Primary HPV testing for cervical cancer -- interim guidelines

Interim guidelines were just released on primary HPV testing in women for cervical cancer (ie, without doing a pap smear at the same time), from the Society of Gynecologic Oncology and the American Society for Colposcopy and Cervical Pathology, and including experts from the American Cancer Society and American College of Obstetrics and Gynecology (see hpv based screening gyn oncol  2015 in dropbox, or doi.org/10.1016/j.ygyno.2014.12.022​).  their conclusions were largely based on a prospective company-sponsored cohort study of 47,208 US women, finding that for primary HPV testing as compared to cytology alone (respectively):

    ​--the sensitivity for >CIN2 was 58.26% vs 42.63%

    --the risk of >CIN2 (positive predictive value) was 12.25% vs 6.47%

    --the risk of >CIN2 in women who were not referred to colposcopy was 0.42% vs 0.59%

    --the false positive rate for >CIN2 was 4.09% vs 6.04%

    --the improvement in the sensitivity and negative predictive values of HPV testing diminished with age and was not significant in those 50 yo and older

    --further analysis found that just doing primary HPV testing in those 25 yo and older was also superior to the current algorithm with cotesting (cytology on everyone starting at age 21 and adding HPV testing in those >30 yo or with abnormal cytology) for both those women with >CIN1 and those >CIN2 (the differences between primary HPV screening and cotesting were still statistically significant though of lesser magnitude than compared with just doing cytology alone, though this analysis compared primary HPV testing every 3 years with cotesting every 5 years​.)

here is the link to the FDA documents leading to the approval of primary HPV testing (open theFDA Executive Summary) --

http://www.fda.gov/AdvisoryCommittees/CommitteesMeetingMaterials/MedicalDevices/MedicalDevicesAdvisoryCommittee/MicrobiologyDevicesPanel/ucm388531.htm . their findings:

--a negative hrHPV (high-risk HPV) screen provides greater reassurance than negative cytology of not having a CIN3 or higher risk, confirmed in several trials (including European trials)

--primary hrHPV screening can be considered an alternative to either cytology alone or cotesting (cytology plus hrHPV in those 30 yo and older)

--those positive for HPV 16 and 18 should be referred for colposcopy

--those positive for the other hrHPV genotypes tested should get cytolology, with referral for colposcopy if abnormal (ASC-US or more) or followup in 12 months if normal cytology

--data on the time to rescreen is a bit muddled: should be no sooner than every 3 years. some data suggests that 5 years is okay, but data are insufficient to justify that recommendation at this time

--initiate hrHPV screening at age 25 (or 3 years after last cytology, for the group who already began screening at age 21). there was concern about the higher false positive rate for HPV testing in those 25-30 (cotesting, in the old recommendation, started at age 30 because of the combination of high false positive HPV testing in younger women and the fact that progression to cancer is quite uncommon in those <30 yo)

--although there are 4 FDA-approved HPV tests, only the cobas test by Roche is approved for primary HPV testing.

--their conclusion: "primary hrHPV​ testing can be considered as an alternative to current US cytology-based cervical cancer screening approaches including cytology alone and cotesting".

so, there are definitely issues here, some elucidated in prior blogs on this issue (see below). 

--one issue is performing hrHPV testing in those 25-30 yo. will there be more false positives leading to more colposcopies, and their attendant morbidities and costs? perhaps we should we continue with cytology til age 30 then switch to primary hrHPV testing? 

--if the testing interval is every 3 years, there will be lots more tests done, leading to both more pelvic exams (patient discomfort) and cost. one issue noted above is the improved performance of hrHPV every 3 years over cotesting every 5 years -- this raises the question as to whether the older recommendations of cotesting 5 years is adequate  -- ie, should cotesting be every 3 years, especially in those at higher risk??? 

--also, since the utility of HPV testing decreases by age 50, should we revert to cytology alone, which is much cheaper?

geoff

link: https://www.dropbox.com/sh/0bmvtita8mzms11/XDTwHySFFg

go to "clinic", then to either "clinical stuff" for articles, or "powerpt presentations" for the powerpoint presentations

please feel free to circulate this to others

here is blog from 4/24/2014:

The FDA approved an HPV DNA test for women >25 yo today  (see http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm394773.htm ). the approved test (cobas test, by Roche) tests for 14 high-risk HPV types, including HPV 16 & 18. the FDA, citing a study by Roche, notes:

    --women positive for HPV 16 or 18 should go directly to culposcopy (HPV 16 or 18 are responsible for 70% of cervical cancer)

    --women positive for the other 12 types should get a regular Pap test to see if need culposcopy

    --although initially approved by the FDA in 2011 as a co-test with Pap, this approval now allows HPV testing to be either part of a co-test with Pap, or a primary screening test.

    --the Roche study involved 40K women > 25yo, doing culposcopy/cervical biopsy on those with positive Pap or HPV as well as some women negative for both, with 3 year follow-up, leading to the FDA approval

so, raises several questions (i have not seen the Roche study, so cannot comment further on that). given that 90% of HPV infections are transient and given the high frequency in women <30yo (as well as the increased likelihood that the infection would be transient), the previous guidelines were to avoid HPV testing til age 30. will lowering it to 25, as per the FDA, lead to more unnecessary invasive procedures in those 25-30 years old? it certainly will help Roche's bottom line...

here is blog from 3/15/14:

NY times reports unanimous vote by FDA subcommittee to replace pap smears with HPV testing alone. prior to this being a new policy, has to be approved by FDA (which almost always happens), though many providers may not use this test alone unless approved by professional societies. the initial test would be done at age 25. for full article: http://www.nytimes.com/2014/03/13/health/an-fda-panel-recommends-a-possible-replacement-for-the-pap-test.html?_r=0 

of note, 

    --approx cost (in houstin texas): $50 for pap, $150 for HPV

    --only one of the hpv tests, the one by roche (not the most common one done), is approved

    --the current guidelines (pap age 21-30 q3yrs, then combo pap/HPV after age 30 q5yrs), have consciously postponed hpv screening til age 30 because in younger sexually active women, these infections most often regress spontaneously and, if picked up by screening, could lead unnecessarily to more invasive followup procedures (eg culposcopy and maybe more).

will also append a prior blog below, sent out last year.

An article this week in the lancet found at that HPV-based cervical cancer screening was better than cytology-based methods (see hpv based screening for cervical cancer lancet 2013 in dropbox, or http://dx.doi.org/10.1016/S0140-6736(13)62028-0).  They looked at four randomized trials in Europe, with 176,000 women aged 20-64 assigned to HPV-based screening (in 3 of these studies it was combination cytology plus HPV) versus cytology screening alone.  Women were followed on average of 6.5 years to assess the development of invasive cervical carcinoma, with a screening interval of 3 years for those with a negative result.  107 invasive cervical cancers were found.  Although there was not any difference in the first 2-1/2 years of followup, thereafter it was found that HPV-based screening provided 60-70% greater protection against invasive cervical carcinoma as compared with cytology. their finding was even more significant in women in the 30-34 age range, an unanticipated result. They felt their data supported HPV screening from age 30 with screening intervals of at least every 5 years. 

 

Results from this study reinforce the decision at our Health Center (as per ACOG and USPSTF) to screen every 5 years for women aged 30 or older with HPV-based screening.  However, as recommended, we do the combination of cytology with HPV screening.  the study authors do raise the issue "because cotesting leads to many unnecessary colposcopy procedures, stand-alone HPV testing also seems recommendable". so, maybe we should be doing only HPV testing in women over the age of 30??

 

this article also raises peripherally the issue of cervical cancer screening in a woman who has never been sexually active, or continued cervical screening on patients who have had negative prior HPV screening and are no longer sexually active (the subtext here is: cervical cancer is basically a sexually transmitted disease caused by certain specific strains of HPV).  i personally do not perform cervical cancer screening in women who have never been sexually active (the issue here, though, is to be certain that the woman really has not been sexually active). for women who have had adequate negative screens (eg, normal HPV screens twice in the preceding 10 years), and have not been in a sexual relationship for several years and with a very high probability of not getting into one (as with many of our patients from some different cultural backgrounds), it seems reasonable to me to stop screening. (and, not so surprisingly, when i ask if they would like a vaginal exam, the answer is almost always a resounding “no”). this is not the recommendation of ACOG or USPSTF, but seems to make sense based on the information at hand.