In an agreement announced this week, Genaissance will help Organon Pharmaceuticals discover genomic markers linked to efficacy and adverse responses during clinical trials of Organon's compounds for psychiatric disorders.
With Roche Diagnostics' CYP450 AmpliChip standing out as a notable counterexample — the company said it planned to market the test first in psychiatry — pharmacogenomic techniques have not yet taken hold in the treatment of behavioral disorders, and they have yet to really prove themselves in drug development. But that hasn't stopped Genaissance and others from giving it a try. Genaissance's deal with Organon, in which it will help investigate modulation of the hypothalamic-pituitary-adrenal axis in disease progression and drug mechanisms in several undisclosed diseases, is one of two efforts the firm has in the field, and it is joined by a handful of other players, such as Nanogen and PerkinElmer (see box, below).
Many neurological disorders and their treatments show a clear genetic component, but that fact, coupled with the existence of several partnering deals, don't necessarily mean that the time for clinical psychiatric pharmacogenomics has come.
To start with, not all approaches to the use of pharmacogenomics in psychiatry have the same advantages and disadvantages. "It depends what type of genes you are talking about — one that is clearly useful is the [CYP450] 2D6 and the 2C19," Jose De Leon, an associate professor of psychiatry at the University of Kentucky, told Pharmacogenomics Reporter. "For other genes, including the genes that psychiatrists are more interested in, including brain receptors, I don't see that happening soon." De Leon is conducting cost-effectiveness research on CYP450 testing in psychiatry.
"The problem with the genetic variations that are related to things happening in the brain — like receptors, transporters — [is that] we are really in the starting steps," said De Leon. "With these types of genes, it's not that I don't do research on them, it's that I don't see them having a clinical value in the next five years — I am 100 percent sure that it is not going to happen."
But despite the unproven nature of the field, Organon views the project as "an opportunity for genetics to provide some means for explaining differential response [to drugs], especially because there's not much to be gained from other means — as far as I know, at least — for explaining variable response," said Gé Ruigt, Organon's director of clinical research, and the company's representative in the US Food and Drug Administration's Pharmacogenomics Working Group. "It may also be that something needs to be improved in the proper diagnosis of these psychiatric disorders, especially because there's a lot of overlap between [mood disorders]," the main focus of reseach, he said. The company has been thinking about such a project for "a couple of years," he added.
The investigation by Genaissance and Organon into the genetic component of HPA-axis-related disease will most likely involve such mood disorders as general anxiety disorder, bipolar disorder, depression, or post-traumatic stress disorder. The project will necessarily involve trying to understand poorly understood molecular mechanisms. "In depressed patients, for example, your HPA indicators are out of whack, so you have higher levels of cortical steroid hormones" and other abnormalities, Krishnan Nandabalan, Genaissance vice president of corporate development, told Pharmacogenomics Reporter. "One way to treat it — at least the hypothesis — is that, if you set right those indicators, you can treat the disorder," he said.
Organon will use Genaissance's database of genetic variation, bioinformatics technology, and its genotyping services "both for research and in the clinic," said Nandabalan. Organon will examine the effects of an undisclosed number of compounds of a class that the firm said are able to modulate the HPA axis, relating genetic findings to disease state and drug response, and it will also likely stratify patients by genomics as well, he said.
In the process, Organon will gauge drug response through the use of standard psychiatric measures of response as well as biomarkers, such as protein and hormone levels, and will then relate the two with genetics. "What's not well understood is how to use [genetic] information in a clinical setting — that's exactly what Organon wants to do," said Nandabalan.
Peculiarities of PGx in Psychiatry
Ground-floor research projects of this type may predominate in psychiatric pharmacogenomics for a while. "The problem with the genetic variations that are related to things happening in the brain — like receptors, transporters — [is that] we are really in the starting steps," said De Leon. "With these types of genes, it's not that I don't do research on them, it's that I don't see them having a clinical value in the next five years — I am 100 percent sure that it is not going to happen." Knowledge about genetics related to the cardiovascular system and oncology far outstrips knowledge about the genetics of the brain, he said.
Nandabalan agrees. In his opinion, the earlier use of pharmacogenomics in "cardiovascular [fields] had to be purely a matter of convenience" — the endpoints are easier to measure and the studies are easier to conduct. "Oncology [pharmacogenomics] happened because it was really topical, patients are dying, and also, really selective classes of drugs are being developed in oncology, so it makes sense to have companion diagnostics in oncology," he said.
The fields of oncology and cardiovascular research also have advantages related to quantitative measurement, whereas in psychiatry, "you are trying to relate a biomarker to these very subjective measurements," Nandabalan said.
"In oncology, they have the advantage of [directly testing] the cells of the tumor — we cannot do that," said De Leon. Blood and its biomarkers are also easily accessed and well understood, including the genetics of response to drugs.
In the case of the D2 dopamine receptors that De Leon studies, for example, only two genetic variations are actually known to affect function, but their effect is not significant, and the accumulation of understanding is slow. "If you think that there are 50 genetic variations [affecting function] in the [CYP450] 2D6, it would be highly surprising that there are only two in the D2 receptor," De Leon said. It is very difficult to measure the effect of genetic variation on receptor function, and the in vitro studies employed to investigate the effect of genetic variation on brain function and drug response are not easily applicable to real patients, he said. In a test of CYP450 genes, on the other hand, a simple urine test will suffice.
In order to get a better understanding of the molecular goings on of the HPA axis, Organon will also be challenging the axis in volunteers and patients using endogenous substances that trigger responses from it. "We will be looking at changes in the axis as a consequence of those challenge tests," said Ruigt.
Depending on the indication for which a particular compound seems most well suited, and depending on the difficulties a promising compound may face during development, Ruigt estimates that it may take as long as six years to complete a study of a specific drug. Now that the agreement has been signed, the companies will begin meeting to discuss how to proceed, he said.
— Chris Womack ([email protected])
Other Recent Targeted-Medicine Efforts in Psychiatry
In addition to the deal with Organon, Genaissance is also trying to resuscitate a Merck KGa drug called vilazodone that was unable to pass Phase II tests due to low efficacy. The company plans to identify patients who will respond.
Last week, Perlegen entered into an agreement with an academic consortium to study bipolar disorder (see the Snippets section of this newsletter or follow this link).
In April, PerkinElmer and Predictive Diagnostics said they had discovered a panel of blood biomarkers that can distinguish Alzheimer's patients from controls. They hope to produce a clinical test that can diagnose the disease in its early stages. Other companies, such as Ciphergen, are also working on developing such a test, according to Pharmacogenomics Reporter sister publication ProteoMonitor.
In March, Nanogen said it had licensed genetic variations from Pathways Diagnostics that predict patients' response to antidepressants and antipsychotic in order to create a diagnostic product to gauge drug dosage. The product will test CYP450 genes and others.
In May 2004, Lynx Therapeutics signed a service contract with the US National Institute of Mental Health to analyze gene expression in a region of the brain cortex that has been implicated in schizophrenia.
In April 2004, a European Union-funded group began a three-year project called Gendep that uses pharmacogenomics technologies to identify genetic differences in patients with depression, in an effort to predict how individuals respond to certain antidepressants. The project involves 16 different medical schools, research centers, and companies in eight European countries.
Also in April 2004, 13 European research groups said they planned to observe and compare gene-expression profiles for mice, rats, and humans in a search for depression drug targets and disease causes.