Metronidazole - shouldn't we be concerned that the question of carcinogenic potential hasn't been settled?

[CharonY]

22 minutes ago, Alfred001 said:

Well, the fact that there may be other drugs that have a suspected link with cancer is not an argument for not investigating metronidazole's suspected link. All of them should be investigated.

No one said that it should not be investigated. In fact, they are still underway. And many have been done in the past (I have seen studies going back to the 70s at least). None of them have found strong evidence for carcinogenic effects in humans so far. 

 

22 minutes ago, Alfred001 said:

So we have a tentative link and strong mechanistic evidence, doesn't that suggest that the link should be further investigated before the drug is prescribed en masse as we are in fact doing?

Required evidence is usually based on existing evidence levels, with clinical data (which is part of the approval process) having the highest weight. The actual evidence for cancer risk in humans is still low, and a bigger worry at this point is neurotoxicity during prolonged use. What ultimately has to be evaluated is the cost/benefit.

Many drugs, including antibiotics, anti-cancer drugs etc. are highly toxic. The question therefore is whether there are groups which ultimately have more benefit than harm. Obviously, if there are alternatives that are lass harmful, they would be preferred. Some other antibiotics that can be used to treat certain conditions (such as Chron's disease) but depending on the person, the side effects can be more or less severe.  There are basically no harmless drugs. Just various levels of harm.

Again, the premise is not that we can only use drugs that have no risk of harming the patient. This is just not viable. It is about finding treatments for conditions that overall have the best risk/benefit ratio for a given condition and a given patient. So far, metronidazole makes the cut and with antibiotics efficacy waning, more toxic ones will be see increased use. 

If the intention was to create the perfect drug before use, well most of us would probably be dead before that happens.

That all being said, I will agree that over prescription can be an issue. But that is not dealt with by kicking out drugs from the portfolio. Rather, we need better and faster diagnostics to ensure that we only prescribe antibiotics that are needed.

 

 

23 minutes ago, Alfred001 said:

But have the studies been done? Is there an adequately powered study with adequate followup that allows us to conclusively say there is no increase in cancer risk or that it is trivially small? Or is it simply that we haven't done the studies so we don't know how much we are potentially increasing people's risk of cancer?

With cancer risk, conclusive evidence under therapeutic limits is very difficult. It is looking for rare cases among rare cases requiring huge cohorts to see an effect. As such having enough statistical power to show no effect is going to be exceedingly difficult. I.e. the rate of stomach cancer is maybe around 10 per 100,000 persons. So even if you enroll 10,000 control cases and 10,000 folks taking the drug, you may have maybe one case, which could be purely stochastic. Even worse, you would need to track them for years which would make it a tremendously expensive study. So realistically you would need many individual studies and try to look at a meta-analysis, or more likely have retroactive surveys. The issue with the latter is that there are a lot of confounding factors beyond the drug under consideration that may  affect cancer risk. So even those might not reach enough folks to figure things out.

That all being said, some of the earlier studies in the 80s tracked over 700 patients for close to 10 years, seeing no effects for patients treated for trichomoniasis for any types of cancers (if adjusted for smoking). Another study was retrospective and looked at matched patient data (i.e. similar groups but one with and another without treatment) for about 10-12 years found no effects for short-term exposure, either. Since then, there have been follow-ups using similar designs but no smoking gun.

That being said, other commonly used antibiotics are starting now also to be suspected to potentially be associated with cancers during long-term treatment. For example one retroactive comparison between clarithromycin compared to metronidazole (there was no untreated patient cohort) showed that clarithromycin had a higher overall death and cancer risk than metronidazole after follow-up. (And interestingly clarithromycin is also under discussion to be used as a potential repurposed drug for cancer treatment at some point...).

Also, it is necessary to look at risk in context of overall risk we are willing to commit. For example, certain diets are associated with cancer, and we are likely to eat more of it and for longer than an antibiotics course. 

 In summary, folks look at the available set of info to assess risk/benefit. Sometimes the assessment is wrong (thalidomide being one of the famous examples), but it always is based on need and empirical evidence. Otherwise, options would be really, really limited.

[exchemist]

2 hours ago, Alfred001 said:

Well, the fact that there may be other drugs that have a suspected link with cancer is not an argument for not investigating metronidazole's suspected link. All of them should be investigated.

So we have a tentative link and strong mechanistic evidence, doesn't that suggest that the link should be further investigated before the drug is prescribed en masse as we are in fact doing?

I actually don't think the study would be that difficult, this drug is widely prescribed, it wouldn't be hard to find subjects to enroll, even a massive number of them, and all you need to do is match them with controls and follow them over, yes, decades, but all that means is check ins every 5-10 years to see who got cancer. But let's set this aside, I want to focus the debate on whether it makes sense to prescribe this drug given the evidence for it causing cancer.

To your last paragraph, I think it's a matter of urgent concern because it has substantial (tho not conclusive) evidence for possibly causing cancer, yet we prescribe the drug widely. Is it the most urgent drug to investigate for such a link? I don't know, I make no claim with regards to that, my claim is simply that prescribing a drug with evidence for causing cancer without disproving a causal connection with an adequate study is insane.

But have the studies been done? Is there an adequately powered study with adequate followup that allows us to conclusively say there is no increase in cancer risk or that it is trivially small? Or is it simply that we haven't done the studies so we don't know how much we are potentially increasing people's risk of cancer?

Checking who got cancer would not be enough. People get cancer anyway from a wide variety of causes, or just by bad luck. So you would need a sample population big enough to detect a statistically significantly higher rate of cancer than an equivalent population who had never been prescribed metronidazole  and then correct that for any causes of cancer (such as recurrent helicobacter pylori infection) which might make the test population more susceptible to cancer than the controls. Not straight forward, it seems to me.

But in any case from the links you provided there is no strong evidence of cancer from this drug as prescribed in human subjects. I quote from your second link:

"A teratogenic effect of metronidazole could not be established (Koss et al. Reference Koss, Baras, Lane, Aubry, Marcus, Markowitz and Koumans2012), but it was found to be carcinogenic in rodents after extended durations of highly dosed treatment. In man, results were less clear and often conflicting (Dobiás et al. Reference Dobiás, Cerná, Rössner and Srám1994). With regard to short-term treatment with metronidazole, originally no correlation between metronidazole intake and cancer was found (Falagas et al. Reference Falagas, Walker, Jick, Ruthazer, Griffith and Snydman1998), but more recent studies report on a limited correlation (Friedman et al. Reference Friedman, Jiang, Udaltsova, Quesenberry, Cha and Habel2009). As a consequence, metronidazole is officially classified as ‘reasonably anticipated to be a human carcinogen’.

 

There is a theoretical risk, no doubt, due to the results of the animal studies, but these involve vastly more intense and prolonged exposure to the drug than occurs in prescribing. It is fairly evident from the passage I have quoted that if there is an effect it cannot be marked.   @CharonY describes the situation 2 posts above this one. 

 

  

 

[CharonY]

In that context one should also to keep in mind that absolute risk, is often not helpful for decision-making. Rather, one probably should think more in terms what the risk of taking a medication is vs not taking it, within a given context. Specifically for stomach cancer, the rates are fairly low, and even if doubled, it is not terribly higher. So if one was a smoker or live in an area with high radon levels, the risks are quite a bit more relevant, for example.

 

[John Cuthber]

On 7/8/2023 at 5:13 PM, John Cuthber said:

Alcohol, sunlight, silica and the fumes from diesel engines are known human carcinogens.
Do you suggest that we ban them?

Or do you think we should consider the benefits as well?

I should have mentioned x-rays too.

The carcinogenicity (and other  risks) of metronidazole are still under investigation- as are those with any other drugs in medical use.

https://yellowcard.mhra.gov.uk/


 

[Alfred001]

On 7/13/2023 at 9:08 PM, CharonY said:

That all being said, some of the earlier studies in the 80s tracked over 700 patients for close to 10 years, seeing no effects for patients treated for trichomoniasis for any types of cancers (if adjusted for smoking). Another study was retrospective and looked at matched patient data (i.e. similar groups but one with and another without treatment) for about 10-12 years found no effects for short-term exposure, either. Since then, there have been follow-ups using similar designs but no smoking gun.

Right, there are some studies that show no correlation and some that show a correlation. I guess we would need to do a systematic review of the literature to really know which direction the evidence is pointing to.

 

On 7/13/2023 at 11:10 PM, exchemist said:

But in any case from the links you provided there is no strong evidence of cancer from this drug as prescribed in human subjects. I quote from your second link:

"A teratogenic effect of metronidazole could not be established (Koss et al. Reference Koss, Baras, Lane, Aubry, Marcus, Markowitz and Koumans2012), but it was found to be carcinogenic in rodents after extended durations of highly dosed treatment. In man, results were less clear and often conflicting (Dobiás et al. Reference Dobiás, Cerná, Rössner and Srám1994). With regard to short-term treatment with metronidazole, originally no correlation between metronidazole intake and cancer was found (Falagas et al. Reference Falagas, Walker, Jick, Ruthazer, Griffith and Snydman1998), but more recent studies report on a limited correlation (Friedman et al. Reference Friedman, Jiang, Udaltsova, Quesenberry, Cha and Habel2009). As a consequence, metronidazole is officially classified as ‘reasonably anticipated to be a human carcinogen’.

No strong evidence does not mean no evidence. I was never claiming there is conclusive evidence. The very text you quoted says "in man, results were less clear and often conflicting," "more recent studies report on a limited correlation," "reasonably anticipated to be a human carcinogen."

Like I said, it points to a potential correlation that should be investigated. Why are we prescribing a drug if the question of carcinogenic potential hasn't been settled. Now, maybe it HAS been settled, like I said, I haven't done a systematic review of the literature, maybe the null finding studies overwhelmingly outnumber the ones that show a correlation, but right now, we (meaning we on the forum discussing this) have a few studies that show a correlation and a few that don't - hardly a settled question.

[exchemist]

2 minutes ago, Alfred001 said:

Right, there are some studies that show no correlation and some that show a correlation. I guess we would need to do a systematic review of the literature to really know which direction the evidence is pointing to.

 

No strong evidence does not mean no evidence. I was never claiming there is conclusive evidence. The very text you quoted says "in man, results were less clear and often conflicting," "more recent studies report on a limited correlation," "reasonably anticipated to be a human carcinogen."

Like I said, it points to a potential correlation that should be investigated. Why are we prescribing a drug if the question of carcinogenic potential hasn't been settled. Now, maybe it HAS been settled, like I said, I haven't done a systematic review of the literature, maybe the null finding studies overwhelmingly outnumber the ones that show a correlation, but right now, we (meaning we on the forum discussing this) have a few studies that show a correlation and a few that don't - hardly a settled question.

This issue has now been adequately answered by several of us. You are becoming a bore. 

[John Cuthber]

12 hours ago, Alfred001 said:

I guess we would need to do a systematic review of the literature to really know which direction the evidence is pointing to.

There are two problems with that.
The first is that's not how testing works.
If you are lucky, it goes like this:
A study of a hundred patients would probably tell you about an effect that happened in 10% of them, but might miss an effect that happened in 1% of them.
If you raise the sample size to a million, you will almost certainly spot any side effect that happens in 0.01% of teh population.

So, at best, if you have a big enough cohort, you can detect an adverse effect that is very rare.
But if the effect is rare enough, you may never be able to get a large enough test pool.

If you are unlucky, you find something like this; the drug raises the lifetime cancer risk from about 30% (or whatever it is) to 30.1%.
You need a huge, well designed study to find an effect like that. Such trials are expensive

And here's the big problem with your suggestion.

What do you do with the result?

The people prescribing metronidazole know it's associated with a small risk of harm from cancer.
But they are  using it to treat a condition with a relatively large risk of harm.

Actually putting a number on the first risk- say it's a 0.1234% higher relative risk- does not change clinical practice.

So you would end up spending a lot of money to confirm something which the doctors already know, and already act on.

Now, if you personally are a billionaire and want to waste your money on such a programme, that's your choice.
But I suspect the rest of us would prefer the healthcare industry to spend its limited resources on things where the outcome will actually make a difference.

Don't you agree?

[Alfred001]

On 7/21/2023 at 10:38 AM, John Cuthber said:

There are two problems with that.
The first is that's not how testing works.
If you are lucky, it goes like this:
A study of a hundred patients would probably tell you about an effect that happened in 10% of them, but might miss an effect that happened in 1% of them.

I don't understand what statistical power has to do with what I said. I said we need to do a systematic review of the literature to know what the totality of evidence says and not look at 4-5 studies.

On 7/21/2023 at 10:38 AM, John Cuthber said:

But if the effect is rare enough, you may never be able to get a large enough test pool.

But don't you understand that you don't have to find it? If you have 100,000 subjects and you don't find an increase in cancer risk then you can say that, if an effect exists, it is smaller than 1 in 100,000 getting cancer. That is sufficient for us to know that the risk, if it exists at all, is miniscule and we can now know the drug is safe or at least safer than not treating H pylori.

On 7/21/2023 at 10:38 AM, John Cuthber said:

You need a huge, well designed study to find an effect like that. Such trials are expensive

I don't understand why people keep bringing up things I've already addressed. It would not be expensive because it would not be a trial, but an observational study.

On 7/21/2023 at 10:38 AM, John Cuthber said:

Actually putting a number on the first risk- say it's a 0.1234% higher relative risk- does not change clinical practice.

OMG, of course it would change clinical practice. If we found out that the risk was higher than the cancer risk conferred by H pylori we would use a different antibiotic to treat the infection. That's the whole point. Again, I've already explained this at least twice.

On 7/21/2023 at 10:38 AM, John Cuthber said:

So you would end up spending a lot of money to confirm something which the doctors already know, and already act on.

Again, it has been stated so many times already - we don't know what the cancer risk associated with metronidazole is. Literally the whole point of this thread.

And, again, when I say we, I mean we here in the thread. Perhaps studies to answer the question have been done, we'd need to do a comprehensive review of the literature. Right now we have a bunch of studies with conflicting results.

[John Cuthber]

30 minutes ago, Alfred001 said:

I don't understand what statistical power has to do with what I said.

I'm sorry; I thought I had made it clear.
For any set of experiments, statistical power is finite.
 

 

31 minutes ago, Alfred001 said:

But don't you understand that you don't have to find it? If you have 100,000 subjects and you don't find an increase in cancer risk then you can say that, if an effect exists, it is smaller than 1 in 100,000 getting cancer.

I understand that.
How did you come to the conclusion that such a test has not been done?
You are the one saying we need more testing; You are also the one saying that we don't need to find the problem

Come back when you have finished arguing with yourself.
 

 

32 minutes ago, Alfred001 said:

. It would not be expensive because it would not be a trial, but an observational study.

As I said, are you offering to pay for it?
But. more importantly, what do you think this is?

 

On 7/14/2023 at 8:38 AM, John Cuthber said:

The carcinogenicity (and other  risks) of metronidazole are still under investigation- as are those with any other drugs in medical use.

https://yellowcard.mhra.gov.uk/

 

 

34 minutes ago, Alfred001 said:

OMG, of course it would change clinical practice. If we found out that the risk was higher than the cancer risk conferred by H pylori we would use a different antibiotic to treat the infection. That's the whole point. Again, I've already explained this at least twice.

AFAICT you have yet to explain why you think we do not already know that the risk from H pylori is greater than that from the drug.
Do you realise that neither estimate of probability needs to be very precise?

 

 

37 minutes ago, Alfred001 said:

- we don't know what the cancer risk associated with metronidazole is.

We know that it is small. (Because, if it was high, it would be noticeable- e.g via the yellow card scheme or through American ambulance chasing lawyers As wiki points out "In 2020, it was the 222nd most commonly prescribed medication in the United States, with more than 2 million prescriptions.^".)

And that's all we actually ned to know.

This is exactly what statistical power has to do with it.
We did tests. They were not powerful enough to be sure of the outcome; they never can be.
But they were good enough to know that the cure was better than the disease.

 

[Alfred001]

On 7/23/2023 at 7:54 PM, John Cuthber said:

How did you come to the conclusion that such a test has not been done?

Again, I've said like 10 times I've not done a systematic review of the literature. Do you know that it has been done?

On 7/23/2023 at 7:54 PM, John Cuthber said:

You are the one saying we need more testing; You are also the one saying that we don't need to find the problem

You did not understand what I wrote. I'm not sure I can make it simpler after so many posts, but let me try:

We have evidence suggestive of carcinogenic potential. We need to test to know how large the risk is. We don't need to know exactly how large it is, knowing that it is smaller than 1 in 100 000 would probably be sufficient. Let me know which part of that you don't follow, I'll explain easier if I know where you're getting lost.

On 7/23/2023 at 7:54 PM, John Cuthber said:

AFAICT you have yet to explain why you think we do not already know that the risk from H pylori is greater than that from the drug.

Show me the meta analysis with adequate statistical power and followup.

On 7/23/2023 at 7:54 PM, John Cuthber said:

Do you realise that neither estimate of probability needs to be very precise?

😄 I can't...

This is exactly what I explained to you in the previous post.

On 7/23/2023 at 7:54 PM, John Cuthber said:

We know that it is small. (Because, if it was high, it would be noticeable- e.g via the yellow card scheme or through American ambulance chasing lawyers As wiki points out "In 2020, it was the 222nd most commonly prescribed medication in the United States, with more than 2 million prescriptions.^".)

Man, use your head. I feel like you're just arguing to be arguing, because there's no way you really believe arguments this bad.

Cancer is a long gestating disease with a million risk factors. Think about it. If a person gets cancer, would they be able to link it to a course of metronidazole they took 10 years ago? Would they even remember taking it. Come on, man.

On 7/23/2023 at 7:54 PM, John Cuthber said:

We did tests. They were not powerful enough to be sure of the outcome; they never can be.
But they were good enough to know that the cure was better than the disease.

Source? No single studies looking at one cancer with short followup and small sample size.

Edited July 24, 2023 by Alfred001

[StringJunky]

Once a drug has been developed and passed  the 3 development phases, the only thing left is feedback from prescribers and patients as it is used. That's $2B plus. From there, the feedback processes guides  the procedures and ongoing prescription advice. I don't know what more they can reasonably do. Your position is coming off like that of a hypochondriac. Stop reading the-side effects. They note every teensy-weensy thing to cover their ass. Just because a patient complains of an effect, doesn't mean it's attributable to the drug.... an apparent correlation is not causation all of the time.

I don't why you are making such a bogeyman of cancer. Cells are little machines and sometimes they malfunction that way, even without a catalyst.

[John Cuthber]

3 hours ago, Alfred001 said:

We don't need to know exactly how large it is, knowing that it is smaller than 1 in 100 000 would probably be sufficient.

OK

Let's start there because we agree about it.
If the risk is low enough, it's not worth worrying about.

Why?
Do you accept that essentially, it's "too small to worry about" because it's "too small to make any (noticeable) difference"?

Well, we have been using the stuff for over half a century.

And nobody noticed the difference.
Even though we have systems in place to check, nobody noticed.


 

[Alfred001]

On 7/25/2023 at 12:51 AM, John Cuthber said:

OK

Let's start there because we agree about it.
If the risk is low enough, it's not worth worrying about.

Why?
Do you accept that essentially, it's "too small to worry about" because it's "too small to make any (noticeable) difference"?

Well, we have been using the stuff for over half a century.

And nobody noticed the difference.
Even though we have systems in place to check, nobody noticed.


 

One of the things that most gets on my nerves in a debate is when a person can't keep track of arguments that were already made and already refuted, but this is even worse. Do you SERIOUSLY not remember that you made this argument FOUR POSTS AGO and I answered it THREE POSTS AGO. Like not on the previous page, THREE POSTS AGO.

[John Cuthber]

52 minutes ago, Alfred001 said:

THREE POSTS AGO.

OK, let's look at what you said 3 posts ago.
 

On 7/24/2023 at 8:11 PM, Alfred001 said:

I've not done a systematic review of the literature.

OK, that's not going to refute anything much.
 

On 7/24/2023 at 8:11 PM, Alfred001 said:

You did not understand what I wrote. I'm not sure I can make it simpler after so many posts, but let me try:

That's simply wrong.
 

On 7/24/2023 at 8:11 PM, Alfred001 said:

We have evidence suggestive of carcinogenic potential.

We pretty much always do- as I said, sunshine, alcohol...
 

On 7/24/2023 at 8:11 PM, Alfred001 said:

We need to test to know how large the risk is. We don't need to know exactly how large it is, knowing that it is smaller than 1 in 100 000 would probably be sufficient. Let me know which part of that you don't follow,

I already said essentially the same thing. As long as you know the risk is small, you don't need to quantify it

On 7/21/2023 at 9:38 AM, John Cuthber said:

Actually putting a number on the first risk- say it's a 0.1234% higher relative risk- does not change clinical practice.

So it's fairly stupid to claim that I don't understand it. Why did you do so?
Is it because you are "a person can't keep track of arguments that were already made"?

 

  

On 7/24/2023 at 8:11 PM, Alfred001 said:

Show me the meta analysis with adequate statistical power and followup.

I already did.
But you failed to understand it.

  

On 7/14/2023 at 8:38 AM, John Cuthber said:

he carcinogenicity (and other  risks) of metronidazole are still under investigation- as are those with any other drugs in medical use.

https://yellowcard.mhra.gov.uk/

What do you think they do with the data from the yellow card scheme?

Do you think they use it as some sort of lottery?
Or, if I make it obvious enough, do you realise that they use it to do analyses of the risks.
It's not a meta-analysis. It's better- it's an analysis of the biggest data set available- the whole uk patient cohort.
 

On 7/24/2023 at 8:11 PM, Alfred001 said:

This is exactly what I explained to you in the previous post.

And I had already made the point (see above "Actually putting a number on the first risk- say it's a 0.1234% higher relative risk- does not change clinical practice.")
So you were failing to read what I had said.

Which makes this

1 hour ago, Alfred001 said:

One of the things that most gets on my nerves in a debate is when a person can't keep track of arguments that were already made and already refuted, but this is even worse. Do you SERIOUSLY not remember that you made this argument FOUR POSTS AGO and I answered it THREE POSTS AGO. Like not on the previous page, THREE POSTS AGO.

really stupid, doesn't it?

 

On 7/24/2023 at 8:11 PM, Alfred001 said:

Source? No single studies looking at one cancer with short followup and small sample size.

Do you really not understand that data- including cancer data - is kept under surveillance?
Were you not aware of things like this?

https://www.kingsfund.org.uk/publications/using-data-nhs-gdpr?gclid=CjwKCAjw_uGmBhBREiwAeOfsd9JFjMNx86yeyws3aIGInpt-FBc5zhgBnR_re34pl9FgvUpo3HG__hoC10gQAvD_BwE

So, as I said, if it was a big enough risk to notice, it would have been noticed.

On the other hand, you have failed to spot the real point I made here (presumably because you were too busy ranting).
 

On 7/24/2023 at 11:51 PM, John Cuthber said:

OK

Let's start there because we agree about it.
If the risk is low enough, it's not worth worrying about.

Why?
Do you accept that essentially, it's "too small to worry about" because it's "too small to make any (noticeable) difference"?

Well, we have been using the stuff for over half a century.

And nobody noticed the difference.
Even though we have systems in place to check, nobody noticed.


 

My point is that "too small to notice" is the same as "too small to notice".

There's one thing which we both agree on- there's a level of risk that's "trivial". Once you know that the risk is less than some cut-off, there's no point putting the resources into measuring it.

And we have systems for monitoring drug safety.


Either our systems are not good enough to spot a problem which is "more than trivial" in which case there's a problem with our systems which has nothing to do with metronidazole.
Or our systems are able to a problem which is "more than trivial" in which case, if the stuff is a problem, then we would spot it.

Which of those conditions are you concerned about here?
 

[studiot]

Good breakdown of your analysis, John.  +1

15 minutes ago, John Cuthber said:

...etc

What do you think they do with the data from the yellow card scheme?

Do you think they use it as some sort of lottery?
Or, if I make it obvious enough, do you realise that they use it to do analyses of the risks.
It's not a meta-analysis. It's better- it's an analysis of the biggest data set available- the whole uk patient cohort.

...etc

 

[Alfred001]

Had to drop this debate due to being busy, but reviving it...

So, here's a review article making my point:

https://pubmed.ncbi.nlm.nih.gov/12052431/

Quote

Metronidazole (MTZ, 1-[2-hydroxyethyl]-2-methyl-5-nitroimidazole), an antiparasitic and antibacterial compound, is one of the world's most used drugs. MTZ is potentially carcinogenic to humans due to the following facts: it is a proven mutagen in bacterial systems, it is genotoxic to human cells and also, it is carcinogenic to animals. However, due to inadequate epidemiological evidence, it is not considered as a risk factor for cancer in humans. As it will be discussed here, the existing population studies are deficient since they have not included sufficient sample size, the follow-up time has not been long enough, and the individual sensitivity to the drug might have been acting as a confounding factor. Due to the increasing use of this drug, more and improved studies are needed to elucidate its mechanism of genotoxicity and its carcinogenic potential.

Another one

https://pubmed.ncbi.nlm.nih.gov/29595104/

Quote

Several retrospective cohort studies have reported MTZ as a potential risk factor for lung cancer (n = 771), cervical cancer (n = 2500), breast cancer (n = 2), cholangiocarcinoma (n = 1), and neuroblastoma (n = 28). So far, all the reported data have confirmed MTZ carcinogenicity in animals; however it is still controversial in humans.

 

Edited November 2, 2023 by Alfred001

[John Cuthber]

5 hours ago, Alfred001 said:

However, due to inadequate epidemiological evidence, it is not considered as a risk factor for cancer in humans.

That's not true.
It is considered as a risk factor in humans
https://en.wikipedia.org/wiki/Metronidazole

says 

"In 2016 metronidazole was listed by the U.S. National Toxicology Program (NTP) as reasonably anticipated to be a human carcinogen."

and "Metronidazole is listed as a possible carcinogen according to the World Health Organization (WHO) International Agency for Research on Cancer (IARC)"

It plainly is considered to be a cancer risk.

Similarly
 

5 hours ago, Alfred001 said:

however it is still controversial in humans.

Says who?
That paper is from 2018 by which time MTZ had been listed as a carcinogen for 2 years or so.


Is anyone saying it's not carcinogenic?

 

[Alfred001]

9 hours ago, John Cuthber said:

Is anyone saying it's not carcinogenic?

???

That's what we'd debated for two pages. You claiming no effect exists or is so small as to not matter, me saying that, as the paper says, there's inadequate evidence to know how big the effect is.

[CharonY]

1 hour ago, Alfred001 said:

That's what we'd debated for two pages. You claiming no effect exists or is so small as to not matter, me saying that, as the paper says, there's inadequate evidence to know how big the effect is.

I think you are still missing the point regarding how risk is evaluated. The reason why it is controversial is because animal studies suggest an effect, but so far no evidence in humans have emerged. And as also said before, the real benchmark is whether a treatment provides a net benefit over a disease, and not whether it is perfectly safe. Almost no drug is. All antibiotics have harmful effects, some rather severe, and sometimes can both, suppress and promote cancer via different mechanisms (chloramphenicol comes to mind). Other drugs, like cisplatin for cancer treatment have a risk to promote secondary cancer, yet without, folks might succumb to the first cancer. 

All it really means is that you should only take the drug when needed. And this is why also vaccines are so important, because for most, the risk of adverse effect is way lower than therapeutic intervention. And besides, there are also chemicals that our body synthesizes and needs, which are anticipated to be carcinogenic, you cannot get rid of those, either. 

[StringJunky]

17 hours ago, Alfred001 said:

Had to drop this debate due to being busy, but reviving it...

So, here's a review article making my point:

https://pubmed.ncbi.nlm.nih.gov/12052431/

Another one

https://pubmed.ncbi.nlm.nih.gov/29595104/

 

Isn't the acid-test for efficacy and safety about how many people it helps vs doesn't?

[Alfred001]

45 minutes ago, CharonY said:

I think you are still missing the point regarding how risk is evaluated. The reason why it is controversial is because animal studies suggest an effect, but so far no evidence in humans have emerged.

BRO, this symptom again? Ignoring things that were said 4-5 posts ago. Let me quote again:

Quote

Several retrospective cohort studies have reported MTZ as a potential risk factor for lung cancer (n = 771), cervical cancer (n = 2500), breast cancer (n = 2), cholangiocarcinoma (n = 1), and neuroblastoma (n = 28). So far, all the reported data have confirmed MTZ carcinogenicity in animals; however it is still controversial in humans.

Just peruse the study at least, just for the love of god, don't ignore what was said within 10 posts ago.

https://pubmed.ncbi.nlm.nih.gov/29595104/

45 minutes ago, CharonY said:

And as also said before, the real benchmark is whether a treatment provides a net benefit over a disease, and not whether it is perfectly safe.

We've been over this a MILLION times. We CANNOT know whether it provides a net benefit if we don't know what it's risk is. You cannot compare the risk of the disease to the risk of cancer from metro if you don't know what the risk of cancer from metro is. Like the review article said, there is inadequate evidence to establish whether metro increases risk of cancer and by how much. Really, I cannot repeat the same argument any more times.

If the risk of driving without a seat belt is x and the risk of taking a mystery pill is unknown, you cannot know which is worse.

But let's just table this point, can I ask you to please do that, just to make the discussion simpler? I ACKNOWLEDGE and RECOGNIZE that it is a matter of risk vs reward, ok, now, let's set that point aside and nevertheless discuss the matter of whether metronidazole in H pylori type course (2 weeks, 500-1500 mg per day) increases risk of cancer.

Edited November 2, 2023 by Alfred001

[StringJunky]

12 minutes ago, Alfred001 said:

BRO, this symptom again? Ignoring things that were said 4-5 posts ago. Let me quote again:

Just peruse the study at least, just for the love of god, don't ignore what was said within 10 posts ago.

https://pubmed.ncbi.nlm.nih.gov/29595104/

We've been over this a MILLION times. We CANNOT know whether it provides a net benefit if we don't know what it's risk is. You cannot compare the risk of the disease to the risk of cancer from metro if you don't know what the risk of cancer from metro is. Like the review article said, there is inadequate evidence to establish whether metro increases risk of cancer and by how much. Really, I cannot repeat the same argument any more times.

If the risk of driving without a seat belt is x and the risk of taking a mystery pill is unknown, you cannot know which is worse.

But let's just table this point, can I ask you to please do that, just to make the discussion simpler? I ACKNOWLEDGE and RECOGNIZE that it is a matter of risk vs reward, ok, now, let's set that point aside and nevertheless discuss the matter of whether metronidazole in H pylori type course (2 weeks, 500-1500 mg per day) increases risk of cancer.

Do you feel this drug is getting too much of a pass regarding the cancer risk?

[CharonY]

You keep repeating the assertion that we need to know absolute risk levels, but as I mentioned many, many times, this is not how it works. You look at whether folks taking a drug have worse or better outcomes, as I and SJ have been saying. 

Studies have shown that H. pylori eradication on average reduces gastric cancer risks. I.e. depending on what resistances are present, treatment with metronidazole or clarithromycin are often indicated. 

https://doi.org/10.7326/0003-4819-151-2-200907210-00009

https://doi.org/10.1053/j.gastro.2016.01.028

In other words, as long as there is no other study suddenly figuring out that metronidazole is somehow responsible for more cancer cases that we have seen so far (I believe we have talked about issue with finding negative results) the risk assessments suggests that treatment is more beneficial.

[Alfred001]

50 minutes ago, StringJunky said:

Do you feel this drug is getting too much of a pass regarding the cancer risk?

Yeah, the situation right now seems to be that there's evidence suggestive of a carcinogenic effect in humans (in vivo). Some studies find no effect, but others do find an effect. As the review article states, it's a controversial question and it seems to me that it's crazy and irresponsible to be so widely prescribing a drug which seems to have a carcinogenic effect of unknown potency.

Right now metro is part of the regimen that most guidelines recommend as the first line treatment for H pylori and the tendency is to prescribe increasingly higher doses (because metro resistance is becoming increasingly prevalent and you need a higher does or duration to overcome it).

[CharonY]

As you might have missed it, matched studies in the 90s short-term treatments in children did not find an effect. Conversely, long-term treatment with a rather wide range of antibiotics have been associated with increased cancer risk (in part because of how they affect our gut microbiome). As such, I am still not sure why you pick out this specific antibiotic, as what we discuss here is applicable to many of the others as well. 

Or again to make the point, no drug is safe, and if you want to be concerned at this level, you should be concerned about all of them.

Perhaps I should summarize it differently. The question is not whether a drug is toxic or potentially harmful. The question is are folks on average better of receiving a treatment or not. In cases of H. pylori infections, especially if folks have other risk factors for gastric cancer, the answer seems to be yes.