Pharmacogenomics & Precision Medicine

June 8, 2022

BioVU celebrates 15 years supporting personalized medicine

BioVU is celebrating its 15th year and has enabled hundreds of studies and publications exploring the genetic underpinnings of a host of conditions including cancer, heart disease and diabetes. 

In 2003, Dan Roden, MD, then director of the Division of Clinical Pharmacology at Vanderbilt University Medical Center, imagined a large-scale biobank integrated with electronic health records to help doctors “personalize” medical care for their patients.

Dan Roden, MD

That vision became BioVU, today one of the world’s largest biobanks, with around 350,000 DNA samples from a single health care system. BioVU is celebrating its 15th year and has enabled hundreds of studies and publications exploring the genetic underpinnings of a host of conditions including cancer, heart disease and diabetes.

VUMC is well-positioned to host a biobank such as BioVU due to the size of its health care system, the thousands of patients who have been treated here, and the expertise that has been developed in analyzing health care records.

Linking DNA samples to the electronic health records (EHRs) of patients enables researchers to answer questions like these: ‘does a given genetic variation increase the risk for type 2 diabetes?” or “Will patients respond differently to treatment because of their genes?”

When undergoing blood testing at VUMC, patients are given the opportunity to participate in BioVU. If they consent, excess samples of their blood may be stored in freezers at temperatures as low as minus 80 degrees Celsius. Not all blood samples are used, or even stored. The samples first must go through a series of processing steps, including scrubbing records of all personal identifying information.

To use BioVU, researchers must submit their study proposals to VUMC’s Institutional Review Board and to the BioVU Review Committee of the Vanderbilt Institute for Clinical and Translational Research. If approved, the researchers can look through the deidentified version of the EHR and request that certain samples be genotyped. Researchers pay for the cost of genotyping, and they agree to make the genetic information they gather available to BioVU for other researchers to access in the future.

In 2016, a concerted effort was made to obtain genotyping data from a large cohort of the BioVU samples. Ultimately about 90,000 samples were genotyped. VUMC-affiliated researchers now can study BioVU samples, often without needing to pay for additional genotyping.

“The genotyping effort opened so many doors for studying genetic relationships,” said Lea Davis, PhD, associate professor in the Division of Genetic Medicine and an investigator in the Vanderbilt Genetics Institute.

Lea Davis, PhD

“A great part of the EHR is (that) it’s longitudinal,” Davis said. “Having medical record data from across patients’ lives allows us to study comorbidities over time.”

For example, Davis may see that a certain genetic variant increases the risk of depression in adolescence and heart disease in middle age. Samples from patients diagnosed with both conditions, one, or neither can be studied to try to understand how these very different traits may be related genetically.

Earlier this year, Nashville Biosciences, a wholly owned VUMC subsidiary, announced an agreement with Illumina, a leading genomic technology company. Through this agreement, Nashville Biosciences and Illumina will work with the pharmaceutical industry to finance whole genome sequencing of hundreds of thousands of de-identified BioVU samples. Resulting sequencing data will be made available to VUMC-affiliated researchers in a comparable manner to other BioVU genomic data.

For the sequencing data to eventually affect patient care, strong genetic associations to disease need to be established. Making the leap from discovery to the clinic is challenging, but the PREDICT program, which stands for Pharmacogenomic Resource for Enhanced Decisions in Care and Treatment, is helping to bridge that gap.

Patients who participate in PREDICT are genotyped, with a focus on verified genetic variants that affect how drugs are metabolized in the body and which can increase the risk for side effects. Clinicians then receive information about the predicted drug responses based on their patients’ genetic information.

 “If a patient is a rapid metabolizer for a medicine, their doctor may choose to treat them with something else, reducing their risk for adverse effects,” said Roden, Senior Vice President for Personalized Medicine and holder of the Sam L. Clark, MD, PhD Endowed Chair in the School of Medicine.

That, he said, is the essence of personalized medicine.