In some patients, tumors contain tertiary lymphoid structures, or TLSs: organized clusters of immune cells that resemble lymph-node-like sites of local immune coordination.
The work helps explain how local anti-tumor immune responses are maintained within tumor tissue and points to potential strategies for improving immunotherapy.
Many tumors contain immune cells, but not all immune infiltrates are functionally productive.
These methods allowed them to map where dendritic cells were located, which immune cells surrounded them, and how those local neighborhoods related to TLS organization.
It is not only about having T cells, B cells, antibodies, or dendritic cells inside the tumor.
Not all immune activity inside a tumor is disorganized. In some patients, tumors contain tertiary lymphoid structures, or TLSs: organized clusters of immune cells that resemble lymph-node-like sites of local immune coordination. Their presence has been associated with improved survival and better responses to immunotherapy in several cancers, making them an increasingly important feature of the tumor microenvironment.
A new study published in Science identifies a specialized immune cell population that helps build and sustain these structures. Researchers at the Icahn School of Medicine at Mount Sinai and collaborating institutions found that type 1 conventional dendritic cells, or cDC1s, act as central organizers of TLSs in cancer. The work helps explain how local anti-tumor immune responses are maintained within tumor tissue and points to potential strategies for improving immunotherapy.
Why TLSs matter in cancer immunity
TLSs are not normal anatomical lymph nodes. They arise in chronically inflamed tissues, including tumors, and can contain organized T-cell zones, B-cell follicles, germinal center-like regions, plasma cells, and specialized stromal networks. In cancer, their presence is often interpreted as evidence that the immune system is not only infiltrating the tumor but organizing a sustained local response.
Many tumors contain immune cells, but not all immune infiltrates are functionally productive. TLSs may provide a site where antigen presentation, T-cell activation, B-cell maturation, and antibody responses occur close to malignant cells. This could help explain why TLS-positive tumors are often associated with more favorable outcomes and greater sensitivity to immune checkpoint blockade.
Until now, however, it has been less clear what drives TLS formation and, just as importantly, what keeps these structures functional once they are established.
cDC1s as local immune architects
Dendritic cells are best known for antigen presentation: they capture tumor antigens and prime T-cell responses. This study expands that role. The researchers found that cDC1s do not only initiate anti-tumor immunity; they help organize the physical and functional immune architecture inside tumors.
“Our goal was to understand how these immune structures develop and persist inside tumors,” said lead author Raphael Mattiuz, PhD. “We found that a distinct subset of dendritic cells acts as the organizer, bringing together different immune cells and keeping the local anti-cancer response active.”
The team analyzed tumor samples from patients with lung, liver, colorectal, kidney, and ovarian cancers using multiplex imaging and spatial gene-expression approaches. These methods allowed them to map where dendritic cells were located, which immune cells surrounded them, and how those local neighborhoods related to TLS organization.
Across tumor types, mature dendritic cells accumulated within TLSs. The mechanistic work then focused on a mouse model of non-small cell lung cancer designed to reproduce mature TLS formation seen in human tumors. In this model, cDC1s were required both during the establishment of TLSs and later for their maintenance.
More than T-cell priming
The study suggests a two-stage role for cDC1s. Early in tumor development, TLS formation depended on IFNγ-driven maturation of cDC1s, migration to tumor-draining lymph nodes, and recruitment of primed T cells back into the tumor. As tumors progressed, however, the biology changed. TLSs could persist even when egress of T cells from tumor-draining lymph nodes was impaired, while cDC1s became retained within intratumoral stromal hubs enriched in CCR7 ligands.
This is clinically interesting because it positions cDC1s as tissue-resident coordinators of ongoing immunity, not merely transient antigen couriers. The researchers found that timed depletion of cDC1s after TLSs had formed disrupted TLS maintenance. Blocking their localization to stromal hubs had a similar effect.
The function of these cells also depended on antigen presentation. Genetic ablation of both MHC class I and II on cDC1s impaired TLS maintenance, T follicular helper cell preservation, germinal center formation, tumor-specific IgG production, and differentiation of progenitor exhausted CD8-positive T cells.
“We were surprised to see that these rare cells become permanent organizers within the tumor itself,” Mattiuz said. “They don’t just activate cancer-killing T cells. They also help coordinate antibody responses, allowing multiple parts of the immune system to work together where they’re needed most.”
Implications for immunotherapy
Checkpoint inhibitors have changed cancer care, but durable responses remain limited to a subset of patients. One reason is that reinvigorating T cells may not be enough if the tumor lacks the local immune organization needed to sustain productive responses. TLS biology offers a complementary framework: the issue may not only be whether immune cells are present, but whether they are organized into functional niches.
This study suggests that cDC1-directed therapies could help strengthen those niches. Potential strategies might include increasing cDC1 abundance, enhancing cDC1 maturation, improving their recruitment or retention in tumors, or combining cDC1 activation with checkpoint blockade, vaccines, radiotherapy, or other immune-modulating approaches.
The findings also add nuance to biomarker development. TLS presence is already being studied as a prognostic and predictive feature, but TLS quality may matter as much as TLS quantity. A tumor with cDC1-rich, antigen-presenting, germinal center-supporting TLSs may behave differently from a tumor with less mature or poorly maintained immune aggregates.
For medical oncology, this could eventually refine how TLSs are interpreted in pathology and translational studies. Rather than treating TLSs as a binary histological feature, future assays may need to assess their cellular composition, dendritic cell state, B-cell organization, and proximity to effector T-cell populations.
Still early, but mechanistically important
The work is not yet a clinical intervention. It does not show that activating cDC1s in patients will reliably generate TLSs or improve immunotherapy outcomes. Tumor type, antigenicity, stromal architecture, prior therapy, and immunosuppressive pathways will likely influence whether this biology can be therapeutically exploited.
Nevertheless, the study provides a clearer mechanistic target in a field that has often treated TLSs as useful but poorly controlled biomarkers. If cDC1s are required to form and maintain functional TLSs, then therapies aimed at dendritic cell biology may become a way to convert poorly organized tumor immune infiltrates into more coordinated anti-tumor responses.
The broader message is that effective cancer immunity is spatial as well as cellular. It is not only about having T cells, B cells, antibodies, or dendritic cells inside the tumor. It is about arranging them in the right place, in the right state, and for long enough to sustain pressure on malignant cells.
By identifying cDC1s as organizers of tumor-associated TLSs, the study offers a more concrete path toward therapies that help the immune system build its own infrastructure inside cancer.
