2024 World Conference on Lung Cancer (WCLC) - Posters-P3.13C.02 A Plasma Proteomics-Based Model for Clinical Benefit Prediction in Small Cell Lung Cancer Patients Receiving Immunotherapy

P3.13C.02 A Plasma Proteomics-Based Model for Clinical Benefit Prediction in Small Cell Lung Cancer Patients Receiving Immunotherapy

Back to course

Pdf Summary

The document describes the development and validation of PROphetSCLC, a predictive model designed to improve treatment outcomes for patients with extensive-stage small cell lung cancer (SCLC) undergoing immune checkpoint inhibitor (ICI) therapy. This disease often has limited treatment options, and the addition of ICIs to chemotherapy offers only modest improvements. Identifying patients likely to benefit from ICIs is therefore crucial.<br /><br />PROphetSCLC builds on previous work with PROphetNSCLC, a model using plasma proteomics to predict outcomes in non-small cell lung cancer (NSCLC) patients. Utilizing a machine learning approach, PROphetSCLC is based on baseline plasma samples and integrates predictions from two distinct models — one derived from SCLC-specific Resistance-Associated Proteins (SCLC RAPs) and the other from a prognostic signature of proteins identified in an NSCLC dataset.<br /><br />The development process involved proteomic profiling of pretreatment plasma samples from 79 SCLC patients using aptamer-based technology, which measured about 7,000 proteins per sample. The two models were combined to form a hybrid model that stratifies patients into 'POSITIVE' or 'NEGATIVE' prognostic groups based on their PROphet score. The POSITIVE group experienced significantly longer overall survival (OS), indicating effective discrimination between survival outcomes.<br /><br />A functional analysis revealed that SCLC RAPs are involved in numerous tumor-promoting processes. The predictive performance of the model was supported by an area under the ROC curve (AUC) of 0.63 and a goodness of fit (R²) of 0.93 between observed and predicted clinical benefit rates.<br /><br />Ongoing clinical trials aim to validate these findings in a larger patient cohort, highlighting the model's potential to inform personalized treatment strategies in SCLC by leveraging insights into biological and clinically relevant biomarkers.

Asset Subtitle

David Gandara

Meta Tag

Speaker David Gandara

Topic SCLC & Neuroendocrine Tumors

Keywords

PROphetSCLC

small cell lung cancer

immune checkpoint inhibitor

plasma proteomics

machine learning

SCLC RAPs

prognostic signature

proteomic profiling

overall survival

personalized treatment