Abstract Introduction: In 2000 we published initial observations from a high-risk cohort of 480 women that cytologic evidence of hyperplasia with atypia in benign breast tissue obtained by random periareolar fine-needle aspiration (RPFNA) was associated with a five-fold increased risk of developing DCIS or invasive breast cancer at a median follow-up of 45 months [Fabian, JNCI 2000]. Few women in the initial cohort had any exposure to prevention agents, as NSABP-P1 was not reported until 1998. We began a new high-risk cohort in 2002 as the tissue processing changed from a filter technique to Thin PrepTM. The superior nuclear detail permitted assessment of cytologic atypia in the absence of high cellularity. Women in the second cohort were told that RPFNA atypia was a risk factor for developing breast cancer and in addition to standard prevention options were offered participation in clinical trials, if applicable. The purpose of this subsequent analysis was to determine if the predictive value of RPFNA atypia was maintained with change in tissue processing and availability of prevention options. Methods: A total of 1,135 high-risk women, eligible on the basis of family history, BRCA1/2 mutation status, prior biopsy indicating LCIS or atypical hyperplasia (AH), prior contralateral breast cancer and/or high mammographic breast density, underwent baseline RPFNA and were enrolled in our second cohort between 2002 and 2015. As in the earlier study published in 2000, if women had 2 aspirations without an intervening intervention within 21 months the worse result was utilized as baseline. In addition to cytomorphology, breast tissue was also categorized by cellularity of the aspirate (10-100, 100-1000, 1000-5000, and >5000 cells per cytology slide). The primary aspirator (CJF) and cytopathologist (CMZ) were the same for both cohorts. Women were censored at the time of prophylactic mastectomy, development of other site cancer, or death. Kaplan-Meier hazard plots and Cox-regression analysis were used to analyze time to development of DCIS or invasive breast cancer and effect of joint variables, respectively. Results: At a median follow-up of 86 months, 79 cases of DCIS or invasive breast cancer had been diagnosed. By univariate analysis, RPFNA atypia at entry was not predictive of subsequent breast cancer (p=0.58; log-rank test). However, breast cancer risk was increased by a prior breast biopsy with LCIS or AH (HR 2.6, 95% CI 1.6-2.4, p<0.001) or by high RPFNA cellularity, defined as >5000 epithelial cells per RPFNA cytomorphology slide (HR 2.1, 95% CI 1.3-3.5, p=0.0034). Thirty-six percent of women subsequently chose to undergo a prevention intervention, including a standard drug, enrollment in a prevention clinical trial of 6-12 months duration, or a prophylactic bilateral salpingo-oophorectomy (BSO) prior to age 45. Prevention interventions were more prevalent in women with RPFNA atypia (49% vs 30%, p<0.001). Not only was uptake of a prevention intervention associated with reduced breast cancer risk on univariate analysis (p=0.043), it was retained on multivariate analysis. The final Cox Regression risk model included prior breast biopsy with LCIS or AH (HR 2.4, 95% CI 1.5-3.9; p<0.001), high RPFNA cellularity (HR 2.2, 95% CI 1.3-3.7; p=0.003) and prevention intervention (HR 0.54, 95% CI 0.33-0.89; p=0.015). Conclusions: In a second large cohort of high-risk women for whom RPFNA acquired cells were processed to slides via ThinPrep™, high RPFNA cellularity (>5000 epithelial cells/slide) but not cytologic atypia predicted short-term breast cancer risk. However, women with atypia were significantly more likely than women without atypia to participate in a prevention intervention, which in turn was associated with reduced breast cancer risk Citation Format: Whitney L Hensing, Carol J Fabian, Carola M Zalles, Priyanka Sharma, Amy L Kreutzjans, Kandy R Powers, Lynn Chollet-Hilton, Bruce F Kimler. Random periareolar fine-needle aspiration (RPFNA) cell number, prior precancerous breast disease and subsequent uptake of a prevention intervention predict short-term breast cancer risk [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P2-10-02.
Read full abstract