I thought I'd add a link to a video discussion of this issue from the recent AUA meeting.
No Diagnosis, No TreatmentDrs. Vickers & Thompson said...
Dr. Thompson: We are moving into less overtreatment today, and so the question is, how do you reduce the amount of overtreatment? How do you take those people out of the equation? There are a couple of ways to do it. First, do your best to avoid finding inconsequential prostate cancer to begin with. Consider a man who is aged 55, white, with no family history, whose prostate-specific antigen (PSA) level goes from 0.5 ng/mL to 1.5 ng/mL, a common reason for a prostate biopsy. His PSA level is zooming up.
Dr. Vickers: The PSA velocity.
Dr. Thompson: Right. But we know that PSA levels go up and down and up and down. If you monitor it enough times, you will get a lot of prompts in these very low-risk men, in whom there may be a 10- to 20-fold greater likelihood that you will find an inconsequential prostate cancer, and a 1%-2% likelihood that you will find a consequential cancer. To that person, you might say, "We want to repeat your PSA level in 6 months or a year and then think about it again" On the flip side, the person who has a higher PSA level, and is older, black, and has a nodule, has much higher likelihood of an aggressive cancer. That man should be biopsied. Randomized trials say that you can make a difference in that man.
So, you don't biopsy the men who are most likely to be low risk to begin with. Biopsy the men who are high risk. Once you look to see what the pathology looks like, if it's really low risk (eg, low volume, Gleason score 3+3), the likelihood that you are going to make a difference is extremely low.
Dr. Vickers: We avoid overtreatment if we don't find the disease in the first place. Then, we don't need to overtreat. Many of the guidelines used to say, "Here is when you should biopsy a man: if he has a high PSA level, a low free:total ratio, a family history, a positive digital rectal examination (DRE), or a recent change in PSA (a high PSA velocity)." There were lots of reasons to biopsy a man. Now, let's restrict that. Let's only biopsy a man if we have a very good reason to. Then, we are not going to pick up these inconsequential cancers and a man won't be at risk for overtreatment because he won't have prostate cancer detected.
Dr. Thompson: The classic example is the man who is referred to the urologist with a prostate nodule. You can feel it, and he's young. He's white, and his PSA level is 0.5 ng/mL. His likelihood of an aggressive prostate cancer is 1%. The likelihood of finding inconsequential prostate cancer is 20%-25%. The likelihood of net harm is high. If you are on active surveillance for low-risk prostate cancer, it's not fun. You undergo repetitive biopsies and PSA levels. You are labeled, you are anxious, and you are a "cancer survivor." We do lots of things, and it costs just as much as many of the usual treatments. It's a net harm. You want to biopsy the man who has a greater likelihood of an aggressive cancer. We are pretty good at it. We can predict it.