Epidemiological studies on breast cancer risk factors and screening

Abstract

This thesis aims to enhance cancer prevention by investigating the factors and outcomes associated with false-positive (FP) mammography recalls, as well as understanding the association between breast cancer risk factors of women and cancer risk among their relatives. Specifically, four studies were conducted using data from Swedish national registers, the Karolinska Mammography Project for Risk Prediction of Breast Cancer (KARMA) cohort, and the Linné-Bröst1 (Libro-1) cohort.

In Study I, we characterized factors associated with FP mammography recalls, comparing women with a FP recall to those who were not recalled and to those who had a true-positive recall (screen-detected cancer). We found that several mammographic and non-mammographic factors, as well as high breast cancer risk scores, were associated with having a FP recall. However, these factors were either equally or more strongly associated with having a truepositive recall.

In Study II, using a matched-cohort design, we examined the risk of subsequent breast cancer among women with a FP mammography recall. We observed a long-term increased breast cancer risk after a FP recall, compared with women who were not recalled. The elevated breast cancer risk differed by age and mammographic density at the matching mammography. In addition, the increased risk for breast cancer diagnosed on the ipsilateral side to the FP recall decreased over time and was highest within the first four years of follow-up.

In Study III, we investigated whether specific breast cancer risk factors in women were associated with their sisters' breast cancer incidence. We found that for women with high breast cancer risk prediction scores, benign breast disease (BBD), and high mammographic density, there was an increased risk of breast cancer for their sisters.

In Study IV, we investigated the associations of both carriership of protein-truncating variants (PTV) in eight genes and breast cancer polygenic risk scores (PRS) in women, with the risk of cancers in their first-degree relatives. We observed an elevated breast cancer risk among female relatives of women with PTVs, and among those with high breast cancer PRS. Additionally, we found a slightly elevated risk of cancers related to hereditary breast and ovarian cancer syndrome (HBOC)—other than breast cancer—among relatives of women with either high PRS or PTVs in the studied genes.

In summary, this thesis provides valuable information for both screening processes and genetic counseling. Although none of the studied factors are viable for interventions aimed at reducing FP recalls—due to the risk of simultaneously missing tumors—our results may aid in tailoring individualized surveillance plans for women with a FP recall. Additionally, our results suggest that women’s breast density and breast cancer risk scores—information that will be available at screening—may be useful for estimating the breast cancer risk in their sisters. Furthermore, PTVs in non-BRCA genes might offer insights into cancer aggregation in families. Overall, this thesis advances evidence-based cancer prevention in the era of precision medicine.