|
||||||||
|
INTERNATIONAL HEALTH American, Korean Experts Gauge Impact of Genomics on Medical PracticeThe maps of the human genome published last February in Science and Nature were less the first shots of a scientific revolution than a signal that global change is already at the doorstep. A symposium titled "Genomics and Proteomics: Impact on Medicine and Health," held on July 3 and 4 in Seoul, Korea, charted the inroads of this revolution in that country and the U.S.Hosted by Asan Medical Center, the event was the third biennial Asan–Harvard Medical International Symposium, which attracted nearly 800 doctors to hear more than a dozen American and Korean investigators. Taken together, the talks trumpeted a new order in biomedicine that is remaking research, medical education, clinical practice, biomedical technology, and public health. Seismic ScienceIn the keynote lecture, Joshua LaBaer, director of the Institute of Proteomics at HMS, echoed some of the insights of the Human Genome Project and outlined the progression from genomics to proteomics. The genome effort, according to Francis Collins, director of the National Human Genome Research Institute at the NIH, was launched primarily "to understand the hereditary contributions to virtually every disease." Yet even before the first drafts were announced in June 2000, it was clear that the scope of the growing data repository was broader than that—part biomedical cornucopia and part Pandora's box."Medical genetics will play a very important role in the future," LaBaer said, listing some of the tools that have arisen along with the genome project: single-nucleotide polymorphisms (SNPs), which may reveal an individual's susceptibility to disease; DNA microarrays, used in high-throughput genetic studies; and model organisms whose sequences can be compared with human homologues. An effort that will become increasingly important for these tools to work, he said, is collecting large samples of tissue and blood that are linked to detailed clinical histories, so scientists can understand which diseases are linked to which genes. This information would fuel drug discovery, an endeavor that has shifted into high gear through genomics. Most drug companies now use genomics to identify target proteins at the gene level, LaBaer said. Having a gene in hand means that without intervening steps, researchers can produce samples of the protein to determine its function. They can then screen small molecules for any interaction and, perhaps, find a cure for disease. "I think one of the things that people need to start thinking about in the coming decade is how we're going to develop high-throughput tools to do ... the elucidation of the function of these proteins," he explained. "And this is really where the area of proteomics comes in. ...The implication of proteomics is that it is the high-throughput study of proteins. It's not studying proteins one at a time but studying proteins a thousand at a time." LaBaer suggested that prediction and prevention of disease will almost immediately be enhanced by the tide of genomics. He also singled out pharmacogenomics—the study of how genetic inheritance influences an individual's response to drugs—as an area of rising importance. Addressing some of the policy puzzles of this new science, LaBaer turned cautionary. Issues include genetic privacy, genetic determinism, the ethics of genetic research, the implications of genetic predictions, and the possibility of genetic customization. "And perhaps a little bit chilling is the idea that humans might at some point decide that they want to take control of human evolution," LaBaer said, "and I think these are issues we need to start thinking about now." Genetic SavvyPerhaps there is no better time to start thinking about medical genetics than in medical school—before practice begins. Bruce Korf, medical director of the Partners Center for Human Genetics at the Harvard Institutes of Medicine, illustrated the fundamental impact of genetics by opening his first talk with a quotation from Francis Collins asserting that at present, genetics can claim to be "the central basic science of medicine." But in clinical practice, Korf said, drawing on a 1998 study (S.J. Hayflick et al., Genetics in Medicine 1:13-21, 1998), utilization of genetic services is low, especially among internists. "So we're starting from a perspective where physicians in practice largely are unaware of even the relatively classical contributions of genetics, never mind the opportunities posed now by this revolution of knowledge ensuing from the Human Genome Project."Two primary forces—information technology and genetics—are now converging and creating "individualized medicine," Korf said. His approach toward educating students and health professionals so they can take advantage of this transformation is to divide the genetic focus between rare and common disorders. The overall objectives are for clinicians to recognize the rare disorders, knowing when to make referrals; and for common disorders like hypertension, diabetes, and cancer, clinicians should be able to identify opportunities for intervention. Gilbert Omenn, executive vice president for medical affairs at the University of Michigan, took up the case of cancer in describing a developing intervention based on genomics—that of molecular profiling. Using specimens from normal prostate tissue and tissue adjacent to cancer cells, two investigators at that university (A.M. Chinnaiyan and M.A. Rubin) devised a molecular classification of prostatic neoplasms using a DNA microarray. A variety of patterns of gene expression emerged as indicators for conditions ranging from normal tissue to localized cancer to metastatic cancer. One clear advancement for treatment, Omenn said, was the ability to identify a set of genes associated with cancers that would be responsive to hormone therapy. A profile of another cancer, hepatocellular carcinoma, was presented by Young-Ki Paik, director of the Yonsei Proteome Research Center at Yonsei University. He described his research using two-dimensional electrophoresis and matrix-assisted laser desorption ionization mass spectrometry. The techniques identify and evaluate molecular targets for diagnosis and treatment. Electrophoresis, for example, showed that several proteins, including glutamine synthetase, were either largely suppressed or completely undetectable in liver cancer, the second deadliest cancer for men in Korea. According to the studies, expression levels of the aldehyde dehydrogenase (ALDH) family vary markedly between normal and cancerous tissues. "What our research tells us," Paik said, "is tumor-ALDH3 and its variants may be directly linked to hepatocarcinogenesis." He and his colleagues are now investigating the biochemical mechanism of this association. Gene Therapy RevisitedOne of the gene-based interventions that is most controversial, particularly after the 1999 death of Jesse Gelsinger at the University of Pennsylvania, is gene therapy. Eleven years after it began, the approach remains a relatively new area of therapeutics, said Katherine High, the William H. Bennett professor of pediatrics at the University of Pennsylvania School of Medicine. "...At this point, there are no licensed gene therapy products."Despite setbacks, High said, there are many promising trials occurring at present, and she described three that illustrate the relative effectiveness of gene therapy for inherited vs. acquired disorders. The studies involved X-linked severe combined immunodeficiency disease, or SCID (Cavazzana-Calvo et al., Science 288:669, 2001); Leber congenital amaurosis, a form of childhood blindness (G.M. Acland et al., Nature Genetics 28:92-95, 2001); and Parkinson's disease (K.S. Bankiewicz et al., Experimental Neurology 164:2-14, 2000). High also presented some of her collaborations on gene transfer for hemophilia B, based on an adeno-associated viral vector. This AAV vector had been shown in the mid-'90s to allow long-term, adequate expression of the transgene, necessary for a lasting benefit in hemophilia. Working in mice, High and her colleagues injected into skeletal muscle a construct for blood coagulation factor IX. Circulating levels rose into the range of 250 to 350 ng/ml, corresponding to a level in humans of from 5 to 7 percent of normal. Expression lasted for about a year. "If you could recapitulate that result in humans," High said, "it would change their phenotype from severe to mild. So this is an exciting result." Extrapolating these findings from the mouse to dog, High and her colleagues saw very long-lasting expression of factor IX in the canine subjects though with lower circulating levels achieved—about 1 and a half percent of normal. The corresponding level in humans, however, would still be therapeutic. The researchers are currently conducting phase I/II clinical trials to determine the safety and efficacy of injecting the gene transfer construct into muscle. And they have just begun a clinical trial for transfer directly into the liver, which yields higher levels of circulating factor IX. Mining MeaningThe discipline that turns the Everest of genomic and proteomic data into a navigable stream is bioinformatics, the application of computational sciences in basic research. According to Ju Han Kim, HMS assistant professor of biomedical informatics at Children's Hospital, many of the challenges in genomics are similar to those in the computational sciences—string sequence homology, pattern recognition, structure prediction, and network analysis. Similarly, features of living organisms like structure, behavior, and development, which are based on arrays of amino acids and nucleic acid bases, also are informatical phenomena.Biomedical informatics, the convergence of bioinformatics and clinical informatics will radically transform our biomedical understanding forever, Kim asserted. As in the study mentioned by Gilbert Omenn (above), data from DNA chips may be clustered based on a variety of algorithms. "Functional clustering is currently the single most popular analysis," Kim said. The result is a distinct pattern in the complex data that gives insight into what certain genes are doing and when. Kim said that this primary analysis should be combined with information from biomedical literature and established databases to shed even greater light on the molecular mechanisms of disease. Another front in the genomics revolution—beyond the symbolic data of informatics—is the proliferation and distribution of actual biological samples. Joshua LaBaer and his colleagues at the HMS Institute of Proteomics are building a repository for all genes in human and model genomes to take up this challenge. Called FLEXGene (for full-length, expression-ready), the repository will contain an expression-ready clone for every gene. The different clones will be packaged in separate, addressable tubes, and they will be sequence-verified, affordable, and flexible enough to be expressed in various vectors. The Population PerspectiveLarge-scale initiatives are one of the hallmarks of the genomics era. Kyuyoung Song, associate professor of biochemistry at the University of Ulsan College of Medicine, described her 21st Century Frontier Project, which characterized 3,000 publicly available SNPs in the Korean population. Among those, she and her colleagues found 176 that were newly identified among Koreans, of which 156 were common. Song said that the comparative pooled DNA sequencing method they used proved to be effective and efficient, and the results suggest that ethnic and population-based differences should be considered in selecting SNPs for the study of complex diseases.Han-Wook Yoo, professor of pediatrics at the University of Ulsan College of Medicine, addressed his development of a mutation database for Korean patients with inherited metabolic disorders to clarify their genetic characteristics. Wilson disease, for example, is thought to be the most common such disorder among Koreans, caused by impaired copper transport into the hepatocyte secretory pathway. Yoo and his colleagues have identified 12 different mutations of the ATP7B gene in affected families. The most frequent mutation, R778L, is present in 38 percent of patients. Halfway around the world, in rural Pennsylvania, the Old Order Amish present an opportunity for research on a closed population. Leslie Biesecker, director of the Laboratory for Genetic Disease Research at the National Human Genome Research Institute of the NIH, speaking at the conference as an independent investigator, conducted a study that converted the Amish families' genealogical record books into a database. Custom-built software enabled the researchers to determine family pedigrees, which they linked to genes for six disorders having a much higher frequency among the Amish than in the outside population. Through positional cloning studies on one of these disorders, Biesecker and his colleagues identified the gene, whose associated protein belongs to the chaperonin family, involved in protein folding. The disease, McKusick–Kaufman syndrome, is characterized by multiple malformations. The researchers also discovered that sometimes a diagnosis of McKusick–Kaufman syndrome was changed as a child grew, and every revised diagnosis was the same: Bardet–Biedl syndrome, previously thought to be totally unrelated. Biesecker's group initially hypothesized an overlap between the genes that cause the two disorders. "What we concluded from this work is that McKusick–Kaufman syndrome and Bardet–Biedl syndrome are, in fact, allelic disorders," Biesecker said. The allele that causes McKusick–Kaufman syndrome appears to be the milder of the two. For the Public GoodJust as it is important to integrate knowledge from genomic research into medical education as Bruce Korf argued (above), it is necessary to make that information part of public health. Allan Noonan, special advisor to the U.S. Surgeon General, who spoke as an independent researcher, said that this effort is based on assessment, policy development, and assurance. "The first challenge to public health in addressing the integration of genomics ... is to look at how genetic tests ... should be utilized," he said. "We need to know the association of the different genotypes and gene variants with specific health outcomes."One government program that Noonan mentioned is the ACCE Project of the Centers for Disease Control; its abbreviation is short for analytical validity, clinical validity, clinical utility, and ELSI (ethical, legal, and social issues). Its overall aim is to develop a model system to evaluate the availability and usefulness of existing data on DNA-based tests and testing algorithms. "The goal of this effort is to facilitate the appropriate transition of genetic tests from the investigational setting to the clinical and public health settings," Noonan said. Addressing ethics in greater detail at the conference was Thomas Murray, president of the Hastings Center, founded in 1969 as the world's first research institute in bioethics. Murray became involved in the genome project even before it was funded, when he was invited in 1988 to speak to a Congressional panel on the project's potential ethical issues. "One of the greatest worries from the beginning was the possible use of genetic information for nonmedical purposes," Murray said. It is important to keep this information confidential so it is not misused for any purpose, but that does not mean it has to be distinguished from other health-related information. In fact, he said, it cannot be effectively separated out since most conditions influenced by genetics are not based on genetics alone. Murray also raised the issue of the fair distribution of the fruits of the genome project such as new technologies for screening, diagnosis, and therapy; advanced newborn screening; and pharmacogenomic services. Framing the TalksIn the symposium's opening address, Changgi Hong, president of Asan Medical Center, struck several cautionary chords as well, amid the general optimism he expressed for postgenomic medicine. Robert Crone, president and CEO of Harvard Medical International, and Je Geun Chi, president of the Korean Academy of Medical Sciences, also gave welcoming remarks. Crone acknowledged the October 2000 renewal of the agreement between Asan Medical Center and Harvard Medical International, a nonprofit subsidiary of Harvard Medical School, which extends the relationship in perpetuity. Harvard Medical International formed its original agreement with Asan in 1996, the first of HMI's international partnerships.The symposium was a collaboration of Asan, the affiliated University of Ulsan College of Medicine, Asan Institute for Life Sciences, Harvard Medical International, and Harvard Medical School, and it was supported by the Korean Academy of Medical Sciences. After a plenary address by Noonan, Mitchell Spellman, director of academic alliances and international exchange programs at Harvard Medical International, gave closing remarks. He outlined the individual advances and issues discussed by the speakers. But he drew from them a holistic take-home message for the audience, that genomics will cause physicians in the future "to subsume, more than our predecessors, the ennobling responsibilities to affirm our common humanity and assure for all equity of access to all that is available to advance health and prevent disease." —Robert Neal |
