2023 World Conference on Lung Cancer (Posters)-EP06.01. A Circulating Tumor Cell-based Artificial Neural Network of Patients with Advanced Non-Small Cell Lung Cancer

EP06.01. A Circulating Tumor Cell-based Artificial Neural Network of Patients with Advanced Non-Small Cell Lung Cancer - PDF(Slides)

Back to course

Pdf Summary

Researchers from the Department of Oncology at Zhongda Hospital, School of Medicine, Southeast University in China have developed an automated predictive model for non-small cell lung cancer (NSCLC) risk assessment. The model is based on an artificial neural network (ANN) constructed from circulating tumor cells (CTCs) in the blood of NSCLC patients. The aim is to assist in clinical decision-making.<br /><br />The study collected blood samples from 71 advanced NSCLC patients before each cycle of anti-tumor treatment. CTCs were isolated from the samples using an inertial microfluidic labyrinth device and analyzed for their correlation with clinical characteristics, treatment efficacy, and patient prognosis. This data was then used to construct the CTC-based ANN model.<br /><br />Results showed that patients with higher CTC counts had an inferior progression-free survival (PFS), suggesting that CTCs are an independent predictor of PFS. The study also found that CTC cluster counts were positively correlated with treatment efficacy. The ANN model, which incorporated CTCs and CTC clusters, demonstrated superior discriminating power compared to the classical Cox model.<br /><br />The researchers concluded that the ANN model based on CTCs can effectively conduct risk stratification, predict tumor progression, and provide a strong basis for the diagnosis and tailored treatment of NSCLC. The model’s high predictive ability was demonstrated by its calibration curve.<br /><br />This study highlights the potential of using CTCs as a biomarker for assessing NSCLC risk and guiding treatment decisions. The automated predictive model could be a valuable tool for clinicians in improving patient outcomes and personalizing treatment strategies for NSCLC patients.

Asset Subtitle

Xuyu Gu

Meta Tag

Speaker Xuyu Gu

Topic Pathology & Biomarkers: Artificial Intelligence in Pathology

Keywords

automated predictive model

non-small cell lung cancer

artificial neural network

circulating tumor cells

CTCs

progression-free survival

CTC counts

CTC cluster counts

risk stratification

personalizing treatment strategies