Cervical Cancer Screening in the 21st Century : Is it Time to Retire the PAP Smear ? Clinical Obstetrics and Gynecology Vol 50/ Number 2/ June 2007 Abstract Introduction Limitations of Cytology Performance of Cytology-based Prevention Programs Human Papilloma Virus DNA Testing Using HPV DNA Testing as an Adjunct to Cervical Cytology Alterative Strategies for Using HPV DNA Testing for Screening Summary Abstract cytology-based cervical cancer prevention programs limited sensitivity expensive to maintain → programs based on testing for high-risk types of human papillomavirus (HPV) HPV testing more sensitive than cytology (either conventional or liquid-based) >30 years old : specificity ↓(slightly) Abstract Combination of cervical cytology and HPV testing to screen : cytology provides little benefit over using HPV testing alone to screen → HPV testing alone to screen → cervical cytology : a way to determine which HPV-positive women require additional follow-up or colposcopy Introduction Cervical cytology the cornerstone of cervical cancer prevention programs highly effective in many countries considered by some cancer control experts to be the single most successful approach to the prevention of any cancer the limitations inherent in cytology-based screening →cytology-based screening programs have had even less of an impact Limitations of Cytology Relatively poor sensitivity the sensitivity of a single Papanicolaou test : 50% for identifying women with CIN 2,3 or cervical cancer (CIN 2+) Meta-analyses that have reviewed the performance of cervical cytology sensitivity : 30% ~ 87% specificity : 86% ~ 100% meta-analysis conducted several years ago reviewed 94 screening studies overall sensitivity of cytology for CIN 2+ : 53% (95% CI 49%-57%) a review of recent European and North American cervical cancer screening studies by Cuzick et al 19% in a German study 77% in a recent large British study Limitations of Cytology liquid-based cytology conventional cytology for detecting significantly more sensitive CIN 2+ Systematic no evidence that liquid-based review cytology improves the sensitivity - publications of cytology through 2003- Positive significantly reduced with a relative predictive PPV of 0.61 (95% CI 0.38-0.97) value (PPV) compared with conventional cytology ASCUS cutoff, 71% (95% CI 58%-81%) 84% (71%-92%). the sensitivity ASCUS cutoff, 9.4% (95% CI 7-12.3) 11.4% (95% CI 8.5-15.0). PPV Limitations of Cytology liquid-based cytology vs conventional cytology No significant differences of sensitivity or PPV for the identification of women with CIN 2+ Limitations of Cytology novel cervical cancer screening strategies alternative, non–cytology-based screening methods Performance of cytology-based Prevention Programs The goal of a cervical cancer prevention programs to identify high-grade cervical cancer precursors (CIN 2,3) to allow the lesions to be treated to prevent the subsequent development of cervical cancer Performance of cytology-based Prevention Programs The effectiveness of cervical cancer prevention programs : depends on several factors + the sensitivity of the screening test Other factors that are important include ; (1) whether or not there is heterogeneity among CIN 2,3 lesions (2) the time it takes for a CIN 2,3 to progress to an invasive cancer (ie, the transit time) (3) the frequency at which screening occurs (4) screening coverage (ie, proportion of the target population that is actually screened) Performance of cytology-based Prevention Programs Long transit time the time for a CIN 2,3 → an invasive cancer ☞ average 10 years the introduction of cervical cytology →reductions in cervical cancer incidence : highly variable between countries ( reduction range 20% ~ 90%) - Epidemiologic studies from Canada, Scandinavia, and the United Kingdom 1970s ~ 1980s : minimal impact on cervical cancer incidence FIGURE 1. Impact of the introduction of organized screening on the incidence of cervical cancer in the United Kingdom. Modified from BMJ. 1999;318:904–908. In 1988, the National Health Service introduced a comprehensive cytology-based screening program that incorporated a comprehensive system of audits. The audits included the clinicians obtaining the cytology specimens laboratories evaluating the specimens specialists treating women with precursor lesions Performance of cytology-based Prevention Programs “call-recall” system every woman in England of the targeted ages receives a regular invitation to be screened screening coverage (1988 ~1995) : 40% → 85% (↑) → incidence of invasive cervical cancer : dropped by 40% death rate from cervical cancer : reduced by almost 50%. (by 2004) Performance of cytology-based Prevention Programs Compensating for the poor sensitivity of cytology → frequent screening In U.S. most women are overscreened some women do not undergo regular screening 10,000 women who develop cervical cancer each year in the U. S. - recent screening (X) : half of them - recent screening (O) : half of them Human Papilloma Virus DNA Testing Human Papilloma Virus DNA Testing invasive cervical cancer : 15 “high-risk” or “oncogenic” types of human papilloma virus (HPV) molecular test for detecting the DNA of high-risk types of HPV in clinical specimens is a solution hybridization test called Hybrid Capture 2 (hc2) (Digene Diagnostics, Gaithersburg, MD) overcome the limitations of cervical cytology for screening lower cost 13 high-risk types of HPV : 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68 Human Papilloma Virus DNA Testing TABLE 1. Performance of Hybrid Capture 2 and Cervical Cytology for Screening in Women 30 Years of Age and Older in Cross-sectional Studies*The target age for the study was 30 years and older for women and 97% of the women enrolled were in the target age.Modified from references J Natl Cancer Inst. 2006;98:765–774; Lancet. 2003;362:1871–1876; Br J Cancer. 2003;88:1570–1577; and Br J Cancer. 2005;93:575–581. presents the results from recent large European trials. Human Papilloma Virus DNA Testing Sensitivity HPV DNA testing > cervical cytology for identifying women with CIN 2,3 or cancer (CIN 2+) negative predictive value : cytology(-) & high-risk HPV DNA (-) →extremely high (> 99.9% in most of the studies) → risk of having an undetected CIN 2+ lesion in women who are screened with both tests : very low (approximately 1 in 1000) Human Papilloma Virus DNA Testing late adolescent females and young women : multiple partners within several years of initiating sexual activity → prevalence of high-risk HPV DNA positivity ↑↑ → multiple sequential infections with different types of HPV Ω most of these infections : transient women 30 years and older older they tend, fewer new sexual exposures resulting in fewer new HPV infections HPV prevalence ↓ Human Papilloma Virus DNA Testing Specificity : cervical cytology > HPV DNA testing > 2 tests in combination women ≥ 30 years old, who are screened using both cytology and high-risk HPV DNA testing → an abnormality on one or the other test → require either workup or additional follow-up --------------------------------------------------------10% Using HPV DNA Testing as an Adjunct to Cervical Cytology Using HPV DNA Testing as an Adjunct to Cervical Cytology FIGURE 2. Management of women using a combination of cervical cytology and HPV DNA testing for primary screening in women 30 years and older. Modified from Obstet Gynecol. 2004;103:304–309. Using HPV DNA Testing as an Adjunct to Cervical Cytology both high-risk HPV DNA and cytology negative → recommended that not require rescreening for 3 years supported by the results of 2 studies Study NCI study followed women enrolled study from France from Kaiser Permanente Portland, OR F/U up to 10 years for 36 months result high-risk HPV DNA (-) cervical cytology (-) → the risk of developing high-risk HPV DNA biopsy-confirmed CIN 3 →biopsy-confirmed CIN 2+ : low : 0.08% Using HPV DNA Testing as an Adjunct to Cervical Cytology How to manage HPV DNA (+), cytology (-) → reflect these women's risk for having, or developing, CIN2+ HPV DNA (+) & Cytology (-) In the French study study from England CIN 2+ risk 4.2% 2.8% HPV DNA (+), cytology (-) Lower risk for having CIN 2+ than are women with ASC-US ( 5~17%) Using HPV DNA Testing as an Adjunct to Cervical Cytology HPV DNA(+), cytology(-) → colposcopy not be performed the 2004 Interim Guidance recommends → retested in 6 to 12 months using both cervical cytology and HPV DNA testing (Fig. 2) In the French & English studies in the Kaiser Permanente Northern California HPV DNA(+) half of women were persistently only 35% were persistent at 12 at 6 months months persistent HPV infections or cytology ≥ low-grade squamous intraepithelial lesions → colposcopy Using HPV DNA Testing as an Adjunct to Cervical Cytology transiently shed HPV DNA → very low risk for having CIN 2+ both repeat tests : cytology(-), HPV DNA (-) → 3 yearly screening Alterative Strategies for Using HPV DNA Testing for Screening HPV testing as an adjunct to cytology : long-standing problems - large numbers of cytotechnicians - the variability in abnormal rate between laboratories - relatively slow turnaround times To maximize the benefits of incorporating HPV testing into screening → identify as many women with CIN 2+ as possible at the initial visit, while limiting the number referred to colposcopy Alterative Strategies for Using HPV DNA Testing for Screening Goal of cervical cancer screening • to reduce the risk of cervical cancer • not eliminate the level of risk reduction the costs potential inconvenience or harm to those being screened Alterative Strategies for Using HPV DNA Testing for Screening common misperception : receive regular screening → completely protected against cervical cancer → develop cancer → their clinician or the cytology laboratory, has failed them → frequent litigation for a failure to diagnose cervical cancer compensatory overscreening by clinicians to try and assure that none of their patients develop cervical cancer → increases the costs of screening → unnecessary interventions such as colposcopy Alterative Strategies for Using HPV DNA Testing for Screening sensitivity of HPV DNA testing : high HPV DNA testing Cytology with Cytology-based screening with cytology adjunctive HPV testing → HPV-based strategy CIN 2+ cytology : no cases 1 ~ 3 cases /10,000 12 ~ 30 additional cases dectection of CIN 2+ that were compared with using / 10,000 missed by HPV HPV testing alone testing In 2 South African In other European screening screening studies studies utilizing HPV DNA testing alone as the primary screening test Alterative Strategies for Using HPV DNA Testing for Screening several potential approaches that can be used for managing HPV DNA positive women : Novel HPV-based screening strategies incorporating either “reflex” cytology HPV genotyping, or a combination of both → a specimen would be obtained from all women using a liquid collection media suitable for both HPV DNA testing and cytology → initially be tested for high-risk types of HPV → HPV DNA (+) processed for cytology or perform HPV genotyping FIGURE 4. Novel HPV-based screening strategies incorporating either “reflex” cytology HPV genotyping, or a combination of both. Alterative Strategies for Using HPV DNA Testing for Screening most of the high-risk HPV DNA positive → developed biopsy-confirmed CIN 3 : HPV 16 or HPV18 In the Kaiser Portland, OR, follow-up study other high-risk types of HPV : slightly increased risk high-risk types of HPV such as HPV 16, 18, 33, 45, or 31 (the 5 most common HPV types found in cervical cancers) → to be referred to colposcopy other high-risk types of HPV → followed-up in 12 months with a repeat HPV DNA test FIGURE 3. Impact of HPV status on the development of CIN 3 in women 30 years and older enrolled from Kaiser, Portland, OR. Solid circles represent women who were HPV 16 positive entry, open circles are those who were HPV 18 positive, solid triangles are women with other high-risk types of HPV, and open triangles are women who were high-risk HPV DNA negative. Modified from J Natl Cancer Inst. 2005;97:1072–1079. HPV16(+) HPV18(+) Other high risk types of HPV High-risk DNA (-) Alterative Strategies for Using HPV DNA Testing for Screening Alternative strategy : screen using HPV DNA testing alone → then use a combination of both reflex cytology and HPV genotyping to identify which HPV DNA positive women need colposcopy - even more promising Summary novel HPV-based screening → overcome many of the limitations of cervical cytology → will maximize the number of women with CIN 2+ identified at the screening visit → make loss to follow-up less of a concern HPV-based screening strategies : sensitivity↑ → safely extend screening intervals from 3 to 5 years → less expensive to the health care system → more convenient for patients Thank you for your attention ! Summary Alterative Strategies for Using HPV DNA Testing for Screening Several studies have now clearly demonstrated that women infected with certain high-risk types of HPV are at much greater risk for developing CIN 2+ than are women infected with other high-risk types of HPV. Alterative Strategies for Using HPV DNA Testing for Screening The problem with eliminating cytology is that when both cytology and HPV DNA testing are utilized, the cytology result allows HPV DNA positive women to be triaged into 2 groups, those requiring colposcopy and those who can simply be followed-up. This triage function is needed since 5% to 8% of women 30 years and older will be found to be high-risk HPV DNA positive (Table 1) Over the next decade, the extraordinary advances that have been made in our understanding of the pathogenesis of cervical cancer are going to allow us to develop novel HPV-based screening strategies that will be able to overcome many of the limitations of cervical cytology. These strategies will maximize the number of women with CIN 2+ identified at the screening visit and make loss to follow-up less of a concern. Because of the increased sensitivity of these HPV-based screening strategies, we will be able to safely extend screening intervals from 3 to 5 years. This will be less expensive to the health care system and more convenient for patients. Already large randomized screening trials are underway in Europe to evaluate whether we can replace cervical cytology with HPV DNA testing. Hopefully such trials will begin in the United States in the near future. Limitations of Cytology When liquid-based cytology was first introduced, a number of studies found that liquid-based cytology was significantly more sensitive than conventional cervical cytology for detecting CIN 2+. These studies had 1 of 2 designs. With 1 design, clinicians collect both a conventional cytology and a liquid-based cytology from the same patient at the same time. This allows a direct comparison of the performance of the 2 types of cytology. The other study design compares the results obtained using liquid- based cytology with results obtained by the same laboratory using conventional cytology before the introduction of liquid-based cytology (ie, historical controls). Unfortunately, in the majority of the studies the end point was a cytologic one (ie, detection of low-grade squamous intraepithelial lesions or high-grade squamous intraepithelial lesions) rather than a histologic one (ie, detection of biopsy-confirmed CIN 2,3). Moreover, both types of studies can be easily biased. Because of these limitations, many in the screening community have remained skeptical that liquid-based cytology is truly more sensitive than conventional cytology. Limitations of Cytology A recent systematic review of publications through 2003 that compared the performance of liquid-based cytology with that of conventional cytology concluded that only 4 of 56 published studies provided sufficient verified data to allow any estimates of sensitivity and specificity and comparisons of test performance to be made. The systematic review concluded that there is no evidence in high-quality studies that liquid-based cytology improves the sensitivity of cytology. The review also concluded that randomized clinical trials were needed. Since the systematic review was completed, several randomized trials have been published comparing the performance of the 2 cytology methods. One large randomized study of Italian women 25 to 34 years of age reported that the sensitivity of liquid-based cytology was not significantly higher than that of conventional cytology. The relative sensitivity of liquid-based cytology for CIN 2+ using an atypical squamous cells of undetermined significance cutoff was 1.24 (95% CI 0.75-2.03) compared with that of conventional cytology. Limitations of Cytology However, the positive predictive value (PPV) of liquid-based cytology was significantly reduced with a relative PPV of 0.61 (95% CI 0.38-0.97) compared with conventional cytology. Another publication from the same group focused on women 35 years of age and older and found similar results. The relative sensitivity of liquid-based cytology in the older population was 1.06 (95% CI 0.72-1.55) compared with conventional cytology. In this study, the PPV of liquid-based cytology was also significantly reduced (relative PPV=0.58, 95% CI 0.33-0.98) compared with conventional cytology. In a recent randomized study from South Africa in which all women underwent colposcopy at the time of screening, we found no significant differences in either the sensitivity or PPV of liquid-based cytology compared with conventional cytology for the identification of women with CIN 2+. At an atypical squamous cells of undetermined significance cutoff, the sensitivity of liquid-based cytology was 71% (95% CI 58%-81%), whereas that of conventional cytology was 84% (71%-92%). The PPV of liquid-based cytology was 9.4% (95% CI 7-12.3) whereas that for conventional cytology was 11.4% (95% CI 8.5-15.0). Performance of cytology-based Prevention Programs In Norway after the introduction of opportunistic (ie, nonorganized) screening, the incidence of cervical cancer actually increased through the mid-1970s and only subsequently began a somewhat slow decline. In large part, the minimal impact was attributable to the fact that relatively few women were being screened (ie, there was low coverage) and the screening frequency was too infrequent. However, in 1995 Norway introduced an organized cytology-based cervical cancer screening program and since its introduction the rates of invasive cervical cancer have dropped considerably. Human Papilloma Virus DNA Testing Recently, Kaiser Permanente of Northern California began offering HPV DNA testing to all women 30 years of age and older. On the basis of the data from over 200,000 women, the overall prevalence of high-risk HPV DNA positivity was just 6.4%. For comparison, 7.8% of the women had some degree of cytological abnormality. 14 In Kaiser, a total of 11.6% of the women had either a cytologic abnormality or were high-risk HPV DNA positive.