Abstract 6469: Associations between serum iron levels and breast cancer tumor size

Background: Iron metabolism is disrupted in cancer cells, with well-established associations between increased iron availability and cancer growth. Excess iron can be toxic, and animal models have demonstrated that excess iron promotes tumor growth, whereas iron deficiency reduces or slows tumor growth. Few observational studies have assessed associations between serum iron levels and human tumor characteristics. The objective of this study is to estimate the associations between serum iron levels and tumor size at diagnosis in breast cancer cases from The Sister Study, a large prospective cohort based in the United States. Methods: Data are from a case-cohort sample of the Sister Study selected in 2016 to study serum iron levels. Of the 3,007 invasive or ductal carcinoma in situ cases included, we used the 2,494 incident breast cancer cases with information on tumor size and iron levels. Serum iron (mcg/dL), ferritin (mcg/dL), and percent transferrin saturation were measured in samples collected at baseline (median time from collection to diagnosis 4.6 years). Prior to modeling, we applied a natural log transformation to the largest tumor size and ferritin levels. To estimate the associations between body iron levels and tumor size, we used Spearman’s rank correlation and linear regression models adjusting for body mass index and age at study entry. If there were multiple tumor measures for one person, we selected the largest tumor size. Sensitivity analyses included: 1) restricting analyses to diagnoses more than 6 months after or less than 4 years after baseline iron measures, 2) excluding women who reported taking iron supplements, and 3) stratifying analyses by tumor subtypes. Results: Associations between all three serum iron levels and tumor size were close to the null, indicating no evidence of an association between iron and tumor size. Spearman’s rank correlations were all close to zero. Adjusted regression slopes (95% confidence interval) were -0.0005 (-0.0010, 0.0005) for serum iron, -0.0163 (-0.055, 0.022) for ferritin, and -0.0009 (-0.004, 0.0021) for percent transferrin saturation. Results from the sensitivity analyses did not substantively differ from our primary findings. Conclusions: We did not find evidence to support the hypothesis of a positive association between breast cancer tumor size at the time of diagnosis and serum iron levels. The conflicting evidence between this study and previous research in animal models suggests that iron in the human tumor microenvironment may operate independently of circulating iron or body iron stores. Citation Format: Ann Von Holle, Rachel L. Thompson, Katie M. O'Brien, Dale P. Sandler, Clarice R. Weinberg. Associations between serum iron levels and breast cancer tumor size [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6469.