Retrieving new clues about a dog breed’s “insane” cancer risk

Golden retrievers, consistently among the most popular dog breeds in the United States, are known as playful and family-friendly companions that are eager to please. A large longitudinal study, now in its 12th year, has revealed an additional, devastating trait: three of every four documented retriever deaths so far have been linked to cancer—by far the highest rate for any breed and among the highest rates of any animal. Of those cancer deaths, 70% are due to hemangiosarcoma, an aggressive blood vessel malignancy that is almost always fatal except for an uncommon cutaneous subtype.

If cuddlier than the unusually cancer-resistant rodents known as naked mole-rats, golden retrievers are at the opposite end of the susceptibility spectrum, making them another focal point of research. Teasing out cancer-associated factors, researchers say, could help to improve the beloved dogs’ longevity—as well as our own.

The Denver-based Morris Animal Foundation launched the biggest research effort to date, the Golden Retriever Lifetime Study, in 2012.¹ Veterinary researchers there enrolled 3044 privately owned dogs throughout the United States (all between 6 months and 2 years of age) to investigate not only the incidence but also environmental and genetic risk factors for cancers and other diseases, such as cognitive decline and osteoarthritis. With the oldest participants now turning 14 years old, Julia Labadie, DVM, PhD, MSPH, the study’s principal investigator, says that studying aging in dogs has emerged as an unanticipated additional goal.

“We have now a cohort of pretty old golden retrievers,” she says, noting that a significant fraction of those survivors could die of non-cancer causes. “So I think there’s a lot of questions that we can answer about the dogs that don’t get cancer and the dogs that live longer than the normal lifespan for golden retrievers that we always quote of about 10 to 12 years.”

Another recent study already is hinting that at least part of the longevity difference may be linked to variants in a gene encoding an epidermal growth factor receptor. Led by Robert Rebhun, DVM, PhD, chair of medical oncology at the University of California Davis School of Veterinary Medicine, the researchers discovered that a variant in a noncoding region of the ERBB4 gene (also known as HER4) grants golden retrievers an extra 2 years of life on average.² Interestingly, ERBB4 appears to have a “good variant” that is associated with longer lifespans and a “bad variant” that is associated with shorter lifespans.

Because longevity in the breed is highly influenced by cancer, the genetic variants are almost certainly associated with cancer as well, says coauthor Michael Kent, DVM, MS, a professor of radiation oncology at the veterinary school. Other research has found that ERBB4 can serve as both a tumor suppressor and an oncogene.

“Everyone thought golden retrievers had a high rate of cancer and died maybe a little younger, but we really hadn’t looked at it,” Dr Kent says. Multiple research groups had previously struggled to find cancer-linked genes amid the background noise of genetic analyses. Dr Rebhun instead flipped the question on its head to ask, “Which dogs live extraordinarily long?”

Beyond their in-house data, the researchers recruited participants from dog shows and through word-of-mouth referrals to collect DNA from exceptionally long-lived retrievers. A genome-wide association study that compared dogs reaching at least 14 years of age to those dying before the age of 12 years uncovered a significant link with the ERBB4 gene. Dr Rebhun and his colleagues have not yet determined the potential mechanistic pathway, although a planned follow-up using samples from the Lifetime Golden Retriever Study could help to validate their findings.

In a separate retrospective study led by Dr Kent, researchers pored over the records of 652 golden retrievers for which the age at death was known and the cause of death had been established through a necropsy examination.³ The study determined that 65% of the dogs had died of cancer—less than the percentage calculated by the Morris Animal Foundation but still an “insane” number, Dr Kent says. Hemangiosarcoma and lymphoid neoplasia accounted for the two most common diagnoses.

Controversy has raged over whether non-neutered males and non-spayed females die of cancer at lower or higher rates. Dr Kent’s study, though, found that intact and neutered male golden retrievers died of cancer at roughly the same rate. Although spayed females died of cancer at significantly higher rates than their non-spayed counterparts, the study found that they also lived significantly longer: 9.5 years versus less than 6 years. Therefore, spayed female retrievers lived long enough to get cancer, Dr Kent says, whereas cancer is relatively uncommon in dogs younger than 5 years. Age, in other words, “appears to have a larger effect on cancer-related mortality than reproductive status,” he and his colleagues concluded in the study.

Longitudinal studies like the one conducted by the Morris Animal Foundation could help to clarify some of the links between cancer, age, and reproductive status. Dr Labadie and Kelly Diehl, DVM, MS, senior director of science and communications at the foundation, say that the power of their study derives in large part from the extensive sampling and data compilation. Beyond blood samples taken at the time of diagnosis and every year before then, the study has collected serum, urine, fecal, hair, and toenail samples from every participant, with biopsies and histopathology performed on most of the tumors. The dog owners complete extensive surveys each year, while electronic medical records capture prescription data and all diagnoses.

Other potential connections are starting to emerge. Lymphoma, the second most common cancer documented by the Golden Retriever Lifetime Study, is a cancer of the immune system, or a hematopoietic malignancy, while new research suggests that hemangiosarcoma may help promote hematopoietic tumors.⁴ This commonality means that a shared genetic mutation could predispose golden retrievers to hematopoietic cell malignancies. “If we can find a biomarker that’s present in these dogs with cancer and then start looking back in time and see how early we can pick that up, we could really make a difference in diagnosing this earlier,” Dr Labadie says. In turn, because dogs have shorter lifespans than humans, potentially damaging effects—including environmental exposures such as PFAS chemicals in water—could appear sooner. Our canine companions, in short, also could be useful sentinels for human health.

For the time being, however, the researchers caution that translating new findings in golden retrievers to humans may be premature. Although dogs and humans mirror each other in age-adjusted cancer incidence curves and develop many of the same tumors, the closest human analogue to hemangiosarcoma is behaviorally distinct angiosarcoma. Ongoing studies in dog breeds, however, may help to reduce the background noise often associated with human cancer genetics. “You can use a much smaller number of dogs, when they’re genetically related, to find something,” Dr Kent says. Comparing golden retrievers in the United States to an external population of dogs in Europe also could help to identify genetic differences between the two populations.

Retrievers more commonly develop T-cell lymphomas and pediatric osteosarcoma than humans do, but Dr Labadie says that both canine cancers are very good analogues to their human counterparts. “It is a unique opportunity to study some of these subtypes that are more rare in people, but common in dogs,” she says. Dr Diehl adds that the ongoing study also could uncover a relatively broad cancer-associated risk factor that triggers separate disease pathways depending on the breed or species. “So maybe it’ll be a hit and cancer, not the hit and hemangiosarcoma or lymphoma,” she says.

Given the improvements in health care and nutrition, of course, more dogs are living long enough to develop cancer. “When humans lived to 30 years old, there wasn’t as much cancer. We live longer now and so we see a lot more cancer, and it’s going to be the same with our pets too,” Dr Kent says. “I think it just makes sense to use what we can learn in dogs and then go apply it to humans.”