2023 World Conference on Lung Cancer (Posters)-P2.13. Pathological Images Machine Learning Predicts Long Term Effects for Immunotherapy in Small-Cell Lung Cancer

P2.13. Pathological Images Machine Learning Predicts Long Term Effects for Immunotherapy in Small-Cell Lung Cancer - PDF(Abstract)

Back to course

Pdf Summary

This presentation discusses the use of machine learning to predict the long-term effects of immunotherapy in small-cell lung cancer (SCLC) using pathological images. The study analyzed the APOLLO study, a prospective cohort study of SCLC patients who received chemo-immunotherapy as the first-line treatment. Tumor tissue samples were stained and annotated, and a non-rigid image registration framework was used to analyze the spatial positioning of various inflammatory and tumor cells. Machine learning was then performed by training a classifier to predict 365-day progression-free survival (PFS). Three models were developed: patient information model, pathological image model, and combination model. The accuracy of the machine learning model was evaluated using the area under the curve (AUC). The results showed that the combination model had the highest AUC (0.868), indicating better predictive accuracy. Patients were then classified into high or low efficacy groups based on the developed predictive model, and the median PFS was compared between the two groups. The results showed that the high efficacy group had longer median PFS compared to the low efficacy group. This study suggests that machine learning analysis of pathological images can predict the efficacy of immunotherapy in SCLC and contribute to biomarker development for immunotherapy. The use of artificial intelligence and machine learning in analyzing pathological images has the potential to provide precise and objective spatial analysis, overcoming the challenges faced by humans in evaluating complex interactions in the tumor immune microenvironment.

Asset Subtitle

Ryota Shibaki

Meta Tag

Speaker Ryota Shibaki

Topic SCLC & Neuroendocrine Tumors: Biomarkers & Radiomics

Keywords

machine learning

immunotherapy

small-cell lung cancer

SCLC

pathological images

APOLLO study

predictive model

progression-free survival

area under the curve

biomarker development