Abstract

Abstract Tertiary lymphoid structures (TLSs) play key roles in tumor adaptive immunity. However, the prognostic value and molecular properties of TLSs in esophageal squamous cell carcinoma (ESCC) patients have not been studied. The prognostic values of the presence and maturation status of tumor-associated TLSs were determined in 394 and 256 ESCCs with hematoxylin and eosin (H&E)-stained slides from Sun Yat-sen University Cancer Center (SYSUCC) and Cancer Hospital of Shantou University Medical College (SUMCCH), respectively. A deep-learning TLS classifier was established with H&E-stained slides using inception-resnet-v2 neural network. Digital spatial profiling (DSP) was performed to determine the cellular and molecular characteristics of TLSs in ESCCs. Immature and mature TLSs were observed in 30.2% and 42.9% ESCCs from SYSUCC by pathological examination. The deep-learning TLSs classifier yielded an area under curve value of 0.975 for patients’ mature TLS detection in the external test set, by which 55.1% ESCCs from SUMCCH were identified with mature TLSs. The presence of mature TLSs associated with an earlier pT-stage and improved survival. Mature TLSs were characterized by increased proportions of proliferating B, plasma, and CD4+ Th cells, and increased B memory, CD4+ T activated and effector memory, and Th17 signatures. TLS-mature ESCCs were with increased intratumoral CD8+ T infiltration. Mature TLSs provide a privileged microniche for B and T cell differentiation and maturation, thereby generating an effective anti-tumor humoral and cellular immunity to improve ESCC prognosis.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call