Tertiary lymphoid structures are organized aggregates of immune cells that form outside the conventional lymph nodes and spleen. In many solid tumors, they are viewed as encouraging signs: evidence that the immune system has mounted a localized, sustained response against the malignancy. These structures typically contain T cells and B cells arranged in patterns that resemble miniature lymphoid organs, sometimes even developing the germinal centers where antibodies are refined. In cancers such as melanoma, lung cancer, and breast cancer, the presence and density of TLS have repeatedly been linked to better responses to immunotherapy and improved survival. The brain, however, has long been considered an immunologically quiet environment, protected by the blood-brain barrier and populated by its own specialized immune cells, so the role of TLS in brain tumors has remained far less clear.

That gap is precisely what a team of neurosurgeons and researchers affiliated with Kunming Medical University and collaborating institutions in China set out to address. In a retrospective study published in the journal Cancer Immunology, Immunotherapy, the investigators analyzed tissue samples from 62 patients with isocitrate dehydrogenase, or IDH, wildtype glioblastoma who were treated at their institution between 2016 and 2023. IDH wildtype status defines the most aggressive and common form of glioblastoma in adults, and it is the subtype for which new prognostic tools are most urgently needed. The study was conducted in accordance with the Declaration of Helsinki, with ethics approval from the First Affiliated Hospital of Kunming Medical University and written informed consent obtained from all participants or their legal representatives.

The methodological heart of the study lies in how the researchers detected and classified these immune structures. Using standard hematoxylin and eosin staining alongside immunohistochemical staining for CD3, a marker of T cells, and CD20, a marker of B cells, the team identified TLS within resected tumor specimens. Rather than simply recording whether TLS were present or absent, the investigators went a step further and built a grading system that captured two dimensions: the quantity of the structures and their spatial distribution within the tumor region. This distinction matters because a TLS sitting in the peritumoral zone, the tissue surrounding the tumor, may reflect a different immunological situation than one embedded deep within the tumor mass itself, where it would be in direct contact with malignant cells.

The results were striking. TLS were present in 58.1 percent of the cohort, meaning that more than half of these patients carried organized immune structures in or around their tumors. When the researchers correlated TLS status with overall survival using Kaplan-Meier curves, they found that TLS presence, TLS grading, and the extent of surgical resection all showed statistically significant associations with survival outcomes, each reaching the threshold of P less than 0.05. The grading system proved particularly informative. Patients with higher TLS grades, and especially those whose structures were located within the tumor itself, corresponding to the highest category, Score 3, experienced significantly prolonged overall survival compared with patients lacking these structures or harboring only peritumoral ones.

To ensure that this association was not simply a statistical artifact driven by other clinical variables, the team turned to multivariate Cox regression analysis, a standard statistical technique that evaluates the independent contribution of each factor while controlling for the others. The analysis confirmed that TLS grading stood as an independent favorable prognostic factor for overall survival in this glioblastoma cohort. In practical terms, this means that even after accounting for variables such as the extent of resection, the grade of the TLS carried its own predictive weight. The extent of resection, which distinguishes total, subtotal, and partial removal of the tumor, also remained significantly correlated with survival, consistent with decades of neurosurgical evidence, but the TLS signal held its ground alongside it.

The implications of these findings extend beyond prognosis into the realm of treatment selection. Glioblastoma has been a notoriously poor responder to immune checkpoint inhibitors, the class of drugs that has revolutionized the treatment of many other cancers. One leading explanation is that the brain tumor microenvironment lacks the pre-existing immune infrastructure that such therapies require to work. TLS could represent exactly that infrastructure: a localized factory for generating and deploying tumor-specific immune cells. If patients whose tumors contain high-grade, intratumoral TLS are the ones with an already-primed immune response, they may be the subset most likely to benefit from intensified immunotherapeutic interventions, while patients without TLS might need strategies that first induce lymphoid neogenesis before checkpoint blockade could have any chance of success.

The authors are appropriately measured in their claims. They emphasize that the proposed TLS grading system provides a potential prognostic stratification tool for glioblastoma that warrants external validation, a crucial caveat in a field where promising single-institution findings sometimes fail to replicate in larger, independent cohorts. With 62 patients, the study is a meaningful proof of concept but not a definitive clinical standard. Standardizing how TLS are identified and scored across pathology laboratories, and testing the grading system in independent datasets from other centers, will be essential steps before it could inform treatment decisions at the bedside. The retrospective design also means that the findings demonstrate association rather than causation, and it remains possible that TLS are a marker of some other underlying biological difference rather than an active driver of better outcomes.

Even with those caveats, the study adds an important piece to the evolving picture of glioblastoma immunology. The tumor microenvironment, once viewed as a hostile wasteland for immune activity, is increasingly understood as a complex ecosystem where organized immune structures can and do form. The fact that the location of these structures, not just their number, carries prognostic weight suggests that the spatial architecture of the anti-tumor immune response matters, a principle that resonates with broader trends in cancer immunology where spatial biology is reshaping how tumors are classified. For a disease in which median survival has barely moved in decades, any new, independently validated prognostic factor is significant, and one that could also guide immunotherapy selection would be doubly valuable.

The research, published as an open-access article, was supported by funding from Kunming Medical University, the Health Commission of Yunnan Province, and related regional research programs. The authors reported no relevant financial or non-financial conflicts of interest. As the field moves forward, the challenge will be to translate this grading system from the research laboratory into routine pathology workflows, where hematoxylin and eosin slides and CD3 and CD20 stains are already standard tools. If external validation confirms what this cohort suggests, neuro-oncologists may one day look at a glioblastoma specimen not only for the features that define its malignancy, but for the quiet architecture of the immune response unfolding within it, and use that architecture to decide which patients need the most aggressive immunological reinforcements available.

Subject of Research: Tertiary lymphoid structures as prognostic biomarkers in IDH-wildtype glioblastoma

Article Title: Tertiary lymphoid structures (TLS) classification: a new perspective on glioblastoma patient’s prognosis

Article References: Meng, M., Li, Y., Dai, X., Yang, G., Li, Z., & Tang, Z. (2026). Tertiary lymphoid structures (TLS) classification: a new perspective on glioblastoma patient’s prognosis. Cancer Immunology, Immunotherapy. https://doi.org/10.1007/s00262-026-04537-2

Image Credits: AI Generated

DOI: 10.1007/s00262-026-04537-2

Keywords: glioblastoma, tertiary lymphoid structures, tumor microenvironment, prognostic markers, immunotherapy, IDH-wildtype, immunohistochemistry, overall survival, neuro-oncology, Cox regression, CD3, CD20