Abstract
Background
Diagnosing atypical glandular cells (AGC) is a significant challenge in cytomorphology.
Methods
A retrospective study was undertaken to assess the prevalence of AGC and the subsequent histological outcomes over 5 years at a single institution.
Results
A total of 159,649 ThinPrep Pap tests, including 395 cases of AGC, were retrieved, of which 330 AGC cases had follow-up histopathology. Among these 330 cases, 43.9% were classified as AGC not otherwise specified, followed by AGC-endocervical cells at 33.3%. The most frequently observed histological findings at follow-up included CIN1 and benign mucosa with reactive changes, followed by high-grade squamous intraepithelial lesion and cervical squamous cell carcinoma. The overall 5-year insignificant AGC rate was 0.12%, and the overall 5-year significant AGC rate was 0.08%. Notably, 36.7% of AGC cases tested positive for high-risk human papillomavirus. Interestingly, the level of experience did not significantly impact the rates for significant or insignificant AGC diagnosis. However, senior cytopathologists had a higher AGC report rate compared to their junior peers.
Conclusion
The AGC diagnostic rate at our institution falls within the range given by the College of American Pathologists. A significant number of cases had follow-up histologic results available, and the overall 5-year insignificant AGC rate was 0.12%.
Keywords: Adenocarcinoma, atypical glandular cells, significant histology, ThinPrep
According to the 2014 Bethesda System for Reporting Cervical Cytology,1 atypical glandular cells (AGC) encompass several subcategories, including AGC not otherwise specified (AGC-NOS), AGC endometrial cells (AGC-EM), AGC endocervical cells (AGC-EC), and AGC favoring neoplastic (AGC-FN). The cytologic evaluation of Papanicolaou smears with glandular abnormalities presents a diagnostic challenge. Differentiating between a benign process, e.g., reactive glandular epithelium, and a neoplastic process is a critical step in the assessment process.2
The reporting rates for AGCs in cytology laboratories participating in laboratory accreditation programs range from 0 to 0.8%, as published by the College of American Pathologists (CAP).1 According to the CAP survey, the 50th percentile rates for AGC specimens in US laboratories stood at 0.1% for conventional Pap smears and 0.2% for both ThinPrep and SurePath.3
Follow-up of AGC cases reveals that high-grade lesions are detected in 10% to 40% of cases, and these are more frequently of squamous origin (high-grade squamous intraepithelial lesion [HSIL]/CIN2-3) than glandular.4 In the United States, cervical glandular cancers continue to pose a significant challenge to cervical cancer prevention efforts. This is especially concerning, as the incidence of cervical squamous cell carcinoma (SCC) has seen a significant decline since the introduction of liquid-based cytology. Moreover, the growing incidence of glandular cancers raises questions about the feasibility and benefits of cytologic screening for endometrial adenocarcinoma, which remains a topic of debate.4
The management of AGC is different from that of squamous lesions. Based on the latest guidelines from the American Society for Colposcopy and Cervical Pathology,5 the recommended management for women diagnosed with any subcategory of AGC and adenocarcinoma in situ (AIS), excluding cases of atypical endometrial cells, entails undergoing colposcopy with endocervical sampling. This recommendation applies irrespective of human papillomavirus (HPV) test results. Additionally, the guidelines advise that women who are 35 years or older should undergo endometrial sampling. For women <35 years of age, the same procedure is recommended if they exhibit clinical indications that suggest a potential risk for endometrial neoplasia. Such indications may include factors like obesity, unexplained vaginal bleeding, and chronic anovulation.
In this study, we conducted a comprehensive review and analysis of liquid-based preparations (ThinPrep Pap smears) over a 5-year period. Our investigation included an assessment of various aspects, including the prevalence of AGC among our patient population, cytomorphology subcategories, outcomes of histological follow-up, significant AGC rate, and cytopathologist performance by years of experience.
METHODS
A retrospective review of all ThinPrep Pap smears was conducted in a multihospital single institution for a 5-year period from May 1, 2016 to April 30, 2021. The liquid-based preparations (ThinPrep) were prepared in accordance with the manufacturer’s specifications and stained using the Pap technique. All cases featuring AGC, including AGC-NOS, AGC-EM, AGC-EC, and AGC-FN, were retrieved for analysis. In cases where coexisting squamous abnormalities were identified alongside AGC on the same Pap test, both abnormalities were reported.
Histopathologic follow-up data were collected from pathology reports of cervical biopsy, diagnostic excision, endocervical curettage, endometrial biopsy, and hysterectomy. Surgical pathology reports for diagnosis of extrauterine tumors were obtained if any. We did not set up a follow-up timeframe, because previous reports showed that AGC is related to a persistently high risk of cervical cancer, particularly adenocarcinoma, for up to 15.5 years.6
The follow-up histopathologic diagnoses encompassed both significant histologies and insignificant histologies. AGC cases with significant histology follow-ups were categorized as “significant AGC.” AGC cases with insignificant histology follow-ups were categorized as “insignificant AGC.” The “significant AGC rate” was calculated as the number of significant AGC cases divided by total Pap smears. The “insignificant AGC rate” was calculated as the number of insignificant AGC cases divided by total Pap smears.
Significant histologies included high-grade squamous intraepithelial lesion (HSIL, CIN2-3), cervical SCC, endocervical AIS, invasive cervical adenocarcinomas (AC), endometrial adenocarcinoma (EAC), endometrial hyperplasia with atypia/endometrial intraepithelial neoplasia (EIN), endometrial hyperplasia without atypia (EH), and other neoplasms. Insignificant histologies included low-grade squamous intraepithelial lesion (LSIL, CIN1) and benign conditions, such as benign cervical (or endometrial) mucosa, cervicitis, and squamous metaplasia. The HPV status of AGC was classified as HPV positive, HPV negative, or not tested, according to the results of high-risk HPV genotypes, i.e., HPV 16, 18, and other high-risk HPV.
Furthermore, we monitored the significant AGC based on the years of experience of the cytopathologists at the time of the cytomorphology diagnosis, categorizing them as either junior (≤5 years) or senior (>5 years)7 in practice. We calculated the following main parameters: AGC report rate (AGC/Pap smears for each cytopathologist), insignificant AGC rate (insignificant AGC/Pap smears for each cytopathologist), and significant AGC rate (significant AGC/Pap smears for each cytopathologist).
Statistical analysis was conducted using GraphPad Prism software, employing the paired or unpaired t test and one-way analysis of variance test. Tukey’s honestly significant difference test was used for post hoc analysis as appropriate.8 A P value < 0.05 was considered as statistical significance.
RESULTS
AGC prevalence, significant AGC rate, and insignificant AGC rate
Over the 60-month study period spanning 5 fiscal years, our institution reported a total of 159,649 Pap tests. Among these, 395 cases (0.25%) exhibited AGC. Out of these 395 patients, 330 cases (83.5%) underwent follow-up histopathology. The prevalence of AGCs in each fiscal year, as well as significant and insignificant AGC rates, is presented in Figure 1. The overall 5-year insignificant AGC rate was 0.12%, while the overall 5-year significant AGC rate was 0.08%. Statistically significant differences were observed in the mean case numbers between significant and insignificant AGC (paired t test, P < 0.05). Significant differences were also present in the means of AGC rates between significant and insignificant AGC across the 5 years (paired t test, P < 0.05).
Figure 1.
AGC prevalence, significant AGC rate, and insignificant rate during 5 years (May 2016–April 2021) at our institution.
Cytomorphological, histological subcategories, and HPV genotyping
Out of the 330 cases with histological follow-up, AGC-NOS was the most common interpretation (43.9%; Figure 2a), followed by AGC-EC (33.3%), AGC-EM (13.0%), and AGC-FN (7.6%). A small percentage of cases presented coexisting HSIL or ASC-H (1.8%), and one case exhibited both AIS and AGC-EM.
Figure 2.
(a) Cytomorphologic classification and (b) histology follow-up results in 330 AGC cases.
Histologic follow-up data revealed that 188 cases (57.0%) showed CIN1 and benign histopathologic changes. In contrast, 65 cases (19.7%) were diagnosed with HSIL and SCC. The remaining histologies accounted for approximately 23.3% of the 330 cases. These included 46 cases (13.9%) of EAC/EIN/EH, 18 cases (5.5%) of AC/AIS, as well as 13 cases (3.9%) of other neoplasms, such as endometrial carcinosarcoma (Figure 2b).
Among the 330 AGC cases, 121 (36.7%) cases tested positive for high-risk HPV, while 155 (47.0%) tested negative. HPV testing was not conducted in 54 (16.3%) patients. Importantly, no statistical differences were observed when comparing the mean case numbers of patients with HPV-positive status to those with HPV-negative status across fiscal years (paired t test, P > 0.05).
Age and histopathology follow-up intervals
In this cohort, the average age of women diagnosed with AGC was 45.9 years, with a range spanning from 21 to 88 years. The median age was 45 years. The mean ages at AGC diagnosis for women with SCC/HSIL, AC/AIS, EAC/EIN/EH, other neoplasms, and CIN1 and benign were 39.4, 42.1, 60.5, 59.5, and 44.0 years, respectively, with statistically significant differences observed (one-way ANOVA, F (4, 328) = 23.64, P < 0.0001). Tukey’s honestly significant difference test for multiple comparisons found that the means of age were significantly different in CIN1 and benign vs. SCC/HSIL (P = 0.04), CIN1 and benign vs. other neoplasms (P = 0.0006), non-neoplasm vs. EAC/EIN/EH (P < 0.0001), SCC/HSIL vs. other neoplasms (P < 0.0001), SCC/HSIL vs. EAC/EIN/EH (P < 0.0001), other neoplasms vs. AC/AIS (P = 0.002), and EAC/EIN/EH vs. AC/AIS (P < 0.0001). No statistically significant difference in the mean ages was revealed in CIN1 and benign vs. AC/AIS, SCC/HSIL vs. AC/AIS, and other neoplasms vs. EAC/EIN/EH.
The histology follow-up intervals ranged from 0 months (concurrent histology and cytopathology) to 25 months. The average histology follow-up interval was 1.9 months, with a median follow-up interval of 1 month. Out of 330 cases with histological follow-up, 303 (91.8%) occurred within 3 months, 6 cases within 4 to 6 months, and 14 cases within 7 to 12 months. Only 7 cases (2.1%) had follow-up histology findings more than 12 months later, but 4 of them had follow-up Pap smears within 12 months.
Out of the 330 cases, 7.9% (26 cases) underwent a loop electrosurgical excision procedure, 5.5% (18 cases) had conization, and 12.7% (42 cases) had hysterectomy as their follow-up methods for obtaining the final histopathology results. The remaining cases (73.9%) had biopsy/curettage for their histology follow-up.
Effect of lockdown during the COVID-19 pandemic
In response to the COVID-19 pandemic, Texas implemented a lockdown period in March and April 2020, with a gradual easing of health restrictions thereafter. This lockdown had a significant impact on the overall volume of Pap smears conducted. Compared with the corresponding months in 2019, there was a slight decrease in case volumes in March (2191 vs. 2672) and May 2020 (2304 vs. 2647). However, in April 2020 there was a dramatic decline, with only 697 cases, representing approximately a 77% decrease compared to 3028 cases in April 2019. In the period from March 2020 to May 2020, 10 cases of AGC were diagnosed, while 15 cases were diagnosed in the same period of 2019. All histological outcomes of AGCs diagnosed between March 2020 and May 2020, as well as those diagnosed between March 2019 and May 2019, were insignificant. Given the very low number of AGC cases, the insignificant AGC rates between these two time periods were not compared.
Effect of cytopathologists’ experience in practice
There were 274 AGC cases diagnosed by senior cytopathologists, compared with 56 AGC cases diagnosed by junior cytopathologists. Significant differences were observed in the mean AGC report rate between junior and senior cytopathologists (junior vs. senior: 1.00% vs. 1.72%, P < 0.05, Figure 3), indicating that senior cytopathologists were more likely (or confident) to report AGC. However, there were no statistically significant differences in the insignificant AGC rate or significant AGC rate between these two groups (unpaired t test, P > 0.05).
Figure 3.
Effect of year of practice on AGC diagnosis.
DISCUSSION
We conducted a retrospective analysis of AGC cases in our institution over a 5-year period. The prevalence of AGC was 0.25%, and 83.5% of cases had corresponding histopathology diagnoses. The overall 5-year insignificant AGC rate was 0.12%. The most frequently observed AGC cytomorphology was AGC-NOS, followed by AGC-EC (33.3%). The predominant histological finding (57.0%) was CIN1 and benign conditions, such as benign mucosa with reactive changes. AC/AIS, EAC/EIN/EH, and other neoplasms accounted for 23.3% of cases, while SCC/HSIL represented 19.7%. Among AGC cases, 36.7% tested positive for high-risk HPV. In this cohort, the mean age of women with AGC was 45.9 years. The average follow-up interval was 1.9 months.
Our cohort study has yielded findings aligned with previous research. The overall insignificant AGC rate in our cohort was 0.12%. Various overall insignificant rates, such as 0.03%9 and 0.28%,10 could be calculated based on data in the literature. In our cohort, 36.4% of the AGC cases tested positive for high-risk HPV. The HPV-positive rate in AGC varies among previous studies, with rates of 1.3%2 and 41%11 reported. These variations likely depend on the institutions and sample sizes.
AGC morphology and the follow-up histological results exhibit variations across institutions. For instance, in a Mayo Clinic study,2 0.2% of 123,763 liquid-based cervical cytological Pap smears were classified as AGC, with a mean age of 54. Among the 246 patients with histology follow-up, AGC-EM was observed in 60 (24.4%) cases, AGC-EC in 36 (14.6%), AGC-FN in 28 (11.4%), and AGC-NOS in 122 (49.6%). Notably, 80 out of 246 patients had a clinically significant lesion, with 48 (60%) cases histologically diagnosed as endometrial adenocarcinoma, and 9 (11%) patients exhibited severe squamous dysplasia (CIN 3). Invasive AC and AIS each accounted for 3 cases (4%). Interestingly, no cases of SCC were diagnosed in this cohort. Patients with neoplasia were significantly older than those without neoplasia (mean age 59.6 vs. 49.2 years). In another study conducted at the University of Pittsburgh Medical Center, an analysis of 662 AGC cases revealed that AGC-EM, AGC-EC, and AGC-NOS accounted for 30.3%, 42.0%, and 25.7%, respectively.12 The histologic outcomes indicated that 8.3% were cervical, 6.3% were endometrial, and 0.6% were ovarian precancerous or malignancies.
The COVID-19 lockdown in Texas from March 2020 to May 2020 significantly decreased our Pap smear volumes in April 2020 compared to the corresponding period in April 2019. This observation aligns with other reports.13 However, it’s noteworthy that the histology outcomes of AGCs diagnosed in March 2020 to May 2020 (10 cases) and March 2019 to May 2019 (15 cases) all indicated insignificant conditions, with no change in the malignancy rate, despite the small sample size in our cohort during the lockdown period. The potential impact of the COVID-19 pandemic and lockdown on the AGC diagnosis and prognosis remains unclear. More data and multicenter studies14 may provide more information.
We also assessed the impact of cytopathologists’ years of experience on AGC diagnoses. In our institute, senior cytopathologists tended to diagnose more cases of AGC, suggesting that the diagnosis of AGC may depend on the cytopathologist’s skills and experience. However, the significant and insignificant AGC rates were similar between our junior and senior cytopathologists.
There are some limitations in this study. First, definitive diagnoses based on specimens from hysterectomy and conization were not available for all of our patients. In our cohort, only 5.5% of patients had conization and 12.7% had hysterectomy for the histology follow-up. However, a previous study on the outcomes of AGC showed that the detection rate of malignant or premalignant lesions can be as high as 87.8% after conization or total hysterectomy, demonstrating the importance of follow-up methods for AGC.15 Second, due to the nature of follow-up biopsy or curettage, it is possible that lesions could be missed due to sampling errors. Third, the sample size is relatively small for cytopathologists in the analysis of cytopathologists’ years of experience with AGC diagnoses. Lastly, we did not conduct interobserver assessments since this was a retrospective study. Other studies have indicated that interobserver agreement is generally poor, particularly in the AGC category, and there is a slight correlation with years of experience.16 Interobserver variability may be a potential explanation for the discordance between cytopathology diagnosis and final histology.
CONCLUSIONS
The AGC diagnostic rate at our institution falls within the CAP report range. A significant number of cases had follow-up histologic results available. The overall 5-year insignificant AGC rate was 0.12%. The majority of AGCs were associated with insignificant histology. Interestingly, senior cytopathologists tended to report AGC, but there was no significant difference in the significant AGC rate or insignificant AGC rate between junior and senior cytopathologists.
Disclosure statement/Funding
The authors report no funding or conflicts of interest.
References
- 1.Nayar R, Wilbur DC.. The Bethesda System for Reporting Cervical Cytology: Definitions, Criteria, and Explanatory Notes. Springer; 2015. https://play.google.com/store/books/details?id=JPJICAAAQBAJ. [Google Scholar]
- 2.Reynolds JP, Salih ZT, Smith AL, Dairi M, Kigen OJ, Nassar A.. Cytologic parameters predicting neoplasia in Papanicolaou smears with atypical glandular cells and histologic follow-up: a single-institution experience. J Am Soc Cytopathol. 2018;7(1):7–15. doi: 10.1016/j.jasc.2017.08.001. [DOI] [PubMed] [Google Scholar]
- 3.Wang T, Zhang H, Liu Y, Zhao C.. Updates in cervical cancer screening guidelines, the Bethesda system for reporting cervical cytology, and clinical management recommendations. J Clin Transl Pathol. 2023;3(2):75–83. doi: 10.14218/jctp.2023.00004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zhao C, Florea A, Austin RM.. Clinical utility of adjunctive high-risk human papillomavirus DNA testing in women with Papanicolaou test findings of atypical glandular cells. Arch Pathol Lab Med. 2010;134(1):103–108. doi: 10.5858/2008-0755-OAR1.1. [DOI] [PubMed] [Google Scholar]
- 5.Perkins RB, Guido RL, Saraiya M, et al. Summary of current guidelines for cervical cancer screening and management of abnormal test results: 2016–2020. J Womens Health (Larchmt). 2021;30(1):5–13. doi: 10.1089/jwh.2020.8918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wang J, Andrae B, Sundström K, et al. Risk of invasive cervical cancer after atypical glandular cells in cervical screening: nationwide cohort study. BMJ. 2016;352:i276. doi: 10.1136/bmj.i276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wang L, Leng B, Rampisela D.. Five-year review of intraoperative pathology consultation in a single institution. Proc (Bayl Univ Med Cent). 2021;34(6):649–653. doi: 10.1080/08998280.2021.1941700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wang T, Wang B, Chen H.. Menthol facilitates the intravenous self-administration of nicotine in rats. Front Behav Neurosci. 2014;8:437. doi: 10.3389/fnbeh.2014.00437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sung JA, Nikas IP, Kim H, Ryu HS, Lee C.. Diagnostic distribution and pitfalls of glandular abnormalities in cervical cytology: a 25-year single-center study. J Pathol Transl Med. 2022;56(6):354–360. doi: 10.4132/jptm.2022.09.05. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Keles E, Ozturk UK, Alınca CM, Giray B, Kabaca C, Cetiner H.. Factors affecting the histopathological outcomes of atypical glandular cells on Pap test. J Cytol. 2021;38(4):210–215. doi: 10.4103/JOC.JOC_54_21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Xiao J, Gong Y, Chen F, Chen L, Xie Y, Sui L.. Clinical diagnostic value of atypical glandular cells in cervical cytology: a single center experience from China. J Low Genit Tract Dis. 2023;27(3):202–206. doi: 10.1097/LGT.0000000000000730. [DOI] [PubMed] [Google Scholar]
- 12.Zhao C, Florea A, Onisko A, Austin RM.. Histologic follow-up results in 662 patients with Pap test findings of atypical glandular cells: results from a large academic womens hospital laboratory employing sensitive screening methods. Gynecol Oncol. 2009;114(3):383–389. doi: 10.1016/j.ygyno.2009.05.019. [DOI] [PubMed] [Google Scholar]
- 13.Vigliar E, Iaccarino A, Bruzzese D, Malapelle U, Bellevicine C, Troncone G.. Cytology in the time of coronavirus disease (COVID-19): an Italian perspective. J Clin Pathol. 2021;74(4):261–263. doi: 10.1136/jclinpath-2020-206614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Vigliar E, Pisapia P, Dello Iacovo F, et al. COVID-19 pandemic impact on cytopathology practice in the post-lockdown period: an international, multicenter study. Cancer Cytopathol. 2022;130(5):344–351. doi: 10.1002/cncy.22547. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shoji T, Takatori E, Takeuchi S, et al. Clinical significance of atypical glandular cells in the Bethesda system 2001: a comparison with the histopathological diagnosis of surgically resected specimens. Cancer Invest. 2014;32(4):105–109. doi: 10.3109/07357907.2014.880453. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Simsir A, Hwang S, Cangiarella J, et al. Glandular cell atypia on Papanicolaou smears: interobserver variability in the diagnosis and prediction of cell of origin. Cancer. 2003;99(6):323–330. doi: 10.1002/cncr.11826. [DOI] [PubMed] [Google Scholar]