Integrating Molecular Biomarker Inputs Into Development and Use of Clinical Cancer Therapeutics

Biomarkers can contribute to clinical cancer therapeutics at multiple points along the patient's diagnostic and treatment course. Diagnostic biomarkers can screen or classify patients, while prognostic biomarkers predict their survival. Biomarkers can also predict treatment efficacy or toxicity and are increasingly important in development of novel cancer therapeutics. Strategies for biomarker identification have involved large-scale genomic and proteomic analyses. Pathway-specific biomarkers are already in use to assess the potential efficacy of immunotherapy and targeted cancer therapies. Judicious application of machine learning techniques can identify disease-relevant features from large data sets and improve predictive models. The future of biomarkers likely involves increasing utilization of liquid biopsy and multiple samplings to better understand tumor heterogeneity and identify drug resistance.

Automated summary

Biomarkers play a crucial role in cancer therapeutics, aiding in screening, classification, survival prediction, and treatment efficacy prediction. The application of machine learning techniques can help identify disease-relevant features and improve predictive models.