Back

The Brain-Lung Immunotherapy Prognostic (BLIP) Score: A Novel Robust Tool for Prognostication in Non-Small Cell Lung Cancer Patients with Brain Metastases

Skribek, M.; Livanou, M.-E.; Vathiotis, I.; Strandman, V.; Thorell, A.; Koulouris, A.; Syrigos, K.; Ekman, S.; Tsakonas, G.

2024-09-24 oncology
10.1101/2024.09.23.24314193 medRxiv
Show abstract

BackgroundLung cancer remains the leading cause of cancer-related mortality, with brain metastases (BMs) significantly worsening prognosis and quality of life. The advent of immune checkpoint inhibitors (ICIs) has revolutionized the treatment landscape for non-small cell lung cancer (NSCLC). However, precise prognostic tools are essential to optimize clinical decision-making in this context. MethodsThe Brain-Lung Immunotherapy Prognostic (BLIP) score was developed based on a retrospective cohort of NSCLC patients treated with ICIs at Karolinska University Hospital, Sweden. Prognostic factors were identified using both univariate and multivariate Cox regression analyses. Internal validation was conducted using bootstrap resampling, penalized Cox regression, k-fold cross-validation, and receiver operating characteristics (ROC) analysis. External validation was performed using an independent cohort from Sotiria Thoracic Diseases Hospital of Athens, Greece. ResultsFrom a total cohort of 1844 patients screened across both study sites, 152 patients from Karolinska University Hospital and 116 from Sotiria Thoracic Diseases Hospital of Athens, Greece, were included in the final analysis. Key prognostic factors influencing outcomes included histology, actionable mutations, age at BM diagnosis, and the number of BMs. The BLIP score effectively stratified patients into two prognostic groups: "Good" and "Poor", with a median overall survival (OS) of 15 and 7 months, respectively (hazard ratio [HR]: 0.4; p < 0.0001). External validation confirmed these findings, showing a significantly lower risk of death for the "Good" group compared to the "Poor" group (HR: 0.49; p = 0.0063). The models robust prognostic performance was confirmed with an area under the ROC curve of 0.87, highlighting its accuracy in predicting survival outcomes. ConclusionThe BLIP score provides a reliable, validated prognostic tool for NSCLC patients with BMs undergoing ICI therapy. By integrating both molecular and clinical variables, it offers significant improvements over existing models. Prospective validation could further support its use in personalized treatment strategies, improving clinical outcomes and patient management. HighlightsO_LIThe BLIP score is a new prognostic tool for NSCLC. C_LIO_LIIt focuses on patients with brain metastases undergoing immunotherapy. C_LIO_LIThe score integrates clinical and molecular factors. C_LIO_LIInternal validation showed strong prognostic power and reliability. C_LIO_LIExternal validation confirmed effectiveness across diverse patient populations. C_LIO_LIKey factors include histology, actionable mutations, age, and brain metastases count. C_LIO_LIThe score stratifies patients into "Good" and "Poor" groups. C_LIO_LIThe BLIP score aids in personalized treatment decision-making. C_LI

Published in British Journal of Cancer (predicted rank #13) · training set

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.