Abstract
Investigating prognostic factors in relapsed or primary refractory classical Hodgkin lymphoma (R/R cHL) patients is essential to optimize risk-adapted treatment strategies. We built a prognostic model using baseline quantitative 18F-FDG PET radiomics features and clinical characteristics to predict progression free survival (PFS) in R/R cHL patients treated with salvage chemotherapy followed by autologous stem-cell transplant (ASCT). Metabolic tumor volume (MTV) and several novel radiomics dissemination features representing inter-lesional differences in distance, volume and standard uptake value (SUV) were extracted from the baseline PET. Machine learning using backward selection and logistic regression were applied to develop and train the model on a total of 113 patients from two clinical trials (NCT02280993 and NCT00255723). The model was validated on an independent external cohort of 69 patients (NCT01508312). In addition, we validated four different PET segmentation methods to calculate radiomics features. We identified a subset of high-risk patients with significant inferior 3-year PFS outcomes of 38.1% versus 88.4% for patients in the low-risk group in the training cohort (p<0.001), and 38.5% versus 75.0% in the validation cohort (p=0.015), respectively. The overall survival was also significantly better in the low-risk group (p=0.022 and p<0.001). We provide a formula to calculate a risk score for individual patients based on the model. In conclusion, we developed a prognostic model for PFS combining radiomics and clinical features in a large cohort of R/R cHL patients. This model calculates a PET-based risk profile and can be applied to develop risk-stratified treatment strategies for R/R cHL patients.
