PURPOSE: Stratifying patients with cancer according to risk of relapse can personalize their care. In this work, we provide an answer to the following research question: How to use machine learning to estimate probability of relapse in patients with early-stage non-small-cell lung cancer (NSCLC)? MATERIALS AND METHODS: For predicting relapse in 1,387 patients with early-stage (I-II) NSCLC from the Spanish Lung Cancer Group data (average age 65.7 years, female 24.8%, male 75.2%), we train tabular and graph machine learning models. We generate automatic explanations for the predictions of such models. For models trained on tabular data, we adopt SHapley Additive exPlanations local explanations to gauge how each patient feature contributes to the predicted outcome. We explain graph machine learning predictions with an example-based method that highlights influential past patients. RESULTS: Machine learning models trained on tabular data exhibit a 76% accuracy for the random forest model at predicting relapse evaluated with a 10-fold cross-validation (the model was trained 10 times with different independent sets of patients in test, train, and validation sets, and the reported metrics are averaged over these 10 test sets). Graph machine learning reaches 68% accuracy over a held-out test set of 200 patients, calibrated on a held-out set of 100 patients. CONCLUSION: Our results show that machine learning models trained on tabular and graph data can enable objective, personalized, and reproducible prediction of relapse and, therefore, disease outcome in patients with early-stage NSCLC. With further prospective and multisite validation, and additional radiological and molecular data, this prognostic model could potentially serve as a predictive decision support tool for deciding the use of adjuvant treatments in early-stage lung cancer.

• MeSH
• lidé MeSH
• lokální recidiva nádoru diagnóza   MeSH
• nádory plic * diagnóza   terapie   MeSH
• nemalobuněčný karcinom plic * diagnóza   terapie   MeSH
• prognóza MeSH
• senioři MeSH
• strojové učení MeSH
• Check Tag
• lidé MeSH
• mužské pohlaví MeSH
• senioři MeSH
• ženské pohlaví MeSH
• Publikační typ
• časopisecké články MeSH
• práce podpořená grantem MeSH

Disclosures: Eduardo Díaz-Rubio: Roche (C/A, RF); Auxiliadora Gómez-España: None; Bartomeu Massutí: Roche (C/A); Javier Sastre: None; Albert Abad: Roche (C/A); Manuel Valladares: Roche (C/A, RF, H); Fernando Rivera: Roche (C/A, RF); Maria J. Safont: None; Purificación Martínez de Prado: None; Manuel Gallén: None; Encarnación González: None; Eugenio Marcuello: None; Manuel Benavides: Roche (C/A); Carlos Fernández-Martos: None; Ferrán Losa: None; Pilar Escudero: None; Antonio Arrivi: None; Andrés Cervantes: Roche (H); Rosario Dueñas: None; Amelia López-Ladrón: None; Adelaida Lacasta: None; Marta Llanos: None; Jose M. Tabernero: Roche, Genentech, Sanofi- Aventis (C/A); Antonio Antón: None; Enrique Aranda: Roche, Merck Serono (C/A). (C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board. Purpose: The aim of this phase III trial was to compare the efficacy and safety of bevacizumab alone with those of bevacizumab and capecitabine plus oxaliplatin (XELOX) as maintenance treatment following induction chemotherapy with XELOX plus bevacizumab in the first-line treatment of patients with metastatic colorectal cancer (mCRC). Patients and Methods: Patients were randomly assigned to receive six cycles of bevacizumab, capecitabine, and oxaliplatin every 3 weeks followed by XELOX plus bevacizumab or bevacizumab alone until progression. The primary endpoint was the progression-free survival (PFS) interval; secondary endpoints were the overall survival (OS) time, objective response rate (RR), time to response, duration of response, and safety. Results: The intent-to-treat population comprised 480 patients (XELOX plus bevacizumab, n = 239; bevacizumab, n = 241); there were no significant differences in baseline characteristics. The median follow-up was 29.0 months (range, 0–53.2 months). There were no statistically significant differences in the median PFS or OS times or in the RR between the two arms. The most common grade 3 or 4 toxicities in the XELOX plus bevacizumab versus bevacizumab arms were diarrhea, hand–foot syndrome, and neuropathy. Conclusion: Although the noninferiority of bevacizumab versus XELOX plus bevacizumab cannot be confirmed, we can reliably exclude a median PFS detriment >3 weeks. This study suggests that maintenance therapy with singleagent bevacizumab may be an appropriate option following induction XELOX plus bevacizumab in mCRC patients.

• MeSH
• analýza přežití MeSH
• deoxycytidin analogy a deriváty   aplikace a dávkování   terapeutické užití   MeSH
• financování organizované MeSH
• fluorouracil analogy a deriváty   aplikace a dávkování   terapeutické užití   MeSH
• klinické zkoušky, fáze III jako téma MeSH
• kolorektální nádory farmakoterapie   sekundární   MeSH
• lidé MeSH
• monoklonální protilátky aplikace a dávkování   terapeutické užití   MeSH
• multicentrické studie jako téma MeSH
• nežádoucí účinky léčiv MeSH
• organoplatinové sloučeniny aplikace a dávkování   terapeutické užití   MeSH
• přežití po terapii bez příznaků nemoci MeSH
• protokoly antitumorózní kombinované chemoterapie terapeutické užití   MeSH
• statistika jako téma MeSH
• Check Tag
• lidé MeSH
• Publikační typ
• přehledy MeSH

• Publikační typ
• abstrakt z konference MeSH