% Free PSA test

Hi Guys,

This is my first time posting to this forum.

My PSA has been under surveillance for over a year and I'm now scheduled for a targeted prostate biopsy. The biopsy is scheduled for Sept 13th and I won't get feedback until the 28th.

Recently I asked my doctor for a % Free PSA test. He effectively said that here in BC Canada we don't do that. So I checked a copy of a Blood Requisition form I have and there is no "% free PSA" box to tick.

I asked for this test because of the strong desire for more information and the extended wait for the biopsy.

So my question is why has the % free PSA test fallen out of favour at least in BC?

Thanks

You're already scheduled for a biopsy, so the free PSA may not provide much more useful information at this point.

Neither the PHI nor the 4K Score are offered. Only the PSA blood test.

Thanks Allen,

I know the PSA test "is a lousy test" but your comment that the % free PSA test isn't much better is the first I've seen. It's odd that the first is frequently criticized but the other blood tests don't seem to be.

I imagine the only reason the PSA test is offered is because the other tests are too expensive, not that the PSA test is inexpensive itself.

They are all fairly inexpensive. PHI was $125 last time I looked.

The problem is that none of them are very good. A good diagnostic test would have a sensitivity of 90+% and a specificity of 90+%. Sensitivity is the percent of all cancers that are correctly picked up by the test. Specificity is the percent of all non-cancers that the test correctly deems non-cancerous. (So a sensitivity of 100% means there are no false negatives, and a specificity of 100% means there are no false positives. False negatives are missed cancers. False positives are sent for unnecessary biopsies.). Sensitivity and specificity are defined relative to a certain cutoff value.

For PSA, at a cut-off of, say 4.0, the sensitivity is only 20% -- 80% of the cancers are missed! The specificity at that cut-off is 95%, so only 5% would be unnecessarily biopsied. If you use a cutoff of, say, 1.0, you will pick up 83% of the cancers, but 61% of men will be sent for an unnecessary biopsy. There is no cut-off that simultaneously has a reasonably good sensitivity and specificity. It's overall accuracy is only 53% - about the same as a coin toss. You can see why it causes so much controversy.

% free PSA is a little better. At a cutoff of 10%, the sensitivity is 45% , and the specificity is 95%. In other words, if only men with % free PSA <10% are biopsied, it will miss more than half the cancers, but almost all the biopsies will be positive. At a cutoff of 25%, the sensitivity is 97% (they will catch almost all the cancers when biopsied), but the specificity is only 13% (87% will be sent for an unnecessary biopsy). There's no cutoff that is reasonably good at both sensitivity and specificity, and most men will fall in the middle (median is 15%). The overall accuracy is 65% - still, not very good.

PHI, at a cutoff of 35, has a sensitivity and a specificity of 65% - overall accuracy of 72%. Still not great, but better than the other two. At least it has a cutoff where both sensitivity and specificity are reasonable.

The problem worsens when you are trying to detect only significant cancers, which would be ideal. Of those tests, PHI is better than the other two, but still not great.

To Mark and Allen,

Thanks for your comments they were both helpful and insightful.

Hey MacroMan - hope you're finding the discussion useful. Sorry for some of the added baggage on your post. We're here to help, though it's not a perfect process!

(An aside to Mr. Mark:

[Edit... deleting comment... never mind... forum dynamics claimed another one... Mark wouldn't see what I wrote anyway. I see he's bailing for good. Very sad... topic for another day...]