Dr. David Goldstein, Director of the Institute for Genomic Medicine (IGM) at New York’s Columbia University used to believe personalized medicine had a “bleak” future. “We were doing genome wide association studies and linking variations to diseases. In my view, we weren’t getting very far in terms of clinical relevance,” he explains. “I’ve always felt that we should be completely transparent, upfront, and accurate with the public. We shouldn’t be saying that we’re making more progress than we really are. We should describe the situation as we see it.”

But all this changed in the early 2010s when a young female patient afflicted by progressive neurodegenerative symptoms came to seek help. The patient’s primary care physician suspected she might have an autoimmune disorder. At the same time, the fact that her condition deteriorated quickly also indicated a clinical potential of precision medicine. So, Dr. Goldstein and his team decided to perform a whole-exome sequencing, a relatively rapid and affordable analysis of the protein-coding components of the genome that are responsible for about 85% of known disease-causing variants.

“We found that she has this super rare genetic disease,” Dr. Goldstein says. “Only 50 or 60 people in the world have it.” The young patient was eventually treated with vitamin B2 supplement. “And now, in one of the most amazing experiences of my career, she’s doing wonderfully.”

Thus far, Dr. Goldstein has authored over 200 papers on the clinical applications of genomic analysis in various diseases including schizophrenia, epilepsy, AIDS, and hepatitis C. “We are doing work very differently now. We are often able to find the exact causes of an individual patient’s disease, and that’s a game changer.”

Dr. Goldstein was named the founding director of Columbia’s Institute for Genomic Medicine in 2015 with the aim of carefully interpreting the genomes of all patients with unresolved or undiagnosed genetic diseases. Dr. Goldstein was also part of the Precision Medicine Initiative, a university-wide initiative founded by President Lee Bollinger, to “go beyond new cures for disease and the practice of medicine”.

The Precision Medicine Initiative encompasses virtually every part of the university, including areas that explore fundamental issues of human self-knowledge and the legal, policy and economic implications of revolutionary changes in our understanding of human biology. Part of the initiative is to collaborate with clinical departments to provide a comprehensive interpretation of patient genomes and act as a permanent resource for clinicians to make effective use of genome data to improve patient care.

One of the areas where Dr. Goldstein believes a difference can be made includes epilepsy and using comprehensive genomic approaches to determine what goes on in children with serious genetic diseases; “If a patient with epilepsy whose genome was sequenced to uncover a mutation in a particular potassium channel gene comes to us, our team can put the protein that gene encodes into a well-established laboratory protocol and test an FDA-approved drug that targets that protein, an approach that has been pursued by a different research team in Melbourne, proving that the effects of a mutation that causes epilepsy can be interrupted in the lab with a particular drug.

“This is why we need an integrated program,” he continues. “If we just get the mutation, we can’t do much with that. We have to understand what that mutation does, biologically, and how it relates to the mutations in other patients, so we can group patients based on the different underlying processes that have gone awry and not just their symptoms.”

But Dr. Goldstein is keen that his research work does not add to the present hype around sequencing, AI, machine learning, and how they influence the development of precision medicine. “There has been a lot of enthusiasm around pharmacogenetics,” he says. “Looking for clues in a patient’s genetic traits or disease profile to predict responsiveness to a particular drug protocol. The reality is that we still have very few real-world examples to understand and even validate the exact underlying molecular cause of a condition and guide how we treat it. There’s no magic. It took us years of halting and intermittent progress to arrive at where we are. But there’s still a long way to go before we find the right medicine for each patient.”