Chin J Cancer Res

Evaluation of triage strategies for high-risk human papillomavirus-positive women in cervical cancer screening: A multicenter randomized controlled trial in different resource settings in China

doi: 10.21147/j.issn.1000-9604.2022.05.09

ObjectiveWe aimed to evaluate the effectiveness of different triage strategies for high-risk human papillomavirus (hrHPV)-positive women in primary healthcare settings in China.MethodsThis study was undertaken in 11 rural and 9 urban sites. Women aged 35−64 years old were enrolled. HrHPV-positive women were randomly allocated to liquid-based cytology (LBC), visual inspection with acetic acid and Lugol’s iodine (VIA/VILI) (rural only) triage, or directly referred to colposcopy (direct COLP). At 24 months, hrHPV testing, LBC and VIA/VILI were conducted for combined screening.ResultsIn rural sites, 1,949 hrHPV-positive women were analyzed. A total of 852, 218 and 480 women were randomly assigned to direct COLP, LBC and VIA/VILI. At baseline, colposcopy referral rates of LBC or VIA/VILI triage could be reduced by 70%−80%. LBC (n=3 and n=7) or VIA/VILI (n=8 and n=26) could significantly decrease the number of colposcopies needed to detect one cervical intraepithelial neoplasia (CIN) 2 or worse and CIN3+ compared with direct COLP (n=14 and n=23). For the 24-month cumulative detection rate of CIN2+, VIA/VILI triage was 0.50-fold compared with LBC triage and 0.46-fold with the direct COLP. When stratified by age, baseline LBC triage+ performed best (P<0.001), peaking among women aged 35−44 years (P_trend=0.002). In urban sites, 1,728 women were hrHPV genotyping test positive. A total of 408, 571 and 568 women were randomly assigned to direct COLP for HPV16/18+, direct COLP for other hrHPV subtypes+, and LBC triage for other hrHPV subtypes+. LBC (n=12 and n=31) significantly decreased the number of colposcopies needed to detect one CIN2+ and CIN3+ compared with direct COLP (n=14 and n=44). HPV16/18+ increased the 24-month cumulative detection rate of CIN2+ (17.89%, P<0.001).ConclusionsLBC triage for hrHPV-positive women in rural settings and direct COLP for HPV16/18+ women and LBC triage for other hrHPV subtype+ women in urban settings might be feasible strategies.

关键词: Cervical cancer screening, hrHPV-positive, LBC, triage strategy, VIA/VILI

Deep learning-based multi-task prediction of response to neoadjuvant chemotherapy using multiscale whole slide images in breast cancer: A multicenter study

doi: 10.21147/j.issn.1000-9604.2025.01.03

ObjectiveEarly predicting response before neoadjuvant chemotherapy (NAC) is crucial for personalized treatment plans for locally advanced breast cancer patients. We aim to develop a multi-task model using multiscale whole slide images (WSIs) features to predict the response to breast cancer NAC more finely.MethodsThis work collected 1,670 whole slide images for training and validation sets, internal testing sets, external testing sets, and prospective testing sets of the weakly-supervised deep learning-based multi-task model (DLMM) in predicting treatment response and pCR to NAC. Our approach models two-by-two feature interactions across scales by employing concatenate fusion of single-scale feature representations, and controls the expressiveness of each representation via a gating-based attention mechanism.ResultsIn the retrospective analysis, DLMM exhibited excellent predictive performance for the prediction of treatment response, with area under the receiver operating characteristic curves (AUCs) of 0.869 [95% confidence interval (95% CI): 0.806−0.933] in the internal testing set and 0.841 (95% CI: 0.814−0.867) in the external testing sets. For the pCR prediction task, DLMM reached AUCs of 0.865 (95% CI: 0.763−0.964) in the internal testing and 0.821 (95% CI: 0.763−0.878) in the pooled external testing set. In the prospective testing study, DLMM also demonstrated favorable predictive performance, with AUCs of 0.829 (95% CI: 0.754−0.903) and 0.821 (95% CI: 0.692−0.949) in treatment response and pCR prediction, respectively. DLMM significantly outperformed the baseline models in all testing sets (P<0.05). Heatmaps were employed to interpret the decision-making basis of the model. Furthermore, it was discovered that high DLMM scores were associated with immune-related pathways and cells in the microenvironment during biological basis exploration.ConclusionsThe DLMM represents a valuable tool that aids clinicians in selecting personalized treatment strategies for breast cancer patients.

关键词: Artificial intelligence, breast cancer, digital pathology, whole slide images

Economic evaluation of cervical cancer screening strategies in urban China

doi: 10.21147/j.issn.1000-9604.2019.06.13

ObjectiveThis study evaluated the feasibility of different cervical cancer screening strategies in urban China.MethodsA Markov model was constructed to simulate a hypothetical cohort of 100,000 females aged 30−59 years in a 20-year period. Screening strategies included liquid-based cytology (LBC) every three years, human papillomavirus (HPV) DNA testing every three and five years, respectively, and a combination of HPV DNA testing and LBC (HPV+LBC) every three and five years, respectively. Model outcomes included cumulative incidence over 20 years, cumulative risk of cervical cancer, costs, life year saved (LYS), quality-adjusted life years (QALYs) and benefits. The cost-effectiveness ratios (CERs), incremental cost-effectiveness ratios (ICERs), cost-utility ratios (CURs), and benefit-cost ratios (BCRs) were used as outcomes in the health economic evaluation analysis. Univariate sensitivity analyses were performed to examine the stability of the results.ResultsThe cumulative incidence of the five screening strategies ranged from 833.02 to 1,158.07 cases per 100,000 females. HPV DNA testing was most effective in reducing the cumulative risk of cervical cancer, saving life years and QALYs and gaining benefits. The CERs of HPV DNA testing every three and five years, and LBC every three years were considered to be very cost-effective if they were below China’s GDP per capita. The CERs of HPV+LBC were considered to be cost-effective if they were below three times GDP per capita. The incremental cost-effectiveness analysis showed that HPV DNA testing every three and five years, LBC every three years and HPV+LBC every five years were dominant strategies.ConclusionsThe findings of this study indicated that HPV DNA testing every five years or LBC every three years should be recommended in urban China.

关键词: Cervical cancer, Markov model, economic evaluation

主管单位：	中国科学院
主办单位：	中国电子学会

版权所有：
通讯地址：
工作邮箱：
办公电话：	020-37627489

Chin J Cancer Res

高级检索