Interpretation of PSMA PET results

[lokibear0803]

So my recent history is this:

Dec 2018: began ADT vacation (intermittent lupron/zytiga) after whole-pelvic EBRT

mid-Aug 2021: PSA has risen enough to resume ADT. Receive F-18 scan in Stanford; among 2-3 other very small loco-regional mets, we have an isolated liver spot, SUV 16.3, w/o CT correlate suspicious for metastasis (emphasis mine).

late-Aug 2021: resumed intermittent lupron/zytiga

Jan 2022: PSA undetectable; rescan, this time with Ga-68 at UC Health in Denver, with intention to track the liver spot. All lymph mets have either resolved or are greatly reduced in size, except for the liver spot — it has remained about the same, with SUV about 6. Conclude this is benign hemangioma, i.e the liver spot is not a concern.

March 2022: PSA still undetectable; start vacation from ADT again.

July 2022: PSA has stayed low until June, when a sudden spike motivates scan. Receive Ga-68 at Rocky Mtn Cancer Center in Denver; nothing is found — completely clear of mets. Liver spot has SUV 2.8, described as persistent but less conspicuous compared to January scan.

Sept 2022: PSA has risen to 3.52, and an incidental 17mm ”T2 hyperintensity” is seen in sacral bone during orthopedic MRI for non-cancer issues. Assuming PSA is now high enough to see PC mets, I receive F-18 scan at UC Health in Denver, and the radiologist report states this: “Redemonstrated radiotracer avid focus in liver … SUV 14.5 … can be further evaluated with MRI of the liver”.

So I’m trying to understand if the liver spot is yet again something to worry about.

Now, I have a bad habit of asking too many questions at one time, and I have about 5-6 questions here. So I’m going to break things up, and ask only 1-2 questions at a time. My first questions are these:

— I assume its apple to oranges to compare SUV levels b/w F-18 and Ga-68, am I right? Something about it’s a relative value, not absolute; and “it depends on what is background liver uptake as baseline”, which can vary from one tracer to the other. As you can tell, biology is not my strength.

— to a certain extent, it’s also apples vs oranges to compare SUV values b/w two scans of the same radiotracer (whether both are F-18 or both are Ga-68), if the scans are done at different locations and therefore using different scanners (which is my situation)…correct?

As always, many thanks in advance. I’ll follow this up with a summary of insights, and then post my next questions.

[tango65]

When you say F18 PET/CT is that a Pylarify, an Axumin , a 18F PSMA 11 PET/CT or a 18F FDG PET/CT?

[lokibear0803 in reply to tango65]

My bad, I should have said Pylarify in both cases. Thanks, tango.

[tango65 in reply to lokibear0803]

You are welcome. The main concern seems to be the persistent PSMA positive area in the liver with a SUV 14.5 in the Pylarify study. Discuss having a biopsy of this persistent finding.

What happened with the17mm ”T2 hyperintensity” in sacral bone ? Was PSMA positive or negative in the last Pylarify PET/CT?

[dhccpa in reply to tango65]

Could you write a short explanation of how those four scans differ? A pain, I know, but looking up scan info on the internet has usually led me to scan centers pushing scans.

Thus far, I've had a PET/CT bone scan, and three Axumin PET/CT scans.

Thank you.

[tango65 in reply to dhccpa]

This article has information about the different ligands and their use. Each one has references to other publication you can obtain if you are interested.

jnm.snmjournals.org/content...

[Seasid in reply to tango65]

Very useful. Thanks

[dhccpa in reply to tango65]

Thanks

[lokibear0803 in reply to tango65]

Thanks tango. For my own purposes, I carve out pieces of articles like this for my own knowledge base. Then, FWIW, I’ll share this here with anyone else who would like a summary vs pouring thru the whole writeup.

Note this article was written in 2017, and much of it is out-of-date; also please note that I’ve extracted (and paraphrased in some cases) only that stuff that is of immediate interest to me. You might read it and find other details for your situation; your read might also show that I’ve paraphrased inaccurately - in which case, please correct things for us:

——————————————————————————————

…if high suspicion of PC, mpMRI helps to rule out clinically significant disease and to guide targeted biopsy, although it can miss aggressive PC lesions.

…18F-FDG is the most widely used radiotracer in oncologic PET/CT imaging; however, only a minority of PC (i.e., aggressive, poorly differentiated, or undifferentiated) shows a high glycolytic rate, limiting its use.

…in Europe, radio-labeled choline derivatives (18F-fluorocholine or 11C-choline) were among the most commonly used PET tracers for PC imaging….high specificity of 95% but a poor sensitivity of 49%…11C-acetate and 18F-fluciclovine, among others, show superiority over choline derivatives.

…most widely used 68Ga-labeled PSMA ligands are 68Ga-PSMA-11 (68Ga-PSMA-HBED-CC) and theranostic agents 68Ga-PSMA- 617 and 68Ga-PSMA-I&T. 18F-labeled agents include 18F-DCFBC, 18F-DCFPyL, and 18F-PSMA 1007…18F has reduced blurring effects, longer half-life, and potential for centralized production and distribution, compared with 68Ga.

…68Ga-PSMA-11 shows superior SUV(max) and tumor-to-background ratios compared to 18F-fluorocholine, and higher detection rates compared to 11C-choline for lymph nodes and bone mets…68Ga-PSMA-11 PET significantly outperformed bone scanning b/c of both high sensitivity and high specificity on a patient and region basis.

…PSMA PET/CT has evolving role in radioligand therapy (RLT = Lu-177, etc), evaluating target expression and therefore potentially predicting response. A rare but potential limitation is absent or low PSMA expression (e.g., in visceral metastases) in advanced disease, which may be related to therapy-induced specific biologic subtypes (e.g., neuroendocrine differentiated PC).

…68Ga- and 18F-labeled PSMA ligands are excreted via kidneys = radiotracer uptake in kidneys and collected urine….18F-PSMA 1007 might have reduced urinary clearance within first 2 h after injection, potentially allowing for improved assessment of the prostate within this time window.

…pitfalls: PSMA ligand uptake: not exclusively specific for PC…increased PET signal has been seen in benign lesions as well as malignant diseases…can be solved in clinical context or by adding further imaging…also, absence of PSMA over-expression in primary tumor or its metastases in up to 10% of patients with primary PC or decreased PSMA expression in advanced disease.

——————————————————————————————

Now to finish a summary addressing dhccpa ’s original question, here’s what I have in my notes:

Pylarify = 18F-DCFPyL, mentioned above

Axumin: Some background: FDG was the most widely used scan type, but it relies on a particular biological process (Warburg effect, if you must know — the “high glycolytic effect” mentioned in article) that most PC cells do not undergo, so uptake was limited. Next came sodium fluoride (NaF), which binds to areas within bone that’s turning over….i.e. the end product of bone remodeling…but, NaF is able to image only bone and not soft tissue…to overcome no Warburg effect and bone-only sensitivity, C11-choline and F-18-labeled FACBC were developed…the latter has trade name Axumin…Axumin allows for fairly sensitive detection, but suffers from low specificity (mechanism of uptake => not specific to PC, rather it’s that of a rapidly dividing cell).

Which brought us to developing the PSMA-based scan technology, which are touched on in the above set of notes.

HTH —

[lokibear0803]

Thank you tango! Yeah, biopsy was discussed when we were uncertain last August, but the lesion wasn’t big enough — if you haven’t got a CT correlate, you can’t guide the needle. Or, something like that.

This report didn’t mention size; so I don’t know yet if it’s growing. If so, perhaps it has grown enough this time to biopsy. I have a question in to my MO already about this.

The 17mm hyperintensity was not seen. So, this is either not PC, or PSMA-negative, or some other type of tumor, or benign, or ….

[Tall_Allen]

What decision are you trying to make?

[lokibear0803]

It’s largely, for starters, a decision around whether or not to freak out.

I’d thought I’d gotten past my liver metastasis thing, but now I’m not clear.

I don’t expect to speak with MO or RO until four days from now. So I reach out to HU for perhaps some insight. I’m trying to ask questions one at a time so the insights can inform my next set of questions. Apologies if that makes my first post seem vague or directionless.

Generally, I’m concerned with the liver spot — is it real, or is it a false positive as was concluded in January. I know HU can’t tell me this; but I’m hoping to get educated enough to shape how I regard it, what questions to ask MO/RO on Tuesday, how many glasses of wine I need tonight, etc.

The angle I’m starting from is wrt the changes in SUV from Aug-2021 (16.3) to Sept-2022 (14.5). Both scans are Pylarify, but done at different clinics (Stanford vs Denver). So I’m first trying to get some education around this SUV metric:

— for this particular situation, is it meaningful to compare two SUV values from same tracer but different clinics? i.e. in both cases where my T has recovered: if we see a slight drop in SUV, can this be thought of as “stable”? This would be consistent with, among other possibilities, a hemangioma (as was concluded in January).

Now, although I had an ADT on-cycle b/w these two SUV measurements, the size of the lesion did not change from Aug-2021 to Jan-2022; now, looking from Aug-2021 to Sept-2022, I don’t yet have size for the latter; but it’s my hope that the two similar SUVs can be regarded as saying that the lesion has not changed in size — am I thinking correctly about this?

— in general, is there any utility in comparing SUV’s b/w a Pylarify vs a Ga-68 scan? This seems to matter b/c my Pyl vs Ga68 liver SUV’s differ by an order of magnitude. I’m thinking this is b/c SUVs are relative to liver background level, but the liver uptake of Pyl might differ from that of Ga-68. Is it correct that there are different baselines b/w different tracers, and if so how useful is it to compare SUV values from one vs the other?

Thanks for taking a moment, TA.

[tango65]

The SUV do not depend on the liver background. SUV of the liver has been used to "define" PSMA expression of the mets. In the Vision trial the mets had to have SUVs equal or higher than the liver.

Here they explain how is SUV calculated

radiopaedia.org/articles/st....

The lesion in the liver has a higher SUV than the liver, if it were PC it has a good PSMA expression.

Discuss having a MRI of the liver and also of the bone lesion to study if the bone marrow is normal. Because of these lesions perhaps you should restart systemic treatment until a definitive diagnosis.

If MRI studies do not resolve these issues, I would insist in biopsies of the 2 areas. Ultrasound can see liver lesions and I believe it could be used to guide the liver biopsy.

[Seasid in reply to tango65]

Wow. Thanks for the explanation.

[lokibear0803 in reply to tango65]

Got it. I conflated liver uptake as a trial baseline with the SUV formula (SUV calculation is independent of liver SUV, but trial eligibility, etc can have a dependency).

I was actually in this same situation almost exactly a year ago. Liver spot detected on Pylarify w/o CT correlate, followup MRI didn’t find it, nor did followup ultrasound. This time, I haven’t done the MRI/US followups just yet — I’m hoping our radiology department can compare the size to what it’s been both Aug-2021 and Jan-2022, and just conclude “yep, it’s still there, it’s still the same size…still a hemangioma”.

Regardless a return to systemic treatment is already on the schedule, as per PSA and its DT. Will revisit possibility for biopsy.

Tango, you da man. Thanks!

[lokibear0803 in reply to tango65]

So to follow up on your reply in the context of my initial questions, here’s my takeaway. Please correct me if I’m off-base:

First, I paraphrase the definition of SUV:

SUV = tissue radioactivity concentration (i.e. uptake per unit volume, units are mCi/mL) at a point in time divided by injected dose of radioactivity (e.g. how much 18-F or Ga-68, units are mCi) per kilogram of body weight (units of kg).

I’ve also seen it expressed this way:

ratio of activity per unit volume of region of interest, to the activity per unit, whole body volume

From discussion with my MO:

SUV avidity is somewhat correlated with size of met

On to my questions then:

…in both cases where my T has recovered: if we see a slight drop in SUV, can this be thought of as “stable”?

I want to knee-jerk with YES on this one…but I’m thinking the SUV has variability instead of remaining in the same general ballpark over time for a given site of uptake. I suspect it does, so my take here would be not necessarily.

…two similar SUVs can be regarded as saying that the lesion has not changed in size — am I thinking correctly about this?

This is actually a more quantitative version of the first question. So, here again I go with not necessarily, assuming again that SUV has noticeable variability at different points in time.

…is there any utility in comparing SUV’s b/w a Pylarify vs a Ga-68 scan?

Same issues. The formula for SUV, i.e. the math, doesn’t care which tracer is used; but if there’s variability over time for same region of interest, then not necessarily. I’m frankly hoping I’m wrong, since my two Pylarify’s were about the same, as were the two Ga-68’s, at different points in time, and I hope this means my liver spot is not growing or shrinking, either with or without ADT = it’s not cancer.

Regardless, if biopsy if possible, I do a biopsy

[tango65 in reply to lokibear0803]

I am not a nucleat medicine person. You need to discuss these questions with your MO who could reach nuclear medicine persons. I have done that a couple of times.

I can give you some guesses based in my comprehension of the subject.

SUV depends on the amount PSMA expressed by a lesion. I asume size is a factor along with the amount of PSMA expressed by each cancer cells.

The size of the mets are determined by the size measured in the CT scan images and not by the SUV.

I think 2 lesions having equal SUV may have different sizes, if a lesion is getting bigger with cells without PSMA expression and maintains a similar population of PSMA expressing cells than in a previous study. The CT scan will tell the size of the lesion.

My MO has compared SUV from Ga 68 with SUV from Pylarify studies.

I wish you the best of luck.

[lokibear0803 in reply to tango65]

Fantastic, thank you. These questions are all definitely coming up when I see MO on Tuesday. Meanwhile, I appreciate your educated guesses.

If I get some useful information from my MO, I’ll follow up here.

[tango65 in reply to lokibear0803]

Good, thanks.

[cesces]

Ask your doc how large is your liver spot and how large does it need to be to do a biopsy.

Then ask what doc at what medical center of excellence might have the tools and technique to do a biopsy at it's current size.

Then get in line to get t ticket to Mayo to get a second opinion from them.

Don't take no for an answer.

[Seasid in reply to cesces]

You are correct. Biopsy could be life saving.

[lokibear0803 in reply to cesces]

Agreed, and she is on that same page already. The Mayo idea is a good one — do you suggest Mayo b/c it’s the most excellent center of excellence?

I searched the other week for centers of excellence, and it’s seems the term has been used somewhat arbitrarily (“you betcha! we’re also a center of excellence!”). Is there some objective definitive list of the “official” centers of excellence?

Thank you, cesces!

[cesces in reply to lokibear0803]

They will have a team looking at you. It's harder to bs in a team environment.

It's harder to say something is impossible just because you are interested in doing something difficult or hard.

[dhccpa]

Great questions that I've had also.