Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster Display session

415P - 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

Date

28 Mar 2025

Session

Poster Display session

Presenters

Alessio Cortellini

Citation

Journal of Thoracic Oncology (2025) 20 (3): S241-S255. 10.1016/S1556-0864(25)00632-X

Authors

A. Cortellini1, E. Garbo2, L. Brunetti1, M. Li3, F. Passiglia4, S. Yendamuri5, E.M. Karapanagiotou6, R. Giusti7, E. Bria8, M. Rost9, G. Minuti10, M. Tiseo11, L. Cantini12, G. Mountzios13, T. Beninato14, F. Aboubakar15, A. Nassar16, B. Ricciuti17, C.M. Caruso18, V. Guarrasi18

Author affiliations

  • 1 Policlinico Universitario Campus Bio-Medico, Rome/IT
  • 2 Dana Farber Cancer Institute, Harvard Medical School, Boston/US
  • 3 The Ohio State University - OSU Wexner Medical Center, Columbus/US
  • 4 Università Degli Studi Di Torino - Orbassano, Orbassano/IT
  • 5 Roswell Park Comprehensive Cancer Center, Buffalo/US
  • 6 Guy's and St. Thomas' Hospital NHS Foundation Trust, London/GB
  • 7 AOU Sant'Andrea, Rome/IT
  • 8 Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome/IT
  • 9 Universitätsklinikum Frankfurt (Johannes-Wolfgang Goethe-Universität), Frankfurt am Main/DE
  • 10 IRCCS Istiuto Nazionale Tumori Regina Elena (IRE), Rome/IT
  • 11 University Hospital of Parma, Parma/IT
  • 12 Fortrea International Holdings B.V., Leiden/NL
  • 13 Henry Dunant Hospital Center, Athens/GR
  • 14 Fondazione IRCCS - Istituto Nazionale dei Tumori, Milan/IT
  • 15 Cliniques Universitaires Saint-Luc (UCLouvain Saint-Luc), 94805 - Woluwe-Saint-Lambert/BE
  • 16 Yale University, New Haven/US
  • 17 Dana Farber Cancer Institute, Boston/US
  • 18 UCBM - Università Campus Bio-Medico di Roma, Rome/IT

Resources

Login to get immediate access to this content.

If you do not have an ESMO account, please create one for free.

Abstract 415P

Background

With nearly one-third of patients with advanced NSCLC and PD-L1 TPS ≥50% surviving beyond five 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 underscore 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 1,050 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 NAIM, a transformer-based AI 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: 0.59–0.72) for the risk of death and an AUC of 0.72 (95%CI: 0.65–0.78) for 5-year survival, identifying ECOG-PS ≥2, increasing age, and metastatic burden as primary risk factors. However, wide confidence intervals for some predictors highlighted statistical instability. NAIM demonstrated robust handling of missing data, with a c-index of 62.98 ± 2.11 for risk of death and an AUC of 60.52 ± 3.71 for 5-year survival. The comprehensive SHAP analysis revealed dynamic, time-dependent patterns, with early mortality dominated by acute factors (e.g., ECOG-PS, steroids) and long-term outcomes increasingly influenced by systemic health (e.g., absence of hypertension/CV comorbidities, BMI). Unexpected insights included the protective role of dyslipidemia (but not statins) and the nuanced impact of smoking status, reflecting evolving disease dynamics and host-tumor interplay.

Conclusions

Our integrative framework illuminates the complexity of long-term outcomes in patients with NSCLC uncovering dynamic, nonlinear prognostication trends. This analysis provides insights into patient trajectories, emphasizing the need for holistic, long-term management strategies.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

A. Cortellini: Financial Interests, Personal, Invited Speaker: AstraZeneca, MSD; Financial Interests, Personal, Advisory Board: Regeneron, Roche, BMS, Amgen, Daiichi Sankyo. R. Giusti: Financial Interests, Personal, Invited Speaker, Writing Engagement: Roche; Financial Interests, Personal, Other, Travel Grant: Pfizer; Financial Interests, Personal, Other, Travel Grant Advisory Board: Takeda. M. Tiseo: Financial Interests, Personal, Advisory Board: AstraZeneca, Boehringer Ingelheim, Roche, Amgen, Takeda, MSD, Merck, BMS, Pfizer, Eli Lilly, Novartis, Janssen, Daiichi Sankyo; Financial Interests, Institutional, Research Grant: AstraZeneca, Boehringer Ingelheim, Roche. All other authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.