A 2011 study - effect of Gleason Type 5 cells ("high risk" G8-10 w/ vs. w/o Type 5 cells)

Hoo boy. Ok. I'm going to share this one, but with a bit of trepidation. This will be a little long.

Let me start with some caveats about the study: it's retrospective, so has inherent selection biases and so on. It's from 2011, so necessarily based on work done up to 10 years before that; hence it's going back a long ways in the universe of radiotherapy. The minimum radiation dose was 75 Gy, so not sure what the actual distribution was beyond that. We're almost universally today using at least 79.2 Gy. Brachy boost was not included in this study as far as I can tell. The dose escalation today gives superior results to what used to be possible. And finally, the title is alarming, with Words We Don't Use.

I read this abstract, and had to get the full article from my U of Mich Cancer Center patient information center. (They're very helpful). This is a long paper, and I have read it several times. Tall Allen was able to reinterpret studies like this, and explain them in easier to understand language. I won't try to do that, for risk of misleading anyone. But I do have just a few observations.

Is that enough foundation? I hope so.

I'm sharing this for really only one reason. I have long chafed at the idea of the equivalence in defining "high risk" as PSA>20, or stage cT3 or T4, or Gleason 8-10. The metastatic capability of those wonky Type 5 cells seems so much out of character to the risks from the other parameters that I've never been comfortable with calling those all the same. The G9/10 (or rare G8 with Type 5, or maybe even G7 with tertiary 5) are so uncommon that they always get lumped in enough others to, in my mind, blur the results. The "G8-10" in as study is likely to be 80% or more G8, so the G9/10 gets lost in there.

This study definitely reveals a significant distinction caused by the Type 5 cells. So much so, that it implies treatment decisions should be made differently due to their presence. This study even suggests that Type 5 cases may benefit from immediate and permanent HT upon diagnosis (oh my...). It also supports more recent studies suggesting a course of docetaxel chemo after radiation therapy. Finally, they also suggest that the probability of systemic spread is so high, that local dose escalation (brachytherapy) is not as likely to be helpful as with others, and that enhanced systemic treatments should be strongly considered (again, chemo, advanced HT options, etc.).

So when they suggest to a G9/10 person that 18 months of HT has been shown to be fine, that's been based on the traditional "high risk" definition. No study I've ever seen has enough G9/10 cases to draw a conclusion like that, for G9/10 cases. It's always lumped in with the G8 and other categories so the researchers can draw conclusions justifying the money spent on the study.

Ok, so here's the link to the study, with the actual title of the study. Please don't freak out about it. With modern treatment options, I don't think things are as clear cut as this study shows. And as I've said before, don't quit buying green bananas. Treat the heck out of this version of PCa, don't shy away from aggressive options, but also don't sit around fussing about it. Life is precious, and maybe things like this help us have a better perspective on just how precious it is.

Gleason Pattern 5 Is the Greatest Risk Factor for Clinical Failure and Death From Prostate Cancer After Dose-Escalated Radiation Therapy and Hormonal Ablation

If there's interest, I'll post a couple of the clearest Kaplan-Meier curves showing the effect of the Type 5 cells. I think that would be ok without violating copyrights. If you've read this far, I'd suggest obtaining a copy of this paper yourself (the actual paper's content is far more dramatic than what they show in the abstract). I have never seen a clearer study regarding this issue.

Since this is now what's being called Gleason Grade Group 5, I'm hoping that in the near future we'll see more studies segregating those results for treatment options and so on.

Thank you Redwing (and Rob for responding) for sharing this. It's going into my online Scrivener file cabinet under "G9 guys" and Redwing's very own file I created along with one for Tall Allen. I keep thinking if my files are bulging but well organized, there's a cure in there! My husband has that diabolical Pattern 5, and it's darn squirrely. He seemed to have responded well to Lupron but his PSA rose alarmingly fast in less than a year after his 33 month Lupron regime was over. Now he's about to go on Zytiga. I've always had a bit of regret that he didn't go the brachy-plus route, something TA had suggested, but our plans were already underway. Now it seems for G9s it's iffier than for others so I can let that go along with other regimes we wouldda, couldda, shouldda tried.

Like you both, my husband and I both get constant advice about prostate cancer. We patiently explain "It's not the good kind. It's the kind you die of not with." "The horse was out of the barn." "No, active surveillance wasn't an option." "Gosh it's way too late for supplements/dietary changes but thank you, etc." Deaf ears. It does help when I explain to female friends anyway that PCa is like some breast cancers that are deadly from the get-go. Well, let's all keep fighting with the information and experiences you are all so generous in sharing. It helps a lot!

30,000 men will die from Prostate Cancer this year. And last year and next year.....The VAST majority of those deaths will occur in men with a G9 or G10 Gleason profile..That's just the way it is. This should not be a big surprise to anyone. Fortunately, less than 20% of men diagnosed with PC have the dreaded Grade 5 cancer cells. Most are Gleason 6 which I consider to be a separate disease from the G-9 and 10 variety..Not even in the same lethal category . The problem seems to be those G-5 cells, aggressive and virulent, escape the prostate capsule and take up residence in other parts of the body, bones mostly, long before the patient discovers he has PC. The horse leaves the barn before anyone knows there was a horse IN the barn...After that, all treatment is palliative whether we want to admit it or not...Yes, there are a lucky few G-9 and G-10 guys who manage to detect it BEFORE the horse leaves the barn and they make up a small group of survivors who give the rest of us hope..

Redwing57 said...
Ok, here's a link to one of the more dramatic sets of Kaplan-Meier curves from this study. The distinction of the results with Type 5 cells is frankly amazing. Particularly for cause specific survival, and overall survival, the presence of Type 5 cells is the only distinguishing factor!

Here's the graphic:
Kaplan-Meier curves Freedom from biochemical failure, freedom from metastasis, cause specific survival, overall survival.

This table shows the Hazard Ratios of the clinical outcomes. I've never seen numbers like these. These ratios are multiples of the baseline risk for G6 as the reference = 1. Check out the G8-10 with GP5, metastasis HR is 17.3, and cause-specific survival HR is 58.4!

I don't know. This study came out in 2011. It seems hugely significant. And yet there is little to no reference to this study anywhere. There are still very few G9 cases, yet they did roll out the Gleason Grade Groups which, among other things, do clearly distinguish G9/10 from G8.

If you have any thoughts about this, please feel free to share them.

Thanks(really, no sarcasm) Redwing for another eyeopening study!

I know. It is never fun for me to look at such figures. But on the hopeful side, I see that freedom from mets- while so vastly worse for us guys with some 5- is still 60% at 5 years, and then did not get any worse up to 10 years. Same with cancer specific survival: again MUCH worse compared to all other categories it was at 50% at 6 years, and then got no worse up to 10 years. ( both of those I find surprising, and wonder why or if there is a problem with the figures?) All of the other categories(except for lowest risk rates of mets and death) continued to get worse with each yer after the 5th.

But I am convinced that there is a problem reading the graphs, because the graphs are not matching the HR tables. For example, another odd finding was the BCR rate for HR with no 5 vs HR with 5. Once again, the 5s stop getting worse after 5 years, while the HR 4s just keep getting worse for the entire 10 years. And as I said, this does not match the HRs, although in the HR table 4 I do not see it divided up by years? There is only one HR level provided for the all years?

Any way, I guess it does not matter, it is what it is and I am a 5(diagnosed 5 years ago) who had a PSA >10 and ended up with with positive margin and SV+ and that is all there is to it. I will just continue to do any dietary and or supplemental and or lifestyle changes that are rumored to be helpful if there is also no evidence of harm, and assuming I can stand to do them. And continue to watch those PSAs and ready to jump to other treatments as soon as there is evidence I should act NOW. (of course, I already know many feel I should have already acted, and they might be right, but I just followed the advice of Dr. Joseph Smith of Vanderbilt on that subject) So there is nothing really to do other than what I have already done, but still I hate to see such numbers. They are discouraging, but treatments do continue to improve.

And of course, we can always hope for those continued improvements in treatment, and who knows if various lifestyle changes might also help.

OTOH, the number for the low risk guys should really ease their minds! But for all of my 5 Bros: hang in there! If I am reading the tables right, even we are still slightly most likely to die of something else!

Yes, I think you have it right: An HR of 2 is twice the rate of whatever.

The only thing that confused, or surprised, me was that with the HRs so WAY higher, that on the graphs showing mets or cancer specific death, as we said, they plateaued at about 5 or 6 years for the highest risk guys, and then got no worse. It continued getting worse over those years for the not quite so high risk group.

I've tried to guess about some possible explanation for this. Maybe the very HR guys are so bad that they are most likely to get the job done quickly? Maybe most of the ones who are going to met or die get er done in the first few years, while the guys with the less aggressive strains take longer to do so? Therefore, even though these VHR guys had HRs through the roof compared to every one else, maybe most of this damage got done for most of these guys during the first 6 years, and those first 6 years account for most of the difference in the HRs? You know, because these way more SOB cancers get after you real quick, and if they fail to get you pretty quick then your odds improve? I have no flipping idea, I am just guessing wild. WAG. Just fishing for an explanation for what seems to be an odd result.

It is weird: even the low risk guys, though they almost never met or die, they continue to have larger BCR rates for the whole 10 years. But the worst of the worst? They have WAY more BCR than anyone else, but from 60 months onward, as bas as it is, it gets no worse. Same thing for mets and death, doesn't get any worse. How odd.

Which is why I am wondering if there is something I am missing from the study. Like maybe they didn't follow the VHR guys after 60 months or something like that? But they give no indication of such that I can find. Strange.

Even overall survival keeps getting worse for all groups for the entire 10 years, but bottoms at 90 months for the VHR guys, and then gets no worse for the next 30 months. Could it be that the guys in this group(the ones alive after 90 months) who are not dead of cancer or other causes by 5 years are also folks who are super healthy otherwise, and super aggressive about doing everything they can for both their cancer battle AND their over all health? Which is one reason(perhaps of many) they didn't succumb early to start with? Again, I have no idea, but I would like to. So if anyone knows the reason for this odd finding, please let me know.