Community-acquired respiratory viruses (CARV), such as influenza-, parainfluenza- or respiratory syncytial virus, pose a significant threat to immunocompromised patients with cancer. Following the COVID-19 pandemic, SARS-CoV-2 has now joined the ranks of endemic respiratory viruses and continues to be a cause of significant morbidity and mortality in patients with cancer. Strategies to protect this vulnerable patient population both by prevention of infection and by early therapeutic intervention in case of infectious disease are therefore of utmost importance. This guideline provides updated evidence-based recommendations on diagnosis, prophylaxis and treatment of CARV infections including COVID-19 in patients with solid tumors or hematologic malignancies to support clinicians in offering optimal care. The guideline is based on a systematic review of currently available data and was developed until the beginning of 2025 by an expert panel of the Infectious Diseases Working Party (AGIHO) of the German Society for Hematology and Medical Oncology (DGHO).
Treatment-free remission (TFR) after discontinuation of ABL tyrosine kinase inhibitors (TKIs) is an important therapeutic goal in chronic myeloid leukemia (CML). Interferon-α (IFN) has been suggested to promote durable TFR. The phase 3 ENDURE trial (NCT03117816; EUDRA-CT 2016-001030-94) prospectively tested this hypothesis in patients with stable deep molecular remission after TKI therapy. A total of 203 patients were randomised 1:1 to receive ropeginterferon alfa-2b (ropeg-IFN; 100 µg subcutaneously every two weeks for 15 months, n = 95) or observation alone (n = 108) after TKI discontinuation. The primary endpoint was molecular relapse-free survival (MRFS), defined as time to loss of major molecular response (MMR) or death. At a median follow-up of 36 months, 25-month MRFS was 56% (95% confidence interval (CI), 45–66) with ropeg-IFN and 59% (95% CI, 49–68) with observation (hazard ratio (HR), 1.02; 95% CI, 0.68–1.55; P = 0.91). Among 83 patients with molecular data after TKI restart, 79 (95%) regained at least MMR, 78 within 12 months (median 3 months, interquartile range: 2-4 months). Ropeg-IFN was well tolerated (median administered dose of 92 µg, range 3–104), and no new safety signals were observed. Ropeg-IFN maintenance did not improve the probability of sustained TFR after TKI discontinuation.
Glioblastoma, IDH-wildtype (GBM-IDHwt) is the most common malignant brain tumor. Histomorphology is a crucial component of the integrated diagnosis of GBM-IDHwt. Artificial intelligence (AI) methods have shown promise to extract additional prognostic information from histological whole-slide images (WSI) of hematoxylin and eosin-stained glioblastoma tissue. Here, we present an explainable AI-based method to support systematic interpretation of histomorphological features associated with survival. It combines an explainable multiple instance learning (MIL) architecture with a sparse autoencoder (SAE) to relate human-interpretable visual patterns of tissue to survival. The MIL architecture directly identifies prognosis-relevant image tiles and the SAE maps these tiles post-hoc to visual patterns. The MIL method was trained and evaluated using a new real-world dataset that comprised 720 GBM-IDHwt cases from three hospitals and four cancer registries in Germany. The SAE was trained using 1878 WSIs of glioblastoma from five independent public data collections. Despite the many factors influencing survival time, our method showed some ability to discriminate between patients living less than 180 days or more than 360 days solely based on histomorphology (AUC: 0.67; 95
Cognitive diagnosis post-stroke is typically restricted to static, in-clinic assessments. We developed and evaluated the Oldenburg Test Battery for Remote Digital Phenotyping of Post-Stroke Cognition (ORPheoS) , which is an item bank and test battery to be used for sampling brief test versions for smartphone-based repeated testing in ecologically valid longitudinal cognitive phenotyping studies to inform stroke rehabilitation. A total of 280 controls and 232 stroke patients underwent ORPheoS during one baseline measurement at ~five days post-stroke, comprising five tasks targeting reasoning, recognition and associative memory, and external and internal interference control. A planned-missingness design with linkage items enabled joint calibration across partially overlapping item subsets. Item properties and domain specific ability scores were estimated using Rasch-family item response theory-based latent trait uni- and multidimensional models, as well as mixed-effects modelling. A bifactor S-1 model was used to examine the taxonomic order of the scaled abilities measured by ORPheoS. Latent trait models demonstrated acceptable to excellent item properties. Domain specific split-half reliability estimates ranged from 0.52–0.88. Known-groups validity was domain-specific. Reasoning showed a large group difference ( g = 0.45, p < 0.001) and the strongest associations with global cognition. Associative memory and mental speed also differentiated groups ( g = 0.25, p = 0.011; and g = 0.91, p < 0.001). Familiarity-based recognition showed no group difference ( g = − 0.03, p = 0.734), while stroke participants outperformed controls on Recollection-based discrimination ( g = − 0.48, p < 0.001). The bifactor S-1 model showed acceptable fit (scaled χ² (47) = 161.12, p < 0.001; robust CFI = 0.941; robust RMSEA = 0.078), confirming interference control and memory as domain specific factors beyond general cognitive ability indicated by reasoning. ORPheoS provides psychometrically calibrated, domain-specific tasks and items within a scalable digital framework, establishing a measurement foundation for longitudinal and personalised monitoring of cognitive functions after stroke. The current dataset and estimated item properties allow for adaptive sampling and item generation for repeated ecological assessments of cognition after stroke, with potential of supporting tailor-made rehabilitation planning and monitoring. This work was supported by the Research Training Group (RTG) 2783, funded by the German Research Foundation (DFG) – Project ID 456732630.
Kurzfassung In der gastroenterologischen Endoskopie wurden erhebliche Fortschritte erreicht, die sie vielfach zur Therapie der ersten Wahl hat werden lassen. Mehrere Leitlinien und Publikationen der DGVS machen Vorgaben zur Qualität der Endoskopie, Mindestzahlen zum Kompetenzerwerb und Kompetenzerhalt sind bislang aber nur partiell adressiert. Daher haben wir Vorschläge für Mindestzahlen für fortgeschrittene endoskopische Prozeduren und zur Implantation eines TIPS erarbeitet. Wenn möglich folgen sie vorliegender Evidenz, ansonsten der Erfahrung der Autorinnen und Autoren. Vorausgesetzt werden Kompetenzen in der diagnostischen Gastroskopie und Koloskopie inklusive einfacher Prozeduren sowie eine strukturierte Weiterbildung durch erfahrene Trainer. Auch ex-vivo Modelle können eingerechnet werden. Wir schlagen zum Kompetenzerwerb 70 Endoskopische Mukosaresektionen von Polypen >2cm (schwache Evidenz), 60 Endoskopische Submukosadissektionen (moderate Evidenz), 15 Vollwandresektionen (keine Evidenz), 200 ERCPs (gute Evidenz), 225 Endosonographien (schwache Evidenz), 50 endoskopische Blutstillungen (schwache Evidenz) und 25 POEM (für ESD erfahrene Endoskopiker) vor. Für TIPS-Anlagen schlagen wir 20 pro Jahr pro Zentrum vor (gute Evidenz). Keine Basis für eine gesonderte Empfehlung sehen wir beim Stenting und bei Verschlusstechniken. Eine breite Implementierung dieser Vorgaben kann zu einer Verbesserung der Versorgungsqualität führen und ist Voraussetzung zur Bildung endoskopischer Zentren.
