Skip to content Skip to navigation

Research & Ideas

Search this site

A new tool helps tailor prescriptions to complement a patient’s genome

The Pharmacogenomics Knowledge Base highlights about 5,500 genetic variants that affect the activity of 600 drugs.

Assortment of pill shapes and colors

Just as genes shape a person’s physical appearance, they can also alter the body’s sensitivity to drugs. | Unsplash/@freestocks

Not everyone responds to drugs in the same way—a drug that’s effective for one person might turn out to be a flop for another.

And while it can be difficult to pin down the exact source of variability in drug responses, a team including Russ Altman, MD, PhD and professor of bioengineering, has built an online database that’s helping to shed light on why a drug might help one person and not another.

It all comes back to an individual's genes, Altman said during a recent talk at the 2018 Beckman Symposium, an annual one-day event that focused this year on technology innovation and the human genome.

Just as genes shape a person’s physical appearance, they also create variation at a molecular level, which can alter the body’s sensitivity to drugs, or the rate at which they’re metabolized—both of which are crucial to the drug’s efficacy overall, Altman explained.

The database is a website that, through careful curation, collects information to help researchers, doctors and patients understand the intersection of genes and drugs, a field of study known as pharmacogenomics. Visitors to the site will find detailed information about molecular drug structures, metabolic pathways, and perhaps most valuably, specific genes that have the potential to influence how certain medications function in the body.

Currently, the website, called, which is short for the Pharmacogenomics Knowledge Base, highlights about 5,500 genetic variants that affect the activity of 600 drugs. Altman and his colleagues have written 27 guidelines for physicians with specific details about how to use genetics to choose and dose drugs. They even integrate their information with Food and Drug Administration labels so users can crosscheck the FDA's take on the genetic variations, too.

“Genetic variation can have profound effects on your clinical response to medications,” Altman said during his address.

“So, the hope, the vision, the expectation, if not the reality now, is that using genetic information paired with environmental variables and lifestyle considerations will improve the specificity and precision of [physician] medication choice and dosing.”

When working in the clinic, Altman said he uses a patient's genomic information to understand which drugs will work best and which drugs to avoid.

Doctors will also be able to spot whether a patient has a variety of known gene variants that could affect drug response and then flag certain drugs the patient should steer clear of, or opt for, in the future.

For instance, one gene variant that’s “a big one,“ as Altman described, stems from the gene CYP2C19. If this gene is tweaked in just the wrong way, it can alter how the patient metabolizes hundreds of drugs. “And that’s a problem, because then, for hundreds of drugs, the average dose is the wrong dose for you,” he said.

Altman said he also uses the database to add potential “allergies” to patient charts. If, for example, a patient has a bad reaction to one drug, the database can help doctors figure out if there are other drugs that might produce a similar response.

Many times, Altman said the information he adds to their charts is not immediately applicable, perhaps a person is predisposed to a disease, but does not have it yet. That’s okay, he said—the goal is to add the right information to the chart to empower the decision of the physician and patient if and when the time comes to navigate the disease.

Get Updates from Stanford Engineering