The development of "automated visual evaluation" for cervical cancer screening: The promise and challenges in adapting deep-learning for clinical testing

Cost-effectiveness analysis of alternative screening strategies for the detection of cervical cancer among women in rural areas of Western Kenya

While the incidence of cervical cancer has dropped in high-income countries due to organized cytology-based screening programs, it remains the leading cause of cancer death among women in Eastern Africa. Therefore, the World Health Organization (WHO) now urges providers to transition from widely prevalent but low-performance visual inspection with acetic acid (VIA) screening to primary human papillomavirus (HPV) DNA testing. Due to high HPV prevalence, effective triage tests are needed to identify those lesions likely to progress and so avoid over-treatment. To identify the optimal cost-effective strategy, we compared the VIA screen-and-treat approach to primary HPV DNA testing with p16/Ki67 dual-stain cytology or VIA as triage. We used a Markov model to calculate the budget impact of each strategy with incremental quality-adjusted life years and incremental cost-effectiveness ratios (ICER) as the main outcome. Deterministic cost-effectiveness analyses show that the screen-and-treat approach is highly cost-effective (ICER 2469 Int$), while screen, triage, and treat with dual staining is the most effective with favorable ICER than triage with VIA (ICER 9943 Int$ compared with 13,177 Int$). One-way sensitivity analyses show that the results are most sensitive to discounting, VIA performance, and test prices. In the probabilistic sensitivity analyses, the triage option using dual stain is the optimal choice above a willingness to pay threshold of 7115 Int$ being cost-effective as per WHO standards. The result of our analysis favors the use of dual staining over VIA as triage in HPV-positive women and portends future opportunities and necessary research to improve the coverage and acceptability of cervical cancer screening programs.

Deep Learning Diagnostic Classification of Cervical Images to Augment Colposcopic Impression

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.

Epidemiology of HPV-associated cancers past, present and future: towards prevention and elimination

Cervical cancer is the first cancer deemed amenable to elimination through prevention, and thus lessons from the epidemiology and prevention of this cancer type can provide information on strategies to manage other cancers. Infection with the human papillomavirus (HPV) causes virtually all cervical cancers, and an important proportion of oropharyngeal, anal and genital cancers. Whereas 20th century prevention efforts were dominated by cytology-based screening, the present and future of HPV-associated cancer prevention relies mostly on HPV vaccination and molecular screening tests. In this Review, we provide an overview of the epidemiology of HPV-associated cancers, their disease burden, how past and contemporary preventive interventions have shaped their incidence and mortality, and the potential for elimination. We particularly focus on the cofactors that could have the greatest effect on prevention efforts, such as parity and human immunodeficiency virus infection, as well as on social determinants of health. Given that the incidence of and mortality from HPV-associated cancers remain strongly associated with the socioeconomic status of individuals and the human development index of countries, elimination efforts are unlikely to succeed unless prevention efforts focus on health equity, with a commitment to both primary and secondary prevention.

HPVsim: An agent-based model of HPV transmission and cervical disease

In 2020, the WHO launched its first global strategy to accelerate the elimination of cervical cancer, outlining an ambitious set of targets for countries to achieve over the next decade. At the same time, new tools, technologies, and strategies are in the pipeline that may improve screening performance, expand the reach of prophylactic vaccines, and prevent the acquisition, persistence and progression of oncogenic HPV. Detailed mechanistic modelling can help identify the combinations of current and future strategies to combat cervical cancer. Open-source modelling tools are needed to shift the capacity for such evaluations in-country. Here, we introduce the Human papillomavirus simulator (HPVsim), a new open-source software package for creating flexible agent-based models parameterised with country-specific vital dynamics, structured sexual networks, and co-transmitting HPV genotypes. HPVsim includes a novel methodology for modelling cervical disease progression, designed to be readily adaptable to new forms of screening. The software itself is implemented in Python, has built-in tools for simulating commonly-used interventions, includes a comprehensive set of tests and documentation, and runs quickly (seconds to minutes) on a laptop. Performance is greatly enhanced by HPVsim's multiscale modelling functionality. HPVsim is open source under the MIT License and available via both the Python Package Index (via pip install) and GitHub (hpvsim.org).

Internal validation of Automated Visual Evaluation (AVE) on smartphone images for cervical cancer screening in a prospective study in Zambia

OBJECTIVES

Visual inspection with acetic acid (VIA) is a low-cost approach for cervical cancer screening used in most low- and middle-income countries (LMICs) but, similar to other visual tests, is subjective and requires sustained training and quality assurance. We developed, trained, and validated an artificial-intelligence-based "Automated Visual Evaluation" (AVE) tool that can be adapted to run on smartphones to assess smartphone-captured images of the cervix and identify precancerous lesions, helping augment VIA performance.

DESIGN

Prospective study.

SETTING

Eight public health facilities in Zambia.

PARTICIPANTS

A total of 8204 women aged 25-55.

INTERVENTIONS

Cervical images captured on commonly used low-cost smartphone models were matched with key clinical information including human immunodeficiency virus (HIV) and human papillomavirus (HPV) status, plus histopathology analysis (where applicable), to develop and train an AVE algorithm and evaluate its performance for use as a primary screen and triage test for women who are HPV positive.

MAIN OUTCOME MEASURES

Area under the receiver operating curve (AUC); sensitivity; specificity.

RESULTS

As a general population screening tool for cervical precancerous lesions, AVE identified cases of cervical precancerous and cancerous (CIN2+) lesions with high performance (AUC = 0.91, 95% confidence interval [CI] = 0.89-0.93), which translates to a sensitivity of 85% (95% CI = 81%-90%) and specificity of 86% (95% CI = 84%-88%) based on maximizing the Youden's index. This represents a considerable improvement over naked eye VIA, which as per a meta-analysis by the World Health Organization (WHO) has a sensitivity of 66% and specificity of 87%. For women living with HIV, the AUC of AVE was 0.91 (95% CI = 0.88-0.93), and among those testing positive for high-risk HPV types, the AUC was 0.87 (95% CI = 0.83-0.91).

CONCLUSIONS

These results demonstrate the feasibility of utilizing AVE on images captured using a commonly available smartphone by nurses in a screening program, and support our ongoing efforts for moving to more broadly evaluate AVE for its clinical sensitivity, specificity, feasibility, and acceptability across a wider range of settings. Limitations of this study include potential inflation of performance estimates due to verification bias (as biopsies were only obtained from participants with visible aceto-white cervical lesions) and due to this being an internal validation (the test data, while independent from that used to develop the algorithm was drawn from the same study).

Enhanced cervical precancerous lesions detection and classification using Archimedes Optimization Algorithm with transfer learning

Cervical cancer (CC) ranks as the fourth most common form of cancer affecting women, manifesting in the cervix. CC is caused by the Human papillomavirus (HPV) infection and is eradicated by vaccinating women from an early age. However, limited medical facilities present a significant challenge in mid- or low-income countries. It can improve the survivability rate and be successfully treated if the CC is detected at earlier stages. Current technological improvements allow for cost-effective, more sensitive, and rapid screening and treatment measures for CC. DL techniques are widely adopted for the automated detection of CC. DL techniques and architectures are used to detect CC and provide higher detection performance. This study offers the design of Enhanced Cervical Precancerous Lesions Detection and Classification using the Archimedes Optimization Algorithm with Transfer Learning (CPLDC-AOATL) algorithm. The CPLDC-AOATL algorithm aims to diagnose cervical cancer using medical images. At the preliminary stage, the CPLDC-AOATL technique involves a bilateral filtering (BF) technique to eliminate the noise in the input images. Besides, the CPLDC-AOATL technique applies the Inception-ResNetv2 model for the feature extraction process, and the use of AOA chose the hyperparameters. The CPLDC-AOATL technique involves a bidirectional long short-term memory (BiLSTM) model for the cancer detection process. The experimental outcome of the CPLDC-AOATL technique emphasized the superior accuracy outcome of 99.53% over other existing approaches under a benchmark dataset.

Squamocolumnar junction visibility, age, and implications for cervical cancer screening programs

Visual assessment is currently used for primary screening or triage of screen-positive individuals in cervical cancer screening programs. Most guidelines recommend screening and triage up to at least age 65 years old. We examined cervical images from participants in three National Cancer Institute funded cervical cancer screening studies: ALTS (2864 participants recruited between 1996 to 1998) in the United States (US), NHS (7548 in 1993) in Costa Rica, and the Biopsy study (684 between 2009 to 2012) in the US. Specifically, we assessed the visibility of the squamocolumnar junction (SCJ), which is the susceptible zone for precancer/cancer by age, as reported by colposcopist reviewers either at examination or review of cervical images. The visibility of the SCJ declined substantially with age: by the late 40s the majority of people screened had at most partially visible SCJ. On longitudinal analysis, the change in SCJ visibility from visible to not visible was largest for participants from ages 40-44 in ALTS and 50-54 in NHS. Of note, in the Biopsy study, the live colposcopic exam resulted in significantly higher SCJ visibility as compared to review of static images (Weighted kappa 0.27 (95% Confidence Interval: 0.21, 0.33), Asymmetry chi-square P-value<0.001). Lack of SCJ visibility leads to increased difficulty in diagnosis and management of cervical precancers. Therefore, cervical cancer screening programs reliant on visual assessment might consider lowering the upper age limit for screening if there are not adequately trained personnel and equipment to evaluate and manage participants with inadequately visible SCJ.

Design of the HPV-automated visual evaluation (PAVE) study: Validating a novel cervical screening strategy

Background

The HPV-automated visual evaluation (PAVE) Study is an extensive, multinational initiative designed to advance cervical cancer prevention in resource-constrained regions. Cervical cancer disproportionally affects regions with limited access to preventive measures. PAVE aims to assess a novel screening-triage-treatment strategy integrating self-sampled HPV testing, deep-learning-based automated visual evaluation (AVE), and targeted therapies.

Methods

Phase 1 efficacy involves screening up to 100,000 women aged 25-49 across nine countries, using self-collected vaginal samples for hierarchical HPV evaluation: HPV16, else HPV18/45, else HPV31/33/35/52/58, else HPV39/51/56/59/68 else negative. HPV-positive individuals undergo further evaluation, including pelvic exams, cervical imaging, and biopsies. AVE algorithms analyze images, assigning risk scores for precancer, validated against histologic high-grade precancer. Phase 1, however, does not integrate AVE results into patient management, contrasting them with local standard care.Phase 2 effectiveness focuses on deploying AVE software and HPV genotype data in real-time clinical decision-making, evaluating feasibility, acceptability, cost-effectiveness, and health communication of the PAVE strategy in practice.

Results

Currently, sites have commenced fieldwork, and conclusive results are pending.

Conclusions

The study aspires to validate a screen-triage-treat protocol utilizing innovative biomarkers to deliver an accurate, feasible, and cost-effective strategy for cervical cancer prevention in resource-limited areas. Should the study validate PAVE, its broader implementation could be recommended, potentially expanding cervical cancer prevention worldwide.

Funding

The consortial sites are responsible for their own study costs. Research equipment and supplies, and the NCI-affiliated staff are funded by the National Cancer Institute Intramural Research Program including supplemental funding from the Cancer Cures Moonshot Initiative. No commercial support was obtained. Brian Befano was supported by NCI/ NIH under Grant T32CA09168.

Artificial intelligence-based image analysis in clinical testing: lessons from cervical cancer screening

Novel screening and diagnostic tests based on artificial intelligence (AI) image recognition algorithms are proliferating. Some initial reports claim outstanding accuracy followed by disappointing lack of confirmation, including our own early work on cervical screening. This is a presentation of lessons learned, organized as a conceptual step-by-step approach to bridge the gap between the creation of an AI algorithm and clinical efficacy. The first fundamental principle is specifying rigorously what the algorithm is designed to identify and what the test is intended to measure (eg, screening, diagnostic, or prognostic). Second, designing the AI algorithm to minimize the most clinically important errors. For example, many equivocal cervical images cannot yet be labeled because the borderline between cases and controls is blurred. To avoid a misclassified case-control dichotomy, we have isolated the equivocal cases and formally included an intermediate, indeterminate class (severity order of classes: case>indeterminate>control). The third principle is evaluating AI algorithms like any other test, using clinical epidemiologic criteria. Repeatability of the algorithm at the borderline, for indeterminate images, has proven extremely informative. Distinguishing between internal and external validation is also essential. Linking the AI algorithm results to clinical risk estimation is the fourth principle. Absolute risk (not relative) is the critical metric for translating a test result into clinical use. Finally, generating risk-based guidelines for clinical use that match local resources and priorities is the last principle in our approach. We are particularly interested in applications to lower-resource settings to address health disparities. We note that similar principles apply to other domains of AI-based image analysis for medical diagnostic testing.

Tritesting in Battor, Ghana: an integrated cervical precancer screening strategy to mitigate the challenges of multiple screening visits and loss to follow-up

Background

Human papillomavirus (HPV) DNA testing is more sensitive than cytology for detecting cervical precancer; however, increasing reports of high-risk HPV (hr-HPV)-negative cases of cervical intraepithelial neoplasia (CIN) and even malignancy motivate the use of combined testing. We present our experience with 'tritesting', defined as the performance of HPV DNA testing, cytology and visual inspection in a single session at the Cervical Cancer Prevention and Training Centre, Ghana. We further determined the prevalence rates of hr-HPV infection, abnormal cytology and cervical lesions among women screened using tritesting.

Methods

This descriptive retrospective cross-sectional study assessed all women screened via tritesting between April 2019 to April 2023. HPV DNA testing was performed using the Sansure MA-6000, GeneXpert or AmpFire platforms. Visual inspection was performed using enhanced visual assessment mobile colposcopy or visual inspection with acetic acid. Liquid-based cytology was performed using cervical samples taken with a Cervex-Brush® and fixed in PreservCyt, while samples for conventional cytology were taken using an Ayre spatula and cytobrush.

Results

Among 236 women screened (mean age, 39.1 years (standard deviation, 10.9)), the overall prevalence rates of hr-HPV infection and cervical lesions were 17.8% (95% confidence interval (CI), 13.1-23.3) and 11.9% (95% CI, 8.0-16.7), respectively. Cytology yielded findings of atypical squamous cells of undetermined significance or worse in 2.5% (95% CI, 0.9-5.5) of women. Histopathology following loop electrosurgical excision procedure revealed CIN I (tritest positive) and CIN III (hr-HPV-positive, visual inspection 'positive', cytology-negative) in one woman each. Factors independently associated with hr-HPV infection among 'tritested' women were age ≥ 39 years, tertiary level of education and current contraceptive use. Twenty-seven out of 39 hr-HPV-positive women (69.2%; 95% CI, 52.4-83.0) showed a type 3 transformation zone and would have needed to be recalled for a cytologic sample to be taken in a 'see and triage' approach with HPV DNA testing and a visual inspection method.

Conclusion

This study brings tritesting into the spotlight, as an alternative to other methods, particularly for women who prefer this due to the advantage of a single visit to a health facility and being more cost-effective, if they have to travel long distances to access cervical screening services.

Reproducible and clinically translatable deep neural networks for cervical screening

Cervical cancer is a leading cause of cancer mortality, with approximately 90% of the 250,000 deaths per year occurring in low- and middle-income countries (LMIC). Secondary prevention with cervical screening involves detecting and treating precursor lesions; however, scaling screening efforts in LMIC has been hampered by infrastructure and cost constraints. Recent work has supported the development of an artificial intelligence (AI) pipeline on digital images of the cervix to achieve an accurate and reliable diagnosis of treatable precancerous lesions. In particular, WHO guidelines emphasize visual triage of women testing positive for human papillomavirus (HPV) as the primary screen, and AI could assist in this triage task. In this work, we implemented a comprehensive deep-learning model selection and optimization study on a large, collated, multi-geography, multi-institution, and multi-device dataset of 9462 women (17,013 images). We evaluated relative portability, repeatability, and classification performance. The top performing model, when combined with HPV type, achieved an area under the Receiver Operating Characteristics (ROC) curve (AUC) of 0.89 within our study population of interest, and a limited total extreme misclassification rate of 3.4%, on held-aside test sets. Our model also produced reliable and consistent predictions, achieving a strong quadratic weighted kappa (QWK) of 0.86 and a minimal %2-class disagreement (% 2-Cl. D.) of 0.69%, between image pairs across women. Our work is among the first efforts at designing a robust, repeatable, accurate and clinically translatable deep-learning model for cervical screening.

Design of the HPV-Automated Visual Evaluation (PAVE) Study: Validating a Novel Cervical Screening Strategy

Objective

To describe the HPV-Automated Visual Evaluation (PAVE) Study, an international, multi-centric study designed to evaluate a novel cervical screen-triage-treat strategy for resource-limited settings as part of a global strategy to reduce cervical cancer burden. The PAVE strategy involves: 1) screening with self-sampled HPV testing; 2) triage of HPV-positive participants with a combination of extended genotyping and visual evaluation of the cervix assisted by deep-learning-based automated visual evaluation (AVE); and 3) treatment with thermal ablation or excision (Large Loop Excision of the Transformation Zone). The PAVE study has two phases: efficacy (2023-2024) and effectiveness (planned to begin in 2024-2025). The efficacy phase aims to refine and validate the screen-triage portion of the protocol. The effectiveness phase will examine acceptability and feasibility of the PAVE strategy into clinical practice, cost-effectiveness, and health communication within the PAVE sites.

Study design

Phase 1 Efficacy: Around 100,000 nonpregnant women, aged 25-49 years, without prior hysterectomy, and irrespective of HIV status, are being screened at nine study sites in resource-limited settings. Eligible and consenting participants perform self-collection of vaginal specimens for HPV testing using a FLOQSwab (Copan). Swabs are transported dry and undergo testing for HPV using a newly-redesigned isothermal DNA amplification HPV test (ScreenFire HPV RS), which has been designed to provide HPV genotyping by hierarchical risk groups: HPV16, else HPV18/45, else HPV31/33/35/52/58, else HPV39/51/56/59/68. HPV-negative individuals are considered negative for precancer/cancer and do not undergo further testing. HPV-positive individuals undergo pelvic examination with collection of cervical images and targeted biopsies of all acetowhite areas or endocervical sampling in the absence of visible lesions. Accuracy of histology diagnosis is evaluated across all sites. Cervical images are used to refine a deep learning AVE algorithm that classifies images as normal, indeterminate, or precancer+. AVE classifications are validated against the histologic endpoint of high-grade precancer determined by biopsy. The combination of HPV genotype and AVE classification is used to generate a risk score that corresponds to the risk of precancer (lower, medium, high, highest). During the efficacy phase, clinicians and patients within the PAVE sites will receive HPV testing results but not AVE results or risk scores. Treatment during the efficacy phase will be performed per local standard of care: positive Visual Inspection with Acetic Acid impression, high-grade colposcopic impression or CIN2+ on colposcopic biopsy, HPV positivity, or HPV 16,18/45 positivity. Follow up of triage negative patients and post treatment will follow standard of care protocols. The sensitivity of the PAVE strategy for detection of precancer will be compared to current SOC at a given level of specificity.Phase 2 Effectiveness: The AVE software will be downloaded to the new dedicated image analysis and thermal ablation devices (Liger Iris) into which the HPV genotype information can be entered to provide risk HPV-AVE risk scores for precancer to clinicians in real time. The effectiveness phase will examine clinician use of the PAVE strategy in practice, including feasibility and acceptability for clinicians and patients, cost-effectiveness, and health communication within the PAVE sites.

Conclusion

The goal of the PAVE study is to validate a screen-triage-treat protocol using novel biomarkers to provide an accurate, feasible, cost-effective strategy for cervical cancer prevention in resource-limited settings. If validated, implementation of PAVE at larger scale can be encouraged.

Funding

The consortial sites are responsible for their own study costs. Research equipment and supplies, and the NCI-affiliated staff are funded by the National Cancer Institute Intramural Research Program including supplemental funding from the Cancer Cures Moonshot Initiative. No commercial support was obtained. Brian Befano was supported by NCI/NIH under Grant T32CA09168. Date of protocol latest review: September 24 th 2023.

Validation in Zambia of a cervical screening strategy including HPV genotyping and artificial intelligence (AI)-based automated visual evaluation

BACKGROUND

WHO has recommended HPV testing for cervical screening where it is practical and affordable. If used, it is important to both clarify and implement the clinical management of positive results. We estimated the performance in Lusaka, Zambia of a novel screening/triage approach combining HPV typing with visual assessment assisted by a deep-learning approach called automated visual evaluation (AVE).

METHODS

In this well-established cervical cancer screening program nested inside public sector primary care health facilities, experienced nurses examined women with high-quality digital cameras; the magnified illuminated images permit inspection of the surface morphology of the cervix and expert telemedicine quality assurance. Emphasizing sensitive criteria to avoid missing precancer/cancer, ~ 25% of women screen positive, reflecting partly the high HIV prevalence. Visual screen-positive women are treated in the same visit by trained nurses using either ablation (~ 60%) or LLETZ excision, or referred for LLETZ or more extensive surgery as needed. We added research elements (which did not influence clinical care) including collection of HPV specimens for testing and typing with BD Onclarity™ with a five channel output (HPV16, HPV18/45, HPV31/33/52/58, HPV35/39/51/56/59/66/68, human DNA control), and collection of triplicate cervical images with a Samsung Galaxy J8 smartphone camera™ that were analyzed using AVE, an AI-based algorithm pre-trained on a large NCI cervical image archive. The four HPV groups and three AVE classes were crossed to create a 12-level risk scale, ranking participants in order of predicted risk of precancer. We evaluated the risk scale and assessed how well it predicted the observed diagnosis of precancer/cancer.

RESULTS

HPV type, AVE classification, and the 12-level risk scale all were strongly associated with degree of histologic outcome. The AVE classification showed good reproducibility between replicates, and added finer predictive accuracy to each HPV type group. Women living with HIV had higher prevalence of precancer/cancer; the HPV-AVE risk categories strongly predicted diagnostic findings in these women as well.

CONCLUSIONS

These results support the theoretical efficacy of HPV-AVE-based risk estimation for cervical screening. If HPV testing can be made affordable, cost-effective and point of care, this risk-based approach could be one management option for HPV-positive women.

Squamocolumnar junction visibility among cervical cancer screening population in Peru might influence upper age for screening programs

Cervical cancer screening and treatment of screen positives is an important and effective strategy to reduce cervical cancer morbidity and mortality. In order to have an accurate cervical cancer screening and evaluation of positives, the entire Squamocolumnar Junction (SCJ) must be visible. Throughout the life course, the position of the SCJ changes and affects its visibility. SCJ visibility was analyzed among participants screened at the League Against Cancer Clinic in Lima, Peru. Of the 4247 participants screened, the SCJ was fully visible in 49.7% of participants, partially visible in 23.1%, and not visible in 27.2%. Visibility decreased with age, and by age 45 years old, the SCJ was not fully visible in over 50% of participants. Our results show that a high percentage of participants at ages still recommended for screening do not have totally visible SCJ, and we may need to reconsider the upper age limit for screening and find new strategies for evaluation of those with a positive screening test and non-visible SCJ.

Toward 70% cervical cancer screening coverage: Technical challenges and opportunities to increase access to human papillomavirus (HPV) testing

The World Health Organization (WHO) has called for the elimination of cervical cancer as a public health problem. Cervical cancer screening through human papillomavirus (HPV) testing is a core component of the strategy for elimination, with a set target of screening 70% of women twice in their lifetimes. In this review, we discuss technical barriers and opportunities to increase HPV screening globally.

Use of risk-based cervical screening programs in resource-limited settings

Cervical cancer screening and management in the U.S. has adopted a risk-based approach. However, the majority of cervical cancer cases and deaths occur in resource-limited settings, where screening and management are not widely available. We describe a conceptual model that optimizes cervical cancer screening and management in resource-limited settings by utilizing a risk-based approach. The principles of risk-based screening and management in resource limited settings include (1) ensure that the screening method effectively separates low-risk from high-risk patients; (2) directing resources to populations at the highest cancer risk; (3) screen using HPV testing via self-sampling; (4) utilize HPV genotyping to improve risk stratification and better determine who will benefit from treatment, and (5) automated visual evaluation with artificial intelligence may further improve risk stratification. Risk-based screening and management in resource limited settings can optimize prevention by focusing triage and treatment resources on the highest risk patients while minimizing interventions in lower risk patients.

Performance of visual inspection of the cervix with acetic acid (VIA) for triage of HPV screen-positive women: results from the ESTAMPA study

VIA is recommended for triage of HPV-positive women attending cervical screening. In the multicentric ESTAMPA study, VIA performance for detection of cervical intraepithelial neoplasia grade 3 or worse (CIN3+) among HPV-positive women was evaluated. Women aged 30-64 years were screened with HPV testing and cytology and referred to colposcopy if either test was positive. At colposcopy visit, study-trained midwives/nurses/GPs performed VIA ahead of colposcopy. VIA was considered positive if acetowhite lesions were observed in or close to the transformation zone. Ablative treatment eligibility was assessed for VIA positives. Performance indicators were estimated. Three thousand one hundred and forty-two HPV-positive women were included. Sensitivity for CIN3+ was 85.9% (95% CI 81.2-89.5) among women <50 years and, although not significant, slightly lower in women 50+ (78.0%, 95% CI 65.9-86.6). Overall specificity was 58.6% (95% CI 56.7-60.5) and was significantly higher among women 50+ (70.3%, 95% CI 66.8-73.5) compared to women <50 (54.3%, 95% CI 52.1-56.5). VIA positivity was lower among women 50+ (35.2%, 95% CI 31.9-38.6) compared to women <50 (53.2, 95% CI 51.1-55.2). Overall eligibility for ablative treatment was 74.5% and did not differ by age. VIA sensitivity, specificity, and positivity, and ablative treatment eligibility varied highly by provider (ranges: 25%-95.4%, 44.9%-94.4%, 8.2%-65.3%, 0%-98.7%, respectively). VIA sensitivity for cervical precancer detection among HPV-positive women performed by trained providers was high with an important reduction in referral rates. However, scaling-up HPV screening triaged by VIA will be challenging due to the high variability of VIA performance and providers' need for training and supervision.

REPRODUCIBLE AND CLINICALLY TRANSLATABLE DEEP NEURAL NETWORKS FOR CANCER SCREENING

Cervical cancer is a leading cause of cancer mortality, with approximately 90% of the 250,000 deaths per year occurring in low- and middle-income countries (LMIC). Secondary prevention with cervical screening involves detecting and treating precursor lesions; however, scaling screening efforts in LMIC has been hampered by infrastructure and cost constraints. Recent work has supported the development of an artificial intelligence (AI) pipeline on digital images of the cervix to achieve an accurate and reliable diagnosis of treatable precancerous lesions. In particular, WHO guidelines emphasize visual triage of women testing positive for human papillomavirus (HPV) as the primary screen, and AI could assist in this triage task. Published AI reports have exhibited overfitting, lack of portability, and unrealistic, near-perfect performance estimates. To surmount recognized issues, we implemented a comprehensive deep-learning model selection and optimization study on a large, collated, multi-institutional dataset of 9,462 women (17,013 images). We evaluated relative portability, repeatability, and classification performance. The top performing model, when combined with HPV type, achieved an area under the Receiver Operating Characteristics (ROC) curve (AUC) of 0.89 within our study population of interest, and a limited total extreme misclassification rate of 3.4%, on held-aside test sets. Our work is among the first efforts at designing a robust, repeatable, accurate and clinically translatable deep-learning model for cervical screening.

Inter- and intra-observer agreement in the assessment of the cervical transformation zone (TZ) by visual inspection with acetic acid (VIA) and its implications for a screen and treat approach: a reliability study

BACKGROUND

In low-resource countries, interpretation of the transformation zone (TZ) using the classification of the International Federation for Cervical Pathology and Colposcopy (IFCPC), adopted by the World Health Organization, is critical for determining if visual inspection with acetic acid (VIA) screening and thermal ablation treatment are possible. We aim to assess inter- and intra-observer agreement in TZ interpretation.

METHODS

We performed a prospective multi-observer reliability study. One hundred cervical digital images of Human papillomavirus positive women (30-49 years) were consecutively selected from a Cameroonian cervical cancer screening trial. Images of the native cervix and after VIA were obtained. The images were evaluated for the TZ type at two time points (rounds one and two) by five VIA experts from four countries (Côte d'Ivoire, Cameroon, Peru, and Zambia) according to the IFCPC classification (TZ1 = ectocervical fully visible; TZ2 = endocervical fully visible; TZ3 = not fully visible). Intra- and inter-observer agreement were measured by Fleiss' kappa.

RESULTS

Overall, 37.0% of images were interpreted as TZ1, 36.4% as TZ2, and 26.6% as TZ3. Global inter-observer reliability indicated fair agreement in both rounds (kappa 0.313 and 0.288). The inter-observer agreement was moderate for TZ1 interpretation (0.460), slight for TZ2 (0.153), and fair for TZ3 (0.329). Intra-observer analysis showed fair agreement for two observers (0.356 and 0.345), moderate agreement for two other (0.562 and 0.549), and one with substantial agreement (0.728).

CONCLUSION

Interpretation of the TZ using the IFCPC classification, adopted by the World Health Organization, is critical for determining if VIA screening and thermal ablation treatment are possible. However, the low inter- and intra-observer agreement suggest that the reliability of the referred classification is limited in the context of VIA. It's integration in treatment recommendations should be used with caution since TZ3 interpretation could lead to an important referral rate for further evaluation. Trial registration Cantonal Ethics Board of Geneva, Switzerland: N°2017-0110. Cameroonian National Ethics Committee for Human Health Research N°2018/07/1083/CE/CNERSH/SP.

Diagnostic accuracy of cervical cancer screening strategies for high-grade cervical intraepithelial neoplasia (CIN2+/CIN3+) among women living with HIV: A systematic review and meta-analysis

BACKGROUND

We systematically reviewed the diagnostic accuracy of cervical cancer screening and triage strategies in women living with HIV (WLHIV).

METHODS

Cochrane Library, Embase, Global Health and Medline were searched for randomised controlled trials, prospective or cross-sectional studies published from database inception to 15 July 2022 reporting diagnostic accuracy of tests in cervical cancer screening and triage of screen-positive WLHIV. Studies were included if they reported the diagnostic accuracy of any cervical cancer screening or triage strategies for the detection of histologically-confirmed high-grade cervical intraepithelial neoplasia (CIN2+/CIN3+) among WLHIV. Summary data were extracted from published reports. Authors were contacted for missing data where applicable. Sensitivity and specificity estimates for CIN2/3+ were pooled using models for meta-analysis of diagnostic accuracy data. Study quality was assessed using the QUADAS-2 tool for the quality assessment of diagnostic accuracy studies. PROSPERO registration:CRD42020189031.

FINDINGS

In 38 studies among 18,737 WLHIV, the majority (n=19) were conducted in sub-Saharan Africa. The pooled prevalence was 12.0% (95%CI:9.8-14.1) for CIN2+ and 6.7% (95%CI:5.0-8.4) for CIN3+. The proportion of screen-positive ranged from 3-31% (visual inspection using acetic acid[VIA]); 2-46% (high-grade squamous intraepithelial lesions, and greater [HSIL+] cytology); 20-64% (high-risk[HR]-HPV DNA). In 14 studies, sensitivity and specificity of VIA were variable limiting the reliability of pooled estimates. In 5 studies where majority had histology-confirmed CIN2+, pooled sensitivity was 56.0% (95%CI:45.4-66.1; I²=65%) for CIN2+ and 65.0% (95%CI:52.9-75.4; I² =42%) for CIN3+; specificity for <CIN2 was 73.8% (95%CI:59.8-84.2, I²=94%). Cytology was similarly variable (sensitivity of ASCUS+ for CIN2+ range: 58-100%; specificity: 9-96%). In 28 studies, sensitivity of tests targeting 14-HR-HPV types was high (91.6%, 95%CI:88.1-94.1; I²=45% for CIN2+ and 92.5%, 95%CI:88.4-95.2; I²=32%) for CIN3+); but specificity for <CIN2 was low (62.2% (95%CI:57.9-66.4;I²=92%). Restriction to 8-HR-HPV increased specificity (65.8%; Relative specificity[RSpec] vs. 14-HR-HPV=1.17; 95%CI:1.10-1.24) with no significant change in sensitivity (CIN2+:85.5%; Relative Sensitivity[RSens]=0.94, 95%CI: 0.89-1.00; CIN3+:90%; RSens=0.96, 95%CI:0.89-1.03). VIA triage of 14-HR-HPV positive women decreased sensitivity for CIN2+ compared to HPV-DNA test alone (64.4% vs. 91.6%; RSens=0.68, 95%CI:0.62-0.75).

INTERPRETATION

HPV-DNA based approaches consistently showed superior sensitivity for CIN2+/CIN3+ compared to VIA or cytology. The low specificity of HPV-DNA based methods targeting up to 14-HR-HPV could be improved significantly by restricting to 8-HR-HPV with only minor losses in sensitivity, limiting requirement for triage for which optimal approaches are less clear.

FUNDING

World Health Organisation; National Cancer Institute; European Union's Horizon 2020 and Marie Skłodowska-Curie Actions programme.

Acting on the call for cervical cancer elimination: Planning tools for low- and middle- income countries to increase the coverage and effectiveness of screening and treatment

Introduction

Accessible planning tools tailored for low-and middle-income countries can assist decision makers in comparing implementation of different cervical cancer screening approaches and treatment delivery scenarios in settings with high cervical cancer burden.

Methods

The Cervical Precancer Planning Tool (CPPT) was developed by PATH for users to explore and compare the accuracy of screening approaches, what treatment equipment to procure, and how best to deploy treatment equipment in a given country. The CPPT compares four screening approaches: 1) visual inspection with acetic acid (VIA), 2) HPV testing, 3) HPV testing followed by a VIA triage, and 4) HPV testing followed by an enhanced triage test. Accuracy of screening outcomes (e.g., true positives, false positives) is based on published sensitivity and specificity of tests to detect cervical precancerous lesions. The CPPT compares five scenarios for deploying ablative treatment equipment: 1) cervical precancer equipment at every location a woman is screened (single visit approach), 2) equipment only at a hospital level, 3) a single unit of equipment in each district, 4) allowing two districts to share a single unit of equipment, and 5) equipment placed at select district hospitals paired with mobile outreach. Users can customize the CPPT by adjusting pre-populated baseline values and assumptions, including population estimates, screening age range, screening frequency, HPV and HIV prevalence, supply costs, and health facility details.

Results

The CPPT generates data tables and graphs that compare the results of implementing each of the four screening and five treatment scenarios disaggregated by HIV status. Outputs include the number and outcomes of women screened, cost of each screening approach, provider time and cost saved by implementing self-sampling for HPV testing, number of women treated, treatment equipment needed by type, and the financial and economic costs for each equipment deployment scenario.

Conclusion

The CPPT provides practical information and data to compare tradeoffs of patient access and screening accuracy as well as efficient utilization of equipment, skilled personnel, and financial resources. Country decision makers can use outputs from the CPPT to guide the scale-up of cervical cancer screening and treatment while optimizing limited resources.

Modelling cervical cancer elimination using single-visit screening and treatment strategies in the context of high HIV prevalence: estimates for KwaZulu-Natal, South Africa

INTRODUCTION

In settings with high HIV prevalence, cervical cancer incidence rates are up to six-fold higher than the global average of 13.1 cases per 100,000 women-years. To inform strategies for global cervical cancer elimination, we used a dynamic transmission model to evaluate scalable screening and treatment strategies, accounting for HIV-associated cancer risks and weighing prevention gains against overtreatment.

METHODS

We developed a dynamic model of HIV-HPV co-infection and disease progression, which we calibrated to KwaZulu-Natal, South Africa. Our baseline scenario reflects the current practice of HPV vaccination with a multi-visit screening and treatment strategy involving cytology and colposcopy triage. We evaluated 13 comparator scenarios with increased vaccination coverage and one-time, two-time or repeat HIV-targeted cervical cancer screening with the following single-visit strategies: HPV DNA testing, HPV genotyping, automated visual evaluation (AVE) and HPV DNA with AVE triage. In all scenarios, HIV antiretroviral therapy, condom use and voluntary male medical circumcision continue at baseline levels. We simulated cancer incidence under each scenario from 2020 to 2120 using the 25 best-fitting parameter sets. We present the median and range of model output from these simulations to account for parameter uncertainty.

RESULTS

We estimate that cervical cancer incidence will decrease by 87% with the continuation of current cervical cancer and HIV prevention strategies, from an age-standardized rate per 100,000 women of 80.4 (range 58.2, 112.1) in 2020 to 10.7 (4.2, 29.9) in 2120. Scenarios scaling up vaccination and single-visit strategies resulted in near- and long-term gains. With repeat HIV-targeted screening, incidence rates were projected to be 29-34% lower in 2030 relative to the baseline scenario, and elimination (incidence <4/100,000) was achieved with HPV DNA testing in 2095 and with AVE in 2114. A strategy of HPV DNA with AVE triage optimized the tradeoff between cancer cases averted and overtreatment.

CONCLUSIONS

Single-visit screening strategies could avert a substantial burden of cervical cancer and accelerate progress towards elimination in settings with a high burden of HIV. Increasing the screening frequency among women with HIV and reducing loss-to-follow-up for treatment will be key components of a successful elimination strategy.

Image Quality Classification for Automated Visual Evaluation of Cervical Precancer

Image quality control is a critical element in the process of data collection and cleaning. Both manual and automated analyses alike are adversely impacted by bad quality data. There are several factors that can degrade image quality and, correspondingly, there are many approaches to mitigate their negative impact. In this paper, we address image quality control toward our goal of improving the performance of automated visual evaluation (AVE) for cervical precancer screening. Specifically, we report efforts made toward classifying images into four quality categories ("unusable", "unsatisfactory", "limited", and "evaluable") and improving the quality classification performance by automatically identifying mislabeled and overly ambiguous images. The proposed new deep learning ensemble framework is an integration of several networks that consists of three main components: cervix detection, mislabel identification, and quality classification. We evaluated our method using a large dataset that comprises 87,420 images obtained from 14,183 patients through several cervical cancer studies conducted by different providers using different imaging devices in different geographic regions worldwide. The proposed ensemble approach achieved higher performance than the baseline approaches.

Point-of-care HPV DNA testing of self-collected specimens and same-day thermal ablation for the early detection and treatment of cervical pre-cancer in women in Papua New Guinea: a prospective, single-arm intervention trial (HPV-STAT)

BACKGROUND

WHO recommends human papillomavirus (HPV) testing and same-day treatment for cervical screening in low-income and middle-income countries (LMICs); however, few published data exist on the validity of the strategy. We aimed to evaluate the clinical performance, treatment completion rates, adverse events profile, and acceptability of a fully integrated strategy, comprising point-of-care HPV DNA testing of self-collected specimens and same-day thermal ablation, for screening of cervical cancer in women in Papua New Guinea.

METHODS

HPV-STAT was a large-scale, prospective, single-arm intervention trial conducted at two clinical sites in Papua New Guinea. Cervical screening clinics with an on-site consultant gynaecologist were selected in consultation with national and provincial health authorities, church health services, and local stakeholders. Eligible participants were women aged 30-59 years attending cervical screening services at the two clinics, who were willing to comply with study procedures and able to provide written informed consent. Women self-collected vaginal specimens for point-of-care GeneXpert testing (Cepheid, Sunnyvale, CA, USA) for oncogenic HPV types. Women testing positive for HPV underwent pelvic examination followed by same-day thermal ablation or referral for gynaecology review. All HPV-positive women and a 15% random sample of HPV-negative women provided a clinician-collected cervical specimen for liquid-based cytology. The primary outcome was clinical performance (ie, sensitivity, specificity, positive predictive value [PPV], and negative predictive value [NPV]) of the strategy for the detection of high-grade squamous intraepithelial lesion (HSIL) or worse. This trial is registered with ISRCTN, ISRCTN13476702.

FINDINGS

Between June 5, 2018, and Jan 6, 2020, we recruited 4285 women, 3638 (84·9%) of whom tested negative for HPV and 647 (15·1%) tested positive for one or more oncogenic HPV type. Sensitivity of the algorithm to detect HSIL or worse was 85·4% (95% CI 81·0-89·6), with specificity 89·6% (88·6-90·6), PPV 35·2% (31·6-39·0), and NPV 98·9% (98·6-99·2). Among HPV-positive women, 602 (93·0%) received same-day thermal ablation and 42 (6·5%) were referred for gynaecology review, 37 (88·1%) of whom attended. Acceptability was high among both HPV-positive and HPV-negative women. Among the 329 HPV-positive women who attended a 3-month follow-up visit, 51 (15·5%) reported mild adverse symptoms that resolved in all cases by the follow-up visit. There were no serious adverse events.

INTERPRETATION

We conducted the first real-world evaluation of a fully integrated point-of-care HPV self-collect, test, and treat strategy for same-day cervical screening in a LMIC and found it to be effective, acceptable, and safe when implemented at scale in primary health-care facilities in Papua New Guinea. Our findings support the introduction and scale-up of HPV screening and treatment for the control and elimination of cervical cancer in LMICs, as recommended by WHO.

FUNDING

Australian National Health and Medical Research Council.

Comparison of accuracy and reproducibility of colposcopic impression based on a single image versus a two-minute time series of colposcopic images

OBJECTIVE

Colposcopy is an important part of cervical screening/management programs. Colposcopic appearance is often classified, for teaching and telemedicine, based on static images that do not reveal the dynamics of acetowhitening. We compared the accuracy and reproducibility of colposcopic impression based on a single image at one minute after application of acetic acid versus a time-series of 17 sequential images over two minutes.

METHODS

Approximately 5000 colposcopic examinations conducted with the DYSIS colposcopic system were divided into 10 random sets, each assigned to a separate expert colposcopist. Colposcopists first classified single two-dimensional images at one minute and then a time-series of 17 sequential images as 'normal,' 'indeterminate,' 'high grade,' or 'cancer'. Ratings were compared to histologic diagnoses. Additionally, 5 colposcopists reviewed a subset of 200 single images and 200 time series to estimate intra- and inter-rater reliability.

RESULTS

Of 4640 patients with adequate images, only 24.4% were correctly categorized by single image visual assessment (11% of 64 cancers; 31% of 605 CIN3; 22.4% of 558 CIN2; 23.9% of 3412 < CIN2). Individual colposcopist accuracy was low; Youden indices (sensitivity plus specificity minus one) ranged from 0.07 to 0.24. Use of the time-series increased the proportion of images classified as normal, regardless of histology. Intra-rater reliability was substantial (weighted kappa = 0.64); inter-rater reliability was fair ( weighted kappa = 0.26).

CONCLUSION

Substantial variation exists in visual assessment of colposcopic images, even when a 17-image time series showing the two-minute process of acetowhitening is presented. We are currently evaluating whether deep-learning image evaluation can assist classification.

False Negative Results in Cervical Cancer Screening-Risks, Reasons and Implications for Clinical Practice and Public Health

False negative (FN) results in cervical cancer (CC) screening pose serious risks to women. We present a comprehensive literature review on the risks and reasons of obtaining the FN results of primary CC screening tests and triage methods and discuss their clinical and public health impact and implications. Misinterpretation or true lack of abnormalities on a slide are the reasons of FN results in cytology and p16/Ki-67 dual-staining. For high-risk human papillomavirus (HPV) molecular tests, those include: truly non-HPV-associated tumors, lesions driven by low-risk HPV types, and clearance of HPV genetic material before sampling. Imprecise disease threshold definition lead to FN results in visual inspection with acetic acid. Lesions with a discrete colposcopic appearance are a source of FN in colposcopic procedures. For FAM19A4 and hsa-miR124-2 genes methylation, those may originate from borderline methylation levels. Histological misinterpretation, sampling, and laboratory errors also play a role in all types of CC screening, as well as reproducibility issue, especially in methods based on human-eye evaluation. Primary HPV-based screening combined with high quality-assured immunocytochemical and molecular triage methods seem to be an optimal approach. Colposcopy with histological evaluation remains the gold standard for diagnosis but requires quality protocols and assurance measures.

HPV prevention and control - The way forward

The global confrontation with COVID-19 has not only diverted current healthcare resources to deal with the infection but has also resulted in increased resources in the areas of testing and screening, as well as educating most of the global public of the benefits of vaccination. When the COVID-19 pandemic eventually recedes, the opportunity must not be missed to ensure that these newly created resources are maintained and redeployed for use in testing and immunisation against other vaccine-preventable infectious diseases. A notable example is infection by human papillomavirus (HPV), the commonest sexually transmitted human virus and the leading cause of a variety of cancers in both men and women, such as cervical, head and neck, anal, vaginal, vulvar and penile cancers. The most important is cervical cancer, the objective of the global elimination goals targeting the vaccination of young female and male adolescents, screening all women and treatment of all infected women. As the campaigns to control SARS-CoV-2, the eradication of HPV-induced cancers also relies on effective prevention and control programs. The lessons learned and the technical, logistical and human resources which have been established to combat COVID-19 by vaccination and testing must be applied to the eradication of other infections which affect the global population. This commentary summarizes the opportunities that the COVID-19 pandemic has created for HPV prevention and control, lists the already available tools for HPV control, and emphasizes the potential public health threats amidst the ongoing COVID-19 pandemic.

Assessment of the implementation context in preparation for a clinical study of machine-learning algorithms to automate the classification of digital cervical images for cervical cancer screening in resource-constrained settings

Introduction

We assessed the implementation context and image quality in preparation for a clinical study evaluating the effectiveness of automated visual assessment devices within cervical cancer screening of women living without and with HIV.

Methods

We developed a semi-structured questionnaire based on three Consolidated Framework for Implementation Research (CFIR) domains; intervention characteristics, inner setting, and process, in Cape Town, South Africa. Between December 1, 2020, and August 6, 2021, we evaluated two devices: MobileODT handheld colposcope; and a commercially-available cell phone (Samsung A21ST). Colposcopists visually inspected cervical images for technical adequacy. Descriptive analyses were tabulated for quantitative variables, and narrative responses were summarized in the text.

Results

Two colposcopists described the devices as easy to operate, without data loss. The clinical workspace and gynecological workflow were modified to incorporate devices and manage images. Providers believed either device would likely perform better than cytology under most circumstances unless the squamocolumnar junction (SCJ) were not visible, in which case cytology was expected to be better. Image quality (N = 75) from the MobileODT device and cell phone was comparable in terms of achieving good focus (81% vs. 84%), obtaining visibility of the squamous columnar junction (88% vs. 97%), avoiding occlusion (79% vs. 87%), and detection of lesion and range of lesion includes the upper limit (63% vs. 53%) but differed in taking photographs free of glare (100% vs. 24%).

Conclusion

Novel application of the CFIR early in the conduct of the clinical study, including assessment of image quality, highlight real-world factors about intervention characteristics, inner clinical setting, and workflow process that may affect both the clinical study findings and ultimate pace of translating to clinical practice. The application and augmentation of the CFIR in this study context highlighted adaptations needed for the framework to better measure factors relevant to implementing digital interventions.