# Integrated Single-Cell and Spatial Transcriptomics Coupled with Machine Learning Uncovers MORF4L1 as a Critical Epigenetic Mediator of Radiotherapy Resistance in Colorectal Cancer Liver Metastasis

## TL;DR

This study identifies MORF4L1 as a key factor in radiotherapy resistance in colorectal cancer liver metastasis using advanced genomic and machine learning techniques.

## Contribution

The study introduces MORF4L1 as a novel epigenetic mediator of radioresistance in CRLM, supported by integrated single-cell and spatial transcriptomics.

## Key findings

- MORF4L1-high tumor cells show enhanced DNA repair and metabolic pathways linked to radioresistance.
- MORF4L1-expressing tumor regions spatially co-localize with immunosuppressive immune cells like Tregs and TAMs.
- Radiotherapy drives tumor cells toward a resistant state marked by increased MORF4L1 expression.

## Abstract

Background and Objective: Colorectal cancer (CRC) liver metastasis (CRLM) represents a major clinical challenge, and acquired resistance to radiotherapy (RT) significantly limits therapeutic efficacy. A deep and comprehensive understanding of the cellular and molecular mechanisms driving RT resistance is urgently required to develop effective combination strategies. Here, we aimed to dissect the dynamic cellular landscape of the tumor microenvironment (TME) and identify key epigenetic regulators mediating radioresistance in CRLM by integrating cutting-edge single-cell and spatial omics technologies. Methods and Results: We performed integrated single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics (ST) on matched pre- and post-radiotherapy tumor tissues collected from three distinct CRLM patients. Employing a robust machine-learning framework on the multi-omics data, we successfully identified MORF4L1 (Mortality Factor 4 Like 1), an epigenetic reader, as a critical epigenetic mediator of acquired radioresistance. High-resolution scRNA-seq analysis of the tumor cell compartment revealed that the MORF4L1-high subpopulation exhibited significant enrichment in DNA damage repair (DDR) pathways, heightened activity of multiple pro-survival metabolic pathways, and robust signatures of immune evasion. Pseudotime trajectory analysis further confirmed that RT exposure drives tumor cells toward a highly resistant state, marked by a distinct increase in MORF4L1 expression. Furthermore, cell–cell communication inference demonstrated a pronounced, systemic upregulation of various immunosuppressive signaling axes within the TME following RT. Crucially, high-resolution ST confirmed these molecular and cellular interactions in their native context, revealing a significant spatial co-localization of MORF4L1-expressing tumor foci with multiple immunosuppressive immune cell types, including regulatory T cells (Tregs) and tumor-associated macrophages (TAMs), thereby underscoring its role in TME-mediated resistance. Conclusions: Our comprehensive spatial and single-cell profiling establishes MORF4L1 as a pivotal epigenetic regulator underlying acquired radioresistance in CRLM. These findings provide a compelling mechanistic rationale for combining radiotherapy with the targeted inhibition of MORF4L1, presenting a promising new therapeutic avenue to overcome treatment failure and improve patient outcomes in CRLM.