A2g is right. You cannot cherry pick 900 samples from known HPV...

A2g is right. You cannot cherry pick 900 samples from known HPV cases with presumably very high virus titres, run the test and proclaim mission accomplished.

It’s more than likely the 900 test samples will contain over 134 CIN 2+ samples, the question is not, can you pick up a bunch of known CIN 2+ samples but can you accurately pick out the 134 CIN 2+ cases in a patient trial which will require approx 6700 patients to have statistical merit.

The 900 patient trial should be regarded as proof of concept rather than as a clinical validation study.
Still it’s important to do this type trial before embarking on larger clinical trials which will have the necessary statistical power to validate the GBI test.

It’s unlikely that HPV genotyping will be used as a primary screen for cervical cancer any time soon.
I agree it will probably happen, but the change from Pap/Cytology testing to HPV testing as a primary screening test for cervical cancer in developed countries, with good health care systems and Pap/Cytology screening programmes is many years off yet.