Healthcare Professionals

  • Personalized Medicine Changes the Game for Clinical Practice

    Beyond the ‘one-size-fits-all’ approach to treatment

    Ever since the Human Genome Project produced the first complete map of a human genome in 2003, futurists have predicted that a doctor could use a patient’s personal blueprint to customize treatment. For a fast-growing number of children, that day has arrived.

    At Cincinnati Children’s, many children who receive kidney transplants and every child who suffers a cancer relapse or needs certain psychiatric medicines already benefits from what some call the “Holy Grail” of advanced healthcare.

    “For thousands of children, personalized medicine isn’t ‘just around the corner’ anymore, it’s here and it’s saving lives,” says Sander Vinks, PharmD, PhD, Director of Clinical Pharmacology.

    Customized chemotherapy makes treatment safer

    When initial rounds of chemotherapy or radiation no longer keep cancers in remission, children from all over the world come to our Cancer and Blood Diseases Institute (CBDI). Taking on complex cases has given CBDI faculty great experience in customized care. Now, advancing technology i taking things to a new level.

    “Systems biology – what we used to call bioinformatics -- is turning out to be at the center of all the advances we are making in cancer therapy,” says John Perentesis, MD, FAAP, CBDI Executive Co-Director. “We are using next-gen sequencing not only to gather genomic information about our patients but also to sequence the genes of the tumors themselves.”

    Perentesis is working with Vinks and Bruce Aronow, PhD, Director, Center for Genome Informatics, on a study exploring the cancer-fighting abilities of sirolimus, an mTOR inhibitor originally developed as an immune suppressor.

    The study uses a gene chip designed to analyze nearly 2000 variants in 225 genes known to influence metabolism to see how patients react to sirolimus. Eventually this work could help predict whether a specific tumor is likely to respond to a specific drug – and at what dose.

    “This is a unique and powerful foray into personalized medicine,” Aronow says. “Nobody else has really nailed this issue of differential drug toxicity and optimization.” To destroy cancerous tumors, doctors
    typically give patients the highest chemotherapy doses they can tolerate. But that approach has its shortfalls.

    “We have drug regimens for Hodgkin’s disease that offer a 97 percent cure rate – without a bone marrow transplant – even for children with stage IV disease,” Perentesis says. “However, the chances of growing up to develop a life-threatening side effect such as heart disease or lung damage is much higher than the 3 percent risk of the drug failing to kill the cancer.”

    At least five multi-drug cocktails can cure Hodgkin’s disease, Perentesis says, but each poses risks. The promise of personalized medicine is that doctors would no longer have to guess at which patients should use which regimens.

    “If we can spot those patients most likely to have a side effect, we can use a lower dose or switch drugs while still achieving a therapeutic result,” Perentesis says. Between the sirolimus study and other projects, the CBDI has sequenced tumors for more than 100 children with relapsed cancers in the past year. 

    Better drug level monitoring to prevent organ rejection

    Obtaining quick, accurate test results to confirm that a drug is working as intended is another crucial part of personalized medicine.

    Vinks is working with a team of clinicians and researchers to develop a webbased decision support tool to help doctors track whether kidney transplant patients are getting ideal doses of the anti-rejection drug mycophenolate mofetil (CellCept).

    “The therapeutic window for this medication is quite narrow. If the levels stay too low, the transplanted organ can be rejected. If they go too high, the patient can suffer side effects,” Vinks says. “But the dose required to stay within that narrow window can vary widely between individuals."

    Cincinnati Children’s pathology lab already conducts drug-level tests for transplant recipients and a few other serious conditions to determine if children are fast, normal or slow metabolizers. But the process is complicated, expensive and not widely available.

    “We want to do all of this in a much more automated fashion, and present the information in an easy-to-use way,” Vinks says.

    The nephrology team plans to begin evaluating a beta version of the decision support tool this fall. If successful, similar tools could be developed for infectious diseases, chronic pain control, cystic fibrosis, lupus and other conditions.

    “So far, our doctors have loved the initial prototype,” says David Hooper, MD, a pediatric nephrologist in the Division of Nephrology and Hypertension. “It logically organizes everything the physician needs to know in a single location and in a way that facilitates clinical decision making. Having this form available has significantly reduced the time needed to plan for medical visits, and has increased efficiency in the clinic.”

    The tool uses color coding to alert doctors to issues of concern, recommends “suggested actions” and allows users to give feedback about the suggestions. In the past several months, more than 80 percent of suggested actions were followed, Hooper says. And over time, the care patients are receiving has become more predictable and requires fewer suggested actions to be made.

    “With this tool, doctors don’t have to rack their brains to remember their patients’ lab test schedules or other routine details. That gives them more time to focus on decisions only they can make at the bedside with their patients,” Hooper says.

    Powerful tests fuel predictive medicine

    In psychiatry, researchers have known for years that several frequently prescribed medications are affected by a few gene variations along a common metabolic pathway. These fairly-simple-to-detect variations can result in big differences in how children respond to treatment.

    Cincinnati Children’s established a genetic pharmacology service in 2004 to run gene tests that help set a child’s starting dose for these medications. More than 10,000 children since have received the “psychiatric panel” as a standard part of care. In 2006, a spin-off company called AssureRX Health Inc. was founded to produce a commercial version of the test.

    Since then, genetic testing for clinical and research purposes has leaped even farther forward.

    In July, the Molecular Genetics Laboratory at Cincinnati Children’s announced ExomeSeq, a whole-exome test developed to diagnose rare and complex conditions by scanning the important coding regions of 20,000 genes. Meanwhile, research scientists here are using even more powerful whole-genome sequencing techniques to hunt for the causes of disease and improved therapies.

    Although the information generated by these powerful tests is already transforming how medical care is delivered, adapting
    them to widespread clinical use will take some time.

    “The amount of data we can put together to analyze is staggering,” Aronow says. “But in terms of having widely available technology to interpret the results of all this high-level analysis, we’re still just scratching the surface.”

 
  • John Perentesis, MD, FAAP.
    John Perentesis, MD, FAAP, Executive Co-Director of the Cancer and Blood Diseases Institute.
  • Powerful, free tools accelerate research

    Anil Jegga, DVM, and Bruce Aronow, PhD, both in the Division of Bioinformatics, have developed a powerful set of systems biology research tools called “ToppGene Suite” and GATACA.

    The software tools – which are free for academic use – enable scientists to rapidly explore the relationships between diseases, drugs and the molecular pathways the affect.

    The software organizes massive libraries of data from about tens of thousands of genes in humans and in mice. Researchers can look up a wide range of “enriched” gene annotation information including disease- gene associations, drug-gene interactions, protein interactions, transcription factor binding sites, miRNA-target genes, corresponding mouse phenotypes and more.

    In addition to detailed lists of information, the software can produce color-coded relationship maps that allow users to visualize communities of diseases and related networks of genes.

    “This is very similar to what Amazon or Netflix does with their recommendation systems. Those consumer services use large collections of descriptors about movies and books to determine if you bought this product, you might be interested in these other products,” Jegga says. “It’s the same with genes. We collect 17 categories of information about gene functions and relationships, which in turn can point to other genes we didn’t think about, but actually may play a role in a particular disease.”

    These tools are increasingly used in combination with next-gen sequencing to accelerate disease gene prediction and drug discovery.

    “For example, if a whole genome or whole exome scan detects 200 genes that are down-regulated among people with a certain condition, then you can use ToppGene to rank them according to those most likely to have clinical importance,” Jegga says.

    Read more information about these tools. Additional information.