Cancer Therapy Vol 3, 435-442, 2005

Aggressive work-up is needed for menopausal women with atypical glandular cells of uncertain significance (AGUS) Pap smear

Research Article

Dong Ock Lee¹, Hoenil Jo^1,2, Youn-kyung Chung¹, Jae Weon Kim^1,2*, Noh-Hyun Park^1,2, Yong-Sang Song^1,2, Soon-Beom Kang^1,2, Hyo-Pyo Lee^1,2

¹Department of Obstetrics and Gynecology,

²Cancer Research Institute, College of Medicine, Seoul National University, Seoul, Republic of Korea

__________________________________________________________________________________

*Correspondence: Jae Weon Kim, M.D., Department of Obstetrics and Gynecology and Cancer Research Institute, Seoul National University, 28 Yungun-Dong, Chongno-Ku, Seoul, 110-744, Korea. Tel: +82-2-760-3511. Fax: +82-2-762-3599. E-mail address: kjwksh@snu.ac.kr.

Key words: Atypical glandular cells of undetermined significance, AGUS, Pap smear

Abbreviations: American Society for Colposcopy and Cervical Pathology, (ASCCP); Atypical Glandular Cells of Undetermined Significance, (AGUS); atypical glandular cells, (AGCs); cervical intraepithelial neoplasia, (CIN); computed tomographic, (CT); endocervical curettage, (ECC); endometrial biopsy, (EMBx); Loop Electrosurgical Excision Procedure, (LEEP); M. D. Anderson Cancer Center, (MDACC); National Comprehensive Cancer Network, (NCCN); Not Otherwise Specified, (NOS); punch biopsy, (p-Bx)

The first two authors contributed equally to this work.

Received: 25 May 2005; Accepted: 16 August 2005; electronically published: August 2005

Summary

This study was performed to evaluate the clinical significance of AGUS Pap smear and to review recent studies on the histological outcomes of AGUS Pap smear. From April 1998 to March 2003, a total of 78,072 Pap smears were performed at Seoul National University Hospital. Among these, 87 were classified as AGUS, an incidence of 0.11%. We reviewed the charts of these patients retrospectively to identify patients’ characteristics, symptoms, and follow-up and management methods. In addition, to comparing our data with others, we searched for studies on histological outcomes of AGUS Pap smear in Korea. Of the 87 AGUS cases, clinically significant diseases were diagnosed in 18 cases, which were; 7 endometrial cancers, 6 cervical adenocarcinomas, 3 squamous intraepithelial lesions, 1 ovary cancer, and 1 metastatic adenocarcinoma. The incidence of AGUS Pap smear was similar to that found in other Korean hospitals (0.09%). At other Korean hospitals, histologically significant diagnoses were reported with frequencies ranging from 9.3% to 80.5%, but the overall mean rate (34.1%) was similar to our data (31.6%). In an analysis of demographic characteristics, menopausal women and women with vaginal bleeding showed histologically significant disease more frequently, but age was not a statistically significant factor. Our study indicates that all patients with an AGUS Pap smear should undergo an aggressive work up. In particular, menopausal women and women with vaginal bleeding should undergo work up for underlying cervical or endometrial disease. The determination of a follow up method according to age does not appear appropriate.

I. Introduction

The 1988 Bethesda System was introduced as a guideline for cytologic cervical and vaginal specimen reports. It was devised to provide effective communication between cytopathologists and referring physicians and to facilitate cytological-histologic correlation. Atypical cells were classified as either ASCUS (Atypical Squamous Cells of Undetermined Significance) or AGUS (Atypical Glandular Cells of Undetermined Significance). AGUS was defined as glandular cells exhibiting changes beyond reactive / reparative changes, but lacking the unequivocal features of invasive adenocarcinoma. In this system, AGUS was sub-divided into endocervical, endometrial, and not otherwise specified (Lundberg, 1989). The 1991 Bethesda System recommended that AGUS should be further sub-divided into; ‘reactive’, a ‘premalignant / malignant process is favored’ or ‘Not Otherwise Specified (NOS)’ (Broder, 1992) However, AGUS subclassifications were not helpful in terms of allowing clinicians to rule out underlying pathology. Up to 10~15% of patients with AGUS-reactive process favored and up to 24~40% of patients with AGUS-NOS had significant findings on follow-up. Patients with AGUS on cervical smears need a thorough evaluation, regardless of subtyping (Veljovich et al, 1998; Reuss et al, 2001). In the 2001 Bethesda System, the classification of glandular abnormalities was significantly revised, reflecting a reappraisal of the strengths and weaknesses of cytology to assess these findings. The term ‘Atypical Glandular Cells of Undetermined Significance’ (AGUS) has been eliminated to avoid confusion with ASCUS, and the term ‘Atypical Glandular Cells’ (AGCs) used instead. The finding of AGCs was considered important clinically because the percentage of cases associated with underlying high-grade disease is higher than for ASCUS. Based on this the qualifier ‘reactive process favored’ was considered misleading and was eliminated, and AGUS was subclassified into ‘AGC’ and ‘AGC, favor neoplastic’ (Solomon et al, 2002)

The incidence of AGUS (AGC) is relatively infrequent with reported frequencies ranging from 0.1% to 0.63%. Although infrequent, AGUS is commonly associated with cervical and endometrial abnormalities in the reported range 29% to 60.9% (Gary et al, 1997; Zweizig et al, 1997; Veljovich et al, 1998; Chin et al, 2000; Chhieng et al, 2001b; Reuss et al, 2001; Hammoud et al, 2002). These studies recommended that all patients with AGUS (AGC) on Pap smear, regardless of their subclassification, should undergo a prompt and aggressive work up because of the high risk of a final pathogenic diagnosis. However, the management of AGUS on Pap smears has not been clearly defined. The American Society for Colposcopy and Cervical Pathology (ASCCP), the National Comprehensive Cancer Network (NCCN), and the M. D. Anderson Cancer Center (MDACC) proposed guidelines for the management of women with atypical glandular cells, but their recommendations differ somewhat.

The purpose of this study was to evaluate the clinical significance of AGUS on Pap smears using data from a single institution and to compare this with similar data on the histologic outcomes of AGUS on Pap smears at other institutions in Korea, and thus, to determine the effects of patient factors such as age or menopausal status or symptoms of vaginal bleeding with respect to the risk of histologically significant findings.

II. Materials and methods

From April 1998 to March 2003, a total of 78,072 Pap smears were obtained at Seoul National University Hospital. Among them 87 Pap smears were classified as AGUS by the Bethesda system. Subclassifications of AGUS were not always used, and thus patient subclassifications were not identified. Cytologic preparations were not submitted for a second review during our study.

Patients with a previous history of gynecologic malignancy were excluded, but all remaining 79 cases were. Cases with a history of squamous intraepithelial lesions were included from the analysis. We reviewed the charts of these cases retrospectively and recorded age, obstetrical characteristics, symptoms, menopausal status, use of hormone replacement therapy, methods for evaluation, and final histologic outcomes. No Pap smear case showed concurrent ASCUS or SIL and no case had a history of a previous hysterectomy.

All subjects were Korean, and Pap smears were collected using a Cytobrush by clinicians. All cervicovaginal smears was prepared on one slide and fixed immediately with 95% alcohol. There was no recommended follow-up method for AGUS Pap smear, so follow-up methods were decided upon by gynecologists. In 11 cases, Pap smear was performed repeatedly every 3 to 6 months. In other cases, colposcopy directed punch biopsy (p-Bx), endocervical curettage (ECC), endometrial biopsy (EMBx), Loop Electrosurgical Excision Procedure (LEEP), or hysterectomy was performed separately or in combination. We reviewed the final histologic outcomes of these cases.

To compare data from our center with other hospitals’ data, we searched for studies on the histologic outcomes of AGUS Pap smear at other single institutions using the KoreaMed Search System, which provides access to medical journals published in Korea. Seven studies conducted from 1998 to 2002 that reported histologic outcomes of AGUS were reviewed, and 6 among these were found suitable for statistical analysis. All studies used the Bethesda system, but subclassification was not always performed. All patients with a preexisting gynecologic malignancy were excluded. Each clinician decided on appropriate methods for the histologic diagnosis of patients with AGUS Pap smear. The incidence of AGUS Pap smear, the age distribution of patients, and histologic outcomes were reviewed and compared with our data.

We regarded both preinvasive and invasive gynecologic neoplasms as clinically significant lesions. These included squamous intraepithelial lesions, cervical adenocarcinoma in situ, endometrial hyperplasia with atypia, cervical squamous cell carcinoma, cervical adenocarcinoma, endometrial adenocarcinoma, and cancers of other sites.

Statistical analysis was performed using chi square test of the standard error of the difference between percentages. Statistical calculations were performed using SPSS software.

III. Results

From April 1998 to March 2003, the incidence of AGUS Pap smear was 0.11% at Seoul National University Hospital. Among the 87 subjects with an AGUS Pap smear, 8 subjects with history of a preexisting gynecologic malignancy were excluded.

For the remaining 79 cases the mean age was 48.2 years (range from 31 to 77), and mean gravity and parity were 4.4±2.9 and 2.8±1.8 respectively. 38 cases had a postmenopausal status (48.1%) and 11 cases were receiving hormone replacement therapy (13.9%). The age distribution of the study subjects is shown in Figure 1. Eleven (13.9%) were receiving hormonal replacement therapy at the time of diagnosis. Among 53 cases (67.1%) had no clinical symptoms, 18 (22.8%) had a presenting symptom of vaginal bleeding, and 3 of these 18 had complained of low abdominal pain, 1 of primary infertility, 1 of dysmenorrhea, 1 of leukorrhea, 1 of uterine prolapse, and 1 of vulvar discomfort. Of the 79 patients, 11 were lost to follow-up, and of the remaining 68, 11 underwent only repetitive Pap smear every 3~6 months. And their follow-up Pap smear results were all within normal limits or showed benign cellular changes.

Table 1. Methods of histologic evaluation

Methods	No. of patients (%)
p-Bx* + EMBx + ECC* + LEEP****	3 (5.3%)
p-Bx + EMBx + ECC	21 (36.8%)
p-Bx + EMBx + LEEP	1 (1.8%)
p-Bx + EMBx	8 (14.0%)
p-Bx + ECC	3 (5.3%)
EMBx + ECC	4 (7.0%)
p-Bx only	7 (12.3%)
EM Bx only	6 (10.5%)
ECC only	1 (1.8%)
LEEP only	1 (1.8%)
Operation	2 (3.5%)
Total	57 (100%)

*p-Bx: colposcopy directed punch biopsy

**EMBx: endometrial biopsy

***ECC: endocervical biopsy

****LEEP: loop electrosurgical excision procedure

Table 2. The histologic outcomes of patients with an AGUS Pap smear

Histologic Finding	No of patients	Frequency (%)
SIL	3	5.3%
Cervix squamous cell carcinoma (Cx SCCA)	0	0%
Cervix adenocarcinoma (Cx adenoCa)	6	10.5%
Endometrial adenocarcinoma (EM adenoCa)	7	12.3%
Ovary cancer	1	1.8%
Other cancer*	1	1.8%
Total histologically significant cases	18	31.6%
Negative/benign	39	68.4%
Total	57	100%

*Other cancer: ACUP (Adenocarcinoma of Unknown Primary) with metastatic adenocarcinoma

Table 3. Incidence and histologic outcome of AGUS Pap smear in other Korean hospitals

	Kim et al, 1998	Kim et al, 2000	Song et al, 2000	Park et al, 2001	Seok et al, 2002	Suh et al, 2002	Total
Incidence	0. 08%	0.13%	0.3%	0.1%	0.09%	0.15%	0.09%
Incidence	(326/407,451)	(87/67,730)	(47/17,744)	(23/19,400)	(104/115,555)	(16/10,807)	(603/638,687)
Mean age	43.0	45.8	47.0	40.9	50.5	42.6	44.6
No. of patients with biopsy	268 (82.2%)	87 (100%)	43 (91.5%)	9 (39.1%)	87 (83.7%)	7 (43.8%)	501 (83.1%)
Histologically significant diagnoses	61/268	30/87	4/43	3/9	68/87	5/7	171/501
Histologically significant diagnoses	(22.8%)	(34.5%)	(9.3%)	(33.3%)	(78.2%)	(71.4%)	(34.1%)
Negative/Benign	207(77.2%)	57(65.5%)	39(90.7%)	6(66.7%)	19(21.8%)	2(28.6%)	330(65.9%)
SIL	28(10.4%)	7(8.0%)	2(4.7%)	2(22.2%)	9(10.3%)	2(28.6%)	50(10.0%)
AIS	5(1.9%)	0	0	0	9(10.3%)	2(28.6%)	320(63.9%)
Cervical glandular atypia/dysplasia	8(3%)	0	0	0	0	0	8(1.6%)
Cervix SCCA	0	2(2.3%)	0	0	10(11.5%)	0	12(2.4%)
Cervix adenoCa	7(2.6%)	14(16.1%)	1(2.3%)	0	21(24.1%)	0	43(8.6%)
Endometrial adenoCa	11(4.1%)	7(8.0%)	1(2.3%)	1(11.1%)	15(17.2%)	1(14.3%)	36(7.2%)
Ovary cancer	0	0	0	0	3(3.4%)	1(14.3%)	4(0.8%)
Other cancer	2(0.7%)*	0	0	0	1(1.1%)**	0	3(0.6%)

*One case of MMMT (Malignant mixed Mullerian tumor) and one case of metastatic adenocarcinoma

** One case of colon cancer

Table 4. The relationships between demographic variables and histologically significant disease

	No of patients	Histologically significant diseases (Rate)	Statistical significance
Menopausal status
Premenopausal	28/57	4/28 (14.3%)	p=0.006
Postmenopausal	29/57	14/29 (48.3%)
Vaginal bleeding
With vaginal bleeding	15/57	10/15 (66.7%)	p=0.001
Without vaginal bleeding	42/57	8/42 (19.0%)
Age
≤ 35	9/57	2/9 (22.2%)	p=0.510
› 35	48/57	16/48 (33.3%)
≤ 40	15/57	3/15 (20%)	p=0.261
› 40	42/57	15/42 (35.7%)

IV. Discussion

The purpose of our study was to evaluate the clinical significance of AGUS Pap smear results from our data and to compare our results with those of others. The incidence of an AGUS Pap smear was 0.11%, which is similar to the overall incidence at other hospitals in Korea (0.09%), but somewhat lower than that in the United States (0.17%~1.8%). This may be explained by the different incidences of endometrial and cervical cancer, which is much lower in Korea than in the United States, although, the rate of cervical adenocarcinoma in Korea (9.2%) is similar to that in the United States (KSGO, 2003).

The rate of pathologically significant diagnoses in patients with an AGUS Pap smear was 31.6% at our center, and 34.1% at other hospitals in Korea. So we concluded that although the incidence of AGUS Pap smear in Korea differs from that of the United States, the meaning of AGUS Pap smear serves to provide the same warning of the possible presence of a gynecologic malignancy or premalignant lesion.

On the basis of our current review, those with an AGUS Pap smear should be considered for aggressive work up. Several studies have reported high rates of premalignant and malignant lesions in patients with an AGUS Pap smear, and authors agree that colposcopy and directed biopsy with endocervical curettage should be performed in patients with an AGUS Pap smear (Chin et al, 2000; Hammond et al, 2002). The question as to whether an endometrial biopsy should be used routinely is controversial, but it is generally recommended in older women (Veljovich et al, 1998). In our data, patients underwent various combinations of follow-up procedures because of the absence of a confirmed follow-up protocol for patients with an AGUS Pap smear. In Korea, procedures for histologic diagnosis are relatively inexpensive. For example, it takes only about 16~42 U.S. dollar for EMBx and 1~3 U.S. dollar for p-Bx. Thus physicians are free to choose procedures for a histologic diagnosis without undue burden.

In the present study, three patients with history of LEEP due to cervical intraepithelial neoplasia (CIN) were included. Their pathologic outcomes were all benign lesion. This may have been due to an alteration of the microscopic anatomy of the endocervix following LEEP (Lee, 1993). Three patients in our study were in a postpartum state; on postpartum days 20, 47, and 50. We performed colposcopy and directed biopsy, endocervical curettage, and endometrial biopsy on these patients, and obtained a benign result in all. This result may be explained by an outgrowth of the endocervix during pregnancy that often causes downgrowth of the endocervical mucosa beyond the external os, where it may be readily sampled (Ostegard, 1979). Chhieng et al. reported that the incidence of AGUS among pregnant and postpartum women was 0.26%, and that the frequency of a significant pathology by biopsy following an AGUS Pap smear in pregnant and postpartum women was 29.4%, which is similar to that in the general population. They recommended women with an AGUS Pap smear should be followed closely with an aggressive work up (Chhieng et al, 2001a).

Stratification of AGUS by subclassification for the assessment of underlying pathologic conditions was attempted, but no impact on the need for working up was found, as the benign subgroup had a 26.3~30% risk of a pathologic condition (Gary et al, 1997; Veljovich et al, 1998). Thus, we did not perform a subclassification of AGUS.

Zweizig et al, evaluated multiple demographic variables and failed to identify any subset of women with AGUS who were at increased risk of cancer or of its precursors. Age, menopause, abnormal bleeding, and current hormone replacement therapy were not found to be correlated with a poorer pathologic outcome. However, no woman <35 years old had endometrial pathologic findings and thus they recommended an endometrial biopsy for the subset > 35 years old (Zweizig et al, 1997). Chin et al. concluded that postmenopausal status and abnormal vaginal bleeding are associated with endometrial or glandular disease. However, they agreed that an evaluation of the endometrium may be warranted only in patients >35 years old, because the majority of glandular lesions of endometrial origin occurred in postmenopausal women (Chin et al, 2000). Meath et al. recommended that an evaluation of the endometrium in all women with AGUS on Pap smears is warranted, until the safety of excluding certain patient groups is established, because more than half of the malignancies identified in their review were of endometrial origin (Meath et al, 2002).

Chhieng et al. reported that the incidence of AGUS in postmenopausal women was 0.51% in their study, a finding that is similar to that in the general patient population (0.56%). The incidence of clinically significant lesions in postmenopausal patients with AGUS Pap smears was 33%, which is similar to that in the general population (29%) (P>0.05) (Chhieng et al, 2001b).

In our data, post-menopausal women and women with vaginal bleeding showed significantly higher rates of an abnormal histologic outcome, but age did not show a statistically significant difference.

At present, no formal recommendation has been issued by the American College of Obstetricians and Gynecologist for the standard management of AGUS Pap smear (Chin et al, 2000). The ASCCP recommended that all patients with an AGUS Pap smear, except those with atypical endometrial cells, should undergo colposcopy with endocervical sampling. The majority of patients with an initial diagnosis of AGUS will have a squamous lesion, CIN 2 or 3, so the initial recommended work-up for all patients should include colposcopy and endocervical curettage (ECC) (Levine et al, 2003). If patients are >35 years old or have a history of abnormal bleeding, endometrial sampling should be added to the work up. Women with atypical endometrial cells should undergo endometrial sampling. Then if the first histologic work up reveals no invasive disease, 4 repeat cytologies at 4~6 month intervals are recommended. And if the initial Pap is “favor neoplasia” and the first histopathologic work up reveals no invasive disease, a diagnostic excisional procedure is recommended.

Other cancers, including those which arise from the fallopian tube, ovary, peritoneum, colon, and pancreas, have been found in those with an AGC Pap smear. Due to the low incidence of these cancers, there is no consensus for the further work-up of women diagnosed with AGC favor neoplasia and with no identifiable lesion in vagina, cervix, or endometrium. Pelvic ultrasound, dilatation and curettage, or hysteroscopy with the addition of the serum tumor markers like CA-125 may be applied in these cases, and a computed tomographic (CT) scan of the abdomen and pelvis can be beneficial in selected patients. The addition of colonoscopy and mammography might be reasonable based on personal and family history (Levine et al, 2003).

Our study supports the opinion that all the patients with an AGUS Pap smear should undergo an aggressive work-up. In particular, menopausal women and women with vaginal bleeding should undergo work up for underlying cervical or endometrial disease. A determination of the follow up method according to age does not seem appropriate.

Acknowledgement

This study was supported by a grant of the Korea Health 21 R&D Project, Ministry of Health & Welfare, Republic of Korea (0412-CR01-0704-0001).

References

Broder S (1992) Rapid communication-The Bethesda system for reporting cervical/vaginal cytologic diagnoses-Report of the 1991 Bethesda workshop. JAMA 267, 1892.

Chhieng DC, Elgert P, Cangiarella JF (2001a) Significance of AGUS Pap smears in pregnant and postpartum women. Acta Cytol 45, 294-299.

Chhieng DC, Elgert P, Cohen JM, Cangiarella JF (2001b) Clinical significance of Atypical Glandular Cells of Undetermined Significance in postmenopausal women. Cancer Cytopathol 93, 1-7.

Chin AB, Bristow RE, Korst LM, Walts A, Lagasse LD (2000) The significance of atypical glandular cells on routine cervical cytologic testing in a community-based population. Am J Obstet Gynecol 182, 1278-1282.

Gary LE, Kenneth BS, Martha AW Pamela PS, Michael TM (1997) Biopsy findings in five hundred thirty-one patients with atypical glandular cells of uncertain significance as defined by the Bethesda system. Am J Obstet Gynecol 177, 1188-1195.

Hammoud MM, Haefner HK, Michael CW Ansbacher R (2002) Atypical glandular cells of undetermined significance; histologic findings and proposed management. J Reprod Med 47, 266-270.

Kim TJ, Lim KT, Chung HW, Lee KG, Lee SH, Park IS, Park CT, Shim JU (1998) Clinical evaluation of follow-up methods and results of atypical glandular cells of undetermined significance (AGUS) detected on cervicovaginal Pap smears. Korean J Gynecol Oncol Colposc 9, 249-258.

Kim CJ, Park TC, Park JS (2000) Atypical glandular cells of undetermined significance (AGUS); Histopathologic results and the significance of the HPV DNA detection. Korean J Obstet Gynecol 43, 1154-1161.

Korean Society for Obstetrics and Gynecology (2003) Annual report of gynecologic cancer registry program in Korean for 2001. Korean J Obstet Gynecol 46, 1849-1887.

Levine L, Lucci JA, Dinh TV (2003) Atypical glandular cells: New Bethesda terminology and management guidelines. Obstet Gynecol Surv 58, 399-406.

Lundberg GD (1989) The 1988 Bethesda system for reporting cervical/vaginal cytological diagnoses. JAMA 262, 931-936.

Lee KR (1993) Atypical glandular cells in cervical smears from women who have undergone cone biopsy: A potential diagnostic pitfall. Acta Cytol 37, 705-709.

Meath AJ, Carley ME, Wilson TO (2002) Atypical glandular cells of undetermined significance; Review of final histologic diagnoses. J Reprod Med 47, 249-252.

Ostegard RD (1979) The effect of pregnancy on the cervical squamoculumnar junction in patient with abnormal cytology. Am J Obstet Gynecol 134, 759-760.

Park HJ, Kim DK, Kim JW (2001) The clinical significance of ASCUS and AGUS in Pap smear. Korean J Gynecol Oncol Colposc 12, 291-301.

Reuss E, Price J, Koonings P (2001) Atypical glandular cells of undetermined significance; subtyping as a predictor of outcome. J Reprod Med 46, 701-705.

Seok WI, Lee KB, Lee JM (2002) Clinical analysis of atypical glandular cells of undetermined significance (AGUS) on Pap smear according to menopausal status. Korean J Obstet Gynecol 45, 967-971.

Solomon D, Davey D, Kurman R (2002) The 2001 Bethesda system. JAMA 287, 2114-2119.

Song ES, Han SH, Park JH (2000) The outcomes of 17,744 cervicovaginal smears in Inha University Hospital. Korean J Obstet Gynecol 43, 363-367.

Suh YH, Ha MN, Park CH (2002) Clinical significance of conventional Papanicolau cervical cytology test: A cytohistologic comparison. Korean J Obstet Gynecol 45, 1537-1553.

Veljovich DS, Stoler MH, Anderson WA, Covell JL, Rice LW (1998) Atypical glandular cells of undetermined significance: A five-year retrospective histopathologic study. Am J Obstet Gynecol 179, 382-390.

Zweizig S, Noller K, Reale F, Collis S, Resseguie L (1997) Neoplasia associated with atypical glandular cells of undetermined significance on cervical cytology. Gynecol Oncol 65, 314-318.