Standardization of 18F-FDG PET/CT According to Deauville Criteria for MRD Evaluation in Newly Diagnosed Transplant Eligible Multiple Myeloma Patients: Joined Analysis of Two Prospective Randomized Phase III Trials

F-18-fluorodeoxyglucose positron emission tomography integrated with computed tomography (FDG-PET/CT) is actually considered as the standard technique to assess and monitor the metabolic response to therapy and to define minimal residual disease (MRD) status outside the bone marrow (BM) in multiple myeloma (MM) patients. In this regard, standardization of image criteria and definition of cut-offs for positivity/negativity is of highly importance. Aim of the present study was to prospectively evaluate FDG-PET/CT at diagnosis and prior to maintenance therapy in a joined analysis of a sub-group of patients with newly diagnosed transplant-eligible MM, enrolled in 2 independent European randomized phase III trials (EMN02/HO95 and IFM2009) (Cavo M et al, Blood 2017 abs; Attal M et al, NEJM 2017). The primary end-point was to standardize PET/CT evaluation by centralized imaging and revision and to define criteria for PET negativity after therapy (MRD definition). 236 patients (102 and 134 from the EMN02 and IFM2009 trial, respectively) were enrolled in the PET/CT imaging sub-studies and followed for a median of 62.9 (IQR: 44.9-67.9) months. By study design, PET/CT scans were performed locally at baseline (B) and prior to the start of maintenance (PM), uploaded in a central website and re-interpreted a-posteriori in a blinded independent central review process, by a panel of expert nuclear medicine physicians. According to the IMPeTUs criteria (Nanni C et al, EJNM 2017), the five-point Deauville scores (between 1 and 5) were applied to the following parameters: bone marrow (BM), focal lesions (FLs), extramedullary disease (EMD). The impact of each parameter on outcomes was evaluated by landmark analyses at PM; the univariate and multivariate analyses were stratified by trial to smooth out differences between the 2 studies. Baseline characteristics of the patients were generally homogeneous between the 2 trials and as follows: median age 59 years, ISS and R-ISS stage III 15.8% and 11.5%, respectively, high-risk cytogenetics (t(4;14) ± del(17p) ± t(14;16) detected by FISH) 14%. Fifty seven percent of the patients were randomized in the transplant arm, and 43% in the bortezomib-intensification arm, with a higher percentage in the IFM2009 vs EMN02 trial (54% vs 24%, respectively). At baseline, 80% of the patients had FLs, with a median maximum standardized uptake value (SUVmax) of 5. Median BM SUVmax was 3. Both median FLs and BM SUVmax were slightly higher in the IFM