Background: Since 2007, U.S. national guidelines recommend cancer-free women at ≥20% lifetime risk of breast cancer (BC) undergo BC screening with mammogram and breast MRI. Little is known about long-term adherence to BC screening among young high-risk women. To address this knowledge gap, we examined predictors of screening adherence over time among women with varying BC risk.

Methods: Eligible women were ≥ 30 years old, had no history of BC/ovarian cancer, an intact breast, are enrolled in the Breast and Ovarian Surveillance Service (BOSS) Cohort, and visited the Johns Hopkins Cancer Genetics Clinic for risk assessment ≥ 2004 but prior to cohort enrollment (N = 374). Screening recommendations were determined from Tyrer-Cuzick BC lifetime risk estimates obtained at the clinic visit. Screening was self-reported at baseline, 4, and 8 years. A subset has been validated. An adherent screen before 2007 was mammography for women <20% risk at age ≥ 40 years or ≥20% risk/BRCA1/2 positive (high-risk). After 2007, an adherent screen for high-risk women was mammography and MRI. Cumulative adherence up to 2017 was also calculated. Associations between estimated risk, patient characteristics, and screening adherence were quantified using logistic regression.

Results: At baseline, the median age was 47 years, 31% had lifetime risk <20%, 60% had risk ≥20%, 8.8% were BRCA1/2 positive, and 69.5% were adherent to population BC screening recommendations. Over a median follow-up of 7.5 years, 9% of women underwent prophylactic bilateral mastectomy and 5% were diagnosed with BC. At each time point, proportion of mammography and MRI was ~80% and ~22.5%, respectively. The ≥20% group had better mammography adherence than the <20% group at 4 years [OR=2.0; 95%CI=1.1-3.7] and 8 years [OR=2.2; 95%CI=1.2-3.8], but was less adherent to risk-appropriate screening than the <20% group at 4 and 8 years [OR=0.2; 95%CI=0.1-0.4] [OR=0.4; 95%CI=0.2-0.6], respectively. Adherence was similar in the ≥20% and ≥30% high-risk groups. Risk-appropriate adherence at 4 years was predictive of adherence at 8 years [OR=11.1; 95%CI=5.5-22.5], adjusted for estimated risk, age, year of risk assessment, parity, other types of screening, comorbidities, and baseline adherence. In a multivariable model of cumulative risk-appropriate adherence, women with ≥20% risk (vs. <20%) or women with comorbidities (vs. no comorbidities) were less likely to consistently screen over 8 years [both OR=0.3; 95%CI=0.1-1.0], while women of older ages (vs. younger) were more likely [OR=1.2; 95%CI=1.1-1.3].

Conclusion: High-risk young women were not adherent to the more intensive screening regimen, and adherence did not improve over time. Other strong predictors of adherence included older age, prior adherence, and fewer comorbidities. New screening alternatives and interventions targeting reasons for non-adherence should be evaluated for high-risk women.

Citation Format: Marcy L. Schaeffer, Betty J. May, Brenna C. Hogan, Mikiaila M. Orellana, Dana Petry, Katie Fiallos, Michelle S. McCullough, Deborah K. Armstrong, Kelly Myers, Kala Visvanathan. Longitudinal examination of breast cancer screening adherence among women in a prospective familial cohort [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2423.