HealthMyne was leveraged for lesion segmentation and automatic radiomic feature extraction. Identification of biomarkers allow for the appropriate stratification and selection of patients and effectively supports precision medicine decisions.
The objective of the retrospective study was to assess the radiomics features obtained by CT examination as biomarkers in order to select patients with lung adenocarcinoma who would benefit from immunotherapy. The analysis looked at the radiomic biomarkers that would predict Overall Survival (OS) or Progression Free Survival (PFS) Time.
- 74 patients that underwent immunotherapy were compared to 50 patients in the control group who received chemotherapy/combination therapy
- 573 radiomic metrics were extracted including 1st, 2nd, 3rd & higher order metrics
- 19 radiomic features were significant for predicting OS and 108 radiomic features for predicting PFS time (morphological, intensity, textural & higher-order statistical metrics were included)
The relationship between radiomics and immunotherapeutic response was demonstrated & specific features can be used to select patients with lung adenocarcinoma who would benefit from immunotherapy