Bone Mineral Density Utilization in Patients with Newly Diagnosed Multiple Myeloma: A Single Center Experience

Introduction: Bone disease is a major cause for morbidity in multiple myeloma (MM), mainly manifested by osteolytic lesions. Bone mineral loss is another aspect of bone involvement, although osteoporosis (with or without compression fractures) in the absence of osteolytic lesions is no more a criterion for treatment initiation in MM.

Methods: We performed a retrospective study to evaluate the use of bone mineral density (BMD) exams by dual-energy X-ray absorptiometry (DXA) among MM patients in a tertiary medical care facility. We included 173 patients with symptomatic MM diagnosed between January 2007 and September 2014 who underwent BMD exam at diagnosis. The T-scores of lumbar spine (LS), left femur neck (FN) and left total hip (TH) were obtained and analyzed. In addition, we have specified the lowest T-score at each site. In patients with follow-up exam, we calculated the relative difference between the baseline and follow-up exams as the percentage change in absolute BMD value expressed in g/cm² as follows: [(BMD at follow-up) - (BMD at baseline)/(BMD at baseline)]*100.

Results: The mean lumbar spine T-score was (-1.3), while the low lumbar spine T-score was (-2.0). At the femur level, the mean FN T-score was (-1.5), while the mean TH T-score was (-1.1) and the mean low femur T-score was (-2.2). There was a strong correlation between spine and femur T-scores (r=0.56-0.61, p<0.0001). Nevertheless, on a multivariate regression analysis, different parameters correlated with the T-scores of the LS and the femur sites. The following variables were encountered as significantly associated with the mean T-score value at the lumbar spine: sex (β coefficient= (-0.17), p=0.05), BMI (β coefficient=0.21, p=0.02) and vertebral fracture(s) (β coefficient= (-0.23), p=0.01). Conversely, the following variables were found to be independent predictors of the FN T-score: age (β coefficient= (-0.37), p<0.0001) and vertebral fracture (s) (β coefficient= (-0.25), p=0.005), whereas vitamin D showed a strong trend towards significance (β coefficient= (-0.15), p=0.08). Patients with light chain only disease had lower T-score values compared with patients with IgG myeloma, reaching significance at the femur sites [FN T-score (-1.7) vs. (-1.3), p=0.06; TH T-score (-1.3) vs. (-0.9), p=0.02; femur low T-score (-2.4) vs. (-1.8), p=0.008]. Patients with vertebral fracture(s) had significantly lower T-scores of the spine compared to patients without vertebral fractures. Sixty-three patients (36.4% of all patients) had a follow-up DXA exam at a median of 25 months from the baseline test (range, 12-82 months). Basically, LS T-scores increased, while femur sites' T-scores decreased during follow-up. This did not include patients who achieved complete response (CR) and/or retained it during follow-up, who had improved BMD results at the femur sites as well (Table).

Conclusion: DXA-based BMD assessment of myeloma bone disease is a valuable tool that illustrates a differential myeloma-induced bone mineral loss regarding the lumbar spine and femur.