% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@ARTICLE{Eckardt:276101,
author = {J.-N. Eckardt and C. Röllig and K. Metzeler and P. Heisig
and S. Stasik and J.-A. Georgi and F. Kroschinsky and F.
Stölzel and U. Platzbecker and K. Spiekermann and U. Krug
and J. Braess and D. Görlich and C. Sauerland and B.
Woermann and T. Herold and W. Hiddemann and C.
Müller-Tidow$^*$ and H. Serve and C. D. Baldus and K.
Schäfer-Eckart and M. Kaufmann and S. W. Krause and M.
Hänel and W. E. Berdel and C. Schliemann and J. Mayer and
M. Hanoun and J. Schetelig and K. Wendt and M.
Bornhäuser$^*$ and C. Thiede and J. M. Middeke},
title = {{U}nsupervised meta-clustering identifies risk clusters in
acute myeloid leukemia based on clinical and genetic
profiles.},
journal = {Communications medicine},
volume = {3},
number = {1},
issn = {2730-664X},
address = {[London]},
publisher = {Springer Nature},
reportid = {DKFZ-2023-01001},
pages = {68},
year = {2023},
abstract = {Increasingly large and complex biomedical data sets
challenge conventional hypothesis-driven analytical
approaches, however, data-driven unsupervised learning can
detect inherent patterns in such data sets.While
unsupervised analysis in the medical literature commonly
only utilizes a single clustering algorithm for a given data
set, we developed a large-scale model with 605 different
combinations of target dimensionalities as well as
transformation and clustering algorithms and subsequent
meta-clustering of individual results. With this model, we
investigated a large cohort of 1383 patients from 59 centers
in Germany with newly diagnosed acute myeloid leukemia for
whom 212 clinical, laboratory, cytogenetic and molecular
genetic parameters were available.Unsupervised learning
identifies four distinct patient clusters, and statistical
analysis shows significant differences in rate of complete
remissions, event-free, relapse-free and overall survival
between the four clusters. In comparison to the
standard-of-care hypothesis-driven European Leukemia Net
(ELN2017) risk stratification model, we find all three
ELN2017 risk categories being represented in all four
clusters in varying proportions indicating unappreciated
complexity of AML biology in current established risk
stratification models. Further, by using assigned clusters
as labels we subsequently train a supervised model to
validate cluster assignments on a large external multicenter
cohort of 664 intensively treated AML patients.Dynamic
data-driven models are likely more suitable for risk
stratification in the context of increasingly complex
medical data than rigid hypothesis-driven models to allow
for a more personalized treatment allocation and gain novel
insights into disease biology.},
subtyp = {Editorial},
cin = {DD01 / HD01},
cid = {I:(DE-He78)DD01-20160331 / I:(DE-He78)HD01-20160331},
pnm = {899 - ohne Topic (POF4-899)},
pid = {G:(DE-HGF)POF4-899},
typ = {PUB:(DE-HGF)16},
pubmed = {pmid:37198246},
doi = {10.1038/s43856-023-00298-6},
url = {https://inrepo02.dkfz.de/record/276101},
}