You are here : Home > News > A new biomarker to predict tumor progression from whole body PET images.

Scientific result | Medical imaging | Positron Emission Tomography | Cancer

A new biomarker to predict tumor progression from whole body PET images.


​The SHFJ/IMIV laboratory (Inserm 1023/CNRS/CEA/UP-Sud), in collaboration with hospital-university teams, has shown that, in patients with advanced-stage large cell lymphoma (B type), automatic quantitative measurements of tumor dissemination detected by whole-body PET imaging have a predictive value of progression-free survival (tumor stabilization) and overall survival.

Published on 16 July 2019

Abstract of the original paper

We assessed the prognostic value of new radiomic features (RF) characterizing the lesion dissemination in baseline 18F-FDG PET and tested whether combining them with baseline metabolic tumour volume (MTV) could improve prediction of progression free survival (PFS) and overall survival (OS) in DLBCL patients. Methods: From the LNH073B trial (NCT00498043), patients with an advanced stage DLCBL and 18F-FDG PET/CT images available for review were selected. MTV and several RF, including the distance between the two lesions that were the furthest apart (Dmaxpatient) were calculated. Receiver operator characteristics analysis were used to determine the optimal cut-off for quantitative variables and Kaplan–Meier survival analyses were performed. Results: With a median age of 46 years, 95 patients were enrolled, half of them treated with R-CHOP14 (rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone), the others with R-ACVBP (rituximab, doxorubicin, cyclophosphamide, vindesine, bleomycin, prednisone), with no significant impact on outcome. Median MTV and Dmaxpatient were 375 cm3 and 45 cm respectively. The median follow-up was 44 months. High MTV and Dmaxpatient were adverse factors for PFS (P = 0.027 and P = 0.0003 respectively) and for OS (P = 0.0007, P = 0.0095). In multivariate analysis, only Dmaxpatient was significantly associated with PFS (P = 0.0014) whereas both factors remained significant for OS (P = 0.037 and P = 0.0029 respectively). Combining MTV (>384 cm3) and Dmaxpatient (>58 cm) yielded 3 risk groups for PFS (P = 0.0003) and OS (P = 0.0011): high with 2 adverse factors (4y-PFS and OS of 50% and 53%, n = 18), low risk with no adverse factor (94% and 97%, n = 36), and an intermediate category with one adverse factor (73% and 88%, n = 41). Conclusion: Combining MTV with a parameter reflecting the tumour burden dissemination further improves DLBCL patient risk stratification at staging.

Top page