2023 World Conference on Lung Cancer (Posters)-P2.13. Integrated Clinical and Radiomics Approach to Assess the Treatment Outcome of Combined Chemoimmunotherapy in Extensive-Stage SCLC

P2.13. Integrated Clinical and Radiomics Approach to Assess the Treatment Outcome of Combined Chemoimmunotherapy in Extensive-Stage SCLC - PDF(Abstract)

Back to course

Pdf Summary

This study aimed to assess the treatment outcome of combined chemoimmunotherapy in patients with extensive-stage small cell lung cancer (ES-SCLC) using an integrated clinical and radiomics approach. The study enrolled 341 ES-SCLC patients treated with combined Atezolizumab or Durvalumab with standard chemotherapy from five medical centers in China. The researchers utilized machine learning techniques to construct a radiomics-score based on radiomics features extracted from lung CT scans before treatment. They also employed COX regression to identify clinical features associated with therapeutic efficacy. The effectiveness of the approach was evaluated using the Log-rank test, time-dependent ROC, and C-index.<br /><br />The results showed that the radiomics model alone was able to separate patients at different risk levels of disease progression. However, the stability of the model was deemed unsatisfactory. Therefore, T stage, N stage, M stage, ECOG, and LIPI index were incorporated into the combined model, resulting in improved predictive efficacy. The low-risk group identified by the combined model had a significantly longer median progression-free survival (mPFS) compared to the high-risk group in both the training cohort and the validation cohorts. External validation yielded high AUC and C-index values.<br /><br />Furthermore, the combined prediction model demonstrated impressive predictive power for overall survival outcomes. The hazard ratio of overall survival between the low-risk and high-risk groups was significantly lower, indicating better survival outcomes for patients in the low-risk group.<br /><br />In conclusion, the integrated clinical and radiomics approach offers a convenient and low-cost prognostic model for decision management in ES-SCLC patients receiving chemoimmunotherapy. This approach could potentially aid in the evaluation of treatment outcomes and help identify a beneficial population for this therapy.

Asset Subtitle

Yang Xia

Meta Tag

Speaker Yang Xia

Topic SCLC & Neuroendocrine Tumors: Biomarkers & Radiomics

Keywords

chemoimmunotherapy

ES-SCLC

radiomics approach

machine learning techniques

therapeutic efficacy

progression-free survival

external validation

overall survival outcomes

decision management

prognostic model