Routine laboratory trajectories predict organ-level complications in cancer

Routine laboratory panels drawn during cancer treatment constitute longitudinal physiological recordings of organ function, yet their temporal structure is discarded by single-timepoint prognostic tools. A transformer trained on 2,777,595 laboratory measurements from 3905 patients with multiple myeloma or ovarian cancer predicted which of 162 treatment-associated complications would be newly recorded within two years, spanning eight clinical categories, achieving 1.5- to 6.1-fold enrichment above prevalence. Across individual endpoints it exceeded non-sequential baselines, including engineered-trajectory models, by +0.024 to +0.070 in mean AUROC. Predictions generalised across both cancers, divergence concentrating in disease-specific complications, and biomarker masking recovered signatures consistent with established pathophysiology. External validation on MIMIC-IV and MMRF CoMMpass showed transferability, strongest for renal and metabolic endpoints, attenuated for infection endpoints. On CoMMpass it added prognostic value to ISS staging. Routine oncological laboratory data carry complication-specific physiological signal months before a complication is recorded, and require no testing beyond standard oncological care.

Authors

Institutions

Publication Details

Journal
npj Digital Medicine
Published
2026-11-17
DOI
https://doi.org/10.1038/s41746-026-03405-8
Primary Topic
Machine Learning in Healthcare
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Routine laboratory trajectories predict organ-level complications in cancer

Marion Högner, Keno K. Bressem, Krischan Braitsch, Tristan Lemke et al.
2 links
npj Digital Medicine
Machine Learning in Healthcare
article

Routine laboratory trajectories predict organ-level complications in cancer

Marion Högner, Keno K. Bressem, Krischan Braitsch, Tristan Lemke, Sebastian Ziegelmayer, Jacqueline Lammert, Florian Gabriel, Christian Wachinger, Anirudh Narayanan, Jannik Lübberstedt, Christof Winter, Friedrich Puttkammer, Hartmut Häntze, Florian Bassermann, Marcus Makowski, Johannes Moll, Fabian Drexel, Andrei Zhukov, Christopher Zirn, Zeineb Ben Chaaben, Jan Kirschke, Su Hwan Kim, Lisa Adams, Markus Graf
article en

Abstract

Routine laboratory panels drawn during cancer treatment constitute longitudinal physiological recordings of organ function, yet their temporal structure is discarded by single-timepoint prognostic tools. A transformer trained on 2,777,595 laboratory measurements from 3905 patients with multiple myeloma or ovarian cancer predicted which of 162 treatment-associated complications would be newly recorded within two years, spanning eight clinical categories, achieving 1.5- to 6.1-fold enrichment above prevalence. Across individual endpoints it exceeded non-sequential baselines, including engineered-trajectory models, by +0.024 to +0.070 in mean AUROC. Predictions generalised across both cancers, divergence concentrating in disease-specific complications, and biomarker masking recovered signatures consistent with established pathophysiology. External validation on MIMIC-IV and MMRF CoMMpass showed transferability, strongest for renal and metabolic endpoints, attenuated for infection endpoints. On CoMMpass it added prognostic value to ISS staging. Routine oncological laboratory data carry complication-specific physiological signal months before a complication is recorded, and require no testing beyond standard oncological care.

npj Digital MedicineVol. 9(1)
TUM Klinikum (DE), National Center for Tumor Diseases (DE), Essen University Hospital (DE), Deutsches Herzzentrum München (DE), Deutsches Konsortium für Translationale Krebsforschung (DE), Technical University of Munich (DE), Charité - Universitätsmedizin Berlin (DE)
Amgen, Bristol-Myers Squibb, Pfizer, Sanofi, Bayerisches Staatsministerium für Gesundheit, Pflege und Prävention
Good health and well-being
Machine Learning in Healthcare
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon