Recently, the impact of patients’ eating habits on both breast cancer (BC) management and inflammation have been proven. Here, we investigated whether inflammatory habits could correlate with baseline bowel [18]F-fluorodeoxyglucose (FDG) uptake and the latter, in turn, with pathological Complete Response (pCR) to neoadjuvant chemotherapy (NAC). We included stage I–III BC undergoing standard NAC at IRCCS Humanitas Research Hospital, Italy. Patients fulfilled a survey concerning eating/lifestyle behaviors and performed a staging [18]F-FDG positrone emission tomography/computed tomography (PET/CT). In the absence of data on the effects of individual foods, we aggregated drink and food intake for their known inflammatory properties. Data were recorded for 82 women (median age, 48). We found positive correlations between colon mean standardized uptake value (SUVmean) and pro-inflammatory drinks (alcohol and spirits; r = +0.33, p < 0.01) and foods (red and cured meats; r = +0.25, p = 0.04), and a significant negative correlation between rectum SUVmean and anti-inflammatory foods (fruits and vegetables; r = −0.23, p = 0.04). Furthermore, colon SUVmean was significantly lower in patients with pCR compared to non pCR (p = 0.02). Our study showed, for the first time, that patients’ eating habits affected bowel [18]F-FDG uptake and that colon SUVmean correlated with pCR, suggesting that PET scan could be an instrument for identifying patients presenting unhealthy behaviors.

Widespread interest in artificial intelligence (AI) in health care has focused mainly on deductive systems that analyze available real-world data to discover patterns not otherwise visible. Generative adversarial network, a new type of inductive AI, has recently evolved to generate high-fidelity virtual synthetic data (SD) trained on relatively limited real-world information. The AI system is fed with a collection of real data, and it learns to generate new augmented data while maintaining the general characteristics of the original data set. The use of SD to enhance clinical research and protect patient privacy has drawn a lot of interest in medicine and in the complex field of oncology. This article summarizes the main characteristics of this innovative technology and critically discusses how it can be used to accelerate data access for secondary purposes, providing an overview of the opportunities and challenges of SD generation for clinical cancer research and health care.

Several multigene assays have been developed to help clinicians in defining adjuvant treatment for patients with hormone-receptor-positive (HR+), human epidermal growth factor receptor-2 (HER2)–negative early breast cancer. Despite the 21-gene assay having been available for decades, it has only recently been included in the healthcare systems of several countries. Clinical optimisation of the test remains of critical interest to achieve a greater impact of genomic information in HR+/HER2- early breast cancer. Although current guidelines recommend the use of the 21-gene assay in early breast cancer at intermediate risk of relapse, the implication of the Recurrence Score (RS) in some grey areas still remains uncertain. Our aim is to critically discuss the role of RS in peculiar circumstances. In particular, we focus on the complex integration of genomic data with clinicopathological factors; the potential clinical impact of RS in node-positive premenopausal women and in the neoadjuvant setting; the significance of RS in special histologies and in male patients; and the management and time-optimisation of test ordering. In the absence of robust evidence in these areas, we provide perspectives for improving the use of the 21-gene assay in the decision-making process and guide adjuvant treatment decisions even in challenging cases.