At The University of Texas MD Anderson Cancer Center in Houston, researchers have just mapped the hidden immune structures that may hold the key to predicting which cancer patients will respond to treatment—and which won't. Led by Linghua Wang, M.D., Ph.D., a team of scientists has created the first comprehensive spatial atlas of tertiary lymphoid structures, or TLSs, revealing that these immune "hubs" are far more complex and informative than anyone previously realized.
TLSs are organized clusters of immune cells—B cells, T cells, antigen-presenting cells, and supporting tissue—that form spontaneously within tumors, working together to coordinate the body's fight against cancer. For years, doctors knew that their presence was a good sign. Patients with mature TLSs often responded better to immunotherapy and had better outcomes. But that's where understanding stopped: either a tumor had them or it didn't. The new research, published in Science, reveals this oversimplification was hiding crucial biological details.
The study examined 340 tumor samples across 12 different cancer types, building a pan-cancer atlas that allowed researchers to examine TLS maturation, cellular composition, and spatial organization in unprecedented detail. Using scalable artificial intelligence frameworks, Wang's team developed tools to detect and classify TLSs not just from specialized spatial omics data but also from routine pathology slides—making the technology immediately applicable in clinical settings. The researchers discovered that TLSs vary substantially between tissues, and as they mature, they undergo coordinated changes in their immune, stromal, and vascular components. Critically, how close these structures sit to tumor cells creates spatial gradients of tumor signaling—information that seems to reflect important features of the tumor's immune microenvironment.
What makes this discovery so significant is the shift in perspective it represents. Instead of asking simply whether a TLS is present, clinicians can now ask far more precise questions: How mature is it? Where exactly is it located relative to the tumor? What cells comprise it? The research team created a composite scoring system that can stratify patients by prognosis and treatment response across different cancer types and treatment contexts—a substantial leap forward in personalized medicine.
"Prior to this study, most of the focus on TLSs as biomarkers was simply on whether or not they were present, and—in some cases—whether they were mature," Wang explained. "Here, we show that we can go much deeper. TLSs in tumor tissues are much more complex than that. Their maturation state, spatial location and composition within tumors can tell us critical information about the tumor immune microenvironment, treatment response and clinical outcomes."
The practical implications ripple outward quickly. Oncologists will soon be able to use AI tools to scan routine pathology slides and extract information that was previously invisible—or at least invisible without expensive specialized testing. This demystifies the tumor microenvironment in ways that could reshape treatment decisions, helping doctors predict not just who will benefit from immunotherapy, but potentially guiding them toward entirely new therapeutic approaches. For patients, it means moving closer to a future where cancer treatment is matched to the specific immunological landscape of their own tumor, rather than applied with a broad brush.