Factors Affecting the Histopathological Outcomes of Atypical Glandular Cells on Pap Test.

BACKGROUND: Glandular cell abnormalities may indicate the presence of pre-malignant or malignant lesions. AIM: This study aimed to investigate the relationship between atypical glandular cells (AGC) and patients' demographics, histopathological outcomes, Human Papillomavirus (HPV) test results.
MATERIAL AND METHODS: Between January 2015 and December 2019, women with AGC on Pap tests were retrieved from the hospital electronic database. The patients with AGC on cervicovaginal smears who underwent further pathological, laboratory, and imaging diagnostic testing and who were followed up at least 1-year were included in the study, while those who had a history of cervical dysplasia or cancer, lost during follow-up, or had missing data were excluded.
RESULTS: Of 85,692 Pap smears, 114 (0.13%) were diagnosed with AGC, of those 88 cases were eligible for final analysis. Gynecological malignancies were detected in 13 (14.8%) patients; including 6 (6.8%) endometrioid endometrial cancers, 3 (3.4%) non-endometrioid endometrial cancers, 2 (2.3%) cervical adenocarcinomas, 1 (1.1%) cervical squamous cell carcinoma, and 1 (1.1%) high-grade tubal serous cancer. Multivariate analysis revealed that presence of concomitant abnormal squamous lesion (P = 0.002), being 50 years and older (P = 0.028), HPV positivity (P < 0.001), and menopause (P = 0.023) were risk factors for significant pathology.
CONCLUSION: The diagnosis of AGC may be related to the preneoplastic/neoplastic processes. A further comprehensive histopathological examination is required in women with AGC, aged 50 years and older, postmenopausal, HPV-positivity and concomitant squamous cell abnormality Clinicians should consider ovarian pathologies when there is no pathological finding on endometrial or cervical histopathological examination. Copyright:

INTRODUCTION

Over the years, successful screening programs have been established for cervical cancer, first, including the Papanicolaou (Pap) test and subsequently Human Papillomavirus (HPV) testing. Despite the declining trend in the diagnosis of cervical cancer due to the success of cancer screening programs, the prevalence of cervical and endometrial glandular lesions, such as endometrial hyperplasia and adenocarcinoma is demonstrating a rising disposition.[1] While squamous lesions constitute most of the abnormal findings in the Pap test, a remarkable number of lesions are related to glandular abnormalities.[2]

Glandular cell abnormalities are rare in the Pap test, and they could indicate the presence of pre-malignant or malignant lesions of the cervix or endometrium. The Bethesda System updated from the definition of “atypical glandular cell of undetermined significance (AGUS)” to the description of “atypical glandular cells (AGC)” and subcategories. AGC was divided into endometrial cells, endocervical cells, and glandular cells based on the source of cells and subdivided into “not otherwise specified” and “favor neoplastic (AGC-NOS, AGC-FN)”.[3]

Many studies reported that cytological examination of glandular cell lesions in Pap test is less effective in diagnosis compared to squamous cell lesions.[4] Additionally, the incidence of atypical glandular cells is quite low. One study reported the incidence of atypical glandular cells in the Pap test as 0.17%.[5] Some studies reported that approximately 9% to 35.5% of women with AGC may have squamous pathologies ranging from low-grade squamous intraepithelial lesion (LSIL) to invasive carcinoma.[56] In a large cohort study with a high number of patients, women with AGC Pap test results had 61 times higher adenocarcinoma risk compared to women with normal Pap test result.[7]

Although there are many studies in the literature questioning the relationship between the cytologic diagnosis of AGC and histological squamous or glandular pathologies, only a few studies investigating the relationship between AGC cytology and HPV result.[589] In the present study, we aimed to investigate the relationship between atypical glandular cells (AGC-NOS, AGC-FN) and patients' demographics, HPV-testing results, endometrial biopsy results, and colposcopy-guided cervical biopsy results.

MATERIALS AND METHODS

Following approval of the research ethics committee (Approval number: 2021/8) of our institution, we retrospectively analyzed the data of women diagnosed with atypical glandular cells at Pap test from January 1, 2015 to December 31, 2019.

Clinical, demographic, and histopathological records were obtained through institutional electronic medical records. ThinPrep (Hologic Inc., Marlborough, Mass., USA) and the Roche Cobas® 4800 HPV assay were used for liquid-based cytologic examinations and HPV testing, respectively.

Patients with AGC cytology, who underwent further examination (histopathological, laboratory, and imaging) were included in our study. Patients who had a history of cervical dysplasia or cancer, lost during follow-up, had missing data such as demographics, HPV status were excluded from the study. Glandular cell abnormalities are classified according to the Bethesda 2014 System. Cervical intraepithelial neoplasia (CIN) 2/3, adenocarcinoma in situ (AIS), cervical squamous cell carcinoma, cervical adenocarcinoma, endometrioid endometrial cancer, non-endometrioid endometrial cancer, and high-grade serous cancer were defined as significant pathologies.

Endocervical curettage (ECC) and cervical biopsy under colposcopy were performed in each patient diagnosed with AGC. Endometrial sampling was also performed in women aged > 35 years or those having high-risk factors for endometrial pathologies such as obesity, polycystic ovary syndrome, and abnormal uterine bleeding. In addition, all patients with glandular abnormalities in Pap smear were evaluated by abdominal ultrasonography, and if necessary, magnetic resonance imaging (MRI) was also done. Demographic, clinical characteristics, histopathological findings, HPV testings, cervicovaginal smears, colposcopy-guided cervical biopsies, endocervical curettages, loop electrosurgical excision procedures or cold knife conizations, endometrial biopsies, and hysterectomies were evaluated within one year after the index test. All specimens were evaluated by experienced gynecologic pathologists.

Statistical analyses were performed using Statistical Package for Social Sciences (IBM SPSS, Version 25.0. Armonk, NY: IBM Corp.). Continuous and normally distributed data were expressed as mean (± standard deviation) and categorical variables as counts and percentage. The χ2 test or Fisher's exact test was used where appropriate. Multivariate analysis was performed using a logistic regression model. Odds ratio (OR) with 95% confidence interval (CI) was calculated. A value of P < 0.05 was considered statistically significant.

RESULTS

Cervical cytology data were obtained for 85,692 cases, and AGC was detected in 114 women. The AGC incidence was 0.13%. The subclassifications of the 114 AGC cases included AGC-NOS in 107 cases (93.9%), AGC-FN in 7 cases (6.1%). Among the 114 women who were assessed for eligibility, 88 women with AGC were enrolled in this retrospective study [Figure 1]. Eighty-four women were assigned to the AGC-NOS group and 4 women were assigned to the AGC-FN group. The mean age of the patients was 46.09 ± 11.44 years. The results of HPV testing of 68 (77.3%) patients with AGC were negative. Nine (10.2%) patients were detected to be positive for HPV 16/18, 7 (8%) for non-16/18 high-risk HPV, and 4 (4.5.%) for HPV 16/18 and non-16/18 high-risk HPV. All demographic characteristics of the groups are shown in Table 1. The rates of significant final pathological results were 40.7% and 18% in menopausal women and premenopausal women, respectively (P = 0.023). Two (2.4%) of 84 patients with AGC-NOS had adenocarcinoma in situ, and 7 (8.3%) of 84 patients with AGC-NOS had high-grade squamous intraepithelial lesion. Gynecologic malignancy was detected in 13 (14.8%) patients. Of the 13 patients, 6 (6.8%) patients diagnosed with endometrioid endometrial cancer, 3 (3.4%) patients diagnosed with non-endometrioid endometrial cancer (endometrial serous papillary carcinoma) [Figure 2], 2 (2.3%) patients diagnosed with cervical adenocarcinoma, 1 (1.1%) patient diagnosed with cervical squamous cell carcinoma, and 1 (1.1%) patient diagnosed with high-grade serous cancer originated in the Fallopian tube [Table 2]. The patient diagnosed with high-grade serous cancer on endometrial biopsy was found to have high-grade serous carcinoma of Fallopian tube origin following definitive surgery. Cervical biopsy results, endocervical canal curettage, and endometrial biopsy results of all patients were shown in Table 2. The concomitant abnormal squamous lesions were detected in 15 patients. Colposcopy-guided cervical biopsy results, endocervical curettage results, and endometrial biopsy results were shown in Table 3. Multivariate analysis revealed that presence of concomitant abnormal squamous lesion (P = 0.002; OR, 6.923; 95%CI, 2.097-22.857), being 50 years or older (P = 0.028; OR, 2.925; 95%Cl, 1.100-7.780), menopause (P = 0.023; OR, 3.125; 95%CI, 1.141-8.560), and HPV positivity (P < 0.001; OR, 41.333; 95%Cl, 10.405-164.192) were risk factors for significant pathology [Table 4].

Figure 1

The flow diagram of the study

Table 1

Baseline characteristics of the patients

	n=88
Age (year)	46.09±11.44
Menopausal status
Premenopausal	61 (69.3%)
Postmenopausal	27 (30.7%)
Gravida	2 (0-6)
Parity	2 (0-6)
HPV negative	68 (77.3%)
HPV positive
HPV 16/18	9 (10.2%)
Non-16/18 hrHPV	7 (8%)
HPV 16/18+ non-16/18 hrHPV	4 (4.5%)
AGC subclassification
AGC-NOS	84 (95.5%)
AGC-FN	4 (4.5%)
CA 125 (U/ml)	43.6±109.4

Values are presented as number (%), mean±±SD or median (range). AGC-FN, atypical glandular cell-favor neoplasia; AGC-NOS, atypic glandular cell-not-otherwise-specified; CA 125, cancer antigen 125; HPV, human papillomavirus; hrHPV, high-risk human papillomavirus

Figure 2

(a) Three-dimensional papillary fragment of tumor cells with irregular outline (PapanicolaouX400). (b) The malignant cells of serous carcinomas, in the form of large irregularly shaped ball of crowded cells (PapanicolaouX100). (c) Histology of the patient's uterine papillary serous carcinoma (Hematoxylin-EosinX400)

Table 2

Biopsy results of the patients with AGC.

	AGC-NOS (n=84)	AGC-FN (n=4)
Cervical biopsy
No dysplasia	55 (65.5%)	3 (75%)
CIN 1	17 (20.2%)	1 (25%)
CIN 2	1 (1.2%)	-
CIN 3	6 (7.1%)	-
AIS	2 (2.4%)	-
Squamous cell carcinoma	1 (1.2%)	-
Adenocarcinoma	2 (2.4%)	-
Endocervical canal curettage
No dysplasia	77 (91.7%)	4 (100%)
CIN 1	2 (2.4%)	-
CIN 3	4 (4.8%)	-
Adenocarcinoma	1 (1.2%)	-
Endometrial biopsy
Benign endometrium	55 (65.5%)	1 (25%)
Endometrial polyp	21 (25%)	-
Nonatypical hyperplasia	1 (1.2%)	-
Endometrioid endometrial	5 (6%)	1 (25%)
cancer	1 (1.2%)	2 (50%)
Non-endometrioid	1 (1.2%)	-
endometrial cancer
High-grade serous cancer

Values are presented as number (%). AGC-FN, atypical glandular cell-favor neoplasia; AGC-NOS, atypic glandular cell-not-otherwise-specified; AIS, adenocarcinoma in situ; CIN, cervical intraepithelial neoplasia, EIN, endometrial intraepithelial neoplasia

Table 3

Biopsy results of the AGC patients with concomitant abnormal squamous lesion who underwent colposcopy-guided cervical biopsy with endocervical curettage and endometrial biopsy

Patient no	Age (year)	Cytology	Biopsy	Procedure	Final pathology
1	40	AGC-NOS + HSIL	CIN 3	Conization	CIN 3
2	43	AGC-NOS + HSIL	CIN 3	Conization	CIN 3
3	31	AGC-NOS + LSIL	CIN 1	-	-
4	34	AGC-NOS + LSIL	Endometrial polyp	-	-
5	37	AGC-NOS + LSIL	CIN 3	Conization	CIN 3
6	32	AGC-NOS + ASC-H	SCC	Hysterectomy	SCC
7	39	AGC-NOS + ASC-H	CIN 3+ Endometrial polyp	Conization	CIN 3
8	53	AGC-NOS + ASCUS	Adenocarcinoma	Hysterectomy	Adenocarcinoma
9	63	AGC-NOS + ASCUS	NILM	-	-
10	32	AGC-NOS + ASCUS	NILM	-	-
11	32	AGC-NOS + ASCUS	CIN 3	Conization	CIN 3
12	43	AGC-NOS + ASCUS	Endometrial polyp	-	-
13	51	AGC-NOS + ASCUS	Endometrial polyp	Hysterectomy	Endometrial polyp
14	67	AGC-NOS + ASCUS	Endometrioid cancer	Hysterectomy	Endometrioid cancer
15	68	AGC-NOS + ASCUS	High-grade serous cancer	Laparotomy	High-grade tubal serous cancer

AGC-NOS, atypic glandular cell-not-otherwise-specified; ASC-H, atypical squamous cells-cannot exclude high-grade squamous intraepithelial lesion; ASCUS, atypical squamous cells of undetermined significance; CIN, cervical intraepithelial neoplasia; HSIL, high grade squamous intraepithelial lesion; LSIL, low grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy; SCC, squamous cell carcinoma

Table 4

Multivariate analysis of significant pathologic results in patients with AGC

Variables	Multivariate analysis
	OR (95% CI)	P
Age (≥ 50 years)	2.925 (1.100-7.780)	0.028
Concomitant abnormal squamous lesion	6.923 (2.097-22.857)	0.002
Menopause	3.125 (1.141-8.560)	0.023
HPV positivity	41.333 (10.405-164.192)	<0.001

AGC, atypical glandular cell; HPV, human papillomavirus

DISCUSSION

The incidence of AGC ranges between 0.05% to 2.1%.[6101112] Our study result is 0.13%, consistent with published reports rates, which was suggested that the diagnosis of AGC in our institution is well managed. Also, in our study, HPV positivity was detected in 20 of the patients with AGC in the Pap test, and premalignant and malignant cervical pathologies were found in most of them (16/20; 80%), which is consistent with the findings of Kuhn's study that there is an increased risk of premalignant and malignant lesions in patients with HPV positivity on the AGC Pap smear.[9]

In the comprehensive study by Zhao et al.,[13] women with AGC aged ≤40 years were significantly associated with cervical preneoplastic and neoplastic pathologies, whereas women aged 50 years and over were associated with endometrial precancerous and malignant pathologies. Similarly, Reynolds et al.[14] revealed that patients with AGC diagnosed with cancer are significantly older compared to younger age groups (59.6 vs 49.2, respectively; P < .001). Harbhajanka et al.[8] reported that endometrial pathologies over 65 years of age were significantly higher than other age groups. In the present study, the mean age of women with AGC was 46 years, cervical in situ and malignant pathologies are frequent in women aged 30 to 45 years, endometrial and ovarian malignant pathologies are prevalent in women aged 50 years and over. We have obtained comprehensive results demonstrating that there is a meaningful difference in women aged 50 years and older to determine significant pathologies. This is in good agreement with previous findings in the literature.[13]

AGC represents cytological differentials that exceed benign cytological changes but do not have the features of adenocarcinoma or precursors of these lesions.[15] Histological results of AGC detected in cervicovaginal smear range from benign lesions to cervical intraepithelial neoplastic lesions and genital tract malignancies, which is difficult for pathologists to determine where it originates. Therefore, the presence of AGC in Pap tests may indicate underlying various degrees of pathological lesions.

Although AGC is less common compared to ASCUS in cervical cytology, it is more associated with malignancies.[10] When AGC diagnosis is categorized into groups as AGC-NOS and AGC-FN, a large-scale specialized hospital study by Zhong et al.[16] reported that AGC-FN was more likely to be associated with significant pathologies than AGC-NOS, both of which required further investigation with endocervical curettage and biopsies. Similarly, Kawano et al.[17] demonstrated the benefits of subcategorizing AGC-NOS and AGC-FN to help predict lesion site, particularly in uterine and extra-uterine lesions. Since AGC may carry a risk for neoplastic changes or more, it is important for clinicians to perform a meticulous clinical evaluation and long-term follow-up of patients with this interpretation.

In the current data, detection rates of premalignant or malignant pathologies for AGC-NOS and AGC-FN vary between 15%–43% and 41%–100%, respectively.[181920] In this study, high-grade intraepithelial lesions were detected in 13 of 84 patients diagnosed with AGC-NOS, while lower genital tract cancers were detected in 10 of 84. In the AGC-FN group, 3 of 4 patients had gynecological malignancies, 2 of whom were endometrial serous carcinoma and 1 of them was endometrioid endometrial carcinoma. Therefore, it can be deduced that physicians should pay attention to the diagnosis of AGC on Pap tests, particularly in AGC-FN, and need to perform a further comprehensive diagnostic assessment including colposcopy-guided cervical biopsy, endocervical curettage, endometrial biopsy, and pelvic examination.

Histological follow-up results of patients with AGC mostly consist of benign lesions, followed by premalignant cervical neoplasias, malignant gynecological and non-gynecological cancers.[16212223] In the comprehensive study of Toyoda et al.,[21] including 1,254 AGC patients, 53.1% (666/1254) of histological diagnosis was benign, 29.0% (364/1254) were malignant lesions. In their histologic follow-up, 209 patients had endometrial carcinoma, 135 patients had cervical cancer, 26 patients had ovarian cancer, two patients had breast cancer, and two patients had malignant lymphoma. A recent large study by Jang et al.[22] demonstrated that of the 540 patients, most of whom observed benign entities (203/540;37.6%), followed by pre-malignant (183/540;33.9%) and malignant pathologies (76/540;33.9%) including 33 cases of endometrial adenocarcinoma, 26 cases cervical adenocarcinoma and 6 cases of ovarian cancer, 5 cases of tubal cancer, 5 cases of metastatic non-gynecologic cancers. Boyraz et al.[23] reported that while most of their patients (53/80; 66.25%) had nonsignificant pathologies, 24 were malignant pathologies (24/80; 30%); 12 of them (51%) were endometrial cancer, 8 (43%) were cervical carcinoma, 2 (2.5%) ovarian cancer and 2 (2.5%) Krunkenberg tumors. The findings of the present study were consistent with previous studies, and the majority of patients were benign cases (62/88) followed by pre-neoplastic lesions (9/88) and neoplastic lesions (13/88), respectively. In our study, the most common gynecological cancers were endometrial cancer, followed by cervical and ovarian cancer, respectively. According to the published studies, the most common histological outcome of AGC was squamous epithelial abnormalities, which appears to have shifted to the predominance of glandular abnormalities in recent research.[24] In current large-scale studies, the higher detection of endometrial cancers in the histopathological results of patients with AGC, as in our research, can be attributed to efficacious cervical cancer screening programs, the more widespread use of liquid-based cytology and the increase in the detection of uterine cancer over the years. In addition, it is worth noting that ovarian pathologies have been observed, albeit very rarely, reported in the literature to be 0.1–0.6%.[13] In the present study, high-grade tubal serous carcinoma was detected in one patient. Here, it could conclude that although there were no histopathological abnormalities in cervical biopsy or endocervical curettage specimens performed in patients with AGC, clinicians should consider that underlying ovarian pathologies may also exist in these patients.

The concurrent HPV positivity and AGC increased the risk of the significant pathology. In countries with national cervical screening program performed as HPV test with reflex cytology triage, AGC may be overlooked, which may result in a low incidence of AGC and also lead to adenocarcinoma cases indicated by AGC being missed. Therefore, it is important to use cervical cytology not only for cervical pathology screening but also to detect significant pathologies at female genital tract.

The small sample size due to the low incidence of abnormal glandular cells in the Pap test can be stated as a limitation of the study. The strengths of the study are the presentation of high-volume hospital experiences in which our institution is one of the largest tertiary referral centers in the country, specializing in gynecology and obstetrics, and the evaluation of histopathology samples by experienced gyneco-pathologists. Therefore, we believe our study is a valuable tool in advancing our knowledge of cytologically diagnosed AGC management and has important implications that are beneficial for clinicians. We propose further research could be undertaken in the development of new biomarkers and molecular approaches for increasing the detection rate of malignant glandular abnormalities in addition to the cytological findings.

In conclusion, it should be emphasized that the diagnosis of AGC may be related to the premalignant and malignant lesions; the diagnosis of AGC should not be neglected, clinical follow-up and advanced diagnostic tests should be performed. The awareness of clinicians should be improved, and further research should be established in patients with AGC, particularly AGC-FN. Since the diagnosis of AGC could be a sign of significant pathologies, women aged >50 years, postmenopausal women, women with concomitant HPV, women with concomitant abnormal squamous lesions should be examined further.

Ethical statement

Database management conformed to legislation on privacy and this study conforms to the ethical guidelines of the Declaration of Helsinki and approval for this retrospective study was obtained by the research ethics committee of our institution.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.