- Email: [email protected]
Screening history in women with cervical cancer in a Danish population-based screening program
Gynecologic Oncology 120 (2011) 68–72
Contents lists available at ScienceDirect
Gynecologic Oncology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y g y n o
Screening history in women with cervical cancer in a Danish population-based screening program Benny Kirschner a,⁎, Susanne Poll a, Carsten Rygaard b, Anne Wåhlin c, Jette Junge b a b c
Department of Gynecology and Obstetrics, Herlev University Hospital, Herlev Ringvej 75, 2730 Herlev, Denmark Department of Pathology, Hvidovre University Hospital, Kettegård Alle 30, 2650 Hvidovre, Denmark Department of Pathology, Hillerød University Hospital, Helsevej 2, 3400 Hillerød, Denmark
a r t i c l e
i n f o
Article history: Received 3 August 2010 Available online 28 October 2010 Keywords: Cervical cytology Cervical screening cervical cancer
a b s t r a c t Objective. The aim of this study was to explore the screening histories of all cervical cancers in a Danish screening population. The intention was to decide suboptimal sides of the screening program and to evaluate the significance of routine screening in the development of cervical cancer. Methods. The study describes the results of a quality control audit, performed on all new cervical cancer cases diagnosed in the years 2008–2009 at two major Danish screening-centers. All relevant cytological and histological cervical samples were reviewed. Results. 202.534 cytological samples were evaluated in the study period, while 112 women were diagnosed with cervical cancer. The histological diagnoses comprised: 62 (55.4%) squamous cell carcinomas, 20 (17.9%) microinvasive squamous cell carcinomas, 25 (22.3%) adenocarcinomas and 5 cancers of different histology. The mean age of study subjects was 46.6 years. 51 (45.5%) women had deficient screening histories, while 45 (40.2%) women had followed the screening recommendations and had normal cervical samples in review. 11 (9.8%) women were diagnosed with false negative cytology, 2 women had false negative histological tests, while pathological review was not feasible for 3 subjects. Conclusions. More than 45% of the cervical cancer cases in our study were due to deficient cervical screening, stressing the importance of increasing the screening-uptake and coverage. 40% interval cancers emphasize the relevance of further cervical testing of women with relevant symptoms, despite of prior normal cervical samples. Finally, 9.8% false negative cytological samples are consistent with previous reports, but still a part of the screening program that should be improved. © 2010 Elsevier Inc. All rights reserved.
Introduction Cervical cytological screening, as developed by George Papanicolau in the 1940s [1], has been described as one of the most successful screening tools for cancer disease in the history of medicine [2]. Several investigators have shown that population based cytological screening has been responsible for a significant decrease in the incidence of cervical cancer [3]. The original cervical smear technique has now been improved with new cytological sample instruments and liquid based screening technology. In addition automated systems have been introduced to make the screening procedures more efficient. In spite of this, it is widely agreed that all cervical cytological screening methods have serious limitations, in particular there are reports of low testsensitivity [4]. Cuzick et al. report test sensitivity as low as 53% [5]. Gay et al. report a false negative rate of 20% with conventional smear ⁎ Corresponding author. E-mail addresses: [email protected] (B. Kirschner), [email protected] (S. Poll), [email protected] (C. Rygaard), [email protected] (A. Wåhlin), [email protected] (J. Junge). 0090-8258/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2010.09.021
technique, and in concurrence with other investigators, find preparation and sampling errors to account for the greater part of these false negative samples [6]. As in other Scandinavian countries, a population-based screening program was introduced in Denmark in the 1960s [7]. The Danish National Board of Health recommended that all women between 23 and 59 years of age should be offered a cervical cytological test free of charge every third year. The program was revised in the year 2007 and cytological testing is now recommended every third year to women between 23 and 50 years of age and hereafter every fifth year until the age of 65 [8]. Furthermore, until 2007 the regional Danish health providers were only recommended to offer cervical screening to its female population. Now this is a mandatory part of the national health program. In spite of this, screening-uptake and coverage in the study period of 2008 and 2009 in the region of Greater Copenhagen, Denmark was only 56% and 76%, respectively. This is consistent with reports from neighboring countries with similar populations [9,10]. It has been reported that close to 50% of women with cervical cancer diagnosis have adequate screening histories within 5 years of disease detection [11]. The aim of this study was to compare this report with data from a Danish population-based screening program,
B. Kirschner et al. / Gynecologic Oncology 120 (2011) 68–72
and to evaluate the significance of routine cervical cytological screening on the development of cervical malignancy.
69
3; 2.7% Regular screening history
Materials and methods With the intention of testing quality and accuracy of the population-based cervical cytological screening program, The Regional Steering Committee for Cervical Cancer Screening in Greater Copenhagen, Denmark, implemented a pilot project with routine audit of all newly diagnosed cervical cancer cases in the year 2008 [8]. This study describes the results of the audit from the years 2008 and 2009 from two screening-centers in greater Copenhagen; the Department of Pathology at Hvidovre University Hospital and the Department of Pathology at Hillerød University Hospital. The two centers provide cytopathological service for a population of approximately 960.000 inhabitants and approximately 323.000 women between 23 and 65 years of age, and evaluate close to 100.000 cytological and 15.000 histological cervical specimens each year. Audit protocol The pilot project includes re-evaluation and investigation of all earlier normal cytological and/or normal histological samples within 5.5 years of disease detection. Furthermore, the relevant physicians are informed of audit results. In the future, additional review of patients charts and final reporting to a central body is planned. All relevant cytological samples are re-evaluated both by the individual who made the original diagnosis (cyto-technician or cytopathologist) and by an independent cyto-pathologist. In case of discrepancy between the two re-evaluations, an additional independent cyto-pathologist is consulted and final diagnosis is determined by majority decision. If this cannot be established, a consensus diagnosis is found by joint microscopy and conference-decision. All relevant histological samples are re-evaluated by the individual who made the original diagnosis (always a pathologist) and an independent pathologist. In case of discrepancy between the two evaluations, an additional independent pathologist is consulted. A final diagnosis is decided by majority decision. If this cannot be established, a consensus diagnosis is found by joint microscopy and conference-decision. The relevant gynecologist is informed regarding final audit-results in each specific case. The audit aims at categorizing the cancer cases in the following groups: ‘Deficient screening history’, ‘False negative cytology’, ‘False negative histology’, ‘Regular screening history’ and ‘Review not possible’. ‘Deficient screening history’ is defined as no cervical cytological screening performed within 3.5 years of cancer diagnosis for women up to 50 years of age, within 5.5 years of diagnosis for women between 50 and 65 years of age and no screening test performed between 55 and 65 years of age for women over the recommended screening age. ‘False negative cytology’ is defined as “normal” cytological sample prior to cancer diagnosis with post-audit diagnosis of High-grade Squamous Intraepithelial Lesion (HSIL) or worse, while ‘False negative histology’ is defined as “normal” histological sample prior to cancer diagnosis with post-audit diagnosis of Cervical Intraepithelial Neoplasia grade 2 or worse (CIN2+). ‘Regular screening history’ is defined as having followed the screening recommendations as decided by the Danish National Board of Health, meaning that there were a registered cytological sample within 3.5 years of cancer diagnosis for women up to 50 years of age, within 5.5 years of diagnosis for women between 50 and 65 years of age and at least one cytological test performed between 55 and 65 years of age for women over the recommended screening age. Furthermore, women in this category had of course no false negative cytological or histological samples found in the audit process. In Denmark, most cervical cytological samples are collected by general practitioners, while only a few are obtained by gynecologists. Cervical histological specimens are typically collected by gynecologists, either in a hospital setting or in private practice.
False negative cytology
45; 40.2%
False negative histology Deficient screening history Audit not possible
51; 45.5% 11; 9.8% 2; 1.8%
Fig. 1. Distribution of screening histories in 112 cases of cervical cancer, absolute numbers and percentage.
For this study, the authors have undertaken a retrospective analysis of audit data of all new cases of invasive cervical cancers diagnosed in 2008 and 2009 in two Danish screening-centers. Both centers are of considerable size, and in the 2 year of the study-period the centers received 202.534 cervical cytological samples and 27.146 histological specimens. Both departments use the Cervex-Brush® system (Rovers Medical Devices, B.V., NL-5347 KV Oss, The Netherlands) for cervical cytological sampling and SurePath® liquid based cytology and SlideWizard computer screening system for cervical cytological processing (BD-TriPath Imaging®, Inc., Burlington, NC, USA). In addition, both departments use automated cytological screening systems (FocalPoint® automated analyzer, BD-TriPath Imaging®, Inc., Burlington, NC, USA), which has been shown to improve both screening sensitivity and productivity [12]. From a nationwide computerized pathology register (The Danish National Pathology Data bank) that contains diagnosis of all pathological material evaluated in Denmark, information on all relevant cervical histology- and cytology-material could be retrieved. In principle, all pathology laboratories in Denmark report to this computerized system, which registers cervical tests taken in all settings: primary cytological samples within the screening-program, opportunistic screening and control specimens. The study includes all diagnosed cervical cancers in the region, and so includes cases that have been detected both as a result of population-based screening, opportunistic testing and as a result of symptomatic disease. Statistical analysis with comparisons of categorical variables including screening history and histological diagnosis were performed by means of Fishers exact test. To adjust for age differences in the groups Mantel–Haenzels test was used. Only results with p b 0.05 were considered statistically significant. Results A total number of 112 women were diagnosed with an invasive cervical cancer in the 2-year period. The mean age of the women in the study was 46.6 ± 15.8 years (range 23–91 years. Screening history The screening history of the women in the study was defined as all cervical pathological samples registered 5.5 years prior to disease detection for women within the screening age,1 and as all cervical pathological samples registered up to 10 years prior to cessation of screening for women older than the recommended screening age.1 1 The screening program as recommended by The Danish National Board of Health was revised in the year 2007. The age-limit for screening was changed from 59 to 65 years of age.
70
B. Kirschner et al. / Gynecologic Oncology 120 (2011) 68–72
17.9% of all the invasive cancers in the study, and 24.4% of all squamous cell carcinomas in the study. Furthermore, 22.3% (25 cases) of the women were diagnosed with adenocarcinoma and 4.5% (5 cases) with a cancer of different histology (3 cases of adenosquamous carcinoma, 1 case of clear cell carcinoma and 1 case of neuroendocrine carcinoma). Histological diagnosis and screening history As mentioned earlier 45.5% (51 subjects) of the women in the study population had not attended regular cytological cervical sampling, meaning that they had a deficient screening history. 66.7% (34 out of 51 subjects) of these women were diagnosed with a squamous cell carcinoma (Fig. 2). 13.7% (7 out of 51 cases) of the women in this group were found to have microinvasive disease, 13.7% were diagnosed with adenocarcinoma and 3 women had disease of different histology (one neuroendocrine- and 2 adenosquamous carcinomas). Squamous cancers (SCC and MI-SCC) were diagnosed in 80.3% (41 out of 51 subjects) of women with deficient screening history. 9.8% of the study population (11 subjects) was diagnosed with a false negative cytological test in the audit procedures. Of these women 54.5% (6 out of 11 cases) had developed squamous cell carcinoma, 18.2% (2 out of 11 cases) were diagnosed with adenocarcinomas, 18.2% had microinvasive disease and one (9.1%) had a clear cell carcinoma. Squamous cancers (SCC and MI-SCC) were diagnosed in 72.7% (8 out of 11 subjects) of women with false negative cytological samples. Only 2 women were found to have false negative histological tests and both of these women were diagnosed with adenocarcinomas. 44.4% (20 out of 45 subjects) of the women, who were diagnosed with cervical cancer in spite of having a regular screening history, were diagnosed with squamous cell carcinomas. 28.9% (13 out of 45 cases) of the women in this group were diagnosed with adenocarcinoma, 11 women (24.4%) were diagnosed with microinvasive disease and only one woman (2.2%) was found to have an invasive cancer of different histology (adenosquamous carcinoma). Squamous cancers (SCC and MI-SCC) were diagnosed in 68.9% (31 out of 45 subjects) of women regular screening history.
Fig. 2. Distribution of histological diagnosis according to screening history in patients with cervical cancer in the Danish screening program, percentage.
45.5% (51 cases) of the women in the study were registered with a deficient screening history (Fig. 1). Due to control-systems in the screening program, it can be assumed that this was not due to suboptimal follow-up by relevant health-personnel. By audit 9.8% (11 subjects) of the 112 women diagnosed with a cervical malignancy were found to have false negative cytological samples in their screening history. Furthermore, 1.8% of the cases in the study, representing 2 women were found to have false negative histological specimen at histopathological review. All of the false negative cytological samples in the study were diagnosed with HSIL in audit procedures, while the two cases of false negative histological samples were both diagnosed with adenocarcinoma in situ (AIS). 40.2% (45 women) of the women in the study had a regular screening history and no registered false cytological or histological sample in the audit procedures. In 3 cases (2.7%), routine audit and cytological or histological review could not be performed as recommended by The Regional Steering Committee for Cervical Cancer in Greater Copenhagen. This was due to loss of relevant cytological or histological material, typically because of destruction of cytological glass slides. Histology
Cancer stage and screening history
The most common histological diagnosis in this study population was squamous cell carcinoma (SCC), which was found in 73.3% (82 cases) of the subjects. 20 of these tumors were classified as microinvasive (MI-SCC) (FIGO stages IA1 and IA2), representing
The clinical stage of cancers at time of diagnosis ranged from FIGO stage lA to stage IV [13] (Table 1).
Table 1 Analysis of cervical cancer cases by screening history. Total cancers
Age bands (years) 20–34 35–49 50–64 N65 Histological diagnosis SCC MI-SCC Adenocarcinoma Other FIGO stage IA IB II III IV Total
Deficient screening history
False negative testa
Regular screening history
Audit not feasible
n
%
n
%
n
%
n
%
27 51 16 18
24.1 45.5 14.3 16.1
11 18 6 16
21.6 35.3 11.8 31.4
3 8 2 –
23.1 61.5 15.4 –
10 26 6 3
22.2 57.8 13.3 6.7
2 – – 1
62 20 25 5
55.4 17.9 22.3 4.5
34 7 7 3
66.7 13.7 13.7 5.9
6 2 4 1
46.2 15.4 30.8 7.7
20 11 13 1
44.4 24.4 28.9 2.2
2 – 1
44 44 6 14 4 N = 112
39.3 39.3 5.4 12.5 3.6 (100%)
16 16 5 10 4 N = 51
31.4 31.4 9.8 19.6 7.9
3 9 – 1 – N = 13
23.1 69.2 – 7.7 –
23 18 1 3 – N = 45
51.1 40.0 2.2 6.7 –
2 1 – – – N=3
SCC; Squamous Cervical Cancer. MI-SCC; Microinvasive Squamous Cervical Cancer. a Includes cytological and histological false negative samples.
B. Kirschner et al. / Gynecologic Oncology 120 (2011) 68–72
62.8% (32 out of 51 subjects) of women with deficient screening history were diagnosed with stage I disease, 9.8% (5 cases) were found to have a stage II cancer, while 19.6% (10 cases) of the women in this group had stage III disease and 7.8% (4 cases) had stage IV cancer. Of the 11 women with a false negative cytological tests in the audit, 90.9% (10 cases) were diagnosed with stage I cancer, while one woman were found to have stage III disease. A total of 91.1% (41 out of 45 subjects) of women with true interval cancers were diagnosed with stage I cancer; 51.1% stage lA (23 cases) and 40.0% stage lB (18 cases). Only one woman in this group had stage II disease and 3 women (6.6%) had stage III cancer. Demographic details In all, 55.4% (62 out of 112 cases) of women were diagnosed with squamous cell carcinoma with a mean age of 46 ± 15.7 years (range 23–91 years). The mean age of the women diagnosed with adenocarcinomas was 44 ± 15.9 years (range 25–84 years) while the same data for women diagnosed with microinvasive squamous carcinoma was 45.0 ± 15.9 years (range 25–89 years. As mentioned earlier, there were 51 women in our study, who had not followed the national screening recommendations. The mean age of this group was 51.0 ± 18.6 years (range 23–91 years), while the mean age for the women with true interval cancers (regular screening history) was 43 ± 11.9 years (range 29–84 years). There were only 11 women with false negative cytological tests and this group had a mean age of 42.1 ± 8.1 years (range 29–59 years). Discussion Over 40% of all invasive tumors diagnosed in our two screeningcenters in the year 2008 and 2009, were found in women with deficient screening-histories. This is similar to earlier studies, and authors like Boulanger et al. report 23.7% of cervical cancer diagnosed in women with no prior screening-history and 43.1% of tumors found in women with only irregular screening [14]. According to Koss, one of the tragedies of cervical cytology is the high incidence of false negative smear reports issued by cytology laboratories [15]. In our study, 9.8% of subjects were found to have false negative cytological tests in the review process, while 1.8% of the women were diagnosed with false negative histology. This concurs with earlier reports, where authors report false negative rates of cytological screening programs ranging from 15% to 65% [16–22]. Some authors argue that the unchanged rate of false negative cytological tests proves the insufficiencies of new and often more expensive screening technologies. On the other hand, studies like our own indicate that these technologies improve screening efficiency and more importantly, decrease the number of false positive tests significantly [23]. Furthermore, with technologies like liquid based cytology, HPV-testing can be done directly from the cytology samples [24]. Investigators who have studied the characteristics of false negative liquid based cytological samples conclude that these typically contain few abnormal cells (b100 cells), few koilocytic cells and few hyperchromatic cells compared to true-positive samples [25]. In our study population, over 40.2% of the cancer-cases were true interval cancers meaning that the women had followed the national screening-recommendations and had no false negative cytological or histological samples discovered in the audit process. Earlier studies have reported highly varying sensitivity for a single cytological test, ranging from 30% to 87% [4]. The average duration of precancerous phase is thought to be 10–12 years before progression to invasive cancer [26]. Taking this into account, one must assume that the majority of the interval cancers in our study are due to sampling error, but rapidly developing neoplasia occurring after cytological sampling cannot be excluded.
71
It must be noticed that cervical cancer in women who have never been screened or who have a deficient screening history must be taken in relation to all the women in the background population who fall into this screening category and not as a proportion of all cervical cancers cases. Like all Danish women, the subjects in our study have received invitations to participate in the population-based screeningprogram for cervical cancer. Furthermore, our data indicates that women with cervical interval cancer typically are diagnosed with lower cancer stage than women with insufficient screening history. The difference is statistically significant, when all the women in the two groups are compared (p = 0.020), but due to insufficient data, not so when adjusting for age (p = 0.16). Similar results are registered, when exploring the difference in histological diagnoses in the two screening groups. There is a trend of adenocarcinomas being diagnosed more frequently in women with interval cancer than in women with deficient screening history, but data is too limited to be conclusive (all women p = 0.24; adjusting for age p = 0.38). The question of early cessation of cervical screening for women with consecutive negative cytological tests and no abnormality by age 50 has been much debated [27]. Authors have studied this issue by analyzing detection rates of preinvasive cervical lesions in these women and in general have registered lower detection rates than in a similarly screened younger population [28,29]. On the other side, there is strong evidence that cervical intraepithelial lesions have higher probability of progressing to invasive disease at older ages [30], and so lower detection rates is not necessarily evidence for lower screening efficiency. In our study, only 2 out of the 45 women who developed invasive cervical disease despite of regular screening history were over the recommended screening age in Denmark,1 and only a total of 6 women in this group were older than 50 years of age. Conclusively, over 45% of women diagnosed with invasive cervical cancers in the Copenhagen area in the year 2008–2009 had not followed the screening recommendations as stated by the Danish National Board of Health. This indicates that despite national guidelines, a free health care system and a relevant screening program, cervical cancer will still be a major health issue, if the problem of noncompliance with screening is not addressed. Consequently, greater focus must be made on increasing screening-uptake and coverage as to reach these women, and as a direct result of this data, measures are currently being undertaken in our institution to improve this part of the screening-program. 40% of the women in our study were found to have a true interval cancer. This emphasizes the importance of further testing of women with relevant symptoms like contact bleeding, despite of normal cervical cytology in earlier screening. This should also be a reminder to the physician that proper cervical sampling is of great significance in cervical screening. We believe that our study confirms the importance of routine cervical cytology and a need for a higher degree of screening participation. Additionally, the data will provide a useful baseline for later screening audits in Denmark. Conflict of interest statement The manuscript has not previously been published and I (Benny Kirschner) do confirm that there were no financial or other relationships that lead to conflict of interest. All authors declare that there is no conflict of interest. Furthermore I do confirm that the original manuscript and the revised paper have been read and approved by all the authors, that the requirements for authorship have been met and that the study represents honest work.
References [1] Papanicolau GN, Traut HF. The diagnostic value of vaginal smears in the carcinoma of the uterus. Am J Obstet Gynecol 1941(42):193–206. [2] Cramer DW. The role of cervical cytology in the declining morbidity and mortality of cervical cancer. Cancer Dec 1974;34(6):2018–27. [3] Vizcaino AP, Moreno V, Bosch FX, Munoz N, Barros-Dios XM, Borras J, et al. International trends in incidence of cervical cancer: II. Squamous-cell carcinoma. Int J Cancer May 1 2000;86(3):429–35.
72
B. Kirschner et al. / Gynecologic Oncology 120 (2011) 68–72
[4] Nanda K, McCrory DC, Myers ER, Bastian LA, Hasselblad V, Hickey JD, et al. Accuracy of the Papanicolaou test in screening for and follow-up of cervical cytologic abnormalities: a systematic review. Ann Intern Med May 16 2000;132(10):810–9. [5] Cuzick J, Clavel C, Petry KU, Meijer CJ, Hoyer H, Ratnam S, et al. Overview of the European and North American studies on HPV testing in primary cervical cancer screening. Int J Cancer Sep 1 2006;119(5):1095–101. [6] Gay JD, Donaldson LD, Goellner JR. False-negative results in cervical cytologic studies. Acta Cytol Nov 1985;29(6):1043–6. [7] Anttila A, Ronco G, Clifford G, Bray F, Hakama M, Arbyn M, et al. Cervical cancer screening programmes and policies in 18 European countries. Br J Cancer Aug 31 2004;91(5):935–41. [8] The Danish National Board of Health. Screening for cervical cancer. The Danish National Board of Health; Sep 7 2007. [9] Gakidou E, Nordhagen S, Obermeyer Z. Coverage of cervical cancer screening in 57 countries: low average levels and large inequalities. PLoS Med Jun 17 2008;5(6):e132. [10] Nygard JF, Skare GB, Thoresen SO. The cervical cancer screening programme in Norway, 1992–2000: changes in Pap smear coverage and incidence of cervical cancer. J Med Screen 2002;9(2):86–91. [11] Sasieni PD, Cuzick J, Lynch-Farmery E. Estimating the efficacy of screening by auditing smear histories of women with and without cervical cancer. The National Co-ordinating Network for Cervical Screening Working Group. Br J Cancer Apr 1996;73(8):1001–5. [12] Wilbur DC, Parker EM, Foti JA. Location-guided screening of liquid-based cervical cytology specimens: a potential improvement in accuracy and productivity is demonstrated in a preclinical feasibility trial. Am J Clin Pathol Sep 2002;118(3): 399–407. [13] Pecorelli S, Zigliani L, Odicino F. Revised FIGO staging for carcinoma of the cervix. Int J Gynaecol Obstet May 2009;105(2):107–8. [14] Boulanger JC, Fauvet R, Urrutiaguer S, Drean Y, Sevestre H, Ganry O, et al. Cytological history of cases of invasive cervical cancer diagnosed in France in 2006. Gynecol Obstet Fertil Sep 2007;35(9):764–71. [15] Koss LG. The Papanicolaou test for cervical cancer detection. A triumph and a tragedy. JAMA Feb 3 1989;261(5):737–43. [16] Andrae B, Kemetli L, Sparen P, Silfverdal L, Strander B, Ryd W, et al. Screeningpreventable cervical cancer risks: evidence from a nationwide audit in Sweden. J Natl Cancer Inst May 7 2008;100(9):622–9. [17] Bulk S, Rozendaal L, Zielinski GD, Berkhof J, Daalmeijer NC, Snijders PJ, et al. High-risk human papillomavirus is present in cytologically false-negative smears: an analysis of “normal” smears preceding CIN2/3. J Clin Pathol Mar 2008;61(3):385–9.
[18] Ejersbo D, Dahl MB, Holund B. False negative Pap smears in a Danish material. Ugeskr Laeger Jun 2 2003;165(23):2391–4. [19] Kenter GG, Schoonderwald EM, Koelma IA, Arentz N, Hermans J, Fleuren GJ. The cytological screening history of 469 patients with squamous cell carcinoma of the cervix uteri; does interval carcinoma exist? Acta Obstet Gynecol Scand Apr 1996;75(4):400–3. [20] Anderson GH, Benedet JL, Le Riche JC, Matisic JP, Thompson JE. Invasive cancer of the cervix in British Columbia: a review of the demography and screening histories of 437 cases seen from 1985–1988. Obstet Gynecol Jul 1992;80(1):1–4. [21] Kristensen GB, Skyggebjerg KD, Holund B, Holm K, Hansen MK. Analysis of cervical smears obtained within three years of the diagnosis of invasive cervical cancer. Acta Cytol Jan 1991;35(1):47–50. [22] van der GY, Vooijs GP, Gaillard HL, Go DM. Screening errors in cervical cytologic screening. Acta Cytol Jul 1987;31(4):434–8. [23] Kirschner B, Simonsen K, Junge J. Comparison of conventional Papanicolaou smear and SurePath liquid-based cytology in the Copenhagen population screening programme for cervical cancer. Cytopathology Aug 2006;17(4):187–94. [24] Ring M, Bolger N, O'Donnell M, Malkin A, Bermingham N, Akpan E, et al. Evaluation of liquid-based cytology in cervical screening of high-risk populations: a split study of colposcopy and genito-urinary medicine populations. Cytopathology Jun 2002;13(3):152–9. [25] Leung KM, Lam KK, Tse PY, Yeoh GP, Chan KW. Characteristics of false-negative ThinPrep cervical smears in women with high-grade squamous intraepithelial lesions. Hong Kong Med J Aug 2008;14(4):292–5. [26] Lonnberg S, Anttila A, Kotaniemi-Talonen L, Kujari H, Melkko J, Granroth G, et al. Low proportion of false-negative smears in the Finnish program for cervical cancer screening. Cancer Epidemiol Biomarkers Prev Feb 2010;19(2):381–7. [27] Rebolj M, van BM, Lynge E, Looman C, Essink-Bot ML, Boer R, et al. Incidence of cervical cancer after several negative smear results by age 50: prospective observational study. BMJ 2009;338:b1354. [28] Baay MF, Smits E, Tjalma WA, Lardon F, Weyler J, Van RP, et al. Can cervical cancer screening be stopped at 50? The prevalence of HPV in elderly women. Int J Cancer Jan 10 2004;108(2):258–61. [29] Flannelly G, Monaghan J, Cruickshank M, Duncan I, Johnson J, Jordan J, et al. Cervical screening in women over the age of 50: results of a population-based multicentre study. BJOG Apr 2004;111(4):362–8. [30] McCredie MR, Sharples KJ, Paul C, Baranyai J, Medley G, Jones RW, et al. Natural history of cervical neoplasia and risk of invasive cancer in women with cervical intraepithelial neoplasia 3: a retrospective cohort study. Lancet Oncol May 2008;9(5):425–34.
Contents lists available at ScienceDirect
Gynecologic Oncology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y g y n o
Screening history in women with cervical cancer in a Danish population-based screening program Benny Kirschner a,⁎, Susanne Poll a, Carsten Rygaard b, Anne Wåhlin c, Jette Junge b a b c
Department of Gynecology and Obstetrics, Herlev University Hospital, Herlev Ringvej 75, 2730 Herlev, Denmark Department of Pathology, Hvidovre University Hospital, Kettegård Alle 30, 2650 Hvidovre, Denmark Department of Pathology, Hillerød University Hospital, Helsevej 2, 3400 Hillerød, Denmark
a r t i c l e
i n f o
Article history: Received 3 August 2010 Available online 28 October 2010 Keywords: Cervical cytology Cervical screening cervical cancer
a b s t r a c t Objective. The aim of this study was to explore the screening histories of all cervical cancers in a Danish screening population. The intention was to decide suboptimal sides of the screening program and to evaluate the significance of routine screening in the development of cervical cancer. Methods. The study describes the results of a quality control audit, performed on all new cervical cancer cases diagnosed in the years 2008–2009 at two major Danish screening-centers. All relevant cytological and histological cervical samples were reviewed. Results. 202.534 cytological samples were evaluated in the study period, while 112 women were diagnosed with cervical cancer. The histological diagnoses comprised: 62 (55.4%) squamous cell carcinomas, 20 (17.9%) microinvasive squamous cell carcinomas, 25 (22.3%) adenocarcinomas and 5 cancers of different histology. The mean age of study subjects was 46.6 years. 51 (45.5%) women had deficient screening histories, while 45 (40.2%) women had followed the screening recommendations and had normal cervical samples in review. 11 (9.8%) women were diagnosed with false negative cytology, 2 women had false negative histological tests, while pathological review was not feasible for 3 subjects. Conclusions. More than 45% of the cervical cancer cases in our study were due to deficient cervical screening, stressing the importance of increasing the screening-uptake and coverage. 40% interval cancers emphasize the relevance of further cervical testing of women with relevant symptoms, despite of prior normal cervical samples. Finally, 9.8% false negative cytological samples are consistent with previous reports, but still a part of the screening program that should be improved. © 2010 Elsevier Inc. All rights reserved.
Introduction Cervical cytological screening, as developed by George Papanicolau in the 1940s [1], has been described as one of the most successful screening tools for cancer disease in the history of medicine [2]. Several investigators have shown that population based cytological screening has been responsible for a significant decrease in the incidence of cervical cancer [3]. The original cervical smear technique has now been improved with new cytological sample instruments and liquid based screening technology. In addition automated systems have been introduced to make the screening procedures more efficient. In spite of this, it is widely agreed that all cervical cytological screening methods have serious limitations, in particular there are reports of low testsensitivity [4]. Cuzick et al. report test sensitivity as low as 53% [5]. Gay et al. report a false negative rate of 20% with conventional smear ⁎ Corresponding author. E-mail addresses: [email protected] (B. Kirschner), [email protected] (S. Poll), [email protected] (C. Rygaard), [email protected] (A. Wåhlin), [email protected] (J. Junge). 0090-8258/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2010.09.021
technique, and in concurrence with other investigators, find preparation and sampling errors to account for the greater part of these false negative samples [6]. As in other Scandinavian countries, a population-based screening program was introduced in Denmark in the 1960s [7]. The Danish National Board of Health recommended that all women between 23 and 59 years of age should be offered a cervical cytological test free of charge every third year. The program was revised in the year 2007 and cytological testing is now recommended every third year to women between 23 and 50 years of age and hereafter every fifth year until the age of 65 [8]. Furthermore, until 2007 the regional Danish health providers were only recommended to offer cervical screening to its female population. Now this is a mandatory part of the national health program. In spite of this, screening-uptake and coverage in the study period of 2008 and 2009 in the region of Greater Copenhagen, Denmark was only 56% and 76%, respectively. This is consistent with reports from neighboring countries with similar populations [9,10]. It has been reported that close to 50% of women with cervical cancer diagnosis have adequate screening histories within 5 years of disease detection [11]. The aim of this study was to compare this report with data from a Danish population-based screening program,
B. Kirschner et al. / Gynecologic Oncology 120 (2011) 68–72
and to evaluate the significance of routine cervical cytological screening on the development of cervical malignancy.
69
3; 2.7% Regular screening history
Materials and methods With the intention of testing quality and accuracy of the population-based cervical cytological screening program, The Regional Steering Committee for Cervical Cancer Screening in Greater Copenhagen, Denmark, implemented a pilot project with routine audit of all newly diagnosed cervical cancer cases in the year 2008 [8]. This study describes the results of the audit from the years 2008 and 2009 from two screening-centers in greater Copenhagen; the Department of Pathology at Hvidovre University Hospital and the Department of Pathology at Hillerød University Hospital. The two centers provide cytopathological service for a population of approximately 960.000 inhabitants and approximately 323.000 women between 23 and 65 years of age, and evaluate close to 100.000 cytological and 15.000 histological cervical specimens each year. Audit protocol The pilot project includes re-evaluation and investigation of all earlier normal cytological and/or normal histological samples within 5.5 years of disease detection. Furthermore, the relevant physicians are informed of audit results. In the future, additional review of patients charts and final reporting to a central body is planned. All relevant cytological samples are re-evaluated both by the individual who made the original diagnosis (cyto-technician or cytopathologist) and by an independent cyto-pathologist. In case of discrepancy between the two re-evaluations, an additional independent cyto-pathologist is consulted and final diagnosis is determined by majority decision. If this cannot be established, a consensus diagnosis is found by joint microscopy and conference-decision. All relevant histological samples are re-evaluated by the individual who made the original diagnosis (always a pathologist) and an independent pathologist. In case of discrepancy between the two evaluations, an additional independent pathologist is consulted. A final diagnosis is decided by majority decision. If this cannot be established, a consensus diagnosis is found by joint microscopy and conference-decision. The relevant gynecologist is informed regarding final audit-results in each specific case. The audit aims at categorizing the cancer cases in the following groups: ‘Deficient screening history’, ‘False negative cytology’, ‘False negative histology’, ‘Regular screening history’ and ‘Review not possible’. ‘Deficient screening history’ is defined as no cervical cytological screening performed within 3.5 years of cancer diagnosis for women up to 50 years of age, within 5.5 years of diagnosis for women between 50 and 65 years of age and no screening test performed between 55 and 65 years of age for women over the recommended screening age. ‘False negative cytology’ is defined as “normal” cytological sample prior to cancer diagnosis with post-audit diagnosis of High-grade Squamous Intraepithelial Lesion (HSIL) or worse, while ‘False negative histology’ is defined as “normal” histological sample prior to cancer diagnosis with post-audit diagnosis of Cervical Intraepithelial Neoplasia grade 2 or worse (CIN2+). ‘Regular screening history’ is defined as having followed the screening recommendations as decided by the Danish National Board of Health, meaning that there were a registered cytological sample within 3.5 years of cancer diagnosis for women up to 50 years of age, within 5.5 years of diagnosis for women between 50 and 65 years of age and at least one cytological test performed between 55 and 65 years of age for women over the recommended screening age. Furthermore, women in this category had of course no false negative cytological or histological samples found in the audit process. In Denmark, most cervical cytological samples are collected by general practitioners, while only a few are obtained by gynecologists. Cervical histological specimens are typically collected by gynecologists, either in a hospital setting or in private practice.
False negative cytology
45; 40.2%
False negative histology Deficient screening history Audit not possible
51; 45.5% 11; 9.8% 2; 1.8%
Fig. 1. Distribution of screening histories in 112 cases of cervical cancer, absolute numbers and percentage.
For this study, the authors have undertaken a retrospective analysis of audit data of all new cases of invasive cervical cancers diagnosed in 2008 and 2009 in two Danish screening-centers. Both centers are of considerable size, and in the 2 year of the study-period the centers received 202.534 cervical cytological samples and 27.146 histological specimens. Both departments use the Cervex-Brush® system (Rovers Medical Devices, B.V., NL-5347 KV Oss, The Netherlands) for cervical cytological sampling and SurePath® liquid based cytology and SlideWizard computer screening system for cervical cytological processing (BD-TriPath Imaging®, Inc., Burlington, NC, USA). In addition, both departments use automated cytological screening systems (FocalPoint® automated analyzer, BD-TriPath Imaging®, Inc., Burlington, NC, USA), which has been shown to improve both screening sensitivity and productivity [12]. From a nationwide computerized pathology register (The Danish National Pathology Data bank) that contains diagnosis of all pathological material evaluated in Denmark, information on all relevant cervical histology- and cytology-material could be retrieved. In principle, all pathology laboratories in Denmark report to this computerized system, which registers cervical tests taken in all settings: primary cytological samples within the screening-program, opportunistic screening and control specimens. The study includes all diagnosed cervical cancers in the region, and so includes cases that have been detected both as a result of population-based screening, opportunistic testing and as a result of symptomatic disease. Statistical analysis with comparisons of categorical variables including screening history and histological diagnosis were performed by means of Fishers exact test. To adjust for age differences in the groups Mantel–Haenzels test was used. Only results with p b 0.05 were considered statistically significant. Results A total number of 112 women were diagnosed with an invasive cervical cancer in the 2-year period. The mean age of the women in the study was 46.6 ± 15.8 years (range 23–91 years. Screening history The screening history of the women in the study was defined as all cervical pathological samples registered 5.5 years prior to disease detection for women within the screening age,1 and as all cervical pathological samples registered up to 10 years prior to cessation of screening for women older than the recommended screening age.1 1 The screening program as recommended by The Danish National Board of Health was revised in the year 2007. The age-limit for screening was changed from 59 to 65 years of age.
70
B. Kirschner et al. / Gynecologic Oncology 120 (2011) 68–72
17.9% of all the invasive cancers in the study, and 24.4% of all squamous cell carcinomas in the study. Furthermore, 22.3% (25 cases) of the women were diagnosed with adenocarcinoma and 4.5% (5 cases) with a cancer of different histology (3 cases of adenosquamous carcinoma, 1 case of clear cell carcinoma and 1 case of neuroendocrine carcinoma). Histological diagnosis and screening history As mentioned earlier 45.5% (51 subjects) of the women in the study population had not attended regular cytological cervical sampling, meaning that they had a deficient screening history. 66.7% (34 out of 51 subjects) of these women were diagnosed with a squamous cell carcinoma (Fig. 2). 13.7% (7 out of 51 cases) of the women in this group were found to have microinvasive disease, 13.7% were diagnosed with adenocarcinoma and 3 women had disease of different histology (one neuroendocrine- and 2 adenosquamous carcinomas). Squamous cancers (SCC and MI-SCC) were diagnosed in 80.3% (41 out of 51 subjects) of women with deficient screening history. 9.8% of the study population (11 subjects) was diagnosed with a false negative cytological test in the audit procedures. Of these women 54.5% (6 out of 11 cases) had developed squamous cell carcinoma, 18.2% (2 out of 11 cases) were diagnosed with adenocarcinomas, 18.2% had microinvasive disease and one (9.1%) had a clear cell carcinoma. Squamous cancers (SCC and MI-SCC) were diagnosed in 72.7% (8 out of 11 subjects) of women with false negative cytological samples. Only 2 women were found to have false negative histological tests and both of these women were diagnosed with adenocarcinomas. 44.4% (20 out of 45 subjects) of the women, who were diagnosed with cervical cancer in spite of having a regular screening history, were diagnosed with squamous cell carcinomas. 28.9% (13 out of 45 cases) of the women in this group were diagnosed with adenocarcinoma, 11 women (24.4%) were diagnosed with microinvasive disease and only one woman (2.2%) was found to have an invasive cancer of different histology (adenosquamous carcinoma). Squamous cancers (SCC and MI-SCC) were diagnosed in 68.9% (31 out of 45 subjects) of women regular screening history.
Fig. 2. Distribution of histological diagnosis according to screening history in patients with cervical cancer in the Danish screening program, percentage.
45.5% (51 cases) of the women in the study were registered with a deficient screening history (Fig. 1). Due to control-systems in the screening program, it can be assumed that this was not due to suboptimal follow-up by relevant health-personnel. By audit 9.8% (11 subjects) of the 112 women diagnosed with a cervical malignancy were found to have false negative cytological samples in their screening history. Furthermore, 1.8% of the cases in the study, representing 2 women were found to have false negative histological specimen at histopathological review. All of the false negative cytological samples in the study were diagnosed with HSIL in audit procedures, while the two cases of false negative histological samples were both diagnosed with adenocarcinoma in situ (AIS). 40.2% (45 women) of the women in the study had a regular screening history and no registered false cytological or histological sample in the audit procedures. In 3 cases (2.7%), routine audit and cytological or histological review could not be performed as recommended by The Regional Steering Committee for Cervical Cancer in Greater Copenhagen. This was due to loss of relevant cytological or histological material, typically because of destruction of cytological glass slides. Histology
Cancer stage and screening history
The most common histological diagnosis in this study population was squamous cell carcinoma (SCC), which was found in 73.3% (82 cases) of the subjects. 20 of these tumors were classified as microinvasive (MI-SCC) (FIGO stages IA1 and IA2), representing
The clinical stage of cancers at time of diagnosis ranged from FIGO stage lA to stage IV [13] (Table 1).
Table 1 Analysis of cervical cancer cases by screening history. Total cancers
Age bands (years) 20–34 35–49 50–64 N65 Histological diagnosis SCC MI-SCC Adenocarcinoma Other FIGO stage IA IB II III IV Total
Deficient screening history
False negative testa
Regular screening history
Audit not feasible
n
%
n
%
n
%
n
%
27 51 16 18
24.1 45.5 14.3 16.1
11 18 6 16
21.6 35.3 11.8 31.4
3 8 2 –
23.1 61.5 15.4 –
10 26 6 3
22.2 57.8 13.3 6.7
2 – – 1
62 20 25 5
55.4 17.9 22.3 4.5
34 7 7 3
66.7 13.7 13.7 5.9
6 2 4 1
46.2 15.4 30.8 7.7
20 11 13 1
44.4 24.4 28.9 2.2
2 – 1
44 44 6 14 4 N = 112
39.3 39.3 5.4 12.5 3.6 (100%)
16 16 5 10 4 N = 51
31.4 31.4 9.8 19.6 7.9
3 9 – 1 – N = 13
23.1 69.2 – 7.7 –
23 18 1 3 – N = 45
51.1 40.0 2.2 6.7 –
2 1 – – – N=3
SCC; Squamous Cervical Cancer. MI-SCC; Microinvasive Squamous Cervical Cancer. a Includes cytological and histological false negative samples.
B. Kirschner et al. / Gynecologic Oncology 120 (2011) 68–72
62.8% (32 out of 51 subjects) of women with deficient screening history were diagnosed with stage I disease, 9.8% (5 cases) were found to have a stage II cancer, while 19.6% (10 cases) of the women in this group had stage III disease and 7.8% (4 cases) had stage IV cancer. Of the 11 women with a false negative cytological tests in the audit, 90.9% (10 cases) were diagnosed with stage I cancer, while one woman were found to have stage III disease. A total of 91.1% (41 out of 45 subjects) of women with true interval cancers were diagnosed with stage I cancer; 51.1% stage lA (23 cases) and 40.0% stage lB (18 cases). Only one woman in this group had stage II disease and 3 women (6.6%) had stage III cancer. Demographic details In all, 55.4% (62 out of 112 cases) of women were diagnosed with squamous cell carcinoma with a mean age of 46 ± 15.7 years (range 23–91 years). The mean age of the women diagnosed with adenocarcinomas was 44 ± 15.9 years (range 25–84 years) while the same data for women diagnosed with microinvasive squamous carcinoma was 45.0 ± 15.9 years (range 25–89 years. As mentioned earlier, there were 51 women in our study, who had not followed the national screening recommendations. The mean age of this group was 51.0 ± 18.6 years (range 23–91 years), while the mean age for the women with true interval cancers (regular screening history) was 43 ± 11.9 years (range 29–84 years). There were only 11 women with false negative cytological tests and this group had a mean age of 42.1 ± 8.1 years (range 29–59 years). Discussion Over 40% of all invasive tumors diagnosed in our two screeningcenters in the year 2008 and 2009, were found in women with deficient screening-histories. This is similar to earlier studies, and authors like Boulanger et al. report 23.7% of cervical cancer diagnosed in women with no prior screening-history and 43.1% of tumors found in women with only irregular screening [14]. According to Koss, one of the tragedies of cervical cytology is the high incidence of false negative smear reports issued by cytology laboratories [15]. In our study, 9.8% of subjects were found to have false negative cytological tests in the review process, while 1.8% of the women were diagnosed with false negative histology. This concurs with earlier reports, where authors report false negative rates of cytological screening programs ranging from 15% to 65% [16–22]. Some authors argue that the unchanged rate of false negative cytological tests proves the insufficiencies of new and often more expensive screening technologies. On the other hand, studies like our own indicate that these technologies improve screening efficiency and more importantly, decrease the number of false positive tests significantly [23]. Furthermore, with technologies like liquid based cytology, HPV-testing can be done directly from the cytology samples [24]. Investigators who have studied the characteristics of false negative liquid based cytological samples conclude that these typically contain few abnormal cells (b100 cells), few koilocytic cells and few hyperchromatic cells compared to true-positive samples [25]. In our study population, over 40.2% of the cancer-cases were true interval cancers meaning that the women had followed the national screening-recommendations and had no false negative cytological or histological samples discovered in the audit process. Earlier studies have reported highly varying sensitivity for a single cytological test, ranging from 30% to 87% [4]. The average duration of precancerous phase is thought to be 10–12 years before progression to invasive cancer [26]. Taking this into account, one must assume that the majority of the interval cancers in our study are due to sampling error, but rapidly developing neoplasia occurring after cytological sampling cannot be excluded.
71
It must be noticed that cervical cancer in women who have never been screened or who have a deficient screening history must be taken in relation to all the women in the background population who fall into this screening category and not as a proportion of all cervical cancers cases. Like all Danish women, the subjects in our study have received invitations to participate in the population-based screeningprogram for cervical cancer. Furthermore, our data indicates that women with cervical interval cancer typically are diagnosed with lower cancer stage than women with insufficient screening history. The difference is statistically significant, when all the women in the two groups are compared (p = 0.020), but due to insufficient data, not so when adjusting for age (p = 0.16). Similar results are registered, when exploring the difference in histological diagnoses in the two screening groups. There is a trend of adenocarcinomas being diagnosed more frequently in women with interval cancer than in women with deficient screening history, but data is too limited to be conclusive (all women p = 0.24; adjusting for age p = 0.38). The question of early cessation of cervical screening for women with consecutive negative cytological tests and no abnormality by age 50 has been much debated [27]. Authors have studied this issue by analyzing detection rates of preinvasive cervical lesions in these women and in general have registered lower detection rates than in a similarly screened younger population [28,29]. On the other side, there is strong evidence that cervical intraepithelial lesions have higher probability of progressing to invasive disease at older ages [30], and so lower detection rates is not necessarily evidence for lower screening efficiency. In our study, only 2 out of the 45 women who developed invasive cervical disease despite of regular screening history were over the recommended screening age in Denmark,1 and only a total of 6 women in this group were older than 50 years of age. Conclusively, over 45% of women diagnosed with invasive cervical cancers in the Copenhagen area in the year 2008–2009 had not followed the screening recommendations as stated by the Danish National Board of Health. This indicates that despite national guidelines, a free health care system and a relevant screening program, cervical cancer will still be a major health issue, if the problem of noncompliance with screening is not addressed. Consequently, greater focus must be made on increasing screening-uptake and coverage as to reach these women, and as a direct result of this data, measures are currently being undertaken in our institution to improve this part of the screening-program. 40% of the women in our study were found to have a true interval cancer. This emphasizes the importance of further testing of women with relevant symptoms like contact bleeding, despite of normal cervical cytology in earlier screening. This should also be a reminder to the physician that proper cervical sampling is of great significance in cervical screening. We believe that our study confirms the importance of routine cervical cytology and a need for a higher degree of screening participation. Additionally, the data will provide a useful baseline for later screening audits in Denmark. Conflict of interest statement The manuscript has not previously been published and I (Benny Kirschner) do confirm that there were no financial or other relationships that lead to conflict of interest. All authors declare that there is no conflict of interest. Furthermore I do confirm that the original manuscript and the revised paper have been read and approved by all the authors, that the requirements for authorship have been met and that the study represents honest work.
References [1] Papanicolau GN, Traut HF. The diagnostic value of vaginal smears in the carcinoma of the uterus. Am J Obstet Gynecol 1941(42):193–206. [2] Cramer DW. The role of cervical cytology in the declining morbidity and mortality of cervical cancer. Cancer Dec 1974;34(6):2018–27. [3] Vizcaino AP, Moreno V, Bosch FX, Munoz N, Barros-Dios XM, Borras J, et al. International trends in incidence of cervical cancer: II. Squamous-cell carcinoma. Int J Cancer May 1 2000;86(3):429–35.
72
B. Kirschner et al. / Gynecologic Oncology 120 (2011) 68–72
[4] Nanda K, McCrory DC, Myers ER, Bastian LA, Hasselblad V, Hickey JD, et al. Accuracy of the Papanicolaou test in screening for and follow-up of cervical cytologic abnormalities: a systematic review. Ann Intern Med May 16 2000;132(10):810–9. [5] Cuzick J, Clavel C, Petry KU, Meijer CJ, Hoyer H, Ratnam S, et al. Overview of the European and North American studies on HPV testing in primary cervical cancer screening. Int J Cancer Sep 1 2006;119(5):1095–101. [6] Gay JD, Donaldson LD, Goellner JR. False-negative results in cervical cytologic studies. Acta Cytol Nov 1985;29(6):1043–6. [7] Anttila A, Ronco G, Clifford G, Bray F, Hakama M, Arbyn M, et al. Cervical cancer screening programmes and policies in 18 European countries. Br J Cancer Aug 31 2004;91(5):935–41. [8] The Danish National Board of Health. Screening for cervical cancer. The Danish National Board of Health; Sep 7 2007. [9] Gakidou E, Nordhagen S, Obermeyer Z. Coverage of cervical cancer screening in 57 countries: low average levels and large inequalities. PLoS Med Jun 17 2008;5(6):e132. [10] Nygard JF, Skare GB, Thoresen SO. The cervical cancer screening programme in Norway, 1992–2000: changes in Pap smear coverage and incidence of cervical cancer. J Med Screen 2002;9(2):86–91. [11] Sasieni PD, Cuzick J, Lynch-Farmery E. Estimating the efficacy of screening by auditing smear histories of women with and without cervical cancer. The National Co-ordinating Network for Cervical Screening Working Group. Br J Cancer Apr 1996;73(8):1001–5. [12] Wilbur DC, Parker EM, Foti JA. Location-guided screening of liquid-based cervical cytology specimens: a potential improvement in accuracy and productivity is demonstrated in a preclinical feasibility trial. Am J Clin Pathol Sep 2002;118(3): 399–407. [13] Pecorelli S, Zigliani L, Odicino F. Revised FIGO staging for carcinoma of the cervix. Int J Gynaecol Obstet May 2009;105(2):107–8. [14] Boulanger JC, Fauvet R, Urrutiaguer S, Drean Y, Sevestre H, Ganry O, et al. Cytological history of cases of invasive cervical cancer diagnosed in France in 2006. Gynecol Obstet Fertil Sep 2007;35(9):764–71. [15] Koss LG. The Papanicolaou test for cervical cancer detection. A triumph and a tragedy. JAMA Feb 3 1989;261(5):737–43. [16] Andrae B, Kemetli L, Sparen P, Silfverdal L, Strander B, Ryd W, et al. Screeningpreventable cervical cancer risks: evidence from a nationwide audit in Sweden. J Natl Cancer Inst May 7 2008;100(9):622–9. [17] Bulk S, Rozendaal L, Zielinski GD, Berkhof J, Daalmeijer NC, Snijders PJ, et al. High-risk human papillomavirus is present in cytologically false-negative smears: an analysis of “normal” smears preceding CIN2/3. J Clin Pathol Mar 2008;61(3):385–9.
[18] Ejersbo D, Dahl MB, Holund B. False negative Pap smears in a Danish material. Ugeskr Laeger Jun 2 2003;165(23):2391–4. [19] Kenter GG, Schoonderwald EM, Koelma IA, Arentz N, Hermans J, Fleuren GJ. The cytological screening history of 469 patients with squamous cell carcinoma of the cervix uteri; does interval carcinoma exist? Acta Obstet Gynecol Scand Apr 1996;75(4):400–3. [20] Anderson GH, Benedet JL, Le Riche JC, Matisic JP, Thompson JE. Invasive cancer of the cervix in British Columbia: a review of the demography and screening histories of 437 cases seen from 1985–1988. Obstet Gynecol Jul 1992;80(1):1–4. [21] Kristensen GB, Skyggebjerg KD, Holund B, Holm K, Hansen MK. Analysis of cervical smears obtained within three years of the diagnosis of invasive cervical cancer. Acta Cytol Jan 1991;35(1):47–50. [22] van der GY, Vooijs GP, Gaillard HL, Go DM. Screening errors in cervical cytologic screening. Acta Cytol Jul 1987;31(4):434–8. [23] Kirschner B, Simonsen K, Junge J. Comparison of conventional Papanicolaou smear and SurePath liquid-based cytology in the Copenhagen population screening programme for cervical cancer. Cytopathology Aug 2006;17(4):187–94. [24] Ring M, Bolger N, O'Donnell M, Malkin A, Bermingham N, Akpan E, et al. Evaluation of liquid-based cytology in cervical screening of high-risk populations: a split study of colposcopy and genito-urinary medicine populations. Cytopathology Jun 2002;13(3):152–9. [25] Leung KM, Lam KK, Tse PY, Yeoh GP, Chan KW. Characteristics of false-negative ThinPrep cervical smears in women with high-grade squamous intraepithelial lesions. Hong Kong Med J Aug 2008;14(4):292–5. [26] Lonnberg S, Anttila A, Kotaniemi-Talonen L, Kujari H, Melkko J, Granroth G, et al. Low proportion of false-negative smears in the Finnish program for cervical cancer screening. Cancer Epidemiol Biomarkers Prev Feb 2010;19(2):381–7. [27] Rebolj M, van BM, Lynge E, Looman C, Essink-Bot ML, Boer R, et al. Incidence of cervical cancer after several negative smear results by age 50: prospective observational study. BMJ 2009;338:b1354. [28] Baay MF, Smits E, Tjalma WA, Lardon F, Weyler J, Van RP, et al. Can cervical cancer screening be stopped at 50? The prevalence of HPV in elderly women. Int J Cancer Jan 10 2004;108(2):258–61. [29] Flannelly G, Monaghan J, Cruickshank M, Duncan I, Johnson J, Jordan J, et al. Cervical screening in women over the age of 50: results of a population-based multicentre study. BJOG Apr 2004;111(4):362–8. [30] McCredie MR, Sharples KJ, Paul C, Baranyai J, Medley G, Jones RW, et al. Natural history of cervical neoplasia and risk of invasive cancer in women with cervical intraepithelial neoplasia 3: a retrospective cohort study. Lancet Oncol May 2008;9(5):425–34.