It depends on what you want to infer from the result. For instance, if you want to infer HSIL-CIN-2/3 from the result, the gold standard that you will be comparing to is the histology. If you want to infer the HPV infection status, your gold standard will be an assay that detect the virus (antigen or nucleic acid). You can even set the colposcopy result as your gold standard if your research question is about visible lesion detected by colposcopy.

Gold standard is not the “truth”. As a researcher, we don’t really know the truth. We just accept or reject our hypotheses in the hope that they will infer the truth. I am a pathologist and I never dare to say that my histologic diagnosis is the truth, but it is just my interpretation that I truly hope it will closely represent the truth as much as possible.

Thank a lot for responding. I just have one clarification or help needed.

I am trying to convince my supervisors that I don't need to send tripple negative ( pap,p16 cytology and HPV) to colposcopy and biopsy because the probability of disease in triple negatives is almost zero. So he is insisting that I won't have true negatives

My research is on p16 cytology as a screening test for cervical precancerous lesions

.

My argument is as follows:

Say for instance when I am doing the screen tests( pap,p16 and HPV) and one of them comes positive,I send them for biopsy.now if p16 is negative and the other two are positive ,we send them for biopsy and histology. On histology the result also comes out negative for the disease. Isn't this true negative?

Example 2 : if p16 comes out positive and the histology is negative. Isn't this false positive?

And so on....

This is for sensitivity and specificity where we want to fill up all four columns of true pos,true neh,false neg,false pos....

Why should u send all 3 negs for biopsy and give the disease free women unnecessary trauma...

Is my argument valid?

Your advice would help.. regards

Nikhat hello,

I just read your argument.

The quick answer is that it is not valid.

The explanation of why that is the long bit.

Here is the supporting material:

http://www.scielo.br/pdf/medical/v4n6/2358-0429-medical-04-06-20170601.pdf

https://www.ncbi.nlm.nih.gov/pubmed/28130839

https://www.ncbi.nlm.nih.gov/pubmed/21840041

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4618010/pdf/ACP-2007-29-1-868952.pdf

https://www.immunopaedia.org.za/immunology/archive/immune-evasion/keratinocyte-immunity/carcinoma-development/human-papilloma-virus-hpv/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4509451/

https://www.nature.com/articles/488S14a

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5576325/

https://onlinelibrary.wiley.com/doi/epdf/10.1002/ijc.28514

https://www.archivesofpathology.org/doi/10.5858/arpa.LGT200570?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%3dpubmed

www.ncbi.nlm.nih.gov/pmc/articles/PMC1581465/#B3

An interesting article:

https://www.pnas.org/content/109/26/10516

I hope that helps :)

Another very good source - a Dutch thesis on bio-markers for cervical neoplasia: https://www.rug.nl/research/portal/files/13997875/Complete_dissertation.pdf

Thankyou ma'am . My question on a brief note was

If a woman has three negative tests( pap,p16 cyto and HPV) she can't be sent for biopsy as it's unethical cz the probability of disease here is zero.

So how do I measure sensitivity and specificity of p16? If I don't have disease negative (true negs) women

I need to have p16 negative women who will come out histology negative to be my true negatives . These would be my disease free women but here it's different because atleast either HPV or Pap would be positive ,opposed to all 3 being negative only p16 is negative

So are these women not my true negs for p16?

I am reading your comment here now.

Yes, Nikhat. Both Tauangtham and I have understood your question the first time around.

It seems though that you did not read carefully the scientific links that I have sent you.

If you had, you would have understood that it is long established that there is no such thing as a 100% fool-proof diagnosis of benign, pre-cancerous or cancerous histology, based on three imperfect tests, the scope of which (esp. sensitivity & specificity) we still explore, and that are, therefore combined to complement each other.

In this format (and without knowing the life style, the medical history or the current physiology of the women that are being tested), these three tests used once will never give a 100% negative testing results. Your supervisor is correct.

You are testing living creatures with complex anatomy and physiology, whose body condition changes dynamically. Anomalous cytology is present at all times, but some of it is successfully cleared by the immune system, while other is not.

It takes time for persistent anomaly to develop to a degree sufficient to diagnose in histology.

Testing negative for diffuse p16 today does not automatically mean health, or that the same woman is going to test negative for diffuse p16 again in three-months' time:

www.ncbi.nlm.nih.gov/pubmed/16462152

Diagnosis would also depend on lab guidelines, as well as patho-histologist's personal judgement and experience.

The last link I have sent you is a dissertation that discusses new DNA methylation markers to try and better the success rate of the three testing tools you work with.

As a physician, I will argue that (and I have already told you via messaging) in the face of:

-- a pronounced lack of certainty on how many diseases develop, and

- imperfect test data on patients' condition,

it is unethical (to recommend to patients or) not to argue in favor of exploring all avenues (even invasive ones) to:

- make an informed diagnosis, (and by doing so)

- save patients more anxiety, suffering and stronger medication (with negative side-effects) due to delayed/incorrectly made diagnosis, and

- ensure that patients stay alive (by detecting pre-cancerous and cancerous conditions early)

I hope this brings further clarification.

Thankyou ma'am.i understood it much better here . :) I now understand where my supervisor is comming from .

I am not sure, if I understand the question properly.

Is the question how to find control group of cases with cytology and histology in p16 a HPV negative patients?

If so, I would try to convince cooperating clinitians to send cytology specimens from patients who are going to hysterectomy from some reasons unrelated with dysplasia (f.e. uterus myomatosus...)