Ex Vivo Treatment Response Prediction in Multiple Myeloma: Assay Formats, Clinical Correlation, and Future Directions

TL;DR

This review examines ex vivo tests for predicting drug responses in multiple myeloma, focusing on those with clinical relevance and potential to improve patient outcomes.

Contribution

The paper uniquely reviews ex vivo tests with demonstrated clinical correlation, offering insights for future assay development.

Findings

2D assays are the most mature but lack full bone marrow microenvironment representation.

3D models better preserve tumor heterogeneity and show emerging clinical relevance.

Dynamic systems are physiologically ambitious but remain technically complex and underdeveloped for broad clinical use.

Abstract

This review focuses specifically on ex vivo tests for multiple myeloma drug response prediction that have demonstrated some level of clinical correlation. This creates a key distinguishing theme when compared to prior reviews in the field. By reviewing these diverse tests, their characteristics, and particularly their maturity and demonstrated clinical utility, we hope to provide a key reference manuscript that researchers and physicians alike will find invaluable as they attempt to make decisions about the design, development, or real-world clinical use of tests like these that have the potential to improve care and survival times in multiple myeloma patients.

Ex vivo functional testing for multiple myeloma is rapidly evolving, yet no single assay has reached the level of reliability and clinical utility needed for routine decision-making. Existing approaches generally fall into three categories: 2D cultures, 3D models, and dynamic systems. Each contributes valuable but incomplete insight into therapeutic response. Among these, 2D assays remain the most mature, with the most extensive clinical correlations to date, though their simplified architecture limits their ability to reflect the full complexity of the bone marrow microenvironment. However, 3D systems, including spheroids and matrix-based organoids, offer improved preservation of tumor heterogeneity and microenvironmental cues. These platforms show emerging clinical relevance and may hold advantages over traditional 2D formats, and validation efforts are developing. Dynamic systems, including microfluidic models and perfused bone-marrow mimetics, represent the most physiologically ambitious category, yet their technical intricacy and early stage of development have so far limited broad clinical correlation. Altogether, the current landscape highlights substantial progress but lacks an optimal assay. In this review, we take the unique approach of examining published ex vivo tests that have demonstrated a level of clinical correlation. We evaluate their respective formats, strengths and limitations, and discuss considerations for what an ideal future assay may encompass.

Figures