PSA tests--Total vs. Ultrasensitive a... - Prostate Cancer N...

Prostate Cancer Network

3,364 members1,927 posts

PSA tests--Total vs. Ultrasensitive after radical prostatectomy

Christopher_E profile image

It has been ten years since I had a radical prostatectomy (with negative margins). At my surgeon's suggestion, I have had yearly "total" PSA tests since then but not since 2019 so I'm overdue. The results so far have always been reported as <0.1 ng/mL without knowing what, if anything, is happening below that level. Quest, who owns the labs in my area, also has a more sensitive "Post Prostatectomy" test which measures down to 0.02 ng/mL but I've never had that test.

Is there an advantage to getting the ultrasensitive PSA instead of the total PSA? Knowing just that the result is less than 0.1 seems a bit like a dummy light on a car that goes on only when things really need attention. Maybe its better to know what, if anything, is happening at a lower level than 0.1?

At what PSA level would treatments be considered? If it's not until the PSA is over 0.1, then it makes sense to stick with a total PSA test and not know anything below that level, even if it is increasing.

I know an ultrasensitive test might cause unnecessary anxiety but, if I see it slowly increasing year after year, then it would be safe to assume there is still prostate cancer somewhere and it will eventually be over 0.1

Thanks for your suggestions.

22 Replies

A recent trial showed that oncological results were no worse compared to immediate salvage radiation if one waited for PSA to rise to 0.1 or 3 consecutive rises. This was among post prostatectomy men who were originally intermediate risk AND had adverse pathology (positive surgical margins, extracapsular extension, or penetration into the seminal vesicles).

For someone like you with no adverse pathology and a long history of undetectable PSA (on a conventional test), I can't think of any reason why you would continue to have annual PSA tests at all. Maybe once every other year if you want to be really prudent.

I had radiation 10 years ago and consider myself cured. My PSA is <0.1. I have the test only because my PSP requests it, but like you, I forgot to get one one year.

Steve507 profile image
Steve507 in reply to Tall_Allen

Dear TA,Please share your disease/treatment history. Gleason scores, etc. Thanks

safari1 profile image
safari1 in reply to Tall_Allen

Hi Tall how much time after SBRT before your nadir was reached? Along the way did it bump up some before descending to nadir?

Thanks in advance. Currently .5 40 months after 42gy sbrt

Tall_Allen profile image
Tall_Allen in reply to safari1

4 years, several bounces.

safari1 profile image
safari1 in reply to Tall_Allen

Understand thanks

You’ve sweated it out for a decade! You are done with PC! PSA monitoring can be stopped now with an extremely low risk of subsequent BCR.

I am now 31-months post-RALP and have had 11-ultrasensitive PSA tests. My very experienced urologist, (retired from doing 25-years of prostatectomies), did not want me to use the ultrasensitive test. This was AFTER getting a surgical path report showing G7 (3+4), a tiny positive margin, and an even smaller extra-prostatic extension.

I insisted on the ultrasensitive test because I am a retired numbers guy who made a career of predicting future events based on a lot of complex statistical data.

I can now tell you that all I have learned from all my data is the ultrasensitive PSA goes up, goes down, stays the same, makes me anxious... and has performed no useful function.

My results, taken every three months are:

0.021 0.018 0.022 0.028 0.035 0.050 0.048 0.055 0.060 0.069 0.052

My urologist wants me to start testing once a year or at most twice a year.

Tall_Allan has it right - I now know I will wait until I have had a least two, and probably three, increases ABOVE 0.10 before I opt for SRT. And, I might not even then because I am sneaking up on 74-years old and there is some research, which shows guys as old as me have a lower overall survival rate with SRT than without, even in the event of increasing post-RP PSA.

The other thing that makes it easy for me to ignore my current meandering µPSA is my Decipher PostRP genomics results, which tell me I have less than a 4% chance of metastasis in 10-years. There are several nomograms that use the path report data, PSA trend data, and the Decipher results, which shows less than a 2% chance of me dying from PCa in the next 15-years, at which time I will be 88-years old.

But, those nomograms only need to know if your post-RP PSA is less than 0.1 or 0.2 and how many months post-RP you are.

Didn't run your numbers, but by just reviewing them I estimated your doubling time somewhere from 1.5 to 2 years. Not bad at all, you can probably skip sRT.

The Johns Hopkins doubling time (based on the slope of the natural log of PSA ) is currently 13.4 months from the nadir in January 2019. That doubling time has increased from 9.4 months after the first four measurements to what it is now.

Using just the last 7 µPSA values (Hopkins recommends the last year of so of measures) gives a value of 49-months because the January 2020 value was 0.050 and the January 2021 value had only increased to 0.052.

The good news is that the Hopkins doubling time using all data points does not get me to BCR (0.20 repeated on a 2nd test) until fall of 2023. And using just the last 7-data points keeps me below BCR thru 2026 (age 79).

Among the 11 values that you posted I do not detect any nadir. It looks to me monotonously upgoing with some random noise superimposed. If you give me the dates I can be more specific tomorrow.

Nadir = lowest point = 0.018 in January 2019.

OR - what do I misunderstand?

My professional life was statistics and predictions since 1972.

A nadir in calculus is relevant to a continuous function, aka the minimum minimorum. If you disperse the idea of superimposed random noise than all these ups and downs translate to a polyonium of higher degree not consistent with the exponential rise of cancer proliferation. Try to smooth out your data with a moving average of 2 or 3 and you will get a better picture of your time series.

Justfor_ profile image
Justfor_ in reply to Justfor_

Since you have done statistics and prediction in your professional life, no need to add comments to the following raw and processed data. I am sure you will see my point.

Raw data:












Regression coefficients: Rlin=0.9277, Rlog=0.9297

Log Doubling Time (assuming equidistant 3 month intervals): 15.7 months

Filtered data by Moving Average of two:











Regression coefficients: Rlin= 0.9785, Rlog=0.9644

Log Doubling Time (assuming equidistant 3 month intervals): 14.2 months

Filtered Data by Moving Average of three:










Regression coefficients: Rlin= 0.9861, Rlog=0.9702

Log Doubling Time (assuming equidistant 3 month intervals): 14.0 months

Hell keep testing so you won't miss all of that anxiety....Good for your toenails.........

Good Luck, Good Health and Good Humor.

j-o-h-n Sunday 03/14/2021 6:16 PM DST

The chance of getting an aggressive recurrence with PSADT < 9 months is extremely small if you are 10 years post RP. There's a rather small but nonzero chance of a more indolent recurrence that might well never need treatment.

My JH onc advised they no longer use ultra sensitive as it offers no further treatment value and serves more to create anxiety. I'm sure there are other opinions like my surgeon in Boston who still uses it.


Mostly for Christopher_E and OldTiredSailor, or anyone else interested…wanted to put up my uPSA numbers and add to the discussion. I am getting in the same boat with you guys, seemingly. Am trying to decide next steps as I cross over the .10 marker. Would like to defer SRT if possible—at age 72 am still thinking I can outlive a recurrence and die of something else. That way, would not face possible urinary/abdominal S.E.s and possible, but unlikely, longer term secondary cancer. Am in very good health and would like to match my grandmother and her sister, who passed at 103 and 104, respectively.

I had RALP in 2012, 3+4=7, one focal 3mm margin that Epstein rated at 3+3=6. Of the nine years since RALP, first five were uPSA undetectable, last four have seen slow, kinda steady rise from .02 to .08 (I think). All tests used LabCorp; tried to control well by same time of day, always fasting, off biotin for 3 days before, similar hydration. Guess you could ballpark my doubling at about 2 years, give or take—haven’t run the numbers/calculated recently--I think my trend is picking up steam.

I put a chart of my numbers here:

I have been monitoring mostly every month or two, recently tried about three months. A couple reasons for this frequency:

(1) am a numbers guy and just wanted enough trend to feel comfortable taking next step, when I thought it was needed;

(2) have been off and on TRT during the period of rising uPSA, and wanted to see if it was affecting my PSA, regardless of “saturation theory” concept. “Trust, but verify” I always say.

(3) live in a small community where it is difficult to get appointments with good docs, e.g., took nearly a year to get in to see an endocrinologist on some related matters; was concerned about getting a sharp PSA rise and not being able to react quickly enough to it, if I was only doing the standard yearly PSA my uro wants.

(4) It is true that some studies have recently indicated that guys like me may be able to put off SRT for a while longer. However, this knowledge wasn’t in place three to five years ago. I have run across a number of guys whose docs, from well known institutions, got them on the SRT radiation bed before they even hit .10 or somewhat over that. Add into this the new PSMA scans and other promising approaches to find and deal with lesions at lower and lower PSAs. So, it seems very unsettled what guys like us should do;

(5) Regardless of my trend, not sure I am constituted such that I could get on board with the “PSA monitoring can be stopped now with an extremely low risk of subsequent BCR”. Yeah, ok, I’m a worrier. Or, maybe, I just want to follow the data where it leads in my case.

(6) Now, having gone ten weeks before most recent test, wouldn’t you know it, I see a big jump. Won’t know if it is valid until a couple more tests. Am wondering if the two Pfizer shots I got in Feb. might have something to do with it? Due to inflammation (I apparently do still have a bit of margin tissue in place)? Or if I had covid in last year and didn’t know it? This study kinda talks around the topic:

“Currently, there is no evidence on the effect of SARS-CoV-2 on prostate-specific antigen (PSA) levels (Fig. 1); the literature is sparse on the effects of any systemic viral infection on PSA, including other CoVs and influenza viruses. A study in a cohort of young military patients affected by infectious mononucleosis (the Epstein-Barr virus from the human herpes family) showed a higher likelihood of a PSA rise than controls, but the absolute change was small (only 5.9% with a rise of >0.50 ng/ml). They also found that patients with nonspecific systemic viral infections were more likely to have a PSA rise than controls, but only >0.2 ng/ml [33]. In addition to the specific viral prostate infection, the effects of a viral immune response with systemic inflammation may affect local vascular permeability or prostate epithelium, resulting in PSA elevation. Thus, a systemic viral infection may produce a small rise in PSA in PCa patients in active surveillance or watchful waiting programs, but no such inference can be made on post–radical prostatectomy patients.”

Justfor_ profile image
Justfor_ in reply to Robert_san

I have run your numbers for 4 time frame starting from:

A-> 7/19, B-> 10/19, C-> 3/20 and D-> 9/20, all up to 3/21.

For all 4 of them, I have done linear and exponential curve fitting for deriving linear and log doubling times (DT) as well as their relevant regression coefficients (R).

The results presented as a sequence of linDT, logDT/ linR, logR are as follows:

A: 21.2, 23.2 months / 0.792, 0.802

B: 18.3, 19.8 months / 0.791, 0.799

C: 15.3, 17.4 months / 0.676, 0.655

D: 8.0, 9.4 months / 0.922, 0.921

The prime take away from this drill is that you have to resume monthly PSA counts in order to verify the latest PSA value which increasingly weights-in to DT with shrinking time frame.

Two additional observations, IMO, of importance:

1) Apart from time frames A,B the linR of time frames C,D are marginally greater than the respective logR. This is NOT typical of a recurrence pattern.

2) Also, in a typical recurrence pattern, linDT > logDT. This does NOT hold true for any examined time frame of your data.

Robert_san profile image
Robert_san in reply to Justfor_


Thank you for lending your expertise to this discussion, and my numbers specifically. Have you, btw, participated in analyzing data or writing up for medical studies or clinical trials? Seems you would be invaluable for that.

Perhaps one idea for the guys posting here, as regards use and frequency of uPSA tests, would be to stick to some reasonable testing interval, like three or six months. If there seems to be a rising trend, or a decreasing trend (we can hope!), then one could drop down to monthly testing for a while and track more closely to try to see what’s going on. That was what I was trying to do with my “on TRT” and “off TRT” stretches.

We all need to be aware of other things possibly affecting uPSA testing, if we choose to do uPSA. Wouldn’t be surprised to hear of the covid vaccines affecting the uPSA test itself in some manner. But we won’t see publications on that for a while.

Also, in my case, at beginning of 2021, I did ramp up my weight-lifting, to almost a daily pattern. I also starting using creatine monohydrate supplements--the standard dose on the label--more regularly; was just haphazard use before. I note there are some studies going on about creatine use in prostate cancer patients, so will be personally interested in the results from those.

“III. To examine associations between creatine supplementation use and changes in biomarkers linked with prostate cancer progression (prostate specific antigen [PSA] and inflammatory markers).”


Bottom line for me is that using the uPSA seems helpful as a post prostatectomy patient, if I can just collect the data and not freak out about it each time. May turn out useful or not, long-term. Don’t know. As least I will have the data.

Justfor_ profile image
Justfor_ in reply to Robert_san

I am a retired el. engineer and have dealt with measurements, employing equipment bearing a price tag from less than 100 Euros to well over10k Euros, for almost 50 years. You wrote: "Trust, but verify”. I say: "Don't trust until verified".

With the manufacturing quality standards of our days the equipment is seldom the root cause of an erroneous measurement. Most of the time it is an operational glitch. So, I check, double check as standard and if there is any unanticipated deviation, triple check.

I will probably be in position to give some tangible info on your query regarding COVID vaccine and PSA interaction. I am scheduled for vaccination during early April. Prior to it, I will take a PSA test and after some time another one. My PSA follows a textbook exponential curve so, if there is a significant deviation either way, I will surely notice.

Robert_san profile image
Robert_san in reply to Justfor_

Will be watching for your April numbers, if you would be so kind as to share them then.

Btw, read your post last month on RYR, possible interaction with PSA, and the gross lab error you got as part of that research. Want to mention that in Nov., I got a LabCorp result of ".306", which you now know was way out of trend for me. I retested in Nov. using Quest (first time using them) on a two-digit uPSA they offered, and result was ".05", well within trend for me. I tried to report the faulty LabCorp reading, to no avail (no interest, and no refund). Health care has other concerns these days (rightfully so). Anyway, had I not had some uPSA trends, can you imagine what the ".306" would have done to my anxiety level? But I called "bs" on it pretty quickly. Cannot imagine how freaked I would have been had I been using standard PSA tests, as my uro recommends, and had gotten a "<.1" in 2019, then a ".3" in 2020!

Thanks again for your insight.

Justfor_ profile image
Justfor_ in reply to Robert_san

This poor soul listened to his uro and now is a bit puzzled:

You may also like...