ABSTRACT
OBJECTIVE: A deep learning classifier that improves the accuracy of colposcopic impression. METHODS: Colposcopy images taken 56 seconds after acetic acid application were processed by a cervix detection algorithm to identify the cervical region. We optimized models based on the SegFormer architecture to classify each cervix as high-grade or negative/low-grade. The data were split into histologically stratified, random training, validation, and test subsets (80%-10%-10%). We replicated a 10-fold experiment to align with a prior study utilizing expert reviewer analysis of the same images. To evaluate the model's robustness across different cameras, we retrained it after dividing the dataset by camera type. Subsequently, we retrained the model on a new, histologically stratified random data split and integrated the results with patients' age and referral data to train a Gradient Boosted Tree model for final classification. Model accuracy was assessed by the receiver operating characteristic area under the curve (AUC), Youden's index (YI), sensitivity, and specificity compared to the histology. RESULTS: Out of 5,485 colposcopy images, 4,946 with histology and a visible cervix were used. The model's average performance in the 10-fold experiment was AUC = 0.75, YI = 0.37 (sensitivity = 63%, specificity = 74%), outperforming the experts' average YI of 0.16. Transferability across camera types was effective, with AUC = 0.70, YI = 0.33. Integrating image-based predictions with referral data improved outcomes to AUC = 0.81 and YI = 0.46. The use of model predictions alongside the original colposcopic impression boosted overall performance. CONCLUSIONS: Deep learning cervical image classification demonstrated robustness and outperformed experts. Further improved by including additional patient information, it shows potential for clinical utility complementing colposcopy.
Subject(s)
Cervix Uteri , Colposcopy , Deep Learning , Uterine Cervical Neoplasms , Humans , Female , Colposcopy/methods , Cervix Uteri/pathology , Uterine Cervical Neoplasms/diagnosis , Uterine Cervical Neoplasms/pathology , Uterine Cervical Neoplasms/classification , Adult , Middle Aged , Sensitivity and Specificity , Image Processing, Computer-Assisted/methods , Young Adult , AgedABSTRACT
Cervical cancer is caused in the vast majority of cases by the human papilloma virus (HPV) through sexual contact and requires a specific molecular-based analysis to be detected. As an HPV vaccine is available, the incidence of cervical cancer is up to ten times higher in areas without adequate healthcare resources. In recent years, liquid cytology has been used to overcome these shortcomings and perform mass screening. In addition, classifiers based on convolutional neural networks can be developed to help pathologists diagnose the disease. However, these systems always require the final verification of a pathologist to make a final diagnosis. For this reason, explainable AI techniques are required to highlight the most significant data to the healthcare professional, as it can be used to determine the confidence in the results and the areas of the image used for classification (allowing the professional to point out the areas he/she thinks are most important and cross-check them against those detected by the system in order to create incremental learning systems). In this work, a 4-phase optimization process is used to obtain a custom deep-learning classifier for distinguishing between 4 severity classes of cervical cancer with liquid-cytology images. The final classifier obtains an accuracy over 97% for 4 classes and 100% for 2 classes with execution times under 1 s (including the final report generation). Compared to previous works, the proposed classifier obtains better accuracy results with a lower computational cost.
Subject(s)
Uterine Cervical Neoplasms , Humans , Uterine Cervical Neoplasms/pathology , Uterine Cervical Neoplasms/diagnosis , Uterine Cervical Neoplasms/classification , Female , Neural Networks, Computer , Deep Learning , AlgorithmsABSTRACT
OBJECTIVE: Neuroendocrine carcinoma of the cervix (NECC) is a rare malignancy with poor clinical prognosis due to limited therapeutic options. This study aimed to establish a risk-stratification score and nomogram models to predict prognosis in NECC patients. METHODS: Data on individuals diagnosed with NECC between 2000 and 2019 were retrieved from the Surveillance Epidemiology and End Results (SEER) database and then randomly classified into training and validation cohorts (7:3). Univariate and multivariate Cox regression analyses evaluated independent indicators of prognosis. Least absolute shrinkage and selection operator (LASSO) regression analysis further assisted in confirming candidate variables. Based on these factors, cancer-specific survival (CSS) and overall survival (OS) nomograms that predict survival over 1, 3, and 5 years were constructed. The receiver operating characteristic (ROC) curve, the concordance index (C-index), and the calibration curve estimated the precision and discriminability of the competing risk nomogram for both cohorts. Finally, we assessed the clinical value of the nomograms using decision curve analysis (DCA). RESULTS: Data from 2348 patients were obtained from the SEER database. Age, tumor stage, T stage, N stage, chemotherapy, radiotherapy, and surgery predicted OS. Additionally, histological type was another standalone indicator of CSS prognosis. For predicting CSS, the C-index was 0.751 (95% CI 0.731 ~ 0.770) and 0.740 (95% CI 0.710 ~ 0.770) for the training and validation cohorts, respectively. Furthermore, the C-index in OS prediction was 0.757 (95% CI 0.738 ~ 0.776) and 0.747 (95% CI 0.718 ~ 0.776) for both cohorts. The proposed model had an excellent discriminative ability. Good accuracy and discriminability were also demonstrated using the AUC and calibration curves. Additionally, DCA demonstrated the high clinical potential of the nomograms for CSS and OS prediction. We constructed a corresponding risk classification system using nomogram scores. For the whole cohort, the median CSS times for the low-, moderate-, and high-risk groups were 59.3, 19.5, and 7.4 months, respectively. CONCLUSION: New competing risk nomograms and a risk classification system were successfully developed to predict the 1-, 3-, and 5-year CSS and OS of NECC patients. The models are internally accurate and reliable and may guide clinicians toward better clinical decisions and the development of personalized treatment plans.
Subject(s)
Carcinoma, Neuroendocrine , Nomograms , SEER Program , Uterine Cervical Neoplasms , Humans , Female , Carcinoma, Neuroendocrine/mortality , Carcinoma, Neuroendocrine/pathology , Uterine Cervical Neoplasms/mortality , Uterine Cervical Neoplasms/pathology , Uterine Cervical Neoplasms/diagnosis , Uterine Cervical Neoplasms/therapy , Uterine Cervical Neoplasms/classification , Retrospective Studies , Middle Aged , Prognosis , SEER Program/statistics & numerical data , Adult , Risk Assessment/methods , Aged , Survival Rate , ROC Curve , Follow-Up Studies , Risk FactorsABSTRACT
BACKGROUND: Although many cervical cytology diagnostic support systems have been developed, it is challenging to classify overlapping cell clusters with a variety of patterns in the same way that humans do. In this study, we developed a fast and accurate system for the detection and classification of atypical cell clusters by using a two-step algorithm based on two different deep learning algorithms. METHODS: We created 919 cell images from liquid-based cervical cytological samples collected at Sapporo Medical University and annotated them based on the Bethesda system as a dataset for machine learning. Most of the images captured overlapping and crowded cells, and images were oversampled by digital processing. The detection system consists of two steps: (1) detection of atypical cells using You Only Look Once v4 (YOLOv4) and (2) classification of the detected cells using ResNeSt. A label smoothing algorithm was used for the dataset in the second classification step. This method annotates multiple correct classes from a single cell image with a smooth probability distribution. RESULTS: The first step, cell detection by YOLOv4, was able to detect all atypical cells above ASC-US without any observed false negatives. The detected cell images were then analyzed in the second step, cell classification by the ResNeSt algorithm, which exhibited average accuracy and F-measure values of 90.5% and 70.5%, respectively. The oversampling of the training image and label smoothing algorithm contributed to the improvement of the system's accuracy. CONCLUSION: This system combines two deep learning algorithms to enable accurate detection and classification of cell clusters based on the Bethesda system, which has been difficult to achieve in the past. We will conduct further research and development of this system as a platform for augmented reality microscopes for cytological diagnosis.
Subject(s)
Image Processing, Computer-Assisted/instrumentation , Uterine Cervical Neoplasms/diagnostic imaging , Vaginal Smears/statistics & numerical data , Algorithms , Deep Learning , Early Detection of Cancer/statistics & numerical data , Female , Humans , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/diagnosisABSTRACT
BACKGROUND: We aimed to analyze the clinicopathological features and outcomes of patients with gastric-type of HPV-independent endocervical adenocarcinoma (GAS HPVI ECA), and compare them with non-GAS HPVI ECA cases. METHODS: Thirty-eight GASs [including 17 minimal deviation adenocarcinoma (MDA), 21 non-MDA GAS] and 17 non-GAS HPVI ECAs were studied. Data of clinical features, pathological characteristics, treatment, and outcomes were evaluated. RESULTS: The median age of patients with GAS and non-GAS HPVI ECA was 46 and 48 years, respectively (p = 0.93). Compared with non-GAS HPVI ECAs, GAS had more common complains of vaginal watery discharge (p = 0.04). GAS cases were also associated with higher clinical stage (p = 0.036), more common in deeper cervical stromal invasion (p = 0.002) and lymphoavascular invasion (p = 0.044). GAS was associated with worse median progression-free survival (PFS) (p = 0.02) and median overall survival (OS) (p = 0.03) over patients with non-GAS HPVI ECAs. MDA had similar clinical and pathological features and prognosis compared with non-MDA GAS. Of note, serum CA19-9 levels were significantly higher in GAS than that in non-GAS HPVI ECA cases. CONCLUSIONS: GAS cases were more likely to have high risk pathological factors and poorer PFS and OS compared with non-GAS HPVI ECAs. Serum CA19-9 may be helpful for diagnosis and screening in patients with GAS.
Subject(s)
Adenocarcinoma/pathology , Uterine Cervical Neoplasms/pathology , Adenocarcinoma/blood , Adenocarcinoma/classification , Adenocarcinoma/mortality , Adult , Aged , CA-19-9 Antigen/blood , Female , Humans , Middle Aged , Neoplasm Staging , Papillomavirus Infections , Prognosis , Progression-Free Survival , Retrospective Studies , Risk Factors , Uterine Cervical Neoplasms/blood , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/mortality , Vaginal DischargeABSTRACT
Cervical cancer is the second most common cancer in women worldwide with a mortality rate of 60%. Cervical cancer begins with no overt signs and has a long latent period, making early detection through regular checkups vitally immportant. In this study, we compare the performance of two different models, machine learning and deep learning, for the purpose of identifying signs of cervical cancer using cervicography images. Using the deep learning model ResNet-50 and the machine learning models XGB, SVM, and RF, we classified 4119 Cervicography images as positive or negative for cervical cancer using square images in which the vaginal wall regions were removed. The machine learning models extracted 10 major features from a total of 300 features. All tests were validated by fivefold cross-validation and receiver operating characteristics (ROC) analysis yielded the following AUCs: ResNet-50 0.97(CI 95% 0.949-0.976), XGB 0.82(CI 95% 0.797-0.851), SVM 0.84(CI 95% 0.801-0.854), RF 0.79(CI 95% 0.804-0.856). The ResNet-50 model showed a 0.15 point improvement (p < 0.05) over the average (0.82) of the three machine learning methods. Our data suggest that the ResNet-50 deep learning algorithm could offer greater performance than current machine learning models for the purpose of identifying cervical cancer using cervicography images.
Subject(s)
Deep Learning , Diagnosis, Computer-Assisted/methods , Machine Learning , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/diagnostic imaging , Algorithms , Diagnosis, Computer-Assisted/statistics & numerical data , Diagnostic Errors , Female , Humans , Neural Networks, Computer , Photography/methods , ROC CurveABSTRACT
Understanding the regional lineages and sublineages of human papillomavirus type 56 (HPV 56) would be of great importance for further evolutionary, epidemiological, and biological investigations. To identify the distribution of lineages and sublineages of HPV 56 in Iran, the sequence variations of the E6 gene were analyzed in normal, premalignant, and malignant samples obtained from the cervix. In total, 58 HPV 56-positive samples were investigated by nested-PCR and followed by bidirectional direct nucleotide sequencing analysis. Both lineages A and B were identified in the studied samples. Lineage B was dominant as it was detected in 88.4% of all samples and the remaining samples belonged to lineage A (11.6%). Sublineages A1 and A2 were detected in 3.3% and 8.3% of all samples, respectively. With regard to the pathological stages of cervical specimens, no statistically significant differences were found in the three studied groups (p > 0.05). In conclusion, our findings showed that lineage B of HPV 56 was prevalent in Iran. However, further studies with a larger sample size are warranted to estimate the pathogenicity risk of HPV 56 lineages/sublineages to the progression of cervical cancer among Iranian women.
Subject(s)
Cervix Uteri/virology , Genetic Variation , Papillomaviridae/classification , Papillomaviridae/genetics , Papillomavirus Infections/virology , Uterine Cervical Dysplasia/virology , Uterine Cervical Neoplasms/virology , Adult , Cross-Sectional Studies , DNA, Viral/genetics , Female , Genotype , Humans , Iran , Middle Aged , Oncogene Proteins, Viral/genetics , Papillomaviridae/isolation & purification , Papillomaviridae/pathogenicity , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/pathology , Uterine Cervical Neoplasms/prevention & controlABSTRACT
High-grade cervical intraepithelial neoplasia (CIN2 and CIN3) represents a heterogeneous disease with varying cancer progression risks. Biomarkers indicative for a productive human papillomavirus (HPV) infection (HPV E4) and a transforming HPV infection (p16ink4a , Ki-67 and host-cell DNA methylation) could provide guidance for clinical management in women with high-grade CIN. This study evaluates the cumulative score of immunohistochemical expression of p16ink4a (Scores 0-3) and Ki-67 (Scores 0-3), referred to as the "immunoscore" (IS), in 262 CIN2 and 235 CIN3 lesions derived from five European cohorts in relation to immunohistochemical HPV E4 expression and FAM19A4/miR124-2 methylation in the corresponding cervical scrape. The immunoscore classification resulted in 30 lesions within IS group 0-2 (6.0%), 151 lesions within IS group 3-4 (30.4%) and 316 lesions within IS group 5-6 (63.6%). E4 expression decreased significantly from CIN2 to CIN3 (P < .001) and with increasing immunoscore group (Ptrend < .001). Methylation positivity increased significantly from CIN2 to CIN3 (P < .001) and with increasing immunoscore group (Ptrend < .001). E4 expression was present in 9.8% of CIN3 (23/235) and in 12.0% of IS group 5-6 (38/316). Notably, in a minority (43/497, 8.7%) of high-grade lesions, characteristics of both transforming HPV infection (DNA hypermethylation) and productive HPV infection (E4 expression) were found simultaneously. Next, we stratified all high-grade CIN lesions, based on the presumed cancer progression risk of the biomarkers used, into biomarker profiles. These biomarker profiles, including immunoscore and methylation status, could help the clinician in the decision for immediate treatment or a "wait and see" policy to reduce overtreatment of high-grade CIN lesions.
Subject(s)
Cyclin-Dependent Kinase Inhibitor p16/metabolism , Cytokines/metabolism , DNA Methylation , Ki-67 Antigen/metabolism , MicroRNAs/genetics , Oncogene Proteins, Viral/metabolism , Uterine Cervical Dysplasia/pathology , Uterine Cervical Neoplasms/pathology , Adult , Biomarkers, Tumor/genetics , Biomarkers, Tumor/metabolism , Cyclin-Dependent Kinase Inhibitor p16/genetics , Cytokines/genetics , Disease Management , Female , Follow-Up Studies , Gene Expression Regulation, Neoplastic , Humans , Ki-67 Antigen/genetics , Oncogene Proteins, Viral/genetics , Prognosis , Prospective Studies , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/genetics , Uterine Cervical Neoplasms/metabolism , Uterine Cervical Dysplasia/classification , Uterine Cervical Dysplasia/genetics , Uterine Cervical Dysplasia/metabolismABSTRACT
Histopathologic classification of endocervical adenocarcinomas (EAC) has recently changed, with the new system based on human papillomavirus (HPV)-related morphologic features being incorporated into the 5th edition of the WHO Blue Book (Classification of Tumours of the Female Genital Tract). There has also been the introduction of a pattern-based classification system to assess invasion in HPV-associated (HPVA) endocervical adenocarcinomas that stratifies tumors into 3 groups with different prognoses. To facilitate the introduction of these changes into routine clinical practice, websites with training sets and test sets of scanned whole slide images were designed to improve diagnostic performance in histotype classification of endocervical adenocarcinoma based on the International Endocervical Adenocarcinoma Criteria and Classification (IECC) and assessment of Silva pattern of invasion in HPVA endocervical adenocarcinomas. We report on the diagnostic results of those who have participated thus far in these educational websites. Our goal was to identify areas where diagnostic performance was suboptimal and future educational efforts could be directed. There was very good ability to distinguish HPVA from HPV-independent adenocarcinomas within the WHO/IECC classification, with some challenges in the diagnosis of HPV-independent subtypes, especially mesonephric carcinoma. Diagnosis of HPVA subtypes was not consistent. For the Silva classification, the main challenge was related to distinction between pattern A and pattern B, with a tendency for participants to overdiagnose pattern B invasion. These observations can serve as the basis for more targeted efforts to improve diagnostic performance.
Subject(s)
Adenocarcinoma/classification , Carcinoma/diagnosis , Papillomaviridae/isolation & purification , Papillomavirus Infections/virology , Pathologists/education , Uterine Cervical Neoplasms/classification , Adenocarcinoma/diagnosis , Adenocarcinoma/pathology , Carcinoma/pathology , Diagnostic Self Evaluation , Education, Distance , Female , Humans , Neoplasm Invasiveness/diagnosis , Uterine Cervical Neoplasms/diagnosis , Uterine Cervical Neoplasms/pathologyABSTRACT
The International Society of Gynecological Pathologists (ISGyP) Endocervical Adenocarcinoma Project aims to provide evidence-based guidance for the pathologic evaluation, classification, and reporting of endocervical adenocarcinoma. This review presents the recommendations pertaining to gross evaluation and intraoperative consultation of specimens obtained from patients in the setting of cervical cancer. The recommendations are the product of review of published peer-reviewed evidence, international guidelines and institutional grossing manuals, as well as deliberation within this working group. The discussion presented herein details the approach to the different specimen types encountered in practice: loop electrosurgical excision procedure, cone, trachelectomy, radical hysterectomy, pelvic exenteration, and lymphadenectomy specimens. Guidelines for intraoperative evaluation of trachelectomy and sentinel lymph node specimens are also addressed. Correlation with ISGyP recommendations on cancer staging, which appear as a separate review in this issue, is also included when appropriate. While conceived in the framework of endocervical adenocarcinoma, most of the discussion and recommendations can also be applied to other cervical malignancies.
Subject(s)
Adenocarcinoma/pathology , Practice Guidelines as Topic , Uterine Cervical Neoplasms/pathology , Adenocarcinoma/classification , Adenocarcinoma/surgery , Evidence-Based Medicine , Female , Gynecology , Humans , Hysterectomy , Lymph Node Excision , Monitoring, Intraoperative , Pathologists , Pelvic Exenteration , Sentinel Lymph Node/pathology , Societies, Medical , Trachelectomy , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/surgeryABSTRACT
There is a lack of consensus regarding the prognostic value of grading endocervical adenocarcinomas and currently, no universally applied, validated system for grading exists. Several grading schemes have been proposed, most incorporating an evaluation of tumor architecture and nuclear morphology and these are often based on the International Federation of Gynecology and Obstetrics (FIGO) system for endometrial endometrioid carcinoma, although some schemes modify the proportion of solid tumor required to separate grades 1 and 2 from 5% to 10%. In the absence of a validated system, we endorse this approach for most human papillomavirus-associated endocervical adenocarcinomas and, based on the available evidence, recommend that tumors with ≤10% solid growth be designated grade 1, 11% to 50% solid growth grade 2 and >50% solid growth grade 3. Tumors should be upgraded in the presence of marked nuclear atypia involving the majority (>50%) of the tumor. Grading is not recommended for human papillomavirus-independent adenocarcinomas, since no validated system has been suggested and most of these neoplasms exhibit intrinsically aggressive behavior regardless of their morphologic appearance. Importantly, grading should not be performed for gastric-type adenocarcinomas, particularly as these tumors may appear deceptively "low-grade" yet still exhibit aggressive behavior. Recently devised, validated and reproducible etiology and pattern-based tumor classification systems for endocervical adenocarcinomas appear to offer more effective risk stratification than tumor grading and, in the future, these systems may render the provision of a tumor grade redundant.
Subject(s)
Adenocarcinoma/pathology , Practice Guidelines as Topic , Uterine Cervical Neoplasms/pathology , Adenocarcinoma/classification , Female , Gynecology , Humans , Neoplasm Grading , Pathologists , Societies, Medical , Uterine Cervical Neoplasms/classificationABSTRACT
The Silva pattern-based classification for human papilloma virus-associated invasive adenocarcinoma has emerged as a reliable system to predict risk of lymph node metastasis and recurrences. Although not a part of any staging system yet, it has been incorporated in synoptic reports as established by the College of American Pathologists (CAP) and the International Collaboration on Cancer Reporting (ICCR). Moreover, the current National Comprehensive Cancer Network (NCCN) guidelines include this classification as an "emergent concept." In order to facilitate the understating and application of this new classification by all pathologists, the ISGyP Endocervical Adenocarcinoma Project Working Group presents herein all the current evidence on the Silva classification and aims to provide recommendations for its implementation in practice, including interpretation, reporting, and application to biopsy and resection specimens. In addition, this article addresses the distinction of human papilloma virus-associated adenocarcinoma in situ and gastric type adenocarcinoma in situ from their invasive counterparts.
Subject(s)
Adenocarcinoma in Situ/classification , Adenocarcinoma/classification , Papillomaviridae/isolation & purification , Papillomavirus Infections/virology , Practice Guidelines as Topic , Stomach Neoplasms/classification , Uterine Cervical Neoplasms/classification , Adenocarcinoma/pathology , Adenocarcinoma in Situ/pathology , Biopsy , Female , Gynecology , Humans , Lymphatic Metastasis , Pathologists , Societies, Medical , Stomach Neoplasms/pathology , Uterine Cervical Neoplasms/pathologyABSTRACT
The incidence of endocervical adenocarcinoma, the second most common cervical cancer in the world, has been on the rise. While most cervical cancers are squamous cell carcinomas and associated with high-risk oncogenic human papillomavirus (HPV), approximately 15% of endocervical adenocarcinomas, which now represent about one quarter of all cervical cancers, are HPV-independent. In this review, we will focus on the shortcomings of historical histologic classification systems of female genital tract tumors as they pertain to endocervical adenocarcinomas, and we will highlight the advantages of the new International Endocervical Adenocarcinoma Criteria and Classification system, which forms the basis for the WHO 2020 classification. We will cover the various histologic types, subtypes, and variants of endocervical adenocarcinoma with regard to morphology, immunophenotype, molecular genetics, HPV status and differential diagnosis, and we will provide International Society of Gynecological Pathologists recommendations for diagnosing these tumors.
Subject(s)
Adenocarcinoma/classification , Papillomaviridae/isolation & purification , Papillomavirus Infections/virology , Practice Guidelines as Topic , Uterine Cervical Neoplasms/classification , Adenocarcinoma/genetics , Adenocarcinoma/pathology , Female , Gynecology , Humans , Immunophenotyping , Neoplasm Grading , Pathologists , Societies, Medical , Uterine Cervical Neoplasms/genetics , Uterine Cervical Neoplasms/pathologyABSTRACT
The International Federation of Gynecology and Obstetrics (FIGO) updated its staging system for cervical cancer in 2018 with changes that affect size criteria for early stage disease, as well as including pathology and radiology in addition to clinical assessment to be used in staging. Lymph node involvement was also included in the staging system. In early stage disease, pathologic findings are crucial in determining stage, which in turn determine treatment and prognosis for the patient. Therefore, it is imperative that there are unified and consistent methods and recommendations for assessing and reporting pathologic parameters for accurate staging. We describe the changes in the revised FIGO staging scheme and discuss controversial issues in cervical cancer staging from a pathologic perspective. We also provide practical recommendations regarding these parameters based on literature review and/or expert opinion/consensus.
Subject(s)
Adenocarcinoma/classification , Papillomaviridae/isolation & purification , Papillomavirus Infections/virology , Practice Guidelines as Topic , Uterine Cervical Neoplasms/classification , Adenocarcinoma/pathology , Female , Gynecology , Humans , Neoplasm Staging , Pathologists , Societies, Medical , Uterine Cervical Neoplasms/pathologyABSTRACT
BACKGROUND: Screening for breast cancer and cervical cancer in the newly independent states of the former Soviet Union is largely opportunistic, and countries in the region have among the highest cervical cancer incidence in the WHO European Region. We aimed to compare the stage-specific distributions and changes over time in breast cancer and cervical cancer incidence in the newly independent states of the former Soviet Union. METHODS: We collected breast cancer and cervical cancer incidence data from official statistics from Armenia, Azerbaijan, Belarus, Georgia, Kazakhstan, Kyrgyzstan, Republic of Moldova, Russian Federation, Ukraine, and Uzbekistan for the years 2008-17 by tumour, node, metastasis (TNM) stage, and by age where population-based cancer registry data were available. We used log-linear regression to quantify the changes over time in age-standardised rates. FINDINGS: During the period 2013-17, more than 50% of breast cancer cases across the analysed countries, and more than 75% of breast cancer cases in Belarus, Kazakhstan, and Ukraine, were registered at stages I-II. The proportion of stage I breast cancer cases was highest in the screening age group (50-69 years) compared with other ages in Moldova and the Russian registries, but was highest in those aged 15-49 years in Georgia and Ukraine. Breast cancer stage-specific incidence rates increased over time, most prominently for stage I cancers. For cervical cancer, the proportions of cancers diagnosed at a late stage (stages III and IV) were high, particularly in Moldova and Armenia (>50%). The proportion of stage I cervical cancer cases decreased with age in all countries, whereas the proportions of late stage cancers increased with age. Stage-specific incidence rates of cervical cancer generally increased over the period 2008-17. INTERPRETATION: Our results suggest modest progress in early detection of breast cancer in the newly independent states of the former Soviet Union. The high proportions of early-stage disease in the absence of mammography screening (eg, in Belarus) provide a benchmark for what is achievable with rapid diagnosis. For cervical cancer, there is a need to tackle the high burden and unfavourable stage-specific changes over time in the region. A radical shift in national policies away from opportunistic screening toward organised, population-based, quality-assured human papillomavirus vaccination and screening programmes is urgently needed. FUNDING: Union for International Cancer Control, WHO Regional Office for Europe, and Ministry of Health of Ukraine.
Subject(s)
Breast Neoplasms/pathology , Uterine Cervical Neoplasms/pathology , Adolescent , Adult , Aged , Breast Neoplasms/classification , Breast Neoplasms/epidemiology , Early Detection of Cancer , Female , Follow-Up Studies , Humans , Incidence , Middle Aged , Neoplasm Staging , USSR/epidemiology , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/epidemiology , Young AdultABSTRACT
Adenocarcinoma admixed with neuroendocrine carcinoma of the uterine cervix is a rare malignancy with a poor prognosis. In the literature, there are few reported cases. Herein, we report a case of a 56-year-old Turkish woman with cervical adenocarcinoma admixed with small cell neuroendocrine carcinoma. Histological examination of endocervical curettage specimens revealed a tumor composed of almost equal areas of small cell neuroendocrine carcinoma and adenocarcinoma. Neuroendocrine differentiation was confirmed by immunohistochemistry for chromogranin-A, synaptophysin, and CD 56. After the adenocarcinoma and small cell neuroendocrine carcinoma association was detected in the curettage material, both cervicovaginal smear and then total abdominal hysterectomy and bilateral salpingo-oophorectomy resection material of the patient were submitted to our pathology department. Histological features of both curettage and resection material were determined by immunohistochemical studies.
Subject(s)
Adenocarcinoma/diagnosis , Carcinoma, Neuroendocrine/diagnosis , Uterine Cervical Neoplasms/diagnosis , Adenocarcinoma/classification , Adenocarcinoma/surgery , CD56 Antigen/genetics , Carcinoma, Neuroendocrine/surgery , Cervix Uteri/pathology , Chromogranin A/genetics , Female , Humans , Hysterectomy , Immunohistochemistry , Middle Aged , Synaptophysin/genetics , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/surgeryABSTRACT
OBJECTIVES: The objective of this study was to determine if there has been an increase in the age of diagnosis of cervical cancer over time, specifically in the proportion of patients over 65 years old, given decreasing rates of hysterectomy. MATERIALS AND METHODS: A retrospective review of a single institution was conducted including cervical cancer patients seen between 1986 and 2016. Data included demographic variables including age of diagnosis, last cervical cancer screening, and cancer information. Cochran-Armitage test was used to assess temporal trends in the proportion of patients diagnosed over 65. RESULTS: A total of 1,019 patients with cervical cancer were reviewed, of whom 116 were over the age of 65. The age of diagnosis increased by 0.2 years per calendar year, with an average age of diagnosis of 43.7 years old in 1986 versus 49.5 years old in 2016 (p<0.01). The proportion of patients diagnosed with cervical cancer over the age of 65 did not significantly differ over time (17.2 % in 1986 vs. 14.8 % in 2016, p=0.39). 19.0 % of women diagnosed with cervical cancer over the age of 65 developed cancer despite exiting screening appropriately. CONCLUSIONS: In our cohort, the age of diagnosis of cervical cancer increased over time, however, there was no significant difference in the percentage of women diagnosed over the age of 65.
Subject(s)
Time Factors , Uterine Cervical Neoplasms/classification , Adult , Aged , Aged, 80 and over , Cohort Studies , Female , Humans , Middle Aged , Retrospective Studies , Uterine Cervical Neoplasms/diagnosis , Uterine Cervical Neoplasms/epidemiologyABSTRACT
Cervical cancer is the fourth most common cancer in women worldwide. Increasing evidence has shown that miRNAs are related to the progression of cervical cancer. However, the mechanisms that affect the prognosis of cancer are still largely unknown. In the present study, we sought to identify miRNAs associated with poor prognosis of patient with cervical cancer, as well as the possible mechanisms regulated by them. The miRNA expression profiles and relevant clinical information of patients with cervical cancer were obtained from The Cancer Genome Atlas (TCGA). The selection of prognostic miRNAs was carried out through an integrated bioinformatics approach. The most effective miRNAs with synergistic and additive effects were selected for validation through in vitro experiments. Three miRNAs (miR-216b-5p, miR-585-5p, and miR-7641) were identified as exhibiting good performance in predicting poor prognosis through additive effects analysis. The functional enrichment analysis suggested that not only pathways traditionally involved in cancer but also immune system pathways might be important in regulating the outcome of the disease. Our findings demonstrated that a synergistic combination of three miRNAs may be associated, through their regulation of specific pathways, with very poor survival rates for patients with cervical cancer.
Subject(s)
MicroRNAs/genetics , Uterine Cervical Neoplasms/genetics , Computational Biology , Female , Gene Expression Regulation, Neoplastic , Humans , MicroRNAs/classification , Prognosis , Transcriptome , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/diagnosis , Uterine Cervical Neoplasms/pathologyABSTRACT
BACKGROUND: Although more than 10 years have passed since HPV vaccination was implemented, first as an interim programme (Emergent vaccine promotion programme) in November 2010, followed by incorporating into the National Immunization Programme in April, 2013 and suspended in June 2013, limited studies have investigated the HPV vaccine effectiveness against high-grade cervical lesions in Japan. METHODS: We collected the matched data of the results of cervical biopsy and history of vaccination from the Japan Cancer Society database. The subjects were women aged 20 to 29 years screened for cervical cancer between April, 2015 and March, 2017, and with information on HPV vaccination status. We estimated the relative risk of developing high-grade cervical lesions in vaccinated subjects using Poisson regression as compared to unvaccinated subjects. RESULTS: Among the 34,281 women screened, 3770 (11.0%) were vaccinated. The prevalence of CIN2+ was statistically significantly lower in the vaccinated women as compared to the unvaccinated women (Vaccine Effectiveness (VE) =76%; RR = 0.24, 95% CI:0.10-0.60). High VE against CIN3+ was also observed (91%; RR = 0.09, 95% CI:0.00-0.42). CONCLUSION: Women aged 20-29 years who received at least one dose of HPV vaccine had a significantly lower risk of high-grade cervical lesions than those not vaccinated. In Japan, HPV vaccination should be resumed in order to reduce the incidence of cervical cancer.
Subject(s)
Papillomaviridae/immunology , Papillomavirus Infections/prevention & control , Papillomavirus Vaccines/immunology , Uterine Cervical Dysplasia/prevention & control , Uterine Cervical Neoplasms/prevention & control , Vaccination , Adult , Cross-Sectional Studies , Female , Humans , Immunization Programs , Incidence , Japan/epidemiology , Papillomavirus Infections/epidemiology , Papillomavirus Vaccines/administration & dosage , Prevalence , Treatment Outcome , Uterine Cervical Neoplasms/classification , Uterine Cervical Neoplasms/virology , Young Adult , Uterine Cervical Dysplasia/classification , Uterine Cervical Dysplasia/virologyABSTRACT
Colposcopy is widely used to detect cervical cancers, but experienced physicians who are needed for an accurate diagnosis are lacking in developing countries. Artificial intelligence (AI) has been recently used in computer-aided diagnosis showing remarkable promise. In this study, we developed and validated deep learning models to automatically classify cervical neoplasms on colposcopic photographs. Pre-trained convolutional neural networks were fine-tuned for two grading systems: the cervical intraepithelial neoplasia (CIN) system and the lower anogenital squamous terminology (LAST) system. The multi-class classification accuracies of the networks for the CIN system in the test dataset were 48.6 ± 1.3% by Inception-Resnet-v2 and 51.7 ± 5.2% by Resnet-152. The accuracies for the LAST system were 71.8 ± 1.8% and 74.7 ± 1.8%, respectively. The area under the curve (AUC) for discriminating high-risk lesions from low-risk lesions by Resnet-152 was 0.781 ± 0.020 for the CIN system and 0.708 ± 0.024 for the LAST system. The lesions requiring biopsy were also detected efficiently (AUC, 0.947 ± 0.030 by Resnet-152), and presented meaningfully on attention maps. These results may indicate the potential of the application of AI for automated reading of colposcopic photographs.