Age‐specific 3‐year cumulative risk of cervical cancer and high‐grade dysplasia on biopsy in 9434 women who underwent HPV cytology cotesting

Age-specific 3-year risk of cervical precancer among HPV-positive women attending screening: a post hoc analysis from a retrospective cohort

This post hoc analysis explored the age-specific risk of cervical precancer in women infected with human papillomavirus (HPV), using data from a cohort of 7263 participants aged 21-71years undergoing cervical screening. We found a slightly varied prevalence of high-risk HPV (hrHPV) in different age, with highest in women under 30 years old (9.28% for 13 hrHPVs tested by HC2-HPV, 10.82% for 14 hrHPVs tested by DH3-HPV). However, the prevalence of cytology abnormalities peaked in age 30-39 years (~ 3.6%). A total of 5840 women completed 3-year follow-up. Among them, 558 were positive for HC2 assay and 583 were positive for DH3-HPV at baseline. Of note, the 3-year cumulative risks for cervical intraepithelial neoplasia grade 2+ (CIN2+) or grade 3+ (CIN3+) in women infected with high-risk HPV did not increase with age but declined (e.g., 41.67%, 27.78%, 26.42%, 15.98%, and 18% for CIN2 + risk in HC2-positive women at year 25-29, year 30-39, year 40-49, year 50-59, and year 60-71, respectively). If stratified by the median age, younger women (25-48 years) positive with HC2-HPV at baseline had a higher 3-year CIN2+/CIN3 + risk than older women (49-71 years) [26.55% (95%CI = 21.8-31.92%) vs. 18.28% (95%CI = 14.11-23.34%), P = 0.019; 15.52% (95%CI = 11.81-20.14%) vs. 9.7% (95%CI = 6.71-13.83%), P = 0.039]. Similarly, for women positive with DH3-HPV at baseline, younger group had a higher 3-year CIN2+/CIN3 + risk than older group [26.44% (95%CI = 21.73-31.75%) vs. 17.01% (95%CI = 13.11-21.78%), P = 0.006; 15.25% (95%CI = 11.6-19.8%) vs. 9.03% (95%CI = 6.24-12.9%), P = 0.021]. These findings indicate the potential value of age-specific risk assessment in cervical cancer screening.

The Immunohistochemical Expression of Epithelial-Mesenchymal Transition Markers in Precancerous Lesions and Cervical Cancer

In the epithelial-mesenchymal transition (EMT) process, cells lose their epithelial phenotype and gain mesenchymal features. This phenomenon was observed in the metastatic phase of neoplastic diseases, e.g., cervical cancer. There are specific markers that are expressed in the EMT. The aim of this study was to determine the localization of and associations between the immunohistochemical (IHC) expression of TWIST, SNAIL, and SLUG proteins in precancerous lesions and cervical cancer. The IHC analysis disclosed higher expressions of EMT markers in precancerous lesions and cervical cancer than in the control group. Moreover, stronger expression of TWIST, SNAIL, and SLUG was observed in cervical intraepithelial neoplasia grade 3 (CIN3) vs. CIN1, CIN3 vs. CIN2, and CIN2 vs. CIN1 cases (p < 0.05). In cervical cancer, IHC reactions demonstrated differences in TWIST, SNAIL, and SLUG expression in grade 1 (G1) vs. grade 2 (G2) (p < 0.0011; p < 0.0017; p < 0.0001, respectively) and in G1 vs. grade 3 (G3) (p < 0.0029; p < 0.0005; p < 0.0001, respectively). The results of our study clearly showed that existing differences in the expression of the tested markers in precancerous vs. cancerous lesions may be utilized in the diagnosis of cervical cancer. Further studies on bigger populations, as well as in comparison with well-known markers, may improve our outcomes.

Clinicopathological characteristics and prognostic risk factors of cervical cancer patients aged ≤35 years old

We aimed to explore the clinicopathological characteristics and prognostic risk factors of cervical cancer in patients aged ≤35 years. A total of 256 cervical cancer patients treated at Anhui Medical University Affiliated Maternity and Child Health Hospital and The First Affiliated Hospital of Anhui Medical University from January 2016 to October 2018 were divided into ≤35-year-old (n = 136) and >35-year-old (n = 120) groups. Their clinicopathological characteristics and 3-year cumulative disease-free survival (DFS) and overall survival (OS) rates were compared. The factors influencing the 3-year cumulative DFS rate of patients in the ≤35-year-old group were analyzed using univariate and multivariate Cox regression models. The human papillomavirus (HPV) infection rate, incidence rate of contact vaginal bleeding, depth of cervical interstitial infiltration, and incidence rates of parametrial metastasis and vascular infiltration were all significantly higher in the ≤35-year-old group than in the >35-year-old group. The 3-year cumulative DFS rates of all patients and those with HPV infection and contact vaginal bleeding were significantly lower in the ≤35-year-old group than in the >35-year-old group (69.12% vs. 77.50%, 68.29% vs. 80.85%, and 66.04% vs. 81.48%) (log-rank χ2 = 7.429, 4.339, and 4.276, P < .05). Depth of cervical interstitial infiltration >4 mm, parametrial metastasis, lymph node metastasis, and vascular infiltration were independent risk factors for the prognosis of cervical cancer patients aged ≤35 years (P < .05). Cervical cancer aged ≤35 years have a worse postoperative prognosis than those aged >35 years, which is affected by the depth of cervical interstitial infiltration >4 mm, parametrial metastasis, lymph node metastasis, and vascular infiltration. Therefore, it is necessary to identify more effective treatment methods for young patients with cervical cancer to improve the therapeutic effect and reduce the risk of recurrence and metastasis.

Classification of cervical biopsy free-text diagnoses through linear-classifier based natural language processing

Routine cervical cancer screening has significantly decreased the incidence and mortality of cervical cancer. As selection of proper screening modalities depends on well-validated clinical decision algorithms, retrospective review correlating cytology and HPV test results with cervical biopsy diagnosis is essential for validating and revising these algorithms to changing technologies, demographics, and optimal clinical practices. However, manual categorization of the free-text biopsy diagnosis into discrete categories is extremely laborious due to the overwhelming number of specimens, which may lead to significant error and bias. Advances in machine learning and natural language processing (NLP), particularly over the last decade, have led to significant accomplishments and impressive performance in computer-based classification tasks. In this work, we apply an efficient version of an NLP framework, FastText™, to an annotated cervical biopsy dataset to create a supervised classifier that can assign accurate biopsy categories to free-text biopsy interpretations with high concordance to manually annotated data (>99.6%). We present cases where the machine-learning classifier disagrees with previous annotations and examine these discrepant cases after referee review by an expert pathologist. We also show that the classifier is robust on an untrained external dataset, achieving a concordance of 97.7%. In conclusion, we demonstrate a useful application of NLP to a real-world pathology classification task and highlight the benefits and limitations of this approach.

Loop Electrosurgical Excision Procedure in Managing Persistent Low-Grade Abnormality or Human Papillomavirus Positivity

OBJECTIVES

The aims of the study were to estimate the rate and to identify predictors of high-grade abnormalities among women with persistent low-grade abnormalities or high-risk human papillomavirus (hrHPV) positivity for at least 2 years stratified by presence (high risk) or absence (low risk) of previous high-grade results or HPV 16/18.

MATERIALS AND METHODS

A retrospective cohort study of patients who underwent a loop electrosurgical excision procedure (LEEP) for persistent low-grade or hrHPV positivity was performed. Patients were stratified based on whether they had a history of high-grade and/or HPV 16/18 positivity. Rates of high-grade or worse abnormalities on LEEP were compared using Fisher exact tests. Logistic regression was used to evaluate the associations between patient characteristics and high-grade results on the LEEP.

RESULTS

Three hundred eleven LEEPs were performed for persistent low-grade or hrHPV positivity. The rates of occult high grade were 12% and 22% among the low- and high-risk groups, respectively. Compared with those 45 years and older, the adjusted odds of high grade was 3.79 (95% CI = 1.19-12.1) for women aged 25-29 years. The odds of high grade was higher among current versus never smokers (6.40; 95% CI = 2.01-20.4) and those with a history of high-grade abnormality (2.23; 95% CI = 1.12-4.43). At 2 years, approximately half had an abnormal cytology and/or hrHPV positivity result independent of whether high grade was identified on their LEEP specimen.

CONCLUSIONS

Patients with persistent low-grade abnormalities or persistent hrHPV should be counseled on the risks and benefits of a LEEP given that 12%-22% have a risk of occult high grade, especially if they have a history of high-grade dysplasia.

Evaluating the performance of three different cervical cancer screening modalities in a large prospective population-based cohort

BACKGROUND

This study aimed to evaluate three different patterns of cervical cancer screening strategies for detection of cervical diseases.

METHODS

In total, 10,186 women aged 21-70 years attending cervical screening program were recruited and cotested by human papillomavirus (HPV) assays and cytology. Three-year histological follow-up data was recorded on women with abnormal screening results, and six clinically common screening algorithms were evaluated.

RESULTS

Significantly better protection against cervical intraepithelial neoplasia 2 or worse (CIN2+) at three-year follow-up was associated with a negative HPV result than by normal cytology at baseline. HPV screening was more sensitive and less specific than cytology screening. Moreover, HPV screening with HPV16/18 and reflex cytology (atypical squamous cells of undetermined significance [ASCUS] threshold) showed a similar sensitivity (94.6% vs. 98.2%, p = 0.125) and a superior specificity as compared to cotesting reflex HPV16/18 and cytology (ASCUS threshold) for CIN2+ (95.8% vs. 95.1%, p < 0.0001), achieving a colposcopy referral rate of 5.4%, and consuming 4.8 colposcopies and 4.4 cytology tests to find one CIN2+.

CONCLUSIONS

HPV screening with triage of HPV-positive women by HPV16/18 genotyping and cytology provided a good equilibrium between screening effectiveness, the number of cytology tests required, and referral rates; HPV testing was similar in sensitivity to cotesting and safer than cytology, thus especially suitable for large population-based screening programs.