The NCCN PCa classification system vs new MMAI model

Posted by handera @handera, Dec 28, 2024

9,787 patients with localized prostate cancer from eight NRG Oncology randomized phase III trials, treated with radiation therapy, androgen deprivation therapy, and/or chemotherapy were evaluated in a multimodal artificial intelligence (MMAI) model for risk stratification.

The MMAI model, which integrated digital histopathology and clinical data, outperformed the National Comprehensive Cancer Network (NCCN) criteria by reclassifying 42% of patients, expanding the low-risk group and refining high-risk identification, as to a patient’s overall 10-year metastasis risk.

This approach would improve prognostic accuracy, reduce over treatment of low-risk cases, and ensure appropriate treatment for high-risk patients.

It holds promise for personalized care and is nearing clinical implementation pending FDA approval.

This new MMAI model has the potential to reduce BOTH over treatment and under treatment cases, upon PCa diagnosis.

This is an important development for the newly diagnosed. I hope it gets adopted quickly.
https://ascopubs.org/doi/10.1200/PO.24.00145

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

Very interesting. Thank you for sharing this.

REPLY

Thanks for sharing the study 😉.

Hopefully this will evolve and prevent under/over treatment for men.

It can be a slippery slope for me to read it because it's easy to think "maybe I've been over/under diagnosed" - obviously focusing on "over", because we all know how easy it is to grasp at straws and false hope. I have to remind myself that is's a preliminary study that could change things or still be proven inaccurate and that the path I have chosen is still a proven and recommended path and that's all I can count on for now.

REPLY
@survivor5280

Thanks for sharing the study 😉.

Hopefully this will evolve and prevent under/over treatment for men.

It can be a slippery slope for me to read it because it's easy to think "maybe I've been over/under diagnosed" - obviously focusing on "over", because we all know how easy it is to grasp at straws and false hope. I have to remind myself that is's a preliminary study that could change things or still be proven inaccurate and that the path I have chosen is still a proven and recommended path and that's all I can count on for now.

Jump to this post

Well…I hope this study doesn’t lead anyone down a path of “second guessing” past treatment decisions.

I posted primarily for the newly diagnosed and as information for those of us who have already moved down a particular path in our PCa journey, as something to encourage others who may soon be joining our “reluctant club”.

We all have had our PCa either over or under treated, to some degree….and men will continue to do so with or without MMAI.

PCa is not the common cold and once diagnosed permanent lifestyle changes are virtually inevitable…no matter what outcome results from our treatment decisions.

I believe one should to do as much research as one can tolerate and then choose a path forward that one is willing to own…no matter the final outcome.

We all know “would have…/should have…” thinking is wearisome and ultimately of no redeeming value.

New and better diagnosis tools and treatment methods will always be “just over the horizon”; therefore, learning decision contentment is just as important as learning about treatment options.

We live in a time where the vast majority of PCa men have a host of suitable treatment options and none of them will be the “perfect solution”.

That said, the prognosis for the largest percentage of PCa men is quite good and, even though it’s something we’ve all heard many times, counting blessings is more constructive than dwelling on disappointments.

REPLY

I just read the article in the link, which is complete, not just the abstract. To summarize, here's what I noted:
The standard guidelines for prognosis from NCCN were used as a standard to compare (and calibrate) an MMAI model for risk stratification (which is not disclosed.) Several phase III clinical trials were used as sources of data (in this last stage of clinical trials there are more people enrolled but large quantities of data are still being collected.) This article reports that the MMAI model increased prognostic accuracy, leading to the possibility of treating larger numbers of PC positive men less aggressively, but a smaller number of high risk PC positive men more aggressively.
Things to note: 1) This is a retrospective look at the data. The MMAI model was not actually used to make treatment decisions. 2) Since the MMAI model is proprietary, my assumption would be that the owners of the model are hoping to make money by selling this to PC positive men and their medical providers. Therefore, this might be viewed as part research and part marketing.
The NCCN guidelines in contrast have to back up in order to gain consensus in a larger open community. I assume that many of the experts who have a hand in formulating (and revising) these guidelines would believe they can more accurately predict outcomes than just by following the guidelines. The MMAI model is a way of anchoring one set of those experts' predictions and applying them more widely.
As a mental health professional, I can provide any number of diagnostic instruments to clients and now almost all of these can be scored by computer and a report generated (especially if they are based on client self-report.) However, I know that these results, while valuable, are not as good as my own expert assessment using the same data together with direct clinical assessment. Still, diagnostic testing is available very quickly with little effort on my part and I also have many colleagues who are less expert than I now am in assessment. (I also was a clinician for many years while I was less expert.) I believe the same will be true as MMAI models become common throughout medicine.

REPLY

I couldn't find any guideline in NCCN about Intermittent ADT. I understand there are trials that validate Intermittent ADT is not inferior to continuous ADT. How long it takes typically to reach a consensus amongst participating centers of NCCN to come up with a guideline for each individual situation? Anyone knows?

REPLY

Md Anderson had a trial that said intermittent was not inferior to continuous. I am preparing to self experiment with it. I was warned to have a petscan before my off time. It seems that 12-15 percent of us can have me etastasis with undectable psa. It might go poorly if that were the case. My plan would be to stay off adt til psa rise then scan to see if there were spot(s)that could be zapped.

REPLY
@billfarm

Md Anderson had a trial that said intermittent was not inferior to continuous. I am preparing to self experiment with it. I was warned to have a petscan before my off time. It seems that 12-15 percent of us can have me etastasis with undectable psa. It might go poorly if that were the case. My plan would be to stay off adt til psa rise then scan to see if there were spot(s)that could be zapped.

Jump to this post

How long you had been on Hormone therapy and what is your current PSA, if you don't mind sharing.
I will be completing 9 months of Hormone therapy by tomorrow after RT. My current PSA is 0.05.
In the beginning I was afraid to go ON hormone therapy. Now, I am afraid to go OFF of it too soon!!

REPLY
@ava11

How long you had been on Hormone therapy and what is your current PSA, if you don't mind sharing.
I will be completing 9 months of Hormone therapy by tomorrow after RT. My current PSA is 0.05.
In the beginning I was afraid to go ON hormone therapy. Now, I am afraid to go OFF of it too soon!!

Jump to this post

Ditto!!

REPLY
@ava11

How long you had been on Hormone therapy and what is your current PSA, if you don't mind sharing.
I will be completing 9 months of Hormone therapy by tomorrow after RT. My current PSA is 0.05.
In the beginning I was afraid to go ON hormone therapy. Now, I am afraid to go OFF of it too soon!!

Jump to this post

I had a postoperative psafailure @5wks 1.8 then a month later 2.3. Got on orgovyx/arbiterone/prednisone. Psa dropped within 4wks to undectable for past 22 months. Fatigue is intensifying as end of 24month planned regimen approaches. Planning a holiday pending scans.

REPLY
@billfarm

I had a postoperative psafailure @5wks 1.8 then a month later 2.3. Got on orgovyx/arbiterone/prednisone. Psa dropped within 4wks to undectable for past 22 months. Fatigue is intensifying as end of 24month planned regimen approaches. Planning a holiday pending scans.

Jump to this post

Thank you for sharing your experience. Wish you the best outcome on your scans.

REPLY
Please sign in or register to post a reply.