PSMA

Tips from Nik Hanson: Basically a CliffsNotes on PSMA PET interpretation

I just finished a PSMA PET/CT and Theranostics conference. This is long, so I put it into bullet points, but I think it is invaluable if you are reading these studies and want to save yourself waffling about certain things and will give you the next step if there are questionable lesions and hopefully save you from missing disease by only looking a the PSMA activity as there is other important clinical information on these scan. This should be useful for people at Tumor Board who don't read these so you don't send people down the wrong road, or rec a biopsy in an inappropriate scenario. I wanted to share some salient points to help us be smarter than the people ordering the scans. Please note that we are using 2 different tracers at the office and the hospital and the scanners are VERY different so definitely pay attention when reading and comparing scans from different sites.

• Isolated rib lesion with activity below liver can essentially be ignored. If you call a met for this and there is no CT correlate, we won’t be able to biopsy and after this conference I have discovered that almost NO academic centers do rib biopsies for bony lesions unless there is a significant soft tissue component so we can safely say no to these low yield procedures. Often the CT will show a lucent lesion with rim sclerosis or other non-aggressive features. These can be followed if their SUV is not high (8+? There is no consensus or standardization on SUV)
• Perirectal nodes are definitely in the drainage pathway for prostate cancer as discovered as a result of this new technology and should be described.
• Contact of primary tumor to capsule for extra capsular extension. This will be hard to differentiate and probably over called by us at the hospital. It will be easier on the digi pet/ct. Basically if you can’t see a thin rim of normal uptake around a peripheral lesion in the gland, it is acceptable to raise concern for extra capsular extension. There may be microscopic disease which we can’t see and I think it can change the radiation treatment fields if Nancy Reyes asks about it. We can’t be definitive, but can raise the question and let her figure out how to treat.
• SUV > 9-19... Positive for prostate cancer almost universally in the context of these exams, even if in an isolated rib lesion. They will often do SBRT without a biopsy in this situation.
• SUV 2-8... Careful evaluation of the pattern of disease and suspicion for Mets (I won’t go into details about primary staging scans for high risk disease vs PSA return in a post prostatectomy patient, but these things matter)
• More bone positive lesions with F-18 tracers (mostly European and Australian experience, not as big a deal with pylarify) These are often inflammatory and can be a problem, again if it is SUV. 9+, I would call cancer/met. If it is 2-8 and an isolated lesion, you may have to hedge and can favor met vs degenerative change based upon presence of CT changes or not and clinical factors.
• SUV comparison across scanners and tracers is not valid for treatment response. We are using Gallium tracers at the office and F18 tracers at the hospital. Even with the same tracers, SUV calculations will vary based upon the scanners. If anyone asks if the SUV Max or Mean is clinically relevant I would just say that this question is still up for debate, but a single lesion that goes from 9 to 15 is meaningless at this point and especially so on different scanners.
• If FDG is hot and PSMA cold -> bad prognosis, not a good candidate for Pluvicto, this should be a rare case for us as we should probably not end up with FDG scans to compare unless the patient was initially staged before they knew the path. Neuroendocrine differentiated tumors and some aggressive pathology (cribriform and intra ductal variants can be low on PSMA expression and can have primarily lytic Mets)
• If you see PSMA positive disease (activity greater than liver visually) you MUST look for lesions that are PSMA negative as these will not respond well to Pluvicto. (THIS IS PROBABLY THE BIGGEST TAKE HOME POINT, PART OF THE VALUE OF THE STUDY IS IDENTIFYING PSMA NEGATIVE LESIONS AS THIS CAN CHANGE MANAGEMENT, ESPECIALLY IN THE LIVER/SOLID ORGANS) sorry to yell there
• Late stage disease, look for skull and brain Mets as these can be clinically significant.
• Look closely at solitary lung lesions, a more common than expected site of prostate Mets discovered as a result of this imaging study, probably need biopsy to differentiate met from second primary
• Inguinal LNs without deeper pelvic LNs are extremely uncommon site of nodal spread and should be looked at with a VERY skeptical eye.
• If you see obturator or int iliac or peri-rectal nodes, can probably assume the next group of nodes have microscopic disease that we can't resolve with even these sensitive exams. Important in case Nancy Reyes asks about treatment planning or a urologist is planning to be curative with a LN dissection.
• Lots of accessory salivary tissue anterior to parotid, around the pharynx. If no other site of disease, this can be essentially ignored.
• Visceral disease is more common than thought, and important prognostically. Mention any solid organ disease and whether or not it is similar in uptake to the primary disease (for primary metastatic work up in high risk disease where you see the lesion in the prostate is significantly advised for the tracer).
• Peritoneal disease is real, look for it! Thankfully it often responds well to systemic treatment.
• Degenerative change and inflammatory processes of bones can be positive.
• Meningiomas and fibro-osseous lesions can be hot. The CT scan often answers the met question. For meningiomas, contrast enhanced MRI should answer the question. Ribs are hard to evaluate (see above) Contrast MRI may also be useful for questionable bone Mets in long bones and spine/pelvis.
• Used for staging of high risk disease, pay attention to local extent of disease, they stage up if seminal vesicle involvement. Probably doesn't change the surgical plan, but important to note prognostically.
• https://recip-criteria.com/ is a training resource to make you understand this staging system. There are many staging systems right now. UCSF and UCLA are trying to create visual scores to save people from measuring anything (ie, low intermediate and high scales instead of measuring SUVs). There are free dosimetry programs and I will see if we can get them for us as they can segment and count all the SUVs for us if we become facile with using them. Great for tumor board and if we ever do trials they will be part of the reporting I am guessing.
• SPECT CT energy wings, probably not relevant to us clinically unless we start doing post treatment scan on a SPECT CT, ignore this for now.

I attached an article with the PROMISE 2 criterion if anyone is interested. I don’t know if this will be used in the future, but it helps clarify what the academics most current thinking is about these scans. We do not use liver directed tracers, so you can ignore those, and Pylarify is on the low end of the spectrum for artificial bone uptake when compared to the European and Australian tracers, but it still definitely has problematic bone uptake requiring a close look by us.

Thanks for your time, this was as short as I could make it and cover everything. As we all become more facile reading these scans, I think they will take less time than the average FDG PET, and given the rapid imaging time, they are great on our new Digi PET.

Tips round 2: Adding some new insights from the Soc of Nuc Med people on the East Coast this time…

Another long one, apologies. I assume most people don’t know much about prostate cancer treatment (I didn’t until about 6 months ago and I am learning something new every day) so this stuff may seem basic if you are already familiar.

1. If our Rad Oncs start moving into the modern era, you may see a non-enhancing, non PSMA avid substance between the rectum and the prostate. This is hydrogel and separates the prostate from adjacent rectum to allow boost doses of radiation to the margin of the prostate without harming the rectum. Just in case you see such a thing, it can also show up on MRI. https://barrigel.com/hcp/barrigel-images

2. We should be using Metal artifact reduction on our attenuation correction scans, I will talk to Becca and see if this is available on the new scanner. If you see crazy artifact from hip arthroplasty, please let me know and I will see if Becca can apply this fix (if we have it).

3. ~5-10% of Prostate CA is PSMA negative. It is VERY important to check for LNs on the CT. If you see questionable LNs, but no PSMA avidity, Fluciclovine (Axumin) PET is still available at the hospital and can help with this situation if it will change management. In this setting, they are not likely to benefit from Pluvicto and PSMA PET will not be useful for follow up. Please mention this in the impression if you see it. We can also go for FDG PET if Axumin not helpful. FDG positive disease with no PSMA avidity is a poor prognostic indicator of more aggressive mutated disease and can be seen after extensive prior chemo and hormone directed therapies.

4. Androgen directed therapy (ADT) can suppress PSA and initially cause a flare effect on PSMA scans. Best to get the staging studies before starting ADT if possible. Non-PSMA avid disease in this setting is a very poor prognostic indicator and a good sign they won’t get full benefit of Pluvicto.

5. After Prostatectomy, it is critical to window PET portion of exam heavily to ensure no residual disease at anastomosis (the most common site of recurrence). Urine is your enemy and if there is high clinical suspicion (rising PSA) and no other disease to explain it, look VERY close to separate urine from residual/recurrent disease. This effect can also be seen on post TURP patients on an initial staging exam where some urine in the TURP defect can really cause confusion, but our CT should be good enough to figure it out.

6. Isolated splenic lesion with PSMA avidity are almost assuredly a hemangioma. Hemangiomas anywhere in the body can be PSMA avid. Multiphase CT/MRI can be helpful in liver and spleen. There should hopefully be obvious bone changes to confirm a hemangioma in vertebral body.

7. Isolated bone lesions, even in pelvis, scapula and especially in ribs, should be carefully looked at. If no obvious ground glass, or enchondroma or paget’s changes. Consider contrast enhanced MRI, or ask an MSK person if that is a ”don’t touch lesion”. Finding old priors can be very useful here for stability. Most investigators don’t mention isolated rib lesions without some other suspicious finding on CT or VERY high SUV (think >10 in a really small lesion, no hard rules here).

8. Pre-sacral LNs vs sacral ganglia - If touches the sacrum on the sagittal reformat, it is probably a ganglia. Paired symmetric stuff should be ganglia as well. Hopefully Becca is sending the sag and cor CT reformats to PACS to make this easier from our office. you would probably have to look at it yourself on the reformatted images at the hospital.

9. If isolated unilateral axillary LNs are hot, ask about vaccine history. COVID and others cause inflammatory changes which lead to PSMA avidity.

10. In setting of high risk disease with biochemical recurrence (Gleason >= 8 and/or PSA doubling of <12 months) post-prostatectomy; be very sensitive to disease outside the pelvis (liver peritoneum lungs etc)

11. Inguinal LNs with low grade PSMA avidity (SUV 2-8?, less than liver? no hard fast rule here) should be considered negative and not mentioned unless you see rectal/anal invasion (or it was invasive on path if post-prostatectomy) or there is high volume disease around the pelvic LNs. I can explain why verbally if anyone cares to know.

12. Always look for left supraclavicular LN (virchow node) and be sure it is not a stellate ganglia. https://www.researchgate.net/figure/Top-panel-68-Ga-PSMA-11-PET-CT-and-PET-CT-Left-cervical-ganglion-arrow-maximum_fig1_350934571

This can be an isolated finding or can be associated with mediastinal LNs. It is a bad prognostic indicator and will make salvage radiation in pelvis ineffective. This node, if you see it, takes patient from a curative treatment pathway to palliative, so definitely look for it and familiarize yourselves with the ganglia.

13. Pelvic LNs, even with minimal activity, are highly Specific for Mets in Biochemical recurrence patients (internal iliac, peri-rectal (esp post-prostatectomy, and obturator nodes). I would call these positive even with relatively low SUV in tiny LNs, like 3mm tiny. They will get salvage radiation for this, and it works pretty well. If you are above the acetabulum, or at periaortic iliac nodes with no intervening nodes, I would be more cautious about calling things as it puts the patients into a palliative systemic treatment pathway.

14. On staging scans, peri-rectal LNs are super important because the Uros don’t usually go after these when they do LN dissections. They will not even try if you don’t mention a hot node down there.

OK, if you got this far you are a champion. There are many other tracers coming out and there are actually already FDA approved markers for ER positive breast cancer and for RCC. I am in contact with the producers of these agents and when they are available in our area I will talk to everyone if they want to add them to our repertoire. PET FAPI tracers are already being touted as replacements for FDG PET in solid organs. FDG may become a lymphoma and infection tracer. There is already data on defining infected cysts in kidneys and livers in adult polycystic disease patients. Girod has already been around asking about evaluating aortic valves with FDG and he says that FDG has replaced Gallium and WBC imaging in the ID literature.

I remain excited about these opportunities and hope we become the owners of the radioactive side of theranostics in the Willamette Valley and can maybe get some research money as the cream on top of all of this effort.

Please feel free to as me about these studies and refer anyone who asks about theranostics to me if you are not sure how to respond to specific questions