Clinical utility of atypical glandular cells (AGC) classification: Cytohistologic comparison and relationship to HPV results

Gynecologic Oncology 104 (2007) 366 – 371 www.elsevier.com/locate/ygyno

Clinical utility of atypical glandular cells (AGC) classification: Cytohistologic comparison and relationship to HPV results Teresa P. Diaz-Montes a,1 , Maryam Armin Farinola b,1 , Marianna L. Zahurak c , Robert E. Bristow a , Dorothy L. Rosenthal b,⁎ a

The Kelly Gynecologic Oncology Service, Department of Gynecology and Obstetrics, The Johns Hopkins Medical Institutions, Baltimore, MD 21287, USA b Department of Pathology, Johns Hopkins Medicine, 600 N. Wolfe Street, Baltimore, MD 21287-6940, USA c Department of Biostatistics, Johns Hopkins Oncology Center, The Johns Hopkins Medical Institutions, Baltimore, MD 21287, USA Received 3 April 2006 Available online 16 October 2006

Abstract Objectives. To determine the utility of the category of atypical glandular cells (AGC) in the management of patients with putative cervical neoplasia and to correlate HPV-DNA test results when available. Methods. The Johns Hopkins Hospital cytopathology records of 50,668 women patients were searched for all liquid-based gynecologic cytology (LBP) results of Atypical Glandular cells of Undetermined Significance (AGUS) and AGC from January 1, 2001 through December 31, 2003, yielding 98 patients (0.19%). Oncogenic HPV-DNA tests were performed on the residual fluids of 43 of these patients, 37 of whom had follow-up biopsy. During the period of January 1, 2000 and December 31, 2002, we identified 237 patients (0.58%) with conventional Pap smears in the AGUS or AGC category, among 41,024 conventional smears collected contemporaneously. To avoid confusion in this paper, AGC will be used to replace those results that originally were AGUS. Results. Following the 98 LBP AGC interpretations, 24 lesions (33.8%) were discovered out of 71 biopsies. HPV-DNA tests were performed on the residual of 43 LBPs, 18 (41.9%) were positive for oncogenic HPV. Only 37 patients had follow-up biopsy and 15 (40.5%) were positive for oncogenic HPV, of which 40% (6/15) had a significant lesion. Of the 237 conventional AGC Paps, 18 lesions (15.3%) were discovered out of 118 biopsies. Conclusions. Comparison of liquid-based and conventional Pap tests revealed a significant difference (33.8% vs. 15.3% respectively) (OR: 2.84, 95% CI: 1.4–5.73, p = 0.004) in the detection of glandular and squamous lesions. HPV testing may prove beneficial to triage AGC patients with negative colposcopic findings and positive HPV results. © 2006 Elsevier Inc. All rights reserved. Keywords: AGC; HPV; Adenocarcinoma in situ; Endocervical adenocarcinoma

Introduction Pap smears are primarily a screening test for squamous lesions of the uterine cervix. The incidence of endocervical adenocarcinoma is increasing and occurs in younger women than does squamous carcinoma of the cervix. The risk factors are similar to squamous cell carcinoma and share an association with human papillomavirus (HPV) [1]. Fifty percent of endocervical lesions have coexisting squamous lesions [1]. ⁎ Corresponding author. Fax: +1 410 614 9556. E-mail address: [email protected] (D.L. Rosenthal). 1 Contributed equally to the preparation of this manuscript. 0090-8258/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2006.09.001

Adenocarcinoma in situ is the precursor to invasive endocervical adenocarcinoma. The sensitivity for cytologic detection of glandular neoplasia is lower than for squamous, limited by problems with sampling and interpretation. The term “atypical glandular cells of undetermined significance” (AGUS) was introduced at the 1988 Bethesda Conference and defined as morphologic changes in glandular cells beyond those that are suggestive of a benign reactive process, but insufficient for the interpretation of adenocarcinoma in situ (AIS). In the new 2001 Bethesda System, the term has been eliminated, the category redefined and renamed “atypical glandular cells” (AGC), with the following subclassifications: not otherwise specified (NOS), favor neoplasia, endocervical

T.P. Diaz-Montes et al. / Gynecologic Oncology 104 (2007) 366–371

AIS, and adenocarcinoma [2]. The category of endocervical glandular cell atypia covers a wide spectrum of changes ranging from reactive to adenocarcinoma in situ. The differential diagnosis of atypical endocervical cells includes tubal metaplasia, lower uterine segment, reactive changes/repair, microglandular hyperplasia, endocervical polyp, HSIL involving endocervical glands, endocervical AIS, and even invasive adenocarcinoma. In order to establish the clinical utility of an AGC result in our institution, we examined three parameters: (a) comparison between cytology and biopsy results; (b) HPV-DNA results when available; (c) performance of conventional Pap smears (CS) versus liquid-based Pap tests (LBP). Methods Approval to conduct this study was obtained from the Johns Hopkins Medical Institutions Clinical Research Committee and Joint Committee on Clinical Investigation. The pathology archives of The Johns Hopkins Hospital cytopathology laboratory were searched from January 1, 2001 to December 31, 2003 for Papanicolaou tests that included the category of AGUS or AGC on liquid-based gynecologic samples. For purposes of this paper, we have collapsed the two categories and will henceforth refer only to AGC results. The samples were processed using the SurePath™ PrepStain™ device (TriPath Imaging, Inc., Burlington NC), following the manufacturer's instructions. Out of the 98 patients with a result of AGC, 71 patients were subsequently biopsied. Oncogenic HPV-DNA tests (Digene, Gaithersburg, MD) were performed on the residual fluids of 43 of these patients, 37 of whom had follow-up biopsy and/or Pap. Hybrid Capture-II HPV-DNA testing has been self-validated by the Johns Hopkins Molecular Virology Laboratory for use with the SurePath preservative fluid, and samples from these patients were likewise run using that methodology based upon the manufacturer's instructions. In order to compare liquid-based Pap test results with conventional Pap smears, The Johns Hopkins Hospital cytopathology records were searched looking for the category of AGC among conventional Paps between January 1, 2000 and December 31, 2002. We chose these dates to avoid demographic bias resulting from lower socioeconomic groups getting conventional Paps before liquid-based Paps were common at our institution. Of the 237 AGC conventional Pap smears, 118 had follow-up tissue diagnosis. The method by which tissue was obtained in the evaluation of these smears varied and included cervical biopsy, endometrial biopsy, endocervical curettage, cervical conization, loop electrosurgical excision procedure and hysterectomy. Clinical information for both groups was gathered by chart review.

Results Patient demographics and characteristics A total of 50,668 liquid-based gynecologic cytology samples were reviewed at The Johns Hopkins Hospital, Division of Cytopathology during 2001–2003. Of these, 98 (0.19%) were reported as AGC. A total of 41,024 conventional Papanicolaou smears were also reviewed at The Johns Hopkins Hospital Division of Cytopathology during 2000–2002. Of these, 237 (0.58%) were reported as AGC. Liquid-based cytology cohort Demographics Among the 98 study patients, the average age was 45.3 years (range 19–86). The majority of the patients were white (53.2%),

367

and 42% reported a history of a sexually transmitted disease (STD). Twelve (17.4%) patients were nulligravid, and 35 (35.7%) patients were postmenopausal. Twenty-one (36.2%) patients reported having used oral contraceptive pills, and 10 (12.4%) patients reported having received hormone replacement therapy (Table 1). Cytohistological comparisons Twenty-four of 71 patients biopsied (33.8%) had a tissueconfirmed clinically significant lesion. The distribution was as follows: 1 (1.4%) was reported as LSIL, 8 (11.3%) were reported as HSIL, 1 (1.4%) was reported as AIS, and 2 (2.8%) were reported as complex hyperplasia. Cancer was identified in 16.9% (12 of 71), of which 6 of 12 (50%) were endometrioid carcinomas, 2 of 12 (16.7%) were malignant mixed Mullerian tumors, 2 of 12 (16.7%) were poorly differentiated squamous cell carcinoma, and 2 of 12 (16.7%) were metastases to the vagina (serous carcinoma of the ovary and poorly differentiated carcinoma from the endometrium). One endocervical adenocarcinoma in situ diagnosis was made with concurrent HSIL. All of the 10 endometrial lesions were predicted by the

Table 1 Demographic characteristics of women diagnosed with AGC via liquid-based Pap tests and conventional Pap smears Variable

Liquid-based n (%) Conventional n (%) p-value

Age (years) Mean Range

45.3 19–86

Ethnic classification White 50 (53.2) African-American 38 (40.4) Asian or Pacific Islander 6 (6.4) Other 0 (0.0) Unknown 4

46.8 19–86

0.40

89 (46.4) 90 (46.9) 12 (6.2) 1 (0.5) 45

0.55

79 (35.1) 146 (64.9) 12

0.92

History of sexual transmitted diseases None 29 (58.0) Yes 21 (42.0) Unknown 48

68 (78.2) 19 (21.8) 150

0.01

Gravity Nulligravid Gravid Unknown

12 (17.4) 57 (82.6) 29

13 (13.4) 84 (86.6) 140

0.48

Hormone replacement therapy Yes 10 (12.4) None 71 (87.6) Unknown 17

13 (6.9) 175 (93.1) 49

0.14

Oral contraception history Yes 21 (36.2) None 37 (63.8) Unknown 40

29 (19.7) 118 (80.3) 90

0.01

Menopause Yes No Unknown

35 (35.7) 63 (64.3) 0

368

T.P. Diaz-Montes et al. / Gynecologic Oncology 104 (2007) 366–371

cytologic impression except for one, which was categorized as AGC-NOS (Table 2).

Table 3 HPV results of biopsy confirmed lesions

HPV-DNA results Of the 98 patients with an AGC result, HPV-DNA tests were performed on the residual fluids of 43 of these patients, 37 of whom had follow-up biopsy (Table 3). Fifteen patients of 37 (40.5%) were positive for oncogenic HPV. Out of the 15 positive patients, 6 had a significant lesion. Only one of the “AGC-favor an endometrial lesion” LBP was tested for HPV, and the result was negative.

Total (37 cases) Benign Premalignant/Malignant

Positive

Negative

15 (40.5%) 9/15 (60.0%) 6/15 (40.0%)

22 (59.5%) 18/22 (81.8%) 4/22 (18.2%)

compared to 18 of 118 patients (15.3%) with the conventional Pap smear (OR: 2.84, 95% CI: 1.4–5.73, p = 0.004) (Table 5). Discussion

Conventional Pap smear cohort Demographics (Table 1) Among the 237 study patients, the average age was 46.8 years (range 19–86). The majority of the patients were African-American (46.9%), and 21.8% reported a history of a STD. Thirteen (13.4%) patients were nulligravid, and 79 (35.1%) patients were postmenopausal. Twenty-nine (19.7%) patients reported having used oral contraceptive pills, and 13 (6.9%) patients reported having received hormone replacement therapy. Cytohistological comparisons Eighteen (15.3%) of 118 patients had biopsy confirmation of a clinically significant lesion. High-grade SIL was discovered on evaluation following an AGC conventional Pap result in 10.2% of patients. Cancer was discovered in 1.7%, of which one was endometrioid carcinoma and the other was a metastasis. One patient was diagnosed with multifocal endocervical adenocarcinoma in situ (Table 4). Comparison of liquid-based and conventional Pap smears There was a significant difference in detection of premalignant/malignant lesions between the two preparation methods. Twenty-four of 71 patients (33.8%) had a tissue confirmed clinically significant lesion using the liquid-based Pap tests Table 2 Histologic diagnosis of AGC on liquid-based Pap test Total

71 cases

Benign Negative Metaplasia Premalignant AIS HSIL LSIL Complex hyperplasia Malignant Endometrioid carcinoma MMMT (carcinosarcoma) Poorly differentiated SCC Metastasis to vagina Total squamous lesions Total glandular lesions

47/71 (66.2%) 42/71 (59.2%) 5/71 (7.0%) 12/71 (16.9%) 1/71 (1.4%) 8/71 (11.3%) 1/71 (1.4%) 2/71 (2.8%) 12/71 (16.9%) 6/71 (8.5%) 2/71 (2.8%) 2/71 (2.8%) 2/71 (2.8%) 11/71 (15.5%) 13/71 (18.3%)

Since the introduction of The Bethesda System for reporting cervical/vaginal cytologic results, atypical cells have been classified into ASC (atypical squamous cells) and AGC (atypical glandular cells of undetermined significance). AGC are defined as glandular cells exhibiting changes beyond those of reactive/reparative, but lacking unequivocal features of invasive adenocarcinoma [3]. Clinical experience with the AGC category is sporadic due to the much lower frequency of glandular abnormalities of the cervix. Consequently, the cytologic features are not well defined and misinterpretation is a danger. The reported frequency of an AGC interpretation in relation to all cervical vaginal cytology results ranges from 0.46% to 2.5% [4–13]. A Q-probe study of the College of American Pathologists involving 348 laboratories reported that 52% of AGC cases were associated with significant lesions; however, only 5.8% were glandular while 40% were squamous [14]. Our study showed that a result of AGC led to identification of a significant pathologic lesion (glandular and/or squamous intraepithelial) in 33.8% of cases. A review of the recent literature on the topic of AGC demonstrates a high percentage of squamous lesions ranging from 8 to 83%, several with endocervical gland involvement on follow-up biopsies after a cytologic result of AGC [4,6,9,14– 19]. The majority of clinically significant AGC lesions are squamous dysplasias, and from 40 to 68% of squamous dysplasias are high-grade lesions [4–13,15,18,20–22]. Consequently, based on these collective data, up to 50% of women Table 4 Histologic diagnosis of AGC on conventional Pap smears Total

118 cases

Benign Negative Metaplasia Atypia Premalignant AIS HSIL LSIL Complex hyperplasia Malignant Endometrioid carcinoma Metastasis Total squamous lesions Total glandular lesions

100/118 89/118 3/118 8/118 16/118 1/118 12/118 2/118 1/118 2/118 1/118 1/118 14/118 4/118

(84.7%) (75.5%) (2.5%) (6.9%) (13.6%) (0.8%) (10.2%) (1.7%) (0.8%) (1.7%) (0.8%) (0.8%) (11.8%) (3.4%)

T.P. Diaz-Montes et al. / Gynecologic Oncology 104 (2007) 366–371 Table 5 Comparison of AGC follow-up results between the liquid-based Pap test and the conventional Pap smear

Benign Premalignant/Malignant

Liquid-based Pap

Conventional Pap

47/71 (66.2%) 24/71 (33.8%)

100/118 (84.7%) 18/118 (15.3%)

with an AGC result have a high-grade squamous dysplasia. Raab et al. [17] reported a SIL rate of 83% after a cytologic interpretation of AGC. Wilbur [23] emphasized the findings of squamous lesions mimicking an endocervical process, particularly when the lesion involved endocervical glands. These results are not surprising since the number of squamous lesions is far greater than the number of endocervical glandular lesions. Glandular lesions of the cervix are diagnostically and therapeutically more challenging to the cytopathologist and clinician because of their relative rarity, relative absence of colposcopic findings, irregular shedding, small size, endocervical location (making lesions less amenable to cytologic sampling) and broader differential diagnosis. Women who have glandular neoplasms generally have an endometrial adenocarcinoma rather than an endocervical adenocarcinoma [11,17]. According to numerous studies, the ratio between adenocarcinoma in situ and cervical intraepithelial neoplasia ranges from 1:26 to 1:237 [23]. An additional confounding factor is that from 24 to 75% of endocervical in situ adenocarcinomas occur in conjunction with squamous dysplasias, and the squamous dysplasia often is correctly diagnosed on the smear, whereas the glandular lesion is not recognized as such [23]. Cullimore et al. [24] reported that, of 51 women with adenocarcinoma in situ, only 53% had atypical glandular cells detected on Pap smear. Andersen and Arffmann [25] reported that, of 36 women with adenocarcinoma in situ, glandular cell atypia was detected on the Pap smear in only half. In our study, we found that a cytologic interpretation of AGC had a slightly higher occurrence of a glandular lesion (18%) than squamous (15%) when LBP tests were considered (Table 2), but the reverse was seen in the results obtained with conventional Pap smears, 11% squamous and 3% glandular (Table 4). Others have found that a squamous intraepithelial lesion is more often identified following an AGC diagnosis. Difference in collection methods (LBP vs. conventional smear) may account for the discrepancies between cell types found in the literature as most of the referenced articles were published before LBP testing became popular. Because of the infrequency of an AGC result and the low numbers of such cases, as in our study, statistical relevance may be difficult to attain. A trend may be all that can be expected. The management guidelines of the American Society for Colposcopy and Cervical Pathology recommend colposcopy with endocervical sampling as the initial confirming step for all women with a diagnosis of AGC [26]. Compared with the follow-up of ASC-US, the recommended investigation following a diagnosis of AGC is more aggressive because AGC is substantially more indicative of a high-grade neoplasm than is ASC-US. Like cervical squamous cell carcinoma and its

369

precursors, cervical adenocarcinoma and AIS are strongly associated with high-risk HPV infection [21,27]. However, while HPV testing now is a preferred method for managing the diagnosis of squamous atypia, its role as a reflex test has not been established for glandular atypia. Ronnett et al. [21] studied 137 women with AGC Pap diagnoses. All of the AIS cases and 92% of CIN 2 and 3 were detected by colposcopy prompted by positive HPV Hybrid Capture-II testing compared to 67% of the same lesions detected following an abnormal repeated cytologic examination. The HPV-DNA testing had 94% sensitivity and 50% specificity [21]. Oliveira et al. [1] demonstrated the detection of high-risk HPV-DNA in 36% of their patients following an initial Pap result of AGC. Repeat Pap results and their HPV results were: HSIL (7% of AGC results) and/or HSIL/AGC (10% of AGC results), (83–88% HPV positive); AIS (6% of AGC results) or SCC (1% of AGC results) (100% HPV positive). Irvin et al. [28] showed that women with HPV positivity were 12 times more likely to have CIN than were women with HPV negativity and that negative HPV test results in the evaluation of AGC provides excellent reassurance that no occult HSIL is present. In the present study, we demonstrated oncogenic HPV testing in 40.5% (15/37) of AGC patients with follow-up biopsy, out of which 40% (6/15) had a confirmed lesion. Only 4 of 22 (18.2%) of the HPV negative cases had a significant lesion on follow-up biopsies. Therefore, the use of HPV testing as an adjunctive test for glandular cell abnormalities may lead to improved management of women, including appropriate referral to colposcopy and the use of diagnostic cone to confirm the diagnosis in suspicious cases. Liquid-based technology was developed to improve the transfer of cells from the collection device to a microscope slide and to provide uniformity of the cell population in each sample. Methods of conventional Pap test processing can result in a variety of characteristic artifacts, including one that mimics the criterion for the diagnosis of AIS, “nuclear feathering” [15]. This feature has not been uniformly reported when liquid-based sampling is used. Due to the immediate fixation after collection, there are fewer artifacts in cellular morphology. There is less obscuring artifact from overlap, cellular debris and inflammatory cells because the cells are deposited in a relative monolayer [29]. In addition, this method provides representative residual material in collection media that can be used for additional testing, such as HPV-DNA. Roberts et al. [30] showed that, in 30 paired cytologic samples, adenocarcinoma in situ was originally predicted in 67% of the conventional smears and in 47% of the ThinPrep samples. Conversely, Ashfaq et al. [31] reported that the ThinPrep method provided more accurate diagnoses of glandular disease. In the present study, our comparison of conventional and liquid-based Pap tests revealed a statistically significant difference in the detection of glandular and/or squamous lesions. The SurePath Pap test identified a lesion in 33.8% of the cases, in comparison to the conventional method, which identified a lesion in 15.3% of cases, with p = 0.004. There are several limitations of the current study that must be considered in interpreting the data. First, we must consider the potential for selection and information bias inherent in most

370

T.P. Diaz-Montes et al. / Gynecologic Oncology 104 (2007) 366–371

retrospective investigations. The fact that only the patients with biopsies that were done or reviewed at Johns Hopkins Medical Institutions were included also introduces selection bias. Different follow-up and management strategies could have been applied to the two cohort groups, further introducing selection bias. Despite these limitations, our study reinforces the previously published concept that atypical glandular cells on Papanicolaou tests represent a cytological category with an associated high incidence of underlying malignant or premalignant lesions on follow-up histologic examination. Thus, an AGC interpretation should be worked up appropriately in accordance with the recently published guidelines of the American Society of Colposcopy and Cervical Pathology [32]. All women with this Pap result should undergo cervical and vaginal colposcopy and endocervical curettage. Those patients with unqualified AGC who have negative colposcopic examinations and endocervical curettage should have repeat Papanicolaou tests every 4 to 6 months until 4 normal smears have been obtained. Those patients with either “AGC-favor neoplasia” with negative colposcopy and endocervical curettage or AGC-NOS with an endocervical curettage positive for neoplasia should undergo cervical conization. Endometrial biopsy, hysteroscopy or curettage is recommended if cells appear to be of endometrial origin [32]. Reflex HPV testing is yet to be established for triage of women with AGC Pap results, but based on our findings and those in the literature, we predict that it may lead to improved management in those cases in which a lesion is suspected but follow-up results of colposcopy and biopsy are inconclusive. Acknowledgments Dr. Rosenthal is a paid member of the TriPath Oncology (a wholly owned subsidiary of TriPath Imaging, Inc.) Scientific Advisory Board. The terms of this arrangement are being managed by The Johns Hopkins University in accordance with its conflict of interest policies. No funds were provided by TriPath for this study.

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16] [17]

[18]

[19]

[20] [21]

References [22] [1] Oliveira ERZM, Derchain SFM, Rabelo-Santos SH, Westin MCA, Zeferino LC, Campos EA, et al. Detection of high-risk human papillomavirus (HPV) DNA by Hybrid Capture II in women referred due to atypical glandular cells in the primary screening. Diagn Cytopathol 2004;31(1):19–22. [2] Solomon D, Davey D, Kurman R, Moriarty A, O'Connor D, Prey M, et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. JAMA 2002;287(16):2114–9. [3] Kurman RJ, Solomon D. The Bethesda System for reporting cervical/ vaginal cytologic diagnoses: definitions, criteria and explanatory notes for terminology and specimen adequacy. New York: Springer-Verlag; 1994. [4] Bose S, Kannan V, Kline TS. Abnormal endocervical cells: really abnormal? Really endocervical? Am J Clin Pathol 1994;101(6):708–13. [5] DiTomasso JP, Ramzy I, Mody DR. Glandular lesions of the cervix. Validity of cytologic criteria used to differentiate reactive changes, glandular intraepithelial lesions and adenocarcinoma. Acta Cytol 1996;40 (6):1127–35. [6] Duska LR, Flynn CF, Chen A, Whall-Strojwas D, Goodman A. Clinical

[23] [24]

[25] [26]

[27]

[28]

evaluation of atypical glandular cells of undetermined significance on cervical cytology. Obstet Gynecol 1998;91(2):278–82. Eddy GL, Ural SH, Strumpf KB, Wojtowycz MA, Piraino PS, Mazur MT. Incidence of atypical glandular cells of uncertain significance in cervical cytology following introduction of the Bethesda System. Gynecol Oncol 1997;67(1):51–5. Goff BA, Atanasoff P, Brown E, Muntz HG, Bell DA, Rice LW. Endocervical glandular atypia in Papanicolaou smears. Obstet Gynecol 1992;79(1):101–4. Kennedy AW, Salmieri SS, Wirth SL, Biscotti CV, Tuason LJ, Travarca MJ. Results of the clinical evaluation of atypical glandular cells of undetermined significance (AGCUS) detected on cervical cytology screening. Gynecol Oncol 1996;63(1):14–8. Nasu I, Meurer W, Fu YS. Endocervical glandular atypia and adenocarcinoma: a correlation of cytology and histology. Int J Gynecol Pathol 1993;12(3):208–18. Raab SS, Bishop NS, Zaleski MS. Effect of cervical disease history on outcomes of women who have a Pap diagnosis of atypical glandular cells of undetermined significance. Gynecol Oncol 1999;74(3):460–4. Veljovich DS, Stoler MH, Andersen WA, Covell JL, Rice LW. Atypical glandular cells of undetermined significance: a five-year retrospective histopathologic study. Am J Obstet Gynecol 1998;179(2):382–90. Zweizig S, Noller K, Reale F, Collis S, Resseguie L. Neoplasia associated with atypical glandular cells of undetermined significance on cervical cytology. Gynecol Oncol 1997;65(2):314–8. Jones BA, Novis DA. Cervical biopsy–cytology correlation. A College of American Pathologists Q-Probes study of 22439 correlations in 348 laboratories. Arch Pathol Lab Med 1996;120(6):523–31. Burja IT, Thompson SK, Sawyer Jr WL, Shurbaji MS. Atypical glandular cells of undetermined significance on cervical smears: a study with cytohistologic correlation. Acta Cytol 1999;43(3):351–6. Mody DR. Agonizing over AGUS. Cancer Cytopathol 1999;87(5):243–4. Raab SS, Isacson C, Layfield LJ, Lenel JC, Slagel DD, Thomas PA. Atypical glandular cells of undetermined significance. Cytologic criteria to separate clinically significant from benign lesions. Am J Clin Pathol 1995;104(5):574–82. Schindler S, Pooley Jr RJ, DeFrias DV, Yu GH, Bedrossian CW. Follow-up of atypical glandular cells in cervical–endocervical smears. Ann Diagn Pathol 1998;2(5):312–7. Soofer SB, Sidawy MK. Atypical glandular cells of undetermined significance: clinically significant lesions and means of patient followup. Cancer Cytopathol 2000;90(4):207–14. Cangiarella JF, Chhieng DC. Atypical glandular cells—an update. Diagn Cytopathol 2003;29(5):271–9. Ronnett BM, Manos MM, Ransley JE, Fetterman BJ, Kinney WK, Hurley LB, et al. Atypical glandular cells of undetermined significance (AGUS): cytopathologic features, histopathologic results, and human papillomavirus DNA detection. Hum Pathol 1999;30(7):816–25. Taylor RR, Guerrieri JP, Nash JD, Henry MR, O'Connor DM. Atypical cervical cytology. Colposcopic follow-up using the Bethesda System. J Reprod Med 1993;38(6):443–7. Wilbur DC. Endocervical glandular atypia: a “new” problem for the cytologist. Diagn Cytopathol 1995;13(5):463–9. Cullimore JE, Luesley DM, Rollason TP, Byrne P, Buckley CH, Anderson M, et al. A prospective study of conization of the cervix in the management of cervical intraepithelial glandular neoplasia (CIGN)—a preliminary report. Br J Obstet Gynaecol 1992;99(4):314–8. Andersen ES, Arffmann E. Adenocarcinoma in situ of the uterine cervix: a clinico-pathologic study of 36 cases. Gynecol Oncol 1989;35(1):1–7. Levine L, Lucci III JA, Dinh TV. Atypical glandular cells: new Bethesda terminology and management guidelines. Obstet Gynecol Surv 2003;58 (6):399–406. Pirog EC, Kleter B, Olgac S, Bobkiewicz P, Lindeman J, Quint WG, et al. Prevalence of human papillomavirus DNA in different histological subtypes of cervical adenocarcinoma. Am J Pathol 2000;157(4):1055–62. Irvin W, Evans SR, Andersen W, Jazaeri A, Taylor P, Stoler M, et al. The utility of HPV DNA triage in the management of cytological AGC. Am J Obstet Gynecol 2005;193(2):559–65.

T.P. Diaz-Montes et al. / Gynecologic Oncology 104 (2007) 366–371 [29] Baker JJ. Conventional and liquid-based cervicovaginal cytology: a comparison study with clinical and histologic follow-up. Diagn Cytopathol 2002;27(3):185–8. [30] Roberts JM, Thurloe JK, Bowditch RC, Humcevic J, Laverty CR. Comparison of ThinPrep and Pap smear in relation to prediction of adenocarcinoma in situ. Acta Cytol 1999;43(1):74–80.

371

[31] Ashfaq R, Gibbons D, Vela C, Saboorian MH, Iliya F. ThinPrep Pap test. Accuracy for glandular disease. Acta Cytol 1999;43 (1):81–5. [32] Wright Jr TC, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ. 2001 Consensus Guidelines for the management of women with cervical cytological abnormalities. JAMA 2002;287(16):2120–9.