Transformer-based AI approach to unravel long-term, time-dependent prognostic complexity in patients with advanced NSCLC and PD-L1 ≥50% : insights from the pembrolizumab 5-year global registry

作者：Alessio Cortellini, Valentina Santo, Leonardo Brunetti, Edoardo Garbo, David J Pinato, Giulia La Cava, Jarushka Naidoo, Artur Katz, M. Loza, Joel W. Neal, Carlo Genova, Scott Gettinger, So Yeon Kim, Ritujith Jayakrishnan, Talal El Zarif, Marco Russano, Federica Pecci, Alessandro Di Federico, J. Alessi, Michele Montrone, Dwight H. Owen, Sara Ramella, Diego Signorelli, Mary J. Fidler, Mingjia Li, Andrea Camerini, Balazs Halmos, Bruno Vincenzi, Giulio Metro, Francesco Passiglia, Sai Yendamuri, Annalisa Guida, Michele Ghidini, Antonio D'Alessio, Giuseppe L Banna, Claudia A M Fulgenzi, Salvatore Grisanti, Francesco Grossi, Armida D'Incecco, E. Josephides, Mieke Van Hemelrijck, Alessandro Russo, Alain Gelibter, Gianpaolo Spinelli, Monica Verrico, Bartłomiej Tomasik, Raffaele Giusti, K. Balachandran, Emilio Bria, Martin Sebastian, Maximilian Rost, Martin Forster, Uma Mukherjee, Lorenza Landi, Francesca Mazzoni, Avinash Aujayeb, Manuel Dupont, Alessandra Curioni-Fontecedro, Rita Chiari, Vincenzo Sforza, Marcello Tiseo, Alex Friedlaender, Alfredo Addeo, Federica Zoratto, Michele De Tursi, Luca Cantini, Elisa Roca, Giannis Mountzios, Danilo Rocco, Luigi Della Gravara, Sukumar Kalvapudi, Alessandro Inno, Paolo Bironzo, Rafael Di Marco Barros, David O'Reilly, Órla Fitzpatrick, Eleni Karapanagiotou, Isabelle Monnet, Javier Baena, Marianna Macerelli, Aida Piedra, Francesco Agustoni, Diego Luigi Cortinovis, Giuseppe Tonini, Gabriele Minuti, Chiara Bennati, L. Mezquita, Teresa Gorría, Alberto Servetto, Teresa Beninato, Giuseppe Lo Russo, Arsela Prelaj, Andrea De Giglio, Jacobo Rogado, L. Moliner, Ernest Nadal, Federica Biello, Frank Aboubakar Nana, A. Dingemans, Joachim G J V Aerts, Roberto Ferrara, Taher Abu Hejleh, Kazuki Takada, Abdul Rafeh Naqash, Marina Chiara Garassino, Solange Peters, Heather A. Wakelee, Amin H Nassar, Biagio Ricciuti, P. Soda, Camillo Maria Caruso, Valerio Guarrasi · 发表于：Archive ouverte UNIGE (University of Geneva) · 年份：2025 · 研究领域：Radiomics and Machine Learning in Medical Imaging、Cancer Immunotherapy and Biomarkers、Ferroptosis and cancer prognosis

Background: With nearly one-third of patients with advanced non-small cell lung cancer (NSCLC) and PD-L1 Tumor Proportion Score≥50% surviving beyond 5 years following first-line pembrolizumab, long-term outcomes challenge traditional paradigms of cancer prognostication. The emergence of non-cancer-related factors and time-dependent trends underscores the need for advanced analytical frameworks to unravel their complex interplay. Methods: We analyzed the Pembro-real 5Y registry, a global real-world dataset of 1050 patients treated across 61 institutions in 14 countries with a long-term follow-up and a large panel of baseline variables. Two complementary approaches were employed: ridge regression, chosen for its ability to address multicollinearity while retaining interpretability, and not another imputation method (NAIM), a transformer-based artificial intelligence model designed to handle missing data without imputation. Endpoints included risk of death at 6, 12, 24, 60 months and 5-year survival. Results: The ridge regression model achieved a c-statistic of 0.66 (95% CI: 0.59 to 0.72) for the risk of death and an area under the curve (AUC) of 0.72 (95% CI: 0.65 to 0.78) for 5-year survival, identifying Eastern Cooperative Oncology Group Performance Status (ECOG-PS)≥2, increasing age, and metastatic burden as primary risk factors. However, wide CIs for some predictors highlighted statistical instability. NAIM demonstrated robust handling of missing data, with a c-index o...