Identification Of Women At High Risk Of Breast Cancer And In Need Of Supplemental Screening - A Cohort Study

Abstract Background. Mammography screening reduces breast cancer mortality, but a large proportion of breast cancers are missed and detected at later stages or develop in between screening intervals. We developed the KARMA model which identifies women who are likely to be diagnosed with breast cancer before or at the next screen. Materials and Methods. The study was based on the prospective screening cohort KARMA including 70,877 participants. We identified 974 incident cancers and sampled 9,376 healthy individuals from the cohort. An image-based risk score was developed using mammographic features (density, masses, microcalcifications), their left-right asymmetries, and age. The lifestyle extended score also included menopausal status, family history of breast cancer, body-mass-index, hormone replacement therapy, and use of tobacco and alcohol. The genetic extended score also included a polygenic risk score including 313 single nucleotide polymorphisms. Relative risks were estimated using age stratified logistic regression. Tumor sub-type specific risks were estimated. Absolute risks were estimated including relative risks and national incidence rates. Results. The image-based model reached an area under the curve (AUC) of 0.73 (95% CI 0.71,0.74). The lifestyle and genetic extended model AUCs were 0.74 (95% CI 0.72,0,75) and 0.77 (95% CI 0.75,0.79) respectively. There was a relative 8-fold difference in risk between the women at high and general risk. High risk women were more likely diagnosed with stage II and >= 20 mm tumors and less likely with stage I and estrogen receptor-positive tumors. The image-based model was validated in two external cohorts. Conclusion. By combining three mammographic features, their left-right asymmetries, and optionally lifestyle factors, family history, and a polygenic risk score we generated a model that identifies women at high likelihood of being diagnosed with breast cancer within two years of a negative screen and in possible need of supplemental screening or preventive intervention. Polygenic Risk Score included 313 SNPs.Mammographic density were adjusted for age and BMI. Table 1. Discrimination performance (AUC) of the risk score in relation to the three models of the study. The 2-year risk Model 1 is compared with the two external validation datasets.ModelAUC (95% CI)1KARMA case-cohort (974 cancers, 9,376 healthy subjects)1. Model 1; mammographic density, microcalcifications, masses, age0.73 (0.71,0.74)2. Model 2; Model 1 + lifestyle and familial risk factors20.74 (0.72,0.75)3. Model 3; Model 2 + PRS30.77 (0.75,0.79)MBTST cohort (104 cancers, 9,745 healthy subjects), Model 10.71 (0.67,0.75)CSAW (613 cancers, 8,489 healthy subjects), Model 10.73 (0.71,0.76)KARMA independent test set (179 cancers, 9,491 healthy subjects), Model 10.73 (0.69, 0.77)Polygenic risk score included 313 SNPs. Table 2. Comparison discrimination performance (AUC) of the PRS, Tyrer-Cuzick, and Gail risk scores with and without mammographic density in KARMA case-cohort.ModelAUC (95% CI)1PRS20.64 (0.62,0.66)PRS2 + mammographic density30.67 (0.65,0.69)Tyrer-Cuzick40.58 (0.56,0.60)Tyrer-Cuzick4 + mammographic density30.62 (0.60,0.64)Gail50.56 (0.54,0.58)Gail5 + mammographic density30.61 (0.60,0.63) Citation Format: Mikael Eriksson, Kamils Czene, Per Hall. Identification of women at high risk of breast cancer and in need of supplemental screening - A cohort study [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS8-01.