims for the consumer DNA testing market, Ancestry.com is continuing to gather massive, massive amounts of information about users and their families, as it has done since the early days. This week, the company unveiled Ancestor Discoveries, a product that automatically identifies ancestors going back to the 1700s, drawing on the company’s treasure trove of archival and crowdsourced historical documents, public records, and photographs. And the company has plans to expand to Mexico and Germany because the data generated in those markets will beef up the offerings in the U.S.

Like 23andMe, Ancestry.com eventually hopes to make money by selling anonymized data about its users to large pharmaceutical and biotech companies. Already, the site has has had some initial conversations with companies that “might value [its] data for purposes of research,” Sullivan said, though they haven’t struck any deals yet.

Those deals, if they happen, risk sparking privacy worries among Ancestry.com’s users, as they did when 23andMe began selling its data to the highest bidder. But Sullivan and Freestone are confident that even if some users grimace at the idea of their genetic information being sold to Genentech or Pfizer, the long-term benefits will make the discomfort worth it. After all, when you go to the doctor, one of the first things they ask for is a medical history of you and your relatives. If Ancestry.com is able to merge its collection of family trees with a large-scale DNA database, it would create a useful, generation-spanning fount of medical information, which could help millions of users see the health problems that await them and take preventative steps.

“As these networks grow, maybe there’s some really interesting leaps we can make,” Freestone said. “Here we’re leaping from genetic markers to potential ancestors. Can we leverage this sort of technology into future health predictions also?”

U.S. proposes effort to analyze DNA from 1 million people

The United States has proposed analyzing genetic information from more than 1 million American volunteers as part of a new initiative to understand human disease and develop medicines targeted to an individual’s genetic make-up.

At the heart of the “precision medicine” initiative, announced on Friday by President Barack Obama, is the creation of a pool of people – healthy and ill, men and women, old and young – who would be studied to learn how genetic variants affect health and disease.

Officials hope genetic data from several hundred thousand participants in ongoing genetic studies would be used and other volunteers recruited to reach the 1 million total.

“Precision medicine gives us one of the greatest opportunities for new medical breakthroughs we’ve ever seen,” Obama said, promising that it would “lay a foundation for a new era of life-saving discoveries.”

The near-term goal is to create more and better treatments for cancer, Dr. Francis Collins, director of the National Institutes of Health (NIH), told reporters on a conference call on Thursday. Longer term, he said, the project would provide information on how to individualize treatment for a range of diseases.

The initial focus on cancer, he said, reflects the lethality of the disease and the significant advances against cancer that precision medicine has already made, though more work is needed.

The president proposed $215 million in his 2016 budget for the initiative. Of that, $130 million would go to the NIH to fund the research cohort and $70 million to NIH’s National Cancer Institute to intensify efforts to identify molecular drivers of cancer and apply that knowledge to drug development.

A further $10 million would go to the Food and Drug Administration to develop databases on which to build an appropriate regulatory structure; $5 million would go to the Office of the National Coordinator for Health Information Technology to develop privacy standards and ensure the secure exchange of data.

The effort may raise alarm bells for privacy rights advocates who have questioned the government’s ability to guarantee that DNA information is kept anonymous.

Obama promised that “privacy will be built in from day one.”

SEQUENCING 1 MILLION GENOMES

The funding is not nearly enough to sequence 1 million genomes from scratch. Whole-genome sequencing, though plummeting in price, still costs about $1,000 per genome, Collins said, meaning this component alone would cost $1 billion.

Instead, he said, the national cohort would be assembled both from new volunteers interested in “an opportunity to take part in something historic,” and existing cohorts that are already linking genomic data to medical outcomes.

The most ambitious of these is the Million Veteran Program, launched in 2011 by the Department of Veterans Affairs. Aimed at making genomic discoveries and bringing personalized medicine to veterans, it has enrolled more than 300,000 veterans and determined DNA sequences of about 200,000.

The VA was a pioneer in electronic health records, which it will use to link the genotypes to vets’ medical histories.

Academic centers have, with NIH funding, also amassed thousands of genomes and linked them to the risk of disease and other health outcomes. The Electronic Medical Records and Genomics Network, announced by NIH in 2007, aims to combine DNA information on more than 300,000 people and look for connections to diseases as varied as autism, appendicitis, cataracts, diabetes and dementia.

In 2014, Regeneron Pharmaceuticals Inc launched a collaboration with Pennsylvania-based Geisinger Health System to sequence the DNA of 100,000 Geisinger patients and, using their anonymous medical records, look for correlations between genes and disease. The company is sequencing 50,000 samples per year, spokeswoman Hala Mirza said.

“NAIVE ASSUMPTION”

Perhaps the most audacious effort is by the non-profit Human Longevity Inc, headed by Craig Venter. In 2013 it launched a project to sequence 1 million genomes by 2020. Privately funded, it will be made available to pharmaceutical companies such as Roche Holding AG.

“We’re happy to work with them to help move the science,” Venter said in an interview, referring to the administration’s initiative.

But because of regulations surrounding medical privacy, he said, “we can’t just mingle databases. It sounds like a naive assumption” if the White House expects existing cohorts to merge into its 1 million-genomes project.

Venter raced the government-funded Human Genome Project to a draw in 2000, sequencing the entire human genome using private funding in less time than it took the public effort.

Collins conceded that mingling the databases would be a challenge but insisted it is doable.

“It is something that can be achieved but obviously there is a lot that needs to be done,” he said.

Collating, analyzing and applying the data to develop drugs will require changes to how products are reviewed and approved by health regulators.

Dr. Margaret Hamburg, the FDA’s commissioner, said precision medicine “presents a set of new issues for us at FDA.” The agency is discussing new ways to approach the review process for personalized medicines and tests, she added.

Toni Clarke and Sharon Begley, Reuters

23andMe’s New Formula: Patient Consent = $

Facebook generates about $8 a year in revenue from each of its users. But what if you offered a company not just your photos and updates, but your entire genome?

Then you could be worth as much as $20,000.

That’s my rough calculation for what Genentech could pay direct-to-consumer gene testing company 23andMe for the chance to trawl the DNA of each of several thousand of its customers for genetic clues to Parkinson’s disease.

The deal between the two companies, announced today, provides some fascinating insights into the evolving DNA business and the commercial prospects for 23andMe, a high-flying company that’s had some problems in the U.S. with regulators. According to detailed coverage over at Forbes, Genentech will pay as much as $60 million for access to 3,000 Parkinson’s patients in 23andMe’s database.

The backstory is that 23andMe pioneered direct-to-consumer genetic tests starting in 2006. It asked consumers to spit in a tube and send it in, and sent back a detailed summary of their risks for common diseases like macular degeneration. But then in 2013 the U.S. Food & Drug Administration banned the test out of concern that the information wasn’t accurate.

That put a big crimp in 23andMe’s business, but it didn’t end it. As Forbes points out, the real business here is mining this data:

Such deals, which make use of the database created by customers who have bought 23andMe’s DNA test kits and donated their genetic and health data for research, could be a far more significant opportunity than 23andMe’s primary business of selling the DNA kits to consumers. Since it was founded in 2006, 23andMe has collected data from 800,000 customers and it sells its tests for $99 each. That means this single deal with one large drug company could generate almost as much revenue as doubling 23andMe’s customer base.

The company hasn’t stopped gathering DNA data either. It still sells its Personal Genome Service health kits in countries like Canada. In the U.S. it continues to offer a more limited genealogy test to people who want to learn what their DNA says about their ancestry and relatives.

The result is that 23andMe may have the largest DNA database anywhere that’s open for medical studies. Of its 820,000 customers, the company says, about 600,000 have also agreed to donate their DNA data for research purposes. According to Forbes:

“I think that this illustrates how pharma companies are interested in the fact that we have a massive amount of information,” says Anne Wojcicki, 23andMe’s chief executive and co-founder. “We have a very engaged consumer population, and these people want to participate in research.”

It’s also a reminder that 23andMe’s real business isn’t selling $99 tests, but selling access to data that it has managed to crowdsource as cleverly as Facebook has gathered other personal details. To some observers, that’s pretty worrisome. In 2013, journalist Charles Seife, writing in Scientific American, called 23andMe intentions “terrifying.”

As the FDA frets about the accuracy of 23andMe’s tests, it is missing their true function, and consequently the agency has no clue about the real dangers they pose. The Personal Genome Service isn’t primarily intended to be a medical device. It is a mechanism meant to be a front end for a massive information-gathering operation against an unwitting public.

Seife’s worry is that the consents customers agree to when they donate their DNA could turn out to be meaningless. Once you are hooked, companies like Google and Facebook often change their privacy policies to expose more and more of your data. Why should DNA be any different?

So far, 23andMe seems more sincere than sinister. Parkinson’s disease is personally important to Wojcicki and her husband, Google founder Sergey Brin: his mother was diagnosed with the disease, which runs in families. Todd Sherer, the head of the Michael J. Fox Foundation for Parksinon’s Research, told me that the couple (now separated) has been the organization’s biggest donor, giving more than $150 million.

But Seife is right about the economics of DNA. It’s collecting free-and-clear data and amassing willing users that counts. According to the Fox Foundation, 23andMe actually gave its testing service away to Parkinson’s patients. That helped it assemble enough of them to create a useful resource it could sell to Genentech to start mining.

As part of its research, Genentech will gain access to the stored spit samples of 3,000 Parkinson’s patients in order to access their full genomic information. That is something that is allowed by the agreements customers signed. But to make sense of the DNA data, Genenetech will also need a lot of extra information about people’s health situations and medical records.

In this case, Forbes reports, the company will be reaching out to them to craft new agreements to access that, too.

Antonio Regalado, MIT Technology Review