What Can Your DNA Actually Tell You — and What It Can't
Dr. RP, MD — Board-Certified, Emergency Medicine & Critical Care Medicine — Founder, Analog Precision Medicine
Published March 1, 2026
I grew up in Canada. That means three things were essentially hardwired into my upbringing: Tim Hortons, apologizing reflexively, and hockey. Always hockey. So when Canada lost the gold medal game to the Americans in Milan last month — 2-1 in overtime, Jack Hughes sliding one through Binnington's five-hole with the whole country watching — I did what any self-respecting Canadian does. I sat quietly on my couch in Hermosa Beach, California, ate something unhealthy, and accepted the outcome with the specific kind of dignity that comes from years of practice. Canada outshot them 42 to 28. Connor Hellebuyck wasn't human. MacKinnon shouldn't have hit the side of the net. It doesn't matter.
Anyway. Genetics.
The analogy is in there somewhere. Carrying a certain set of genetic variants — even ones that look alarming on paper — doesn't mean you're destined for the outcome everyone fears. Biology, like hockey, plays out in ways that raw statistics don't fully capture. And the public conversation around genetic testing has gotten muddled enough that it's worth mapping out what the science actually supports, versus what the marketing implies.
Where Genetics Delivers Real Signal
Some variants operate more or less like biological certainties. BRCA1 and BRCA2 are the textbook examples — pathogenic BRCA1 variants carry lifetime breast cancer risks of 57–72% against a population average of roughly 12%, and ovarian cancer risks of 40–44%.[1] These numbers are not marketing estimates. They come from large prospective cohorts and directly inform decisions around surveillance, chemoprevention, and risk-reducing surgery.
Familial hypercholesterolemia is the underdiagnosis story that bothers me most. Variants in LDLR, APOB, or PCSK9 cause LDL levels of 300–400 mg/dL from birth — not from lifestyle, not from dietary choices, but as a factory setting. The CASCADE FH Registry estimates fewer than 10% of the approximately 1.3 million Americans with FH have ever been accurately diagnosed.[2] The other 90% are likely on a modest statin that isn't doing the job, told their cholesterol is “a little high” and to try eating better. Identifying an FH variant changes everything: treatment intensity, treatment targets, and the decision to pursue cascade testing in first-degree relatives.
The Most Underutilized Application: Pharmacogenomics
If I had to identify the single area of genomic medicine most ready for clinical primetime — and most underutilized — it's pharmacogenomics. The Clinical Pharmacogenomics Implementation Consortium (CPIC) has published evidence-based dosing guidelines for over 100 gene-drug pairs. The FDA has added pharmacogenomic biomarkers to more than 200 drug labels.[3] This is not emerging science.
CYP2C19 and clopidogrel. Plavix is one of the most prescribed drugs in cardiology — the standard antiplatelet agent after a stent, after an acute coronary syndrome, after the kind of cardiac event that brings someone into my ER at two in the morning. What most patients don't know, and what too few cardiologists routinely check, is that clopidogrel is a prodrug. It's inert until the liver activates it, and that activation runs almost entirely through one enzyme: CYP2C19. Somewhere between 25–30% of people carry loss-of-function variants in that gene. Their livers convert the drug poorly. They're walking around thinking they're protected, their cardiologist thinks they're protected, and the pill is doing a fraction of what everyone believes it's doing. A 2009 NEJM study of patients following PCI found that CYP2C19 poor metabolizers had a 53% higher rate of major cardiovascular events — not because they weren't taking their medication, but because the medication wasn't doing anything meaningful once they did.[4] The genetic test that identifies this costs less than dinner.
“The genetic test that identifies this costs less than dinner.”
The PREPARE study, a randomized controlled trial published in The Lancet in 2023 enrolling over 6,900 patients, found that pre-emptive pharmacogenomic panel testing reduced clinically relevant drug toxicity by 30% compared to standard care.[5] That's a prospective, randomized result. Not a hypothesis.
What Your DNA Can't Tell You
A few hard limits worth knowing:
A high-risk result is not a prognosis. Most common diseases emerge from the interaction of genetics, environment, lifestyle, and probability. Polygenic risk scores — which aggregate small effects from hundreds of variants — carry real predictive value at the population level, but a high score is not a verdict and a low score is not protection.[6]
A negative result isn't a clean bill of health. A negative BRCA test in someone with a strong family history eliminates tested variants, not hereditary risk entirely. Context matters.
Consumer tests are not clinical-grade. The FDA has authorized direct-to-consumer companies to report a narrow set of variants and explicitly notes these tests are not diagnostic.[7] Raw consumer data reinterpreted through third-party tools carries meaningfully elevated false-positive rates.
I ordered a consumer genetic test a few years back out of curiosity. It came back with elevated risk for a handful of things and informed me I have wet earwax. As revelations go, I'd had better days.
What it didn't give me was clinical context — a physician who understood both the findings and me, and could connect the results to what I should actually do differently. That gap between a data output and a clinical interpretation embedded in someone's full health picture is precisely where genetic information either becomes useful or doesn't.
Used well, it changes trajectories. At Analog Precision Medicine, genomic analysis is a clinical instrument — ordered deliberately, interpreted rigorously, and placed alongside biomarkers, phenotype, family history, and everything else we know about a person. DNA is a chapter, not the whole story. Getting value out of it requires someone who knows how to read the rest of the book.
And maybe — though I'll admit the timing still stings — who also handles overtime losses better than I do.
References
- 1.Kuchenbaecker KB, et al. Risks of breast, ovarian, and contralateral breast cancer for BRCA1 and BRCA2 mutation carriers. JAMA. 2017;317(23):2402–2416. doi:10.1001/jama.2017.7112
- 2.Knowles JW, et al. Development and validation of a model to identify individuals with heterozygous familial hypercholesterolemia using the electronic medical record. JACC. 2017;70(11):1323–1330. doi:10.1016/j.jacc.2017.07.714
- 3.Relling MV, Klein TE. CPIC: Clinical Pharmacogenomics Implementation Consortium of the Pharmacogenomics Research Network. Clinical Pharmacology & Therapeutics. 2011;89(3):464–467. doi:10.1038/clpt.2010.279
- 4.Mega JL, et al. Cytochrome P-450 polymorphisms and response to clopidogrel. New England Journal of Medicine. 2009;360(4):354–362. doi:10.1056/NEJMoa0809171
- 5.Swen JJ, et al. A 12-gene pharmacogenomic panel to prevent adverse drug reactions: an open-label, multicentre, controlled, cluster-randomised crossover implementation study. The Lancet. 2023;401(10374):347–356. doi:10.1016/S0140-6736(22)01841-4
- 6.Khera AV, et al. Genome-wide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nature Genetics. 2018;50(9):1219–1224. doi:10.1038/s41588-018-0183-z
- 7.U.S. Food and Drug Administration. Direct-to-consumer tests. FDA.gov. Updated 2023. https://www.fda.gov/medical-devices/in-vitro-diagnostics/direct-consumer-tests
Dr. RP, MD is dual board-certified in Emergency Medicine and Critical Care Medicine and is the founder of Analog Precision Medicine, a precision medicine practice serving executives, athletes, and health-conscious adults in Southern California. This article is for educational purposes and does not constitute medical advice or establish a physician-patient relationship.
← Back to Home