HPV based screening is slowly but steadily replacing cytology as the primary cervical screening test. It offers much higher sensitivity but lower specificity, largely due to transient infections with minimal progressive potential, so that some form of immediate triage on the same specimen is desirable to better identify those women who are most in need of direct referral to colposcopy. Where good cytology is available one option is reflex cytology. When this shows high grade changes, immediate referral for colposcopy is warranted, but lower grade cytological abnormalities still carry a high false positive rate even for HPV positive women.
A range of ‘molecular’ tests are now under evaluation to try to improve discrimination. Of these some form of HPV genotyping has been most fully investigated, but usually this has been limited to types 16 and 18. There is emerging evidence that fuller typing provides useful additional information and that types 31 and especially 33 carry a much higher risk that other types, and that types 39,56,59,66 and 68 carry lower risk and could usefully be considered ‘intermediate risk’ types. Types 18 and 45 do not have a high PPV for CIN2+, but are more related to invasive cancer and lesions in the endocervical canal, and deserve a different management with more emphasis on repeat testing to establish persistence. New evidence suggests that measures of viral load have type specific relevance and can add information about the likelihood of a high-grade precursor lesion. In addition measures of DNA methylation in both host and viral genes look promising, as does use of immunohistochemical and possibly RNA measures of p16, while tests of E6 and E7 protein levels still appear to lack sensitivity.
Data on these markers will be reviewed from a range of different screening based studies.
Molecular testing has much to offer to improve efficency of HPV related primary screening.