Dr. Salari on family history, germline genetic risk in favorable-risk prostate cancer

In this video, Keyan Salari, MD, PhD, describes the background and notable findings from the recent Journal of Urology study “Impact of Family History and Germline Genetic Risk Single Nucleotide Polymorphisms on Long-Term Outcomes of Favorable-Risk Prostate Cancer.” Salari is a urologic oncologist at the Massachusetts General Hospital and Harvard Medical School.

Transcription:

Please describe the background for this study.

Prostate cancer, as you know, is a high disease burden condition. It's the second leading cause of cancer death in the United States. But three quarters of patients who are diagnosed today are diagnosed with localized disease, many of whom have favorable prognosis that may not actually warrant treatment. But we know that there's substantial clinical heterogeneity among patients with even low- or intermediate-risk prostate cancers. Many of the patients that we initially monitor with what we call active surveillance ultimately end up requiring treatment in the long run. There has been growing evidence that family history and germline genetic risk variants contribute to the risk of aggressive prostate cancer. And so in this study, we set out to investigate the impact of both family history as well as germline genetics in a cohort of patients with prostate cancer who had favorable prognosis as defined by standard clinical parameters.

What were some of the notable findings? Were any of them surprising to you and your coauthors?

One of the notable things just to describe about the cohort is that this is called the Health Professionals Follow-Up Study, which is a long-standing prospective cohort study run by the Harvard T.H. Chan School of Public Health. It started in the 80s, and the median follow-up for the patients with prostate cancer in the study is nearly 15 years, which is quite a long time and really allowed us to explore and look at a long-term important outcome, which is prostate cancer death, the likelihood of dying of prostate cancer. So we were able to look at the role of family history. And one of the things we did was not just limited to the family history of prostate cancer, but we also explored family history of breast and pancreatic cancer, other cancer types that we know are genetically related to prostate cancer. We had previously shown in one of our active surveillance cohorts at Mass General here that patients who have this expanded family history of prostate, breast, ovarian, or pancreatic cancer, have a higher risk of having progression of their disease on active surveillance with higher grade disease showing up on future biopsies. So we wanted to look at this more expanded definition of family history in the Health Professionals Follow-Up Study. And what we found is that over a third of patients, 36%, had positive family history of either prostate, breast, or pancreatic cancer. And if you had this type of a family history, there was nearly a two-fold risk of dying of prostate cancer in the long term. And each first-degree relative that you have with one of these family histories added about a 50% increased risk of dying of prostate cancer. That was one of the first notable findings. And then we moved on to start looking at germline genetic risk variants in the study. What we first looked at was something called a polygenic risk score, which is basically a score that aggregates the effect of many, many, many SNPs or single nucleotide polymorphisms, a common type of genetic variation in the human genome, that each individually may contribute only a small amount of risk, but in aggregate together can potentially confer a large risk on your risk of developing prostate cancer or dying of prostate cancer. What we found was that the most commonly used modern polygenic risk score that aggregates 269 different SNPs actually was not associated with prostate cancer death in our cohort. In some ways, this actually wasn't a surprising finding, just because the initial era of polygenic risk scores were really designed from genome-wide association case control studies where the cases were any prostate cancer vs healthy controls. They weren't really designed to distinguish aggressive prostate cancers from indolent prostate cancers. And so it wasn't that surprising that this polygenic risk score in a cohort of our patients who all have prostate cancer didn't distinguish those who died vs didn't die. But there are newer polygenic risk scores that are being developed today that are in progress that are starting to look more closely at risk of aggressive vs indolent disease. And so after that, we started to drill down into some specific genetic variants that have been associated with prostate cancer survival, and we found 1 SNP on chromosome 19Q that was associated significantly with the risk of dying of prostate cancer—about an 80% increased risk if you had the risk allele. And then we looked together, combining that with family history and a multivariable model and found that family history and germline genetics - this specific germline genetic SNP—each independently increased the risk of dying of prostate cancer by about 80%. And if you limited the cohort to the patients with really low-risk disease, those that would be eligible for active surveillance under stricter criteria, it actually was almost a three-fold risk of dying of prostate cancer; the hazard ratio was 2.8. And so, it was quite interesting and just notable to see that both family history and this germline genetic risk SNP were independent contributors to that risk of dying of prostate cancer.

This transcription was edited for clarity.