What if your genome could help select a treatment?
Your DNA can influence how your body handles certain medicines, and a tumour's DNA can reveal which drugs may work against it. Genetic information is already shaping some prescribing decisions. But not every variant carries a clear message, and most drug choices still rely on much more than genes.
⚠ This is a thought experiment. It explores a possibility; it does not report a result.
Original illustration — speculative and artistic, not a depiction of a real device, patient or result.
What exists today
Genetics is already used to guide some treatments. In cancer, tumours are commonly tested for changes that predict whether a targeted drug is likely to help, and some of those tests are FDA-approved companion diagnostics tied to specific medicines. Separately, pharmacogenomics looks at inherited variants that affect how people process drugs, which can influence dose or drug choice for some medicines, such as certain blood thinners, antidepressants and cancer treatments. The FDA maintains a public table of these gene-drug associations. Many clinics do not test routinely, and practice varies between countries, hospitals and insurers.
What has been demonstrated
Precision oncology has clearly shown that matching some drugs to specific tumour mutations can improve results for patients whose cancers carry those changes, which is why companion diagnostics exist. In pharmacogenomics, researchers have documented gene variants that change drug levels in the body, and for some pairings there is enough evidence for regulators to mention genetic testing on drug labels. Standards such as the ACMG/AMP variant-interpretation guidelines help laboratories classify genetic findings more consistently. What remains less clear is how much routine, broad genetic testing improves outcomes across whole populations, compared with testing targeted at specific situations.
What remains difficult
Most genetic variants have no clear link to drug response. Many tests produce variants of uncertain significance, which cannot yet be used to make decisions. Drug response is also shaped by age, kidney and liver function, other medicines, diet and adherence, so a gene rarely tells the whole story. Genetic databases have historically over-represented people of European ancestry, which can make results less reliable for others. Results need to reach the right clinician at the right time, often through electronic health records, and clinicians need training to use them. Privacy, cost and insurance coverage also shape whether people get tested.
What would change if it worked
Wider, well-validated use of genetic information could help doctors choose the right drug or dose sooner, rather than relying on trial and error. It could reduce some serious side effects and avoid prescribing drugs unlikely to work. In cancer, broader tumour sequencing might identify more patients eligible for targeted therapies or clinical trials. A single test done once could, in principle, inform decisions throughout a lifetime. But the benefits would be uneven unless databases become more diverse, testing becomes affordable, and health systems build the tools to act on results responsibly.
What evidence would convince us
Convincing evidence would come from randomised trials comparing genotype-guided care with standard care, measuring outcomes such as side effects, hospital admissions, treatment success and cost. Results would need to hold across diverse populations and routine clinical settings, not just specialist centres. For tumour testing, trials showing that matched treatment improves survival or quality of life, not only tumour shrinkage, would carry the most weight. Clear professional guidelines on which genes to test, when, and how to act on results would signal that the evidence has matured. Honest reporting of when testing changed nothing would matter too.
Should genetic tests that guide medication choices be offered more routinely?
Polls measure reader opinion for fun and discussion. They are not scientific evidence.
How should a doctor use a genetic result that is 'uncertain' — and how would you want that uncertainty explained to you?
Try it hands-on:
Play Decode a Genome →Related reporting
Genome + CellsDecode the Genome: What Sequencing Can Tell Us—and What It Cannot
From the $3 billion Human Genome Project to routine sequencing, reading DNA has become easy. Here is what a genome can reveal today, and where uncertain variants, risk scores and missing context set hard limits.
Cancer FrontiersThe New Cancer Toolkit: Immunotherapy, Targeted Drugs, Cell Therapy, and Precision Oncology
Targeted drugs, checkpoint inhibitors, CAR-T cells and biomarker testing have joined surgery, radiation and chemotherapy. Each works differently, helps some patients dramatically and others not at all.
Genome + CellsGene Editing, Gene Therapy, and RNA Medicines: Three Different Ways to Change an Outcome
In December 2023 the FDA approved Casgevy, the first CRISPR-based therapy, for sickle cell disease. Here is how gene editing, gene therapy and RNA medicines differ, and what that approval does and doesn't prove.
