Video
A Urologist’s Commentary on Biomarker Test Use in Clinical Practice
Author(s):
Dr Andriole shares insight into how he incorporates pre-biopsy biomarker tests into prostate cancer risk assessment in his clinical practice.
Dr. Gerald Andriole: So, I'm a fervent believer that we should only biopsy men who are at risk for prostate cancer. And in addition to looking at the PSA alone in isolation, I think urologists are best advised to also look at PSA in relation to the size of the prostate, what's been referred to as PSA density, because that can help us understand whether the PSA elevation is due to cancer or just benign enlargement of the prostate. But then, PSA density is helpful, but it's not quite as good as biomarkers would be, and it's certainly not as good as an MRI would be. Long answer to a specific question; I never proceed to biopsy without either an MRI or a biomarker showing me that a biopsy is indeed necessary.
It's been well-recognized that among the biomarkers that are recommended, for example by the NCCN guidelines, that sometimes one gets conflicting information; some may be elevated and worrisome for cancers, others may be reassuring. And that's one area that we as a specialty need to do a better job in the coming years, which is to compare and contrast the available biomarkers to determine which ones are best in which particular clinical situation. But in my practice, I rely as I indicated, for the most part, on the 4Kscore and/or MRI to guide my biopsy decisions.
I think to rely exclusively on an elevated PSA as an indication for a biopsy, particularly if the biopsy form would be a random biopsy of the prostate, is a strategy that we started doing in the late 1980s and early 1990s and we know just how bad it is; that it results in performing way too many biopsies and it results in missing way too many cancers. So, the whole strategy in the last several years has been to do the proper type of biopsy in the proper patient, and you can identify the proper patient with good biomarker testing and/or MRI.
There are multiple statistical analyses of biomarker performance. Their sensitivity, in other words how, what proportion of the cancers will the marker identify? There's negative predictive value, in other words, how reliable is it to interpret a biomarker result that can safely tell me that I don't need to do a biopsy of the prostate; that's the negative predictive value. In my way of looking and thinking about the problem, I like biomarkers that have a high negative predictive value, meaning they're very reliable in telling me that the man in front of me does not need a biopsy because it's very unlikely that he has prostate cancer, so hence I know I won't be missing any clinically relevant cancers in that patient. And secondarily, like a marker that's sensitive, but I think in comparing the different statistical parameters that a high negative predictive value is a very important one.
