cervical cancer screening webinar

8/19/2019 Cervical Cancer Screening Webinar

1/53

CERVICAL CANCERSCREENING


2/53

MARCH 15, 2012

• The USPSTF and the ACS (inconjunction with ASCCP and ASCP)released updated cervical cancerscreening recommendations

• Not a coincidence

• Independently developed• Remarkably similar

conclusions/guidelines


3/53


4/53

CERVICAL CANCER

• Histologic types

• Squamous cell• 70% of all cases (primary target of

cytological screening)

• Arises at squamocolumnar junction(transformation zone)

• Primary target of cytology screening

• Adenocarcinoma• ~18%

• Mixed adenosquamous and other


5/53

A reminder: squamocolumnar junction


6/53

CERVICAL CANCERMORTALITY (PER 100,000)

White Non-white Combined

1950(unadjusted)

10.2 18.0

2007(adjusted)

2.2 4.3 2.4

This dramatic decline has beenattributed to the implementation and

dissemination of screening.


7/53

CERVICAL CANCERINCIDENCE & MORTALITY


8/53

BURDEN OF ILLNESS

• SEER data:

• “It is estimated that 12,710 women willbe diagnosed with and 4,290 women

will die of cancer of the cervix uteri in 2011.”

• For comparison, for every woman whowill die of cervical cancer

• 5 will die of colon cancer

• 8 will die of breast cancer

• 15 will die of lung cancer


9/53

INADEQUATE SCREENING

• About half of all cervical cancer deathsare in women who have not beenscreened or who have had incompletefollow-up to screening and treatment

• If we could assure adequate screeningof the entire population, the residualpreventable burden would be small

• What goals should we have for achange in prevention strategy,whether immunization or a change inscreening approach?


10/53

POSSIBLE GOALS FOR NEWCERVICAL CANCER PREVENTION

STRATEGIES (INCLUDING IMMUNIZATION)

• Further reduction in mortality

• Caveat:the elimination of cervicalcancer and/or cervical cancermortality is not a realistic goal ofscreening

• Reduction in the burden and/orharms of screening and treatmentof screen-detected disease


11/53

Draft University of

Missouri

Department of

Family Medicine

updated clinicalalgorithm for

cervical cancer

screening

http://fcm-algo.umh.edu/Algorithms/CCS.htm


12/53

Screening

Evaluation ofabnormal

screen

Follow up post

colposcopy


13/53

SCREENING


14/53

SCREENING


15/53

SCREENING


16/53

HPV INFECTION

• “It is well recognized that infectionwith oncogenic HPV types is anecessary, although not sufficient,

cause of virtually all cervicalcancer.25 “

• Results from a large international

collection of cervical tumorspecimens revealed the presence ofHPV DNA in 99.7 percent of cases.3

http://preview.ncbi.nlm.nih.gov/bookshelf/booktest/br.fcgi?book=es86&part=references.rl1http://preview.ncbi.nlm.nih.gov/bookshelf/booktest/br.fcgi?book=es86&part=references.rl1http://preview.ncbi.nlm.nih.gov/bookshelf/booktest/br.fcgi?book=es86&part=references.rl1http://preview.ncbi.nlm.nih.gov/bookshelf/booktest/br.fcgi?book=es86&part=references.rl1


17/53

HPV INFECTION: NATURALHISTORY

• From HPV infection to cervicalcancer

• HPV transmission,• Acute HPV infection,

• Persistent HPV infection leading toprecancerous changes, and

• ICC.45

http://preview.ncbi.nlm.nih.gov/bookshelf/booktest/br.fcgi?book=es86&part=references.rl1http://preview.ncbi.nlm.nih.gov/bookshelf/booktest/br.fcgi?book=es86&part=references.rl1


18/53

HPV TRANSMISSION

• Primarily as a result of skin-to-skinor mucosa-to-mucosa contact


19/53

HPV INFECTION ANDPERSISTENCE

• A high proportion of sexually activewomen become infected with HPV,but only a small proportion of HPVinfections become persistent

• 91 percent of prevalent HPVinfections clear within 24 months

(including infections with high risksubtypes)


20/53

PREVALENCE OF HPVINFECTION


21/53

WHY NOT SCREEN BEFOREAGE 21?

• Cervical cancer is rare in theyounger age group

Per 100,000 women


22/53

WHY NOT SCREEN BEFOREAGE 21?

• HPV infection is common andresults in transient abnormalities ofthe cervix

• Detection and Rx of thoseabnormalities leads to harm


23/53

WHAT ABOUT SEXUALHISTORY?

• Young women with multiple sexualpartners are the most susceptible tothe harms of screening

• The possibility of benefit isvanishingly close to zero

• Just say no to screening for cervicalcancer before age 21.


24/53

SCREENING


25/53

SCREENING INTERVAL FOR

CYTOLOGY IN WOMEN AGE 21-65

• RCTs of screening programs atdifferent intervals never exist

• e.g. no one has done an RCTcomparing colonoscopy for coloncancer screening every 5 years to

every 10 years or 20 tears (decided toleave in the typo)

• Task Force has used modeling


26/53

WHAT OUTCOME,PARTICULARLY FOR HARMS?

• False positives

• Colposcopies

• CIN 2-3

• Cancer cases, cancer deaths


27/53

HARMS:COLPOSCOPIES

• Pain, bleeding

• Sentinel measure for downstream

harms• Similar to using number of

colonoscopies as sentinel measure ofharm in modeling of colon cancerscreening


28/53

HARMS: OVER-DIAGNOSIS

• CIN2 can/does regress – over-diagnosis and over-treatment arereal risks

• CIN3 can also regress

• Standard of care currently to Rx all

CIN2+


29/53

TREATMENT OF CIN2+

• Common treatments include LEEP orcervical conization

• Short term harms of pain (67%),

bleeding (87%), discharge (63%)• Increased risk of adverse pregnancy

outcomes

• Perinatal mortality, preterm delivery,low birth weight

• Evidence on specific procedures isincomplete and retrospective

MODEL: ENORMOUSLY


30/53

MODEL: ENORMOUSLYCOMPLICATED – EVEN IF YOU LIKE

MATH


31/53

CYTOLOGY STARTING AGE 21,FOLLOWED FOR LIFE (PER 1000)

Q1 Q2 Q3 q5

False

positives

951 515 350 214

Colposcopy 1931 1084 758 483

CIN 2-3 91 88 80 66

Cancer

cases

2.5 5.8 8.5 12.7

Cancer

deaths

0.3 0.9 1.5 2.7

Quote from Task Force member:

“All models are wrong, some are

useful.”


32/53

OBSERVATIONAL STUDY

• Lancet Oncology, vol 12, July 2011

• “Cervical cancer risk for womenundergoing concurrent testing forhuman papillomavirus and cervicalcytology: a population based studyin routine clinical practice”

• Katki et al followed 313,818 womenin Kaiser Permanente NorthernCalifornia


33/53

OBSERVATIONAL STUDY ONINCIDENCE

• 319,177 (96.2%) of women hadnormal Pap at baseline

• CIN3+ at:• 3 years 0.17%,

• 5 years 0.36%

• Risk of invasive cancer at five years

after normal cytology was 7.5 per100,000 women (0.0075%)


34/53

SCREENING INTERVAL FORCYTOLOGY IN WOMEN AGE 21-65

• Cytology every 3 years demonstratesa good balance of benefits and harms

• “Pap smears every three years aresafe and effective at reducing cervicalcancer, while minimizing the risks offalse positive results and the harms

associated with treating disease thatwill go away without treatment.”


35/53

SCREENING


36/53

WHAT IS THE ROLE FOR HPVTESTING IN SCREENING?


37/53

RECALL: PREVALENCE OFHPV INFECTION


38/53

HPV SCREENING BEFOREAGE 30

• Recommend against

• Prevalence is high, therefore falsepositive rate is high

• False positive in this context does notmean the test is positive in someonewho does not have HPV

• False positive means identifying

someone “in need of intervention” toprevent cervical cancer who does notneed that intervention because herdisease will regress spontaneously

Just say no to screening

for cervical cancer withHPV before age 30.


39/53

HPV SCREENING FOR CERVICALCANCER FOR AGE > 30 YEARS

• Multiple studies of varied designdemonstrate that HPV testing is…

• More sensitive than cytology forCIN2+

• Less specific than cytology

• The Task Force had the challenge of

being “moderately certain” aboutthe balance of benefits and harms.


40/53

RCTS OF HPV SCREENINGFOR CERVICAL CANCER

• EPC reviewed and presented theresults of 6 European RCTs thatincluded HPV in some way in the

experimental group• Inconsistent design, varying

protocols, incomplete reporting andperhaps most importantlyincomplete follow-up through tworounds of testing


41/53

DRAFT RECOMMENDATION

• Insufficient evidence to determinethe balance of benefits and harmsof HPV screening


42/53

POST DRAFT

• Two important publications

• Completed follow-up of the secondround of the RCT in the Netherlands

• Kaiser observational data notedearlier in presentation


43/53

POBASCAM

• 44,938 women age 30-56 randomizedto screening with conventionalcytology vs. co-testing with HPV andconventional cytology

• Round two testing in five years – bothgroups received co-testing

• Complex protocol for referral forcolposcopy – does not reflect currentstandard of care in the US

• e.g. immediate referral only for HSIL


44/53

POBASCAM RESULTS

Cumulative CIN2

Cytology round one

127

Co-testing round one

168

Cumulative CIN3 252 243

Round one cancer 6 12

Round two cancer 14 4

Cumulative cancer 20 16

Recall denominator in each group ~20,000


45/53

POBASCAM APPLICABLE TOUS?

• We are more aggressive in use ofcolposcopy, so detection of CIN2+likely to be higher

• Safe to conclude that co-testingevery five years as good as(better?) than cytology every five

years in an RCT

• Reported harms (CIN2) modest


46/53

CYTOLOGY EVERY 3 YEARS VSCO-TESTING EVERY 5 YEARS

• Kaiser observational data

• Further exploration in the model to

try to fill in gaps in evidence fromPOBASCAM and Kaiser


47/53

KAISER DATA

• Cumulative incidence of CIN3+ the same(0.17%) …

• three years after normal cytology and

• five years after double negative co-testing

• Other analyses confirm increased

sensitivity and decreased specificity ofHPV testing relative to cytology

• Did not report total colposcopies


48/53

MODEL DATA

False

positives

Colpos-

copies

CIN2-3 Cancers Cancer

deaths

Cytology

q3 years

350 758 80 8.5 1.55

Cytologyq3 years

until age

30 then

co-testing

q5 years

255 575 84 7.44 1.35

Note: model assumed women with

normal colposcopy immediately returned

to usual screening

WHAT IS THE ROLE FOR HPV


49/53

WHAT IS THE ROLE FOR HPVTESTING IN SCREENING?


50/53

SCREENING


51/53

AGE 65 YEARS OR OLDER

• Potential for benefit in thoseadequately screened in the pastwhose screening tests are normal is

very low, potential for harm at leastsmall

• Note women who have had CIN2+should continue to be screened for 20

years• Consider screening women who do nothave a history of adequate screening


52/53

TWO IMPORTANT CHANGES

• USPSTF did not addressmanagement of abnormal results –but ACS/ASCCP did make two

specific recommedations

• ASCUS/HPV negative – Rx as normal

• Negative Cytology/+HPV

• Repeat in one year and colpo if either ispositive, or…

• Test for HPV 16/18 and colpo if positive


53/53

QUESTIONS?

cervical cancer screening webinar

Documents