Summary The incidence, characteristics and risk factors for the development of osteonecrosis of the jaw (ONJ) were evaluated among 303 myeloma patients. Only patients who received bisphosphonates developed ONJ (28/254; 11%). Zoledronic acid produced 9·5‐fold greater risk for developing ONJ than pamidronate alone (P = 0·042) and 4·5‐fold greater risk than subsequent use of pamidronate + zoledronic acid (P = 0·018). Use of thalidomide and number of bisphosphonate infusions also increased the risk for ONJ by 2·4‐fold (P = 0·043), and 4·9‐fold respectively (P = 0·012). ONJ developed earlier among patients receiving zoledronic acid. Our data indicates that administration of zoledronic acid for more than 2 years or in combination with thalidomide requires caution in myeloma.

Bortezomib reduces serum dickkopf-1 and receptor activator of nuclear factor-jB ligand concentrations and normalises indices of bone remodelling in patients with relapsed multiple myeloma Bortezomib is a proteasome inhibitor that is currently indicated for patients with multiple myeloma (MM) who have received at least one prior therapy. Although the anti-myeloma effect of bortezomib has been clearly demonstrated, its effect on bone metabolism is still unclear. In addition to increased bone resorption, bone formation is suppressed in patients with myeloma (Silvestris et al, 2004). There are recent reports that bortezomib may increase serum alkaline phosphatase activity, which is consistent with enhanced osteoblast function (Shimazaki et al, 2005;Zangari et al, 2005). However, knowledge of the molecular mechanisms that are involved and the effect bortezomib also has on bone resorption is still lacking. Dickkopf-1 (DKK-1) protein is an inhibitor of Wnt signalling that participates in osteoblast dysfunction in MM (Tian et al, 2003). The receptor activator of nuclear factor-kappa B ligand (RANKL), RANK and osteoprotegerin (OPG) system is considered the final mediator of osteoclastogenesis as well as osteoclast activation in myeloma [Terpos et al, 2003a]. Therefore, this study aimed to evaluate the effect bortezomib has on RANKL and DKK-1 serum levels in MM patients and determine if this correlates to markers of bone metabolism. Patients and methods PatientsThirty-four myeloma patients who had relapsed after previous therapies were evaluated ( SummaryThe effect of bortezomib on bone remodelling was evaluated in 34 relapsed myeloma patients. At baseline, patients had increased serum concentrations of dickkopf-1 (DKK-1), soluble receptor activator of nuclear factor-jB ligand (sRANKL), sRANKL/osteoprotegerin ratio, C-telopeptide of type-I collagen (CTX) and tartrate-resistant acid phosphatase isoform-5b (TRACP-5b); bone-alkaline phosphatase and osteocalcin were reduced. Serum DKK-1 correlated with CTX and severe bone disease. Bortezomib administration significantly reduced serum DKK-1, sRANKL, CTX, and TRACP-5b after four cycles, and dramatically increased bone-alkaline phosphatase and osteocalcin, irrespective of treatment response. This is the first study showing that bortezomib reduces DKK-1 and RANKL serum levels, leading to the normalisation of bone remodelling in relapsed myeloma.

ONJ healed in 75% of patients. Patients with spontaneous ONJ have a higher risk of nonhealing and recurrence. Reinitiating bisphosphonates after healing of ONJ is a reasonable option in patients experiencing relapse who are at risk of skeletal complications. Further studies of the pathogenesis and healing of ONJ are needed.

Plasma cell leukemia (PCL) is a rare and aggressive plasma cell disorder, with poor outcome. Bortezomibbased regimens (BBR) are highly effective in myeloma, but there is limited information about their efficacy and safety in PCL. Thus, we retrospectively collected data from 42 consecutive PCL patients (25 with primary PCL-pPCL and 17 with secondary PCL-sPCL) to explore the role of BBR in this entity. BBR were administered in 29 of 42 patients, while 6 of 25 patients with pPCL underwent autologous transplantation. Objective response (partial response) was significantly higher in patients treated with BBR versus conventional therapies (69% vs. 30.8%, P 5 0.04); 27.5% of patients treated with BBR achieved at least very good partial response (vgPR). The highest ORR was observed in pPCL patients treated with BBR (88.9%; vgPR: 33.3%). In BBR-group, grade 3 of 4 hematological, neurological and renal toxicity and neutropenic infections were observed in 41.4%, 7%, 3.4%, and 31%, respectively. With a median follow-up of 51 months, median overall survival (OS) for patients treated with BBR versus conventional therapies was 13 versus 2 months (P < 0.007). Median OS of patients with pPCL and sPCL treated with BBR was 18 and 7 months, respectively (P < 0.001). In the multivariate analysis normal PLTs, treatment with BBR and high quality response were the only powerful predictors for survival. Our study carrying the longest reported median follow-up, demonstrated that treatment of PCL with BBR induces high response rates and prolongs survival over conventional therapies, regardless of additional autologous transplantation rescue or established high risk features, with manageable toxicity.

No abstract