Glucocorticoids Attenuate the Stimulatory Effect on Bone Formation by Bortezomib

Multiple myeloma (MM) is characterised by extensive osteolysis both due to an increased osteoclast activity and a decrease of bone formation by osteoblasts. Treatment with bisphosphonates and chemotherapy can inhibit the osteoclast activity and thereby reduce osteolysis, but until recently no drugs have been able to overcome the inhibition of osteoblasts in MM. In 2005 it was found that bortezomib (bz) has a positive effect on the bone formation marker bone specific alkaline phosphatase (BAP) in MM patients. Other studies have confirmed this finding. However, bz is increasingly used in combination with other drugs especially glucocorticoids (gc), which may have negative effects on bone formation. So far there has been conflicting results as to how the addition of gc to bz treatment may affect bone. Secondly, it remains unclear to what extent the effect of bz on bone is dependent on the response to treatment. A possible reason for these conflicting results may be that the patients in these studies have been treated previously and also have received bisphosphonate treatment which is known to have a profound and long lasting effect on bone. It is still not known what the effect may be in previously untreated patients. Finally the kinetics of bone markers in MM patients treated with bz have never been investigated.

We have initiated a phase 2 clinical trial including only newly diagnosed, previously untreated MM patients. The patients in this study receive one 3-weeks cycle of bz 1.3 mg/m2 by i.v. on day 1, 4, 8 and 11 as monotherapy followed by three cycles of bz and gc in combination with dexamethasone 20 mg/day given on the day of bz treatment and the following day. Blood and urine samples were collected at sixteen different time-points during the course of treatment so the effects of single agent and combined treatment on biochemical markers of bone formation could be evaluated in each individual patient in a consecutive manner. Bisphosphonate treatment was not permitted. In addition to BAP we also measured PINP, a novel marker that measures bone formation activity, and the osteoblast inhibitor Dkk-1. As a supplement to the clinical study, the effect of single and combined pulse treatment with bz and gc was tested in an in vitro model of osteoblast like cells using immortalised and primary human mesenchymal stem cells (hMSC). Response was evaluated by measuring osteopontin and osteocalcin using Q-PCR.

The planed 20 patients have been included in the trial and 12 patients are mature for evaluation. In patients with VGPR or better a 90 percent increase of BAP was observed already after one treatment cycle of bz given as monotherapy. After the addition of gc to the treatment, BAP was still increased but only to a value 46 percent above the pre-treatment level. PINP showed a similar pattern increasing by 91 percent after the first cycle followed by an attenuated response after addition of gc to a value of 33 percent above the pre-treatment level. However in addition PINP temporary decrease on the days when gc were given only to partially recover afterwards. No significant changes were observed in BAP or PINP in patients with less than VGPR. Dkk-1 showed a rapid and sustained 17 fold decrease in patients with PR or better in a gc independent matter. Treatment of hMSC in vitro with bz alone resulted in a significant increase of both osteopontin and osteocalcin. The effect of single agent bz was diminished but not completely abrogated by the addition of gc in vitro.

In conclusion, we show that BAP and PINP increase after one treatment cycle with bz and diminishes after the addition of gc. Our in vitro data support that the decrease may be caused by the addition of gc. Secondly, our data indicate that only patients responding to treatment show an increase in BAP and PINP, which may indicate that the direct effect on OB is masked by myeloma cell inhibition in non-responders. Thirdly, PINP in general shows the same long term pattern as BAP but may give additional information about the short term inhibitory effect of gc.