Predictive Medicine: How DNA Testing Is Influencing Healthcare

Predictive medicine is growing in its influence on healthcare in the United States and around the world. Yet few healthcare professionals can correctly articulate just what it is, its anticipated (and likely revolutionary) influence on how medical care functions, and how it will greatly benefit patients and professionals.    

BioSupply Trends Quarterly recently conducted an interview with Brandon Colby, MD, founder and CEO of Sequencing.com, the world’s largest platform for DNA testing and analysis. The Sequencing.com team consists of medical doctors, geneticists, bioinformatics experts and software developers who share a passion for helping patients live healthier, happier and longer through DNA analysis. Dr. Colby is trained in clinical genomics and internal medicine, and he has a bird’s-eye view of predictive medicine that few can rival. 

BSTQ: How would you best describe predictive medicine?

Dr. Colby: Predictive medicine uses patients’ biomarkers and genetic information to understand what diseases they are at risk for throughout life. That information is also used to implement proactive, preventive measures that will help mitigate the risks of those diseases even before they arise.

BSTQ: How prevalent is predictive medicine today, and what is it anticipated to be in the future? 

Dr. Colby: Today, predictive medicine is used in pieces by different physicians. For instance, oncologists may utilize genetic testing to look at people’s risk for certain types of diseases. Internal medicine physicians and pharmacologists may use particle genomics, which is a way to look at people’s genes to determine what medications may pose risk of an adverse reaction for them. Different specialties are using genetic testing in different ways to understand risk. Those are small pieces of predictive medicine: personalized proactive prevention put into place by predicting what a person is at risk for and using that information to steer some type of treatment plan or preventive plan. So, it is not very prevalent in terms of being its own specialty. Rather, it is about predictive medicine having more and more of an impact upon each individual specialty as healthcare providers become comfortable using genetic information and integrating it into their practice. 

BSTQ: Can you provide a recent example of a case in which predictive medicine was used successfully?

Dr. Colby: When it comes to the genes we’re familiar with, such as the BRCA1 and BRCA2 genes that cause a very significant increased risk of breast cancer, there are many ways to implement personalized, preventive measures throughout a person’s life, even early on for adolescents. A recent case was a young girl whose BRCA1 mutation was detected that had the potential to dramatically increase her risk of cancer. While there was no really drastic preventive measure at that point in her life, we were able to advise her parents of the importance of avoiding radiation exposure to their daughter’s chest throughout her life. So, when she goes to the dentist, the importance of making sure the lead vest is covering her chest and her thyroid extremely well should be communicated. If she has a cough and goes to her pediatrician, that pediatrician should understand to hold off on a chest X-ray unless it is very urgent she have one. Any radiation exposure to her chest is like throwing gasoline on a fire, increasing the risk of breast cancer. So while there’s no medication that can be given — and, of course, we’re not going to perform surgery on someone so young — there are still steps we were able to provide to these parents. With this knowledge, there was a type of empowerment in that they understood the risks, and they understood that their daughter will likely have to deal with breast cancer down the line. But, at least today, they’re able to take some steps that will help limit her overall risk throughout her life. 

BSTQ: Why is predictive medicine important in healthcare? 

Dr. Colby: It’s quintessential to the survival of healthcare. It helps to reformulate healthcare from being about “sick care,” or focusing only on people who have already become sick. Healthcare providers know there’s a losing battle in waiting for someone to get sick. Predictive medicine is changing the whole paradigm of healthcare from that “sick care” model to one that is truly healthcare focused on personalized prevention of disease. There is no other way to predict disease risk without predictive medicine that focuses on patients on a personal level, understanding their environmental risks, as well as their genetic risks, including what’s already happening inside their bodies today. With predictive medicine, we can go beyond and focus on the personalized prevention of disease, which is really when we start to prevent disease from ever occurring. That way, we’re getting ahead of the problem, and that’s when our healthcare system is going to be able to survive, both in terms of bandwidth and cost. It is a lot more cost-effective to stop a disease from occurring than to treat it once it has arisen. 

Oncologists are now experts at using BRCA1 and BRCA2 testing as part of their practices. For psychiatrists, pharmacogenomics continues to play an increasing role in their practices. The daily use of genetic information by physicians is also quickly moving into internal medicine to identify risk and prescribe optimal treatments for chronic diseases.

And, studies have shown that BRCA1 and 2 testing have very clear economic benefits for identifying those people who are at risk for breast cancer (including men), and then monitoring and preventing the cancer before it occurs. If we look at the cold numbers, predictive medicine offers a significant advantage to the medical system and the entire world in terms of detecting disease at its earliest stages when treatment is usually most effective, or preventing disease from occurring in the first place. 

BSTQ: What are some of the complications of predictive medicine? 

Dr. Colby: In terms of understanding the risk, we’re not usually talking about absolutes. We’re talking about risks such as odds ratios and relative risks for increasing or decreasing the risk of disease in individuals. So, it can be a little bit of a gray area, for instance, if a person may have an increased risk for liver cancer, but that increased risk in terms of absolute risk is not very significant. What is really needed is a guide to understanding these risks to help steer whether what has been detected on a genetic level is or is not important. 

BSTQ: What are the tools used in predictive medicine? 

Dr. Colby: The predictive tools I use daily are genetic testing results, which allow me to understand changes in genes and/or genetic variances a person has and conditions for which they are at risk. I also use biomarkers, which are usually blood tests that will indicate serum levels of a substance. Even something as simple as vitamin D has a lot of predictive value in terms of risks for certain diseases, so that’s another example of a simple biomarker I test for frequently when I treat patients. Genetic risk doesn’t change from the day a person is conceived until the day he or she dies, and the risks are able to be identified for the long term. You can look at a baby and understand his or her risk of developing Alzheimer’s disease 30, 40, 60 years into the future. Biomarkers, on the other hand, give us a much better idea of what’s going on within that person’s body today, as well as within the next six months to a year. So, genetic testing and biomarkers together provide a much clearer predictive picture than either alone — the biomarkers in terms of what’s going on in the short term and the genetics for what’s going on in the long term. 

BSTQ: With regard to predictive tools, are false-positives and false-negatives a problem? 

Dr. Colby: They are definitely a problem. Whenever we’re looking at multivariant analysis, there are many different data points being obtained for a test. There are always going to be some that are lower quality, and there are always going to be some false-positives and false-negatives within that data set. For instance, for a data set in which we’re simply looking at vitamin D levels, it is much easier to get the correct result. But when we start to look at genes using something like 23andMe or other such companies, they review about 600,000 data points. Genetic variances within a person’s genome for whole genome sequencing is what I focus on — sequencing a person’s entire genome — which includes more than three billion data points. So, when analyzing an entire genome, it is important to understand there are going to be false-positives and false-negatives and how to implement different quality controls to limit those false-positives and -negatives. If we come back to the BRCA1 and BRCA2 testing example, when that is detected, it’s crucial to have a follow-up genetic test to confirm the results. That’s part of the counseling a physician provides: making sure the initial test is correct when a result is going to potentially have a very profound impact. A follow-up test ensures that something like a false-positive is not guiding our approach so it first must be validated. 

BSTQ: Are there any ethical objections to predictive medicine?

Dr. Colby: For those of us who employ predictive medicine, ethics are not as impactful when testing people who are not pregnant but may be thinking about getting pregnant. When it comes to genetic testing, usually the ethical debates are limited. Most physicians, patients and others are OK with people learning about their genes. 

However, there is a different type of ethics surrounding predictive medicine when it comes to various companies that perform genetic testing such as 23andMe, Ancestry.com and other similar types of companies. One of the ethical debates surrounding this type of testing is what those companies do with the genetic data after they obtain it. For example, are they selling it? Or, what else might they be doing with it?

BSTQ: Is there anything else you would like to add regarding predictive medicine? 

Dr. Colby: One of the most important things is there’s a true revolution in the field of genetic testing, and that surrounds whole genome sequencing. As I mentioned, companies such as 23andMe and Ancestry.com look at around 600,000 genetic variants, but that is less than 0.1 percent of a person’s genome. Whole genome sequencing (Figure) obtains data on all three billion genetic variants, which is 100 percent of a person’s genome. While the cost of whole genome sequencing in a single person was more than $1 billion 20 years ago, the cost today is now less than $500 a person! Studies have projected that within the next several years, we’re going to start to see hundreds of millions of people having their genome sequenced.

Governments around the world are already evaluating the recommendation that all newborns have their whole genome sequenced at the time of birth. In the United States, there’s the All of Us Research Program from the National Institutes of Health, which has sequenced the genome of a million people. Similar programs exist in the United Kingdom, European Union and China. So, we’re about to see this major influx of genetic information into the medical field, where everyone will start to have their genome sequenced and will want their medical providers to utilize the data as part of their care using predictive medicine.