- Top of page
- Methods and materials
Arsenic trioxide (ATO) induces apoptosis of malignant plasma cells through multiple mechanisms, including inhibition of DNA binding by nuclear factor kappa-B, a key player in the development of chemoresistance in multiple myeloma (MM). This activity suggests that ATO may be synergistic when combined with other active antimyeloma drugs. To evaluate this, we examined the antimyeloma effects of ATO alone and in combination with bortezomib, melphalan and ascorbic acid (AA) both in vitro and in vivo using a severe combined immunodeficient (SCID)-hu murine myeloma model. Marked synergistic antimyeloma effects were demonstrated when human MM Los Angeles xenograft IgG lambda light chain (LAGλ-1) cells were treated in vitro with ATO and any one of these agents. SCID mice bearing human MM LAGλ-1 tumours were treated with single-agent ATO, bortezomib, melphalan, or AA, or combinations of ATO with either bortezomib or melphalan and AA. Animals treated with any of these drugs alone showed tumour growth and increases in paraprotein levels similar to control mice, whereas animals treated with ATO-containing combinations showed markedly suppressed tumour growth and significantly reduced serum paraprotein levels. These in vitro and in vivo results suggest that addition of ATO to other antimyeloma agents may result in improved outcomes for patients with relapsed or refractory MM.
Multiple myeloma (MM) is a malignancy characterised by the proliferation of a single clone of plasma cells derived from B cells in the bone marrow (Berenson, 2004). The American Cancer Society estimated that approximately 16 570 new cases of myeloma were diagnosed in 2006 in the United States (American Cancer Society, 2006). This disease is associated with serum monoclonal protein or immunoglobulin, development of osteolytic lesions, anaemia and renal failure. At present, the disease is incurable; and although survival has improved, there remains a need for new therapeutic options for patients with this B-cell malignancy.
Arsenic trioxide (ATO) (Mallinckrodt Baker, Inc., Phillipsburg, NJ, USA) is a novel anticancer agent with established activity in relapsed or refractory acute promyelocytic leukaemia (APL) (Niu et al, 1999; Soignet et al, 2001). Although ATO is currently approved for use in relapsed or refractory APL, the mechanism of action underlying the robust antiproliferative activity of ATO supports its use in other haematological malignancies, including MM. ATO has been shown to induce apoptosis of plasma cells through a number of mechanisms, including the downregulation of Bcl-2 expression (Chen et al, 1996; Akao et al, 1998; Zhang et al, 1998) and the inhibition of DNA binding by nuclear factor kappa-B (NF-κB) (Friedman et al, 2002). Established human MM cell lines and freshly isolated primary MM patient cells are sensitive to ATO in vitro at clinically relevant concentrations (Rousselot et al, 1999). Importantly, ATO shows anti-MM activity in the clinical setting. In two separate clinical studies, single-agent ATO, when administered daily, produced clinical responses in 21% and 33% of patients with relapsed or refractory MM (Munshi et al, 2002; Hussein et al, 2004).
Although more widely known for its antioxidant properties, ascorbic acid (AA) or vitamin C (MAC), also possesses pro-oxidant properties. In the plasma, AA is oxidised to dehydroascorbic acid before being transported into the cell, where AA is regenerated through a reaction that converts intracellular free glutathione (GSH) to GSH disulfide (May et al, 2001). This reaction depletes intracellular GSH, the molecule that eliminates reactive oxygen species (ROS), thereby increasing hydrogen peroxide production and sensitising MM cells to chemotherapeutic agents (Grad et al, 2001). As the antitumour activity of ATO is in part dependent upon the generation of ROS, removing GSH with AA enhances the antimyeloma effects of ATO (Grad et al, 2001). Elevated intracellular levels of GSH and GSH-related enzymes in MM cells have been shown to confer drug resistance to alkylating agents such as melphalan (Gupta et al, 1989; Bellamy et al, 1991; Grad et al, 2001). Thus, the combination of melphalan with ATO and MAC may overcome resistance to melphalan in MM cells, which provides the rationale for evaluating MAC in this B-cell malignancy.
In addition to inhibiting DNA binding by NF-κB, ATO has been shown to inhibit NF-κB activation by blocking the degradation of the NF-κB inhibitor IκBα (Mathieu & Besançon, 2006). Because elevated NF-κB activity has been implicated as a key factor involved in resistance of myeloma cells to chemotherapy (Feinman et al, 1999), these mechanisms of ATO suggest that this agent may overcome chemoresistant cancer cells by reducing NF-κB activation (Kapahi et al, 2000).
Bortezomib, a proteasome inhibitor approved for the treatment of patients with relapsed/refractory multiple myeloma, also inhibits the activation of NF-κB, leading to the accumulation of the inhibitor complex IκB. Bortezomib has been shown to overcome chemoresistance in MM cells both in the laboratory (Ma et al, 2003; Hideshima et al, 2005) and in the clinic (Berenson et al, 2006). The mechanism by which bortezomib prevents degradation of IκB differs from that of ATO. Bortezomib inhibits the proteasome, resulting in the accumulation of IκB, whereas ATO prevents phosphorylation of the IκB-complex, blocking its ubiquitinylation and subsequent proteasomal degradation. Thus, these different mechanisms of IκB inhibition provide the rationale for combining these agents to most effectively accomplish the deactivation of NF-κB. Furthermore, other differing mechanisms of action between bortezomib and ATO support the combination of these agents to maximise cytotoxicity against MM cells. Both ATO and bortezomib have been shown to induce their cytotoxic effects on MM cells via caspase-dependent pathways (McCafferty-Grad et al, 2003), including caspase-3, caspase-8 and caspase-9 signaling (Hayashi et al, 2002; Hideshima et al, 2003), and inhibiting activation of signal transducer and activator of transcription 3 (STAT3)- and interleukin-6-induced Janus kinase-STAT signaling; and bortezomib inhibits DNA repair machinery and suppresses adhesion of MM cells to bone marrow stromal cells (Hideshima et al, 2001; Mitsiades et al, 2003).
Our animal model of human MM, LAGλ-1, was initially generated from the intramuscular implantation of a fresh bone marrow biopsy from a MM patient (Campbell et al, 2006) and has undergone more than 25 passages during the past 3 years. This tumour retains the morphological and immunological features of the patient's original myeloma cells, including the expression of CD38, CD138 and hIgGλ (Campbell et al, 2006). Furthermore, LAGλ-1 tumour cells implanted in mice grow locally and do not metastasise to the bone marrow. These xenografts also show consistent growth as indicated by increases in both the human paraprotein and the size of the intramuscular tumour. The LAGλ-1 tumour provides an excellent preclinical model to rapidly assess and optimise the development of new therapies to more effectively treat and expand the treatment options for patients with MM. Additionally, because LAGλ-1 maintains the resistance to melphalan that was observed clinically in the patient from which it was derived (Campbell et al, 2006), it is an important tool for preclinical studies to help elucidate the mechanisms that lead to chemoresistance and evaluate novel treatment combinations to overcome drug resistance.
Here, we report our findings on the in vitro and in vivo effects of combining ATO with bortezomib, melphalan and AA on the growth of MM cells. These data provide the rationale for combining ATO with these agents in the treatment of MM patients who have disease that is resistant to these same drugs alone or when used in other combinations.
- Top of page
- Methods and materials
During the course of their disease, MM patients’ tumour cells inevitably develop a chemoresistant phenotype when treated with chemotherapeutic agents. This is important to consider when developing new therapies and regimens that must overcome the chemoresistant threshold to effectively improve overall treatment. Here, the results from these studies involving human myeloma cells in vitro and in vivo in our SCID-hu animal model of myeloma, LAGλ-1, provide the rationale for combining ATO with several currently active antimyeloma drugs for patients with this incurable B-cell malignancy.
Multiple myeloma cells cultured in vitro in the presence of clinically achievable concentrations of ATO (0·01–3 μmol/l) were found to show significant inhibition of cell growth, which suggests a favourable therapeutic index for the use of this drug in myeloma patients. These in vitro results prompted the evaluation of ATO in vivo in our SCID-hu animal model of myeloma, LAGλ-1. SCID mice implanted intramuscularly with this tumour showed marked anti-MM effects at the highest dose administered (6·0 mg/kg) daily for 5 d a week but not at several lower doses. The dose that produced the anti-MM effect was comparable with other in vivo preclinical xenograft studies using ATO (Kito et al, 2003; Rousselot et al, 2004).
ATO has previously been shown to induce apoptosis in vitro (Friedman et al, 2002; Han et al, 2005). Although RPMI8226 cells cultured with either media or AA alone showed no significant apoptotic effects in our study, cells cultured with either melphalan or ATO showed an increase in cells undergoing apoptosis. However, a significant increase in caspase-3 activation was observed in cells treated with the combination of ATO and melphalan as compared with control and/or single agent treatment.
We have previously shown that LAGλ-1 in SCID mice responds to high doses (0·5 mg/kg) of the proteasome inhibitor bortezomib (Campbell et al, 2006). Both ATO and bortezomib have been shown to reduce NF-κB activation in tumour cells (Friedman et al, 2002; Mitsiades et al, 2002). Although ATO and bortezomib share some similar potential mechanisms by which they inhibit myeloma (Miller et al, 2002; Chauhan et al, 2005), these agents differ in the way they inhibit NF-κB and functionally modulate diverse signaling pathways, such as those involving mitogen-activated protein kinase and phosphoinositide-3 kinase/Akt (Cavigelli et al, 1996; Chauhan et al, 2005), and affect Bcl-2 levels (Chen et al, 1996; Akao et al, 1998). Thus, it is possible that combinations of these two agents may be more effective than either alone for the treatment of myeloma.
Exposure of RPMI8226 and U266 cells to low concentrations of ATO and bortezomib produced significantly more growth inhibition than treatment with either of these agents alone. These results suggest that treatment with this combination is synergistic in its antimyeloma effects and that some of the anti-MM effects of these two agents are different from each other. In support of these in vitro findings, when we combined doses of ATO (1·25 mg/kg) and bortezomib (0·25 mg/kg) that showed no anti-MM effects when each was administered alone to the LAGλ-1-bearing animals, significant tumour growth inhibition and decreased serum paraprotein levels were observed. Such a low-dose combination may have the potential to limit toxicity for patients who are elderly or have other comorbidities that preclude higher-dose or other aggressive chemotherapy-based regimens. Indeed, results from a recent multicenter, open-label, phase I dose-escalation study have shown that an ATO/bortezomib/AA regimen was well tolerated and produced significant clinical benefits in a heavily pretreated study population even among patients receiving lower doses of these drugs (Berenson et al, 2007).
We investigated, whether low doses of ATO were capable of chemosensitising LAGλ-1 to melphalan. Our studies showed that the addition of ATO could overcome the melphalan resistance of LAGλ-1, as demonstrated by the nearly 50% reduction in tumour volume in the LAGλ-1-bearing mice treated with this combination compared with those animals treated with ATO, melphalan, or vehicle alone. Interestingly, a reduction in tumour volume and levels of serum paraprotein was also observed among mice treated with the combination of AA and melphalan. As noted above, this observation supports the concept that AA can enhance the antimyeloma effects of melphalan, perhaps by depleting intracellular GSH. It is important to note that the most profound anti-MM effects on LAGλ-1 growth in vivo were produced when the mice were treated with all three agents (MAC) together. Indeed, in a clinical study of the MAC combination in patients with relapsed or refractory MM, objective responses were observed in 48% of patients, including those who had previously been treated with melphalan and bortezomib (Berenson et al, 2006).
The data described here, provide the preclinical rationale for combining ATO with other available antimyeloma agents. These preclinical results have recently been translated into clinical trials and show the efficacy of some of these ATO-containing combinations for the treatment of patients with relapsed or refractory MM (Berenson et al, 2006; Berenson et al, 2007). In addition to the promising results of the phase I trial combining bortezomib and AA with ATO (Berenson et al, 2007), recent results from a large multicenter phase II study showed that nearly half of 65 heavily pretreated MM patients showed responses to the MAC combination, with a median duration of response of 12 months, and that this regimen was well tolerated (Berenson et al, 2006). Clearly, there is much to learn about combination therapy in MM as well as dosing and scheduling of multiple agents. For example, recent studies from our laboratory show that more frequent administration of lower doses of liposomal doxorubicin produces a much greater anti-MM effect on LAGλ-1 and is better tolerated by the mice than are less frequent higher doses of this drug (Campbell et al, 2006). Future efforts will be directed toward optimising the dose and schedule of these combination treatments for patients with MM. In addition, other agents known to be active in MM patients will also be evaluated with ATO, including anthracyclines, cyclophosphamide, thalidomide and lenalidomide. These studies should rapidly provide support for clinical trials evaluating the most promising ATO-containing regimens from the results of these preclinical studies. Ultimately, these efforts should help to increase the number of therapeutic options for patients with this incurable B-cell malignancy.