Eduard Vrdoljak

Background: The treatment landscape in HR+/HER2− metastatic breast cancer (mBC) is continuously evolving, with evidence on new agents and combinations published almost every year. Despite updated therapeutic guidelines, second-line (2L) selection may be challenging due to clinical factors, biomarker status, and available agents. Methods: A two-round Delphi consensus was organized in July 2024, gathering input from 10 experts in research, diagnosis, and treatment of HR+/HER2− mBC on optimal 2L and beyond choice, considering the available biomarkers and results from published clinical trials. Consensus was defined as 70% agreement or disagreement. Results: The experts considered initially a list of 39 statements, structured into the following four sections: biomarker testing; selection of 2L treatment at progression of disease on first line endocrine therapy (ET) + CDK4/6i at ≥6 months after initiation of ET for mBC; selection of 2L treatment at disease progression on ET + CDK4/6i, at <6 months after initiation of ET for mBC, whilst on ET; and selection of post-2L treatment options. After a discussion, the experts decided to remove four statements, refine ten, and include three new ones. The final list consisted of 38 statements, and consensus was achieved in 37. Conclusions: The panel recommends next-generation sequencing as the method of choice for genomic characterization at disease progression on first line. The optimal agent or treatment class is indicated depending on the presence of specific mutations; however, the panel admits that the strategy is different in clinical practice, where novel therapies might not be available or reimbursed.

Background: Due to recent changes in breast cancer treatment strategy, significantly more patients are treated with neoadjuvant systemic therapy (NST). Radiological methods do not precisely determine axillary lymph node status, with up to 30% of patients being misdiagnosed. Hence, supplementary methods for lymph node status assessment are needed. This study aimed to apply and evaluate machine learning models on clinicopathological data, with a focus on patients meeting NST criteria, for lymph node metastasis prediction. Methods: From the total breast cancer patient data (n = 8381), 719 patients were identified as eligible for NST. Machine learning models were applied for the NST-criteria group and the total study population. Model explainability was obtained by calculating Shapley values. Results: In the NST-criteria group, random forest achieved the highest performance (AUC: 0.793 [0.713, 0.865]), while in the total study population, XGBoost performed the best (AUC: 0.762 [0.726, 0.795]). Shapley values identified tumor size, Ki-67, and patient age as the most important predictors. Conclusion: Tree-based models achieve a good performance in assessing lymph node status. Such models can lead to more accurate disease stage prediction and consecutively better treatment selection, especially for NST patients where radiological and clinical findings are often the only way of lymph node assessment.