The work, published online July 16, 2026 in Science (DOI:10.1126/science.ady1678), addresses a long-standing question in cancer immunology: patients whose tumors contain TLSs tend to live longer and respond better to immunotherapies, yet the mechanisms that create and sustain these structures were unclear.

Using advanced multiplex imaging and spatial gene analysis, the team examined tumor samples spanning lung, liver, colorectal, kidney, and ovarian cancers. They visualized where dendritic cells were positioned within the tumor microenvironment and mapped which immune cell populations they physically and functionally interacted with.

To test causality, the researchers engineered a new mouse model that closely reproduces TLS features seen in human cancers. With this system, they selectively removed, activated, or genetically modified dendritic cells at defined stages of tumor progression, allowing the investigators to evaluate how TLS formation and persistence depend on these cells over time.

The results show that dendritic cells are essential not only for establishing TLSs, but also for maintaining their integrity as tumors evolve. Rather than performing a transient “alarm” role, dendritic cells remain in place and continuously orchestrate immune activity within the tumor.

Once TLSs are established, these cells coordinate interactions between T cells and B cells, supporting both cellular and humoral arms of anti-tumor immunity. In this model, the tumor becomes a self-contained immunological hub, promoting sustained immune pressure.

The authors emphasize that dendritic cells do more than activate individual immune components: they synchronize multiple pathways of defense, effectively enabling coordinated attacks where they are most needed.

This mechanism also points toward translational opportunities. Therapeutic strategies that increase dendritic cell abundance or boost dendritic cell type 1 function could strengthen TLS formation, potentially improving responses to existing immunotherapies—especially in patients who currently show limited benefit.

The study concludes that TLSs can become durable immune platforms driven by dendritic cell organizers, offering a roadmap for therapies aimed at longer-lasting control of cancer and reducing recurrence.

Subject of Research: Human tissue samples
Article Title: Dendritic cells control tertiary lymphoid structure development and maintenance in cancer
News Publication Date: 16-Jul-2026
Web References: http://dx.doi.org/10.1126/science.ady1678
References: Mattiuz et al., Science (2026) DOI:10.1126/science.ady1678
Image Credits: Mattiuz et al., Science (2026)

Keywords: Cancer immunology, tertiary lymphoid structures, dendritic cells, tumor microenvironment, spatial gene analysis, T cells, B cells, immunotherapy

Traditional approaches to predicting immunotherapy success have largely relied on assessing single molecular markers like PD-L1 expression. While PD-L1 assays have been the clinical standard for identifying candidates likely to respond to checkpoint inhibitors, their predictive accuracy remains limited, hovering around 63%. The new Moffitt study demonstrates that incorporating spatial information—examining how tumor cells and immune cells interact within their microenvironment—can achieve predictive accuracy of up to 87.5%.

The research team employed a cutting-edge combination of multiplex imaging methods, spatial statistical analyses, and machine learning algorithms. These techniques enabled them to profile paired pre-treatment and on-treatment biopsy samples from patients enrolled in a clinical trial testing a combination of the HDAC inhibitor vorinostat with the PD-1 inhibitor pembrolizumab. By moving beyond single-cell marker expression and instead focusing on the architectural patterns of cell neighborhoods, the investigators uncovered tumor-immune ecologies that stratify patients into distinct prognostic groups.

At the core of their analysis was multiscale spatial analysis (MSA), a method that quantifies cellular interactions across multiple scales—from individual cells to larger tissue compartments known as quadrats—within multiplex-stained tissue images. In essence, MSA captures the rich tapestry of cellular positioning and neighborhood relationships that define the immunological “climate” of the tumor. The researchers found that patients whose tumors exhibited a suppressive immune architecture before treatment—characterized by tight spatial clustering of FoxP3-positive regulatory T cells and PD-1-expressing immune cells near tumor cells—were more likely to experience disease progression despite therapy.

Conversely, patients with stable disease demonstrated immune-permissive ecosystems where cytotoxic CD8-positive and helper CD3-positive effector T cells were closely colocalized with tumor cells, indicating an active anti-tumor immune response. These intricate spatial relationships were largely captured prior to the initiation of immunotherapy, suggesting that this ecosystem-based profiling could inform clinical decision-making from the outset.

In an illuminating Q&A, the study’s co-authors elucidated how viewing the tumor as a dynamic and interactive ecosystem marks a paradigm shift in oncology. Treating the microenvironment as a complex community of interacting cells rather than isolated molecular targets allows for a more holistic understanding of treatment response mechanisms. This ecosystem perspective explains why PD-L1 alone is insufficient: it reflects only the presence of a single checkpoint molecule rather than the spatial context of immune cell function and tumor cell susceptibility.

The implications for patient care are profound. Integrating spatial immune profiling into diagnostic workflows could enable clinicians to stratify patients with greater precision—identifying those whose tumors are “immune hot” and likely to respond well to checkpoint blockade, versus those with “immune cold” or suppressive environments who may require combination therapies or enrollment in clinical trials for novel agents. This stratification would help optimize therapy selection, reduce unnecessary side effects, and improve overall survival rates for lung cancer patients.

Although multiplex immunohistochemistry and digital spatial profiling technologies are still emerging in routine clinical practice, their adoption is accelerating. The study authors highlight the potential for developing streamlined computational platforms that can automate ecosystem feature extraction and predictive modeling, paving the way for real-world implementation in pathology laboratories.

Beyond advancing immunotherapy, the methodology showcased in this research opens doors for applying spatial ecology principles to a wide range of cancer types and treatment modalities. Understanding the spatial choreography of tumor-immune interactions could guide personalized approaches in targeted therapies, chemotherapy, and radiation by revealing how local cellular ecosystems modulate therapeutic efficacy.

This landmark study was made possible by support from the National Cancer Institute and the Moffitt Cancer Center’s Centers of Excellence in Evolutionary Therapy and Lung Cancer. It exemplifies the power of interdisciplinary research combining oncology, computational biology, and advanced imaging to unravel complex biological systems.

As the oncology field moves rapidly toward precision medicine, this research heralds a new era where spatial biology will supplement—and in some cases surpass—traditional biomarker diagnostics. Harnessing the spatial context of tumors holds promise to transform not only lung cancer treatment, but the broader landscape of cancer care, bringing us closer to truly personalized and effective interventions.

The innovative framework developed by the Moffitt team offers a compelling vision for the future: one in which digital pathology and artificial intelligence converge to decode the tumor microenvironment’s spatial language, unlocking novel biomarkers and therapeutic targets. As these insights are validated and deployed clinically, patients may benefit from more accurate prognoses and tailored therapies that reflect their tumor’s unique ecosystem.

In conclusion, this study underscores the critical importance of the tumor microenvironment’s spatial structure in shaping treatment outcomes. By shifting the focus from individual molecular markers to multicellular spatial networks, researchers and clinicians can gain a deeper understanding of cancer biology and devise smarter ways to combat this devastating disease. The evolving concept of the tumor as an ecosystem charts a promising path forward for lung cancer immunotherapy and precision oncology at large.

Subject of Research: Human tissue samples

Article Title: Distinct Tumor-Immune Ecologies in Patients with Lung Cancer Predict Progression and Define a Clinical Biomarker of Therapy Response

News Publication Date: March 1, 2026

Web References:

• Cancer Research article DOI

Image Credits: Sandhya Prabhakaran, Chandler Gatenbee, and Alexander Anderson/Moffitt Cancer Center

Keywords: Lung cancer, tumor microenvironment, immunotherapy, spatial biology, multiplex imaging, non-small cell lung cancer, predictive biomarkers, tumor-immune ecologies, machine learning, PD-L1, spatial statistics