Genomics is simply the study of DNA’s effects on and interactions with all the other systems within a biological organism. What makes this field especially exciting right now is the sheer number of breakthroughs and practical applications making their way into human medical practice—and slowly but surely into veterinary practices. Genomics has been around since Gregor Mendel and his pea plants, but took off in the mid-1950s with the discovery of the structure of the DNA molecule by Watson and Crick; and then exploded in 2002 with the first full sequencing of the human genome.
Veterinary medicine is no stranger to genomics, but the principal application for this field has been less about curing diseases than about helping with breeding decisions. This is important, but really left many veterinarians a little cold, especially companion animal clinicians in a country where around 85 percent of pets are spayed or neutered. That is all about to change.
In the 20 years since the human genome was sequenced (with the canine genome following two years later), the hype around these advances that were going to change everything in medicine, has quieted somewhat. However, the quiet belies a slow, steady improvement in the understanding of genomics alongside advances in computing power and gene sequencing technologies that now make genomics at the clinical level both practical and cost-effective.
Today, we are able to sequence and analyze substantial parts of the genome for a couple hundred dollars per sample. Similar to how computers being priced low enough to move into our homes changed the way we live, as genetic tests move into our clinics, we will achieve new insights at an exponential rate. Alas, what can these tests do?
Genomics as prognostics
Imagine being able to have the following discussion at your next new puppy exam:
“Mrs. Nelson, congratulations on bringing this beautiful ball of joy into your home and family. We appreciate the cheek swab you got to us a couple weeks ago, and the physical exam findings are close to perfect. She is a practically perfect little pup. There were two findings in her genetic screening that make me want to make a couple additional recommendations. She has a couple risk variants—our term for mutations—in genes called SLC7A9 and SLC3A1, which have been associated with cystinuria, essentially crystals of cysteine that can form in her bladder.
Now, we don’t know for sure if anything will come of these mutations, but to be safe, I’m going to recommend you bring us a urine sample every six months, and we’ll be able to catch anything before it develops into bladder stones or chronic infections. The second mutation we picked up was in a gene called VWF, which controls one of her blood clotting factors. It’s usually not very serious, but can cause her clots to form more slowly than dogs without that mutation. Because of that, I’d like to delay her spay a couple of months in order to make sure she’s big enough to handle the surgery better and we’ll also have some plasma available in case we need it to help her through.”
With a simple cheek swab test, you have gotten a clue into two potentially serious conditions in this dog, established a plan to deal with them and, not incidentally, generated an opportunity for more frequent interactions with this client. You can deliver better medicines and treatments for your client’s puppy moving forward—and this is not science fiction.
In a 2023 paper published in PLOS Genetics, a research and development team has combined the data analysis tools of big tech with a database of over a million DNA tests to show common mutations we can find in pet dog populations.1 These kinds of tests can be clinical gold mines that allow us to proactively seek out likely diseases, catch them and treat them early, and provide families with treatment options and the knowledge they are being excellent parents.
Yet, it does not stop with identifying disease risk based on the mutations our patients are born with.
Genomics as diagnostics
Genomics is commonly central to the active clinical management of diseases. Nowhere is this perhaps more obvious than in cancer’s clinical management because cancer cells are the result of an accumulation of mutations over an animal’s lifetime. Imagine a hypothetical cancer that needs four genetic “switches” flipped in a particular sequence in order for those cells to start dividing uncontrollably. Every animal is born with a certain number of those switches already flipped the wrong way.
Breeds that appear to be predisposed to cancer (golden retrievers or boxers) may have one or two of those switches in the wrong position. During a lifetime of cells dividing, other mistakes in copying the DNA are made, and more switches flip for that cell and every cell that comes after it. As those daughter cells divide, more mistakes are made until one day, in one cell, the last switch flips and the cell loses its inhibitions against uncontrolled division—and a tumor is born.
This accumulation theory of oncogenesis is widely accepted in human medicine, and billions of dollars have gone into studies to understand it better and exploit it to treat and cure cancer. The opportunity that viewing cancer development in this way allows is that if we know exactly which mutations have occurred in a tumor, and, therefore, which molecular pathways have been disrupted, then we can develop drugs to undo that disruption, and that specific cancer can be stopped. Additionally, the combination of those mutations are likely only occurring in the tumor cells, and not the normal cells in the body, which might hold a key to treatment options.
When the human genome was first sequenced, the scientific community—naively, in retrospect—believed a few “mountains” in the genomic landscape would be quickly discovered; that there would be a few mutations common to all or most cancers and with those mountains fully explored, we would soon find the keys to curing cancer. Well, the genomic landscape turned out to be more complicated. The few mountains turned out to be a multitude of hills and valleys that took decades to map. And while that mapping process is continuing to add higher resolution, the maps we have today have already become foundational tools in human oncology in which the identification of specific mutations in cancer tissue samples is often incorporated into the standard diagnostic workup. Specific mutations are sometimes uniquely enriched in specific tumor types, particularly amongst a set of differentials, and so clinical genetics and genomics laboratories are critical components of the human oncology infrastructure.
Genomics for therapeutic planning
There are over 145 targeted therapies approved by the U.S. Food and Drug Administration (FDA) for use in humans. They are “targeted” to inhibit specific pathways that have been perturbed by mutations. This is a fundamentally different approach to treating cancer than our traditional approach of killing all rapidly dividing cells and hoping that more cancer cells die than do normally dividing cells. Given they target mutations that are likely to only be found in tumor cells, these drugs are the closest thing in our arsenal to “silver bullets” that kill only the cells in the tumor and spare all non-tumor cells. How do we know which genes have been mutated in a given tumor? By testing the DNA in those tumor cells.
Cancer genomics and the behavior of cancer cells is complicated. However, cells cannot act in a way their DNA does not direct them to. Understanding the mutations in the DNA of a cancer cell is the first step in understanding how to stop that specific cancer in that specific patient. Sometimes similar cancers carry similar mutations, but what has become clear in the past 20 years is that each cancer, in each patient, is unique. To understand that tumor and how to stop it requires sequencing the DNA of the tumor cells.
There are a number of these tests available to veterinarians. They take a piece of the tumor (obtained through needle aspirate or biopsy), extract the DNA of the tumor cells, and analyze that DNA for mutations in cancer genes. Once the mutation pattern is established, then treatment recommendations can be made. While we do not have 145 targeted drugs available in veterinary medicine, we do have some, and many more are available off-label. Luckily, we know quite a bit about the safety of these drugs in dogs, and we are learning more about efficacy with every case that is followed.
Genomics in veterinary medicine has evolved from a tool primarily used in breeding decisions to one that has immediate and practical clinical applications. Especially in companion animals, where our knowledge of disease pathways is deeper and our ability to translate science from humans to species more central to our profession. Genetic testing can help us understand which diseases an animal may be predisposed to, and when diseases develop, we can understand which specific pathways are involved, and, therefore, which pathways need to be corrected through targeted therapies.
We have made huge advances, and the tools are available now, and this is only the beginning. As knowledge accumulates and markets mature, these tests will become more precise and applicable and one day will be as ubiquitous and central to our diagnostic process as small animal biochemical profiles and complete blood counts.
David Haworth, DVM, PhD, is president and co-founder of Vidium Animal Health, a veterinary genomic diagnostics company. Dr. Haworth was previously president of PetSmart Charities in the U.S. and Canada, and prior to that, president/CEO of Morris Animal Foundation. Haworth started his professional career with 11 years at Pfizer Animal Health (now Zoetis, Inc.) after a short stint in private practice. He earned his veterinary and graduate degrees at Colorado State University (CSU), where he also completed a postdoctoral fellowship at the Flint Animal Cancer Center.
Reference
- Donner J, Freyer J, Davison S, Anderson H, Blades M, Honkanen L,
et al. (2023) Genetic prevalence and clinical relevance of canine Mendelian disease variants in over one million dogs. PLoS Genet 19(2): e1010651. https://doi.org/10.1371/journal.pgen.1010651