Appropriate Technology For Breast Cancer Diagnosis And Screening In Low-and Middle-income Areas

ObjectivesWe aimed to provide more reliable and applicable scientific evidence for the optimization of the breast cancer diagnosis and screening process in China,especially in low-and middle-income areas1.In the multicenter clinical diagnostic study,to evaluate and compare the diagnostic performance of automated breast ultrasound system(ABUS)and handheld ultrasound(HHUS)both separately and in combination with mammography(MG)in female outpatients;2.Establish a multicenter,long-term breast cancer screening trial,and describe the preliminary findings.Methods1.In the diagnostic study,outpatients aged 40-69 years old were recruited from breast surgery in five tertiary hospitals.All participants underwent ABUS,HHUS,and MG examinations.Additional magnetic resonance imaging(MRI)or biopsy was performed to avoid verification bias,according to the patients' preference in those with Breast Imaging-Reporting And Data System(BI-RADS)category 3,as well as 10%of those with BI-RADS categories 1-2 that were selected randomly.Image interpretation was performed separately.Women diagnosed with BI-RADS categories 4-5 on any imaging modality or BI-RADS category 3 along with an abnormal MRI had a core biopsy performed within three months of visible abnormality.2.In the national multicenter breast cancer screening trial,women aged 35-64 years old were recruited for breast cancer screening.All eligible participants underwent ABUS and HHUS screening,while women aged 45 years old or older also underwent MG screening.The most suspicious finding by ABUS,HHUS,and MG was taken for the referral.Women classified as BI-RADS 4-5 by any modality were considered to be screening positive and referred for biopsy,women classified as BI-RADS 3 will be followed-up within 6-12 months.Additional examinations will be referred for women classified as BI-RADS 0,while women with negative findings will be screened 3 years later.Results1.Among all the 1245 participants in the diagnostic study,a total of 321 breast cancer cases were diagnosed.MG detected 282 of them,with a sensitivity of 87.85%and a specificity of 91.56%.In MG-negative women,ABUS detected 35 additional breast cancers and caused 33 false-positive results.The sensitivity and specificity of the combination of ABUS with mammography was 98.75%and 87.99%,respectively.When using HHUS in MG-negative women,35 cases of breast cancer can be detected,and 49 cases of false-positive results were caused.The sensitivity and specificity of breast cancer diagnosis for the combination of HHUS and MG was 98.75%and 86.26%,respectively.ABUS and HHUS had a high consistency among MG negative women(percent agreement:95.93%,Kappa:0.74).Similar results were obtained in MG-negative women with dense breasts.When using ABUS in MG-negative women with dense breasts,the sensitivity and specificity of ABUS combined with MG was 96.88%and 88.85%,respectively;that for HHUS combined with MG was 96.88%and 87.34%,respectively.The additional cases detected by ultrasound were mainly distributed in women whose MG results were BI-RADS 1 or 3 with dense breasts.2.A total of 13606 women aged 35-64 years old were breast cancer screened.7.50%of them had abnormal breast examination findings,mainly with palpable masses.36.84%of them had abnormal ABUS findings,and 3.30%of them had positive ABUS screening results.35.15%of the participants had abnormal HHUS findings,and 2.96%had positive HHUS screening results.Approximately 57.70%of all 7208 women aged 45-64 who completed MG screening had dense breasts,and this proportion decreased with age.30.52%of them had abnormal MG findings,and 2.14%of the MG results were positive for breast cancer screening.There was moderate consistency between ABU S and HHUS,and weak consistency between ultrasound(both ABUS and HHUS)and MG.Conclusions1.For breast outpatients,both ABUS and HHUS as adjuncts to mammography can significantly improve the breast cancer detection rate,and there is a strong correlation between them,with less false-positive findings caused by ABUS than HHUS.2.In the screening population,ABUS and HHUS have similar detection rates of breast abnormalities,and they had moderate consistency.Both ABUS and MG had a weak correlation with MG.Given the high prevalence of dense breasts and the multiple advantages of ABUS over HHUS,such as less operator dependence,the separation of image acquisition and interpretation,ABUS showed great potential for being used in breast cancer early detection,especially in resource-limited areas.