How to predict relapse in leukemia using time series data: A comparative in silico study.

Risk stratification and treatment decisions for leukemia patients are regularly based on clinical markers determined at diagnosis, while measurements on system dynamics are often neglected. However, there is increasing evidence that linking quantitative time-course information to disease outcomes can improve the predictions for patient-specific treatment responses. We designed a synthetic experiment simulating response kinetics of 5,000 patients to compare different computational methods with respect to their ability to accurately predict relapse for chronic and acute myeloid leukemia treatment. Technically, we used clinical reference data to first fit a model and then generate de novo model simulations of individual patients' time courses for which we can systematically tune data quality (i.e. measurement error) and quantity (i.e. number of measurements). Based hereon, we compared the prediction accuracy of three different computational methods, namely mechanistic models, generalized linear models, and deep neural networks that have been fitted to the reference data. Reaching prediction accuracies between 60 and close to 100%, our results indicate that data quality has a higher impact on prediction accuracy than the specific choice of the particular method. We further show that adapted treatment and measurement schemes can considerably improve the prediction accuracy by 10 to 20%. Our proof-of-principle study highlights how computational methods and optimized data acquisition strategies can improve risk assessment and treatment of leukemia patients.

Contribution of Adventitia-Derived Stem and Progenitor Cells to New Vessel Formation in Tumors.

Blocking tumor vascularization has not yet come to fruition to the extent it was hoped for, as angiogenesis inhibitors have shown only partial success in the clinic. We hypothesized that under-appreciated vascular wall-resident stem and progenitor cells (VW-SPCs) might be involved in tumor vascularization and influence effectiveness of anti-angiogenic therapy. Indeed, in patient samples, we observed that vascular adventitia-resident CD34+ VW-SPCs are recruited to tumors in situ from co-opted vessels. To elucidate this in detail, we established an ex vivo model using concomitant embedding of multi-cellular tumor spheroids (MCTS) and mouse aortic rings (ARs) into collagen gels, similar to the so-called aortic ring assay (ARA). Moreover, ARA was modified by removing the ARs' adventitia that harbors VW-SPCs. Thus, this model enabled distinguishing the contribution of VW-SPCs from that of mature endothelial cells (ECs) to new vessel formation. Our results show that the formation of capillary-like sprouts is considerably delayed, and their number and network formation were significantly reduced by removing the adventitia. Substituting iPSC-derived neural spheroids for MCTS resulted in distinct sprouting patterns that were also strongly influenced by the presence or absence of VW-SPCs, also underlying the involvement of these cells in non-pathological vascularization. Our data suggest that more comprehensive approaches are needed in order to block all of the mechanisms contributing to tumor vascularization.

Use of sedating medications around nursing home admission in Denmark.

PURPOSE: To examine use of sedating medications around the time of nursing home admission in Denmark. METHODS: We conducted a register-based drug utilization study, describing patterns of commonly used medications with sedative effects leading up to and after nursing home admission using data from 94 Danish nursing homes between 2015 and 2017. RESULTS: We identified 5179 residents (median age 84 years, 63% female) and described monthly incidence and total use of benzodiazepines (BZDs), Z drugs, mirtazapine/mianserin, quetiapine, promethazine, and melatonin. The proportion of unique users of sedating medications was similar before and after admission (42% before vs. 40% after) despite an increase in total use after admission. The overall incidence of sedating medications peaked in the 6 months before and 6 months after admission (peaking at 4.6 per 100 person-months 1 month after admission). The most commonly initiated medications were mirtazapine/mianserin, followed by BZDs and Z drugs. Total use of sedating medications increased leading up to admission (peaking at 1001 defined daily doses per 100 residents per month 1 month after admission) and decreased gradually after admission. CONCLUSIONS: Sedative medication initiation increases sharply leading up to admission in Danish nursing homes. Mirtazapine/mianserin is a commonly used agent in nursing homes, despite limited evidence on benefits and harms. Efforts to promote rational use of these medications in nursing homes remain warranted.

LOX-catalyzed collagen stabilization is a proximal cause for intrinsic resistance to chemotherapy.

The potential of altering the tumor ECM to improve drug response remains fairly unexplored. To identify targets for modification of the ECM aiming to improve drug response and overcome resistance, we analyzed expression data sets from pre-treatment patient cohorts. Cross-evaluation identified a subset of chemoresistant tumors characterized by increased expression of collagens and collagen-stabilizing enzymes. We demonstrate that strong collagen expression and stabilization sets off a vicious circle of self-propagating hypoxia, malignant signaling, and aberrant angiogenesis that can be broken by an appropriate auxiliary intervention: Interfering with collagen stabilization by inhibition of lysyl oxidases significantly enhanced response to chemotherapy in various tumor models, even in metastatic disease. Inhibition of collagen stabilization by itself can reduce or enhance tumor growth depending on the tumor type. The mechanistical basis for this behavior is the dependence of the individual tumor on nutritional supply on one hand and on high tissue stiffness for FAK signaling on the other.