External validation of a prediction model for bleeding events in anticoagulated cancer patients with venous thromboembolism (PredictAI)

• María Carmen Viñuela-Benéitez ^[1] ; Claudia Iglesias Pérez ^[2] ; Laura Ortega Morán ^[3] ; Ignacio García Escobar ^[15] ; Diego Cacho Lavín ^[4] ; Rut Porta i Balanyà ^[5] ; Silvia García Adrián ^[6] ; Marta Carmona Campos ^[16] ; Gretel Benítez López ^[17] ; José Antonio Santiago Crespo ^[7] ; Miriam Lobo de Mena ^[8] ; Javier Pérez Altozano ^[18] ; Enrique Gallardo Díaz ^[9] ; Julia Tejerina Peces ^[10] ; Pilar Ochoa Rivas ^[11] ; María José Ortiz Morales ^[12] ; Victoria Eugenia Castellón Rubio ^[19] ; Carmen Díez Pedroche ^[13] ; María Rosales Sueiro ^[20] ; Felipe Gonçalves ^[21] ; Manuel Sánchez-Cánovas ^[22] ; Miguel Ángel Ruiz ^[14] ; José Muñoz-Langa ^[23] ; Pedro Pérez Segura ^[10] ; Eva Martínez de Castro ^[4] ; Alberto Carmona-Bayonas ^[22] ; Paula Jiménez-Fonseca ^[2] ; Andrés Jesús Muñoz Martín ^[3]
  1. [1] Universidad Complutense de Madrid

     Universidad Complutense de Madrid

     Madrid, España

     
  2. [2] Hospital Universitario Central de Asturias

     Hospital Universitario Central de Asturias

     Oviedo, España

     
  3. [3] Instituto de Investigación Sanitaria Gregorio Marañón

     Instituto de Investigación Sanitaria Gregorio Marañón

     Madrid, España

     
  4. [4] Hospital Universitario Marqués de Valdecilla

     Hospital Universitario Marqués de Valdecilla

     Santander, España

     
  5. [5] Hospital Universitari de Girona Dr. Josep Trueta

     Hospital Universitari de Girona Dr. Josep Trueta

     Gerona, España

     
  6. [6] Hospital de Móstoles

     Hospital de Móstoles

     Móstoles, España

     
  7. [7] Hospital Virgen de La Luz de Cuenca

     Hospital Virgen de La Luz de Cuenca

     Cuenca, España

     
  8. [8] Hospital General Universitario de Valencia

     Hospital General Universitario de Valencia

     Valencia, España

     
  9. [9] Universitat Autònoma de Barcelona

     Universitat Autònoma de Barcelona

     Barcelona, España

     
  10. [10] Hospital Clínico San Carlos de Madrid

      Hospital Clínico San Carlos de Madrid

      Madrid, España

      
  11. [11] Hospital Central de la Defensa

      Hospital Central de la Defensa

      Madrid, España

      
  12. [12] Hospital Universitario Reina Sofia

      Hospital Universitario Reina Sofia

      Cordoba, España

      
  13. [13] Hospital Universitario 12 de Octubre

      Hospital Universitario 12 de Octubre

      Madrid, España

      
  14. [14] Universidad Autónoma de Madrid

      Universidad Autónoma de Madrid

      Madrid, España

      
  15. [15] Department of Medical Oncology, Hospital Universitario de Toledo, Toledo, Spain
  16. [16] Department of Medical Oncology, Hospital Clínico Universitario de Santiago de Compostela, A Coruña, Spain
  17. [17] Department of Medical Oncology, C.H.U. Insular-Materno Infantil de Gran Canaria, Las Palmas, Spain
  18. [18] Department of Medical Oncology, IMED Hospital Elche, Alicante, Spain
  19. [19] Department of Medical Oncology, Hospital Universitario Torrecárdenas, Almería, Spain
  20. [20] Department of Immunohemotherapy, Portuguese Institute of Oncology of Porto (IPO Porto), Porto, Portugal
  21. [21] Department of Immunohemotherapy, Portuguese Institute of Oncology of Lisbon Francisco Gentil (IPO Lisboa), Lisboa, Portuga
  22. [22] Department of Medical Oncology, Hospital Universitario José Maria Morales Meseguer, Murcia, Spain
  23. [23] Department of Medical Oncology, Hospital Arnau de Villanova de Valencia, Valencia, Spain

  Mostrar afiliaciones +
• Localización: Clinical & translational oncology, ISSN 1699-048X, Vol. 27, Nº. 10, 2025, págs. 4031-4039
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Objective The objective of this study was to validate the PredictAI models for predicting major bleeding (MB) in patients with active cancer and venous thromboembolism (VTE) with anticoagulant (ACO) therapy, within 6 months after primary VTE, using an independent cohort of patients from the TESEO database.

    Methods This study conducted an external validation of the PredictAI models using the international, prospective TESEO registry from July 2018 until October 2021. Data from 40 Spanish and Portuguese hospitals recruiting consecutive cases of cancer-associated thrombosis under anticoagulant treatment and without missing values regarding the model outcome or predictors were used. Patients with baseline MB or unknown MB status during follow-up were excluded for the validation analysis. Logistic regression (LR), decision tree (DT), and random forest (RF) approaches were used to validate the models.

    Results Included patients from the TESEO cohort (2179 patients) had similar key demographics and clinical characteristics to the PredictAI cohort (21,227 patients). During the 6-month follow-up period, 10.9% (n=2314) and 5.9% (n=129) of patients experienced at least one MB event in the PredictAI and TESEO cohorts, respectively. Hemoglobin, metastasis, age, platelets, leukocytes, and serum creatinine were described as predictors for MB in PredictAI; the external validation results in TESEO showed statistical significance by LR and RF approaches, with ROC-AUC values of 0.59 and 0.56, respectively (both p<0.05).

    Conclusion PredictAI models for predicting MB in anticoagulant-treated cancer patients within the first 6 months following VTE diagnosis have been externally validated. These models may be considered as a tool to guide objective decisions regarding the indication or extension of anticoagulant therapy in this population.