S. Schallenberg

An explainable transformer model learning from entire treatment timelines for pan-cancer risk profiling across healthcare systems

Cancer outcomes vary widely between individual patients, each accumulating an irregular record of treatments, diagnoses, measurements, and complications. Current prognostic models reduce this complexity into a single snapshot, focus on narrow clinical settings, and rarely generalize across hospitals. Here we introduce Chronicle, an explainable transformer that learns from entire patient trajectories, predicts diverse clinical outcomes throughout the disease course while capturing both short- and long-term temporal dependencies. Trained on 53.7 million longitudinal data points from 51,711 patients spanning 67 cancer types, Chronicle operates natively on irregular data without imputation and jointly predicts eight endpoints within a flexible framework adaptable to additional outcomes. Chronicle outperformed cross-sectional models for overall survival prediction (C-index 0.84 vs 0.76-0.79), stratified patients more accurately than established prognostic systems, including TNM stage, and predicted seven adverse event and transfusion endpoints (AUC 0.80-0.92). Applied without retraining to 69,341 patients in Germany, Switzerland, and the United States, Chronicle generalized across healthcare systems and improved further with local fine-tuning. Integrated explainability traced each risk update to patient-specific clinical factors, revealing distinct temporal persistence of prognostic information, with relevance half-lives ranging from weeks for therapies to nearly one year for baseline characteristics. These findings demonstrate that learning from hospital-wide patient trajectories enables interpretable and continuously updated predictions, providing a scalable framework to support individualized treatment decisions.

CD19 CAR T cell therapy for treatment-refractory seropositive rheumatoid arthritis: a phase 1 trial.

Immune evasion in locally advanced mismatch repair-deficient microsatellite instability-high colorectal cancer: Reduced T-cell infiltration and upregulation of epithelial IDO1 expression

ABSTRACT Early-stage colorectal cancer (CRC) with confirmed microsatellite instability generally has a favorable prognosis associated with pronounced immune infiltration. However, some patients still develop metastases. The underlying mechanisms, particularly those related to the tumor immune microenvironment, remain incompletely understood. The study included tissue samples from 217 patients with dMMR/MSI-H CRC, comprising 89 stage III/IV and 128 stage I/II cases. Tissue microarrays and immunohistochemical analyzes were performed for all cases. We evaluated immune markers identifying T cells, B cells, dendritic cells, natural killer cells, macrophages, immunosuppressive markers, and immune checkpoint targets in epithelial and stromal compartments, and additionally performed a cohort-derived immune infiltration score (IIS). The survival analysis assessed the prognostic impact of immune markers stratified by tumor stage. Stage I/II dMMR/MSI-H CRCs showed significantly higher CD3⁺ T-cell and natural killer cell levels, higher IIS metrics across all regions, and higher CD4⁺ T helper cell levels in the stroma. Stage III/IV cases exhibited increased epithelial expression of indoleamine 2,3-dioxygenase 1. Given the limited number of stage IV patients, an additional stage III vs. stage I/II comparison was performed, revealing that these immune differences were already evident at the level of nodal progression. In addition, CD3⁺, CD4⁺, and CD8⁺ T cells and the IIS showed varying prognostic associations across tumor stages. These findings suggest that the overall immunogenicity and prognostic relevance of immune markers in dMMR/MSI-H CRC depend on tumor stage. Immune profiling could be used for early-stage patient stratification and also suggests the potential benefit of early immunotherapeutic intervention.