|
BACKGROUND/RATIONALE:
The clinical utility of genetic testing to guide medication selection and dosing decisions for antidepressants is an active area of research. Enzymes in the cytochrome P450 (CYP450) system, in particular 2D6 and 2C19, metabolize many psychotropic agents, including antidepressant medications. There is substantial polymorphism in the genes that code for these enzymes, with some alleles producing inactive or less active forms of the enzymes. The variation for 2D6, an important metabolic enzyme for psychotropic medications, is most pronounced. Research has shown that 7% of Caucasians, <2% of Asians, and 2-4% of African Americans are poor metabolizers for 2D6 (with both alleles coding for inactive enzyme). Poor metabolizers for a given antidepressant medication will have unusually high plasma levels after a single dose and are at increased risk for medication side effects. Also, a small proportion of the population are ultra-rapid metabolizers, for instance <2% of northern European and 10% of southern European populations, and may not respond to recommmended doses of these medications. Genetic testing could be used to personalize antidepressant medication treatment and, thereby, reduce the risk of side effects, increase medication adherence, and ultimately improve treatment outcomes.
OBJECTIVE(S):
Objective 1.
Incorporate probabilities for CYP450 polymorphisms, costs, and outcomes into an existing depression treatment decision analysis model.
Objective 2.
Estimate the cost-effectiveness of conducting genetic testing for CYP450 polymorphisms at initiation of antidepressant treatment and after first antidepressant treatment failure.
Objective 3.
Conduct sensitivity analyses for a range of probabilities for CYP450 polymorphisms, side effects, adherence outcomes, and genetic testing cost.
METHODS:
For the decision model, we chose fluoxetine as the first-line antidepressant treatment because it is commonly used, recently became generic, and is one of the least expensive antidepressants available to the VA. We will use four levels of metabolism for the 2D6 enzyme (poor, intermediate, extensive [normal], and ultra-rapid) and three levels of metabolism for the 2C19 enzyme (poor, intermediate, and extensive [normal]). We also have developed a matrix for the probabilities for each of the polymorphism combinations based on results from the STAR*D depression treatment trial. The matrix includes antidepressant options to maximize tolerability using the results of the genetic testing. The current cost of conducting genetic testing for 2D6 and 2C19 is $880 per patient. The outcomes will be measured in quality-adjusted life years (QALYs) from the literature based on level of treatment response.
The original deterministic model will be adapted to specifically model the costs and effects of genetic testing vs. usual care (no genetic testing). We also will test the effect of conducting genetic testing after the first failed antidepressant trial. We are using TreeAge Pro to build the decision model, which will incorporate the costs and prevalence estimates for the polymorphisms of the genetic tests. The modeled treatment algorithms will be tailored based on the results of the genetic tests in that arm of the model. Simulations of 10,000 hypothetical subjects will be used to assess the variability of costs and effects over the stochastic parameters in the model and a series of one way sensitivity analyses will identify key variables that influence the results. The sensitivity analyses will include a range of probabilities for CYP450 polymorphisms, side effects, adherence, outcomes, and genetic testing cost. A break-even analysis will be conducted to identify the price at which the genetic test will become cost effective. This will be important because the costs of genetic testing will most likely decrease over time, and this analysis can inform the specific price point in which the tests becomes cost effective.
FINDINGS/RESULTS:
No results at this time.
IMPACT:
Advances in genomic medicine--and the capability to personalize treatment by assessing genetic profiles--will lead to a new era of pharmacotherapy for mental illnesses and improve outcomes for veterans. Decision analysis models can inform future research directed at taking advantage of genetic testing.
PUBLICATIONS:
None at this time.
DRA:
Chronic Diseases, Health Services and Systems, Mental Illness
DRE:
Diagnosis and Prognosis, Technology Development and Assessment, Pathophysiology
Keywords:
Cost effectiveness, Depression, Genomics
MeSH Terms:
none
|
