Introduction to the Currently Confusing State of US Cervical Screening
There is widespread and growing confusion and controversy1–3 about cervical screening in the United States. Women and their providers face an expanding number of competing screening and triage approaches.4 Although a wealth of testing options can now provide unprecedented, “precision” prediction of cervical precancer risk as a surrogate endpoint for cancer risk,5,6 the loss of simplicity and uniformity of our public health message (“Get your annual Pap”) is hindering movement to improved testing technology and strategies.
In the current state of flux, national guidelines and less formal guidance offer an overabundance of options and rather complex algorithms, based on cytology alone, cytology combined with human papillomavirus (HPV) testing (cotesting), or most recently stand-alone HPV testing.7–10 Stand-alone HPV testing is only approved for 1 specific HPV test,10 but more Food and Drug Administration (FDA) approvals for this indication are very likely to follow.
There is debate regarding both negative and positive screening results, complicated by the coexistence of the several different test options. Most prominently, there is considerable resistance to the 5-year screening interval recommended for a negative cotest (both cytology and HPV negative).1 The management of positive screening results is equally unresolved. Although it is widely acknowledged that positive HPV tests require triage rather than universal, immediate colposcopic referral, the optimal triage methods are undecided and quite varied. Possible triage methods include cytology3,9–13 and related methods (e.g., p16/Ki67 dual staining),14 HPV genotyping in various configurations,8,10,15 and other promising novel technologies including redesigned automated cytology (as presented by Schiffman et al. at the 30th International Papillomavirus Conference in 2015) and biomarkers such as methylation.16–18 The multiple screening/triage combinations need to be better validated in large prospective studies and, in the absence of sufficient information, they have not been addressed in formal guidelines.
Moreover, cervical screening is a lifelong process rather than a single testing visit, further raising the level of complexity.5 For the subset of women needing management of positive findings, repeated testing with varying test methods makes interpretation of results more challenging. Most clinicians likely will find it increasingly difficult to integrate in “real time” the full detail of available different kinds of test results, over multiple rounds of testing. In short, the effectiveness of cervical screening in the United States could be threatened by excessive and increasing complexity.
Although the current state of confusion is widely recognized, achieving more unified and widely accepted approaches to cervical screening in the United States will take considerable, concerted effort. Many important practical factors (e.g., cost, societal emphasis on safety, established laboratory and clinical practice patterns, and outdated quality metrics based on annual cytology) influence how US women are screened. Scientifically demonstrated effectiveness is only 1 consideration in choosing between available strategies. Nonetheless, a discussion of relevant scientific evidence is a good place to begin; accordingly, the following discussion offers some research-backed suggestions for simplification as HPV testing enters US screening programs.
Societal Acceptance of a Low but Nonzero Level of Cancer Risk
Screening simply cannot provide complete safety against cervical cancer, even if we were to adopt frequently repeated HPV-cytology cotesting starting in adolescence, at the cost of massive overtreatment.7,8 For example, there are rare rapidly progressive cancers in very young women,19 uncommon cotest-negative cancers,20 and deep canal lesions that escape detection and progress, despite careful protocols. In deciding on screening strategies, we must explicitly accept low, nonzero risk levels. No group holds the sole expertise in setting the acceptable level. To reiterate a fundamental point, societal understanding that a small number of cancers will unfortunately still occur in the presence of good cervical screening is a precondition to devising rational screening and management guidelines that balance safety and overtreatment.
The Goal of Cervical Screening Programs: Diagnosing and Treating Precancer to Prevent Cancer
Strictly speaking, “screening” applies to women at presumed low general population risk of disease, whereas “management” applies to the actions taken after a positive screen. An HPV or cytologic test can be performed in either setting. Most discussions of screening programs, including this one, address both population screening and management of abnormal results.
The primary goal of cervical cancer screening programs in the United States is not to find invasive cancer. Rather, the goal is to prevent cervical cancer by detecting and treating true cancer precursors, which for this presentation will be called “precancer.” Precancer is the subset of high-grade intraepithelial lesions, or cervical intraepithelial neoplasia (CIN) 3/adenocarcinoma in situ (AIS), or whatever terms are used in the prevailing classification scheme, which would progress to invasive cancer if left untreated. Treatment of a lesion not destined to progress to morbid cancer is overtreatment.21 Just as perfect safety is unachievable, overtreatment is unavoidable, although we hope to limit it while maintaining acceptable safety. Deciding on the right balance is critical to setting screening intervals (because overtesting can lead to overtreatment) and making management guidelines.
For example, 1 ongoing debate is how long to wait for the next routine visit after a negative screen,1,7,8 which is a nonscientific question of societal or personal tolerance of small risks, and the value assigned to avoiding such risks. An ideal cervical screening interval would lead to meaningfully large numbers of detected true precancers but virtually no cancers. Finding very few true precancers per screen implies that the screening interval may be too short. Finding too many invasive cancers implies that the interval may be too long. The exact balance must be discussed and decided.
Simplified View of HPV Natural History and the Stages in Cervical Cancer Development
It is important to agree on how to view the disease process that cervical screening is designed to interrupt. Many diagnostic terms are used in various cervical screening programs; however, the key stages in cervical carcinogenesis can be conceived of simply as normal cervix, high-risk HPV infection, precancer, and cancer (see Figure 1, top row).22 Each stage of cervical carcinogenesis has a characteristic age distribution; the characteristic age curves can be viewed as proxies for time since cervical HPV acquisition, which occurs mainly in adolescence and early adulthood.22 Each stage logically links to a corresponding clinical action. In the absence of diagnostic error, the stages can be viewed conceptually as follows: women without current HPV infection of one or more of the dozen high-risk (also called carcinogenic or oncogenic) HPV types23 are at virtually zero risk of having concurrent cervical cancer and are at ultra low risk for the next several years.24–26 Human papillomavirus negativity should lead to continued routine screening at extended intervals. Only women testing positive for current infection from carcinogenic HPV types (HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) require intensified management. The high-risk types are genetically related and restricted to a few species groups within the alpha genus of HPV.27 These infections as a group are nonetheless extremely common and almost always “clear” within 1 to 2 years.28,29 At any age, newly appearing cervical HPV infections are most likely to clear, whereas the longer an HPV infection persists (remains detectable on one of the standard DNA/RNA assays), the more likely the infection will continue to persist, associated with a high long-term risk of precancer.25,29,30 The theoretically ideal response to HPV infection without precancer would be to retest at an interval that permits the major proportion of infections to clear, e.g., 2 years. In practice, it is difficult for woman and their providers to wait that long and prolonged follow-up intervals risk loss to follow-up; therefore, return testing at approximately 1 year is currently recommended.8
Predicting the immediate risk and subsequent outcome of infections is currently the major challenge in cervical screening. Two principles are worth emphasizing: as part of minimizing overdiagnosis and overtreatment, the carcinogenic type group must be defined as those types causing invasive cancer, not precancer, which can be caused by a larger group of types.31,32 Second, within the high-risk group, the 13 carcinogenic HPV types vary greatly in carcinogenic strength, from the uniquely strong (HPV 16) to the marginal (e.g., HPV 51).31,33
The task of defining which infections place a woman at risk, whereas most are benign, is made more challenging by our difficulty in defining true precancers to treat. We presently do not have a way to distinguish a precancer that will invade from one that will not.34 Pragmatically, all currently used definitions of precancer represent some overdiagnosis with resultant overtreatment, by including a majority that will not progress to cancer. Thus, cervical precancer remains a heterogeneous classification as defined by current colposcopic biopsy methods,35 and although it is the best diagnostic method we currently have, it remains an imperfect surrogate endpoint for cancer.
Clinical Test Results Compared With the 4 Underlying Stages of Carcinogenesis
We need to understand the strengths and weaknesses of each cervical testing method (histopathology from colposcopically directed biopsy, cytology, and HPV testing) in determining whether the true state of the cervix is normal, HPV infection, precancer, or cancer.
Moving from top to bottom in Figure 1, we rely primarily on histopathologic diagnosis from colposcopically directed biopsy to define whether precancer or cancer is present. Histopathology remains our reference diagnostic method but, nonetheless, it is misclassified on multiple levels. Histopathologic definition of precancer is especially prone to false-positive, “high-grade lesions” (CIN 2 and even CIN 3, noting that we know much less about AIS) that would never lead to clinically diagnosed cancers even if untreated, as evidenced most compellingly by the very large ratios of CIN 2 and 3 found in screening programs compared with observed lifetime risks of invasive cancer in previously unscreened populations.36 Moreover, error in colposcopic targeting of biopsies leads to underdiagnosis and underestimates the prevalence of possibly precancerous lesions in a woman.37 At the low-grade end of the spectrum, the distinction between histopathologic diagnoses of HPV infection (CIN 1 or low-grade squamous intraepithelial lesion [LSIL]) and noninfection (benign biopsy) is not reproducible, sensitive, or specific, severely limiting the usefulness of this category.38
Cytologic definition of the state of the cervix began in the original Papanicolaou classification as a probability of underlying cancer but has become increasingly “diagnostic.” The Bethesda System has terms corresponding to normal (negative for intraepithelial lesion of malignancy), infection (LSIL), precancer (high-grade squamous intraepithelial lesion [HSIL]/AIS), and cancer39; however, the accuracy of cytology at “the low end” for defining high-risk HPV infection is very low. Most high-risk infections have concurrent normal cytologic results,40 and many minor cytologic abnormalities are caused by low-risk HPV types or are unrelated to HPV.41 Moreover, a large proportion of histopathologic precancers are found among women with LSIL or equivocal results (i.e., atypical squamous cells of undetermined significance [ASC-US]) rather than HSIL.42
Human Papillomavirus Test Results
High-risk type-specific or pooled HPV testing provides the reference definition of high-risk infections. However, HPV testing cannot distinguish infection from precancer or cancer. Even genotype-specific typing yields only risk associations, not unequivocal distinctions.
Phasing Out Unneeded and Unhelpful Tests, Terms, and Distinctions
In striving for simplified essential screening strategies, it might be time to retire some familiar but ultimately distracting details of current HPV tests, cervical cytology classifications, and histologic nomenclatures. Some might be viewed as holdovers from earlier classification schemes that predated our improved understanding of HPV natural history.
Simplifying HPV Testing
First, there is no clinical reason to test for HPV infections other than the 13 high-risk types.13,43 For the core group of proven carcinogenic types, the amount of detail given by the testing laboratory to the clinician and woman is worth careful consideration. Different HPV infections act independently and one might persist while another clears.44 There is no evidence at present that tracking them at the individual type level is clinically useful. There is some evidence for distinguishing the highest-risk HPV types such as HPV 16. Defining the optimal amount of typing detail is an important and current research goal.
Simplifying Cytologic Results
There is strong evidence supporting the combination of LSIL and HPV-positive ASC-US into a diagnostic term indicating “cytologic evidence of HPV infection.”5 Atypical squamous cells of undetermined significance represent the majority of nonnormal cytologic results, but it is an unreliable characterization, has limited biological meaning (it expresses uncertainty between NILM and LSIL), and can be divided by HPV testing into normal cervix (uninfected with high-risk HPV types, with risk similar to a negative cytology result) versus HPV infected (equivalent in risk of precancer to LSIL).45,46 Combining LSIL and HPV-positive ASC-US would create a relatively large category encompassing the great majority of all nonnormal results, in accordance with the high prevalence of HPV infection compared with the uncommon occurrence of precancer. If HPV testing is used instead of cytology to define infection, the most useful cytologic distinctions are the severe ones that suggest underlying precancer; HSIL (with the corresponding equivocal result of ASC-H) indicates a high risk of underlying precancer, but less than 1% of women receive this result.47 According to current guidelines, women with HSIL can be treated immediately as an option.8 Of note, almost no women are classified in US screening programs as having cytologic results of invasive cancer.5 Similarly, glandular results of atypical glandular cells and AIS are useful when found but rare.
Simplifying Histopathologic Diagnoses
The recently proposed Lower Anogenital Squamous Terminology (LAST) nomenclature abandons altogether CIN and older nomenclatures,35 which have been shown not to reflect accurately the stages of cervical carcinogenesis. If only women who are HPV positive were referred to colposcopy, pathologists could concentrate on whether the histology revealed precancer (or cancer, with such details as glandular vs squamous pathology). They would not need to focus on reporting histologic signs of HPV infection. Accordingly, the LAST proposal accepts CIN 3 as true precancer (as a practical compromise until a more specific biomarker is found) and recommends p16 staining to adjudicate equivocal precancers (including CIN 2). The LAST proposal will help if used as recommended to clarify CIN 2 diagnoses. However, there is a risk of overusing p16 staining in CIN 1 cases, with the result of upgrading and overtreating lesions that represent infections, because many CIN 1 lesions are p16 positive. Consequently, judicious use of p16 staining will be necessary to bring histologic diagnosis of high-grade lesion into closest possible agreement with true underlying precancer.
Gradual Replacement of Cervical Cytology by HPV-Based Strategies for Primary Screening
Simply put, HPV testing is the most sensitive method to distinguish the normal cervix from a high-risk HPV infection indicating appreciable risk of precancer/cancer. All major comparisons, except for studies with serious ascertainment biases,2,3 have reported that HPV is more sensitive than cytology, predicting lower risks when negative and holding prognostic value for longer subsequent periods.20,48,49
The superior sensitivity and negative predictive value of HPV testing will likely lead to its adoption as the main primary screening method. As an interim strategy in the transition to HPV testing, cotesting might be used, but that is an expensive strategy not under consideration outside of the United States. It has been shown quite convincingly that among women who are HPV negative, the marginal increase in reassurance achieved by adding cytology is small, approximately 0.003% lower risk of cancer for 5 years.20 Although HPV-negative cancers do rarely occur, HPV test-negative, cytology-positive cotest results mainly represent HPV testing error or cytologic abnormalities (mainly ASC-US caused by low-risk HPV types or as “look-alikes”) unrelated to cervical cancer risk.50 Eventually, the impact of HPV vaccination is likely to push screening away from cytology or cotesting to stand-alone HPV screening that includes partial typing.51,52
If we accept that the United States will eventually rely on HPV-based screening rather than cytology, then it is important to validate the best possible FDA-approved tests and to understand how they compare in distinguishing infections from precancer. The available cross-sectional and the scant prospective comparative data indicate gross comparability of the approved tests when negative.53,54 In other words, the most highly validated tests that have obtained or are seeking FDA approvals are roughly comparable in analytic sensitivity and prediction of low risk of underlying or subsequent precancer when negative. Thus, the similarities of the assays outweigh the differences at the first-cut level of determining whether the primary screening result should be classified as normal or abnormal (including HPV infected/precancer/cancer). This agreement across assays is much higher than the typically reported reproducibility of cervical cytology.33,38,53–55 If the similar reassurance provided by approved tests is confirmed by large-scale prospective data, it will mean that it does not matter much to clinical management of negative results, which validated test is used, and medical histories will not need to distinguish which HPV test was negative.
The main uncertainties regarding HPV screening relate to the management of the less than 10% of women in the targeted age group that test positive. The high prevalence of HPV relative to the low risk that infection represents diagnosable precancer/cancer led the recent ASCCP-sponsored guidelines committee to recommend against sending all women who are HPV positive to colposcopic biopsy.5,8,56 The results of worldwide testing of invasive cancers, including adenocarcinomas, suggest that genotyping for HPV 16 and 18 (and perhaps HPV 45) is worthwhile in guiding management.31,33 The value of identifying more types, perhaps combined in groups according to risk of developing precancer, requires study. The newer test methods tend to include partial genotyping57–59; in the United States, women with the highest-risk genotypes (HPV 16 and 18 in current guidelines) are sent immediately to colposcopy, whereas the rest are retested in a year.8
Even with partial genotyping, HPV screening demands a paired triage test if one wishes to avoid high rates of referral to colposcopy with low yield of precancer/cancer. The combinations include cytology (perhaps grouped in a few categories) and some promising, novel triage approaches, such as a immunocytochemical dual stain on a cytology preparation that is scored as positive or negative,14 and an automated cytologic method that is programmed to provide a severity score possibly useful for triage, (as presented by Schiffman et al. at the 30th International HPV Conference in 2015). Methylation methods would not require the preparation of a cytology slide but are early in development.16
The marketplace will not necessarily provide the optimal answer to which of the increasing number of available assays should be used for screening and triage. The required strictness and specificity of FDA approvals are creating competing “systems,” each approved with detailed specifications of collection device (swab, brush, broom), collection buffer, and testing technology. Each company with FDA-approved or pending tests has its own approach, several of which do not include conventional cytologic screening and interpretation.
Because new assays are introduced in the market, these combinations may change, but the companies are naturally inclined to seek increased market share with their proprietary combinations. Understanding the best approaches will require unbiased, publicly financed, and rigorous comparisons of HPV screening and paired triage methods. The needed head-to-head comparisons will require very large, prospective cohorts with biospecimen collections and long-term follow-up. The ability to (1) test for multiple assays from the same specimens and (2) obtain excellent disease ascertainment in long-term follow-up permits us to address important questions about screening and management in observational studies, without requiring randomized controlled trials for every decision. Linking the performance data to costs will enable decision makers to choose between alternatives.
As one such effort, the US National Cancer Institute (NCI) is conducting a trial to provide the “big data” needed to compare HPV screening and triage tests. In collaboration with Kaiser Permanente Northern California (KPNC), NCI is collecting residual cervical biospecimens from approximately 70,000 women (including 50,000 HPV positive) cotested over the next few years. This random sample of the KPNC population will be followed actively with repeat collections for 5 years and then passively by linkage to the KPNC electronic medical records for another 5 years. The NCI expects to compare all FDA-approved screening and triage assays using the stored aliquots. The sampling and testing schemes are beyond the scope of this discussion, but the aim is simple: to provide the comparative effectiveness data needed for the next set of cervical screening guidelines. The youngest members of the cohort will include many vaccinated women, permitting consideration of that increasingly important risk modifier. A major cost-effectiveness component is incorporated. Collaboration with many other, similar projects underway throughout the world will be a priority, to achieve the sample sizes and data/specimen exchanges needed for international comparisons.
Adoption of a Risk-Based Approach to Cervical Screening and Management
Early cervical screening and management guidelines were based in large part on expert opinion, which invariably relied on the experts' implicit sense of risk. In other words, the experts' cumulative experiences led them to internal risk estimates that they used to argue for more or less intensive interventions. As a step forward, later guidelines have incorporated more transparent and explicit principles of “equal management of equal risk,” “risk thresholds,” and “benchmarking.”60 Equal management of equal risk implies that regardless of the test results leading to a particular risk, management should be the same. A good example is HPV-positive ASC-US and LSIL, which have equivalent risks and should be managed the same (or, as we propose, should be combined).
Thresholds of risk of precancer or preferably cancer are set for each level of management. Decisions based on estimates of immediate versus 3- or 5-year risk tend to agree, with a few important exceptions; HPV-positive NILM tends to have low immediate risk of detecting precancer, but the risk rises sharply with follow-up because lesions grow to the level of colposcopic detectability. The current options in the United States could be limited to 4 risk bands. Recognition that any screening result will inevitably place a woman into 1 of these 4 bands is another part of simplifying cervical screening. The bands are, in order of descending risk/intensity of management: treatment of precancer/cancer, colposcopic biopsy, enhanced (shorter interval) screening, and routine screening. When a new screening or triage test is introduced, the risks after combinations of positive and negative result can be computed, and the proper management (assignment of 1 of the 4 risk/action bands) for each combination can be benchmarked against other test combinations yielding similar risk.
Here, we are not addressing the remaining fifth action band, exiting (when risk of cervical cancer morbidity and mortality approaches zero, e.g., after benign hysterectomy or among older women after a lifetime of normal screening). We still require more prospective evidence to determine how best to exit screening in the HPV testing era; at present, enough follow-up time has not yet passed.
As another simplifying point, in the context of new screening and management guidelines, it is unclear whether any of the known etiologic cofactors for the development of cervical cancer among women with HPV infection22 are clinically important enough to influence management. Ongoing studies are designed to address whether details of smoking status, hormonal contraceptive use, multiparity, or coinfection with other STIs (note that we are not discussing women with HIV infection) warrant altering clinical management of cervical screening results.
Impact of HPV Vaccination
Although the first cohort of vaccinated young women is just now entering the age of screening, it is already evident that vaccination will speed the transition from cytology to HPV testing, specifically to testing with partial typing.61 Vaccination decreases the incidence of subsequent precancer but only modestly reduces the total number of abnormal cytologic results (or pooled HPV typing results). Vaccine-targeted types represent the minority of the total pool of infections, including the pool of types that cause ASC-US or LSIL, which are much more common than HSIL/AIS (the specific cytologic signs of precancer). With “fewer needles in the haystack,” the positive predictive value of cytology or pooled HPV testing suffers in a vaccinated cohort.22,62 Because vaccine penetration increases and time passes, future populations will require adjustments to whatever screening is conducted. Screening below the age of 25 years among vaccinated cohorts might not be needed given extremely low risk of cancer in early ages. For future public health monitoring of vaccine effectiveness and durability of protection in the United States, using HPV tests that distinguish between vaccine-targeted and nontargeted high-risk types would be useful.
Summary of a New Cervical Screening Paradigm: Simple Causal Model, Complex Risk Estimation, Simplified Public Health and Clinical Practices
The major premise of this commentary is that to escape the increasing state of confusion over cervical screening, the prevention community should use the new HPV-based test methods to define simple risk categories that correspond to the stages of cervical carcinogenesis. In moving to a simple, biologically accurate classification scheme linked to clinical actions, the radical part might be the elimination of some current cytologic and even some histologic distinctions and terms that are no longer helpful.
An important goal of guidelines groups, in active dialog with advocates representing those affected, should be to decide something partly nonscientific: the thresholds of acceptable risk linked to each of the management options. It will be necessary to address difficult issues of equipoise when risk estimates from test results are close to established risk thresholds, when there is no obviously correct choice between alternative choices (HPV-positive NILM with certain non-HPV 16 types provides an example of a difficult “judgment call” as to whether colposcopy is warranted.) Once societal norms are established, regardless of tests used, the clinicians can note into which risk band a woman falls, suggesting which of the 4 actions (or 5 actions if exiting is included) they will take. As always, they can modify actions on the basis of special circumstances, recognizing that guidelines function as suggestions not fixed policy.
Extension to Women After Colposcopy
Although beyond the scope of this commentary, similar issues pertain to women after colposcopy if, as is usually the case, a treatable precancer is not found. Colposcopy that does not find precancer will be an increasing occurrence because women are referred for persistent HPV positivity; such women as a group are at risk of precancer that, even if truly present, might be too early and small for visual detection. To avoid creating an even larger group of women attending colposcopy clinic, one important answer will be to increase the reassurance of negative colposcopy by increasing colposcopic sensitivity, e.g., by multiple biopsies targeting even faintly acetowhite lesions.37 At the same time, although there is no existing or (to our knowledge) pending FDA-approved indication for HPV testing of women postcolposcopy, clinical guidelines will likely recommend relying on the reassurance of sequential negative stand-alone HPV testing or some form of cotesting to return women to general screening.63
Finally, it is noteworthy that the next generations of HPV-based tests are already in sight. If we adapt to the principles of equal management of equal risk, risk thresholds, and benchmarking, we will be ready to adjust to the inevitability of further technical change in the spirit of “continuous improvement.”
Mark Schiffman, MD, MPH
Nicolas Wentzensen, MD, PhD, MS
Clinical Genetics Branch
Division of Cancer Epidemiology and Genetics
National Cancer Institute
1. Kinney W, Wright TC, Dinkelspiel HE, et al. Increased cervical cancer risk associated with screening at longer intervals. Obstet Gynecol
2015; 125: 311–5.
2. Blatt AJ, Kennedy R, Luff RD, et al. Comparison of cervical cancer screening results among 256,648 women in multiple clinical practices. Cancer Cytopathol
2015; 123: 282–8.
3. Giorgi-Rossi P, Arbyn M, Meijer CJ. Cervical cancer screening by human papillomavirus testing followed by cytology triage. JAMA Intern Med
2015; 175: 1068.
4. Schiffman M, Wentzensen N, Wacholder S, et al. Human papillomavirus testing in the prevention of cervical cancer. J Natl Cancer Inst
2011; 103: 368–83.
5. Katki HA, Kinney WK, Fetterman B, et al. Cervical cancer risk for women undergoing concurrent testing for human papillomavirus and cervical cytology: a population-based study in routine clinical practice. Lancet Oncol
2011; 12: 663–72.
6. Zhao FH, Hu SY, Zhang Q, et al. Risk assessment to guide cervical screening strategies in a large chinese population. Cancer Epidemiol Biomarkers Prev
7. Saslow D, Solomon D, Lawson HW, et al. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer. J Lower Gen Tract Dis
2012; 16: 175–204.
8. Massad LS, Einstein MH, Huh WK, et al. 2012 updated consensus guidelines for the management of abnormal cervical cancer screening tests and cancer precursors. J Lower Gen Tract Dis
2013; 17 (5 Suppl 1): S1–S27.
9. Rijkaart DC, Berkhof J, van Kemenade FJ, et al. Evaluation of 14 triage strategies for HPV DNA-positive women in population-based cervical screening. Int J Cancer
2012; 130: 602–10.
10. Huh WK, Ault KA, Chelmow D, et al. Use of primary high-risk human papillomavirus testing for cervical cancer screening: interim clinical guidance. J Lower Gen Tract Dis
2015; 19: 91–6.
11. Naucler P, Ryd W, Tornberg S, et al. Efficacy of HPV DNA testing with cytology triage and/or repeat HPV DNA testing in primary cervical cancer screening. J Natl Cancer Inst
2009; 101: 88–99.
12. Bergeron C, Ronco G, Reuschenbach M, et al. The clinical impact of using p16 immunochemistry in cervical histopathology and cytology: An update of recent developments. Int J Cancer
2015; 136: 2741–51.
13. Zhao C, Moriarty AT, Ghofrani M, et al. Human papillomavirus testing and reporting rates in 2012: results of a College of American Pathologists national survey. Arch Pathol Lab Med
2015; 139: 757–61.
14. Wentzensen N, Fetterman B, Castle PE, et al. p16/Ki-67 dual stain cytology for detection of cervical precancer in hpv-positive women. J Natl Cancer Inst
15. Cox JT, Castle PE, Behrens CM, et al. Comparison of cervical cancer screening strategies incorporating different combinations of cytology, HPV testing, and genotyping for HPV 16/18: results from the ATHENA HPV study. Am J Obstet Gynecol
2013; 208: 184 e1–e11.
16. Wentzensen N, Sun C, Ghosh A, et al. Methylation of HPV18, HPV31, and HPV45 genomes and cervical intraepithelial neoplasia grade 3. J Natl Cancer Inst
2012; 104: 1738–49.
17. Bierkens M, Hesselink AT, Meijer CJ, et al. CADM1 and MAL promoter methylation levels in hrHPV-positive cervical scrapes increase proportional to degree and duration of underlying cervical disease. Int J Cancer
2013; 133: 1293–9.
18. Louvanto K, Franco EL, Ramanakumar AV, et al. Methylation of viral and host genes and severity of cervical lesions associated with human papillomavirus type 16. Int J Cancer
2015; 136: E638–45.
19. Hildesheim A, Hadjimichael O, Schwartz PE, et al. Risk factors for rapid-onset cervical cancer. Am J Obstet Gynecol
1999; 180 (3 Pt 1): 571–7.
20. Gage JC, Schiffman M, Katki HA, et al. Reassurance against future risk of precancer and cancer conferred by a negative human papillomavirus test. J Natl Cancer Inst
21. Marcus PM, Prorok PC, Miller AB, et al. Conceptualizing overdiagnosis in cancer screening. J Natl Cancer Inst
22. Schiffman M, Castle PE, Jeronimo J, et al. Human papillomavirus and cervical cancer. Lancet
2007; 370: 890–907.
23. Bouvard V, Baan R, Straif K, et al. A review of human carcinogens—Part B: biological agents. Lancet Oncol
2009; 10: 321–2.
24. Kjaer S, Hogdall E, Frederiksen K, et al. The absolute risk of cervical abnormalities in high-risk human papillomavirus-positive, cytologically normal women over a 10-year period. Cancer Res
2006; 66: 10630–6.
25. Chen HC, Schiffman M, Lin CY, et al. Persistence of type-specific human papillomavirus infection and increased long-term risk of cervical cancer. J Natl Cancer Inst
2011; 103: 1387–96.
26. Schiffman M, Glass AG, Wentzensen N, et al. A long-term prospective study of type-specific human papillomavirus infection and risk of cervical neoplasia among 20,000 women in the Portland Kaiser Cohort Study. Cancer Epidemiol Biomarkers Prev
2011; 20: 1398–409.
27. Schiffman M, Herrero R, Desalle R, et al. The carcinogenicity of human papillomavirus types reflects viral evolution. Virology
2005; 337: 76–84.
28. Plummer M, Schiffman M, Castle PE, et al. A 2-year prospective study of human papillomavirus persistence among women with a cytological diagnosis of atypical squamous cells of undetermined significance or low-grade squamous intraepithelial lesion. J Infect Dis
2007; 195: 1582–9.
29. Rodriguez AC, Schiffman M, Herrero R, et al. Longitudinal study of human papillomavirus persistence and cervical intraepithelial neoplasia grade 2/3: critical role of duration of infection. J Natl Cancer Inst
2010; 102: 315–24.
30. Kjaer SK, Frederiksen K, Munk C, et al. Long-term absolute risk of cervical intraepithelial neoplasia grade 3 or worse following human papillomavirus infection: role of persistence. J Natl Cancer Inst
2010; 102: 1478–88.
31. Guan P, Howell-Jones R, Li N, et al. Human papillomavirus types in 115,789 HPV-positive women: a meta-analysis from cervical infection to cancer. Int J Cancer
2012; 131: 2349–59.
32. Schiffman M, Clifford G, Buonaguro FM. Classification of weakly carcinogenic human papillomavirus types: addressing the limits of epidemiology at the borderline. Infect Agent Cancer
2009; 4: 8.
33. Schiffman M, Burk RD, Boyle S, et al. A study of genotyping for management of human papillomavirus-positive, cytology-negative cervical screening results. J Clin Microbiol
2015; 53: 52–9.
34. Wentzensen N, Schiffman M, Dunn T, et al. Multiple human papillomavirus genotype infections in cervical cancer progression in the study to understand cervical cancer early endpoints and determinants. Int J Cancer
2009; 125: 2151–8.
35. Darragh TM, Colgan TJ, Cox JT, et al. The Lower Anogenital Squamous Terminology Standardization Project for HPV-Associated Lesions: Background and Consensus Recommendations From the College of American Pathologists and the American Society for Colposcopy and Cervical Pathology. J Lower Gen Tract Dis
2012; 16: 205–42.
36. Schiffman M, Rodriguez AC. Heterogeneity in CIN3 diagnosis. Lancet Oncol
2008; 9: 404–6.
37. Wentzensen N, Walker JL, Gold MA, et al. Multiple biopsies and detection of cervical cancer precursors at colposcopy. J Clin Oncol
2015; 33: 83–9.
38. Stoler MH, Schiffman M. Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic estimates from the ASCUS-LSIL Triage Study. JAMA
2001; 285: 1500–5.
39. Nayar R, Wilbur D. The Bethesda System for Reporting Cervical Cytology. Definitions, Criteria, and Explanatory Notes
. 3rd ed. Cham, Switzerland: Springer International Publishing; 2015.
40. Kovacic MB, Castle PE, Herrero R, et al. Relationships of human papillomavirus type, qualitative viral load, and age with cytologic abnormality. Cancer Res
2006; 66: 10112–9.
41. Zuna RE, Wang SS, Rosenthal DL, et al. Determinants of human papillomavirus-negative, low-grade squamous intraepithelial lesions in the atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesions triage study (ALTS). Cancer
2005; 105: 253–62.
42. Kinney WK, Manos MM, Hurley LB, et al. Where's the high-grade cervical neoplasia? The importance of minimally abnormal Papanicolaou diagnoses. Obstet Gynecol
1998; 91: 973–6.
43. Thomsen LT, Frederiksen K, Munk C, et al. High-risk and low-risk human papillomavirus and the absolute risk of cervical intraepithelial neoplasia or cancer. Obstet Gynecol
2014; 123: 57–64.
44. Liaw KL, Hildesheim A, Burk RD, et al. A prospective study of human papillomavirus (HPV) type 16 DNA detection by polymerase chain reaction and its association with acquisition and persistence of other HPV types. J Infect Dis
2001; 183: 8–15.
45. Results of a randomized trial on the management of cytology interpretations of atypical squamous cells of undetermined significance. Am J Obstet Gynecol
2003; 188: 1383–92.
46. Gage JC, Katki HA, Schiffman M, et al. The low risk of precancer after a screening result of human papillomavirus-negative/atypical squamous cells of undetermined significance papanicolaou and implications for clinical management. Cancer Cytopathol
2014; 122: 842–50.
47. Katki HA, Schiffman M, Castle PE, et al. Five-year risks of CIN 3+ and cervical cancer among women with HPV-positive and HPV-negative high-grade Pap results. J Lower Gen Tract Dis
2013; 17 (5 Suppl 1): S50–5.
48. Ronco G, Dillner J, Elfstrom KM, et al. Efficacy of HPV-based screening for prevention of invasive cervical cancer: follow-up of four European randomised controlled trials. Lancet
2014; 383: 524–32.
49. Sankaranarayanan R, Nene BM, Shastri SS, et al. HPV screening for cervical cancer in rural India. N Engl J Med
2009; 360: 1385–94.
50. Katki HA, Schiffman M, Castle PE, et al. Five-year risks of CIN 3+ and cervical cancer among women with HPV testing of ASC-US Pap results. J Lower Gen Tract Dis
2013; 17 (5 Suppl 1): S36–42.
51. Schiffman M. Integration of human papillomavirus vaccination, cytology, and human papillomavirus testing. Cancer
2007; 111: 145–53.
52. Franco EL, Cuzick J. Cervical cancer screening following prophylactic human papillomavirus vaccination. Vaccine
2008; 26 (Suppl 1): A16–23.
53. Cuzick J, Cadman L, Mesher D, et al. Comparing the performance of six human papillomavirus tests in a screening population. Br J Cancer
2013; 108: 908–13.
54. Szarewski A, Ambroisine L, Cadman L, et al. Comparison of predictors for high-grade cervical intraepithelial neoplasia in women with abnormal smears. Cancer Epidemiol Biomarkers Prev
2008; 17: 3033–42.
55. Stoler MH, Ronnett BM, Joste NE, et al. The Interpretive Variability of Cervical Biopsies and Its Relationship to HPV Status. Am J Surg Pathol
2015; 39: 729–36.
56. Katki HA, Schiffman M, Castle PE, et al. Five-year risks of CIN 3+ and cervical cancer among women who test Pap-negative but are HPV-positive. J Lower Gen Tract Dis
2013; 17 (5 Suppl 1): S56–63.
57. Schiffman M, Vaughan LM, Raine-Bennett TR, et al. A study of HPV typing for the management of HPV-positive ASC-US cervical cytologic results. Gynecol Oncol
2015; 138: 573–8.
58. Reid JL, Wright TC Jr, Stoler MH, et al. Human papillomavirus oncogenic mRNA testing for cervical cancer screening: baseline and longitudinal results from the CLEAR study. Am J Clin Pathol
2015; 144: 473–83.
59. Schiffman M, Wentzensen N. Transitioning to a new era in cervical cancer screening. Gynecol Oncol
2015; 136: 175–7.
60. Katki HA, Schiffman M, Castle PE, et al. Benchmarking CIN 3+ risk as the basis for incorporating HPV and Pap cotesting into cervical screening and management guidelines. J Lower Gen Tract Dis
2013; 17 (5 Suppl 1): S28–35.
61. Franco EL, Tsu V, Herrero R, et al. Integration of human papillomavirus vaccination and cervical cancer screening in Latin America and the Caribbean. Vaccine
2008; 26 (Suppl 11): L88–95.
62. Rodriguez AC, Solomon D, Herrero R, et al. Impact of human papillomavirus vaccination on cervical cytology screening, colposcopy, and treatment. Am J Epidemiol
2013; 178: 752–60.
63. Katki HA, Gage JC, Schiffman M, et al. Follow-up testing after colposcopy: five-year risk of CIN 2+ after a colposcopic diagnosis of CIN 1 or less. J Lower Gen Tract Dis
2013; 17 (5 Suppl 1): S69–77.