Optimizing secondary prevention of cervical cancer: Recent advances and future challenges

Although human papillomavirus (HPV) vaccines offer enormous promise for the ultimate control and possible elimination of cervical cancer, barriers to uptake and coverage of the vaccine both in high‐ and low/middle‐income settings mean that advances in secondary prevention continue to be essential to prevent unnecessary deaths and suffering from cervical cancer for decades to come. While cytology (the Pap smear) has reduced cervical cancer incidence and prevalence in jurisdictions where it has been systematically implemented in population‐based programs—mainly in high‐income settings—limitations inherent to this method, and to program delivery, leave many women still vulnerable to cervical cancer. Recent evidence has confirmed that screening based on HPV testing prevents more invasive cervical cancer and precancerous lesions, and offers innovative options such as self‐collection of specimens to improve screening uptake broadly. In this paper, we review key advances, future opportunities, and ongoing challenges for secondary prevention of cervical cancer using HPV‐based testing.


| BACKGROUND
With the approval of the 9-valent human papillomavirus (HPV) vaccine,we move a step closer to the elimination of cervical cancer. 1 However, eliminationwill only be possiblewith high rates of vaccine uptake foryoung girls across the globe, in all regions and countries.Todate,addressingbarrierstoHPVvaccineuptake,such as acceptability, cost, and program infrastructure, remain a significant challenge for most countries, particularly in low-and middleincomecountries(LMICs).ArecentreviewofHPVvaccinationfound that globally, only 32.1% of girls aged 10-20years in high-income settings and approximately 0.3% in low-and middle-income settings-where cervical cancer rates remain highest-have had an HPVvaccinationseries. 2 Anadditionalchallengeisthatthevaccine preventsviral infection at relativelyyoung ages, and this infection resultsincancersthatoccurdecadeslater. 3 Thus, any woman who did not receive thevaccine prior to infectionwith oncogenic HPV typesmaybeatriskfordevelopingcervicalcancer,andsherequires accesstoeffectivesecondaryprevention-i.e.screening,withtreatmentwhenneeded.Withouteffortstoimproveboththequalityof, andaccessto,screening,itisestimatedthatthenumberofwomen developing cervical cancer annually worldwide will rise to more than 700 000 by 2030. 4   While co-testingwith both cytology and HPV tests is an option for screening programs, studies have confirmed that there is limited benefit from adding cytology to HPV screening. Long-term studies from Kaiser Permanente that included over 1 million women found thatHPVtestinghasaveryhighnegativepredictivevalueforprecancerous lesions. 11

| IMPLEMENTING HPV TESTING IN LOW-RESOURCE SETTINGS
While there is now consensus that the primary screening method forcervicalcancershouldbeHPVtestingwherepossible,numerous questions remain regarding the implementation and optimization of theseprograms.InLMICs,whileHPVtestingofferstheopportunity for very accurate once-in-a-lifetime screening, there are significant infrastructure requirements to implement molecular-based testing effectively.Regionsneedtoexaminehowtobestfacilitateaccessto thisveryeffectivetoolforcancerprevention.
As noted above, HPV testing has a very high negative predictive value,whichmeansthatscreeningintervalscansafelybeextendedfor women who are HPV negative. In fact, there are unintended consequencesifscreeningisdonetoofrequently,asincidentHPVinfections that are likely to clearwithin a 3-4-year time period may be detected andleadtounnecessaryandpotentiallyharmfultreatment. 16 As such, it isessentialthatprogramsensurethatscreeningwithHPVtestingisnot donetoofrequently.The2014guidelinesfromWHOrecommendHPV screening every 5 years, 12 and the 2016ASCO guidelines recommend 2-3screeningsinthelifetimeusingHPVtestingifresourcesarelimited. 13

| SELF-COLLECTION OF SAMPLES
As with HPV vaccination, high coverage with screening is required inthepopulationatrisktoreducetheincidenceandmortalityrates for cervical cancer. Unfortunately, in many jurisdictions, even with longstanding screening programs, coverage rates remain very low (10%-20%inthepoorestwealthdeciles),resultinginsocioeconomic, geographical,cultural,andeducationaldisparitiesevidentinscreening rates and in follow-up for evaluation and treatment. 17 A critical advantage of HPV testing over cytology is the capability for using self-collected vaginal samples, which represents an opportunity to offer solutions to these coverage disparities. For many women, in bothhigh-andlower-incomesettings,attendingclinicalappointments can be challenging owing to childcare, work schedules, and transportation requirements. In addition, for many women, undergoing apelvicexaminationispersonallydifficultowingtoculturalbarriers suchasexposingprivategenitalareastoanunknownevaluator,pain experienced in previous evaluations, or past history of abuse. 18,19 Self-collection has already proved to have substantial acceptability for women across a variety of jurisdictions and has demonstrated impressive diagnostic accuracy compared with clinician-collected specimens. 20,21 Even though there is no test yet approved for use with selfcollectedvaginalsamples,dataaboutthesensitivityandacceptability ofself-collectionamongwomenhavepromptedseveralcountriesto moveaheadwiththatstrategy. 20 Inameta-analysis,self-collectionhas been found to significantly increase participation in cervical cancer screeninginwomenwhodidnotroutinelyattendscreeningprograms previously. 22 Self-collectionoffersanopportunitytoengagewomen, asitcanbeofferedinavarietyofsettings,includingathome,communitycenters,placesofgathering,andclinicalsites.Self-collection with the self-collection strategy were never screened in the past. 27 ExperienceinUgandawithcommunity-basedself-collectionhasalso confirmedsubstantialacceptability,withuptakeratesofgreaterthan 95%comparedwithattendanceof48.4%forscreeningbyVIA. 4

| PERFORMANCE OF HPV TESTING
While HPV testing offers vastly superior sensitivity than cytology for detection of precancerous lesions, its lower specificity means that there could be an increased number of unnecessary diagnostic or treatment procedures. 28,29 Additional tests can triage HPV posi-tiveresultstoimprovespecificitybydelineatingwhichHPV-positive womenrequirefollow-up. 18

| QUALITY ASSURANCE
Comprehensive quality assurance programs, particularly in lowresourcesettings,areacrucialconsiderationfortheimplementation ofHPV-basedcervicalcancerscreening.Suchprogramsshouldmirror bestpracticesforotherlaboratoryprogramsandencompassaspects such as training, laboratory processing, and overall program implementation. All relevant tiers of healthcare personnel should receive appropriate training, including how to collect and report specimens, transportandstorespecimens,conductlaboratorytesting,anddeliver results. Once trained, healthcare personnel should undergo regular proficiency assessment to ensure they are performing their work in accordancewithtestrequirementsandnationalguidelines.
Test instruments, reagents, laboratory supplies, and specimens needtoperformproperlyinthetestenvironment,whichisaparticularly important consideration for low-resource settings. Screening programs should ensure that the laboratory environment meets the manufacturer requirements for their specific test (e.g. consistent supply of electricity, lab temperature, and humidity requirements).
Programsneedtodeveloptoolstoensureadherencetostorageand handling instructions and to procure the correct supplies. Quality assurance of overall program implementation should consider the entiretestcontinuumandmonitorkeystepsandprocessestoensure high-quality testing and accuracy of results delivered to patients.
Standardoperatingprocedures,checklists,andothertoolswillassist program managers in identifying and monitoring these areas of importance.

| COMMUNITY AND PROVIDER EDUCATION
Cervicalcancerscreeningwithcytology,asalong-establisheddisease preventionactivity,iswellentrenchedinmanyjurisdictions.Theshift from cytology to HPV testing will be a significant change-from an oncologic screening paradigm to a communicable disease paradigm.

| SCREENING PROTOCOLS IN SPECIFIC POPULATIONS
Although there is considerable room for improvement, jurisdictions globally are starting to implement population-level HPV vaccination programs for girls. 2  Ultimately,toachieveourgoalofcervicalcancerelimination,we need to reach the point when HPV vaccination coverage is equitablyhighacrossallregionsandallbirthcohorts.Overtime,moreand more women in the age span appropriate for cervical cancer screening willhavebeenvaccinatedagainstHPV,soscreeningpolicieswillhave to be revisited when this ideal scenario is reached. 33 The reduction intransmissionofthevaccine-targetedHPVtypesinthepopulation and the consequent rarity of cervical lesions caused by these types willleadtoachangeinthebalanceofbenefitstoharmsfromcervical cancerscreening,evenwiththeimprovedmoleculartechnologiesnow coming of age. 34 Afuturechallengetothisendisdefiningtheresearch agendafordecidingontheoptimaltechnologyandintensityofscreeningtomakecervicalcanceraveryrarediseasewhileminimizingthe possibleharmsfromscreeningforaveryrarecondition.

ACKNOWLEDGMENTS
The authors would like to thank Heather Pedersen for her assis-tanceineditingandcompilingthereferencesforthispaperandRose SlavkovskyatPATHforherinputsonHPVtestingqualityassurance.

CONFLICTS OF INTEREST
The authors have no conflicts of interest to declare. In terms of involvement with industry, GO has not received industry grants, funds, or honoraria. However, her co-investigators have received