ESTRO 2024 - Abstract Book

S1791

Clinical - Lung

ESTRO 2024

importance of personalized treatment strategies in managing limited stage SCLC, underscoring the potential benefits of bi-fractionation radiotherapy coupled with appropriate prophylactic measures.

Keywords: SCLC, BID, PCI

References:

(1) Faivre-Finn et al. The Lancet Oncology 2017 Concurrent once-daily versus twice-daily chemoradiotherapy in patients with limited-stage small-cell lung cancer (CONVERT): an open-label, phase 3, randomised, superiority trial

(2) Bogart et al. Journal of clinical Oncology 2023 High-Dose Once-Daily Thoracic Radiotherapy in Limited-Stage Small-Cell Lung Cancer: CALGB 30610 (Alliance)/RTOG 0538

(3) Arriagada R, Ann Oncol 2002 Patterns of failure after prophylactic cranial irradiation in small-cell lung cancer: analysis of 505 randomized patients

2908

Digital Poster

Research on implementing AI to predict the efficacy of immunotherapy in NSCLC based on CT images

Xin Yang, Yanyan Long, Bin Feng, Fu Jin, Huanli Luo

Chongqing University Cancer Hospital, Department of Radiation Therapy, Chongqing, China

Purpose/Objective:

Based on pre-treatment CT (computed tomography) images, deep learning methods are used to construct a prediction model for the efficacy of anti-PD-1/PD-L1 immune checkpoint inhibitors in the first-line treatment of stage IV non-small cell lung cancer, to help radiotherapists analyze clinical factors, radiation Combination of omics features and CT images predicts patient prognosis before treatment.

Material/Methods:

The basic clinical information and treatment CT images of a total of 116 patients with advanced non-small cell lung cancer treated with anti-PD-1/PD-L1 immunotherapy in our hospital from January 2019 to December 2022 were retrospectively collected. Recent efficacy and survival data. Through pyradiomics, a total of 1425 effective radiomics features of the lung and GTV in the patients' pre-treatment enhanced and non-enhanced CT images were extracted, and the improved least absolute shrinkage and selection operator (LASSO) method was used to construct a radiomics efficacy prediction model. Patients were randomly divided into a training set (90%) and a validation set (10%), and 5-fold cross-validation was used to test the generalization ability of the radiomics efficacy prediction model in new data sets.