Publications by authors named "A HUTTMANN"
Next Generation Sequencing-based subtyping and interim- and end of treatment positron emission tomography (i/eot-PET) monitoring have high potential for upfront and on-treatment risk assessment of diffuse large B-cell lymphoma patients. We performed Dana Farber Cancer Institute (DFCI) and LymphGen genetic subtyping for the HOVON84 (n = 208, EudraCT-2006-005174-42) and PETAL (n = 204, EudraCT-2006-001641-33) trials retrospectively combined with DFCI genetic data (n = 304). For all R-CHOP treated patients (n = 592), C5/MCD- and C2/A53-subtypes show significantly worse outcome independent of the international prognostic index.
View Article and Find Full Text PDFArticle Synopsis
- The study aimed to validate a deep learning model for predicting treatment outcomes in diffuse large B-cell lymphoma patients across 5 clinical trials, comparing it to the international prognostic index (IPI) and radiomic models.
- The deep learning model, trained on PET/CT scans, demonstrated a higher predictive performance (AUC of 0.66) than IPI (AUC of 0.60) and performed well across all trials.
- While the deep learning and clinical PET models showed similar performance (AUC of 0.69), the PET model achieved the highest AUC (0.71), although the deep learning model provided outcomes without requiring tumor delineation.
Background: Intensified systemic chemotherapy has the highest primary cure rate for advanced-stage, classical Hodgkin lymphoma but this comes with a cost of severe and potentially life long, persisting toxicities. With the new regimen of brentuximab vedotin, etoposide, cyclophosphamide, doxorubicin, dacarbazine, and dexamethasone (BrECADD), we aimed to improve the risk-to-benefit ratio of treatment of advanced-stage, classical Hodgkin lymphoma guided by PET after two cycles.
Methods: This randomised, multicentre, parallel, open-label, phase 3 trial was done in 233 trial sites across nine countries.
JCO The primary analysis of the Ro-CHOP phase III randomized controlled trial (ClinicalTrials.gov identifier: NCT01796002) established that romidepsin (Ro) plus cyclophosphamide, doxorubicin, vincristine, prednisone (CHOP) did not yield an increased efficacy compared with CHOP alone as first-line treatment of peripheral T-cell lymphoma. We report the planned final analysis 5 years after the last patient enrolled.
View Article and Find Full Text PDFConvolutional neural networks (CNNs) may improve response prediction in diffuse large B-cell lymphoma (DLBCL). The aim of this study was to investigate the feasibility of a CNN using maximum intensity projection (MIP) images from F-fluorodeoxyglucose (F-FDG) positron emission tomography (PET) baseline scans to predict the probability of time-to-progression (TTP) within 2 years and compare it with the International Prognostic Index (IPI), i.e.
View Article and Find Full Text PDF