# Proof of Concept for Tumor Mutational Burden Prediction Through Biophysical Analysis Based on UHF-Dielectrophoresis

## TL;DR

This paper explores using electromagnetic sensing to predict tumor mutational burden, a key biomarker for cancer immunotherapy, as a faster and cheaper alternative to sequencing.

## Contribution

The study introduces UHF-dielectrophoresis as a novel biophysical method for assessing tumor mutational burden.

## Key findings

- EMS measurements showed an upward shift correlated with higher tumor mutational burden in cancer cell lines.
- ENU-induced mutagenesis was reliably detected through changes in electromagnetic signatures.
- The method can detect both intrinsic and experimentally induced mutational load differences.

## Abstract

Tumor Mutational Burden (TMB) is a critical biomarker used to determine patient eligibility for immunotherapy with immune checkpoint inhibitors. However, its gold-standard assessment via whole exome sequencing is limited by high costs, technical complexity, and lengthy processing times. To address these challenges, we investigated whether Ultra-High-Frequency (UHF) electromagnetic wave sensing could serve as an alternative method for evaluating TMB. We analyzed the dielectrophoresis crossover frequency spectrum and corresponding electromagnetic signature (EMS) of cancer cells using a lab-on-a-chip biosensor that integrates microfluidics with dielectrophoresis-based electro-manipulation. Across seven solid tumor cell lines exhibiting diverse TMB levels, EMS exhibited an upward shift correlated with higher TMB, suggesting a relationship between mutational load and electromagnetic behavior. To further explore this connection, we artificially increased the somatic variant burden by exposing cells to the mutagen N-ethyl-N-nitrosourea (ENU). EMS measurements reliably detected the induced increase in variant load in ENU-treated cells. Overall, these findings demonstrate that EMS can detect both intrinsic TMB differences and experimentally induced increases in mutational burden, enabling refined categorization of cancer cells. Although further validation is required, this work lays the foundation for developing complementary, rapid, and accessible tools to support cancer cell stratification and guide immunotherapy decision-making.