https://orcid.org/0000-0003-3778-2571
Figures
Abstract
Objective
To validate the colposcopy indication proposed by the 2019 ASCCP Risk-Based Management Consensus Guidelines for abnormal cervical cancer screening tests (the 2019 ASCCP guidelines).
Methods
Clinical data of 1404 patients who underwent colposcopy in single center in China were reviewed. Based on history and current cervical screening (HPV & cytology), corresponding recommendations were given according to the 2019 ASCCP guidelines. The agreement and discrepancy of colposcopy indication were analyzed between the Chinese consensus and the 2019 ASCCP guidelines.
Results
Colposcopy indication was matched in about 80% patients. The left 20% were recommended with follow-up by the 2019 ASCCP guidelines. The discrepancy mainly focused on patients having a current result of HPV-positive NILM without unknown history. The ratio of observed CIN3+ in our database over estimated CIN3+ by the 2019 ASCCP guidelines was 6.2 (31/5). The ratio was even higher in patients with HPV16/18-positive NILM (7, 28/4), compared with those with other types of high-risk HPV-positive NILM (3, 3/1). The 2019 ASCCP guidelines had a relatively high sensitivity (83.1%), a low specificity (21.5%), a low positive predictive value (14.1%) and a high negative predictive value (89.1%) for prediction of CIN 3+.
Conclusions
We could try to apply the 2019 ASCCP guidelines in Chinese population. The classification of HR-HPV was strongly recommended during risk assessment. For patients with HPV16/18 infection, colposcopy should be recommended. Perspective multi-center randomized controlled trial with reliable follow-up should be performed in the future to confirm the feasibility.
Citation: Gui T, Chen Z, Chen F (2021) Validation of the indication for colposcopy proposed by the 2019 ASCCP risk-based management consensus guidelines: A single-center study in China. PLoS ONE 16(7): e0253493. https://doi.org/10.1371/journal.pone.0253493
Editor: Zhi-Ming Zheng, National Cancer Institute, UNITED STATES
Received: February 5, 2021; Accepted: June 6, 2021; Published: July 1, 2021
Copyright: © 2021 Gui et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper.
Funding: This study was financially supported by Innovative Medicine Program of Chinese Academy of Medical Sciences (No. 2017-I2M-1-002) and National Clinical Research Program for Obstetric & Gynecologic Diseases (No. FL2021). The funders played a role in study design.
Competing interests: The authors declare that they have no competing interests.
Introduction
Cervical cancer is the most common malignancy among women in China. The incidence rate of cervical cancer is decreasing in recent years largely due to effective screening. When abnormalities are identified during screening, colposcopy will be adopted to further confirm any cervical lesions. Colposcopy plays a role of connecting bridge between screening and final treatment. The number of unnecessary blinded biopsies and conization has been reduced substantially due to the application of colposcopy. However, colposcopy examination seems to be overused in recent years in China. Appropriate application of colposcopy has great clinical and economic significance. From a public health perspective, reduction in the number of unnecessary invasive procedures will definitely be beneficial for both patients and the society as a whole.
The screening guidelines for cervical cancer continue to be reevaluated and updated, with the overall goal of improving diagnosis and economizing time and resources. The original consensus guidelines was proposed by American Society of Colposcopy and Cervical Pathology (ASCCP) in 2001 [1], and was subsequently updated in 2006 [2] and 2012 [3]. The 2019 ASCCP guidelines are the fourth version for management of cervical cancer screening abnormalities [4]. The 2012 consensus guidelines introduced the principle of “equal management for equal risk”, which was a conceptual breakthrough. Specifically, patients with similar risks for pre-cancer lesions or cancer could be managed similarly, in despite of the variety of screening time points and methods [5]. With a more in-depth understanding of how previous results affect the risk, assessement, the key difference between 2019 ASCCP guidelines and previous versions was the change from “test results-based algorithms” to “algorithms based on quantitative risk estimate values”, officially defined as clinical action thresholds (CATs) [4]. For instance, colposcopy is recommended for patients with HPV-positive and cytology result ≥ atypical squamous cells of undefined significance [ASC-US] by previous guidelines. According to the 2019 ASCCP guidelines, colposcopy should be recommended to patients whose screening results (considering both previous and current) yielded a > 4.0% probability of identifying cervical intraepithelial neoplasm 3+ (CIN3+). Risk estimates used in 2019 ASCCP guidelines were generated from a prospective longitudinal cohort of over 1.5 million patients with over ten-year follow-up at Kaiser Permanente Northern California (KPNC) [6]. The feasibility of these risk estimates in other regions in United States has been validated in other data sets from several screening programs and clinical trials [7]. However, there is no relative data in Asian countries including China.
Therefore, the aims of the present study were to 1) compare the indication for colposcopy in Chinese consensus and the clinical action threshold for colposcopy adopted by the 2019 ASCCP guidelines; 2) assess the applicability of the clinical action threshold for colposcopy adopted by the 2019 ASCCP guidelines in China.
Materials and methods
Study design
We retrospectively reviewed the medical records of women who underwent colposcopic examination and cervical biopsy at the colposcopy clinic of Peking Union Medical College Hospital (PUMCH) in China between June 2018 and Jan 2020, and then further assessed the applicability of the clinical action threshold for colposcopy adopted by the 2019 ASCCP guidelines in our Chinese database.
Subjects
Women included in our research had no history of pelvic radiation or hysterectomy, no sexual intercourse for 3 days before the examination, and no confirmed or clinically suspected immunosuppression or other chronic disease that might have compromised their immune system.
Patients’ information possibly related to cervical intraepithelial lesions were collected, including age, gravidity, parity, number of sex partner, age of first sex intercourse, menstrual status, method of contraception, clinical manifestations including vaginal bleeding and increased vaginal discharge, cervical surgery history, types of transformation zone and the screening results of HPV and cytology.
All data in this study were collected from the hospital’s archived database. This study did not influence the diagnosis or treatment of the patients. This study was approved by the ethics committee of PUMCH. Since all data were deidentified, written informed consent was not necessary due to the retrospective nature of the study.
Colposcopy indication after abnormal screening in Chinese consensus
- Continuous high-risk HPV infection + NILM (annual visit >2 consecutive years)
- HPV-16/18 infection + NILM
- HPV-positive +ASC-US
- Abnormal cytological test results greater than ASC-US including LSIL, ASC-H, HSIL, ASC, AGC, invasive cervical cancer
Clinical action threshold leading to recommendation of colposcopy in 2019 ASCCP risk-based management consensus guidelines [4]
Guidelines: Colposcopy is recommended to patients having an immediate risk of CIN3+ of 4.0% or greater, considering both history and current results.
Rationale: Among patients referred directly to colposcopy, the immediate risk of CIN 3+ was reported to range from 3%-7% [8–11]. The clinical action threshold for colposcopy referral, a 4% immediate risk of CIN3+, was proposed after balancing the benefits and harms. For patients without screening history, only those with HPV-positive and ASC-US or LSIL cytology would be referred to colposcopy, while patients with HPV-positive and NILM cytology would not.
Colposcopy and biopsy procedure
All colposcopies were performed by two expert colposcopists and were recorded by using the VIZ-YD system. This was a video exoscope-based system (optical electronic integration colposcopy, Beijing SWSY technology Co., Ltd, China), allowing full high density video documentation of the colposcopic examination process.
Steps in colposcopic assessment of the cervix included the following: 1) clean the cervix with normal saline; 2) assess the cervix about 1 min after the application of 5% acetic acid, and observe margins of the lesion, epithelial color and vascular patterns; and 3) assess the cervix after the application of diluted Lugol’s iodine solution, and observe iodine staining.
All colposcopically detected abnormal areas were biopsied. If the colposcopic examination found no cervix lesions, biopsy specimens would be obtained randomly at the squamo-columnar junction in four quadrants at 3, 6, 9, or 12 o’clock. An endocervical curettage was performed after the cervical biopsy. All biopsy specimens were fixed in formaldehyde and embedded in paraffin routinely.
Statistical analyses
Continuous variables were recorded as mean±standard deviation (SD) if normally distributed, or median with interquartile range if not normally distributed. Categorical variables were expressed in terms of frequency and percentage. Continuous variables were compared by using the student’s t test. Frequencies were compared using the χ2 test or Fisher’s exact test. Univariate analysis was used to analyze the effect of each variable on susceptibility of CIN3+, and variables with p < 0.05 in univariate analysis were adopted for further multivariate analysis. The diagnostic performances were analyzed and compared in terms of sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). The receiver operating characteristic (ROC) curves and the areas under the ROC curves (AUC) were calculated by the 2×2 contingency table and chi-square test, and McNemar test was used to compare the differences in performance. All data were analyzed by using SPSS 23.0 version software (IBM, USA), and p<0.05 was considered statistically significant.
Results
Flowchart of patients (Fig 1)
From Jun 2018 to Dec 2019, a total of 1927 patients underwent colposcopy in PUMCH in China. During the process of data preparation, we found 159 patients did not have complete HPV & cytology data record (current and history), 294 patients missed pathological results, 16 patients were diagnosed with cervical cancer and underwent radical surgery or radiation, and 54 patients received hysterectomy for other diseases. Finally, 1404 patients were included in our research, among whom 1215 patients were diagnosed as inflammation or CIN 1 or CIN 2, and 189 patients were diagnosed as CIN3+ (CIN 3 or adenocarcinoma in situ of cervicx (AIS) or invasive cervical cancer).
Abbreviations: HPV, human papillomavirus; TCT, liquid-based cytology test; CIN, cervical intraepithelial neoplasia; CIN3+, CIN3/AIS/CC; AIS, adenocarcinoma in situ; CC, invasive cervical cancer.
Basic characteristics of patients (Table 1)
Patients included in our research were classified into two groups according to the final pathological diagnosis after colposcopy examination and biopsy, the < CIN3 group and the ≥ CIN3+ group.
Our research showed that CIN3+ rates were higher in patients who had more sex partners and had their first sexual intercourse in younger age, with p = 0.046 and p = 0.014, respectively. CIN3+ rates were also higher in patients with HPV infection, especially HPV16/18 infection (p < 0.001, for both previous and current infection), and with higher grade of cytology result (p = 0.05 and p < 0.001 for previous and current cytology status, respectively).
However, statistical analyses showed no significant differences between the two groups in the following terms including the patients’ age, the gravidity, the parity, the menstruation status, the contraception method, and the cervical surgery history. In addition, no significant differences were observed regarding the clinical manifestations and the status of transformation zone of cervix.
Univariate analyses of high-risk factors for CIN 3+ (Table 2)
Univariate analyses showed that patients with present HPV 16/18 infection had a nearly threefold increased risk of CIN3+ compared with the patients with a negative HPV status (OR = 2.73, 95%CI [1.07–6.97]). Furthermore, patients with present cytology of ASC-H and HSIL increased the CIN3+ risk by nearly eight-fold (OR = 7.81, 95%CI [4.02–15.17]) and fifteen-fold (OR = 14.58, 95%CI [8.72–24.38]), compared with patients who had normal cytology screening test.
Multivariate analyses of high-risk factors for CIN 3+ (Table 3)
All variates with P < 0.05 in the univariate analyses were included in the multivariate regression analysis. Patients, who were infected with other types of high-risk HPV(HR-HPV) and HPV 16/18, showed an increased risk for CIN3+ (OR = 5.81,0R = 16.93, respectively), although both of which were not statistically significant. After adjusting for age, gravidity, parity, number of sex partner, menstruation status and contraception method, patients with current cytology of ASC-H and HSIL had a twenty-fourfold (adjusted OR = 23.90, 95%CI [5.50–103.92]) and approximate ten-fold (adjusted OR = 10.32, 95%CI [4.10–25.99]) risk for CIN3+, compared with patients who had a normal cytology test.
Risk estimates and management recommendation according to the 2019 guidelines (Tables 4 and 5)
In our study, a total of 1404 patients received colposcopy examination due to cervical screening abnormality in PUMCH in accordance with the Chinese consensus. According to the risk estimate tables supporting the 2019 ASCCP guidelines [6] and the management recommendation flow [4, Fig 1], our research found that the immediate CIN3+ risk and the 5-year CIN3+ risk were both significantly higher in the ≥ CIN3+ group compared with the < CIN 3 group (p<0.001 both). Specifically, 1 of 1404 (0.07%) patient was recommended 5-y regular screening, 1 of 1404 (0.07%) patient was recommended 3-y regular screening, 291 of 1404 (20.66%) patients were recommended 1-y regular screening, and 1111 of 1404 (79.20%) patients were recommended colposcopy/immediate treatment. Among the 293 patients recommended for follow-up, 261 (89.07%) patients were graded lower than CIN3 (166 inflammation, 47 CIN 1, 48 CIN 2), while 32 (10.93%) patients were graded CIN3+ (26 CIN3, 6 AIS/cancer).
Comparison of management recommendation between Chinese consensus and 2019 ASCCP guidelines (Tables 6 and 7)
About 80% of the patients admitted to colposcopy clinic in PUMCH in China were also recommended colposcopy procedure according to the 2019 ASCCP guidelines, while the left 20% were recommended follow-up, among whom 32 patients were graded as CIN 3+ accounting for 2.3% of the total study population.
The discrepancy mainly focused on one scenario in Table 1A provided by Egemen et al [6], in which patients had a current test result of HPV-positive & NILM and an unknown/undocumented history, with 1-year follow-up recommendation on the basis of immediate CIN3+ risk of 2.1% (< 4.0%) under this circumstance. The ratio of CIN3+ patients observed in PUMCH database over CIN3+ patients expected according to the immediate CIN3+ risk (O/E) in this scenario was 6.2 (31/5). When the HPV genotypes of this subset of patients were further categorized to HPV16/18 and 12 other genotypes of high-risk HPV (other HR-HPV), the CIN3+ ratio was even higher in patients with the result of HPV16/18-positive NILM (O/E = 7, 28/4), compared with patients with the result of other HR-HPV-positive NILM (O/E = 3, 3/1).
Accuracy of management recommendation by the 2019 ASCCP guidelines in predicting Cervical Intraepithelial Neoplasia (CIN) (Table 8)
The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for prediction of CIN3+ were shown in Table 8. The 2019 ASCCP guidelines had a relatively high sensitivity (83.1%, 95% CI [77.0%-88.1%]) and a low specificity (21.5%, 95% CI [19.2%-23.9%]) for prediction of CIN 3+. Besides, our statistical analyses showed a low PPV (14.1%, 95% CI [12.1%-16.3%]) and a high NPV (89.1%, 95% CI [84.9%-92.4%]) for prediction of CIN 3+.
In addition, we also analyzed the sensitivity, specificity, PPV and NPV of 2019 ASCCP guidelines in differentiating a cervical lesion of CIN2/3 from CIN 1. Table 8 showed that the sensitivity was 10.5% (CI [7.7%-13.9%]), the specificity was 90.1% (CI [88.0%-91.9%]), the PPV was 31.2% (CI [23.6%-39.6%]), and the NPV was 70.2% (CI [67.6%-72.8%]).
Discussion
The 2019 ASCCP guidelines highlight that detection and treatment of pre-cancer lesions still remains the main aim of cervical cancer prevention, and this version of guidelines comprehensively use and expand upon the principle of “equal management for equal risks”.
The 2019 ASCCP guidelines take both the current and the previous screening tests into consideration, and make recommendations based on immediate CIN3+ risk which is the probability of patients currently having CIN3+, and 5-year CIN3+ risk which is the probability of developing CIN3+ over the ensuing 5 years. It is worth noting that, CIN 3+, instead of CIN 2+, is chosen as main clinical endpoint for risk estimates, because CIN 3+ is more pathologically reproducible [12], and has a more similar HPV-type distribution to that of invasive cervical cancers [13], while CIN 2 has a greater tendency for regression even without any treatment [14–16].
Statisticians have conducted extensive data analyses effort to produce risk estimates for all combinations of screening tests, mainly based on Kaiser Permanente Northern California (KPNC) database, which was the largest and most comprehensive data set in the United States [6]. Three additional databases were also analyzed to ensure that results would be applicable to patients of diverse racial, ethnic, and socioeconomic strata, including the Onclarity HPV trial [17, 18], the New Mexico State HPV Pap Registry [19, 20], and the US Centers for Disease Control and Prevention’s National Breast and Cervical Cancer Early Detection Program [21].
To date, our research is the first study of investigating the applicability of risk-based recommendation for colposcopy in China, by comparing the colposcopy indication between Chinese consensus and 2019 ASCCP guidelines.
In our study, 1404 patients were included in our research. Patients were classified into two groups, the < CIN3 group (1215 (86.5%)) and the ≥ CIN3+ group (189 (13.5%)). Patients who had more sex partners and had their first sexual intercourse in younger age were more susceptible to CIN3+, concordant with previous published studies. Final multivariate analyses showed patients with current cytology of ASC-H and HSIL had approximate twenty-six fold and ten fold risk for CIN3+ with statistical significance, compared with patients who had a normal cytology test. However, the HPV status (previous and present) and the previous cytology result were not independent risk factors for CIN 3+ statistically. This result could probably be accounted by the selection bias of our research population. Patients selected in our research all had underwent colposcopy procedure, with an overwhelming majority of HPV positive and such a tiny part of HPV negative (32 vs. 3 or 127 vs. 5). Besides, it is widespread accepted that HPV test has relatively higher sensitivity and negative predictive value, while cytology test has relatively higher specificity and positive predictive value.
We further analyzed the applicability of the 2019 ASCCP guidelines for predicting cervical lesions in China. Our results showed that the agreement of colposcopy indication was perfectly matched in about 80% of the patients, suggesting that the clinical action threshold for colposcopy adopted in 2019 ASCCP guidelines was applicable in China in general. Nevertheless, more attention should be paid to the left 20% who received colposcopy examination in PUMCH in China while the 2019 ASCCP guidelines recommended follow-up. Among the 20% patients, 32 patients were graded as CIN 3+ whom might be misdiagnosed subsequently, accounting for about 2.3% of the total study population. The discrepancy mainly focused on patients having a current test result of HPV-positive NILM and an unknown history, with 1-year follow-up recommendation on the basis of immediate CIN3+ risk of 2.1% (< 4.0%). Here we introduced a ratio O/E, that was the number of actually observed CIN3+ patients after final pathological diagnosis over the number of expected CIN3+ calculated by the total number of patients × immediate CIN 3+ risk 2.1%. In our study, the ratio O/E was 6.2 (31/5), suggesting that CIN3+ could be probably misdiagnosed if 1-year follow-up recommendation was given according to the 2019 ASCCP guidelines. Additionally, when the HPV types were further categorized to HPV16/18 and others HR-HPV, the O/E ratio was even higher in patients with HPV16/18-positive NILM, compared with patients with others HR-HPV-positive NILM. Therefore, the classification of HR-HPV was strongly recommended during the risk assessment. For patients with current HPV-positive NILM without prior documented HPV status, 1-year follow-up could be recommended for patients with others HR-HPV infection, but colposcopy should be recommended for patients with HPV16/18 infection.
In the present study, we also found that the 2019 ASCCP guidelines had a high sensitivity (83.1%), capable of differentiating a cervical lesion of CIN 3+ from CIN 1/2; however, the specificity was low (21.5%). This would probably lead to over-estimation and thereby increase patients’ cost. In addition, the 2019 ASCCP guidelines had a low PPV of 14.1% and a high NPV of 89.1%, indicating that if the guidelines do not recommend colposcopy, the probability for a patient of developing CIN 3+ would be about 10%.
In cervical cancer screening, once screening abnormalities are identified, colposcopy would probably be recommended to confirmed any cervical lesions. The primary task for colposcopy examination is to improve the accuracy of diagnosing high-grade lesions and prevent misdiagnosis. Generally, an overestimated risk leads to an unnecessary colposcopy procedure and cervical biopsy, whereas an underestimated risk leads to missed diagnosis, possibly high-grade lesions or even carcinomas.
Sensitivity is the percentage of all patients who test positive. During risk assessment, most clinicians would prefer higher sensitivity, since the significant negative consequence of missing high grade cervical lesions are far more pressing, despite the potential for over-treatment. In addition, with a higher NPV, the likelihood that a test-negative person has no disease, screening-negative women are at a lower risk for developing cervical lesions over multiple years. A high sensitivity and NPV for colposcopy referral maximize the detection of cervical lesions and minimized the misdiagnosis.
In our research, the study population was all patients who had already been referred to colposcopy, and the final pathological diagnosis of high grade cervical lesions was used to evaluate the indication for colposcopy. It is worth noting that, the prevalence of disease in a population has a major impact on the specificity, the percentage of all people without certain disease who test negative. In a referral population, the higher prevalence of disease causes a decrease in specificity. This is due to the increased prevalence, which diminishes the denominator in the calculation for the specificity. Besides, in a referral population, as screening abnormalities are quite common and not all abnormalities would cause disease and have a high tendency for spontaneous regression, the PPV, the likelihood that a test-positive person having disease, would be relatively low.
Therefore, the sensitivity and NPV have greater significance than the specificity and PPV in cervical cancer screening.
Our study had several limitations. First, this research was a single-center study. Second, the selection bias was inevitable. Third, this research was a retrospective review without long-term follow-up. In future study, a perspective multi-center randomized controlled trial with reliable follow-up should be performed.
Conclusions
We could try to apply the 2019 ASCCP guidelines in Chinese population. The classification of HR-HPV was strongly recommended during risk assessment. For patients with current test result of HPV-positive NILM and without prior documented HPV status, 1-year follow-up could be recommended for patients with others HR-HPV infection, but colposcopy should be recommended for patients with HPV16/18 infection. Perspective multi-center randomized controlled trial with reliable follow-up should be performed in the future.
References
- 1. Wright TC, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ. 2001 consensus guidelines for the management of women with cervical cytological abnormalities. J Low Genit Tract Dis. 2002; 6: 127–43. pmid:17051012
- 2. Wright TC, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D, et al. 2006 consensus guidelines for the management of women with abnormal cervical cancer screening tests. Am J Obstet Gynecol. 2007; 197: 346–55. pmid:17904957
- 3. Massad LS, Einstein MH, Huh WK, Katki HA, Kinney WK, Schiffman M, et al. 2012 updated consensus guidelines for the management of abnormal cervical cancer screening tests and cancer precursors. Obstet Gynecol. 2013; 121: 829–46. pmid:23635684
- 4. Perkins RB, Guido RS, Castle PE, Chelmow D, Einstein MH, Garcia F, et al. 2019 ASCCP risk-based management consensus guidelines for abnormal cervical cancer screening tests and cancer precursors. J Low Genit Tract Dis. 2020; 24: 102–31. pmid:32243307
- 5. Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam SL, Cain J, et al. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer. CA Cancer J Clin. 2012; 62: 147–72. pmid:22422631
- 6. Egemen D, Cheung LC, Chen X, Demarco M, Perkins RB, Kinney W, et al. Risk estimates supporting the 2019 ASCCP risk-based management consensus guidelines. J Low Genit Tract Dis. 2020; 24: 132–43. pmid:32243308
- 7. Cheung LC, Egemen D, Chen X, Katki HA, Demarco M, Wiser AL, et al. A 2019 ASCCP risk-based management consensus guidelines: methods for risk estimation, recommended management, and validation. J Low Genit Tract Dis. 2020; 24: 90–101. pmid:32243306
- 8. Schiffman M, Solomon D. Finding to date from the ASCUS-LSIL triage study (ALTS). Arch Pathol Lab Med. 2003; 127: 946–9. pmid:12873166
- 9. Wright TC, Stoler MH, Behrens CM, Sharma A, Zhang G, Wright TL. Primary cervical cancer screening with human papillomavirus: end of study results from the ATHENA study using HPV as the first-line screening test. Gynecol Oncol. 2015; 136: 189–97. pmid:25579108
- 10. Stoler MH, Wright TC Jr, Sharma A, Apple R, Gutekunst K, Wright TL, et al. High-risk human papillomavirus testing in women with ASC-US cytology: results from the ATHENA HPV study. Am J Clin Pathol. 2011; 135: 468–75. pmid:21350104
- 11. Arbyn M, Roelens J, Simoens C, Buntinx F, Paraskevaidis E, Martin-Hirsch PP, et al. Human papillomavirus testing versus repeat cytology for triage of minor cytological cervical lesions. Cochrane Database Syst Rev. 2013; 3: CD008054. pmid:23543559
- 12. Carreon JD, Sherman ME, Guillen D, Solomon D, Herrero R, Jeronimo J, et al. CIN 2 is a much less reproducible and less valid diagnosis than CIN 3: results from a histological review of population-based cervical samples. Int J Gynecol Pathol. 2007; 26: 441–6. pmid:17885496
- 13. Wentzensen N, Wilson LE, Wheeler CM, Carreon JD, Gravitt P, Schiffman M, et al. Hierarchical clustering of human papillomavirus genotype patterns in the ASCUS-LSIL triage study. Cancer Res. 2010; 70: 8578–86. pmid:20959485
- 14. Silver MI, Gage JC, Schiffman M, Fetterman B, Poitras NE, Lorey T, et al. Clinical outcomes after conservative management of cervical intraepithelial neoplasia grade 2 (CIN 2) in women ages 21–39 years. Cancer Prev Res (Phila) 2018; 11:165–70.
- 15. Moscicki AB, Ma Y, Wibbelsman C, Darragh TM, Powers A, Farhat S, et al. Rate of and risks for regression of cervical intraepithelial neoplasia 2 in adolescents and young women. Obstet Gynecol. 2010; 116: 1373–80. pmid:21099605
- 16. Tainio K, Athanasiou A, Tikkinen KAO, Aaltonen R, Cardenas J, Hernandes , et al. Clinical course of untreated cervical intraepithelial neoplasia grade 2 under active surveillance: systematic review and meta-analysis. BMJ. 2018; 360: k499. pmid:29487049
- 17. Wright TC, Stoler MH, Behrens CM, Apple R, Derion T, Wright TL. The ATHENA human papillomavirus study: design, methods, and baseline results. Am J Obstet Gynecol. 2012; 206: 46. e1–11.
- 18. Stoler MH, Wright TC, Parvu V, Vaughan L, Yanson K, Eckert K, et al. The onclarity human papillomavirus trial: design, methods, and baseline results. Gynecol Oncol. 2018; 149: 498–505. pmid:29681462
- 19. Gage JC, Schiffman M, Hunt WC, Joste N, Ghosh A, Wentzensen N, et al. Cervical histopathology variability among laboratories: a population-based statewide investigation. Am J Clin Pathol. 2013; 139: 330–5. pmid:23429369
- 20. Wheeler CM, Hunt WC, Cuzick J, Langsfeld E, Pearse A, Montoya GD, et al. A population-based study of human papillomavirus genetype prevalence in the United States: baseline measures prior to mass human papillomavirus vaccination. Int J Cancer. 2013; 132: 198–207. pmid:22532127
- 21. Ekwueme DU, Uzunangelov VJ, Hoerger TJ, Miller JW, Saraiya M, Benard VB, et al. Impact of the National Breast and Cervical Cancer Early Detection Program on cervical cancer mortality among uninsured low-income women in the U.S., 1991–2007. Am J Prev Med. 2014; 47: 300–8. pmid:25015564