I came across this study from 2011 where Partin et al were researching the reliability of a 5th generation digital immunoassay for PSA using single molecule arrays. This particular test measured to less than 0.1 picograms, so more than 1/1000th more sensitive than the standard testing sensitivity of 0.1 ng/ml.
In addition to proving that the 5th generation test was accurate and effective, a conclusion of the paper was that this 5th generation assay was was able to show bifurcation in eventual BCR likelihood in men from these extremely low levels of PSA. What I found most interesting were the closely followed PSA levels over the first 18 months or so post-surgery and the implications these had for eventual BCR. If I interpreted the information correctly, no one recurred that had a reading of less than 0.003 ng/ml during those first 18mos. Other interesting things I noticed is that there were a number of Gleason 3+2 patients that had surgery and that a Gleason 8 and Gleason 9 were among those that didn't recur (although most of those that recurred, did have unfavorable pathology, staging etc.)
This begs the question: why are these super-super-sensitive assays not widely available? Cost? Something else? It seems like the recurrence fears for many men could be put to rest with a test like this. I, for one, would like to know what my "true" PSA is post surgery. While I'm <0.006 and blessed for now, this data would suggest that knowing if I'm <0.003 has (maybe a lot of) value. This study, while small, is yet another data point that increases my belief in the value of the uPSA.