Pelvic nodal RT in salvage prostate

[Radonc90]

I have a somewhat similar case but want to ask the group re "Rise in PSA and Value of Pelvic LN RT"...

I wonder who here would treat the pelvic nodes (in addition to prostate bed) in the setting below, which is very typical
in a radonc clinic...

------------

- 70 yoM s/p radical prostatectomy 2 yrs ago for Gleason 8, negative margins, 6 pelvic nodes negative. Now PSA slowly rising to 0.4 ---> 0.6 ng/ml level...

1. This ACR Appropriateness criteria was last updated in 2014:

https://acsearch.acr.org/docs/69400/Narrative/

"...The addition of pelvic RT to prostate fossa radiation is generally discouraged, although it may be appropriate in certain clinical situations (eg, absence of lymph node dissection, evidence of nodal involvement at prostatectomy or on imaging studies)...."

2. The SPPORT trial was reported here:

The ASCO Post

ascopost.com

"...At 5 years following treatment, freedom from progression (FFP) rates in the interim analysis group were 71.1% for PBRT alone, 82.7% for PBRT+ADT, and 89.1% for PLNRT+PBRT+ADT. The FFP rate was highest for the arm combining all three treatments (P < .0001). Freedom from progression (I think they meant "FFP end point") was defined as a PSA nadir of +2, clinical failure or death from any cause..."

---

Members don't see this ad.

 

[elementaryschooleconomics]

I wonder who here would treat the pelvic nodes (in addition to prostate bed) in the setting below, which is very typical
in a radonc clinic...

------------

- 70 yoM s/p radical prostatectomy 2 yrs ago for Gleason 8, negative margins, 6 pelvic nodes negative. Now PSA slowly rising to 0.4 ---> 0.6 ng/ml level...

Ah, nodal coverage in every site except head and neck - how I loathe thee:

1) Was his PSA ever undetectable? I take it from the wording of this case (only two years ago, PSA already 0.4) that he always had some level of post-op PSA?

2) When you say "slowly", what does that mean? Did that increase take place over a year? Was his post-op PSA 0.4 and this is his first rise thus far?

3) cT1c or was there SV involvement etc?

Regardless, for any salvage consult that comes my way, I always order a CT A/P and bone scan. I will try to get an Axumin scan, but sometimes I'll get pushback from insurance depending on the benefit manager (looking at you, any Medicare Advantage Plan, literally ever). I'll eventually switch over to PSMA but that machine isn't up and running in my hospital yet.

If there are any suspicious nodes on these scans, that makes things easier. Chunky regional nodes, I'll cover the consensus pelvic nodal levels. If absolutely everything is negative - I'll probably just zap the prostate bed. Because his PSA is >0.5, I'll add in ADT (or at least talk about it with him).

Again, I absolutely hate nodes. One of my partners will treat nodes in this situation almost every time. Another of my partners almost never treats nodes. Then, there's the question of dose, where the pendulum seems to keep swinging between 6600 cGy and 7020 cGy (or 7200 cGy if gross disease, which there isn't in your case, correct?).

So...yeah. Depending on his clinical history and the results of these scans, I would consider covering nodes, and since his PSA is above 0.5, I would add in ADT.

*insert Star Trek gif here with Shatner yelling KHAN except he's yelling NODES instead*

 

[evilbooyaa]

Moved to a new thread, no need to revive an old one with a separate question.

Gleason 8, negative margin, 6 nodes negative. I'm going to assume initial post-op PSA was undetectable.
First and foremost, why was patient not treated shortly after increase above 0.2 (if not climbing, but less than 0.2)?

PSA is now over 0.34 which showed bPFS benefit of adding pelvic nodal RT. Yes, arbitrary cut-off from 0534 is arbitrary, but was the median PSA for trial enrolees.

First I would get an Axumin or (preferably) PSMA PET/CT to look for nodal disease. If there is a target, boost as feasible based on nearby OAR tolerance. If the uptake is in the prostate bed, get an MRI to confirm it's a nodule. If it's a node, blast away to surrounding OAR tolerance, goal of as close to definitive dose to prostate as feasible.

Even if PET scan is negative, I would treat nodes.

Getting CT A/P and bone scan in this setting... is silly and mostly useless, but demanded by most insurances, likely. But, wouldn't really change my practice. Would still shoot for PET imaging. Maybe if there was a visible chunky node and I couldn't get PET imaging even with that finding.

 

[Chartreuse Wombat]

I have a somewhat similar case but want to ask the group re "Rise in PSA and Value of Pelvic LN RT"...

I wonder who here would treat the pelvic nodes (in addition to prostate bed) in the setting below, which is very typical
in a radonc clinic...

------------

- 70 yoM s/p radical prostatectomy 2 yrs ago for Gleason 8, negative margins, 6 pelvic nodes negative. Now PSA slowly rising to 0.4 ---> 0.6 ng/ml level...

1. This ACR Appropriateness criteria was last updated in 2014:

https://acsearch.acr.org/docs/69400/Narrative/

"...The addition of pelvic RT to prostate fossa radiation is generally discouraged, although it may be appropriate in certain clinical situations (eg, absence of lymph node dissection, evidence of nodal involvement at prostatectomy or on imaging studies)...."

2. The SPPORT trial was reported here:

The ASCO Post

ascopost.com

"...At 5 years following treatment, freedom from progression (FFP) rates in the interim analysis group were 71.1% for PBRT alone, 82.7% for PBRT+ADT, and 89.1% for PLNRT+PBRT+ADT. The FFP rate was highest for the arm combining all three treatments (P < .0001). Freedom from progression (I think they meant "FFP end point") was defined as a PSA nadir of +2, clinical failure or death from any cause..."

STADT + XRT to PB and pelvis

Pet (Axumin, PSMA) if possible to guide XRT delivery (SIB to LN if avid)

18F-fluciclovine-PET/CT imaging versus conventional imaging alone to guide postprostatectomy salvage radiotherapy for prostate cancer (EMPIRE-1): a single centre, open-label, phase 2/3 randomised controlled trial

Molecular imaging is increasingly used to guide treatment decisions and planning in prostate cancer. We aimed to evaluate the role of 18F-fluciclovine…

www.sciencedirect.com

 

[DoctwoB]

Moved to a new thread, no need to revive an old one with a separate question.

Gleason 8, negative margin, 6 nodes negative. I'm going to assume initial post-op PSA was undetectable.
First and foremost, why was patient not treated shortly after increase above 0.2 (if not climbing, but less than 0.2)?

PSA is now over 0.34 which showed bPFS benefit of adding pelvic nodal RT. Yes, arbitrary cut-off from 0534 is arbitrary, but was the median PSA for trial enrolees.

First I would get an Axumin or (preferably) PSMA PET/CT to look for nodal disease. If there is a target, boost as feasible based on nearby OAR tolerance. If the uptake is in the prostate bed, get an MRI to confirm it's a nodule. If it's a node, blast away to surrounding OAR tolerance, goal of as close to definitive dose to prostate as feasible.

Even if PET scan is negative, I would treat nodes.

Getting CT A/P and bone scan in this setting... is silly and mostly useless, but demanded by most insurances, likely. But, wouldn't really change my practice. Would still shoot for PET imaging. Maybe if there was a visible chunky node and I couldn't get PET imaging even with that finding.

Agree with most of the above. Given the high grade disease can cover the nodes. PSMA sadly isn't sufficiently sensitive to rule out nodal disease (see recent study about sensitivity in the preop setting being <50%), or rule out fossa disease. The main benefit would be if you get a PSMA and a few treatable bone lesions pop up, could delay time to ADT.

 

[elementaryschooleconomics]

Getting CT A/P and bone scan in this setting... is silly and mostly useless, but demanded by most insurances, likely. But, wouldn't really change my practice. Would still shoot for PET imaging. Maybe if there was a visible chunky node and I couldn't get PET imaging even with that finding.

I have been burned several times by this. It reminds me of the algorithm that you can't get an MRI for back pain without a basic xray first (is that still a thing? Maybe that's gone away).

There's a few private carriers which have allowed me to get a PET without the CT and bone scan but...it's not a guarantee. And it goes without saying, Evicore ABSOLUTELY won't let you get away with this.

 

[medgator]

I have been burned several times by this. It reminds me of the algorithm that you can't get an MRI for back pain without a basic xray first (is that still a thing? Maybe that's gone away).

There's a few private carriers which have allowed me to get a PET without the CT and bone scan but...it's not a guarantee. And it goes without saying, Evicore ABSOLUTELY won't let you get away with this.

They like spending extra money when it comes to radiology studies... Very paradoxical when you consider how they treat rad onc

 

[elementaryschooleconomics]

They like spending extra money when it comes to radiology studies... Very paradoxical when you consider how they treat rad onc

It's absolutely insane. Just let me get the Axumin scan!! I don't need the other things if I have the Axumin!!

 

[seper]

SPPORT abstract is the most compelling evidence to treat nodes in prostate Ca. So most people would treat nodes here.
What is the evidence that PSMA is not helpful in case like that ?

 

[thecarbonionangle]

What is the best paper to look at sensitivity and specificity of PSMA in the multiple preop/post op settings for ranges of PSAs?

 

[communitydoc13]

It's absolutely insane. Just let me get the Axumin scan!! I don't need the other things if I have the Axumin!!

Since EMPIRE-1 came out, we've been getting approval for Axumin in salvage setting pretty uniformly. Maybe regional?

 

[elementaryschooleconomics]

Since EMPIRE-1 came out, we've been getting approval for Axumin in salvage setting pretty uniformly. Maybe regional?

Probably - the patients in my area with Medicare Advantage plans often have one which is managed by Evicore, so...yeah.

 

[medgator]

Probably - the patients in my area with Medicare Advantage plans often have one which is managed by Evicore, so...yeah.

Evilcore, aim or healthhell.... It's such a growth industry

 

[maxxor]

Agree with most of the above. Given the high grade disease can cover the nodes. PSMA sadly isn't sufficiently sensitive to rule out nodal disease (see recent study about sensitivity in the preop setting being <50%), or rule out fossa disease. The main benefit would be if you get a PSMA and a few treatable bone lesions pop up, could delay time to ADT.

Can you link this specific paper?

I’m currently designing my psma petmr protocol (literally just got a request for one).

My fluciclovine scans come out superb. I want to see what the pitfalls will be for PyL.

 

[Radonc90]

I think this is the largest surgical series from Amsterdam...

The Predictive Value of Preoperative Negative Prostate Specific Membrane Antigen Positron Emission Tomography Imaging for Lymph Node Metastatic Prostate Cancer:

American Urological Association

"...Overall sensitivity, specificity, positive predictive value and negative predictive value of prostate specific membrane antigen positron emission tomography for the detection of pelvic lymph node metastases were 37.9%, 94.1%, 64.3% and 84.4%, respectively. The negative predictive value of prostate specific membrane antigen positron emission tomography in patients with intermediate risk prostate cancer was 91.6% (95% CI 86–97), compared to 81.4% (95% CI 77–86) in patients with high risk prostate cancer. When only assessing patients with <rT3 disease on multiparametric magnetic resonance imaging, 51/52 patients with intermediate risk prostate cancer had a true negative prostate specific membrane antigen positron emission tomography (negative predictive value=98.1%; 95% CI 94–100).

 

[DoctwoB]

I think this is the largest surgical series from Amsterdam...

The Predictive Value of Preoperative Negative Prostate Specific Membrane Antigen Positron Emission Tomography Imaging for Lymph Node Metastatic Prostate Cancer:

American Urological Association

"...Overall sensitivity, specificity, positive predictive value and negative predictive value of prostate specific membrane antigen positron emission tomography for the detection of pelvic lymph node metastases were 37.9%, 94.1%, 64.3% and 84.4%, respectively. The negative predictive value of prostate specific membrane antigen positron emission tomography in patients with intermediate risk prostate cancer was 91.6% (95% CI 86–97), compared to 81.4% (95% CI 77–86) in patients with high risk prostate cancer. When only assessing patients with <rT3 disease on multiparametric magnetic resonance imaging, 51/52 patients with intermediate risk prostate cancer had a true negative prostate specific membrane antigen positron emission tomography (negative predictive value=98.1%; 95% CI 94–100).

Key part of the stats here to note is PPV is much lower then specificity due to low prevalence. Based on high specificity you might think that “well it misses a lot, but if it’s there it’s accurate,” but if your nomogram gives a 10% chance of LN Mets there’s a very good chance it’s a false positive.

 

[maxxor]

Key part of the stats here to note is PPV is much lower then specificity due to low prevalence. Based on high specificity you might think that “well it misses a lot, but if it’s there it’s accurate,” but if your nomogram gives a 10% chance of LN Mets there’s a very good chance it’s a false positive.

I downloaded the paper, but it unfortunately does not give imaging protocols. They used 3 different tracers and 3 different types of imaging (PET/CT ACO, PET/CT + Contrast, ?PET/MR?). I'll need to dig more into this to figure out why it underperformed so much.

I don't recall seeing a table of false positives vs true positives. They state 56/434 PSMA PET positive for nodal mets and 95 with positive nodes on path. Naively, isn't that a 58.9% sensitivity?

I wonder if there's some other technical factor at play here. Maybe they used crappy scanners without resolution recovery reconstruction algorithms that have a reliable quantification threshold of 5-6 mm and missed all nodes below that. They don't give data so I have no idea. There is a difference in largest node size between the positive and negative PSMA scans, which makes me think there's something going on due to that.

I have been reliably finding 3-4 mm nodes with fluciclovine on my machine. I'll likely do the same scan parameters for the first few PSMA to see how they turn out.

 

[Chartreuse Wombat]

Key part of the stats here to note is PPV is much lower then specificity due to low prevalence. Based on high specificity you might think that “well it misses a lot, but if it’s there it’s accurate,” but if your nomogram gives a 10% chance of LN Mets there’s a very good chance it’s a false positive.

Agree.

Humans are horrible at intuiting the meaning of a test. Doctors are not immune. Conditional probability did not influence evolution.

Interpretation by physicians of clinical laboratory results - PubMed More than decades ago

The Inability to Calculate Predictive Values: an Old Problem that Has Not Gone Away - PubMed No better (maybe worse) in 2020