Some patients are now getting DNA tests before starting antidepressants because roughly 60 percent of people with depression do not benefit from the first medication they try. These tests, known as pharmacogenomic or PGx tests, analyze specific genes that control how quickly a person’s body breaks down certain drugs. The idea is straightforward: rather than spending months cycling through medications that may not work or cause harsh side effects, a simple cheek swab can flag which antidepressants are more likely to help and which ones a patient should probably avoid. For someone caring for a loved one with dementia who also struggles with depression, that kind of shortcut matters — months lost to the wrong medication can accelerate cognitive decline and erode quality of life in ways that are difficult to reverse.
The science behind these tests has been building for years. The largest controlled study to date, the GUIDED trial published in 2019, enrolled 1,167 patients and found that people whose prescriptions were guided by genetic testing achieved remission at significantly higher rates than those treated the usual way. But the story is more complicated than a simple before-and-after. Testing is not recommended for everyone, insurance coverage is inconsistent, and some professional guidelines urge caution about using these panels as routine screening tools. This article walks through what pharmacogenomic testing actually measures, who stands to gain the most from it, what the clinical evidence shows, and the practical realities of cost and access that patients and caregivers should understand before requesting a test.
Table of Contents
- Why Do So Many Antidepressants Fail on the First Try?
- What Pharmacogenomic Tests Actually Measure — and What They Cannot Tell You
- What the Clinical Evidence Actually Shows
- Who Should Actually Get a Pharmacogenomic Test?
- The Messy Reality of Cost and Insurance Coverage
- What the Test Process Looks Like in Practice
- Where Pharmacogenomic Testing Is Headed
- Conclusion
- Frequently Asked Questions
Why Do So Many Antidepressants Fail on the First Try?
Major depressive disorder affects roughly one in five Americans during their lifetime, and medication remains the most common treatment. Yet the landmark STAR*D study — one of the largest real-world depression trials ever conducted — showed that only about one-third of patients achieve full remission on their initial antidepressant. The rest are left adjusting doses, switching medications, or stacking drugs on top of each other in a process clinicians sometimes call “trial and error.” Between 10 and 30 percent of patients with major depression ultimately fall into the category of treatment-resistant, meaning standard medications fail them entirely. The reason for these failures is partly genetic. Everyone metabolizes drugs at a different speed, and that speed is determined in large part by variations in liver enzymes encoded by genes like CYP2D6, CYP2C19, and CYP2B6. A person who metabolizes sertraline unusually fast, for example, may never reach a therapeutic blood level at standard doses. A person who metabolizes it unusually slowly may accumulate too much of the drug and suffer side effects that drive them to quit. Neither outcome means the diagnosis was wrong — it means the medication was a poor biochemical fit.
Traditional prescribing had no reliable way to know this in advance, so clinicians relied on clinical judgment, patient feedback, and patience. Pharmacogenomic testing is an attempt to replace some of that guesswork with data. Consider a 72-year-old woman recently diagnosed with both mild cognitive impairment and moderate depression. Her doctor starts her on citalopram. After six weeks she feels no better, and the sedation is worsening her memory complaints. She switches to venlafaxine. Two more months pass. Had a PGx test revealed she was a CYP2C19 ultra-rapid metabolizer, her physician might have chosen a different starting medication or adjusted the dose from day one, potentially saving her nearly four months of ineffective treatment — time that matters enormously when cognitive function is already on a downward slope.
What Pharmacogenomic Tests Actually Measure — and What They Cannot Tell You
Pharmacogenomic panels work by collecting a DNA sample, usually through a cheek swab, and analyzing specific gene variants that influence drug metabolism. The 2023 Clinical Pharmacogenetics Implementation Consortium guideline, which remains the current standard in 2026, covers 13 antidepressants including widely prescribed medications like citalopram, escitalopram, fluoxetine, paroxetine, sertraline, duloxetine, and venlafaxine. The genes with the strongest evidence are CYP2D6, CYP2C19, and CYP2B6. Based on a patient’s variants, the test classifies them as a poor, intermediate, normal, rapid, or ultra-rapid metabolizer for each enzyme, and the prescriber uses that information to choose medications and dosages that align with the patient’s metabolic profile. However, these tests have clear limits. The same CPIC guideline reviewed two other genes — SLC6A4 and HTR2A — that some commercial panels include, and concluded the evidence does not support their clinical use for antidepressant prescribing. This is an important distinction because some testing companies market broad gene panels without clearly separating which results are clinically actionable from which are speculative.
The FDA maintains a continuously updated Table of Pharmacogenomic Biomarkers in Drug Labeling, but the agency has been careful to note that inclusion on that list does not automatically mean testing is required before prescribing. In other words, a PGx test can tell you how fast your body processes certain drugs, but it cannot tell you which antidepressant will make you feel better. Depression is influenced by genetics, environment, life circumstances, and neurochemistry in ways that a metabolic panel does not capture. If your loved one has dementia-related depression, an additional layer of complexity exists. Many PGx panels were validated primarily in populations with major depressive disorder, not in elderly patients with comorbid neurodegenerative conditions. The metabolic information still applies — a slow metabolizer is a slow metabolizer regardless of the underlying diagnosis — but the clinical decision about which antidepressant to choose involves considerations beyond metabolism, including anticholinergic burden, fall risk, and drug interactions with dementia medications like donepezil or memantine. A PGx test is one input, not the whole answer.
What the Clinical Evidence Actually Shows
The strongest piece of evidence in favor of pharmacogenomic-guided prescribing comes from the GUIDED trial, published in 2019. This was the largest randomized controlled study of its kind, enrolling 1,167 patients who had already failed at least one antidepressant. Patients whose care was guided by PGx testing showed significantly higher response rates — 26.0 percent versus 19.9 percent — and higher remission rates — 15.3 percent versus 10.1 percent — compared to patients who received treatment as usual. The most striking finding was among patients who switched from a medication flagged as gene-incongruent to one classified as gene-congruent: this subgroup saw 33.5 percent symptom improvement versus 21.1 percent for those who stayed on incongruent drugs. Meta-analyses that pool data across multiple studies have found that PGx-guided patients were 41 to 78 percent more likely to achieve remission and 20 to 49 percent more likely to respond than patients receiving standard treatment.
A 2025 real-world study analyzing over 20,000 adults with major depressive disorder who underwent GeneSight testing showed tangible benefits in medication use patterns and healthcare resource utilization, lending the kind of large-scale, real-world support that randomized trials sometimes lack. That said, the evidence base is not uniform. Results vary depending on which testing panel was used, how “guided care” was defined, and whether the study was industry-funded. The GUIDED trial, for example, was sponsored by Myriad Genetics, the company that makes GeneSight. This does not invalidate the findings — industry funding is common across pharmaceutical research — but it does mean the results should be read alongside independent studies, which tend to show smaller effect sizes. For caregivers weighing whether to pursue testing, the honest summary is this: PGx-guided prescribing appears to help, particularly for people who have already failed at least one medication, but it is not a guarantee of finding the right drug on the next try.
Who Should Actually Get a Pharmacogenomic Test?
Professional guidelines are fairly specific about who benefits most from PGx testing. The International Society of Psychiatric Genetics recommends testing primarily for patients who have already experienced an inadequate response to a previous antidepressant or who suffered significant adverse reactions. This is not positioned as a routine first-line screening tool for everyone who walks in with new-onset depression. The reasoning is practical: routine preemptive testing for all patients is not considered cost-effective because only a relatively small proportion of the population has genetic phenotypes extreme enough to necessitate medication changes. For dementia caregivers, this recommendation maps fairly cleanly onto real life. If your family member was recently started on an SSRI and it seems to be working — even modestly — there is little reason to pursue testing.
But if they are on their second or third medication, or if they experienced unusual side effects on a standard dose, a PGx test could provide useful information. The February 2026 consensus from the American Society of Clinical Psychopharmacology went further, stating that responsible psychotropic deprescribing — the careful process of reducing or stopping medications — should explicitly consider pharmacogenomic data alongside pharmacokinetics, adherence history, and shared decision-making. This is particularly relevant in elderly patients who are often on multiple medications and where every unnecessary drug adds risk. The tradeoff is between the cost of testing and the cost of continued trial-and-error prescribing. For a patient who has been through two failed antidepressants — each requiring six to eight weeks to evaluate — that is three to four months of suffering, potential cognitive worsening, and caregiver stress. A PGx test takes days to return results. Whether the financial cost is justified depends on the individual situation, but the time cost of not testing can be substantial.
The Messy Reality of Cost and Insurance Coverage
The financial landscape for PGx testing is uneven and, in some cases, frustrating. GeneSight, the most widely used panel, lists a self-pay price of $330, and its manufacturer Myriad Genetics states that 98 percent of patients pay $330 or less out of pocket. Medicare Part B covers the test with zero out-of-pocket cost for beneficiaries, which is directly relevant for the older population most affected by dementia-related depression. For many elderly patients, this makes the test effectively free. Commercial insurance is a different story. As of January 1, 2025, UnitedHealthcare updated its policy to generally exclude coverage of multi-gene PGx panels like GeneSight for commercial plan members. Coverage varies significantly by insurer, and policies change frequently.
Other major panels on the market include Genomind’s Genecept Assay, Tempus, CNSDose, and Neuropharmagen — each with its own pricing structure and insurance relationships. Before ordering a test, patients or caregivers should call their insurance company and ask specifically whether pharmacogenomic testing with the intended panel is covered under their plan. Getting a prior authorization denial after the test has already been run is a common and avoidable problem. A broader concern is that the low adoption rate suggests systemic barriers beyond cost. A study of a large managed care cohort of over 438,000 patients found that fewer than one percent received a CYP2D6 or CYP2C19 test through insurance within a year of their depression diagnosis during the 2013 to 2018 period. Even accounting for the years that have passed since, adoption remains heavily underutilized relative to what the evidence supports. Part of this is physician awareness — many primary care doctors, who write the majority of antidepressant prescriptions, have limited training in pharmacogenomics and may not know to offer testing or how to interpret results.
What the Test Process Looks Like in Practice
The testing itself is simple. A clinician or patient orders a kit, which arrives by mail or is available in the office. The patient swabs the inside of their cheek, the sample is sent to a laboratory, and results typically come back within a few days. The report categorizes medications into groups — often color-coded as green, yellow, and red — indicating which drugs are expected to work normally, which may need dose adjustments, and which should be used with caution or avoided based on the patient’s genetic profile.
What matters more than the test itself is what happens next. The report is a tool for the prescribing clinician, not a prescription in itself. A medication flagged as red does not mean it is absolutely contraindicated — it means the patient’s metabolism may require a non-standard dose or closer monitoring. Conversely, a green-flagged medication can still fail for reasons unrelated to metabolism. For families navigating dementia care, the most productive approach is to bring the test results to the prescribing psychiatrist or geriatrician and have a conversation about how the genetic information fits into the broader clinical picture, including other medications, kidney and liver function, and the specific symptoms being treated.
Where Pharmacogenomic Testing Is Headed
The field is moving toward a model where genetic testing becomes a standard part of the medical record — done once and referenced throughout a patient’s life, much like a blood type. Several health systems are already piloting preemptive PGx testing programs, where patients are genotyped during routine care and the information is embedded in their electronic health record for future prescribing decisions. If these programs demonstrate cost savings and better outcomes at scale, the insurance coverage landscape could shift substantially.
For brain health specifically, the integration of pharmacogenomic data with other emerging tools — such as blood-based biomarkers for Alzheimer’s disease and digital monitoring of medication adherence — could eventually allow for genuinely personalized treatment plans that account for both the neurological disease and the psychiatric symptoms that so often accompany it. That future is not here yet, but the building blocks are being laid. In the meantime, for patients and caregivers dealing with depression that has not responded to the first medication, asking about a PGx test is a reasonable and evidence-supported step.
Conclusion
Pharmacogenomic testing is not a crystal ball, and it will not eliminate the complexity of treating depression — especially in patients who also have dementia or other neurodegenerative conditions. But the evidence, including the GUIDED trial’s results across 1,167 patients and meta-analyses showing significantly higher remission rates with guided prescribing, supports its use as a practical tool for reducing the months of trial-and-error that too many patients endure. The test is most valuable for people who have already tried an antidepressant without adequate results, and it is most accessible for Medicare beneficiaries, who can receive it at no cost.
For caregivers, the actionable takeaway is this: if your loved one’s antidepressant is not working after a reasonable trial period, ask their prescriber whether pharmacogenomic testing might help inform the next choice. Come prepared to discuss insurance coverage, and understand that the test results are one piece of a larger clinical puzzle. The goal is not to find a perfect medication — it is to narrow the field and reduce the time spent on drugs that were never a good biochemical match to begin with.
Frequently Asked Questions
How long does it take to get pharmacogenomic test results back?
Most commercial PGx tests return results within three to five business days after the lab receives the sample. The sample collection itself takes less than a minute — it is a simple cheek swab that can be done at home or in a clinic.
Does a pharmacogenomic test need to be repeated if my family member switches doctors?
No. Your DNA does not change over time, so the results from a single test remain valid indefinitely. The report can be shared with any new prescriber. Some health systems are beginning to embed PGx results directly into electronic health records so the information follows the patient.
Is pharmacogenomic testing covered by Medicare?
Yes. Medicare Part B covers GeneSight testing at zero out-of-pocket cost to the patient. Commercial insurance coverage varies significantly by insurer — as of January 2025, UnitedHealthcare generally does not cover multi-gene PGx panels for commercial plan members, so patients should verify their specific coverage before ordering a test.
Can a pharmacogenomic test tell me which antidepressant will definitely work?
No. The test identifies how your body metabolizes certain drugs, which helps rule out poor fits and adjust dosing. But depression treatment response depends on many factors beyond metabolism, including the specific neurochemistry of the illness, other medications, and individual health conditions. A green-flagged medication can still be ineffective for non-genetic reasons.
Should someone get a PGx test before ever trying an antidepressant?
Current guidelines from the International Society of Psychiatric Genetics do not recommend PGx testing as routine first-line screening. The strongest evidence supports testing for patients who have already tried at least one antidepressant without adequate response or who experienced significant side effects. Routine preemptive testing for all patients has not been shown to be cost-effective.
Are pharmacogenomic tests relevant for dementia medications like donepezil or memantine?
PGx testing panels designed for psychiatric medications focus primarily on antidepressants, antipsychotics, and related drugs. Donepezil is metabolized by CYP2D6, so a PGx test that includes this gene could provide relevant information. However, the clinical guidelines and most of the evidence base center on antidepressant prescribing rather than dementia-specific medications.
