[ejstronge]

> You can develop cervical cancer via some other route, but the vast, vast majority of cervical cancers are caused by HPV infection.

What are these other ways? There's an intuition that bodies are like computer programs that can fail in unpredictable ways, but this is usually false and belies a failure to see links between 'novel' and previously described mechanisms.

[necovek]

Computer programs always fail in predictable — but usually unpredicted — ways.

Human bodies, not so much, mostly because we lack the capability to monitor, measure and emulate the behaviour of such complex systems. As such, we gain medical knowledge using statistics, usually covering most common "failure modes" first, but we increasingly learn that those are never as clear cut either as our observation technology improves (as it does with science otherwise too — eg. Newtonian mechanics is completely true up to some error bars and constraints achievable in that period).

[epcoa]

Spontaneous mutations? Which no matter how much you carve out modifiable risk factors will always be a thing. At least 5% of cervical cancers are HPV negative, so it's not even all the uncommon (not I would call "vast, vast" IMHO)

[ejstronge]

> Spontaneous mutations? Which no matter how much you carve out modifiable risk factors will always be a thing. At least 5% of cervical cancers are HPV negative

Random mutations causing cervical cancer essentially does not happen - as a sibling commenter writes, well-studied cases of this are so rare that they’re below our sensitivity of detection/technical error rates.

This is what I mean when I say we try to apply our intuitions to medicine - they’re not reliable and the truth is idiosyncratic.

Because our prior for cervical cancer being caused by HPV is so incredibly high, we would require overwhelming evidence to reject the hypothesis that any new case is due to HPV. There are ways to do this, and, should they be attained, would be published in a reputable journal based on their novelty.

[epcoa]

> as a sibling commenter writes, well-studied cases of this are so rare that

And yet not a single cite in sight. A random commenter on orange site is not evidence

However I know the paper they are referring to - it is from 1999 in J Pathology, famous at the time, and it is woefully out of date.

> they’re below our sensitivity of detection/technical error rates.

Hogwash.

https://www.mdpi.com/2076-0817/14/7/668

> There are ways to do this, and, should they be attained, would be published in a reputable journal based on their novelty.

There are plenty of papers on HPV independent cervical cancer based on actual gene expression methods published in reputable journals in the last 30 years.

[ejstronge]

> > as a sibling commenter writes, well-studied cases of this are so rare that

> And yet not a single cite in sight. A random commenter on orange site is not evidence

At a certain point, one has to take on the obligation of self-education.

> Hogwash.

From the website link you pasted: "Test-Specific Technical Issues - Standard tests (such as PCR or in situ hybridization) may not detect HPV types not included in the test panel (rare or classic non-oncogenic variants)."

> There are plenty of papers on HPV independent cervical cancer based on actual gene expression methods published in reputable journals in the last 30 years.

You don't understand the field, so you don't know why the gene expression does not establish the presence or absence of HPV but instead can be used as an argument for the absence of HPV.

EDIT: Also, a fundamental point of misunderstanding - cervical cancer is not a 'new' disorder, so unless you have an explanation for why the 1999 article is wrong, we can't simply discard its result and favor a next-generation sequencing article.

[epcoa]

Well to bring this back to the original question: "What are these other ways?", how does GCA not qualify?

[tialaramex]

Where did you see 5%? I'm not an expert (though I think I know one) but the study I'm looking at says 0.3% - hence my "vast, vast". That's 3 patients out of 1000.

In that study they take cancer cells and check for HPV, get about 7% negatives which I'm guessing is what you're describing too, but they take those negatives and PCR them to figure out, well, OK, what was wrong with these cells and when you take the cells to pieces very often your assay goes oh, these instructions are HPV. So, you know, the cancer cells aren't "infected" with HPV but well the genetics are just HPV, the replication has gone haywire and tangled parts of HPV with the human cell instructions and now it's cancer.

Crucially we can assume that if you don't get infected with HPV this wouldn't happen. So HPV was still causal.

[epcoa]

> Crucially we can assume that if you don't get infected with HPV this wouldn't happen. So HPV was still causal.

Nope. This is literally “correlation does not equal causation” 101. Based on the 0.3% I’m gonna guess you’re (either directly or indirectly) citing a famous, 1999 paper in J Pathology (Walboomers et al). It’s outdated, missing a control *, and it’s pretty well accepted that just finding a bystander HPV DNA fragment around somewhere is not conclusive of causality. We have much more sophisticated assays of gene expression. Try looking for review articles in the last 3 to 4 years rather than 30, the prevalence of truly HPV independent cervical cancer is not precisely characterized but it’s almost certain much greater than 0.3%.

https://www.mdpi.com/2076-0817/14/7/668

https://journals.lww.com/md-journal/fulltext/2024/10110/rese.... (3% to 8%)

[tialaramex]

That 3-8% cites two sources but...

One of those cited resources just keeps citing other people for a 5% risk, if they came up with their own number I didn't find it and I wonder why they'd cite somebody else in their own abstract without even mentioning they don't agree if somewhere in their work they do get a different number.

The other citation from the second link seems to be a paper which doesn't say 3% it says, and I'll quote: "The main explanation for HPV-negative cervical cancer was a false diagnosis, followed by cancers associated with non-HR-HPV types, and false-negative HR-HPV results. Truly HPV negative seem to be very rare in Caucasian populations".

If they're to be taken for 3% the only way to get there is by disregarding that conclusion and deciding that false negatives count as true negatives. Reviewers should ideally catch that but didn't here.

As I said, I don't doubt it exists, but 8% seems insane and these citations did not persuade me I was wrong to say 0.3% based on the paper you don't like.

[epcoa]

This has nothing to do with "liking" or appeals to emotion. The fact is that paper does not have a control and the methods are not sophisticated enough to determine causation.

"The other citation from the second link seems to be a paper which doesn't say 3% it say". Yes it does dude. You need to read the actual results from the paper more carefully: "Overall, 340/350 cases of primary cervical cancer confirmed by surgical staging tested HC2 positive (97.2%)." Ie 2.8% (~3%) were considered true HPV negative by this testing.

They're going from 8.8% (in that particular admitted biased dataset) to still 2.8%, the wording of the conclusion is wonky, but the results of the paper are overall consistent. You're taking that quote out of context.

In that same paper it says: "Our results are in accordance with The Cancer Genome Atlas Research Network (CGARN) ‘Integrated Genomic and Molecular Characterization of Cervical Cancer Study’, which used next-generation sequencing to characterize primary cervical cancers. The CGARN study found 95% of primary cervical cancers were HPV-positive and 5% HPV-negative."

Which is one of the primary sources for the 5% figure.

https://pmc.ncbi.nlm.nih.gov/articles/PMC5354998/

In any case these are both order of magnitude more than 0.3%

I'll agree 8% worldwide is high (though since environmental factors and genetics both play a role in both HPV-positive and negative cases), these incidences can vary throughout the world or within certain subgroups, and if you manage those groups that matters.

https://pmc.ncbi.nlm.nih.gov/articles/PMC11075765/

GAS is a clearly known cervical adenocarcinoma not related to HPV, and it accounts for 20% of all cervical adenoca in Japan, which overall places it close to 5% of all cervical cancer diagnoses there, just for this subtype.

[tialaramex]

> "Overall, 340/350 cases of primary cervical cancer confirmed by surgical staging tested HC2 positive (97.2%)." Ie 2.8% (~3%) were considered true HPV negative by this testing.

That study literally explains each of the ten excess you say are "true negative", and only puts one in that bucket. You've made the same error as the author you cited previously. You're conflating HC2 negative ("it wasn't seen in this common test") with HPV negative ("it wasn't present")

They start with 371 smears and access to the patients, in all 371 cases there was also surgical biopsy.

* 340 of the smears say HPV with HC2. So we're agreed those are HPV positive.

* 21 of the biopsies aren't Cervical cancer. If cancer cells from other nearby tissues spread and and we find that in a cervical biopsy that's extremely bad news but it is NOT cervical cancer.

* 5 have non high-risk HPV. HPV-53, HPV-70, HPV-73

* 4 do have high-risk HPV even though the HC2 assay was negative

* Which leaves just one true negative case.

Now the 5 aren't proof of anything, and you can (and doubtless will) argue that it's just a coincidence. Uncleared infections with HPV aren't rare, so we can't rule that out, but equally it is wrong to insist it must be a coincidence. You'd need a mechanism, present only in the high risk variants to explain why these are innocent bystanders or else that's correlation.

The four high risk cases are even more clear cut, that's simply a false negative. Insisting you get to count these as true negative is crazy. HC2 was not adequate to detect the virus here, these patients were HC2-negative but they had high risk HPV anyway, the HC2 assay just didn't see it.