Myeloma Progression Without Osteolysis: Rethinking Whole-Body Low-Dose CT Criteria for Bone Disease

Abstract

Whole-body low-dose CT (WBLDCT) is recommended as the first-line modality for assessing osteolytic bone disease in multiple myeloma [1]. Under current criteria, myeloma-related bone disease requires at least one osteolytic lesion of 5 mm or more; focal hyperattenuating medullary lesions in the appendicular skeleton, however large or numerous, do not satisfy this criterion [2]. MRI and PET/CT apply a different standard, in which focal bone marrow lesions meeting modality-specific criteria are considered positive for myeloma activity, irrespective of location [3, 4]. We report a patient in whom progressive hyperattenuating femoral medullary lesions were visible on two consecutive annual WBLDCT examinations, neither of which fulfilled the criteria for myeloma-related bone disease in the absence of osteolysis; symptomatic myeloma was established on subsequent MRI.

A 46-year-old man with smoldering multiple myeloma underwent WBLDCT at initial diagnosis, which showed no osteolytic lesions; whole-body MRI, performed per current recommendations, showed no focal lesions. He entered annual WBLDCT surveillance per institutional practice. Two years later, in his third annual WBLDCT, no osteolytic lesions were identified, whereas a single 5 mm hyperattenuating lesion was present in the mid-diaphyseal medullary cavity of the left femur (Figure 1). The finding did not fulfill the criteria for myeloma-related bone disease, and surveillance was continued. At the next annual examination, osteolytic lesions were again absent, whereas the left femoral lesion had enlarged to approximately 3.1 cm, and multiple new focal hyperattenuating lesions had developed in both femoral medullary cavities (Figure 2); the examination again did not establish myeloma bone disease under current criteria, despite clear interval progression of the imaging findings. Subsequently, whole-body MRI was performed. It confirmed the femoral focal lesions and showed additional focal lesions in the axial skeleton (Figure 3a). Retrospective review of the WBLDCT showed no osteolysis at the sites of the MRI focal lesions in the axial skeleton (Figure 3b,c). The patient was reclassified to symptomatic myeloma on the basis of the presence of more than one focal lesion greater than 5 mm on MRI. No other CRAB-SLiM criteria were met during the 2-year surveillance period. The patient initiated treatment with daratumumab, bortezomib, lenalidomide, and dexamethasone, and underwent autologous transplantation, achieving an MRD-negative complete response.

This case illustrates several points regarding the imaging criteria for myeloma-related bone disease. Although myelomatous involvement predominates in the axial skeleton, a minority of patients harbor focal lesions confined to the appendicular skeleton. The precise frequency of truly appendicular-confined disease is unknown, and existing figures are both dated and methodologically non-comparable; they depend on the imaging modality, on anatomic coverage, and on differing definitions of “axial” versus “extra-axial.” The often-quoted figure of approximately 10% derives from an early whole-body versus spinal MRI study [5] in which the axial skeleton was defined as the spine and sacrum alone, so that all other bones, including marrow-rich central sites, were counted as extra-axial; this classification likely overestimates the fraction of exclusive extra-axial disease. Our patient, whose disease was most conspicuous in the femurs, exemplifies this appendicular-predominant presentation, the true frequency of which remains undefined.

The imaging criteria for focal bone marrow lesions on MRI and FDG-PET/CT apply irrespective of anatomic location: a focal lesion fulfilling modality-specific criteria in a long bone is weighted identically to one in the spine or pelvis or other axial skeleton sites. The criterion for WBLDCT, by contrast, is confined to osteolysis. This distinction is consequential because osteolysis, by definition the destruction of mineralized bone, cannot arise within the spacious, largely trabecula-free medullary cavities of the appendicular long bones. In this compartment myelomatous infiltration may nonetheless be demonstrable on WBLDCT as increased medullary attenuation. Indeed, hyperattenuating medullary lesions in the femurs and humeri have long been described and correlate with marrow involvement, yet they are not incorporated into the diagnostic criteria [6, 7]. In our patient, such a lesion enlarged from 5 mm to 3.1 cm, with further lesions appearing in both femurs, while no osteolytic lesions developed at any skeletal site in the interim. The same femoral lesions that fulfilled a myeloma-defining event on MRI carried no diagnostic weight on CT.

Taken together, these observations suggest that the current CT criteria merit reconsideration, so that unequivocal non-osteolytic medullary lesions in the appendicular skeleton can contribute to establishing myeloma-related bone disease. The range of attenuation values for appendicular medullary involvement has been described [7], indicating that such lesions can be defined objectively rather than subjectively. At a minimum, the identification of hyperattenuating medullary lesions on WBLDCT at any point during the course of imaging surveillance should prompt complementary whole-body MRI or PET/CT [8]. Finally, this case underscores that focal lesions identified on MRI in the axial skeleton need not have an osteolytic counterpart on CT; on retrospective review, none of the axial focal lesions in our patient showed osteolysis on WBLDCT. This reflects the superior sensitivity of MRI for bone marrow involvement that has not produced bone destruction and reinforces the complementary, rather than interchangeable, roles of the two modalities in the assessment of myeloma-related bone disease.

The authors have nothing to report.

The authors have nothing to report.

The patient provided informed consent for his data to be used for research and publication purposes.

The authors declare no conflicts of interest.

The data that support the findings of this study are available from the corresponding author upon reasonable request.