There’s the toddler who took an antibiotic to treat an ear infection. Within days, his eyes swelled shut, and his skin peeled from his body in sheets. The pain was intense. His mother said he looked as though he’d been burned in a house fire. In another case, a boy given medication to control his epileptic seizures soon found himself on life support, his skin shedding “rather like a snake’s.” It’s not a problem limited to kids, either. Last year, after taking a commonly used antibiotic, 31-year-old Wayne Auzenne Jr. of Lake Charles, Louisiana, began burning, as his wife described it, from the inside out.
Stevens-Johnson syndrome (SJS) is a severe adverse skin reaction triggered in certain people after treatment with commonly used drugs, including the antibiotic amoxicillin, the anti-seizure drug carbamazepine, even over-the-counter ibuprofen and cold medicines. The condition typically begins with flu-like symptoms, quickly followed by a painful rash and blisters. Next, the skin begins detaching. If those peeling layers get severe enough that a person loses more than 30 percent of his or her skin, it’s called toxic epidermal necrolysis (TEN). Either way, it’s a medical emergency, often landing people in intensive care or burn units and in a medically induced coma. In about 10 percent of cases, people suffering from SJS die of the disease. For those with SJS/TEN, up to 50 percent die.
Fortunately, SJS/TEN is pretty uncommon in the U.S. The Mayo Clinic calls SJS both rare and unpredictable. Sometimes the precise cause isn’t ever found. But, as Surakameth Mahasirimongkol from the Ministry of Public Health (MOPH) in Nonthaburi, Thailand, well knows, SJS/TEN isn’t always so rare or so unpredictable as that. In an effort that many point to as a success story for genomic and precision medicine around the world, Thailand is showing that SJS can sometimes be prevented.
For a long time, Mahasirimongkol’s tropical home country of Thailand, with a population of about 67 million, was experiencing hundreds of SJS/TEN horror stories every year.
While little remarked upon, Thailand had one of the highest rates of SJS/TEN in the world. Strangely, it often seemed to occur in people taking carbamazepine for seizures or allopurinol for gout. Why? The reason for Thailand’s unusually high incidence of a rare syndrome is genetic: Thai people are more likely to carry a gene variant that makes their immune systems more prone to produce violent and potentially deadly reactions to drugs used to treat seizures, gout, infections, or pain.
And there are other factors making adverse drug reactions more frequent in Thailand, Mahasirimongkol points out. For one thing, for a drug to gain approval in Thailand, its safety isn’t tested first in Thai people.
“If a drug is approved in Japan, the U.S., or Europe, you can use the data to get approval in Thailand,” he explains. The so-called HLA-B*15:02 variant that leaves many Thai people susceptible to SJS/TEN rarely occurs in those of European or Japanese descent. As a result, when carbamazepine and other drugs known to spark SJS/TEN came into use in Thailand and other parts of Southeast Asia many years ago, no one knew that a genetic difference would put many Asian people taking those drugs at greater risk of suffering a severe adverse drug reaction. Even as that reality became clearer, abandoning carbamazepine for newer anti-seizure medications wasn’t an easy answer. That’s because the older drug is still safe and effective for the majority of people. And in comparison to other alternatives, carbamazepine is also cheap, which makes it particularly attractive to Thai patients, whose average monthly wage is less than $500.
Fortunately, in 2004, Thailand had a prime minister who was interested in new technologies. It soon became clear that the Thai people’s genetic misfortune presented an opportunity for the small country to serve as a global leader in the implementation of pharmacogenomics, the study of genetic differences influencing drug responses. If doctors in Thailand began screening their patients for the presence of the HLA-B*15:02 variant before prescribing particular drugs, they could protect those at greater risk from developing SJS/TEN by prescribing alternative drugs.
There were some obstacles to overcome at first. The test needed to be inexpensive and paid for by the Thai health system. Thailand met those requirements thanks in part to genetic testing labs in the country that already offered premarital screening for gene variants linked to thalassemia, a hereditary form of anemia that is common in Thailand. Those labs were readily equipped to begin offering pharmacogenomic testing too, meaning that no new infrastructure was needed.
Since 2009, doctors test for HLA-B* 15:02 before writing a prescription for Thai patients who need anti-seizure medications for the first time. Patients who’ve been tested are given a small purple card printed clearly with their test results to carry around in their wallets. “We used to think this disease was an unfortunate, unpredictable event,” Mahasirimongkol says. “After the genetic test was introduced, it became a predictable and avoidable adverse event. It changed practice.”
Mahasirimongkol says he thinks one reason the Thai example stands out is because “we’re still an upper middle-income country. A few years ago, we were just a middle-income country, and [yet] we’re already providing a pharmacogenomic testing service.” Teri Manolio, a physician, an epidemiologist, and the director of the Division of Genomic Medicine at the National Human Genome Research Institute (NHGRI) in Rockville, Maryland, took special notice of Thailand’s focused approach, at a meeting in Washington, D.C., in January 2014. “It captured our imaginations,” she says.
Capturing imaginations is exactly what the meeting was designed to do. Sponsored by NHGRI and the National Academy of Science, the meeting aimed to inspire new global collaborations and directions by sharing efforts in genomic medicine already under way around the world. It often feels as though “we live in a cocoon,” says Geoff Ginsburg, the director of Duke University’s Center for Applied Genomics and Precision Medicine in Durham, North Carolina, and Genome’s editor-at-large. “It’s not terribly surprising. We tend to be myopic about our [research] space, interacting mostly with colleagues in the U.S. and perhaps some counterparts in Europe.”
The meeting was an opportunity to widen the circle. It took place just as a polar vortex was delivering the most bitterly cold days Washington had experienced in decades. Many of the meeting’s participants had come from places where people don’t own sweaters. Despite the frigid weather, 90 leaders representing 25 countries on five continents showed up, paying their own way, for the sole purpose of relaying experiences and aspirations in genomic medicine.
As the participants explained in a perspective piece written after the meeting and published in Science Translational Medicine last year, the most common efforts around the world focus on cancer genomics and sequencing of large numbers of individuals representing various populations. For instance, there’s an effort in England to sequence the complete genomes of 70,000 people by the year 2017. (For more examples of global efforts in genomic and precision medicine, see “Taking Stock,” page 46.) The focus initially was on people with cancer and infectious and rare diseases. On the Genomics England website, you can meet a few participants. One of them is Arthur, a young, bespectacled redhead with vision problems related to a poorly understood form of albinism. The genomes in England’s 100,000 Genomes Project are being analyzed so the DNA data can be used to improve care for Arthur and others like him.
Belgium is building a national genome-sequencing pipeline too, as is the small country of Estonia. The Estonian Health Insurance Fund is covering the cost of sequencing the exomes (the 1.5 percent of the genome that codes for proteins) of 50,000 Estonians, with the goal of expanding to 500,000 of the country’s roughly one million citizens. DNA and tissue samples collected in the process are housed in the Estonian Biobank and are linked to the country’s electronic medical records system, accessible to individuals through a national identity smart card. A story about the effort last year in The Atlantic noted that “Estonia’s small size, sophisticated technological infrastructure, and relatively [homogeneous] populace make it an ideal place to put this ambitious idea to the test.”
The Estonian Human Genes Research Act, passed in 2000, has helped to garner public support for the effort. The law ensures the “voluntary nature of gene donation and the confidentiality of the identity of gene donors.” It protects participants from the potential misuse of their genetic data and from discrimination based on interpretation of their DNA sequences and any associated disease risks. As Andres Metspalu of the University of Tartu explained to his audience at the 2014 meeting in Washington, D.C., “by law, we have to use the genetic results to improve public health.”
asked why it’s important to build global awareness of genomic and precision medicine, Manolio suggests it is just common sense. She says that each country doesn’t need to do this individually: “It’s terribly inefficient; we all run into the same problems and come up with the same solutions. Why do that?” The U.S., she admits, doesn’t have the most nimble healthcare system. “Some argue we don’t have a ‘system’ at all.” In many smaller countries, with more centralized healthcare systems, it’s possible to implement new approaches to medicine and “try things we could never do,” Manolio says.
Ginsburg says there’s really no question that single payer health systems, in which governments pay for health care rather than many private insurance companies, have distinct advantages when it comes to the coordinated implementation and coverage of new approaches to health care. And if countries implementing these new approaches around the world can find ways to work together, it can also help in overcoming what the Science Translational Medicine authors identified as one of the biggest barriers: lack of evidence that genomic interventions are effective and worth their price tags.
“I think everybody, myself included, sees a huge opportunity if we can figure out how to organize,” Ginsburg says. “Building the evidence [for genomics in medicine] is something that is partly a numbers game. The more groups engaged [in it] — more providers, patients, and participants — the more likely it is that you can generate data on a scale that proves the value of genomic medicine.”
“The genome is so big,” Manolio says. “We need to build on [each other’s] efforts rather than duplicate them.”
to continue building on ongoing efforts and keep the international conversation on genomic medicine alive, participants at that first 2014 meeting have since launched the Global Genomic Medicine Collaborative (G2MC), described as an “action collaborative” among global leaders in the implementation of genomic medicine in clinical care. A second meeting has already taken place in Singapore; the next stop is Athens, Greece, in 2017. The goals of G2MC include community engagement and the development of a catalog of genomic medicine projects and programs that span the globe. The group also intends to serve as a forum for global policy on the integration of genomic data into medicine.
Robyn Ward, the deputy vice-chancellor of research at the University of Queensland, Australia, is co-chairing G2MC, along with Ginsburg. Ultimately, she says, what makes sense in one country, based on specific healthcare needs, healthcare systems, and other factors, will not always make sense in another. So far, it looks like genome sequencing can play an important role in the early diagnosis of rare genetic disorders in young children in Australia, just as it has been shown to do in other parts of the world. On the other hand, given the ethnic backgrounds represented in Australia, SJS/TEN isn’t the big issue there that it is in Southeast Asia.
Despite those differences, she says, global collaboration remains key. “The fact that everyone gets up at ridiculous hours of the morning and night to sit on teleconferences from across the world tells you there must be a perceived value-add,” she says. “I think there’s the sense of an almost overwhelming task ahead and a keenness to make sure that what we are doing is responsible, so we’re not ending up with inappropriate applications or uses of genomics in areas where we’re not going to deliver any value.”
Perhaps the most specific of the G2MC’s goals to come out of its first meeting is the global eradication of preventable SJS/TEN. In fact, Thailand isn’t the only country to have adopted pharmacogenomic screening to avoid it. Singapore has also found success in implementing genetic testing. The small country is home to 5.5 million people, with ancestries tracing primarily to China, Malaysia, and India. Singaporean Malays and Chinese often carry the HLA-B*15:02 gene variant as well, putting them at increased risk for SJS/TEN when starting carbamazepine.
Before pharmacogenomic screening could be put in place in Singapore, it had to be shown to be cost effective. An analysis found that it was, although the price was initially a bit high (about $300 U.S. dollars) and the turnaround a bit slow. Singaporean health officials found a way to batch the tests at a centralized facility; this cut the costs and the time in half. For the last three years, screening has been the norm.
It’s clearly working. “We haven’t had a case of SJS/TEN in new users of carbamazepine who have taken the test,” says Cynthia Sung, a faculty member at Duke-NUS Medical School and visiting consultant to the Health Sciences Authority in Singapore. She says that means there are likely about 50 people in Singapore right now whose doctors prescribed them an alternative anti-seizure medication after they tested positive for the HLA risk variant and “who don’t even know that they’ve dodged a bullet.”
Following on that success, Singapore is developing an active surveillance network for adverse drug reaction monitoring and discovery of predictive genomic biomarkers. Ongoing clinical trials are focusing on adverse events related to cholesterol-lowering statins and common chemotherapeutic drugs.
Ideas about the value of pharmacogenomics seem to be catching on in Southeast Asia. Mahasirimongkol is exploring genetic factors involved in other adverse drug reactions, including drug-induced hepatitis in people undergoing treatment for tuberculosis. He says Malaysia and Indonesia may begin testing people for HLA-B*15:02 this year. Pharmaco-genomic testing is also under consideration in Vietnam and the Philippines.
The Thai example led Manolio to recognize a gap in research efforts on severe adverse drug reactions such as SJS/TEN. It remains largely unclear what factors contribute to risk in people without Asian ancestry. (The Food and Drug Administration does recommend genetic testing for anyone of Asian ethnicity.) According to the Centers for Disease Control and Prevention, about 120,000 Americans are hospitalized each year for adverse drug events. To address gaps in understanding those reactions and identify potentially useful biomarkers for predicting those reactions, the National Institutes of Health has issued a call for research proposals.
“I’m enamored with the whole notion of a pharmacogenomics card,” like the one being offered in Thailand, that you could present to a pharmacist when filling a prescription, Ginsburg says. “It sounds simple in concept. [In reality] it’s more complex … but that’s the point. How do we take lessons from a successful implementation that has public health impact and replicate that in other regions of the globe?”