Embodying (In)justice and Embodied Truths

Chapter 2 discusses application of ecosocial theory to analyze the health impacts of Jim Crow and its legal abolition, racialized and economic breast cancer inequities involving the breast cancer estrogen receptor, and the joint health impacts of physical and social hazards at work (including racism, sexism, and heterosexism) and relationship hazards (involving unsafe sex and violence). It also uses ecosocial theory to develop and apply measures of structural injustice, including historical redlining (1930s US government policies imposing racial residential segregation) and contemporary racialized economic segregation. The chapter additionally explains the construct of “emergent embodied phenotypes” and the different types of histories involved in disease processes: societal, individual (lifecourse), pathological/cellular, and evolutionary. It concludes by providing selected examples of how others, in diverse disciplines and settings worldwide, have employed the ecosocial theory of disease distribution to conceptualize and analyze embodiment of (in)justice across a wide range of exposures and outcomes.

Deep learning-enabled breast cancer hormonal receptor status determination from base-level H&E stains

For newly diagnosed breast cancer, estrogen receptor status (ERS) is a key molecular marker used for prognosis and treatment decisions. During clinical management, ERS is determined by pathologists from immunohistochemistry (IHC) staining of biopsied tissue for the targeted receptor, which highlights the presence of cellular surface antigens. This is an expensive, time-consuming process which introduces discordance in results due to variability in IHC preparation and pathologist subjectivity. In contrast, hematoxylin and eosin (H&E) staining—which highlights cellular morphology—is quick, less expensive, and less variable in preparation. Here we show that machine learning can determine molecular marker status, as assessed by hormone receptors, directly from cellular morphology. We develop a multiple instance learning-based deep neural network that determines ERS from H&E-stained whole slide images (WSI). Our algorithm—trained strictly with WSI-level annotations—is accurate on a varied, multi-country dataset of 3,474 patients, achieving an area under the curve (AUC) of 0.92 for sensitivity and specificity. Our approach has the potential to augment clinicians’ capabilities in cancer prognosis and theragnosis by harnessing biological signals imperceptible to the human eye.

Use of skincare products and risk of cancer of the breast and endometrium: a prospective cohort study

Background Concerns have been raised that extensive use of personal care products that contain endocrine disrupting compounds increase the risk of hormone sensitive cancers. Objective To assess the effect of skincare product use on the risk of pre- and postmenopausal breast cancer, estrogen receptor positive (ER+) and negative (ER-) breast cancer and cancer of the endometrium. Methods We used data from 106,978 participants in the population-based Norwegian Women and Cancer cohort. Participants were categorized into non-, light, moderate, frequent and heavy users of skincare products based on self-reported use of hand and facial cream and body lotion. Cancer incidence information from the Cancer Registry of Norway was linked to individual data through the unique identity number of Norwegian citizens. Multivariable Cox proportional hazard regression was used to assess the effect of skincare product use on the risk of cancer of the breast and endometrium. We used multiple imputation by chained equations to evaluate the effect of missing data on observed associations. Results We found no associations between use of skincare products and incidence of premenopausal breast cancer (frequent/heavy versus non−/light use: hazard ratio [HR] =1.10, 95% confidence interval [CI]: 0.92–1.32), postmenopausal breast cancer (heavy versus light use: HR = 0.87, 95% CI: 0.65–1.18, frequent versus light use: HR = 0.97, 95% CI: 0.88, 1.07) or endometrial cancer (frequent/heavy versus non−/light use: HR = 0.97, 95% CI: 0.79–1.20). Use of skincare products did not increase the risk of ER+ or ER- breast cancer and there was no difference in effect across ER status (0.58 ≤ pheterogeneity ≤ 0.99). The magnitude and direction of the effect estimates based on complete case analyses and multiple imputation were similar. Conclusion Heavy use of skincare products, i.e. creaming the body up to two times per day during mid-life, did not increase the risk of cancer of the breast or endometrium.