New spatial map reveals treatment vulnerabilities in bladder cancer

Researchers at The University of Texas MD Anderson Cancer Center have developed a spatial map of muscle-invasive bladder cancer, revealing how tumor cell states, immune environments and therapeutic vulnerabilities are organized within tumors. The study, published in Cancer Discovery, provides a new framework for understanding why patients with bladder cancer may respond differently to treatment.

The research was led by Linghua Wang, M.D., Ph.D., professor of Genomic Medicine, executive director and head of the Center for Cellular Language Intelligence, associate member of the James P. Allison Institute™, and focus area co-lead with the Institute for Data Science in Oncology; together with Jianjun Gao, M.D., Ph.D., professor of Genitourinary Medical Oncology, and co-first authors Kai Yu, Ph.D., postdoctoral fellow in the Wang laboratory, and Jianfeng Chen, M.D., Ph.D., instructor of Genitourinary Medical Oncology.

Traditional molecular subtyping often classifies bladder cancers as either luminal or basal, but our spatial analyses show that this binary view is incomplete. Within a single patient's tumor, luminal and basal-like programs can coexist in highly organized spatial patterns, and those patterns are closely tied to immune activity, lineage-specific treatment vulnerabilities, and how different tumor regions may respond to treatment."

Linghua Wang, M.D., Ph.D., professor of Genomic Medicine, executive director and head of the Center for Cellular Language Intelligence, associate member of the James P. Allison Institute

What does the study reveal about bladder cancer biology?

Muscle-invasive bladder cancer is a clinically heterogeneous disease, and current biomarkers do not fully explain why patients respond differently to treatment. To address this, the researchers integrated spatial transcriptomics from 22 pretreatment tumors with matched whole-exome sequencing and bulk RNA sequencing. They also used single-cell RNA sequencing and single-cell spatial transcriptomics from additional tumors to validate the findings at higher resolution.

The spatial map revealed that cancer cells can gradually change from one cell type (luminal) to another (basal) along a continuous differentiation axis within individual tumors. Luminal-like tumor cells were more often located in tumor cores and were enriched for FGFR3 and NECTIN4, two clinically relevant therapeutic markers.

In contrast, basal-like tumor cells were found more often near invasive tumor margins, where they showed higher EGFR signaling, epithelial-mesenchymal transition programs, chromosomal instability and immune infiltration, which are all signs of tumor cell aggressiveness.

These findings suggest that bladder cancer is not defined solely by a dominant molecular subtype across the whole tumor but, instead, contains multiple tumor cell states that occupy different physical regions and interact with different immune environments.

"An important message from this work is that effective treatment may need to account for both luminal and basal components within the same tumor, as well as their spatial organization," Gao said. "Rather than treating bladder cancer as a single uniform disease state, spatial information may help guide rational combinations or sequencing of therapies that target distinct tumor regions and cell states."

How can this guide treatment strategies for patients with bladder cancer?

The study identified lineage-specific treatment vulnerabilities. For example, luminal tumor regions showed high NECTIN4 expression, supporting NECTIN4 as a relevant therapeutic target in tumors or tumor regions enriched for luminal features. Basal-like regions, by contrast, were associated with immune-rich microenvironments and greater chemotherapy sensitivity.

Further testing in preclinical models supported this, as NECTIN4 overexpression increased sensitivity to enfortumab vedotin - a NECTIN4-targeted antibody-drug conjugate - in vitro and shifted tumor cells toward a more luminal-like state. Chemotherapy and enfortumab vedotin appeared to act differently, suggesting spatial information may help inform how therapies are combined or sequenced to address distinct tumor compartments.

The researchers also found that FGFR3 and EGFR marked opposite ends of the luminal-basal spectrum, which can help identify tumor aggressiveness based on where tumor cells fall along that spectrum. FGFR3 was linked to luminal states, while EGFR was enriched in basal-like and more plastic tumor states. These patterns were validated across more than 3,000 tumors from independent bladder cancer cohorts.

The study supports the development of spatially informed biomarkers that could help researchers and clinicians better understand which therapies are most likely to affect specific tumor regions and tumor cell states. For example, future approaches could evaluate NECTIN4-directed therapies for immune-cold luminal tumor cores, while using chemotherapy or immunotherapy-based treatments may be a better strategy to target inflamed basal-like margins.

Future studies will need to validate these findings in larger, prospective clinical cohorts, including post-treatment samples, to determine how therapy reshapes tumor architecture over time.