Although conventional radiography is still the standard staging procedure for newly diagnosed and relapsed multiple myeloma (MM) patients in clinical practice, magnetic resonance imaging (MRI) using T2- and T1-weighted contrasted enhanced images has proved to have advantages over a conventional skeletal survey. Fluoro-desoxy-glucose positron emission tomography (18F-FDG PET) is not yet validated as an imaging method for MM evaluation. The purpose of this study was to prospectively compare MRI and 18F-FDG PET for short-term evaluation of the response in newly diagnosed MM patients and to correlate the imaging assessment with the biological response.
A group of 15 newly diagnosed myeloma patients were treated in a single center. After induction with bortezomib-lenalidomide-dexamethasone (VRD), they all received high dose cyclophosphamide and G-SCF prior to peripheral stem cell collection. The patients were then randomized to either autologous stem cell transplantation (ASCT) with 2 consolidation VRD cycles or 5 VRD cycles. All patients subsequently received maintenance treatment with low dose lenalidomide for 1 year. Imaging was performed at diagnosis, after the first three VRD cycles and immediately before initiating maintenance treatment. The spinal and pelvic MRI protocol consisted of sagittal T1-weighted (T1-w) spin echo and fat-suppressed T2-weighted (T2-w) STIR sequences of the spine, axial and coronal T1-w spin echo and fat-suppressed T2-w STIR sequences of the pelvis and upper parts of the lower limbs and finally a fat-suppressed T1-w sequence after intravenous gadolinium covering the spine, pelvis and upper parts of the lower limbs. The biological response was evaluated according to IMWG criteria. The presence or absence of diffuse bone marrow infiltration (DI), diffuse patchy bone marrow infiltration (DPI) and focal bone marrow lesions (FL) was recorded prospectively. A hypointense T1, hyperintense STIR and enhanced fat-suppressed T1 lesion were scored as FL on MRI scans. The presence of extra-medullary disease (EMD) and the SUVmax of FDG-avid focal lesions were assessed. Complete response (CR) and partial response (PR) were determined by MRI and 18F-FDG PET after induction and before maintenance. CR was defined as the absence of diffuse infiltration or focal lesions for MRI and as the absence of diffuse and/or focal uptake for 18F-FDG PET. The response was considered to be partial if the image differed from CR. We compared the results of the two techniques with the biological response for every patient.
The 15 patients (5 females, 10 males; age: 55 ± 6 years (mean, SD)) were classified according to the ISS (1, n=6; 2, n=8; 3, n=1) and DSS (IIIA, n=14; IIIB, n=1). At baseline, MRI showed DI, DPI and FL in 3, 7 and 189 cases respectively; the 18F-FDG PET uptake displayed a diffuse homogeneous pattern in 4 patients and a diffuse heterogeneous pattern in 11 patients with only 55 focal uptakes. Five cases of extra-medullary uptake were observed on baseline 18F-FDG PET scans and remained at the follow-up. On the basis of the biological response, the patients were classified as MR (n=1), PR (n=7), VGPR (n=5) and CR (n=2) after induction and as MR (n=1), PR (n=6), VGPR (n=3) and CR (n=5) before maintenance treatment. Differences in focal lesion detection were observed between MRI and 18F-FDG PET for small lesions (<10 mm). Two false positive focal uptakes were observed (pulmonary infection, sacrum abnormality).
The status assessment showed concordant results between the two techniques in only 5/15 (CR, n=2; PR, n=3) and 8/15 (CR, n= 2; PR, n=6) cases at induction and before maintenance respectively. MRI showed residual disease in 7/15 patients before the initiation of maintenance treatment.
At diagnosis and during treatment evaluation, both 18-FDG PET and MRI provide information about the extent of the disease, allowing for a more comprehensive assessment of persisting or recurrent myeloma. The morphologic MRI can often be falsely positive due to persistent non-viable lesions. Hence 18-FDG PET might be more suitable than morphologic MRI for determination of the remission status.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.