# TUSCAN: Tumor segmentation and classification analysis in spatial transcriptomics

## TL;DR

TUSCAN is a new tool that improves tumor identification in tissue samples by using copy number variations and images, offering better accuracy and insights into cancer evolution.

## Contribution

TUSCAN introduces a novel method for tumor segmentation using spatial copy number variation profiles combined with histological images.

## Key findings

- TUSCAN outperforms existing methods in accurately delineating tumor regions in spatial transcriptomics data.
- The method provides interpretable insights into clonal evolution and tumor heterogeneity.
- Combining copy number variations with histological images enhances tumor identification accuracy.

## Abstract

The identification of tumor cells is pivotal for understanding tumor heterogeneity and the tumor microenvironment. Recent advances in spatially resolved transcriptomics (SRT) have revolutionized the way that transcriptomic profiles are characterized and have enabled the simultaneous quantification of transcript locations in intact tissue samples. SRT is a promising alternative method to study gene expression patterns in spatial domains. Nevertheless, the precise detection of tumor regions within intact tissue remains a great challenge. A common strategy for identifying tumor cells is via tumor-specific marker gene expression signatures, which are highly dependent on marker accuracy. Another effective approach is through aneuploid copy number alterations, as most types of cancer exhibit copy number abnormalities. Here, we introduce a novel computational method, called TUSCAN (TUmor Segmentation and Classification ANalysis in spatial transcriptomics), which constructs a spatial copy number variation profile to improve the accuracy of tumor region identification. TUSCAN combines gene information from SRT data and hematoxylin-and-eosin-staining image to annotate tumor sections and other benign tissues. We benchmark the performance of TUSCAN and several existing methods through the application to multiple datasets from different SRT platforms. We demonstrate that TUSCAN can effectively delineate tumor regions, with improved accuracy compared to other approaches. Additionally, the output of TUSCAN provides interpretable clonal evolution inferences that may lead to novel insights into disease development and potential druggable targets.

In our research, we addressed a fundamental challenge in understanding cancer: a tumor is not a single entity, but rather a complex and evolving ecosystem of different cells. To understand how cancer grows and becomes clinically aggressive, scientists need an accurate map of where cancerous cells are within a tissue sample. Traditional methods for creating such maps often rely on gene markers that can be unreliable or vary from patient to patient, making it difficult to see the full picture. We developed a new computational tool called TUSCAN that instead looks for a more universal signature of cancer: copy number variations (CNVs), a hallmark of most tumors. By combining this genetic information with standard tissue images, TUSCAN can more accurately distinguish tumor areas from normal tissue. We tested our method across several datasets and found that it consistently outperformed existing tools. Beyond identifying tumor regions, TUSCAN also provides insights into how cancer cells evolve over time, offering a clearer view of tumor diversity. This could help researchers better understand the development of the disease and reveal new opportunities for treatment.