EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Application of machine learning to identify risk factors for bloodstream infections caused by different antimicrobial resistance phenotypes in chemotherapy patients

Anwendung von Machine Learning zur Identifikation von Risikofaktoren für Blutstrominfektionen durch verschiedene Antimikrobiell-Resistenz-Phänotypen bei Chemotherapiepatient:innen

Adhemar Villani Junior, Maristela P Freire, Luis Alberto De Padua Covas Lage, Felippe Lazar Neto, Maura Salaroli Oliveira, Edson Abdala, Fatima L S Nunes, Anna Sara S Levin

Supportive Care in Cancer · 2026-11 · 34 (11)

DOI: 10.1007/s00520-026-11379-2 ↗

Auf einen Blick

In 1.435 bestätigten BSI-Episoden wurden mit Machine-Learning-Modellen Merkmale verschiedener Erregerprofile untersucht. Die Trennschärfe war für CR-GNB am höchsten (AUC-ROC 83,25 %; 95 %-KI 73,91 %–91,64 %).

Fragestellung

Welche Merkmale sind bei Chemotherapiepatient:innen mit unterschiedlichen mikrobiologischen Profilen von Blutstrominfektionen assoziiert?

Methode

Retrospektive Analyse von 1.435 bestätigten BSI-Episoden bei 1.220 Patient:innen; Multi-Label-Klassifikation mit logistischem Regressions-, Random-Forest-, Neural-Network- und Gradient-Boosting-Modell; Kreuzvalidierung und unabhängiger Testdatensatz.

Ergebnisse

Die meisten BSIs traten im ersten Chemotherapiezyklus auf und waren nicht-MDRO oder fungal bedingt (1.193 Episoden; 83,1 %). CR-GNB wurde in 109 Episoden (7,6 %) identifiziert. Wichtige Merkmale waren Hospitalisierungstage, hämatologische Malignität, frühere MDRO-Infektion, Monocytopenie und palliative Chemotherapie.

Schlussfolgerung der Autor:innen

Multi-Label-Machine-Learning-Ansätze könnten helfen, die komplexen Ursachen von BSIs bei Krebspatient:innen durch Assoziationen zwischen Wirtsfaktoren, früherer MDRO-Infektion und Chemotherapieklasse zu ordnen und künftige Hypothesen zu generieren.

Kernaussagen

  • CR-GNB: höchste Trennschärfe mit AUC-ROC 83,25 %
  • Häufigste Gruppe: non-MDRO/non-fungal mit 83,1 %
  • Frühere CR-GNB-Infektion war stärkster Prädiktor für CR-GNB-BSI

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Supportive Care in Cancer, online erschienen am 20.10.2026. Verfasst von Adhemar Villani Junior und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Arts, Sciences and Humanities, University of Sao Paulo, Sao Paulo, Brazil. Hauptthemen laut MEDLINE: Machine Learning, Neoplasms, Antineoplastic Agents, Bacteremia. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

Bloodstream infections (BSIs) remain a major cause of morbidity and mortality in patients undergoing chemotherapy. The rising incidence of multidrug-resistant organisms (MDROs) further challenges the adequacy of empirical antimicrobial therapy. Therefore, this study aimed to identify features associated with different microbiological profiles of BSIs in chemotherapy patients using machine learning models.

Methods

We retrospectively analyzed 1435 confirmed BSI episodes among 1220 patients treated at a tertiary cancer center from 2017 to 2022. A multi-label classification approach was used to model BSIs associated with carbapenem-resistant Gram-negative bacilli (CR-GNB), methicillin-resistant staphylococci, third-generation cephalosporin-resistant Enterobacterales (3GCRE), fungal pathogens, and non-MDRO/non-fungal. Classifier chain models based on logistic regression, random forests, neural networks, and gradient boosting were trained and evaluated using cross-validation with an independent test set. Model outputs were interpreted using SHapley Additive exPlanations (SHAP) to estimate feature contributions.

Results

Most BSIs occurred during the first chemotherapy cycle and were due to non-MDROs or fungal agents (1193 episodes; 83.1%), CR-GNB was identified in 109 (7.6%). Model performance varied across pathogen groups, with the highest discrimination observed for CR-GNB (AUC-ROC 83.25%; 95% CI 73.91%–91.64%). Key features included hospitalization days, hematologic malignancy, prior MDRO infection, monocytopenia, and palliative chemotherapy. Chemotherapy class contributed differently across pathogen groups, with alkylating agents contributing to candidemia, antitumor antibiotics to CR-GNB, and antimetabolites to 3GCRE. Prior CR-GNB infection was the strongest predictor for CR-GNB BSI, while palliative chemotherapy correlated with non-MDRO/non-fungal BSI.

Conclusion

Multi-label machine learning approaches may help disentangle the complex etiologies of BSIs in cancer patients by identifying associations between host factors, prior MDRO infection, and chemotherapy class within confirmed BSI episodes, informing future hypothesis generation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningNeoplasmsAntineoplastic AgentsBacteremiaHumansRetrospective StudiesRisk FactorsFemaleMaleMiddle AgedAgedClassification AlgorithmsPredictive Learning ModelsRandom ForestPhenotypeAnti-Bacterial AgentsDrug Resistance, Multiple, BacterialBoosting Machine Learning Algorithms

Stichwörter der Autor:innen

Antitumor antibioticsCarbapenem-resistant Gram-negativeEnterobacteralesMulti-label machine learning classifiersMultidrug resistance

Substanzen

Antineoplastic AgentsAnti-Bacterial Agents

Autor:innen und Institutionen

  • School of Arts, Sciences and Humanities, University of Sao Paulo, Sao Paulo, Brazil.
  • Infectious Diseases Service, Cancer Institute, University of São Paulo School of Medicine, Sao Paulo, Brazil. maristelapf@uol.com.br.
  • Department of Infectious Diseases, University of São Paulo School of Medicine, Dr Eneas de Carvalho Aguiar Street, 255, Sao Paulo, SP, 05403-900, Brazil. maristelapf@uol.com.br.
  • Department of Hematology, Cancer Institute, University of Sao Paulo School of Medicine, Sao Paulo, Brazil.
  • Department of Oncology, Cancer Institute, University of Sao Paulo School of Medicine, Sao Paulo, Brazil.
  • Department of Infectious Diseases, University of São Paulo School of Medicine, Dr Eneas de Carvalho Aguiar Street, 255, Sao Paulo, SP, 05403-900, Brazil.

Zitieren

APA

Villani Junior, A., Freire, M. P., De Padua Covas Lage, L. A., Lazar Neto, F., Oliveira, M. S., Abdala, E., Nunes, F. L. S., & Levin, A. S. S. (2026). Application of machine learning to identify risk factors for bloodstream infections caused by different antimicrobial resistance phenotypes in chemotherapy patients. Supportive Care in Cancer, 34(11). https://doi.org/10.1007/s00520-026-11379-2

Vancouver

Villani Junior A, Freire MP, De Padua Covas Lage LA, Lazar Neto F, Oliveira MS, Abdala E, et al. Application of machine learning to identify risk factors for bloodstream infections caused by different antimicrobial resistance phenotypes in chemotherapy patients. Supportive Care in Cancer. 2026;34(11). doi:10.1007/s00520-026-11379-2

BibTeX
@article{VillaniJunior2026,
  author = {Villani Junior, Adhemar and Freire, Maristela P and De Padua Covas Lage, Luis Alberto and Lazar Neto, Felippe and Oliveira, Maura Salaroli and Abdala, Edson and Nunes, Fatima L S and Levin, Anna Sara S},
  title = {Application of machine learning to identify risk factors for bloodstream infections caused by different antimicrobial resistance phenotypes in chemotherapy patients},
  journal = {Supportive Care in Cancer},
  year = {2026},
  volume = {34},
  number = {11},
  doi = {10.1007/s00520-026-11379-2}
}
Kostenlos starten →