NEW YORK (GenomeWeb) – Psychiatric pharmacogenomics startup CNSDose announced this week that it is developing a genomic test to predict psychiatric drug dose requirements specifically for Asian populations, in addition to the panel it is already developing primarily for Caucasian populations.
Ajeet Singh, the company's founder and CEO, told GenomeWeb this week that the Australian firm has also recently inked a new distribution partner, Atlanta-based Alpha Genomix, through which it plans to launch the latter test for patients in the US this fall.
Both of CNSDose's tests analyze mulitple genes involved in the metabolism of antidepressants in the liver, or in how readily medications cross into the brain via the blood-brain barrier.
In that study, they prospectively compared genetically guided antidepressant prescribing using CNSDose to traditional unguided prescribing in 148 Caucasian patients recruited in Victoria, Australia with a principle DSM-5 diagnosis of major depressive disorder. They found that subjects receiving genetically guided prescribing had an approximately 2.5-fold greater chance of remission than those dosed using traditional trial-and-error methods.
At the Royal Australian and New Zealand College of Psychiatrists (RANZCP) meeting in Hong Kong this week, University of Melbourne researchers who led an independent retrospective follow-up to this study shared some of their results.
In this study of 119 subjects, the researchers compared the eventual effective dose reached in patients after 10 weeks of standard trial and error prescribing with what the CNSDose report would have predicted. According to Singh, the sensitivity and specificity of CNSDose — in predicting the effective dose reached by gold standard trial-and-error prescribing — were both above 85 percent.
Singh said that CNSDose hopes to publish this study in full in a peer-reviewed journal later this year, as publishing clear clinical utility data for its tests is a central focus for the young company.
Not all psychiatric PGx companies have made robust data publication as much of a cornerstone, and even those that have still met with resistance from payors.
In a personal perspective piece published in The Lancet late last month, Chad Boseman — who helped lead the University of Melbourne- and University of Toronto-based replication study of CNSDose — highlighted the varying evidence for different available psychiatric genetic tests.
According to Boseman, as more and more of these tests enter the market, there remains considerable uncertainty among clinicians about their validity and usefulness. Based on his Lancet analysis, the only company other than CNSDose to have published a randomized controlled trial of its test is US-based Assurex for its GeneSight Psychotropic PGx panel.
Though it has not done an RCT, Boseman also cited Genomind for having published two supporting studies, one an open-label cohort study, and the other a cost-savings analysis.
And while Boseman claimed in his analysis that PGx firm Genelex lacks any published evidence for its psychiatric drug prescribing test, that company did also publish a cost effectiveness and decision impact study late last year showing that patients who received PGx testing had fewer hospitalizations and emergency room visits than patients in the comparison cohort. Using mean cost estimates, researchers led by the University of Utah's Diana Brixner estimated potential savings of $218 per patient who received PGx testing.
However, based on Boseman's data, more than a dozen other companies offer testing to guide psychiatric prescribing in the US and elsewhere for which they have published little or no utility data.
"The mere existence of these tools should not be mistaken for evidence of clinical usefulness. Our evaluation … suggests research on validity, reliability, clinical usefulness, and cost-effectiveness is needed for most these tools before universal adoption into clinical practice," he concluded.
If it can raise the required funds, Singh said that CNSDose would like to support a second, validation, in the form of a prospective RCT conducted by independent researchers, to bolster the results of its initial trial and the newer independent retrospective follow-up study.
Ideally he said, this trial would take place in Asia, and would serve to specifically validate the company's newly announced Asia-specific panel.
The difference between this new assay and the firm's current panel will be in the number of variants it covers in specific polymorphic genes. For example, Singh said, there are certain phase I and phase II hepatic metabolism enzymes for which there are variants that are rare in Western populations that have not been widely studied, but which may be much more important and predictive in Asian ethnicities.
"I can't go into details because of commercial confidences, but we feel there are certain variants that have been overlooked, and we are hoping to include those to increase the utility in an Asian setting," Singh said.
A second major thrust for CNSDose is inking deals with distribution partners that have their own sales engines to drive commercial adoption, something the company can't realistically fund on its own at this early stage.
In and around the Hong Kong RANZCP meeting, Singh said that CNSDose has been in early discussions with potential distribution partners, focusing on labs in Hong Kong and Singapore.
"Being a lean startup, our business model is to leverage labs with sales forces and customer acquisition infrastructure … [since] we are not in a position to have a field sales team yet."
The company's first agreement of this kind is with Alpha Genomix, Singh said, which plans to launch the CNSDose test for US patients this fall.
CNSDose is also available in Australia through two labs, Singh said. However, these labs don't have their own sales architectures, and so growth for the test in its country of origin may have to come on the heels of progress abroad.
As it moves forward in the US, CNSDose also plans to conduct an ongoing Phase IV-style trial in which it will collect and analyze commercially tested patient outcomes.
"The thing about depression is that people rarely have a pure depression," Singh said. "They often have additional problems like substance misuse or anxiety or trauma, so following a real-world sample and seeing the efficacy will be another type of data we can build without specific R&D spending."
"Then if the company begins to scale we can then work towards the second RCT we would like to do," he added.