Letter to the editor: Ethnic and age differences in right-left breast asymmetry in a large population-based screening population

To the Editor,

We read with great interest the recent article by Hudson and colleagues in the November issue of British Journal of Radiology entitled “Ethnic and age differences in right-left breast asymmetry in a large population-based screening population.”¹ The authors should be commended for the largest population-based analysis of breast volume using an automated assessment of mammography in healthy females. Our biggest conclusion of this important study is that breast volume asymmetry is common among various age and ethnic populations and not an indication of underlying pathology. The question of breast asymmetry and an association with breast cancer is challenging as there are several ways to measure breast asymmetry and multiple confounding factors that can contribute to differences from one breast to the other.

Our group is interested in breast volume and symmetry from another perspective, one of the reconstruction after breast cancer treatment.² We agree that there is a paucity of normative data and predict that perfect symmetry is more of a hypothetical goal than a realistic expectation. Our methods included clinical measurements (e.g., nipple position, breast ptosis) and volume measurements on three-dimensional surface images (i.e., three-dimensional photographs); however, we found similar trends in our pre-operative breast cancer patients. Of particular interest, more than 50% of our patients exhibited a volume difference greater than 50 ml. This is consistent with Hudson et al.’s observation that the median (25th, 75th centiles) absolute difference in breast volume was 60.6 (26.6, 117.8) cm³, with this trend held at every age and ethnic group in the study.

Our data also showed that race and BMI were associated with breast asymmetry. We used the absolute volume ratio for asymmetry and adjusted for age, BMI, and cancer status. The multiple linear regression model results indicated race is a factor for breast asymmetry, with Caucasians having more symmetry than non-Caucasians (p = 0.043). Similarly, the authors show that race is a significant factor after adjusting breast volume (p for homogeneity <0.01).

Interestingly, we found that tumor size did not impact overall symmetry. It would be critical to know which patients in Hudson et al.’s study went on to develop breast cancer, and to what degree they exhibited breast volume asymmetry or density volume asymmetry. We suspect that overall volume asymmetry will be relatively common, while density asymmetry will carry more clinical relevance.

Ethnic and age differences in right–left breast asymmetry in a large population-based screening population

¹Sue M Hudson BSc MSc, ²Louise S Wilkinson, ³Rachel Denholm, ⁴Bianca L De Stavola and ¹Isabel dos-Santos-Silva

¹Department of Non-Communicable Disease Epidemiology, London School of Hygiene and Tropical Medicine, London, UK.

²Oxford Breast Imaging Centre, University of Oxford Hospitals NHS Foundation Trust, Oxford, UK

³Centre for Academic Primary Care, Bristol Medical School, University of Bristol, Bristol, UK

⁴Population, Policy and Practice Programme, Great Ormond Street Institute of Child Health, University College London, London, UK

REPLY TO THE LETTER TO EDITOR

We would like to thank Dr Hanson and her colleagues for their interest in our recently published paper.¹ It is reassuring to note that their data on pre-operative breast cancer patients also show that left–right breast volume asymmetry is common among different ages and ethnic groups. In particular, the magnitude of the right–left breast volume asymmetry values appears to be remarkably similar across our two studies despite differences in the methodologies used. We used a fully automated algorithm to estimate the volume of each breast from two-dimensional digital mammographic raw images¹ while in Hanson et al. the breast volumes were based on clinical measurements and three-dimensional photographs.

The authors of the letter expressed an interest in knowing which patients in our study went on to develop breast cancer, and to what degree they exhibited right–left asymmetry in breast volume or in density volume (i.e., in the volume of radiologically dense tissue). They suspected that overall volume asymmetry will be relatively common, while density asymmetry will carry more clinical relevance. We agree with Hanson et al that these are important research questions which need to be clarified. We have just completed a new investigation within the same population of our previous study¹ to examine whether left–right asymmetries in breast volume and in density volume are associated with the risk of being diagnosed with cancer at the same or subsequent screens, or between screens as an interval cancer. A manuscript reporting the findings from this new research is currently undergoing peer review. Hopefully, this new study will address the issues raised by Hanson et al. Of particular interest, Hanson et al. stated that they “found that tumor size did not impact overall symmetry” (sic). In our new study, we have also found that tumor size, as measured at surgical excision, was not associated with degree of breast volume asymmetry.