HealingWell
Search Close Search
☰

Thoughts about uPSA nadir of < 0.008 following RP to independently predict BCR risk?

Support Forums
>
Prostate Cancer
✚ New Topic ✚ Reply
❬ ❬ Previous Thread |Next Thread ❭ ❭
Posted 1/24/2020 8:19 AM (GMT 0)
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4106725/

"In the multivariate analysis, statistically significant differences were observed only in pathological nodel stage within the PSA nadir <0.008 ng/ml group (P=0.0107). For this reason, postoperative follow-up using ultra-sensitive PSA is considered to be of value, since the use of high-sensitivity PSA to confirm a reduction to below postoperative measurement threshold value (PSA nadir <0.008 ng/ml) may avert administering unnecessary additional treatment, regardless of pathological reccurrence factors."
Posted 1/24/2020 10:53 AM (GMT 0)
Huh? OK, I'm reading this first thing in the morning, I'm not sure I'm all the way awake. It sounds like this is saying that if you go undetectable after RP, that your chances of BCR are really low, and that post RP, one should be getting usPSA.

This is standard of care. It's what we do.
Posted 1/24/2020 11:08 AM (GMT 0)
I think the results are not dissimilar from this 2005 study, which found a 4% BCR rate for nadir <0.01 (a 2-decimal test) using 0.1 as the BCR point.

Keep in mind the OP's study of Japanese men used an Abbott uPSA (limit = 0.008) for which we do not have the standard deviation at the testing levels. Nor can I make out how they "compensated" for the "background noise" of non-prostatic PSA sources.

PSA nadir appears to be a useful parameter to factor in with the other well-known parameters, but using a very low predictor level may run into difficulty with assay accuracy and reliability.

Djin
Posted 1/24/2020 12:15 PM (GMT 0)
djin, exactly. As an analytical chemist, I've never been real comfortable with the advice to pull the trigger on follow up therapy at 0.02--or whatever the latest number is. Heck, when I was first treated and living in rural small town america, the only ultrasensitive test I could get was a the Urologist's office..and HIS machine only went down to <0.04.

Don't get me wrong, I'm really happy getting a string of <0.006's. And if it starts to move, I may feel differently about it. BUT--I'd also really like to know what the analytical bias is for the test.
Posted 1/24/2020 3:51 PM (GMT 0)
but, could you express that in a poem?
Posted 1/24/2020 7:37 PM (GMT 0)
djin halbert thanks for engaging

I wish I was still at < 0.006 as that ended 18 months post RP

been going downhill for the last 30 months with a steadily rising uPSA to 0.012

my background is Physical Chemistry with some biochemistry, Applied Math and Physics despite the fact I made my living in software automating the jobs of others in silicon valley

BigData and machine learning was my last area of expertise which can trivially answer these kinds of questions as to what are the odds of a post RP uPSA of 0.012 ever going to BCR if the data is made available

I value my health more than my privacy so my uPSA data is open

it is frustrating that no one has processed the uPSA data into a model that can factor in uPSA numbers and uPSA velocity and predict BCR risk at 5 year, 10 year, 15-year time windows

I believe the LabCorp uPSA test has a standard deviation of +/- 0.002 which means guys that are consistently below < 0.006 are for all practical purposes forever at or below 0.003. when you add in the natural variance of uPSA in the body these guys are really at 0.001 or less

so my uPSA level is 12x of those guys and rising, i am in a different zipcode

the standard useless reply is when I cross a uPSA of 0.03 then BCR is likely

well I would greatly appreciate knowing my odds of ever having a uPSA of 0.03 given my uPSA of 0.012 and upward trend velocity

even better my odds of BCR given my uPSA of 0.012 and upward trend velocity

these questions can be answered hopefully now to some degree as the data is there

if you are the smartest guy in the room you are in the wrong room

assume = to make an ass out of you and me
Posted 1/24/2020 8:57 PM (GMT 0)
The independent lab that looked at the Roche assay that Labcorp uses found that (in near ideal lab conditions) for the 0.01 target solution, repeat testing produced a S.D. of +/- 0.001. Generally 2 S.D. are used, since that covers about 95% of results. This gives the +/- 0.002. (Note that for the 0.1 target, 2 S.D. was higher, about +/- 0.005.)

As I mentioned in another post about the uPSA, we mortals don't generally operate in ideal conditions; nonetheless, from my own results over 28 months, I think the test is fairly accurate and precise. There are many intrinsic and extrinsic factors. One study found that PSA (before RP) varies with the time of day! I don't know how it varies day to day, or at our small numbers after treatment. Nor whether or how much our non-prostatic PSA varies over time ("background noise"). Extrinsic factors outside those of the assay itself are things like sample prep., equipment calibration, reagent prep (if any are involved), etc. We know that PSA varies from one assay method to another, too. And 2 S.D. means that about 5% of results may be even higher or lower (I believe 3 S.D. takes you to about 98% -- still not 100%).

My own values have been:

11-28-17 (3 m ) 0.010
02-26-18 (6 m ) 0.009
05-30-18 (9 m ) 0.007 (nadir)
08-27-18 (1 yr.) 0.018 (huh?)
09-26-18 (13 m) 0.013 (retest)
11-26-18 (15 m) 0.012
02-25-19 (18 m) 0.015
05-22-19 (21 m) 0.015
08-28-19 (2 yr. ) 0.016
12-18-19 (28 m) 0.015

I don't think you can take any one value too seriously -- for example it looked like my PSA more than doubled in 3 months at one point! Can I take to the bank that my(Labcorp) nadir was just below the paper's "magic" value for the Abbott assay's undetectable, <0.008? That would be nice, but I don't think so. If everyone were followed with a 3-digit uPSA we'd have a ton of data by now, but that's not the case.

Djin

Post Edited (DjinTonic) : 1/24/2020 2:07:14 PM (GMT-7)

✚ New Topic ✚ Reply