Imatinib and chronic myeloid leukemia: close to the bone

Imatinib mesylate and, more recently, the second- generation tyrosine kinase inhibitors (TKIs), nilotinib and dasatinib, have been used for several years with excellent results in patients with chronic myeloid leukemia (CML). The effect of TKIs on bone metabolism has been the aim of several studies during the last decade. Initially, Berman and colleagues described that approximately 50% of patients with CML who received imatinib had low serum phosphate levels [1]. This cohort of patients had been given higher doses of imatinib (median dose 600 mg) than patients who had normal phosphate levels. Subsequent studies confirmed this finding, and noted that imatinib-treated patients developed a secondary hyperparathyroidism, due to decreased serum calcium and increased serum 1,25-hydroxyvitamin D₃ [2]. The decreased serum phosphate and calcium level is possibly the result of the removal of these minerals to the bones through an uncoupling of bone remodeling. Normally osteoclasts and osteoblasts balance bone resorption and formation under the regulation of several local and systemic factors. In vitro data have shown that imatinib inhibits osteoclastogenesis induced by the major osteoclast regulators, such as receptor-activator of nuclear factor-kappaB ligand, macrophage-colony stimulating factor and 1,25-hydroxyvitamin D₃ [3]. This reduction in osteoclastogenesis is also supported by other studies that showed that imatinib inhibits c-fms, c-kit and carbonic anhydrase II, which are stimulators of osteoclast activity [2,4]. Imatinib also promotes osteoblast differentiation by inhibiting platelet-derived growth factor signaling and by enhancing the expression of osteogenic genes, including RUNX2, osteocalcin and bone morphogenic protein-2 [3,5]. Although the above in vitro and animal model data suggest that imatinib alters bone remodeling in favor of bone formation, the outcome in patients has not been fully delineated.

In one of the first studies in the field, Fitter and colleagues reported that approximately 50% of patients with CML (eight of 17) increased their trabecular bone volume more than two-fold when they were treated with imatinib [6]. Similarly, Jönsson and colleagues showed that patients with CML had higher bone mineral density (BMD) of the hip and the lumbar spine (assessed by dual energy X-ray absorptiometry [DXA] scan) and higher radial and tibial cortical BMD (assessed by peripheral quantitative computed tomography) compared to age- and gender-matched healthy controls after 24–73 months of imatinib therapy [7].

Here, the elevation in BMD correlated with decreased serum phosphate, calcium and bone formation markers [7]. The authors repeated the analysis of the same group of patients prospectively after 4 more years of imatinib therapy, and found that patients’ parathyroid hormone increased significantly within the 4 years of the study period, and 7/17 patients developed secondary hyperparathyroidism; however, this secondary hyperparathyroidism did not affect their volumetric BMD, which was significantly higher in their cortical compartment when compared with controls [8]. In the first prospective study in the field, O’Sullivan and colleagues observed that there was a significant increase in the BMD of the lumbar spine of nine patients with CML who were treated with imatinib over a 24-month period, but there was no change in proximal femur BMD and only a trend toward an increase in BMD of the total body [9]. Based on the above studies, we can conclude that in the short term there is increased bone formation in patients with CML who are treated with imatinib. But does this effect continue with long-term imatinib therapy? Are these patients protected by osteopenia and osteoporosis?

In this issue of Leukemia and Lymphoma, O’Sullivan and colleagues report the results of 4 years of follow-up of their previous prospective study. They found that after 48 months of imatinib therapy, a mild secondary hyperparathyroidism was present, but did not worsen between 18 and 48 months of the study [10]. BMD of the lumbar spine, left proximal femur and total body was similar to that at baseline, while the major bone formation and resorption markers were below the baseline values. The study suggests that there is a mild alteration of bone remodeling of patients with CML by imatinib which does not affect their BMD long-term. This result is in accordance with a recent study in healthy rats in which the administration of imatinib did not increase BMD and did not alter biochemical markers of bone resorption [11]. Thus, it seems that patients with CML experience initially an increase in bone formation with imatinib therapy, which does not continue after the first 1 or 2 years of treatment. But is there any explanation of this finding? What is the “real” effect of imatinib on bone cells? A possible explanation is that imatinib has different roles in the different phases of bone formation. A recent study showed that although imatinib promoted early osteoblast differentiation, it also reduced mineralization, and this effect was more profound at low concentrations of imatinib [12]. Furthermore, in the same study, the expression of RUNX2 (a master gene that encodes a protein involved in the osteogenic differentiation process from mesenchymal precursors) was enhanced by imatinib at early time points during differentiation, while the expression of osterix (an osteoblast-specific transcription factor which activates a repertoire of genes during differentiation of preosteoblasts into mature osteoblasts and osteocytes) was reduced by imatinib late during differentiation [12]. Therefore, the effect of imatinib on osteoblast differentiation depends also on the maturation stage of the cells, and this finding might partly explain the different results of imatinib in human BMD within the first 1–2 years of treatment and afterwards.

It would appear that imatinib mesylate alters bone remodeling in favor of bone formation and this results in a positive balance on BMD, mainly during the first 18–24 months of therapy; this effect seems not to continue during long-term treatment, and thus BMD is not different from baseline after 4 years of imatinib. A mild secondary hyperparathyroidism occurs on imatinib therapy but has no effect on BMD, after predominantly 2 years of treatment. “Routine” assessment of BMD is unusual in the general CML population treated with imatinib. It is reassuring that after 2 years of imatinib, regular biochemical assessment of the bone profile for the purpose of bone metabolism assessment is unlikely to be necessary.

Potential conflict of interest:

Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.