CCPP Module 29, Section 3: Selecting Patients for Genetic Testing: Clinical and Ethical Considerations
Module 29: Applied Pharmacogenomics in Collaborative Practice

Section 3: Selecting Patients for Genetic Testing: Clinical and Ethical Considerations

A practical guide to “case finding.” You will learn the key clinical indicators for PGx testing, focusing on high-risk drugs and patient populations. We will explore scenarios involving therapeutic failure, unexpected adverse effects, and pre-emptive testing, alongside a crucial discussion of the ethical considerations surrounding genetic information.

SECTION 29.3

Selecting Patients for Genetic Testing: Clinical and Ethical Considerations

From Shotgun Approach to Precision Targeting.

29.3.1 The “Why”: Maximizing Value and Minimizing Noise

You have now mastered the language of pharmacogenomics and the evidence-based playbook of the CPIC guidelines. The logical and perhaps most challenging next question is: “Who should be tested?” While it may one day be feasible to perform pre-emptive, panel-based testing on every patient, we do not yet operate in that reality. For now, PGx testing remains a valuable but finite resource. Your role as a collaborative practice pharmacist is to become an expert in identifying patients for whom this resource will provide the maximum clinical benefit, while also navigating the complex ethical landscape that surrounds genetic information.

Simply ordering a test “just in case” is not a strategy. It is a shotgun approach that can lead to information overload, clinically irrelevant findings (noise), and unnecessary costs to the healthcare system and the patient. The goal is not to test everyone, but to test the right person at the right time for the right reason. This requires a new layer of clinical acumen: the ability to perform expert “case finding.” You must learn to recognize the subtle and overt clinical signals that suggest an underlying pharmacogenomic issue is contributing to a patient’s poor therapeutic outcome. You are moving from being a medication manager to a clinical detective, using clues from the patient’s medication history, side effect profile, and planned treatments to pinpoint where a genetic test can solve a difficult clinical problem.

Furthermore, this is not merely a scientific or clinical decision. A pharmacogenomic test is fundamentally different from a serum creatinine test. A genetic result is immutable, lifelong information that has implications not only for the patient but also for their family. Therefore, the decision to test is a shared one, steeped in ethical considerations that you must be prepared to navigate. It requires a thoughtful process of pre-test counseling and informed consent that respects patient autonomy and addresses potential concerns about discrimination, privacy, and the handling of unexpected findings. This section will equip you with a dual skillset: the clinical framework to identify high-yield testing opportunities and the ethical framework to ensure that testing is performed responsibly, transparently, and always in the patient’s best interest.

Pharmacist Analogy: The Master Electrician’s Diagnostic Toolkit

Imagine a large, complex office building where some lights flicker, some outlets don’t work, and one wing keeps tripping its breaker. How would a true professional diagnose the problem?

  • The DIY Approach (Shotgun Testing): A novice might walk in and start randomly flipping every breaker in the main panel off and on, hoping to stumble upon the solution. This is disruptive, inefficient, and doesn’t identify the root cause. This is akin to ordering broad, untargeted genetic panels on every patient without a clear clinical question, creating noise and potential for misinterpretation.
  • The Master Electrician’s Approach (Targeted, Guideline-Driven Testing): A master electrician arrives with a specific toolkit and a diagnostic process.
    1. Interview (Gathering Clinical Clues): First, they ask targeted questions. “Which lights are flickering? When did the outlet stop working? Does the breaker trip when you use the microwave?” This is you, the pharmacist, identifying a clinical problem: “The patient’s depression isn’t responding to a second SSRI,” or “The patient had a stent thrombosis despite being on clopidogrel.”
    2. Consulting the Blueprint (The CPIC Guideline): The electrician looks at the building’s electrical schematic. They see that all the failing components are on “Circuit 7.” This schematic is your CPIC guideline, telling you that SSRIs and clopidogrel are both on the “CYP2C19 circuit.” You now have a prime suspect.
    3. Using the Multimeter (The PGx Test): The electrician doesn’t replace the whole circuit. They go to the junction box for Circuit 7 and use a multimeter to test the voltage. This specific, targeted test gives a definitive answer: “There’s a faulty connection in this box.” The PGx test is your multimeter. It directly tests the function of the suspected gene (CYP2C19) and provides a definitive answer: “This patient is a Poor Metabolizer.”
    4. The Repair (Actionable Recommendation): With the precise diagnosis, the electrician makes a specific repair—tightening the faulty wire. You, armed with the PGx result and the CPIC guideline, make a specific recommendation—”Switch from clopidogrel to ticagrelor.”

Your role is to be the master electrician. You don’t test blindly. You use your clinical expertise to identify a problem, consult the evidence-based “schematics” to form a hypothesis, and then recommend a specific diagnostic “tool” to confirm that hypothesis and guide the ultimate “repair.”

29.3.2 The Reactive Approach: Case Finding in Patients with Existing Problems

The most common and highest-yield opportunities for PGx testing in current practice are reactive. You are identifying patients who are actively experiencing a problem—either the drug isn’t working, or it’s causing unacceptable side effects. Your existing skills in medication therapy management (MTM) and patient counseling make you uniquely positioned to spot these clinical red flags. This is about connecting the dots between a patient’s complaint and a potential underlying genetic cause.

Indicator 1: Documented Therapeutic Failure

A patient has been prescribed an appropriate dose of a medication for an appropriate indication, has been adherent, yet is experiencing no or minimal therapeutic benefit. While many factors can contribute to therapeutic failure, an inherited variation in drug metabolism or transport is often a prime, overlooked suspect.

Masterclass Table: Recognizing PGx-Related Therapeutic Failure
Clinical Scenario Drug Class Example Drug Pharmacist’s Observation / Patient’s Complaint Potential PGx Link (Gene & Phenotype) Actionable Next Step
Cardiology / Post-PCI Antiplatelets (Prodrug) Clopidogrel (Plavix) A patient experiences an in-stent thrombosis, MI, or stroke despite documented adherence to clopidogrel. CYP2C19 Poor or Intermediate Metabolizer. The patient cannot effectively activate the prodrug, leading to inadequate platelet inhibition. Recommend Testing & Immediate Intervention. “Dr. Smith, given this breakthrough event on clopidogrel, the patient is at high risk for being a CYP2C19 poor metabolizer. The CPIC guidelines strongly recommend switching to prasugrel or ticagrelor. I also recommend we order the CYP2C19 test to confirm for future therapies.”
Psychiatry / MDD SSRIs/TCAs (Active Drug) Sertraline, Escitalopram, Amitriptyline Patient has trialed two or more antidepressants at therapeutic doses for adequate duration (>6-8 weeks) with no meaningful improvement in depression scores (e.g., PHQ-9). CYP2D6 or CYP2C19 Ultrarapid Metabolizer. The patient is clearing the antidepressant so quickly that they never reach a therapeutic concentration in the brain. Recommend Testing to Guide Future Selection. “Dr. Jones, this patient has now failed two SSRIs. Before we trial a third, it would be valuable to get a PGx panel. If they are an ultrarapid metabolizer of 2D6 or 2C19, we could either select an agent that avoids those pathways, like vortioxetine, or know that we might need to push the dose of another agent much higher than usual.”
Pain Management Opioids (Prodrug) Codeine, Tramadol Patient reports absolutely no pain relief from a standard dose of codeine or tramadol. They are often unfairly labeled as “drug-seeking.” CYP2D6 Poor Metabolizer. Codeine is a prodrug for morphine; tramadol is a prodrug for its more potent M1 metabolite. A PM cannot perform this activation and derives no analgesic effect. Recommend Testing & Immediate Change in Therapy. “The patient’s report of zero effect from codeine is a classic sign of a potential CYP2D6 poor metabolizer. This is not a behavioral issue; it’s a genetic one. We must switch to an active opioid that doesn’t require CYP2D6 activation, such as morphine, hydromorphone, or oxycodone.”
Gastroenterology / GERD Proton Pump Inhibitors (Active Drug) Omeprazole, Pantoprazole Patient with severe GERD or erosive esophagitis has persistent symptoms despite BID dosing of a standard PPI. CYP2C19 Ultrarapid Metabolizer. The patient is clearing the PPI so rapidly that the duration of acid suppression is too short to be effective, especially overnight. Recommend Testing to Justify Dose/Drug Change. “This patient’s refractory GERD on BID pantoprazole could be due to ultrarapid CYP2C19 metabolism. Testing could confirm this and provide a strong rationale for either increasing the dose significantly or switching to a PPI less dependent on CYP2C19, like rabeprazole or dexlansoprazole.”

Indicator 2: Unexpected or Severe Adverse Drug Reactions (ADRs)

This is the other side of the coin. A patient experiences a severe or unusual side effect at a standard, or even low, dose of a medication. This often points to a “Poor Metabolizer” phenotype for an active drug, leading to accumulation and toxicity, or a specific immune-response gene.

Masterclass Table: Investigating PGx-Related ADRs
Clinical Scenario Drug Pharmacist’s Observation / Patient’s Complaint Potential PGx Link (Gene & Phenotype) Actionable Next Step
Rheumatology / Gout Allopurinol A patient of Han Chinese or Thai descent develops a severe, blistering rash, fever, and mucosal involvement within weeks of starting allopurinol. HLA-B*58:01 Positive. This allele is strongly associated with allopurinol-induced Stevens-Johnson Syndrome (SJS) / Toxic Epidermal Necrolysis (TEN), a life-threatening hypersensitivity reaction. This is a Medical Emergency. The drug must be stopped immediately. Testing is done to confirm the diagnosis and to prevent re-challenge. For future gout management, febuxostat or probenecid must be used. CPIC recommends pre-emptive testing in high-risk ancestries.
Cardiology / Dyslipidemia Simvastatin Patient develops severe muscle pain (myalgia), weakness, and has a markedly elevated creatine kinase (CK) level after starting simvastatin 40 mg. SLCO1B1 variant (e.g., *5/*5 genotype). The SLCO1B1 gene codes for the OATP1B1 transporter, which helps move statins from the blood into the liver. A low-functioning transporter leads to higher systemic concentrations of the statin, increasing the risk of myopathy. Recommend Testing to Guide Statin Choice. “This patient’s severe myopathy on simvastatin is highly suggestive of a SLCO1B1 variant. Testing is recommended. Regardless of the result, we should switch to a statin less affected by this transporter, such as pravastatin or rosuvastatin, and start at a low dose.”
Psychiatry / MDD Amitriptyline An elderly patient started on a low dose (e.g., 25 mg) of amitriptyline develops profound sedation, confusion, dry mouth, and constipation. CYP2D6 Poor Metabolizer. The patient cannot clear the active drug, leading to toxic accumulation and exaggerated anticholinergic and antihistaminic side effects. Recommend Testing & Drug Change. “The patient’s extreme sensitivity to a low dose of amitriptyline strongly suggests they are a CYP2D6 poor metabolizer. We should stop the drug and switch to an agent that is not a primary 2D6 substrate. A PGx test would confirm this and help guide all future psychiatric prescribing.”
Pain Management Codeine A breastfeeding mother taking codeine for post-partum pain reports that her infant has become excessively sleepy, limp, and has difficulty feeding. CYP2D6 Ultrarapid Metabolizer (in the mother). The mother is converting codeine to morphine at an extremely high rate. The high levels of morphine pass into her breast milk, causing opioid toxicity in the infant. This is a Medical Emergency. Instruct mother to stop codeine and stop breastfeeding immediately and seek emergency care for the infant. This is a black-box warning for codeine. Testing can confirm the mother’s UM status to prevent future exposures.

29.3.3 The Proactive Approach: Pre-emptive Testing Before Problems Arise

While reactive testing is a powerful problem-solving tool, the ultimate goal of pharmacogenomics is to be pre-emptive—to use genetic information to select the best drug and dose from the very beginning, preventing therapeutic failure and adverse reactions before they have a chance to occur. This is a paradigm shift from “trial and error” to “test and target.” Your role here is to identify clinical situations where the risk of a drug-gene interaction is so high and the consequences so severe that ordering a test before the first dose is the most prudent course of action.

Strategy 1: Focus on High-Risk Drugs

Certain medications have well-established, high-stakes interactions with common genetic variants. When you see one of these drugs about to be prescribed, it should trigger a mental alert to consider pre-emptive PGx testing.

The Pharmacist’s “Top 5” Pre-emptive Testing Checklist

When you encounter a new prescription for one of these drugs, it should prompt a conversation about the value of PGx testing.

  1. Allopurinol in High-Risk Ancestries: Before initiating allopurinol in a patient of Han Chinese, Thai, Korean, or other high-risk Southeast Asian descent, strongly recommend testing for HLA-B*58:01 to prevent SJS/TEN. This is a CPIC Level A recommendation and is becoming a standard of care.
  2. Clopidogrel in High-Risk Scenarios: Before a patient undergoes a planned percutaneous coronary intervention (PCI) where clopidogrel is the intended P2Y12 inhibitor, recommend CYP2C19 testing. Preventing a stent thrombosis is far better than reacting to one.
  3. 5-Fluorouracil or Capecitabine in Oncology: Before a patient starts therapy with these chemotherapeutic agents, testing for variants in the DPYD gene (which codes for the DPD enzyme) is critical. DPD deficiency can lead to life-threatening toxicity. This is another CPIC Level A recommendation.
  4. Warfarin Initiation: Before giving the first dose of warfarin, consider testing for CYP2C9 and VKORC1. While many institutions use clinical dosing algorithms, genotype-guided dosing has been shown to reduce bleeding events, especially in the initial phase of therapy.
  5. Abacavir for HIV: This is a classic, universally accepted example. Before any patient ever receives a dose of abacavir, testing for HLA-B*57:01 is mandatory to prevent a potentially fatal hypersensitivity reaction. This is a black-box warning and a standard of care.

Strategy 2: Focus on High-Risk Patient Populations

Beyond specific drugs, some patient populations are at higher risk for medication-related problems due to the nature of their illness or the typical prescribing patterns within that specialty. In these areas, advocating for broader, panel-based pre-emptive testing can be highly valuable.

  • Psychiatry Patients: The treatment of major depressive disorder, anxiety, and other mental health conditions is notoriously characterized by trial and error. Patients often cycle through multiple medications over months or years, suffering from side effects and prolonged illness. A pre-emptive panel covering key genes (CYP2D6, CYP2C19, SLC6A4, HTR2A) can provide a roadmap to guide therapy, potentially shortening the time to an effective regimen. It helps clinicians avoid drugs likely to cause side effects (due to PM status) and identify drugs that might be ineffective (due to UM status).
  • Polypharmacy Patients: Any patient, particularly the elderly, on a large number of medications is at high risk for both adverse events and drug-drug interactions. Many drug-drug interactions are phenoconversion events (e.g., a strong CYP2D6 inhibitor making a Normal Metabolizer behave like a Poor Metabolizer). Knowing the patient’s baseline genetic metabolism can help you anticipate and manage these interactions with far greater precision. For example, knowing a patient is a CYP2D6 IM at baseline means that adding even a moderate inhibitor like duloxetine could render them a full PM, increasing their risk of toxicity from other CYP2D6 substrates.
  • Chronic Pain Patients: Similar to psychiatry, chronic pain management often involves trialing multiple agents. A PGx test can provide objective, biological data that can be incredibly helpful. It can identify a CYP2D6 PM who will not respond to codeine/tramadol, or a CYP2C9 PM who may be at higher risk of GI bleeding from NSAIDs. This can help tailor therapy and, importantly, validate a patient’s reported experience.

29.3.4 The Ethical Framework: Navigating the Nuances of Genetic Information

A pharmacogenomic test result is not just another lab value. It is a piece of a person’s fundamental blueprint, and as such, it carries a unique set of ethical, legal, and social implications (ELSI). As the medication expert facilitating this testing, you have a professional and ethical obligation to be well-versed in these issues and to guide the patient through them. This is a core competency of a PGx-enabled pharmacist.

Pillar 1: Meaningful Informed Consent

Informed consent for PGx testing is more than just getting a signature on a form. It is a dedicated counseling session to ensure the patient understands the test’s purpose, benefits, limitations, and potential consequences. Your role is to facilitate this conversation.

The Pharmacist’s Pre-Test Informed Consent Checklist
Discussion Point Key Information to Convey Sample Scripting for the Pharmacist
Purpose of the Test Clearly state why the test is being recommended for *this* patient and *this* drug. Avoid jargon. “The reason we’re discussing this test is because the medication we’re about to start, clopidogrel, works differently in different people based on their genetics. This test helps us predict if it will be the most effective and safest choice for you.”
Potential Benefits Explain the “best-case scenario”—how the test could improve their care. “If the test shows that clopidogrel might not work well for you, we can choose a different medication from the start, which could lower your risk of having a heart attack or stroke.”
Potential Risks & Limitations Be transparent about what the test *doesn’t* do and potential downsides. “It’s important to know this test only looks at how your body processes certain medications. It’s not a test for disease risk. Also, while it gives us powerful information, it’s one piece of the puzzle, and we still need to consider all your other health factors.”
Nature of Genetic Information Explain that the result is lifelong and may have implications for blood relatives. “Unlike a blood sugar reading that changes, this genetic result will not change over your lifetime. Because you share genes with your family, your result might also give some information about how they might respond to these medications.”
Potential for Incidental Findings If using a broad panel, discuss the possibility of finding information unrelated to the immediate question. “Because this test looks at several genes at once, there is a small chance it could find something we weren’t looking for. The lab has a policy on what types of serious, actionable findings it will report back. We can discuss what you would want to do in that situation.”
Cost and Insurance Be as clear as possible about potential out-of-pocket costs. “We’ve run a benefits investigation, and it looks like your insurance will cover this test with a $50 copay. If for any reason the cost is higher than that, the lab will contact you before running the test.”

Pillar 2: Genetic Information Nondiscrimination Act (GINA)

Patient fear of genetic discrimination is a major barrier to the uptake of PGx testing. You must be able to explain the protections offered by the landmark 2008 federal law, GINA, as well as its limitations.

What GINA Protects

GINA makes it illegal for:

  • Health Insurers: To use a person’s genetic information to make decisions about their eligibility, coverage, or premiums. They cannot require a genetic test or use your PGx result to deny you a health plan.
  • Employers: (with 15 or more employees) To use genetic information to make decisions about hiring, firing, promotion, or other terms of employment.

What GINA Does NOT Protect

GINA’s protections do not apply to:

  • Life Insurance
  • Disability Insurance
  • Long-Term Care Insurance
These insurers can still ask about and use genetic information when you apply for a new policy. This is a critical counseling point for patients.

The “Manifest Disease” Loophole

GINA prevents discrimination based on your genetic information, but not based on a current, diagnosed health condition (a “manifest disease”). If a pharmacogenomic test revealed you were a `SLCO1B1` poor function carrier (the genetic info), an insurer couldn’t use that. But if you later developed statin-induced myopathy (the manifest disease), they could potentially use that diagnosis in their underwriting, as it’s now a pre-existing condition. This is a nuanced but important distinction.

Pillar 3: Privacy, Data Security, and Health Equity

You must also be prepared to address patient concerns about who will have access to their genetic data and to be mindful of the broader societal implications of this technology.

  • Data Security: Explain that the genetic testing lab and the health system are bound by HIPAA, just like for any other piece of protected health information. The results are placed in their secure medical record. Discuss the lab’s specific policies on data de-identification and use in research, if applicable.
  • Health Equity: Be aware that the frequency of many pharmacogenomic alleles varies significantly across different ancestral populations. For example, the `CYP2C19` no-function alleles are more common in individuals of East Asian descent, while the `CYP2D6` poor metabolizer phenotype is more common in those of European descent. This means the clinical impact of certain drugs will be disproportionately felt by different groups. As a pharmacist, you must advocate for equitable access to testing and be cautious not to let “race-based” prescribing become a crude, inaccurate substitute for actual genetic testing. Always remember: ancestry informs risk, but it does not replace the test.