There's a lot out there about the potential benefits of NSAIDs in PC, both for prevention and as part of treatment. In regard to adding them to an anti-inflammatory diet/supplement regime, some suggest a single LD-aspirin a day, some a 200-400 celebrex a day, and some suggest both (or neither).
Very little info seems to be available on how or if to use ibuprofin, even though some research indicates there may be anti-PC benefits to its use. (I think I recall reading something about it actually countering the benefit of aspirin if used together.) I have read that NSAID use right before and after radical surgery may help lessen the activation of micro-metastases (in mice, anyway).
Things that are cheap and readily available are nice to have, but only if it can reasonably be asserted they will be of actual benefit (or at least of no harm). I am interested in opinions/articles about specific types and doses that seem reasonable, even if not of proven benefit. What are you guys doing with these, if anything? When might they be contraindicated, even in tiny doses?
Written by
noahware
To view profiles and participate in discussions please or .
NSAID should be taken when needed to control pain or fever and for the shortest time possible. They are associated with GI bleeding, cardiovascular events (heart attacks and stroke), hypertension, atrial fibrillation and renal damage (interstitial nephritis).
Well, as with so many things, it seems like the potential benefit(s) with an NSAID might work through the same channel as any potential adverse outcome(s). In the case of celebrex, inhibition of COX-2 may help your prostate cancer, but may also be a danger to proper kidney function (among other things).
So I don't see it as that much different in principle from many other interventions (like chemo) where potential benefits ALWAYS come with potential costs. Trying to weigh risk/benefit. That's why I'm posting.
My impression w/ NSAIDs is that potential benefits can come with small and infrequent dosing, while most potential dangers are with larger and more frequent intake. So I already know that popping 20 pills of ibuprofen a day for years on end is a bad idea. But how bad or good might it be to take a single pill daily?
The general rule that "NSAID should be taken when needed to control pain or fever and for the shortest time possible" may not apply to LD-aspirin depending on an individual's risk profile. (After a stroke, you don't need to control pain or fever.)
But that said, can we always know the possible individual risk low dose has for potential downsides when taken daily for years, given that bleeding from a baby aspirin is pretty unlikely for most of us? I assume there are ways to tell if one is at higher risk for aspirin-related bleeding, but not sure what they are.
Okay, I understand now... it's just that since your list of associated risks was not PC-specific, I thought your comment about how to use NSAIDs was also not PC-specific.
You said "In the case of celebrex, inhibition of COX-2 may help your prostate cancer, but may also be a danger to proper kidney function (among other things)."
This is why my doctor would not prescribe it to me when I asked him about it. He looked up the Stampede trial data right when I asked him during the apppointment and said he was not impressed with it. He did not think it was worth taking an NSAID daily.
Was it because it was a subgroup analysis? (It was pre-planned) A 22% increase in overall survival (and a 36% increase in prostate cancer-specific survival) is nothing to sneeze at, but I can understand that patients with kidney disease or other contraindications to NSAIDs might be excluded.
That's a valid criticism, although it was a very large subgroup (n=755). Median survival was 55 months for zometa+celebrex+SOC, but only 43 months for SOC. 5-year survival was 47% for zometa+celebrex, but only 37% for SOC.
When planned subgroup analysis is a consideration, good research technique is to look at other subgroups that seem to align with the hazard ratio (HR). If that were the only subgroup, it would raise suspicion. In this case, several other large subgroups: men with N1, GS8-10, <70, no NSAID use, radiotherapy planned, and the total all have similar HRs. This shows it's not just a random fluke. Also, other endpoints show good improvement with the combination in men with metastases: failure-free survival, and prostate cancer-specific survival.
They also looked at adverse events for those taking the combo. Serious adverse events were slightly lower among those taking the combo (32% for combo vs 36% for SOC) . Most specific adverse events were similar but lower for the combo than for SOC (hot flashes, pain, renal, fatigue, cardiac, and respiratory). Only GI and ONJ were higher (5% vs 3% and 2% vs 0% respectively). It's probably prudent to take daily Prilosec with any daily NSAID, including Celebrex.
Since there was no significant hazard attributable to taking the combination, what's the downside?
By "Level I evidence" you probably mean a randomized prospective preferably double-blinded trial. An of course no one is ever going to do that huge an dexpensive undertaking to establish an additional indication for a dirt-cheap generic medicine available worldwide. So we have to look to lower levels of evidence which do suggest a significant benefit. (One was posted here this week.) So then you consider the potential risks (extremely low) vs. potential benefits and make your decision. To me that is a reasonable and practical approach.
There have been Level One trials that include Metformin and there are others in progress. You should ignore all lower level of evidence studies when higher level is available.
This small one showed there was no benefit to adding metformin to ADT in men with relapsed or advanced PC:
Only the third citation, (docetaxel +/- Met study) carries any weight for me (for numerous design reasons in the first two). Failure to show benefit in those weak studies does not rise to showing that no benefit exists.
Short term use of Met while undergoing Taxotere gives no added benefit seems clear.
As the author wrote:
"For patients with mCRPC without diabetes, the addition of metformin to docetaxel chemotherapy does not improve PSA50, objective response rates, mPFS, or OS. This could suggest that either metformin is not clinically active for patients with CRPC, or that docetaxel may not be the right partner drug to be used in combination with metformin, or that the CRPC setting may not be the right disease state. The authors also suggested that perhaps if metformin had been continued at progression instead of discontinued at the same time as chemotherapy, this may have made an effect as the benefits may be more long term."
Apparently, you are are willing to believe observational studies that conform to your preconceived notions over an actual randomized clinical trial, if the RCT is small. And you seem to ignore the study that explains why detection bias occurs in observational studies. It sounds like confirmation bias on your part. I agree that the small RCTs are not definitive, but are at least as good as any large observational study. They are both only hypothesis-generating. I remain agnostic, although it so far doesn't seem like metformin confirms any benefit. (This is the default position). As I said, the STAMPEDE trial will be definitive.
So if observational trials are your cup of tea, what do you make of these?
• "We did not find a connection between antidiabetic drugs and PCa aggressiveness or progression. "
• "The addition of metformin to abiraterone for patients with metastatic castration-resistant prostate cancer and PSA progression while receiving abiraterone therapy does not affect further progression and has no meaningful clinical benefit. A higher-than-expected gastrointestinal toxicity attributed to metformin was observed."
"Metformin use was not associated with improved biochemical survival or cancer specific survival in this cohort of men treated with prostate brachytherapy."
"This large study did not find an association between metformin use and risk of prostate cancer among older men with diabetes, regardless of cancer grade or method of diagnosis."
"Among diabetic men with a negative pre-study biopsy who all underwent biopsies largely independent of PSA, metformin use was not associated with reduced risk of prostate cancer diagnosis."
OK Allen. I will work through these, maybe tonight. Feels like a barrage, or are you trying to give me a whipping?
As for "Do I have confirmation bias going on?" Absolutely Yes. I am looking specifically for possible therapies and specific applications that may be of benefit in some settings for prostate cancer treatment. Only about treatment.
So the prevention and detection oriented studies are outside of my criteria. And therefore so is the detection bias explanation which is valid in that arena.
I do not like to generalize negative studies as it is nearly impossible to prove a negative, as you well know. To say that adding metformin while starting ADT (vs ADT alone) in 36 subjects randomized for 28 weeks shows no benefit. Well I do not find that compelling. It lacks any statistical power and the premise is unreasonable to begin with. Of course the new ADT is going to dominate the clinical picture and drown any incremental benefit from a metabolic agent affecting mTOR, cyclin D-1 and AMPK effects, etc.
The hazard ratios should have confidence intervals to show that there is the statistical power to demonstrate anything.
The Docetaxel with randomized metformin was much better BTW. They made their case but only in that specific setting and duration. So it says maybe elsewhere. Or longer.
Something seems to be going on with metformin (see below) and hopefully STAMPEDE will ferret it out. Appreciate your views as always Allen.
"In the clinical setting, retrospective data is mixed. Two large retrospective studies have not found any prostate cancer-specific survival benefit. However, a recent large meta-analysis which included 30 different studies and 1.6 million patients suggested that metformin use improved overall survival (HR = 0.72), prostate cancer-specific survival (hazard ratio (HR)= 0.78), and recurrence-free survival (HR=0.60)." - From Jason Zhu MD, TAXOMET report
Everyone is looking for something that will help. But the only way to look at any research study is through the lens of skepticism. In fact, the only useful research hypothesis (called the "null hypothesis") is that the treatment makes no difference. So contrary to what you state, the only valid way to do a study is to disprove the null hypothesis. A p value < 0.05 means that the null hypothesis could have occurred by chance less than 5% of the time, which means the null hypothesis (i.e., it makes no difference) is rejected. As the philosopher Karl Popper pointed out, science can only prove falsification.
Those small RCTs met that statistical threshold, showing that the null hypothesis (no effect) could not be rejected. They do lack power to detect a small effect, however. And you may be right that longer trials in more patients may detect some effect. Observational studies of the kind that you post are worthless because of detection bias. Even a thousand observational studies in a million people cannot overcome detection bias. There are techniques such as propensity score matching or inverse probabilty weighting that simulate randomization and do a better job of getting around selection bias. We have to rely on those techniques when observational studies are all we have.
Thank you for the reminders. The double negative of rejecting the null hypothesis indicating positive effect notwithstanding. I am not familiar with propensity score matching / inverse probability weighting are too obscure for my understanding (lies damn lies and statistics territory). Everyone has selection bias - All we can do is be aware and beware of it. And I am very familiar with selection bias. My own clinical trial fell on those hard rocks some time back. Keep on turning over the rocks. You may find a jewel but only if you look.
In a small randomized trial, aspirin did not decrease the incidence of prostate cancer occurrence, delay castration-resistant prostate cancer transformation, or reduce tumour-associated death.
Yes, sorry... NSAID before surgery is obviously a no-no. I should have been more careful in my comment, because I don't mean to imply anyone should do that... for mice only! I was just referencing one study of many on possible anti-proliferation and metastasis-slowing properties. Most articles I've seen, though, seem to be on cancers other than PC.
Well, that may be true, but the ADT standard-of-care all stems from the work of Huggins, and he had said the progress of his work on hormones and cancer was at least partly due to that of Lacassagne... who worked with mice!
One can get a hypothesis and even a justification for an effect observed in humans from mice, but they tell one nothing about the safety or efficacy of the medicine in humans. Huggins demonstrated dramatic success of castration for survival in humans. Patients would do well to completely ignore mouse studies in the literature (I do) - the purpose of those studies (if reputable) is never to guide human therapy, only to guide clinical research. Unconscionable snake-oil salesmen (like Life Extensions) make millions pretending that mouse studies are applicable to humans.
I can't say for a fact that there is the same combination benefit with denosumab (Prolia or Xgeva) as there is with Zoledronate (Zometa). Denosumab inhibits the enzyme responsible for osteoclast activation (the cells that break down bone and allow it to be resorbed), thereby inhibiting bone resorption and increasing bone mass. Zoledronate binds to bone and inhibits its resorption by osteoclasts. So they both accomplish the same thing, but in different ways.
My guess is the survival benefit doesn't rely on the mechanism of action. I suspect that prostate cancer itself has osteoclast stimulating behavior in the tumor microenvironment (which may be inhibited by either bone medicine), and the destruction of bone releases inflammatory cytokines that result in a build-up of fibroblasts in the tumor supportive tissue. The inflammation that causes those cytokines is inhibited by the Celebrex acting in the tumor stroma (the supportive tissue). I think both actions (cell destruction and consequent cytokine response) are necessary to build the tumor stromal compartment. If I'm correct, the combination may reduce the stromal build-up that is necessary for large tumor growth. Just a guess.
Thank you for this more complete and well reasoned analysis Allen.
Years back I started Celebrex 400mg and was on a different bisphosphonate at the time when I had BCR with PSADT < 3months. After starting the Celebrex my PSADT slowed to over 18 months which remained stable for two years.
Tell us more about yourself... Are you from noahware or somewhere? age? location? scores (psa/gleason) treatments to date? treatment center(s)? type of treatment(s)? doctor"s names(s)? All info is voluntary but helps us help you and helps us too. Thank you! If you do respond you may want to add it to your home page for future reference.
Yeah, I'll get that all in a profile one of these days...
DX in 05/19 at age 59 w/ PSA of 20 (first test after about 15 years), Gleason 3+4 w/ a few mets in spine. Went on lo-cal / lo-protein keto and lost 40 extra pounds over 3 mos, w/ a psa drop from 20 to 13. Added back protein and calories to maintain weight/muscle, then added back carbs, and psa back up... to 26 in 11/19, and now to 31.
Conclusions: a restrictive diet is difficult (for me), and maybe you CAN starve cancer... by starving yourself?
Plan is to start some form of ADT pretty soon, still debating what and when and with whom.
Unfortunately, I had some very profound advice but lost it writing the above!!! And that is the result of ADT!! I no longer know if I’m coming or going!!! The rest of the side effects have gone now but this lingers on!!! The new normal? Who knows?!!!! Not insinuating that you’ll get the SE’s like this, likely not, but you’ll get some!!!
I only mention that because it appears you intend to apply ADT as your first line of defense. With what little I can recall of the wealth of knowledge I acquired here, I wish you would run “that” by the group!!!
(By the way, that was the profound advice!!!(it came back))
A light in my fog says I would zap them bugs and maybe us some lesser ADT to mop up!!!
Maybe tell me to FO but it just kind of rubbed me funny!!!
Content on HealthUnlocked does not replace the relationship between you and doctors or other healthcare professionals nor the advice you receive from them.
Never delay seeking advice or dialling emergency services because of something that you have read on HealthUnlocked.
List of drugs by their metastatic prostate cancer cancer fighting properties 
Imipramine Inhibits Migration and Invasion in Metastatic_Castration-Resistant_Prostate_Cancer
