Optimization Of Breast Cancer Screening Programme Of Female Based On Risk Levels In Beijing

Background:Breast cancer has become the most common malignancy of female in the world.The incidence rate of breast cancer in Beijing is also increasing year by year.On the other hand,the mortality rate of breast cancer is decreased.The decreasing of mortality benefits from the progress of breast cancer treatment,as well as,the early diagnosis of breast cancer.Breast cancer screening is an important way of early diagnosis.The United States and other western countries have greatly improved the early diagnosis rate of breast cancer and reduced the mortality rate of breast cancer through the breast cancer screening program form the late 1970s.They also established a variety of breast cancer risk prediction models to guide breast cancer screening strategies.The characteristics of female breast cancer in China are different from those in western countries and we have lack experience on breast cancer screening.The breast density of Chinese women is generally higher than that of Western women.The average age at which Chinese women diagnosed breast cancer was 45-55 years old,about 10 years earlier than Western women.And the incidence rate of breast cancer in China is still lower than incidence rate in western countries.The manpower and material resources consumed by the general survey of breast cancer will cause huge burden to the national economy.The Western strategies of breast cancer screening based on mammography may not be suitable for China.Objective:To analyze the impact of breast cancer risk factors and risk levels on breast cancer incidence rate among opportunistic screening patients in Beijing,and to validate the PUMCH model and than optimize it.To analyze the influence of risk factors and risk levels on the test efficiency of different screening strategies,and optimize the screening program for female breast cancer in Beijing.Methods:Through extensive publicity and education,women aged 35-75 were recruited to participate in opportunistic screening.From January 1,2016 to January 1,2019,women who underwent opportunistic screening for breast cancer in 10 centers in Beijing entered the cohort.All the women were examined clinically,and the breast volume information(bra cup)was recorded,and the follow-up card was issued.They were randomly divided into ultrasound group or ultrasound combined with mammograph group.The screening results and breast density information were recorded.The patients were followed up by telephone and outpatient medical records to record the breast biopsy and biopsy pathology.All patients were followed up until January 1,2020.Logistic regression model was used to analyze the correlation between risk factors and the incidence of breast cancer.Receiver operating curve(ROC),AUC and R2 were used to evaluate the predictive ability of PUMCH model.Decision curve analysis(DCA)was used to evaluate the clinical value of the model.Hosmer-lemeshow test and calibration curve were used to test the model fitting.Nomogram was used to show the modified PUMCH model with breast density.The sensitivity,specificity,accuracy,positive predictive value,negative predictive value and test accuracy of different screening methods and groups were calculated.Receiver operating characteristic(ROC)curve was used to analyze the difference between groups.The diagnostic cost of breast cancer is calculated as(the cost of screening program+the cost of biopsy*the positive rate of screening)/(sensitivity of the screening scheme*incidence rate of screening breast cancer within 1 year).Results:A total of 10537 eligible women were enrolled in 10 centers in Beijing.As the largest center,our hospital enrolled 5241.The average age of the enrolled women was 47.48±8.32 years,the median age was 47 years,and the median follow-up time was 38 months.After data quality control,a total of 10381 cases in 10 centers were included in the statistical analysis,and 161 cases of breast cancer occurred.A total of 5228 cases of single center data were included in the statistical analysis,a total of 78 cases of breast cancer.Risk factors analysis showed that age 55-65,long-term mental depression,contraceptive/MHT use,high density of breast increased the risk of breast cancer(P<0.05).The risk of breast cancer in the middle risk group is 2-3 times that of the low risk group,the high risk group is 2.5-4 times,and the most highest risk group is 3.4-5 times.It is verified that PUMCH model has good discrimination,calibration and clinical application value.The PUMCH nomogram model optimized by adding breast density factor can improve the prediction ability of the original model.There was no significant difference between automatic ultrasound group and manual ultrasound group(Delong test p>0.05).The diagnostic efficiency of automatic ultrasound+mammography group was better than that of manual or automatic ultrasound group(Delong test p<0.05).Compared with the three different screening method,the sensitivity of breast ultrasound was higher than that of mammography,the specificity was slightly lower,and the detection efficiency of ultrasound was better than that of mammography(Delong test,P<0.05).In multi center data,the efficiency of breast ultrasound combined with mammography examination was significantly higher than that of breast ultrasound alone(Delong test,P<0.05).For women aged 35-55 years,women with small breast volume,women with high breast density,and women with low risk,ultrasound combined with mammograph can not improve the diagnostic efficiency(Delong test,P>0.05),and breast ultrasound alone can be used for screening.Ultrasound combined with mammography can improve the diagnostic efficiency of 55-75 years old women,women with large breast volume,women with low breast density,and women with high risk(Delong test,P<0.05),Ultrasound combined with mammography can be used for screening.According to the cost of breast ultrasound,mammography and biopsy in Beijing,the annual cost of ultrasound for one case of breast cancer is about 25000-30000 yuan/year,that of ultrasound combined with mammography is about 48000-53000 yuan/year,and that of mammography is about 59000-65000 yuan/year.Conclusion:PUMCH model has good discrimination,calibration and clinical application value.Adding breast density into PUMCH nomogram model can improve the prediction ability of PUMCH model.Age and breast density have influence on the positive rate of screening results.There was no significant difference in diagnostic efficiency between automatic ultrasound group and manual ultrasound group.The diagnostic efficiency of automatic ultrasound combined with mammography group was better than that of manual or automatic ultrasound group.The sensitivity of breast ultrasound is higher than that of mammography,and the specificity is slightly lower.For women aged 35-55,women with small breast volume,women with high breast density and women with low risk,ultrasound combined with molybdenum target screening can not improve the diagnostic efficiency,and breast ultrasound alone can be used for screening.Ultrasound combined with mammography can improve the diagnostic efficiency of 55-75 years old women,women with large breast volume,women with low breast density,and women with medium or high risk.The screening cost of ultrasound is the lowest,mammography is the highest,while ultrasound combined with mammography cost is in the middle.