Multi-resolution deep learning characterizes tertiary lymphoid structures and their prognostic relevance in solid tumors.

BACKGROUND: Tertiary lymphoid structures (TLSs) are dense accumulations of lymphocytes in inflamed peripheral tissues, including cancer, and are associated with improved survival and response to immunotherapy in various solid tumors. Histological TLS quantification has been proposed as a novel predictive and prognostic biomarker, but lack of standardized methods of TLS characterization hampers assessment of TLS densities across different patients, diseases, and clinical centers. METHODS: We introduce an approach based on HookNet-TLS, a multi-resolution deep learning model, for automated and unbiased TLS quantification and identification of germinal centers in routine hematoxylin and eosin stained digital pathology slides. We developed HookNet-TLS using n = 1019 manually annotated TCGA slides from clear cell renal cell carcinoma, muscle-invasive bladder cancer, and lung squamous cell carcinoma. RESULTS: Here we show that HookNet-TLS automates TLS quantification across multiple cancer types achieving human-level performance and demonstrates prognostic associations similar to visual assessment. CONCLUSIONS: HookNet-TLS has the potential to be used as a tool for objective quantification of TLS in routine H&E digital pathology slides. We make HookNet-TLS publicly available to promote its use in research.

Tertiary lymphoid structures (TLS) are dense accumulations of immune cells within a cancer. They have been associated with patient survival and treatment effectiveness. Quantification of TLS in cancer microscopy images may therefore aid clinical decision-making. However, no consensus for defining TLS in such images exists leading to inconsistent and variable findings across different labs and studies. We developed a computational tool for automated and objective TLS quantification in cancer images. The tool, called HookNet-TLS, integrates information from multiple image resolutions, which resembles the process of how a pathologist would identify these structures using a microscope. HookNet-TLS detected TLS similarly to trained researchers in three different tumor types. We provided access to HookNet-TLS to facilitate its development and use for TLS assessment in clinical decision-making and research into the role of TLS in cancer.

Human CtIP: A 'double agent' in DNA repair and tumorigenesis.

Human CtIP was originally identified as an interactor of the retinoblastoma protein and BRCA1, two bona fide tumour suppressors frequently mutated in cancer. CtIP is renowned for its role in the resection of DNA double-strand breaks (DSBs) during homologous recombination, a largely error-free DNA repair pathway crucial in maintaining genome integrity. However, CtIP-dependent DNA end resection is equally accountable for alternative end-joining, a mutagenic DSB repair mechanism implicated in oncogenic chromosomal translocations. In addition, CtIP contributes to transcriptional regulation of G1/S transition, DNA damage checkpoint signalling, and replication fork protection pathways. In this review, we present a perspective on the current state of knowledge regarding the tumour-suppressive and oncogenic properties of CtIP and provide an overview of their relevance for cancer development, progression, and therapy.

Molecular, Immunological, and Clinical Features Associated With Lymphoid Neogenesis in Muscle Invasive Bladder Cancer.

Lymphoid neogenesis gives rise to tertiary lymphoid structures (TLS) in the periphery of multiple cancer types including muscle invasive bladder cancer (MIBC) where it has positive prognostic and predictive associations. Here, we explored molecular, clinical, and histological data of The Cancer Genome Atlas, as well as the IMvigor210 dataset to study factors associated with TLS development and function in the tumor microenvironment (TME) of MIBC. We also analyzed tumor immune composition including TLS in an independent, retrospective MIBC cohort. We found that the combination of TLS density and tumor mutational burden provides a novel independent prognostic biomarker in MIBC. Gene expression profiles obtained from intratumoral regions that rarely contain TLS in MIBC showed poor correlation with the prognostic TLS density measured in tumor periphery. Tumors with high TLS density showed increased gene signatures as well as infiltration of activated lymphocytes. Intratumoral B-cell and CD8+ T-cell co-infiltration was frequent in TLS-high samples, and such regions harbored the highest proportion of PD-1+TCF1+ progenitor-like T cells, naïve T cells, and activated B cells when compared to regions predominantly infiltrated by either B cells or CD8+ T cells alone. We found four TLS maturation subtypes; however, differences in TLS composition appeared to be dictated by the TME and not by the TLS maturation status. Finally, we identified one downregulated and three upregulated non-immune cell-related genes in TME with high TLS density, which may represent candidates for tumor-intrinsic regulation of lymphoid neogenesis. Our study provides novel insights into TLS-associated gene expression and immune contexture of MIBC and indicates towards the relevance of B-cell and CD8+ T-cell interactions in anti-tumor immunity within and outside TLS.