Artificial intelligence methods are been increasingly used for analysis of pathology slides. In this issue of the Journal, Sandbank et al. describe the validation and utility of a robust second reader system that can distinguish in situ and standfirst invasive carcinomas from non-neoplastic lesions of the breast.

The combination of an endocrine agent with a CDK4/6 inhibitor is the standard of care in the first-line setting for patients with hormone receptor-positive, HER2-negative metastatic breast cancer. Randomized trials have demonstrated similar and significant improvements in progression-free survival using the three available CDK4/6 inhibitors and led to regulatory approval. However, mature overall survival data now suggest potential differences among the various agents, suggesting an evolution in selection preferences.