I asked Northwestern University's William Catalona, MD, the developer of PSA testing and a genetic research pioneer, what he thought of this new development.
Here's his response:
Short answer: This approach has merit and I believe ultimately will become an adjunct to PSA testing.
This topic is controversial, and there likely will be pushback on this presentation, because some previous studies involving fewer prostate cancer-risk SNPs (single nucleotide polymorphisms) have concluded prematurely that prostate cancer polygenic risk scores (PRS) can distinguish between prostate cancer and no prostate cancer but cannot distinguish between aggressive and indolent prostate cancer. In truth, as shown in our studies and others, there are some prostate cancer risk SNPs associated with more aggressive prostate cancer.
Rosalind Eeles and Zsofia Kote-Jarai in the UK have led the way in their PRACTICAL Consortium by genotyping hundreds of thousands of men to discover many of the first 269 prostate cancer risk SNPs (I was a collaborator with deCODE Genetics that discovered the first risk SNP on chromosome 8q24). In these so-called GWAS (genome-wide association studies) studies, 40-50% of the aggressive cases have PRS scores in the top two PRS deciles, and few aggressive cases are in the bottom deciles. So, there’s no doubt that patients with aggressive tumors tend to have higher PRS scores. In fact, patients whose PRS are in the very highest deciles have a risk comparable to BRCA2 pathogenic mutation carriers.
Recent studies led by the USC group , including Christopher Haiman, David Conti, and Burcu Darst involving nearly 1 million US veterans have expanded the number of prostate cancer risk SNPs to 451, which has improved the discrimination for aggressive disease. Our prostate SPORE project, led by John Witte (Stanford) and me was the first to demonstrate that the 269 PRS was significantly associated with conversion from active surveillance to treatment.
In that study, we found that SNPs for a man’s genetically-regulated PSA secretion levels also are significantly associated with conversion from active surveillance. John Witte’s research group has a large NIH grant to further study PSA SNPs. This is important because PSA testing is confounded because PSA levels are increased not only in men who are genetically high PSA secreters, but also in men with BPH and/or prostatitis and men with prostate cancer. This confounding reduces the clinical specificity of PSA testing. However, in recent as-yet unpublished studies, we have found that adjusting the 451 PRS for PSA secretion SNPs is significantly associated with grade reclassification in men on AS (in the Canary PASS active surveillance consortium who were genotyped in our SPORE study).
There are also SNPs predicting the development of BPH. We are currently investigating whether adjusting the 451 PRS for the BPH SNPs will further improve the accuracy of PRS testing for prostate cancer aggressiveness. These studies are a major component of the NCI Program Project proposal that our PCASP (Prostate Cancer Active Surveillance Project) group is now negotiating with the NCI staff for permission to submit.
A concerning issue for me is that PRS testing with earlier PRS SNP panels and without adjustment for PSA SNPs have recently been released commercially by a few companies. These tests should be inexpensive – shouldn’t cost more that ~$250. And they can be done simply by having patients spit into a saliva tube and dropping the tube in the mail. Less accurate commercial tests could be aggressively marketed to select patients for PSA screening or active surveillance and yield inferior results that could adversely affect patient outcomes.
Remind me again why is it helpful to avoid “diagnosis leading to AS.” With my family history of PC, I’m very glad that I’m going to be getting PSA 2-3 times a year rather than once, as well as periodic MRIs and biopsies.
It all depends on your individual situation. Some of of us need mfore intense surveillance. Like those with a family history or those with certain genetic mutations. Getting on the AS train may be a lifesaver for some. Others shouldn't rush toward an unnecessary diagnosis and AS that carry potential financial toxicity, stigma, and emotional distress.
Things have changed since the early days of AS when we were facing down unnecessary treatment our doctors were pushing in a major public health disaster. Now those men with lower-risk disease ought to tread lightly before getting on an AS treadmill. In my opinion. HW
I asked Northwestern University's William Catalona, MD, the developer of PSA testing and a genetic research pioneer, what he thought of this new development.
Here's his response:
Short answer: This approach has merit and I believe ultimately will become an adjunct to PSA testing.
This topic is controversial, and there likely will be pushback on this presentation, because some previous studies involving fewer prostate cancer-risk SNPs (single nucleotide polymorphisms) have concluded prematurely that prostate cancer polygenic risk scores (PRS) can distinguish between prostate cancer and no prostate cancer but cannot distinguish between aggressive and indolent prostate cancer. In truth, as shown in our studies and others, there are some prostate cancer risk SNPs associated with more aggressive prostate cancer.
Rosalind Eeles and Zsofia Kote-Jarai in the UK have led the way in their PRACTICAL Consortium by genotyping hundreds of thousands of men to discover many of the first 269 prostate cancer risk SNPs (I was a collaborator with deCODE Genetics that discovered the first risk SNP on chromosome 8q24). In these so-called GWAS (genome-wide association studies) studies, 40-50% of the aggressive cases have PRS scores in the top two PRS deciles, and few aggressive cases are in the bottom deciles. So, there’s no doubt that patients with aggressive tumors tend to have higher PRS scores. In fact, patients whose PRS are in the very highest deciles have a risk comparable to BRCA2 pathogenic mutation carriers.
Recent studies led by the USC group , including Christopher Haiman, David Conti, and Burcu Darst involving nearly 1 million US veterans have expanded the number of prostate cancer risk SNPs to 451, which has improved the discrimination for aggressive disease. Our prostate SPORE project, led by John Witte (Stanford) and me was the first to demonstrate that the 269 PRS was significantly associated with conversion from active surveillance to treatment.
In that study, we found that SNPs for a man’s genetically-regulated PSA secretion levels also are significantly associated with conversion from active surveillance. John Witte’s research group has a large NIH grant to further study PSA SNPs. This is important because PSA testing is confounded because PSA levels are increased not only in men who are genetically high PSA secreters, but also in men with BPH and/or prostatitis and men with prostate cancer. This confounding reduces the clinical specificity of PSA testing. However, in recent as-yet unpublished studies, we have found that adjusting the 451 PRS for PSA secretion SNPs is significantly associated with grade reclassification in men on AS (in the Canary PASS active surveillance consortium who were genotyped in our SPORE study).
There are also SNPs predicting the development of BPH. We are currently investigating whether adjusting the 451 PRS for the BPH SNPs will further improve the accuracy of PRS testing for prostate cancer aggressiveness. These studies are a major component of the NCI Program Project proposal that our PCASP (Prostate Cancer Active Surveillance Project) group is now negotiating with the NCI staff for permission to submit.
A concerning issue for me is that PRS testing with earlier PRS SNP panels and without adjustment for PSA SNPs have recently been released commercially by a few companies. These tests should be inexpensive – shouldn’t cost more that ~$250. And they can be done simply by having patients spit into a saliva tube and dropping the tube in the mail. Less accurate commercial tests could be aggressively marketed to select patients for PSA screening or active surveillance and yield inferior results that could adversely affect patient outcomes.
Best regards,
Bill
Remind me again why is it helpful to avoid “diagnosis leading to AS.” With my family history of PC, I’m very glad that I’m going to be getting PSA 2-3 times a year rather than once, as well as periodic MRIs and biopsies.
It all depends on your individual situation. Some of of us need mfore intense surveillance. Like those with a family history or those with certain genetic mutations. Getting on the AS train may be a lifesaver for some. Others shouldn't rush toward an unnecessary diagnosis and AS that carry potential financial toxicity, stigma, and emotional distress.
Things have changed since the early days of AS when we were facing down unnecessary treatment our doctors were pushing in a major public health disaster. Now those men with lower-risk disease ought to tread lightly before getting on an AS treadmill. In my opinion. HW