By Monica Heger
This story was originally posted March 9.
In an example of how sequencing can be used to study not only the genetic underpinnings of disease but also the genetics of drug response, researchers at the HudsonAlpha Biotechnology Institute have teamed up with clinicians at the University of Alabama, Birmingham, to perform transcriptome sequencing on 200 brain tissue samples, from patients and healthy individuals, as well as whole-exome sequencing on the genomes of 100 Parkinson's patients enrolled in a clinical trial for the drug levodopa.
The goal of the project is two-pronged. In the RNA-seq study, the researchers hope to discover disease-causing genes. Exome sequencing of Parkinson's patients enrolled in the clinical trial, meantime, will be used to uncover a gene or genes responsible for eliciting an adverse reaction to the drug levodopa, which, if successful, could have immediate applications in the clinic.
"If we could identify a gene, or small set of genes, which determine whether you are at risk for side effects of levodopa, it would really change our [treatment] approach," David Standaert, a neurologist at the University of Alabama, told Clinical Sequencing News.
Levodopa, which is considered the most effective treatment for Parkinson's, often causes the severe side effect dyskinesia, a movement disorder characterized by wild, uncontrollable movements. About half of patients given the drug develop the disorder, which can be disabling, while other patients never develop it, despite long-term use of the drug. A genetic component is believed to be involved, and if that gene or genes can be uncovered it could help guide treatment.
The researchers will use whole-exome sequencing to study 100 patients — 50 who have developed an extreme form of dyskinesia, and 50 who have shown no side effects to the drug. The patients are all enrolled in a clinical trial at the University of Alabama.
Standaert said having relevant clinical data on each patient is a key component of the project because it allows them to select patients at opposite ends of the spectrum — both those who have developed the most severe forms of dyskinesia as well as those who never develop the disorder.
"While there are collections of DNA from Parkinson's patients, for the most part, the DNA gets disconnected from the patient and you don't have detailed information about that patient," Standaert said. "If we just started sequencing Parkinson's patients randomly and tried to correlate with drug effects, that would never work."
Identifying a genomic signature for patients with an adverse reaction to the drug could help not only those patients, but also patients who do not develop dyskinesia. Currently, clinicians are conservative in their initial dosages of levodopa and often exhaust other treatments before resorting to the drug because there is no way to tell how a patient will react. But, said Standaert, if the clinician could identify which patients were at low risk of developing the disorder, they could be started on the drug earlier in treatment and potentially at higher dosages.
"Right now, we're often very conservative and don't give [levodopa] out, but there are probably many patients who could take it safely," he said.
Mark Hallett, chief of the medical neurology branch at the National Institute of Neurological Disorders and Stroke said that understanding the genetic basis of levodopa response could have immediate impacts on treatment.
"Where dyskinesias come from is not at all clear," he said. "If there is a genetic difference that could be identified, that would be very helpful."
For instance, he said he recently saw a patient who is starting to show symptoms of Parkinson's. The patient was reluctant to start levodopa treatment right away because of concerns about the side effects. However, if there were a genetic test that could determine the patient's risk of developing dyskinesia, that would likely influence treatment decision, said Hallett.
"It could influence how soon you start levodopa, or whether to use an alternative medication. It might influence therapy from the get go," he said.
Standaert is in the midst of recruiting patients and gaining regulatory approval for the study. He said that patients will have to sign a consent form indicating that they understand that because exome sequencing is not proven to have value in clinical care, the results will not be returned.
"It's a very slippery slope when you start to give patients information that may or may not be correct," Standaert said. "That kind of information can be misleading in clinical care."
He added, however, that he has been working with the institutional review board to evaluate how best to deal with the exome sequencing results, particularly since there is very little precedent on how to incorporate sequencing into clinical trials.
Rick Myers' team at the HudsonAlpha Institute will do the exome sequencing, which Myers said he expects will begin toward the middle of the year. Myers' team will also perform the transcriptome sequencing of the 200 brain tissue samples, which are being provided by neurologist John Trojanowski at the University of Pennsylvania. Myers said that most of the sequencing for both projects will be done on the Illumina HiSeq instrument, but added that they might do some microRNA sequencing on Life Technologies' SOLiD. Aside from the HiSeq and SOLiD, the institute also operates Roche's 454 GS FLX platform and Ion Torrent's PGM.
Myers said his team has already begun the transcriptome sequencing portion of the project, the goal of which is to identify disease genes. While Parkinson's is believed to have genetic roots, so far only a handful of genes have been linked to the disease, and they collectively account for only about 10 percent of all cases, said Standaert.
Ideally, Myers said, the team would discover biomarkers that would allow for early diagnosis of the disease, as well as a way to predict what form the disease would take in terms of age of onset, severity, and drug response. In addition, he said, those genes could then serve not only as diagnostic tools, but also potentially as new drug targets.
The project isfunded by a $500,000 donation from philanthropist John Jurenko, and Myers said that depending on costs, it could be expanded to include methylome sequencing, additional samples, or even whole-genome sequencing. The sequencing is expected to be completed by the end of 2012.
While it is still early days for combining sequencing technology with clinical trials, the project is not the first of its kind. Washington University researchers are sequencing breast cancer patients enrolled in an aromatase inhibitor drug trial, for instance (IS 8/10/2010), and both Standaert and Myers said they expect to see more trials like this in the future.
"We're already seeing dozens of examples of this," Myers said. "And, I think sequencing is going to be applied clinically more and more."
Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.