Kyung Serk Cho, Yunhe Liu, Guangsheng Pei, Jianfeng Chen, Yibo Dai, Yang Liu, Tieling Zhou, Antoine Bougoüin, Alejandra Serrano, Khalida Wani, Akshaya Jadhav, Jimin Min, Sharia Hernandez, Wei Lu, Daiwei Zhang, Jiahui Jiang, Diana Shamsutdinova, Enyu Dai, Fuduan Peng, Ansam Sinjab, Paola A Guerrero, Idania Carolina Lubo Julio, Kai Yu, Helen Clark, Dipen Maru, Mingyao Li, Andrew Futreal, Sanghoon Lee, L M Solis Soto, Lulu Shang, P Msaouel, Jaffer A. Ajani, Hannah Beird, Amir A. Jazaeri, Alexander J. Lazar, Catherine Sautes-Fridman, Wolf H. Fridman, Anirban Maitra, Humam Kadara, J Gao, Padmanee Sharma, L L W Wang
Tertiary lymphoid structures (TLSs) are critical regulators of antitumor immunity, yet their spatial organization, maturation, and clinical relevance remain incompletely defined across cancers. We analyzed spatial transcriptomics spanning 12 cancer types to construct a pan-cancer TLS atlas and characterized TLS spatial architecture and maturation states. TLS maturation was accompanied by coordinated remodeling of distinct niche cell populations and distance-dependent gradients in tumor programs, orthogonally supported by ultrahigh-plex single-cell spatial profiling. To enable scalable TLS profiling, we trained an artificial intelligence framework that predicts TLS maturation states directly from hematoxylin and eosin-stained images and evaluated it across TCGA and independent therapy cohorts. We further derived a maturation-aware composite score capturing intratumoral TLS state composition, which robustly stratifies patients across cancer and treatment contexts, outperforming conventional TLS metrics.