Whole-body low-dose CT recognizes two distinct patterns of lytic lesions in multiple myeloma patients with different disease metabolism at PET/MRI

Renato Zambello,Carmelo Lacognata,Albana Lico,Alberto Ponzoni,Antonio Branca,Stefania Vio,Filippo Crimì,Marco Pizzi,Pietro Zucchetta,Ercole De Biasi,Vanna Scapin,Tamara Berno,Lucia Checuz,Gianpietro Semenzato,Cristiano Varin,Roberto Vezzaro,Gregorio Barilà,Annamaria Guolo

doi:10.1007/s00277-018-3555-7

Renato Zambello, Carmelo Lacognata + Show 16 more

Open Access

https://doi.org/10.1007/s00277-018-3555-7

Copy DOI

Abstract

We evaluated differences in density and 18F-FDG PET/MRI features of lytic bone lesions (LBLs) identified by whole-body low-dose CT (WB-LDCT) in patients affected by newly diagnosed multiple myeloma (MM). In 18 MM patients, 135 unequivocal LBLs identified by WB-LDCT were characterized for inner density (negative or positive Hounsfield unit (HU)), where negative density (HU < 0) characterizes normal yellow marrow whereas positive HU correlates with tissue-like infiltrative pattern. The same LBLs were analyzed by 18F-FDG PET/DWI-MRI, registering DWI signal with ADC and SUV max values. According to HU, 35 lesions had a negative density (− 56.94 ± 31.87 HU) while 100 lesions presented positive density (44.87 ± 23.89 HU). In seven patients, only positive HU LBLs were demonstrated whereas in eight patients, both positive and negative HU LBLs were detected. Intriguingly, in three patients (16%), only negative HU LBLs were shown. At 18F-FDG PET/DWI-MRI analysis, negative HU LBLs presented low ADC values (360.69 ± 154.38 × 10−6 mm2/s) and low SUV max values (1.69 ± 0.56), consistent with fatty marrow, whereas positive HU LBLs showed an infiltrative pattern, characterized by higher ADC (mean 868.46 ± 207.67 × 10−6 mm2/s) and SUV max (mean 5.04 ± 1.94) values. Surprisingly, histology of negative HU LBLs documented infiltration by neoplastic plasma cells scattered among adipocytes. In conclusion, two different patterns of LBLs were detected by WB-LDCT in MM patients. Both types of lesions were indicative for active disease, although only positive HU LBL were captured by 18F-FDG PET/DWI-MRI imaging, indicating that WB-LDCT adds specific information.

Highlights

In multiple myeloma (MM) patients, the detection of lytic bone lesions (LBLs) represents a criterion to define symptomatic MM requiring treatment, even in the absence of other clinical symptoms [1,2,3]
To better define the specific contribution to whole-body lowdose CT (WB-LDCT) to skeletal survey in a series of newly diagnosed MM patients, we retrospectively evaluated the features of LBLs detected by WB-LDCT and compared the pattern with 18F-FDG PET/DWI-MRI, which fuse together two very highly sensitive functional techniques suggested by IMWG to assess bone involvement in MM patients [10, 11]
Lesions with positive Houdsfield unit (HU) showed a trend toward a prevalent spine distribution as compared to those with negative HU (60% vs 38%, respectively (p < 0.05), whereas lesions with negative HU were more represented at the site of pelvis, not significantly (45% vs 26%, p = 0.067)

Summary

Introduction

In multiple myeloma (MM) patients, the detection of lytic bone lesions (LBLs) represents a criterion to define symptomatic MM requiring treatment, even in the absence of other clinical symptoms [1,2,3]. By CT imaging evaluation, this feature results in lower CT attenuation values (CTav) than the density of water (which corresponds to HU 0) When this space is infiltrated by neoplastic cells as in MM patients, as a consequence of the destruction of mineralized bone, the marrow lesion densities are characterized by a positive value (HU > 0), consistent with the presence of solid (myelomatous) tissue [8]. To better define the specific contribution to WB-LDCT to skeletal survey in a series of newly diagnosed MM patients, we retrospectively evaluated the features of LBLs detected by WB-LDCT and compared the pattern with 18F-FDG PET/DWI-MRI, which fuse together two very highly sensitive functional techniques suggested by IMWG to assess bone involvement in MM patients [10, 11]

Methods

Results

Conclusion