American Cancer Society Guidelines for the Early Detection of Cancer, 2004

2002;52;342-362 CA Cancer J ClinSmith, Harmon J. Eyre and Carmel Cohen

Debbie Saslow, Carolyn D. Runowicz, Diane Solomon, Anna-Barbara Moscicki, Robert A. Cancer

American Cancer Society Guideline for the Early Detection of Cervical Neoplasia and

This information is current as of January 18, 2007

http://caonline.amcancersoc.org/cgi/content/full/52/6/342the World Wide Web at:

The online version of this article, along with updated information and services, is located on

http://caonline.amcancersoc.org/subscriptions/individuals only): , go to (USCA: A Cancer Journal for CliniciansTo subscribe to the print issue of

ISSN: 0007-9235. Online ISSN: 1542-4863. Clifton Road, NE, Atlanta, Georgia 30329. (©American Cancer Society, Inc.) All rights reserved. Print

is owned, published, and trademarked by the American Cancer Society, 1599CANovember 1950. Lippincott Williams & Wilkins. A bimonthly publication, it has been published continuously since

is published six times per year for the American Cancer Society byCA: A Cancer Journal for Clinicians

by guest on January 18, 2007 (©A

merican C

ancer Society, Inc.)

caonline.amcancersoc.org

Dow

nloaded from

http://caonline.amcancersoc.org/cgi/content/full/52/6/342

http://caonline.amcancersoc.org/subscriptions/

http://caonline.amcancersoc.org:80

342 CA A Cancer Journal for Clinicians

ACS Guideline for the Early Detection of Cervical Neoplasia and Cancer

INTRODUCTION

Cervical cancer mortality in the United States has decreased over the last fivedecades by over 70 percent in large part attributable to the introduction of thePapanicolaou (Pap) test. Cervical cancer, once the number one cancer killer ofwomen, now ranks 13th in cancer deaths for women in the United States.1 Ascervical cytology screening has become more prevalent, preinvasive lesions of thecervix are detected far more frequently than invasive cancers. In 2002, an estimated13,000 cases of invasive cervical cancer will be diagnosed, and an estimated 4,100women will die from this disease.Women with preinvasive lesions have a five-yearsurvival rate of nearly 100 percent.When cervical cancers are detected at an earlystage, the five-year survival rate is approximately 92 percent.1

Between 93 and 100 percent of squamous cell carcinomas of the cervix containDNA from high-risk types of human papillomavirus (HPV), which are transmittedduring sexual activity.2,3 Studies of the natural history of cervical cancer have shownthat infection with high-risk HPV types may lead to low-grade or high-gradeintraepithelial lesions. High-grade lesions may progress to cervical carcinoma if not

American Cancer Society Guideline for the Early Detection of CervicalNeoplasia and CancerDebbie Saslow, PhD; Carolyn D. Runowicz, MD; Diane Solomon, MD;Anna-Barbara Moscicki, MD; Robert A. Smith, PhD; Harmon J. Eyre, MD; Carmel Cohen, MD

ABSTRACT An update to the American Cancer Society (ACS) guideline regarding screening

for the early detection of cervical neoplasia and cancer, based on recommendations from a

formal review and recent workshop, is presented. The new screening recommendations

address when to begin screening, when screening may be discontinued, whether to screen

women who have had a hysterectomy, appropriate screening intervals, and new screening

technologies, including liquid-based cytology and HPV DNA testing. (CA Cancer J Clin

2002;52:342-362.)

Dr. Saslow is Director, Breast andGynecologic Cancer, AmericanCancer Society, Atlanta, GA.

Dr. Runowicz is Vice Chairman,Department of Obstetrics andGynecology, St. Lukes-RooseveltHospital Center, New York, NY.

Dr. Solomon is Project Officer,ASCUS/LSIL Triage Study, NationalCancer Institute, Rockville, MD.

Dr. Moscicki is Director, TeenColposcopy Clinic and AssociateDirector, Division of AdolescentMedicine, University of California,San Francisco, CA.

Dr. Smith is Director of CancerScreening, American Cancer So-ciety, Atlanta, GA.

Dr. Eyre is Chief Medical Officer andExecutive Vice President forResearch and Cancer Control,American Cancer Society, Atlanta,GA, and Editor in Chief of CA.

Dr. Cohen is Professor and Director,Division of Gynecologic Oncology,and Vice Chairman, Department ofObstetrics, Gynecology, and Repro-ductive Science, The Mount SinaiMedical Center, New York, NY.

This article is available online athttp://CAonline.AmCancerSoc.org

Author disclosures: R. Marshall Austin, MD, PhD, has served as a consultant and/or speaker without accepting personal compensation for AutoCyte Inc., CytycCorporation, Digene Corporation, Morphometrix Technologies Inc., NeoPath, Inc., Neuromedical Sciences Inc., and Veracel Inc. J.Thomas Cox, MD, has con-sulted for Cytyc Corporation, Digene Corporation, 3M Pharmaceuticals, Inc., and Merck & Co., Inc., and is on the speaker’s bureau of Cytyc Corporation,and 3M Pharmaceuticals, Inc. Jack Cuzick, PhD, is a consultant to Digene Corporation. Eduardo Franco, MPH, DrPH, has occasionally been invited by 3MPharmaceuticals, Inc., GlaxoSmithKline, Digene Corporation, and F. Hoffmann-La Roche Ltd. to serve as a temporary consultant or visiting speaker.WalterKinney, MD, has received research support from 3M Pharmaceuticals and serves on the speaker’s bureau for Cytyc Corporation and Digene Corporation. EvanMyers, MD, MPH, has received research support from and is a consultant to Merck Research Laboratories. Ellen Sheets, MD, was appointed as a vice presidentof Cytyc Corporation in May 2002, following the ACS guideline workshop.Thomas C.Wright, Jr., MD, has received research support from Digene Corporationand Cytyc Corporation and is on the speaker’s bureau of Cytyc Corporation and TriPath Imaging, Inc.


merican C



Dow

nloaded from


treated.4,5 Most HPV* infections, however, aretransient, resulting either in no symptoms orcellular changes or producing low-gradeintraepithelial lesions.6,7,8 The purpose ofscreening, in addition to detecting cervicalcancers at an early stage, is to detect andremove high-grade lesions and thus preventpotential progression to cervical carcinoma.

The very success of the Pap test in cervicalcancer screening has fostered an unrealisticexpectation that the test is perfect. It is not. Paptest sensitivity for high-grade cervicalintraepithelial neoplasia (CIN) is in the rangeof 70 to 80 percent.9 Factors that limit testsensitivity include: small size of a lesion,inaccessible location of the lesion, the lesionnot being sampled, the presence of only a fewabnormal cells on the slide, small size of theabnormal cells, or the presence ofinflammation and/or blood obscuring cellvisualization. False-negative results occur evenin optimized screening programs and cannotbe entirely eliminated.

Approximately half of the cervical cancersdiagnosed in the United States are in womenwho have never been screened, and anadditional 10 percent of cancers occur inwomen who have not been screened withinthe past five years.10 While the new AmericanCancer Society screening guideline includes areview of new cervical cancer screening tests,perhaps the largest gain in reducing cervicalcancer incidence and mortality could beattained by increasing screening rates(regardless of the test used) among womenwho are currently unscreened or screened onlyinfrequently. However, new technologies mayoffer advantages over what is already asuccessful screening test, when utilized.

Guideline Development

The ACS guideline for the early detectionof cervical cancer was last reviewed in 1987.Atthat time it was recommended that all women

who are or have been sexually active, or havereached the age of 18, have an annual Pap testand pelvic examination; after a woman has hadthree or more consecutive, technicallysatisfactory, normal annual examinations, thePap test may be performed less frequently atthe discretion of her physician.11 Thisrecommendation was accepted as policy inidentical or similar wording by the NationalCancer Institute, the American College ofObstetricians and Gynecologists, the AmericanMedical Association, the American Academy ofFamily Physicians, the American MedicalWomen’s Association, and the AmericanNurses Association.

In 2001 to 2002, the ACS convened anexpert panel to review the existing guideline inthe context of evidence that has accumulatedsince the last revision. The panel was dividedinto working groups to review the evidenceand develop recommendations regarding (1)when to start screening; (2) when to dis-continue screening; (3) screening of womenwho have had a hysterectomy; (4) screeningintervals; and (5) screening tests.

During the current guideline review,published articles related to cervical cancerscreening, including new screening tests, wereidentified using MEDLINE (National Libraryof Medicine), bibliographies of identifiedarticles, personal files of panel members,and unpublished manuscripts. Expert panelmembers reviewed articles using specifiedcriteria (Appendix A) and discussed themduring a series of conference calls. Each workgroup developed recommendations, rationale,and evidence summaries, and reviewed thesummaries developed by the other workgroups prior to an April 2002 workshop.Whenevidence was insufficient or lacking, the finalrecommendations incorporated the expertopinions of the panel members. Relevantunpublished manuscripts were distributed toworkshop attendees prior to the meeting.During the conference calls and workshop,

Volume 52 • Number 6 • November/December 2002 343

CA Cancer J Clin 2002;52:342-362

*Unless otherwise specified, HPV refers to high-risk types only throughout this document.


merican C



Dow

nloaded from


consensus was reached on thekey issues within theguideline recommendations.Following the workshop,ACS Gynecologic CancerAdvisory Group membersdeliberated over the guidelinemodifications. Each workgroup member and workshopattendee reviewed the draft ofthis manuscript.

Several organizations re-viewed this manuscript, pro-vided comments, and indi-cated their support of the new recommendations. Theseorganizations include theAmerican College of Obstet-ricians and Gynecologists,the American Social HealthAssociation, the AmericanSociety of Colposcopy andCervical Pathology, theAssociation of ReproductiveHealth Professionals, theGynecologic Cancer Foun-dation, the National Asso-ciation of Nurse Practitionersin Women’s Health, and theSociety of Gynecologic On-cologists.

RECOMMENDATIONS,

RATIONALE, AND EVIDENCE

When to Start Screening

Recommendation

Cervical cancer screeningshould begin approximatelythree years after the onset ofvaginal intercourse. Screeningshould begin no later than 21years of age. It is critical thatadolescents who may notneed a cervical cytology test

obtain appropriate preventive health care,including assessment of health risks,contraception, and prevention counseling,screening and treatment of sexually transmitteddiseases.The need for cervical cancer screeningshould not be the basis for the onset ofgynecologic care.

Rationale

The published and unpublished data on theincidence of cervical cancer and the naturalhistory of HPV infection and of low- andhigh-grade cervical lesions suggest that there islittle risk of missing an important cervicallesion until three to five years after initialexposure to HPV (Evidence section). Thus,cervical cytology screening in adolescents isunlikely to add appreciable benefits within thefirst three years following onset of vaginalintercourse. It is the intent of the ACS thatoffering screening “approximately” three yearsafter the onset of sexual intercourse will avoiddenial of health insurance coverage for teensand young women who undergo their firstcytology test prior to the suggested three years.However, the concern is that screening beforethe three-year-period may result in anoverdiagnosis of cervical lesions that willregress spontaneously, leading to inappropriateintervention, which may result in more harmthan good. Because the risk of HPVtransmission to the cervix is low for othertypes of sexual activity, the onset of vaginalsexual intercourse has been selected as thehistorical marker for initiating cervicalcytology screening.

An upper age limit for when to initiatescreening is needed for providers who don’task patients about their sexual history and foradolescents who are unable or unwilling todisclose prior consensual and/or noncon-sensual intercourse. Such an upper age limitensures that young women, including victimsof sexual abuse, are protected.

In cases where a history of sexual abuse hasbeen established, there is a lack of evidence tosupport earlier cervical screening for victims of



Gynecologic Cancer Advisory Group: Carmel Cohen, MD, (Chair), Professor and Director,Division of Gynecologic Oncology, and ViceChairman, Department of Obstetrics, Gynecology,and Reproductive Science, The Mount Sinai MedicalCenter, New York, NY; Diane M. Harper, MD, MPH,Associate Professor, Departments of Community andFamily Medicine and Obstetrics and Gynecology,Dartmouth Medical School, Lebanon, NH; Joan G.Jones, MD, Professor of Pathology and ClinicalObstetrics, Gynecology, and Women's Health, AlbertEinstein College of Medicine, and Director ofSurgical Pathology, Weiler-Einstein Division,Montefiore Medical Center, Bronx, NY; Heyoung LeeMcBride, MD, Radiation Oncologist, John StoddardCancer Center, Iowa Methodist Medical Center, DesMoines, IA; William McGuire, MD, Director, FranklinSquare Hospital Cancer Center, Baltimore, MD;Edward Partridge, MD, Professor and Vice Chairman,Department of Obstetrics and Gynecology, Division ofGynecologic Oncology, The University of Alabama atBirmingham, Birmingham, AL; Stephen Rubin, MD,Professor and Director, Department of Obstetrics andGynecology, Division of Gynecologic Oncology,University of Pennsylvania Medical Center,Philadelphia, PA; Carolyn D. Runowicz, MD, ViceChairman, Department of Obstetrics and Gynecology,St. Lukes-Roosevelt Hospital Center, New York, NY;Debbie Saslow, PhD, Director, Breast andGynecologic Cancer, American Cancer Society,Atlanta, GA; Diane Solomon, MD, Project Officer,ASCUS/LSIL Triage Study, National Cancer Institute,Rockville, MD.When to Start Screening Work Group:Anna-Barbara Moscicki, MD, (Chair), Director TeenColposcopy Clinic and Associate Director, Division ofAdolescent Medicine, University of California, SanFrancisco, CA; S. Jean Emans, MD, Chief, Division ofAdolescent Medicine, Children’s Hospital, Boston,and Professor of Pediatrics, Harvard Medical School,Boston, MA; Sue J. Goldie, MD, MPH, AssociateProfessor of Health Policy and Decision Science,Harvard Center for Risk Analysis, Department ofHealth Policy and Management, Harvard School ofPublic Health, Boston, MA; Paula Adams Hillard, MD,Professor, Departments of Obstetrics/Gynecologyand Pediatrics, Division of Adolescent Medicine,Children’s Hospital Medical Center, and Director,Women’s Health, University of Cincinnati College ofMedicine, Cincinnati, OH; Luella Klein, MD, Professor,Department of Gynecology and Obstetrics, EmoryUniversity, Atlanta, GA; Mary-Ann Shafer, MD,Professor, Department of Pediatrics, and AssociateDirector, Division of Adolescent Medicine, Universityof California, San Francisco, CA; Debbie Saslow, PhD,Director, Breast and Gynecologic Cancer, AmericanCancer Society, Atlanta, GA; Robert A. Smith, PhD,Director of Cancer Screening, American CancerSociety, Atlanta, GA.Work Group on Interval, Older Women,and Hysterectomy:Carolyn D. Runowicz, MD, (Chair), Vice Chairman,Department of Obstetrics and Gynecology, St. Lukes-Roosevelt Hospital Center, New York, NY; Carol AnnArmenti, Executive Director, Center for CervicalHealth, Toms River, NJ; David Atkins, MD, MPH, ChiefMedical Officer, Center for Practice and TechnologyAssessment, Agency for Healthcare Research andQuality, Rockville, MD; R. Marshall Austin, MD, PhD,Medical Director and Director of Cytopathology andGynecologic Pathology Services, Coastal PathologyLaboratories, SC, and Clinical Associate Professor ofPathology, Medical University of South Carolina, Mt.Pleasant, SC; J. Thomas Cox, MD, Director,Colposcopy and Gynecologic Clinic, Health ServicesDepartment, University of California, Santa Barbara,CA; Jack Cuzick, PhD, Head, Department ofMathematics, Statistics, and Epidemiology, Cancer


merican C



Dow

nloaded from


prepubescent sexual abuse. Abuse victims whohave had vaginal intercourse, especially post-puberty, may be at increased risk of HPVinfection and cervical lesions and should bereferred for screening once they arepsychologically and physically ready (i.e., post-puberty) by a provider who has experience and sensitivity for working with abusedadolescents.

Provider discretion and patient choicefollowing counseling should be used to guidethe initiation of cervical cytology screening inyoung women aged 21 and older who havenever had vaginal sexual intercourse and forwhom the absence of a history of sexual abuseis certain.

Young women who are infected with HIVand/or are immunocompromised shouldfollow the US Public Health System Guide-lines, i.e., obtain a Pap test twice in the firstyear after diagnosis of HIV infection and, if theresults are normal, annually thereafter.12

Evidence

Cervical cancer in women less than 19 yearsof age is rare. Mount, et al.13 provided thehighest reported prevalence (3.77 percent) ofsquamous intraepithelial lesions (SIL) among10,296 cytology smears from patients aged 10to 19 years, 18 percent of SILs were high grade,and no cases of invasive cancer were detected.The National Cancer Institute’s Surveillance,Epidemiology, and End Results (SEER)program reported that the incidence rate ofinvasive cervical cancer was 0/100,000/yearfor ages 10 to 14 years; 0/100,000/year for ages15 to 19 years and 1.7/100,000/year for ages20 to 24 years from 1995 to 1999.14 Based onthe SEER data, the potential implied cost ofmissing high-grade SIL (HSIL) in 15- to 19-year-olds (by not screening or treating theselesions) is at most 1.7 cases of cervical cancerper 100,000/year (the incidence rate for 20- to24-year-olds). It is not known, however, howmany cervical cancer cases among the 20- to24-age group were found in women who wereimmunosuppressed and/or HIV positive, how

many of the women werescreened, or at what stage thecancers were diagnosed.

The goal of cervicalscreening in the United Statesis to identify and removesignificant precancerous lesionsin addition to preventingmortality from invasive cancer.HSIL is considered a significantprecancerous lesion, whereaslow-grade SIL (LSIL) isconsidered a much morebenign lesion since most ofthese lesions regress. Data onregression and the naturalhistory of LSIL and HPVinfection in young womenaged 13 to 22 have shown thatmost HPV infections aretransient, with 70% regressionrates of high-risk HPV typeswithin three years and over90% regression for low-risktypes.7 HPV regression rates ofover 90 percent within twoyears have been reported incollege populations.8 Re-gression of LSIL is morecommon in adolescents than inadults. Fifty to eighty percentof LSIL in adult women willregress,15,16,17 whereas 90percent of LSIL in adolescentsand young women (aged 13 to21) will regress.18 In one study,when HPV type was known,81 percent of LSIL in youngwomen with a high-risk HPVDNA type regressed, whereassix percent progressed.18 Thisdiscrepancy may reflect a morebenign natural history for HPVinfection in adolescents orreflect an increased accuracy ofcytological diagnosis at entry,since LSIL greatly outnumbersHSIL in adolescents, reducing



Research UK, London, United Kingdom; Diane Fink,MD, Chief Cancer Control Officer, American CancerSociety, California Division, Inc., Oakland, CA; DianeM. Harper, MD, MPH, Associate Professor,Departments of Community and Family Medicine andObstetrics and Gynecology, Dartmouth MedicalSchool, Lebanon, NH; Ira Horowitz, MD, the WillafordRansom Leach Professor and Vice Chairman,Department of Gynecology and Obstetrics, andDirector, Division of Gynecologic Oncology, EmoryUniversity, Atlanta, GA; Herschel W. Lawson, MD,Medical Advisor, Program Services Branch, DCPC,NCCDPHP, Centers for Disease Control andPrevention, Atlanta, GA; Martin C. Mahoney, MD,PhD, Director, Cancer Prevention and DetectionCenter, Division of Cancer Prevention and PopulationSciences, Roswell Park Cancer Institute, Buffalo, NY;Jeanne Mandelblatt, MD, MPH, Director, CancerControl Program, Lombardi Cancer Center,Georgetown University Medical Center, Washington,DC; Kenneth L. Noller, MD, Obstetrician-in-Chief,Department of Obstetrics and Gynecology, Tufts-NewEngland Medical Center, Boston, MA; George F.Sawaya, MD, Assistant Professor, Departments ofObstetrics, Gynecology, and Reproductive Sciences,Epidemiology and Statistics, University of California,San Francisco, CA; Debbie Saslow, PhD, Director,Breast and Gynecologic Cancer, American CancerSociety, Atlanta, GA; Robert A. Smith, PhD, Director ofCancer Screening, American Cancer Society, Atlanta,GA.Work Group on Technologies:Diane Solomon, MD (Chair), Project Officer,ASCUS/LSIL Triage Study, National Cancer Institute,Rockville, MD; Diane D. Davey, MD, Professor,Department of Pathology and Laboratory Medicine,and Director, Cytopathology, University of KentuckyMedical Center, Lexington, KY; Eduardo L. Franco,MPH, DrPH, Professor of Epidemiology and Oncology,and Director, Division of Cancer Epidemiology, McGillUniversity, Montreal, Canada; Katherine Hartmann,MD, PhD, Director, North Carolina Program forWomen’s Health Research, and Assistant Professorof Obstetrics and Gynecology and Epidemiology,University of North Carolina, Chapel Hill, NC; IraHorowitz, MD, the Willaford Ransom Leach Professorand Vice Chairman, Department of Gynecology andObstetrics, and Director, Division of GynecologicOncology, Emory University, Atlanta, GA; Joan G.Jones, MD, Professor of Pathology and ClinicalObstetrics, Gynecology, and Women's Health, AlbertEinstein College of Medicine, and Director of SurgicalPathology, Weiler-Einstein Division, MontefioreMedical Center, Bronx, NY; Walter Kinney, MD,Gynecologic Oncologist, Division of GynecologicOncology, The Permanente Medical Group, Inc.,Sacramento, CA, and Associate Clinical Professor ofObstetrics and Gynecology, University of California,Davis, CA; Evan Myers, MD, MPH, AssociateProfessor, Department of Obstetrics and Gynecology,Duke University Medical Center, Durham, NC; MarkSchiffman, MD, MPH, Chief, Interdisciplinary StudiesSection, Division of Cancer Epidemiology andGenetics, National Cancer Institute, Department ofHealth and Human Services, Rockville, MD; EllenSheets, MD, Associate Professor, Department ofObstetrics, Gynecology, and Reproductive Biology,Harvard Medical School, and Vice President, Cytyc,Corp., Boston, MA; Edward J. Wilkinson, MD,Professor, Department of Pathology and LaboratoryMedicine, University of Florida College of Medicine,Gainesville, FL; Thomas C. Wright, Jr., MD, AssociateProfessor of Pathology, College of Physicians andSurgeons, Columbia University, New York, NY; DebbieSaslow, PhD, Director, Breast and GynecologicCancer, American Cancer Society, Atlanta, GA; RobertA. Smith, PhD, Director of Cancer Screening,American Cancer Society, Atlanta, GA.


merican C



Dow

nloaded from


the misclassification rate compared to adultwomen. The high regression rate of LSILparallels the high regression rates reported forHPV DNA detection, supporting the benignnature of most LSIL in adolescents and youngwomen.

There are limited data on the progressionrate of LSIL in adolescents. Moscicki, et al.18

found that only three percent of adolescentswith LSIL progressed to HSIL within threeyears. On the other hand, five percent ofadolescents who developed an HPV infectionin an earlier study developed a HSIL, half ofwhom did not have a preceding detectableLSIL.7 Based on the prevalence of HSIL oncytology smears and the number of patientswith HSIL (CIN2/3) found on histology whohad only LSIL on cytology, approximately 1.7percent of adolescents are estimated to harboran HSIL. However, the risk of progression ofthese high-grade lesions among adolescentsremains unknown. Nasiell, et al.19 observed thatthe time of progression for women with CIN2to a carcinoma in situ or cancer over age 51was 70 to 80 months, 41 to 42 months forwomen aged 26 to 50 years, and 54 to 60months for women under 25 years of age.Therelevance of these findings to young womentoday is unclear given that sexual behavior,contraceptive use, and smoking habits have allchanged from the time when this and othercohort studies were performed in the 1960s to1980s. The 54- to 60-month estimatedprogression for women under 25 years of age isthe only available data.The difference from 41months (for women aged 26 to 50) could bethe result of study design rather than biology.Therefore the more conservative period of 41months was used as the basis forrecommending that screening beginapproximately three years from potentialexposure to HPV.

In support of this recommendation,mathematical modeling20 suggests that the mostcost-effective strategy (excluding HPV DNAtesting) is to start screening three years after age

of onset of vaginal intercourse, with an age capat 25 (this modeling used the assumption thatcytology would be performed every three yearsfollowing three annual, consecutive, technicallysatisfactory normal/negative cytology results).This strategy was modified to reflect theopinion of an expert panel of clinicians, whichheld that an age cap of 25 might result in aproportion of women not being screened untiltheir late 20s in the absence of initiation basedon an accurate sexual history. Consequentlysome high-grade lesions could progress.Furthermore, the SEER data indicate thatinvasive cervical cancer does occur in womenunder 25 years of age. The revised policyanalysis showed that a sexual activity-basedstrategy with an age cap of 21 was also cost-effective, and particularly more cost-effectivethan universal age-based screening at age 18.The expert panel suggested that an age cap of21 would represent a more practical andrealistic age than 25 for compliance and accessto patients, and would be more acceptable thanpostponing screening until age 25 in theUnited States. Of note, a sexual activity-basedscreening initiation criterion rather that anage-based criterion was superior even whenscreening was conducted every two years orevery year.

When to Discontinue Screening

Recommendation

Women who are age 70 and older with anintact cervix and who have had three or moredocumented, consecutive, technically satis-factory normal/negative cervical cytologytests, and no abnormal/positive cytology testswithin the 10-year period prior to age 70 mayelect to cease cervical cancer screening.Screening is recommended for women whohave not been previously screened, women forwhom information about previous screeningis unavailable, and for whom past screening isunlikely. Women who have a history of




merican C



Dow

nloaded from


cervical cancer, in utero exposure todiethylstilbestrol (DES), and/or who areimmunocompromised (including HIV+)should continue cervical cancer screening foras long as they are in reasonably good healthand do not have a life-limiting chroniccondition. Until more data are available,women aged 70 and older who have testedpositive for HPV DNA should continuescreening at the discretion of their health careprovider. Women over the age of 70 shoulddiscuss their need for cervical cancer screeningwith their health care provider based on theirindividual circumstances (including thepotential benefits, harms, and limitations ofscreening) and make informed decisions aboutwhether to continue screening. Women withsevere comorbid or life-threatening illnessesmay forego cervical cancer screening.

Rationale

There is general consensus that theincidence of cervical cancer in older women isalmost entirely confined to the unscreened andunderscreened. Screening in the unscreenedpopulation can reduce morbidity and mortalityfrom cervical cancer.

Evidence suggests there is very low risk ofcervical cancer for women aged 50 and olderin countries with organized screeningprograms. Data do not exist to supportscreening over age 65 to 70 in women whohave been regularly screened. Since screeningthe unscreened has been shown to affectmortality rates, there is rationale to screen theelderly unscreened population.

In the United States, cervical cancer is rareamong older screened women. Furthermore, itmay be difficult to get satisfactory samples fromolder women due to conditions such asatrophy and cervical stenosis.There is evidencethat screening is associated with potentialharms, including anxiety and discomfortduring cytology sampling of some olderwomen, and the invasive procedures, anxiety,

and higher health care costs due to false-positive cytology results.

The choice of exact age at which to ceasescreening is arbitrary. The choice of age 70 isbased on the opinions of the expert panelmembers in an effort to balance the benefitsand harms of screening older women. Whilesome organizations have chosen the age of 65(which is also arbitrary) in their guidelines, theACS set the age at 70 based on mathematicalmodeling21 (Jeanne Mandelblatt, MD, MPH,unpublished data.) and demographic trendsthat may increase the likelihood of olderwomen having new sexual partners and thusnew exposures to HPV. Older women whochoose to discontinue screening shouldcontinue to obtain appropriate preventivehealth care.

Evidence

Several studies have shown a low efficiencyof cytological screening in women over the ageof 50; the vast majority of cervical cancers inolder women occurred in those who were notpreviously screened or who did not have threeconsecutive normal cytology results.22,23,24,25,26,27,28

Few studies provide data on women aged 65and older. Sigurdsson24 reported decreasingrates of high-grade intraepithelial lesions inwomen aged 60 to 69 with increasing numberof prior normal cytology tests. Sawaya, et al.28

observed low rates of LSIL and HSIL inwomen aged 65 and over with at least oneprevious normal cytology result within the lastthree years.

Screening After Hysterectomy

Recommendation

Screening with vaginal cytology testsfollowing total hysterectomy (with removal ofthe cervix) for benign gynecologic disease isnot indicated. Efforts should be made toconfirm and/or document via physical exam




merican C



Dow

nloaded from


and review of the pathology report (whenavailable) that the hysterectomy was performedfor benign reasons (the presence of CIN2/3 isnot considered benign) and that the cervix wascompletely removed.Women who have had asubtotal hysterectomy should continue cervicalcancer screening as per current guidelines.Women with a history of CIN2/3 or forwhom it is not possible to document theabsence of CIN2/3 prior to/or as theindication for the hysterectomy should bescreened until three documented, consecutive,technically satisfactory normal/negativecervical cytology tests and no abnor-mal/positive cytology tests within a 10-yearperiod are achieved.Women with a history ofin utero DES exposure and/or with a historyof cervical carcinoma should continuescreening after hysterectomy for as long as theyare in reasonably good health and do not havea life-limiting chronic condition.

Rationale

Use of cytology tests in women who havehad their cervix removed for benign reasonsscreens the vaginal cuff. Vaginal cancer is anuncommon gynecologic malignancy with anincidence rate of 1 to 2/100,000/year.1

Abnormal vaginal cytologic smears areuncommon and rarely of clinical importance.However, women with a history of in uteroDES exposure are at increased risk for vaginal aswell as cervical cancer.Women who have had ahysterectomy for cervical intraepithelial lesionsmay be at increased risk of vaginal cancer, butthe data are limited. Women who have had asubtotal hysterectomy retain their cervix andshould continue cervical screening. Womenwho discontinue screening should continue toobtain appropriate preventive health care.

For women in whom CIN2/3 was theindication for the hysterectomy, follow-upcytology is recommended every four to sixmonths. Three documented, consecutive,technically satisfactory normal/negativevaginal cytology tests and no abnor-

mal/positive cytology tests should be achievedwithin an 18- to 24-month period followinghysterectomy before discontinuing cytologyscreening. Women with a history of CIN2/3prior to/but not as the indication forhysterectomy should be screened until three documented, consecutive, technicallysatisfactory normal/negative vaginal cytologytests and no abnormal/positive cytology tests(within a 10-year period) are achieved.

Evidence

A retrospective cohort study of vaginal cuffcytologic smears (VCS) in 5,862 women whohad undergone a hysterectomy for benigndisease found abnormal VCS among 79women (1.1 percent of all smears). The meanlength of time from hysterectomy to abnormalcytology result was 19 years. The positivepredictive value for detection of vaginal cancerwas 0 (95 percent CI 0 to 33 percent).29 A 10-year retrospective study among 697 womenafter hysterectomy for benign disease foundthat 663 VCS were needed to detect one caseof vaginal dysplasia.30 A retrospective study of220 women selected at random from 2,066women who had a previous hysterectomy forbenign conditions and followed for an averageof 89 months identified seven patients (threepercent) who had intraepithelial cytologicabnormalities, but no vaginal cancers. Four ofthese patients underwent successful excision orlaser treatment of the lesions, and dysplasticlesions in the remaining three patientsregressed without any treatment. No benefit inpatient outcomes was observed.31 A cross-sectional study of 5,330 screening cytologytests in women who had had a hysterectomyfound one case of dysplasia and no cancers.32

Videlefsky, et al.31 reported two cases ofsevere dysplasia among 44 women (4.5percent) after hysterectomies for HSIL. Thedifference in risk compared with womenwithout a history of abnormal cytology priorto hysterectomy (3.6 percent) was notstatistically significant. In a study of 193




merican C



Dow

nloaded from


women with CIN at hysterectomy, theincidence of abnormal vaginal cuff cytologicsmears at least two years after hysterectomy was0.7 per 1,000, and at 20 years 96.5 percent ofthe women continued to have normal smears.33

Screening Interval

Recommendation

After initiation of screening, cervicalscreening should be performed annually withconventional cervical cytology smears ORevery two years using liquid-based cytology; ator after age 30, women who have had threeconsecutive, technically satisfactory normal/negative cytology results may be screenedevery two to three years (unless they have ahistory of in utero DES exposure, are HIV+,or are immunocompromised by organtransplantation, chemotherapy, or chroniccorticosteriod treatment).

Rationale (for Conventional Cervical Cytology)

The difference in absolute risk of animportant lesion progressing to invasive diseaseis small when comparing one-, two-, andthree-year screening intervals withconventional cervical cytology. While mostdata in this area come from countries withorganized screening programs, data from theUnited States are similar. The difference inrelative risk of an important lesion progressingto invasive disease between two- or three-yearscreening intervals compared with a one-yearinterval is significant; however, it is importantto note that the probability of disease is quitesmall even among women screened every threeyears. The number of high-grade lesions thatmight progress during a screening intervallonger than three years is considered to beunacceptably high in the United States.Whilemore frequent screening increases sensitivity,it also increases patient harm and costs.Sensitivity, patient harm, and cost were allfactors in determining the ACS guideline.

Screening interval recommendations applyto women who have been screened previouslyand received a technically satisfactorynormal/negative result. Specimen adequacyand quality indicators should be consideredwhen determining the timing of repeatscreening for both conventional Pap smearsand liquid Pap tests. If endocervicalcells/transformation zone elements are absentor if there are partially obscuring factors, anannual repeat may be considered, and selectedwomen may benefit from an earlier repeattest.34,35 Women who have had a recentabnormal cytology result would be classified asunder surveillance rather than undergoingscreening. A history of consecutivenormal/negative cytology results has beenassociated with decreased risk of HSIL andcervical carcinoma.

Preliminary data suggest the mostappropriate screening interval is age-dependent and younger women may benefitmore from a shorter screening interval (JackCuzick, PhD, unpublished data.) e.g., one-yearintervals rather than two- to three-yearintervals for women under the age of 30.Women who are immunocompromised shouldfollow the CDC guideline on screening HIV-infected women.12 Women who have beenexposed to DES in utero have an increased riskof cervical and vaginal cancers and should bescreened annually. Other risk factors, such asearly age of onset of sexual activity or multiplesexual partners, should not be used as arationale for more frequent screening.36 Thereare insufficient data to support recom-mendations for or against a more frequentscreening interval (i.e., annual screening evenafter age 30) for women who smoke cigarettes,although one small study suggests that therewould not be a benefit.37

There is some concern that screening lessthan annually with conventional cytology willmiss a proportion of endocervical adeno-carcinomas.There is little data on the efficacyof cervical cytology as a screening anddetection tool for nonsquamous cell




merican C



Dow

nloaded from


carcinomas. Use of the endocervical brush aswell as new technologies including liquid-based cytology tests may increase the sensitivityto detect this type of cervical cancer.

Evidence for Screening With ConventionalCytology

The rigorous criteria established for theACS guideline review process (Appendix A)resulted in exclusion of many studies. In mostcases, excluded studies did not address whetherthe subjects had any prior history of cytologytests. Most of the case-series studies did notinclude information about prior screening nordid they have controls. Some measuredincident cancer cases three to six years after thelast screening test; however, the guidelinereview panel agreed that the risk of screeningless frequently than three-year intervals wasunacceptable, and therefore the panel wasinterested only in cases identified within threeyears. The number of such cases is very smalleven over a long period of time. Negative ornormal results as presented across the studieswere not consistent, standardized, or defined.

While the format and definitions employedin measuring relative risk of invasive cervicalcancer for one-, two-, and three-year screeningintervals were often inconsistent, most studiessuggest that compared to annual screening, therelative risks with a two-year or a three-yearscreening interval are in the range of one totwo and two to three, respectively, above annualscreening.38,39,40,41,42,43 Risk increases with longerscreening intervals of 4 to 10 years.38,39,40,41,42,44

Because most studies do not provide data onthe number of women screened, measurementsof absolute risk are even more limited. Sawaya,et al.28 provided data from a prospective cohortstudy of 128,805 women in the United Statesand determined the age-adjusted incidencerate of HSIL, carcinoma-in situ, or invasivecarcinoma within three years of normalcytology results to be 25 per 10,000 for

women screened at 9 to 12 months, 29 per10,000 for women screened at 13 to 24months, and 33 per 10,000 for womenscreened at 25 to 36 months following anormal cytology result. The differences inincidence were not statistically significant.28 Acase control study reported that a two-year orthree-year interval led to doubling the relativerisk of being diagnosed with invasive cancer,but only a slight increase in absolute risk.43 Thisfinding was based on 482 cases observed over13 years in a health maintenance organizationwith an enrollment of about one millionwomen each year. Fifty-three of the casesoccurred within three years of three or morenegative/normal Pap smears. In thispopulation, the absolute risk of invasivesquamous cell cancer within the three yearsfollowing three or more normal Pap smearswas estimated at less than 5 per 100,000women per year.

Several studies have shown that a history ofnormal/negative results has a protective effecton cervical cancer incidence.39,40,45,46 One casecontrol study in Italy found a decrease in therisk of invasive cervical cancer of 90 percentamong women with three or morenormal/negative cytology smears compared towomen with no previous cytology test.45 Acohort study in Denmark found that womenwith two- to four-previous normal/negativecytology results had a negligible risk ofdeveloping cancer within two years, and aslower increase in risk compared to womenwith only one previous negative result.40

Sawaya, et al.46 estimated the absolute risks ofcervical cancer (within 18 months of the lastnegative smear) following one, two, and threeor more consecutive negative cytology smearsas 3.09, 2.56, and 1.43 per 100,000 women,respectively, based on 2.4 million long-termmembers of a prepaid health plan. Further, it isimportant to note that there is an irreduciblerisk of cervical cancer in the United States. Ifall women in the United States were screened




merican C



Dow

nloaded from


annually, it is estimated that there would be 1.5cases of cervical cancer per 100,000 womenhaving a negative cytology result within 0 to18 months and at least three prior consecutivenormal/negative results.47

Patient harm is very difficult to measure, andstudying patient harm resulting from screeningfor a rare outcome such as cervical cancerposes particular challenges. Sawaya, et al.27

reported that in approximately 2,561 post-menopausal screened women, 110 requireddiagnostic work ups for incident abnormalities(one or two years after a negative smear)involving 231 total interventions, and onewoman was found to have a mild to moderateintraepithelial lesion.

The recommended screening interval forolder women is rarely addressed in the availablepublished literature. One exception is theHeart and Estrogen/Progestin ReplacementStudy (HERS) study,27 which included womenup to age 80. This study concluded thatcervical cytology screening of previously-screened postmenopausal women at a ratemore often than every two years had a positivepredictive value of 0 to 2.7 percent (PPVvaried depending on whether women withoutavailable biopsy results were included orexcluded) and therefore supports not screeningpostmenopausal women within two years ofnormal cytologic results.

NEW TECHNOLOGIES

Recently, interest has been focused on newtechnologies that enhance the accuracy ofcervical cancer screening. The expert reviewpanel considered several cervical screeningtechnologies with sufficient published clinicaldata and with potential application within theUnited States.The panel excluded technologiesthat are currently under development or thatmay have utility only in low-cost screeningsettings, i.e., aided visualization, cervicography,

computer-assisted screening devices, opticalprobe devices, self-collected vaginal samples forHPV DNA testing, and spectroscopy/electronicdetection devices.

When evaluating and comparing the utilityof the technologies, several points must beemphasized. Data on sensitivity are generallyderived from research settings with optimizedtesting conditions; the same results may not beachievable in actual practice. Certain researchdesigns by their nature may also over- orunderestimate test sensitivity.

Test sensitivity must be distinguished fromprogram sensitivity. Test sensitivity is theprobability that a single test performed at aspecified point in time will detect the presenceof underlying disease. Program sensitivity is theprobability that tests repeated at specifiedintervals will detect underlying disease over aperiod of time. Repeat screening at regularintervals therefore compensates somewhat forthe limitations of the sensitivity of the technique.

For any test, sensitivity for detection ofdisease can be improved by lowering the testthreshold for a “positive” result but only withconcomitant loss in specificity, resulting inmore false-positive results. When screening apopulation for a very low-prevalence disease,even a small percentage change in specificityaffects a large number of women because thevast majority of women screened do not havedisease. For example, if disease prevalence is 10percent and test specificity is 95 percent, then afive percent decrement of specificity for 50million screening tests represents an increase of2.25 million women (from 2.25 million to 4.5million) who will receive a “false-positive”result.

Prevailing management paradigms, medico-legal issues, economic factors, and societalexpectations are all factors in determining thebalance between sensitivity and specificity for a screening program. Risk perception,understanding, and acceptability all varyamong individual patients, care providers, and




merican C



Dow

nloaded from


policy makers.48,49,50,51 Criteria for the ACSreview and evaluation of new technologyarticles are summarized in Appendix B.

Liquid-based Pap Technology

Recommendation

As an alternative to conventional cervicalcytology smears, cervical screening may beperformed every two years using liquid-basedcytology; at or after age 30, women who havehad three consecutive, technically satisfactorynormal/negative cytology results may bescreened every two to three years (unless theyhave a history of in utero DES exposure, areHIV+, or are immunocompromised).

Rationale

Currently, two liquid-based Pap (LBP)technologies have been approved by the FDAas being at least equivalent to the conventionalPap smear in their ability to detect cervicalprecancerous lesions and cancers. Most studieshave shown improved sensitivity for LBPcompared to conventional smears. LBPspecificity has been difficult to calculate butgenerally is thought to be comparable ormoderately decreased compared toconventional cytology when performed lessfrequently. Although the ACS and others haverecommended since before 1980 thatconventional cytology can be safely performedup to every three years for most women, in theUnited States, annual screening is expected bymost women and health care providers. Alonger interval for LBP, i.e., every two to threeyears, compared to annual conventional Paptesting compensates for the decrease inspecificity. Conversely, screening with LBP atthe same interval as conventional cytology willlikely lead to significant increases in thedetection of atypical squamous cells -uncertain significance (ASC-US) and low-grade abnormalities, with subsequent increasesin referral of women to colposcopy

unnecessarily, risking the potential forovertreatment and increased health care costs.

Several problems with conventionalcytology smears are addressed by liquid-basedmethods. With LBP, the sampling device isdirectly placed into a liquid fixative instead ofbeing spread onto a glass slide providingimmediate fixation, thereby decreasing air-drying artifact and thus improving specimenadequacy.Additionally, LBP are felt to improvecellular sampling, distribute the cells moreevenly over the slide and reduce celloverlapping, and decrease obscuringbackground factors (such as blood andinflammatory cells) often seen in conventionalcytology (factors that may also affect testaccuracy). Another advantage of LBP is theavailability of residual material, whichpotentially may be used for ancillary testing(e.g., for oncogenic HPV DNA).

There are currently no data to support aparticular screening interval for LBP. Thisrecommendation was based on modeling bytwo independent researchers52,53 (Evan Myers,MD, MPH, unpublished data.) as well as expertopinion based on the existing data describedbelow.

Evidence

Compared with conventional cytology, LBPprovides at least equivalent percentages ofsatisfactory specimens.54,55,56 The key criteria forevaluating the available published literaturewere the sample size of the study, the type of population utilized (with a screeningpopulation similar in risk to the US populationpreferred), comparable medical practices tothose of the United States, comparable patientdemographics for historical cohort studies, anacceptable method of addressing inter/intra-observer variability, a way to address false-negative outcomes, and histology outcomes ina significant portion of the patients screened(Appendix B).

All of the available studies have importantflaws, and the results are all presented in




merican C



Dow

nloaded from


different ways. Taken as a whole, the availableevidence supports the conclusion that LBP isan acceptable option for cervical screening,that LBP is somewhat more sensitive but lessspecific for high-grade lesions, and that it mayincrease the number of ASC-US referrals andpossibly the proportion of samples lacking anendocervical component. Of the studiesreviewed54,55,56,57,58,59,60 all but one60 showedincreased sensitivity. Three studies showed asignificant increase in the ASC-US detectionrate,57,58,59 while others showed similar orslightly decreased rates.54,55,56,60 In the studiesthat provided data on specificity, the specificitywas significantly decreased in two,57,58 wassimilar in two,56,60 and was uncertain in one.59

One study found improved sample adequacy56

while three found an increased rate of sampleslacking an endocervical component.54,55,60

Importantly, in the one study that found nosignificant differences in sensitivity betweenLBP and conventional cytology, sampling forconventional testing was greatly enhanced byuse of a specialized collection device, removalof mucus and cellular debris from the cervicalsurface prior to sampling, and colposcopically-guided sampling to verify harvesting of cellsfrom the endocervical canal and full cir-cumference of the cervical surface.60

Although there have been questionsregarding the detection of glandular lesions byLBP, recent studies62,63,64 have shown animprovement in sensitivity and specificity forbiopsy-proven adenocarcinoma. Ashfaq, et al.62

demonstrated similar rates of glandularabnormalities on LBP as conventionalcytology; however, the positive predictive valuefor LBP was higher and there were fewer false-negative LPB reports preceding biopsy-confirmed adenocarcinomas and adeno-carcinoma in situ (AIS). Bai, et al.63 reported a50 percent lower rate for glandular atypia onLBP, but the positive predictive value for AISwas five-fold higher.Thus it is felt that LBP arecapable of more precisely predictingAIS/adenocarcinoma. This is particularlyimportant given the recent literature indicating

an overall rise in the rate of invasive cervicaladenocarcinoma.

The preponderance of the data consideredin this review of LBP utilized the first of thetwo FDA-approved techniques (ThinPrep).The studies and abstracts of the other LBPmethod (SurePath, previously Autocyte)suggest equivalent performance, but the fewstudies limit opportunities for comparison.

HPV DNA Testing With Cytology for the Screening

of Cervical Cancer and Its Precursor Lesions

Preliminary Recommendation

HPV DNA testing with cytology forprimary cervical cancer screening has not beenapproved by the FDA. Based on the availabledata, both published and unpublished, the ACSguideline review panel found this technologyto be promising. Should the FDA approveHPV DNA testing for this purpose, it wouldbe reasonable to consider that for women aged30 and over, as an alternative to cervicalcytology testing alone, cervical screening maybe performed every three years usingconventional or liquid-based cytologycombined with a test for DNA from high-riskHPV types. Frequency of combined cytologyand HPV DNA testing should NOT be moreoften than every three years. Counseling andeducation related to HPV infection is a criticalneed. Consensus guidelines for themanagement of women with a technicallysatisfactory normal/negative cytology resultand a HPV DNA test result that is positive forhigh-risk HPV types would need to bedeveloped.

Rationale

Since the mid-1990s there has beensubstantial interest in the use of HPV DNAtesting as a cervical cancer screening tool basedon the premise that standardized moleculartesting of exfoliated cervical cells for thecausative agent of cervical cancer could have




merican C



Dow

nloaded from


acceptable diagnostic performance while beingmore reproducible and more easily adapted forclinical practice than conventional cytology.

There have been several studies assessing therelative utility of both a cytology test and HPVDNA testing compared to the cytology testalone as a primary cervical cancer screeningtool. Most studies have used the HybridCapture® (HC) system, the only HPV DNAtest currently approved by the FDA. Thecurrently marketed system is Hybrid CaptureII® (HC2). It detects the presence of 13 typesof HPV that have been associated with cervicalcancer. Most of the studies reviewed by ourpanel indicate that women with a concurrentnormal cytology test and a negative HC2 resulthave substantially decreased risk of high-gradelesions on colposcopy relative to those forwhom the only screening information is anormal conventional cytology result. On theother hand, a positive HC2 result is not anabsolute indicator that high-grade lesions existor will develop; the prognostic value of apositive test result, especially in the absence ofa cytologic abnormality, has not been fullyvalidated in prospective studies.

HPV DNA testing has greater sensitivitythan cytology for detecting clinically relevantlesions. Restricting screening to women aged30 and older reduces the number of women tobe referred to colposcopy due to transientHPV infection. The high negative predictivevalue resulting from concomitant screeningwith cytology and HPV DNA testing couldsafely permit increasing screening intervals,thus lowering costs. While definitive evidenceof efficacy is still needed from long-termfollow-up studies with CIN2/3 as an outcomeand from randomized controlled trials, theopinion of the majority of expert panelmembers was that the available evidencesupports consideration of the use of HPVDNA testing as an adjunct to cervical cytologywith a screening interval no more frequentlythan every three years. More frequentscreening would not significantly improvesensitivity, but would likely result in over-

evaluation and potential overtreatment ofmany women for transient HPV infections.53

Only testing for high-risk HPV types wouldbe of value. Testing for low-risk HPV types is not useful, and may have a negativepsychological impact on the patient.

HPV DNA testing for the triage of patientswith a cytology result of ASC-US wasconsidered outside the scope of this screeningguideline review. Consensus recommendationsfor the management of women with abnormalcytology tests were developed through aprocess sponsored by the American Society forColposcopy and Cervical Pathology, andpublished in April 2002.65

Evidence

Only studies using the HC2 test were takeninto account for the purpose of this guidelinereview.59,66,67,68,69,70,71,72,73 All of them assessed thediagnostic performance for existing lesionsusing cross-sectional or short-term follow-upinvestigations of European,59,66,67 African,69,72,73

Asian,68 Latin American,71 and North American(Canada)70 populations. Lesion definitionvaried across studies and included either CINof all grades or CIN2/3 or worse lesions,diagnosed by histology on specimens obtainedby colposcopy-guided biopsy. In some studies,the colposcopic result was used if no biopsywas taken. None of these studies were based onlong-term follow-up for more relevantendpoints, such as incidence of/or mortalityfrom invasive cervical cancer. No results fromrandomized controlled trials have yet beenpublished; all studies were based onconcomitant testing. The review panel tookinto account the statistical issues that make itnecessary to use different criteria to judgesensitivity and specificity when two tests areused together.

HPV DNA testing at the manufacturer’srecommended threshold of positivity of onepg/ml (equivalent to 5,000 viral copies) hasbeen shown to have greater sensitivity thancytology (on average, 25 percent higher in




merican C



Dow

nloaded from


absolute terms) but lower specificity (onaverage, 10 percent lower in absolute terms) fordetecting high-grade lesions.59,66,67,68,69,70,71,72,73

Screening of women aged 30 or older or 35 orolder tended to improve the specificity of HPVDNA testing considerably because viralinfections in this age group are less commonand are less likely to be of a transient naturethan those in younger women. An importantfinding of most studies was the realization thatthe combination of cytology and HPV DNAtesting attained very high negative predictivevalues (approaching 100 percent). Althoughnone of these studies has been on an Americanpopulation, it is reassuring that ongoinganalyses from the NIH-funded Portland cohortcorroborate the enhanced value of HPV DNAtesting in screening for high-grade cervicallesions as compared with cytology.74

ADDITIONAL RECOMMENDATIONS

The expert panel made several additionalrecommendations: (1) The ACS and othersshould educate women, particularly teens andyoung women, that a pelvic exam does notequate with a cytology (Pap) test, and thatwomen who may not need a cytology test stillneed regular health care visits, includinggynecologic care and STD screening andprevention. (2) The current guideline reviewdid not address the potential usefulness ofpelvic and/or rectal examinations. Pelvicexams are not effective in detecting cervicalcancer, however both pelvic and rectal examsmay facilitate identification of other types ofcancer and of other gynecologic conditions.Women should discuss the need for theseexams with their provider. (3) Referrals ofwomen with low-grade lesions for colposcopymay be less necessary for adolescents given theself-limited nature of many LSILs in this agegroup. Detection and treatment of HSILshould be the goal of adolescent screening andreferral. (4) Health insurance payers should notexclude adolescents or women of any age from

coverage for cervical health on the basis offalse-positive cytology results and/or mildabnormalities on cervical cytology. (5) Healthinsurance coverage for new cervical screeningtechnologies is not uniform. Providers shouldconfirm coverage before ordering tests such asLBP and HPV DNA testing, including use fortriage of patients with ASC-US.

CONCLUSION

Compared to the previous ACS guidelinepublished in 1988,11 this guideline representsmore comprehensive recommendations basedon the body of accumulated evidence. Changeswere made concerning what age to startscreening and, to a lesser extent, the screeningintervals. New recommendations weredeveloped to address when screening may bediscontinued, screening of women who havehad a hysterectomy, and the use of newscreening technologies. Changes to thescreening recommendations are unlikely tohave a significant impact on the relatively lowincidence and mortality associated withcervical cancer in the United States. Yet thelarge number of women currently beingscreened each year (50 million) who receive anabnormal result (two million ASC-US andASC-H, 1.25 million LSIL, 300,000 HSIL, inaddition to 13,000 cancers) and undergoadditional procedures represents a potential fora large impact in reducing patient discomfort,anxiety, and inconvenience as well as healthcare costs.

The guideline continues to emphasize theimportance of flexibility for women and theirproviders in the context of informed decision-making. Individual patients will have differentperceptions of risk and risk tolerance that mayaffect their choice of screening interval,screening test, and whether to discontinuescreening after a certain age. Ideally thesedecisions should be based on discussions of thebenefits, risks, and limitations of cervical cancerscreening between women and their providers.




merican C



Dow

nloaded from


Screening interval remains a controversialissue in the United States.While the evidencesupports the conclusion that conventionalcytology can be safely performed at two- tothree-year intervals, many women and

providers in the United States may be morecomfortable with annual screening. A keyfactor is the limited sensitivity of theconventional Pap test. A significant proportionof false-negative conventional cytology results



Evidence Grading 1 = Strong evidence. 2 = Limited evidence. 3 = No evidence/exclude.

Study Type Code (from US Preventive Services Task Force)

Criteria for Evidence Grading

(1) Strong Evidence (2) Limited Evidence (3) No Evidence/Exclude

Key Data to Abstract for Each Article that Meets Inclusion Criteria

CountrySample sizeSample description (age, risk,ethnicity, screening history)Time period

Biases (selection, verification,observer)Issue addressedEndpointsLength of follow up (e.g., average,individual, person-years)

Major flawsMajor strengthsAuthors’ conclusionsReviewer’s conclusions

Primary Reports of New Data Collection

Class A: Randomized, controlled trial.Class B: Prospective cohort study; Case-control study

nested within a prospective cohort study.Class C: Non-randomized trial with concurrent or

historical controls; Case-control study (exceptfor the preceding); Retrospective cohortstudy; Study of sensitivity and specificity of a diagnostic test; Population-based descriptivestudy.

Class D: Cross-sectional study; Case series;Case reports.

Reports That Synthesize or Reflect Upon Collections of Primary Reports

Class M: Meta-analysis; Decision analysis; Cost-benefitanalysis; Cost-effectiveness study.

Class R: Review article; Consensus statement;Consensus report.

Class X: Medical opinion.

APPENDIX A

• Evidence is useful to our task(reviewer’s conclusion may bedifferent from authors’).

• Sample size is adequate to givestatistical power.

• Unbiased or biases addressed.• Endpoint defined as CIN2/3.

• Conclusions/assumptions are notsupported by data, but someuseful data is provided.

• Sample size insufficient to givestatistical power to observe a trueeffect.

• Flaws or biases that could negateconclusions.

• Study design weakens conclusions(reviewer should provideexplanation).

• Review article with a newperspective.

• No relevant data (e.g., reviewarticle).

• Symptomatic women.• Shortcomings negate conclusions.• Articles not in English.


merican C



Dow

nloaded from


are due to inadequate sampling; improvementsin the ability to obtain an adequate samplewould increase the sensitivity and effectivenessof conventional cytology. In addition, manyexperts believe that the use of newtechnologies such as liquid-based Pap tests andHPV DNA testing (in combination withcytology), when performed at a less frequentinterval, offers several advantages overconventional cytology smears alone. Theseadvantages may include increased sensitivity,lower long-term costs, and facilitation of thetriage of ASC-US results. If used at the sameinterval though (e.g., annual LBP compared toannual convention cytology), new technologieswould significantly increase the number ofwomen referred for colposcopy unnecessarily,greatly increasing health care costs and harmsto patients, with little or no benefit.

It is important to reiterate that the biggestgain in reducing cervical cancer incidence andmortality would be achieved by increasingscreening rates among women who have notbeen screened or who have not been screenedregularly. Missed opportunities for screeningabound, particularly among unscreened olderwomen, women of low income and/or loweducation, and women who are uninsured orunderinsured. Clinicians, hospitals, healthplans, and public health officials should seek toidentify and screen these women, and to ensurecontinued screening at regular intervals.

APPENDIX B

Criteria for Evaluating Studies of Cervical

Screening Technologies

The purpose of these suggested criteria isnot to exclude any study because it is missingany of the following criteria, nor to grade aparticular study based on the criteria, but togive a sense of weight of the evidence for agiven technology. At least one study shouldprovide evidence that supports use of thetechnology, and meet the individual evidence

criteria. For example, if every study ofTechnology A suggests it is more sensitive thanTechnology B, and at least one study addressesat least one of these methodological issues,then Technology A would be considered moresensitive. Issues of study design come into playin trying to estimate how much more sensitive,but trying to come up with quantitativeestimates was beyond the scope of thisguideline review.

Reference Standard Specified

The minimum criteria for inclusion is thatthe reference standard is explicitly stated.Timing is an important issue to be considered(e.g., how soon after the cytology was thecolposcopy done). Suggested acceptablereference standards included conebiopsy/hysterectomy specimen, colposcopy/biopsy, and (under defined circumstances)comparison of cytological results.

Test Results Directly Comparable: Should BeApplied to the Same Patient/Specimen or BeRandomly Assigned

Studies that do not involve direct comparisonof the results of two or more tests in either thesame patients, or in patients drawn from thesame population and randomly assigned to aparticular test, are less preferable than studieswhere either the new technology was randomlyassigned or where it was applied to the samepopulation of patients and/or slides.

Tests Performed in the Context of Screening,Not Follow-up of Abnormal Results

When an imperfect reference standard(colposcopy/histology) is used, testcharacteristics are dependent on the prevalenceof the disease. Studies that report testperformance in a general screening populationare preferable. Any study should provideadequate characterization of the studypopulation.




merican C



Dow

nloaded from


Study Reports Results of Detection ofHSIL/CIN2/3

Because no study will be able todemonstrate reduction in the incidence ofcervical cancer, the next best alternative is theability to improve detection of high-gradeprecursors. In order to infer that a newtechnology will result in a decrease in cervicalcancer, evidence about its ability to detectCIN2/3 must be provided.

Verification Bias Addressed

If the reference standard is applied only topositive screening tests, then the sensitivity willalways be overestimated. Ideally, at least onestudy of a technology will have a design thatavoids verification bias.

Chance Gains in Sensitivity Have BeenConsidered

Adding an adjunct screening test in parallelto Pap cytology will always yield an increasedcombined sensitivity that may not be greaterthan that contributed by an unrelated adjuncttest in the same screening setting. Ideally,studies should consider sensitivity gains ofcombined testing only after taking intoaccount this chance increase in sensitivity.75

Observer Variability Addressed

Inter- and intraobserver variability may havesignificant impacts on the consistency of testsensitivity and specificity.All other things beingequal, a more consistent test is preferable.Ideally, at least one study of a technology willaddress inter- and intraobserver variability inthe interpretation of results.

Statistical/Sample Size Issues Addressed

Because most comparisons of cervicalscreening tests are based on categoricaloutcomes, statistical power is inherently

somewhat reduced compared to continuousoutcomes. Sample size can also affect studyinterpretation in other ways. In addition, use ofmultivariate analytic techniques invariablyresults in some loss of power. Powercalculations, preferably a priori, should bestated. Studies that present results that allowquantification of uncertainty (point estimatesof sensitivity and specificity with 95%confidence limits) are in general preferable tostudies that provide only the results ofsignificance testing (chi-square comparison oftwo different sensitivities).

APPENDIX C

Collection of Cervical Cytology Samples

It is estimated that at least one third or moreof false-negative cytology tests (negative resultswhen a woman has a high-grade cervicallesion) are related to sampling issues—specifically that abnormal cells are not presenton the cytologic slide examined in thelaboratory.76,77,78,79 Sampling problems can occurif abnormal cells are not collected onto thesampling instrument or if collected cells arenot transferred to the slide to be examined, butrather remain on the sampling instrument.Regardless of the collection device, only arelatively small percentage of cells aretransferred to the slide to be submitted ascompared to the total number of cells collectedon the device.80,76,81

Obtaining an Adequate Cervical CytologicSample

An adequate cervical cytologic specimeninvolves circumferential sampling of theectocervix adjacent to the transformationzone, the endocervix, and the cervicaltransformation zone (T-zone). A vaginalcytologic sample is not necessary for cervicalcytologic sampling, and is not recommendedfor cervical cytologic sampling.Vaginal samples




merican C



Dow

nloaded from


are recommended for the detection of vaginalcarcinoma in women exposed to DES in utero.

Collection Instruments

Using a combination of the extended tipspatula and the endocervical brush providessampling of the ectocervix, T-zone, andendocervix, and has the lowest false-negativerate when compared to less thoroughsampling.82,83,84 Both samples are usually placedon a single slide, rather than using two slides.85

The extended tip spatula is superior to theconventional Ayres spatula for sampling theectocervix and the T-zone.81 The spatula hasbeen traditionally made of wood; however,plastic spatulas perform equally well forconventional smears and perform better whenused in liquid-based systems because the cellscan be washed from the plastic more readilythan from wood spatulas.

To collect the endocervical cytologicsample, it is necessary to insert the collectiondevice within the endocervical canal andcollect cells by gently scraping or brushing theendocervical mucosa. When using theendocervical brush, the brush should beinserted until it has moved into the endocervixand until the bristles most proximal to thehandle are approximately even with theapparent external cervical os.The brush is thenrotated 180 degrees (one-half turn) in thecanal. Additional rotation does not improvesampling and may cause bleeding. In the non-pregnant patient, the use of an endocervicalbrush provides a cellular sample that isgenerally rich in endocervical and/or high T-zone cervical cells.This device is used after thespatula sampling of the ectocervix and T-zone.The endocervical brush can be used to prepareconventional cervical smears, as well as withliquid-based cervical cytology. The endo-cervical brush collects more diagnostic cellularmaterial than the swab, as studied in patientswho had undergone prior treatment to the T-zone, including cryotherapy, laser ablation, andconization.84,86,87 Although the endocervical

brush is not generally recommended by themanufacturer for use in pregnant womenbecause of the potential risk of inadvertentperforation of the amniotic sac, there isconsiderable clinical experience with the useof the brush in pregnant women with noapparent complications.88

Cervical broom instruments and othersingle sampling instruments are designed tosimultaneously sample the ectocervix, T-zone,and endocervical areas, and are fairlycomparable to the combination of extendedtip spatula and endocervical brush.The cervicalbroom may be used for conventional cytologysmear samples as well as for liquid-basedsystems, and may be used in pregnant women.

Use of a Small Swab for Endocervical Sampling

An endocervical swab is less sensitive thanthe endocervical brush and its use isdiscouraged for endocervical cytologicsampling.81 It may be considered for pregnantwomen when there is concern about usingother endocervical sampling devices. It shouldbe used in combination with an ectocervicaland T-zone sampling instrument such as theextended tip spatula. The swab should bemoistened with a small amount of saline priorto use in order to prevent cellular desiccationon contact with the swab, and to release cellsfrom the swab more easily when making asmear. A swab is not recommended withliquid-based systems. For liquid-based systems,the cervical broom is recommended in thepregnant patient, rather than the spatula andswab collection method.

Patient Advice Prior to a Cytology Test

The patient can be advised that there areseveral actions she can take to optimize thecytology test (based on American Society ofCytopathology 2001 guidelines available at:http://www.cytopathology.org/guidelines/cervical-cytologyiii.php).These include:

1.Try to schedule the appointment when it




merican C



Dow

nloaded from


is not during the menstrual period.2. Do not douche 48 hours prior to the

cytology test.3. Refrain from intercourse 48 hours prior

to the test.4. Do not use tampons, birth control foams,

jellies, or other vaginal creams or vaginalmedications for 48 hours prior to the test.

Schedule the test to avoid menses if possible.Ideally, screening is best performed in theabsence of heavy menstrual flow but shouldnot be deferred in the event of abnormalbleeding (i.e., bleeding between periods, postcoital bleeding, postmenopausal bleeding) or ifaccessibility for return examination is difficult.

Collection of a Specimen With the CervicalBroom Device

There are a number of single-samplingdevices that sample both the endocervix andthe T-zone at the same time, with onecollection scraping of the cervix.The cervicalbroom is the most commonly used amongsuch devices. It can be used instead of theextended tip spatula and the endocervicalbrush, and can be used in pregnant women.The long central bristles of the broom areinserted into the endocervical os while thebroom is pressed against the cervix so that theouter bristles bend.The broom is then rotatedin one direction for five complete rotations.(Rotating in one direction, and then the otheris not recommended because it can result inloss of cells.) The cervical broom can be usedwith conventional cervical cytology, as well aswith liquid-based cervical cytology. Forconventional cytology, the broom sample issmeared on a slide (labeled with the patient’sname) by stroking the broom on the usablesurface of the slide (excluding the frosted label)from the label margin toward the other end ofthe slide.The sample is then quickly fixed witha spray fixative or by immersing the slide in95% ethanol. For liquid cytology, one

manufacturer recommends that the broom berinsed as free of cells as possible in thepreservative medium. Another manufacturerrecommends snapping off the handle of theinstrument and leaving the broom end of theinstrument in the fixative solution to be sent tothe laboratory.

Typical Procedure for Making the CervicalCytologic Smear When Employing theExtended Tip Spatula and the EndocervicalBrush Using a Single Slide Technique

Using a glass slide labeled with the patient’sname (a slide with a frosted label space on oneend is preferred, so the patient’s name can bewritten on the slide with pencil) theendocervical brush with the cellular sample isrolled onto the surface of the glass slide. Thebrush sample can be rolled on the slideimmediately adjacent to the frosted label area,usually using about a two cm portion of theslide.The extended tip spatula sample can thenbe smeared on the same slide, smearing thesample immediately adjacent to theendocervical sample and, using a circularmotion, smearing the cellular sample onto theglass slide.The slide is then rapidly fixed with aspray fixative, or immersed in 95% ethanol. If aliquid-based Pap test is used, the brush and theextended tip spatula are carefully rinsed in avial of preservative, and the container is labeledwith the patient’s name.

Submitting Samples Using Liquid-basedTechniques

For liquid cytology, both the endocervicalbrush sample and the extended tip spatulacytologic sample are submitted in a singlecontainer with appropriate fixative solution andlabeled with the patient’s name. A plasticextended tip spatula is recommended forliquid-based techniques. One manufacturerrecommends that the spatula and the




merican C



Dow

nloaded from


REFERENCES1. American Cancer Society, Cancer Facts &Figures 2002.2. Bosch FX, Manos MM, Munoz N, et al.Prevalence of human papillomavirus in cervicalcancer: A worldwide perspective. Internationalbiological study on cervical cancer (IBSCC)Study Group. J Natl Cancer Inst 1995;87:796-802.3.Walboomers JM, Jacobs MV, Manos MM, et al.Human papillomavirus is a necessary cause ofinvasive cervical cancer worldwide. J Pathol1999;189:12-19.4. Park TW, Fujiwara H, Wright TC. Molecularbiology of cervical cancer and its precursors.Cancer 1995;76:1902-1913.5. Holowaty P, Miller AB, Rohan T, et al. Naturalhistory of dysplasia of the uterine cervix. J NatlCancer Inst 1999;91:252-258.6. Hildesheim A, Schiffman MH, Gravitt PE, et al.Persistence of type-specific human papillomavirusinfection among cytologically normal women. JInfect Dis 1994;169:235-240.7. Moscicki AB, Shiboski S, Broering J, et al.Thenatural history of human papillomavirus infectionas measured by repeated DNA testing in adoles-cent and young women. J Pediatr 1998;132:277-284.8. Ho GY, Bierman R, Beardsley L, et al. Naturalhistory of cervicovaginal papillomavirus infectionin young women. N Engl J Med 1998;338:423-428.9. Sherman ME, Schiffman M, Herrero R, et al.Performance of a semiautomated Papanicolaousmear screening system: Results of a population-based study conducted in Guanacaste, Costa Rica.Cancer 1998;84:273-280.10. Cervical cancer. NIH Consens Statement1996 Apr 1-3;14:1-38.11. Fink DJ. Change in American Cancer SocietyCheckup Guidelines for detection of cervical can-cer. CA Cancer J Clin 1988;38:127-128.12. Centers for Disease Control. CDC guidelinefor immunocompromised women; USPHS/IDSAGuidelines for the prevention of opportunisticinfections in persons infected with humanimmunodefiency virus: A summary. Morb MortalWkly Rep 1995;44(RR-8):1-34.13. Mount SL, Papillo JL.A study of 10,296 pedi-atric and adolescent Papanicolaou smear diagnosesin northern New England. Pediatrics 1999;103:539-545.14. Ries LAG, Eisner MP, Kosary CL, et al. (eds).

SEER Cancer Statistics Review, 1973-1999.National Cancer Institute. Bethesda, MD, 2002.Available: http://seer.cancer.gov/csr/1973_1999.15. Syrjanen K, Kataja V,Yliskoski M, et al. Naturalhistory of cervical human papillomavirus lesionsdoes not substantiate the biologic relevance of theBethesda System. Obstet Gynecol 1992;79:675-682.16. Nasiell KV, Roger V, Nasiell M. Behavior ofmild cervical dysplasia during long-term follow-up. Obstet Gynecol 1986;67:665-669.17. Nash JD, Burke TW, Hoskins VJ. Biologiccourse of cervical human papillomavirus infec-tion. Obstet Gynecol 1987;69:160-162.18. Moscicki AB, Hills N, Shiboski S. High regres-sion rate of LSIL in adolescents.Abstract. PediatricAcademic Society Annual Meeting, Baltimore,MD, (5/4-5/7/02).19. Nasiell K, Nasiell M,Vaclavinkova V. Behaviorof moderate cervical dysplasia during long-termfollow-up. Obstet Gynecol 1983;61:609-614.20. Goldie SJ, Kim J, Moscicki AB. Alternativepolicies for the initiation of cervical cancerscreening. Plenary presentation at the NationalMeeting of the Society for Medical DecisionMaking. San Diego, CA. October 2001. Abstractavailable: http://www.smdm.org/.21. Mandelblatt JS, Lawrence WF,Womack SM, etal. Benefits and costs of using HPV testing toscreen for cervical cancer. JAMA 2002;287:2372-2381.22. Gustafsson L, Sparén P, Gustafsson M, et al.Low efficiency of cytologic screening for cancerin situ of the cervix in older women. Int J Cancer1995;63:804-809.23.Van Wijngaarden WJ, Duncan ID. Rationale forstopping cervical screening in women over 50. BrMed J 1993;306:967-971.24. Sigurdsson K.Trends in cervical intra-epithe-lial neoplasia in Iceland through 1995: Evaluationof targeted age groups and screening intervals.Acta Obstet Gynecol Scand 1999;78:486-492.25. Lawson HW, Lee, NC, Thames SF, et al.Cervical cancer screening among low-incomewomen: Results of a national screening program1991-1995. Obstet Gynecol 1998;92:745-752.26. Mandelblatt J, Gopaul I, Wistreich M.Gynecological care of elderly women: Anotherlook at Papanicolaou smear testing. JAMA1986;256:367-371.27. Sawaya GF, Grady D, Kerlikowske K, et al.Thepositive predictive value of cervical smears in pre-viously screened postmenopausal women: The

Heart and Estrogen/progestin Replacement Study(HERS).Ann Intern Med 2000;133:942-950.28. Sawaya GF, Kerlikowske K, Lee NC, et al.Frequency of cervical smear abnormalities within3 years of normal cytology. Obstet Gynecol2000;96:219-223.29. Pearce KF, Haefner HK, Sarwar SF, et al.Cytopathological findings on vaginalPapanicolaou smears after hysterectomy forbenign gynecologic disease. N Engl J Med1996;335:1559-1562.30. Piscitelli JT, Bastian LA, Wilkes A, et al.Cytologic screening after hysterectomy for benigndisease. Am J Obstet Gynecol 1995;173:424-430;discussion 430-432.31. Videlefsky A, Grossl N, Denniston M, et al.Routine vaginal cuff smear testing in post-hys-terectomy patients with benign uterine condi-tions:When is it indicated? J Am Board Fam Pract2000;13:233-238.32. Fox J, Remington P; Layde P, et al.The effectof hysterectomy on the risk of an abnormalscreening Papanicolaou test result. Am J ObstetGynecol 1999;180:1104-1109.33. Wiener JJ, Sweetnam, PM, Jones JM. Longterm follow up of women after hysterectomy witha history of pre-invasive cancer of the cervix. Br JObstet Gynaecol 1992;99:907-910.34. Davey D, Austin RM, Birdsong G, et al.ASCCP patient management guidelines: Pap testspecimen adequacy and quality indicators. J LowerGenital Tract Disease 2002;6:195-199.35. Solomon D, Davey D, Kurman R, et al. The2001 Bethesda System:Terminology for reportingresults of cervical cytology. JAMA 2002;287:2114-2119.36. Frame PS, Frame JS. Determinants of cancerscreening frequency:The example of screening forcervical cancer. J Am Board Fam Pract1998;11:87-95.37. Hildesheim A, Hadjimichael O, Schwartz PE,et al. Risk factors for rapid-onset cervical cancer.Am J Obstet Gynecol 1999;180:571-577.38. Sasieni PD, Cuzick J, Lynch-Farmery E.Estimating the efficacy of screening by auditingsmear histories of women with and without cer-vical cancer. The National Co-ordinatingNetwork for Cervical Screening Working Group.Br J Cancer 1996;73:1001-1005.39. International Agency for Research on CancerWorking Group on Evaluation of CervicalCancer Screening Programmes. Screening forsquamous cervical cancer: Duration of low risk



endocervical brush (or the cervical broom, ifused) be rinsed as free of cells as possible in thefixative solution.Another manufacturer recom-mends the sampling devices be separated or cutfrom the handle and placed into the fixative

solution for transport to the laboratory.

Acknowledgments: The authors thank Kim Andrews Sawyerand Connie Lim for their assistance in the preparation of thismanuscript.

CA


merican C



Dow

nloaded from




after negative results of cervical cytology and itsimplication for screening policies. Br Med J293;659-664.40. Lynge E, Poll P. Risk of cervical cancer fol-lowing negative smears in Maribo CountyDenmark 1966-1982. IARC Sci Publ1986;76:69-86.41. Sato S, Makino H, Yajima A, et al. Cervicalcancer screening in Japan. A case-control study.Acta Cytol 1997;41:1103-1106.42. Shy K, Chu J, Mandelson M, et al.Papanicolaou smear screening interval and risk ofcervical cancer. Obstet Gynecol 1989;74:838-843.43. Miller M, et al.The protective effect conferredby different Pap smear screening intervals in ahealth maintenance organization: A case-controlstudy. Obstet Gynecol. In press.44.Viikki M, Pukkala E, Hakama M. Risk of cer-vical cancer after a negative Pap smear. J MedScreen 1999;6:103-107.45. Parazzini F, Negri E, La Vecchia C, et al.Screening practices and invasive cervical cancerrisk in different age strata. Gynecol Oncol1990;38:76-80.46. Sawaya GF, Sung HY, Kinney W, et al. Theeffect of multiple negative screening Pap smearson cervical cancer risk in longterm members of aprepaid health plan. American Society ofColposcopy & Cervical Pathology March 2002.Oral presentation.Abstract.47. Sawaya GF, Sung HY, Kinney, et al. Irreduciblecervical cancer risk following multiple negativescreening Pap smears in longterm members of aprepaid health plan. American Society ofColposcopy & Cervical Pathology March 2002.Oral presentation.Abstract.48.Dominitz JA,Provenzale D.Patient preferencesand quality of life associated with colorectal can-cer screening. Am J Gastroenterol 1997;92:2171-2178.49. Lawrence VA, Gafni A, Kroenke K. Evidence-based vs emotion-based medical decision-making:Routine preoperative HIV testing vs universalprecautions. J Clin Epidemiol 1993;46:1233-1236.50. Ward J. Population-based mammographicscreening: Does ‘informed choice’ require any lessthan full disclosure to individuals of benefits,harms, limitations, and consequences? Aust N Z JPublic Health 1999;23:301-304.51. Marteau TM, Senior V, Sasieni P. Women’sunderstanding of a “normal smear test result”:Experimental questionnaire based study. Br Med J2001;322:526-528.52. Kim JJ,Wright TC, Goldie SJ. Cost-effective-ness of alternative triage strategies for atypicalsquamous cells of undetermined significance.JAMA 2002;287:2382-2390.53. Goldie SJ, Kim JJ, Wright TC. Decision ana-lytic modeling to inform US national health pol-icy: New guidelines for cervical screening. Oralpresentation at the National SMDM Meeting2002. Available at http://66.40.168.196/abstracts.htm. Last accessed August 25, 2002.54. Diaz-Rosario LA, Kabawat SE. Performanceof a fluid-based thin-layer papanicolaou smearmethod in the clinical setting of an independentlaboratory and an outpatient screening populationin New England. Arch Pathol Lab Med1999;123:817-821.

55. Lee KR, Ashfaq R, Birdsong GG, et al.Comparison of conventional Papanicolaou smearsand a fluid-based thin- layer system for cervicalcancer screening. Obstet Gynecol 1997;90:278-284.56. Marino JF, Fremont-Smith M. Direct-to-vialexperience with AutoCyte PREP in a small NewEngland regional cytology practice. J Reprod Med2001;46:353-358.57.Hutchinson ML,Zahniser DJ, Sherman ME, etal. Utility of liquid-based cytology for cervicalcarcinoma screening: results of a population-basedstudy conducted in a region of Costa Rica with ahigh incidence of cervical carcinoma. Cancer1999;87:48-55.58. Belinson J, Qiao YL, Pretorius R, et al. ShanxiProvince Cervical Cancer Screening Study: Across-sectional comparative trial of multiple tech-niques to detect cervical neoplasia. GynecolOncol 2001;83:439-444.59. Clavel C, Masure M, Bory JP, et al. Humanpapillomavirus testing in primary screening forthe detection of high-grade cervical lesions: Astudy of 7932 women. Br J Cancer 2001;84:1616-1623.60. Obwegeser JH, Brack S. Does liquid-basedtechnology really improve detection of cervicalneoplasia? A prospective randomized trial com-paring the ThinPrep Pap Test with the conven-tional Pap Test including follow-up of HSIL cases.Acta Cytol 2001;45:709-714.61. Belinson JL, Pan QJ, Biscotti C, et al. Primaryscreening with liquid-based cytology in anunscreened population in rural China with anemphasis on reprocessing unsatisfactory samples.Acta Cytol 2002;46:470-474.62. Ashfaq R, Gibbons D,Vela C, et al. ThinPrepPap Test. Accuracy for glandular disease. ActaCytol 1999;43:81-85.63. Bai H, Sung CJ, Steinhoff MM.ThinPrep PapTest promotes detection of glandular lesions of theendocervix. Diagn Cytopathol 2000;23:19-22.64. Hecht JL, Sheets EE, Lee KR. Atypical glan-dular cells of undetermined significance in con-ventional cervical/vaginal smears and thin-layerpreparations. Cancer 2002;96:1-4.65.Wright TC Jr, Cox, JT, Massad LS, et al. 2001Consensus Guidelines for the management ofwomen with cervical cytological abnormalities.JAMA 2002;287:2120-2129.66. Cuzick J, Beverley E, Ho L, et al. HPV testingin primary screening of older women. Br J Cancer1999;81:554-558.67. Clavel C, Masure M, Bory JP, et al. HybridCapture II-based human papillomavirus detec-tion, a sensitive test to detect in routine high-grade cervical lesions: A preliminary study on1518 women. Br J Cancer 1999;80:1306-1311.68. Belinson J, Qiao Y, Pretorius R, et al.Prevalence of cervical cancer and feasibility ofscreening in rural China: A pilot study for theShanxi Province Cervical Cancer ScreeningStudy. Int J Gynecol Cancer 1999.;9:411-417.69. Kuhn L, Denny L, Pollack A, et al. Humanpapillomavirus DNA testing for cervical cancerscreening in lowresource settings. J Natl CancerInst 2000;92:818-825.70. Ratnam S, Franco EL, Ferenczy A. Humanpapillomavirus testing for primary screening ofcervical cancer precursors. Cancer EpidemiolBiomarkers Prev 2000;9: 945-951.

71. Schiffman M, Herrero R, Hildesheim A, et al.HPV DNA testing in cervical cancer screening:Results from women in a high-risk province ofCosta Rica. JAMA 2000;283:87-93.72. Wright TC Jr, Denny L, Kuhn L, et al. HPVDNA testing of self-collected vaginal samplescompared with cytologic screening to detect cer-vical cancer. JAMA 2000;283:81-86.73. Blumenthal PD, Gaffikin. L, Chirenje ZM, etal. Adjunctive testing for cervical cancer in lowresource settings with visual inspection, HPV, andthe Pap smear. Int J Gynaecol Obstet 2001;72:47-53.74. Sherman ME, Lorincz AT, Scot, et al.Cytopathology and human papillomavirus testingto determine risk for cervical neoplasia: A ten-year cohort analysis of 20,810 women. J NatlCancer Inst. In Press.75. Franco EL, Ferenczy A.Assessing gains in diag-nostic utility when human papillomavirus testingis used as an adjunct to papanicolaou smear in thetriage of women with cervical cytologic abnor-malities.Am J Obstet Gynecol 1999;181:382-386.76. Hutchinson ML, Isenstein LM, Goodman A, etal. Homogeneous sampling accounts for theincreased diagnostic accuracy using the ThinPrepProcessor.Am J Clin Pathol 1994;101:215-219.77. Joseph MG, Cragg F,Wright VC, et al. Cyto-histological correlates in a colposcopic clinic:A 1-year prospective study. Diagn Cytopathol1991;7:477-481.78. Kristensen GB, Skyggebjerg KD, Holund B, etal. Analysis of cervical smears obtained withinthree years of the diagnosis of invasive cervicalcancer.Acta Cytol 1991;35:47-50.79.Van der Graaf Y,Vooijs GP. False negative ratein cervical cytology. J Clin Path 1987;40:438-442.80. Rubio C. The false negative smear. II. Thetrapping effect of collecting instruments. ObstetGynecol 1977;49:576-580.81. Martin-Hirsch P, Lilford R, Jarvis G, et al.Efficacy of cervical-smear collection devices: asystematic review and meta-analysis. Lancet1999;354:1763-1770.82. Alons-van Kordelaar JJM, Boon ME.Diagnostic accuracy of squamous cervical lesionsstudied in spatula- cytobrush smears. Acta Cytol1988;32:801-804.83. Chakrabarti S, Guijon FB, Parakevas M. Brushvs spatula for cervical smears: Histologic correla-tion with concurrent biopsies. Acta Cytol1994;84:168-173.84. Boon ME, de Graaff Guilloud JC, RietveldWJ. Analysis of five sampling methods for thepreparation of cervical smears. Acta Cytol1989;33:843-848.85. Pinto M, Waradzin M. Papanicolaou smearsprepared using combined cytobrush-spatula sam-pling.Acta Cytol 1989;33:705.86. Germain M, Heaton R, Erickson, et al. Acomparison of the three most commonPapanicolaou smear collection techniques. ObstetGynecol 1994;84:168-173.87. Partoll L, Javaheri G. Cervical cytology aftercryosurgery laser ablation and conization:A com-parison of the cotton swab and the endocervicalbrush.Acta Cytol 1993;37:876-878.88. Orr JW, Barrett JM, Orr PJ, et al.The efficacyand safety of the cytobrush during pregnancy.Gynecol Oncol 1992;44:260-262.


merican C



Dow

nloaded from


In the November/December 2002 issue, inthe article “Cancer Statistics for AfricanAmericans” (Ghafoor A, Jemal A, Cokkinides V,et al. CA Cancer J Clin 2002:52;326-341), anerror appeared in the data for Tables 2 and 3

(pages 330 and 331, respectively). Table 2 andTable 3 have been modified with the correcteddata and are reproduced here.We apologize forthese errors and any confusion they may havecaused.


Errata

*Rates are per 100,000 and age adjusted to the 2000 US standard population.

†Site selected if cases greater than 100.

Source: Surveillance, Epidemiology, and End Results program, Division of Cancer Control and Population Sciences, National Cancer Institute,2002.

African American to White Incidence Rate Ratios, US, 1995 to 1999*

Males Females

Cancer Type† African American White African Cancer Type† African American White African American/White American/White

Rate Rate Ratio Rate Rate Ratio

Myeloma 13.1 6.5 2.0 Myeloma 10.3 4.2 2.5Stomach 19.3 10.7 1.8 Stomach 10.3 4.6 2.2Esophagus 12.8 7.6 1.7 Esophagus 4.4 2.0 2.2Larynx 12.2 7.1 1.7 Small Intestine 2.9 1.5 1.9Prostate 266.8 163.2 1.6 Larynx 2.9 1.6 1.8Liver and Uterine Cervix 13.6 8.1 1.7

Intrahepatic Bile Duct 10.3 6.4 1.6 Pancreas 15.2 9.5 1.6Lung and Bronchus 125.6 84.4 1.5 Liver and Pancreas 18.2 12.3 1.5 Intrahepatic Bile Duct 4.0 2.5 1.6Oral Cavity and Pharynx 21.8 16.5 1.3 Kidney and Renal Pelvis 9.8 7.8 1.3Kidney and Renal Pelvis 18.4 15.6 1.2 Colon and Rectum 56.1 47.1 1.2Colon and Rectum 69.0 65.0 1.1 Soft Tissue (including heart)2.6 2.4 1.1Non-Hodgkin’s Lung and Bronchus 54.4 53.3 1.0

Lymphoma 19.9 24.5 0.8 Breast 123.7 140.9 0.9Leukemia 12.6 16.7 0.8 Urinary Bladder 7.7 10.2 0.8Myeloid Leukemia 5.7 6.9 0.8 Leukemia 7.6 9.8 0.8Acute Myeloid Leukemia 3.4 4.5 0.8 Uterine Corpus 17.3 26.1 0.7Hodgkin’s Lymphoma 2.7 3.4 0.8 Hodgkin’s Lymphoma 2.0 2.8 0.7Lymphocytic Leukemia 5.3 7.4 0.7 Ovary 12.0 18.1 0.7Chronic Lymphocytic Non-Hodgkin’s

Leukemia 4.1 5.5 0.7 Lymphoma 11.2 16.5 0.7Brain 4.6 8.3 0.6 Thyroid 5.6 10.0 0.6Urinary Bladder 19.3 39.8 0.5

All Cancers 695.3 560.1 1.2 All Cancers 412.9 433.5 1.0

TABLE 2

Errata


merican C



Dow

nloaded from


African American to White Mortality Rate Ratios, US, 1995 to 1999*

Males Females

Cancer Type† African American White African Cancer Type† African American White African American/White American/White

Rate Rate Ratio Rate Rate Ratio

Larynx 5.8 2.4 2.4 Stomach 6.8 3.0 2.3Prostate 72.8 31.2 2.3 Myeloma 6.8 3.0 2.3Stomach 14.2 6.3 2.3 Uterine Cervix 6.2 2.8 2.2Myeloma 9.2 4.5 2.0 Esophagus 3.5 1.7 2.1Oral Cavity and Pharynx 8.3 4.2 2.0 Larynx 0.9 0.5 1.8Esophagus 12.9 7.2 1.8 Uterine Corpus 3.2 2.0 1.6Liver and Soft Tissue

Intrahepatic Bile Duct 9.2 5.9 1.6 (including heart) 1.9 1.3 1.5Lung and Bronchus 109.1 79.7 1.4 Colon and Rectum 25.4 18.0 1.4Pancreas 16.2 12.0 1.4 Pancreas 13.0 9.0 1.4Small Intestine 0.5 0.7 1.4 Liver and Colon and Rectum 34.4 25.8 1.3 Intrahepatic Bile Duct 3.9 2.8 1.4Kidney and Breast 37.1 28.2 1.3

Renal Pelvis 6.2 6.2 1.0 Urinary Bladder 3.1 2.3 1.3Leukemia 9.3 10.6 0.9 Lung and Bronchus 40.2 41.7 1.0Hodgkin’s Lymphoma 0.7 0.7 1.0 Kidney and Renal Pelvis 2.9 2.9 1.0Lymphocytic Leukemia 3.0 3.3 0.9 Leukemia 5.6 6.1 0.9Myeloid Leukemia 3.7 4.5 0.8 Myeloid Leukemia 2.5 2.8 0.9Non-Hodgkin’s Lymphoma 7.8 11.2 0.7 Ovary 7.6 9.3 0.8Urinary Bladder 5.7 8.0 0.7 Non-Hodgkin’s Lymphoma 4.7 7.5 0.6Brain 3.2 6.0 0.5 Brain 2.3 4.1 0.6Melanoma 0.5 4.4 0.1 Melanoma 0.5 2.1 0.2

All Cancers 359.2 253.0 1.4 All Cancers 203.5 169.8 1.2

TABLE 3

*Rates are per 100,000 and age adjusted to the 2000 US standard population.

†Site selected if number of deaths is greater than 100.

Source: Underlying mortality data provided by the National Center for Health Statistics. Available at: http://www.cdc.gov/nchs.

In the November/December 2002 issue, inthe article “American Cancer SocietyGuideline for the Early Detection of CervicalNeoplasia and Cancer” (Saslow D, RunowiczCD, Solomon D, et al. CA Cancer J Clin2002:52;342-362), an error appeared in theauthor byline on page 342.

The byline should have read “Debbie Saslow,PhD; Carolyn D. Runowicz, MD (for the Work

Group on Interval, Older Women, andHysterectomy); Diane Solomon, MD (for theWork Group on Technologies); Anna-BarbaraMoscicki,MD (for the When to Start ScreeningWork Group); Robert A. Smith, PhD; HarmonJ. Eyre, MD; Carmel Cohen, MD (for theGynecologic Cancer Advisory Group).”

We apologize for this error and anyconfusion it may have caused.

Volume 53 • Number 2 • March/April 2003 127



merican C



Dow

nloaded from


American Cancer Society Guidelines for the Early Detection of Cancer, 2004

Documents

Transcript of American Cancer Society Guidelines for the Early Detection of Cancer, 2004