Postop pathology to ChatGPT & asked about BCR. Anyone done this?

Posted by esperling @esperling, Dec 14, 2025

I used ChatGPT for the first time and input my post-op pathology details into it and then asked about recurrence. To my surprise, it stated that for men with “several high-aggressiveness factors” like mine,I.e.,
* Seminal vesicle invasion (pT3b)
* Multiple positive surgical margins
* Cribriform + Intraductal carcinoma
* Gleason 4+3
* Decipher score 0.89 (high-risk)
I should consider “ultra early” radiation therapy. It went on to state that a Sustained detectible PSA as low as 0.03-0.05 ng/mL should be used as a threshold to initiate conversations with your oncologists about radiation early salvage treatment and possibly, ADT. It specifically said,
“ PSA 0.03–0.05 ng/mL and rising (confirmed on two tests)”

And emphasized,
“Do NOT wait until PSA reaches 0.1. Patients with these features progress quickly once microscopic recurrence begins, and outcomes are best when salvage radiation is given at very low PSA levels.”

Referencing the AUA/ASTRO/SUO 2024 Guideline:
* Recommends early salvage radiation when PSA first becomes detectable.
* Strongly favors PSA ≤0.2, but many experts recommend ≤0.1 or even ≤0.05 for high-risk patients.
*
2. Multiple large studies
Consistently show:
* Best long-term cancer control happens when salvage RT is delivered with PSA ≤0.05
* Outcomes are significantly worse once PSA rises above 0.1
* Waiting until 0.2 (the “traditional” definition of recurrence) is now considered too late for high-risk patients.

It went on to explain that while the broad consensus guidance still defines biochemical recurrence at PSA >=0.2 ng/mL, the studies this was based on did not include a statistically relevant number of higher risk patients or account for their specific tendency for greater metastasis velocity.

It certainly puts new emphasis for me on watching my PSA very very closely and I will be discussing with my doctors on the coming weeks depending on the outcome of my next PSA (blood draw on Tuesday 12/16/25).

Interested in more discussions like this? Go to the Prostate Cancer Support Group.

Adjunct radiation does make sense for people that have very serious cases of prostate cancer. Here is a report on recommendations for early radiation from very highly respected doctor.

You are both PT3B and a high decipher, which is what are required in order to say that you should get this treatment. The article below really covers it. Definitely agrees with what AI told you.

Dr. Efstathiou concluded as follows:
* Early salvage radiotherapy is favored over adjuvant radiotherapy in most patients
* Consider adjuvant radiotherapy in otherwise fit, motivated, very high-risk patients with ≥2 of the following risk factors:
* pT3b-4
* Gleason score 8-10
* pN+ Lymph node Metz
* Decipher score >0.6
* In high-risk patients, use lower thresholds to initiate ‘ultra-early salvage or adjuvant-plus’ radiotherapy
* If giving adjuvant radiotherapy, it implies high-risk disease. Thus, Dr. Efstathiou would recommend treating the prostate bed and pelvic lymph nodes, in addition to short-term versus long-term ADT, depending on risk factors
* May consider genomic classifiers or artificial intelligence tools to help with informed decision-making
* The goal is to avoid (or delay) radiotherapy in those who we can, without missing a window to cure patients who are guaranteed to recur

Here is a link to the article supplied by @surftohealth88 originally
https://www.urotoday.com/conference-highlights/apccc-2024/151546-apccc-2024-debate-how-to-best-manage-a-fit-patient-with-high-risk-localised-and-locally-advanced-prostate-cancer-how-to-select-patients-for-adjuvant-therapy-after-radical-prostatectomy-and-how-to-treat-them.html

REPLY
Profile picture for jeff Marchi @jeffmarc

Adjunct radiation does make sense for people that have very serious cases of prostate cancer. Here is a report on recommendations for early radiation from very highly respected doctor.

You are both PT3B and a high decipher, which is what are required in order to say that you should get this treatment. The article below really covers it. Definitely agrees with what AI told you.

Dr. Efstathiou concluded as follows:
* Early salvage radiotherapy is favored over adjuvant radiotherapy in most patients
* Consider adjuvant radiotherapy in otherwise fit, motivated, very high-risk patients with ≥2 of the following risk factors:
* pT3b-4
* Gleason score 8-10
* pN+ Lymph node Metz
* Decipher score >0.6
* In high-risk patients, use lower thresholds to initiate ‘ultra-early salvage or adjuvant-plus’ radiotherapy
* If giving adjuvant radiotherapy, it implies high-risk disease. Thus, Dr. Efstathiou would recommend treating the prostate bed and pelvic lymph nodes, in addition to short-term versus long-term ADT, depending on risk factors
* May consider genomic classifiers or artificial intelligence tools to help with informed decision-making
* The goal is to avoid (or delay) radiotherapy in those who we can, without missing a window to cure patients who are guaranteed to recur

Here is a link to the article supplied by @surftohealth88 originally
https://www.urotoday.com/conference-highlights/apccc-2024/151546-apccc-2024-debate-how-to-best-manage-a-fit-patient-with-high-risk-localised-and-locally-advanced-prostate-cancer-how-to-select-patients-for-adjuvant-therapy-after-radical-prostatectomy-and-how-to-treat-them.html

Jump to this post

@jeffmarc thanks again for this reference. I’m convinced this is the route I will be going down. I just want to get a confirmed PSA reading and that happens this week.

REPLY

Totally agree with ALL recommendations. I think there is evidence - even without AI - that you should proceed to adjunct radiation treatment/ADT. Best,
Phil

REPLY
Profile picture for heavyphil @heavyphil

Totally agree with ALL recommendations. I think there is evidence - even without AI - that you should proceed to adjunct radiation treatment/ADT. Best,
Phil

Jump to this post

@heavyphil thanks Phil. I’m getting my blood pulled today for post op PSA#2. The results will be very telling on what’s next.

REPLY

I got similar advice from ChatGPT just last week (also one of the few times I've queried that source), which I take with a large grain of salt but which will guide my questions at next meting with the doc at Mayo on 2/3/26.

My biopsy on 8/26/25 showed Gleason 4+3, intraductal, large cribiform & extracapsular extension. My radical prostatectomy on 10/29/25 (with several pelvic lymph node removed) showed Gleason 4+5, intraductal & extracapsular extension. But surgery showed negative margins and no cancer found in lymph or seminal vesicles. I had a PSA of 8.4 before surgery, and PSA of 0.23 thirty days post surgery. I assume the 30 day number is residual. Will have 90 day post surgery PSA on 2/3/26 & will discuss adjuvant radiation vs. very early salvage with my surgeon at that time. Will also try to get appointment with an oncologist to get a wider perspective.

I talked with a local radiologist who said he does not recommend adjuvant radiation & said I should get back to him if two PSA readings of 0.2 or above. ChatGPT said early salvage at 0.02 PSA (So, some conflict between 0.2 and 0.02 threshold, which could just be someone's typo). Assuming ChatGPT is at all correct, I am considering adjuvant radiation (if medically approved) out of concern that there might be significant delay between a detectable PSA and commencement of very early salvage. But early days.

REPLY
Profile picture for tomsaville @tomsaville

I got similar advice from ChatGPT just last week (also one of the few times I've queried that source), which I take with a large grain of salt but which will guide my questions at next meting with the doc at Mayo on 2/3/26.

My biopsy on 8/26/25 showed Gleason 4+3, intraductal, large cribiform & extracapsular extension. My radical prostatectomy on 10/29/25 (with several pelvic lymph node removed) showed Gleason 4+5, intraductal & extracapsular extension. But surgery showed negative margins and no cancer found in lymph or seminal vesicles. I had a PSA of 8.4 before surgery, and PSA of 0.23 thirty days post surgery. I assume the 30 day number is residual. Will have 90 day post surgery PSA on 2/3/26 & will discuss adjuvant radiation vs. very early salvage with my surgeon at that time. Will also try to get appointment with an oncologist to get a wider perspective.

I talked with a local radiologist who said he does not recommend adjuvant radiation & said I should get back to him if two PSA readings of 0.2 or above. ChatGPT said early salvage at 0.02 PSA (So, some conflict between 0.2 and 0.02 threshold, which could just be someone's typo). Assuming ChatGPT is at all correct, I am considering adjuvant radiation (if medically approved) out of concern that there might be significant delay between a detectable PSA and commencement of very early salvage. But early days.

Jump to this post

@tomsaville
You did have the PSA test a little early, but I would not want to wait until February For the next test. Normally the first test after surgery is done around seven or eight weeks. At that point PSA should be undetectable. If it’s above .2 that’s a problem since there’s something hanging around besides the prostate itself. You did have clear margins so it sounds like it’s not that specific area but your first PSA is definitely a little high. A PSA test right now would be about the right time to find out whether or not you really are undetectable. That could really give you an answer as to whether you need further treatment. Waiting till February could be a mistake if your PSA is rising. Ask your doctor about getting the next test now, not in February.

It is possible you could get a PSMA pet test To see if there’s spread somewhere else in your body. If the next test is higher, the pet test would Give real useful information.

ASCO American Society of Clinical Oncology, the world's leading professional organization for cancer doctors doesn’t recommend salvage radiation at .02 but at .2 it is highly recommended.

From Ascopubs about what PSA to do salvage radiation.

≤0.2 ng/mL:
Starting at this level maximizes disease control and long-term survival. Patients treated at PSA < 0.2 ng/mL achieve higher rates of undetectable post-SRT PSA (56-70%) and improved 5-year progression-free survival (62.7-75%).
Delaying SRT beyond PSA ≥0.25 ng/mL increases mortality risk by ~50%.

0.2–0.5 ng/mL:
Still effective, particularly for patients with low-risk features (e.g., Gleason ≤7, slow PSA doubling time). The Journal of Clinical Oncology recommends SRT before PSA exceeds 0.25 ng/mL to preserve curative potential.

0.5–1.0 ng/mL:
Salvage radiation remains beneficial but may require combining with androgen deprivation therapy (ADT) for higher-risk cases.

This article discusses the above;
https://ascopost.com/news/march-2023/psa-level-at-time-of-salvage-radiation-therapy-after-radical-prostatectomy-and-risk-of-all-cause-mortality/

REPLY

UPDATE: As the originator of this topic posting, my Ultrasensitive PSA result received on 12/16/25 was “< 0.006 ng/mL” - that is below the lab test detection limiting 0.006 ng/mL. So that was a great early Christmas present for me! It means at least for the present time, the cancer is essentially in remission - with positive margins, Gleason (4+3), SVI, EPE, IDC and Decipher of .89, I will continue to very very closely monitor PSA using the ultrasensitive test. I will NOT be waiting until my PSA reaches 0.2 to act - I will be reacting much earlier - two detected readings, where level is increasing will be enough for me to start salvage treatment. I’ve already consulted with radiation and medical oncologists. I am visiting with my urology oncologist surgeon At Johns Hopkins on 1/06/26 and will discuss my current prognosis and future response timing etc.
Happy New Year and continued health and remission for all in this battle with Prostate Cancer.

REPLY

Yes, AI can be used as a "tool" and as valid information that can be used as talking point for future discussions with your doctor.

As you saw at the end, every case is different and even you with all those "bad" feature so far had great results.

There is just no way that one can predict with 100% certainty what will happen or not happen in the future. There are patients who have low gleason and negative margins who have BCR in 2 years and patients with high gleason and EPE and "all that jazz" and do not have BCR in 7 years or more so one really has to take one day at a time and just take all of the precautions to discover BCR early and treat it early for the best possible results.

My husband is high risk patient and we decided to have PSA tested every month for the first 6 mos. even though our doctors do not think it is necessary. We go to WalkInLab and results are the same as are in the hospital where he had RP. His hospital measured PSA as undetectable if it is less than 0.014. Since he is high risk his MO suggested paying attention if PSA ever reaches 0.05 and than doing treatment at PSA 0.1 and he was opposed (as was a surgeon) to adjuvant radiation therapy in our case. We intend to follow his advice.

Wishing you forever 0.006 PSA !!!

REPLY

@tomsaville and @esperling, I agree with @surftohealth88 that AI tools like ChatGPT can be useful tools to start researching. Fact checking is key. AI tools do not replace human judgment or oversight. Any text, image, or video generated by AI should be used only as a starting point, not as verified information. It may contain inaccuracies, biases, and other problems. Generative AI tools can sometimes generate plausible-sounding answers that are wrong.

Here's further guidance on using AI wisely (and safely).
- What is Generative AI? What does this mean on Mayo Clinic Connect? https://connect.mayoclinic.org/blog/about-connect/newsfeed-post/what-is-generative-ai-artificial-intelligence-what-does-this-mean-on-mayo-clinic-connect/

Did you also know that free AI tools use different levels of data for free accounts vs paid accounts? Ugh!

REPLY

Thank you Colleen for those wise precautionary statements! The reality is that no man or machine is omnipotent and we must carefully weigh and filter all information we gather and validate it against our already accumulated knowledge regarding our cancer condition and treatment plan. We each only have the one life to work with and making assumptions or blinding believing anything on first glance is foolish at best and deadly at worst. Thanks for the comment response!!
PS - my second post-op ultrasensitive PSA Result on 12/15/25 was below the lab’s detection limit (< 0.006 ng/mL). My wife and I could not have asked for a better Christmas Present!

REPLY
Please sign in or register to post a reply.