(2004) 18, 1518–1521 & 2004 Nature Publishing Group All rights reserved 0887-6924/04 $30.00 www.nature.com/leu A clinical and pharmacological study of in advanced multiple myeloma patients

P Rousselot1, J Larghero1, B Arnulf1, J Poupon2, B Royer1, A Tibi3, I Madelaine-Chambrin1, P Cimerman4, S Chevret5, O Hermine6, H Dombret1, J Claude Brouet1 and J Paul Fermand1

1Department of Immunology, Hematology, Cell Therapy and Institute of Hematology, Hoˆpital Saint-Louis, Paris, France; 2Department of Biochemistry-Toxicology, Hoˆpital Fernand Widal, Paris, France; 3Etablissement Pharmaceutique des Hoˆpitaux de Paris, Paris, France; 4De´le´gation a` la Recherche Clinique, Assistance Publique Hoˆpitaux de Paris, Paris, France; 5DBIM, Hoˆpital Saint-Louis, Paris, France; and 6Department of Hematology, Hoˆpital Necker, Paris, France

We previously showed that arsenic trioxide (ATO) and melar- months after ATO discontinuation.10 After these encouraging soprol may inhibit the growth of multiple myeloma (MM) cells in results, we decided to evaluate the efficacy and the safety of vitro and in vivo. We report here the administration of arsenic derivatives in 12 relapsing or refractory secretory MM patients. ATO administration in patients with refractory MM. A total of 10 patients received ATO (eight in a continuous schedule, two discontinuously) and two received melarsoprol. The melarsoprol arm was prematurely closed due to toxicity. In the ATO arm, median duration of treatment was 38 days (9–54). Patients and methods Hepatic toxicity was grade 3 and 2 in one and eight patients, respectively. Other toxicities included neuropathy (n ¼ 2, grade Patients 2), encephalitis (n ¼ 1, grade 3) and leuconeutropenia (n ¼ 4, grade 3). At 2 weeks after treatment initiation, mean serum concentration of arsenic was 1.1170.16 lmol/l. No complete or Patients aged 18 years or more with secretory MM in failure after partial remission was observed. A minor response (25–49% at least two lines of treatment were eligible. Patients may have reduction of M protein in serum) and a stabilization of the M- been previously treated by conventional (using protein level were observed in three and four patients, repeated courses of akylating agents, and/or respectively. After ATO discontinuation, these responses were corticosteroids) and by an intensive strategy including a high- of short duration in all cases. ATO as a single agent did not produce any significant response in advanced MM patients dose therapeutic regimen supported with autologous bone despite sufficient arsenic exposure. Strategies to improve marrow transplantation. The study was approved by the Ethics biodistribution, pharmacokinetic and efficacy of the drug as Committee of Hoˆpital Pitie´-Salpe´trie`re (Paris, France) and well as treatment combinations are needed. Ministe`re de l’Emploi et de la Solidarite´. All patients gave Leukemia (2004) 18, 1518–1521. doi:10.1038/sj.leu.2403424 signed informed consent. Published online 22 July 2004 Keywords: arsenic trioxide; myeloma; pharmacokinetic; clinical trial

Treatments

Introduction Study design consisted of allocating 10 patients to continuous ATO and 10 patients to melarsoprol. In order to avoid patients’ In 1992, arsenic trioxide (As2O3, ATO) was reported by Chinese selection, we decided to randomize treatment arm. ATO was investigators to have a substantial activity in acute promyelo- manufactured by the Pharmacie Centrale des Hoˆpitaux de Paris cytic leukaemia (APL).1 This observation, that was confirmed by 2–4 (Paris, France). The formulation process was validated as various investigators, prompted screening of ATO as a new previously described.4 ATO was administrated at the dosage biologic agent in various haematological malignancies. of 0.15 mg/kg/day by a 3-h intravenous infusion. Melarsoprol Despite the use of intensive chemotherapeutic regimens, was obtained from manufacturer as Arsobals (Aventis, Paris, multiple myeloma (MM) remains an incurable disease with a France) and administered in three intravenous perfusions at the median survival still limited to 3 years. Patients usually respond dosage of 1 mg/kg at day 1 and 2.2 mg/kg/day thereafter. To to treatment but experience recurrences featured by the 5 prevent potential arsenic-related neurotoxicity, all patients emergence of resistant plasma cells. received vitamin B1 (250 mg/day) and clobazam (10–30 mg/ We and others have previously investigated the biological 6–9 day) during treatment. ATO and melarsoprol were maintained activity of ATO in MM. We found that arsenic derivatives for a maximum of 56 days and stopped in case of severe toxicity such as melarsoprol and ATO induced apoptosis in myeloma (OMS grade 2–4, depending of the organ concerned) or if the À5 cell lines and in plasma cells derived from MM patients in vitro. arsenic serum concentration was 10 M or greater. For the last In an attempt to further document the interest of the arsenicals in two patients in the ATO group, the trial was amended and vivo, we treated SCID mice transplanted with human myeloma patients received ATO during 14 days, one cycle every month, cells by melarsoprol or ATO and observed significant responses, using an increased daily dosage (0.2 mg/kg/day) as compared to including apparent complete remission persisting up to 5 the continuous ATO regimen. The melarsoprol arm of treatment was stopped because of the occurrence of serious, grade III–IV Correspondence: Dr J Paul Fermand, Hoˆpital Saint-Louis, Service adverse events in the two first treated patients. Subsequently, all ´ d’Immunologie et d’Hematologie, 1 avenue Claude Vellefaux, 75475 included patients received ATO. Paris cedex 10, France; Fax: þ 33 1 42 49 96 76; E-mail: [email protected] The study was initiated in September 1998 and terminated in Received 11 February 2004; accepted 25 May 2004; Published online May 1999 after the inclusion of 10 patients in the ATO arm, as 22 July 2004 initially planned. Arsenic trioxide for multiple myeloma P Rousselot et al 1519 Response criteria and renal function tests were assessed at least twice a week and an ECG was obtained weekly. Adverse events were graded on a Complete response (CR) was defined as no monoclonal scale of 0–4 using the World Health Organization toxicity immunoglobulins (M protein) in serum and/or 100 times criteria. concentrated urine by immunofixation analysis and less than 5% plasma cells in bone marrow aspiration when evaluated. Partial response (PR) was defined as a 50% or more reduction of Results M protein in serum and/or a 75% or more reduction of the Bence–Jones (BJ) proteinuria. Minor response (MR) was defined Patient characteristics as a 25–49% reduction M protein in serum and/or a 50–74% reduction of the BJ proteinuria. From October 1998 to April 1999, 12 patients with relapsing or Failure to treatment was defined by less than a 25% reduction refractory secretory MM were treated. Two patients received in M-protein serum level and/or by less than a 50% reduction of melarsoprol and 10 patients received ATO. Patient’s character- BJ proteinuria. Progressive disease was defined by any of the istics and previous treatments are listed in Table 1. The median following criteria: increase of 25% or more in serum M-protein time from diagnosis to inclusion was 86.2 months. level on two successive dosages, increase of 50% or more in BJ proteinuria, onset of hypercalcemia, plasmocytoma or new bone lesions. Completion of allocated treatment None of the eight patients allocated to the continuous ATO Pharmacokinetic (PK) studies regimen completed the 56-day planned treatment. The median treatment duration was 35 days (range 5–44). The treatment was Arsenic concentrations were evaluated at baseline, day 4, day 7 interrupted because of failure or progression in four patients, and then once a week until the end of the treatment. Serum toxicity in three patients and total serum arsenic concentration À5 residual total arsenic concentrations were determined just 41 Â 10 in one patient. ATO daily dose was reduced (by before the following injection. Total arsenic in serum was 75%, during 2 weeks) in only one patient who developed grade determined as previously described.4 III hepatitis. Both two patients that received the discontinuous ATO regimen completed the planned treatment. Safety evaluation The first patient treated by melarsoprol experienced grade IV neurological toxicity at day 9. In the second patient, systematic During ATO administration, physical examination was per- monitoring of serum arsenic revealed over dosage at day 3 and formed daily, complete blood count with differential, hepatic the patient also experienced neurologic toxicity (see below).

Table 1 Characteristics of patients in ATO group

Patient Age Sexe MIg Previous treatments OMS Time from PC Creatinine b2M MIg level Treatment Best (years) type Dg to ATO aspirate (mM/L) (mg/ml) (g/l) duration response (months) (%) (days) during therapy

1 62 M IgDl HDT (with TBI) and 0 107 8 187 5.5 12 44 Minor ABSCT; IFN; A.A. 2 58 M IgGk HDT (with TBI) and 0 37 25 89 2.8 42 37 Stabilization ABSCT; VAD-like; Fludarabine 3 51 M IgDl HDT (with TBI) and 0 60 41 93 3 6.2 33 Stabilization ABSCT; A.A.; Fludarabine 4 57 M IgAk A.A.; IFN; A.A.; VAD-like 1 129 85 120 NA 36 30 Progression regimen; A.A. 5 59 F IgGl HDT (with TBI) and 0 73 NA 90 2.5 41 9 Nonevaluable ABSCT; A.A. 6 50 F IgGk A.A.; IFN; HDT (with TBI) 1 91 NA 92 4.8 78 39 Minor and ABSCT; A.A. 7 54 M IgGk A.A.; HDT (with TBI) and 0 89 20 80 2.2 26 38 Minor ABSCT; VAD-like; A.A.; HDT and ABSCT; A.A. 8 51 M IgGk HDT (with TBI) and 1 104 NA 71 NA 19 5 Nonevaluable ABSCT; A.A. 9 49 F IgGk HDT (with TBI) and 0831194NA8042a Stabilization ABSCT; IFN; A.A. 10 65 F IgGk A.A.; IFN; A.A.; VAD- 0 107 12 103 NA 43 42a Stabilization like; A.A. HDT, high-dose therapy; ABSCT, autologous blood stem cell transplantation; TBI, total body irradiation; A.A., alkylating agent-based regimen; VAD- like, , , -like regimen; IFN, interferon a; MIg, monoclonal immunoglobulin; b2M, beta-2 microglobulin; Dg, diagnosis; ATO, arsenic trioxide; PC, plasma cell. aThree cycles of 14 days each.

Leukemia Arsenic trioxide for multiple myeloma P Rousselot et al 1520 Accordingly, the survey committee of the protocol advised the reduction of M protein in serum). In two of these, ATO must be interruption of the melarsoprol arm. stopped due to toxicity. The later patient developed a plasmocytoma of the clavicle while on therapy. In all patients, disease stabilization or minor responses were of 1–3 weeks PK studies duration. After ATO administration, all surviving patients were treated In the eight patients who received the continuous ATO regimen by thalidomide alone or combined with dexamethasone. (0.15 mg/kg/day), the mean residual serum total arsenic con- centration progressively increased to reach a maximum of 1.1170.16 mmol/l at day 14 (Figure 1a). In the two patients Adverse events (patient 9 and 10) treated with ATO 14 days per month (0.20 mg/kg/day), residual values of arsenic at day 1 of cycle 1, 2 As mentioned, the melarsoprol arm of the study was rapidly and 3 were 0, 0.42 and 1 mmol/l, respectively, providing interrupted because of high toxicity for central nervous system. evidence for arsenic accumulation. However, maximal residual Indeed, the first patient experienced refractory generalized serum total arsenic concentrations were not significantly grand mal seizure during the second week of therapy, and the different after completion of one cycle as compared to the second patient rapidly presented with tremor and lethargy. previous one (Figure 1b). Serum arsenic concentration profiles Treatment by ATO also produced significant toxicity. Non- did not differ between responding and nonresponding patients haematological adverse events were responsible for premature (data not shown). treatment discontinuation in three cases with encephalopathy with confusion and somnolence (grade 3, n ¼ 1) and hepatitis cytolysis (grade 3, n ¼ 2). In two cases, infections lead to a Patient outcome transient interruption of ATO therapy. Finally, other reported grade 1 and 2 adverse events were increased transaminase No CR or PR was observed in any patient. A transient levels (n ¼ 8), herpes zoster (n ¼ 3), nausea (n ¼ 2), skin rash stabilization of the M-protein level was observed in two patients (n ¼ 2) and peripheral neuropathy (n ¼ 2). in the continuous regimen and two in the discontinuous In addition to its extrahaematological toxicity, ATO also regimen. Three patients showed a minor response (25–49% produced cytopenias. Without taking into account the two patients who had grade IV at treatment initiation, four of the eight patients who received more than 15 days of ATO presented a grade 4 neutropenia leading to a transient treatment arrest. In contrast, one patient with grade 3 neutropenia at treatment initiation improved to grade 1 during therapy. Grade 3 thrombocytopenia occurred in three patients and grade 4 in one other.

Discussion

The trial was designed to treat heavily pretreated myeloma patients either with ATO or with melarsoprol. The melarsoprol arm had to be rapidly stopped due to severe encephalitic complications, confirming the previously reported direct toxicity of the drug for the central nervous system in the treatment of trypanosomiasis.11 ATO, which was initially administered continuously, according to the regimen currently used in acute promyelocytic leukaemia (0.15 mg/kg/day), was modified to a discontinuous schedule (0.20 mg/kg/day 14 days/month) be- cause of a significant haematological toxicity. Unexpectedly, increased transaminase levels were observed in all but one patient. Thus, treatment toxicity, particularly hepatic and haematological, was relatively high, leading to a poor benefit/ risk ratio. Whatever the ATO regimen used, only a minor response was noted in three out of 10 patients even in cases of prolonged administration (median cumulative duration of ATO: 37.5 days; range 5–44). Arsenic pharmacokinetic studies in our patients showed that the mean residual arsenic serum concentrations was of 1.1170.16 mmol/l after 2 weeks. These concentrations were twice higher than those observed in equivalently treated APL patients4 and relate to the more important extrahaematological adverse events and the poorer haematological tolerance. The Figure 1 Pharmacokinetic studies. (a) Mean residual total arsenic dose required in a myeloma cell environment for inducing cell concentration7s.e.m. during the first 5 weeks of continuous ATO (0.15 mg/kg/day) treatment (n ¼ 8). (b) Pharmacokinetic of ATO in death leads probably to an unacceptable toxicity. This effective patients 9 and 10 who received ATO (0.20 mg/kg/day) discontinuously concentration is still questioned. The serum concentration (14 day cycles, every month). obtained should have been satisfactory as in our previous

Leukemia Arsenic trioxide for multiple myeloma P Rousselot et al 1521 in vitro studies, a concentration of 1 mmol of ATO sufficed to 4 Raffoux E, Rousselot P, Poupon J, Daniel MT, Cassinat B, Delarue induce apoptosis of plasma cell from MM patients.6 Yet, we R et al. Combined treatment with arsenic trioxide and all-trans have also noted, in our SCID-MM model, that effective bone retinoic acid in patients with relapsed acute promyelocytic marrow concentrations may be difficult to achieve.10 leukemia. J Clin Oncol 2003; 21: 2326–2334. 5 Drach J, Kaufmann H, Urbauer E, Schreiber S, Ackermann J, Huber Two other phase II trials have recently been reported in H. The biology of multiple myeloma. J Cancer Res Clin Oncol 12 advanced-phase MM patients. In the study of Munshi et al, 14 2000; 126: 441–447. patients were treated with a similar ATO continuous protocol 6 Rousselot P, Labaume S, Marolleau JP, Larghero J, Noguera MH, (0.15 mg/kg up to 60 days). In the study of Hussein et al,13 ATO Brouet JC et al. Arsenic trioxide and melarsoprol induce apoptosis regimen consisted in i.v. infusion of 0.25 mg/kg 5-day/week for in plasma cell lines and plasma cells from myeloma patients. 2 weeks followed by no therapy for 2 weeks, in repeated 4-week Cancer Res 1999; 59: 1041–1048. 7 Park WH, Seol JG, Kim ES, Hyun JM, Jung CW, Lee CC et al. cycles. Objective and minor responses were observed in 3/14 Arsenic trioxide-mediated growth inhibition in MC/CAR myeloma and 9/21 patients in each study, respectively. In both studies, cells via arrest in association with induction of cyclin- responses were of short duration. These data did not compare dependant kinase inhibitor, p21, and apoptosis. Cancer Res 2000; favourably with the results obtained with other new drugs, such 60: 3065–3071. as and Revimid, when used as single agent in highly 8 Hayashi T, Hideshima T, Akiyama M, Richardson P, Schlossman pretreated patients.14,15 RL, Chauhan D et al. Arsenic trioxide inhibits growth of human multiple myeloma cells in the bone marrow microenvironment. Thus, combination of ATO with other drugs may be required Mol Cancer Ther 2002; 1: 851–860. to enhance ATO efficacy in MM. Candidates could be 9 Liu Q, Hilsenbeck S, Gazitt Y. Arsenic trioxide-induced apoptosis conventional cytotoxic drugs, high-dose steroids or sensitizing in myeloma cells: p53-dependant G1 G2/M cell cycle arrest, agents, such as ascorbic acid to produce a glutathione activation of caspase-8 or caspase-9, and synergy with APO2/ depletion.16 Preliminary data using such combinatorial strategy TRAIL. Blood 2003; 101: 4078–4087. showed some results. Indeed, in one study, the combination of 10 Rousselot P, Larghero J, Labaume S, Chopin M, Dosquet C, Marolleau JP et al. Arsenic trioxide is effective in the treatment of ATO at low dosage (0.25 mg/kg twice weekly), oral multiple myeloma in SCID mice. Eur J Haematol 2004; 72: and intravenously administered ascorbic acid produced four 166–171. responses in 10 treated patients.17 However, all these responses 11 Burri C, Nkunku S, Merolle A, Smith T, Blum J, Brun R. Efficacy of were of short duration. In another study, a regimen combining new, concise schedule for melarsoprol in treatment of sleeping ATO (0.25 mg/kg 5 days in week one and two times a week for sickness caused by Trypanosoma brucei gambiense: a randomised weeks 2–10), dexamethasone and ascorbic acid was used in 16 trial. Lancet 2000; 355: 1419–1425. 12 Munshi NC, Tricot G, Desikan R, Badros A, Zangari M, patients, producing response and disease stabilization in one Toor A et al. Clinical activity of arsenic trioxide for the 18 and six patients, respectively. Finally, in a phase I/II trial, treatment of multiple myeloma. Leukemia 2002; 16: Bahlis et al19 reported two PRs and four stable disease in 1835–1837. patients receiving ascorbic acid 1000 mg/day in combination 13 Hussein MA, Mason J, Saleh NM, Rifkin RM, Ravandi F. Arsenic with two different doses of ATO (0.15 or 0.25 mg/kg/day). trioxide (Trisenox) in patients with relapsed or refractory multiple In conclusion, ATO administration in MM patients is feasible myeloma (MM): final report of a phase II clinical study. Blood 2002; 100: 5138a. but has a limited efficacy and a significant toxicity. Additional 14 Richardson PG, Barlogie B, Berenson J, Singhal S, Jagannath S, research to improve the biodistribution of the drug and its Irwin D et al. A phase 2 study of bortezomib in relapsed, refractory efficacy are needed. myeloma. N Engl J Med 2003; 348: 2609–2617. 15 Richardson PG, Schlossman RL, Weller E, Hideshima T, Mitsiades C, Davies F et al. Immunomodulatory drug Acknowledgements CC-5013 overcomes drug resistance and is well tolerated in patients with relapsed multiple myeloma. Blood 2002; 100: This study was supported by Grant no. P970708 and AOM 97088 3063–3067. 16 Grad JM, Bahlis NJ, Reis I, Oshiro MM, Dalton WS, Boise LH. from Le Programme Hospitalier de Recherche Clinique, Ministe`re Ascorbic acid enhances arsenic trioxide-induced cytotoxicity in de l’Emploi et de la Solidarite´, France. multiple myeloma cells. Blood 2001; 98: 805–813. 17 Borad M, Swift RA, Sadler K, Yang H, Berenson JR. Melphalan, arsenic trioxide and ascorbic acid (MAC) is effective in the References treatment of refractory and relapsed multiple myeloma (MM). Blood 2003; 102: 827a. 1 Sun HD, Ma L, Hu XC. Ai-Lin 1 treated 32 cases of acute 18 Birch R, Schwartzberg LS, Lawrence V, Schnell FM, Tongol JM, promyelocytic leukemia. Chin J Integr Chin West Med 1992; 1: Prill SJ et al. A phase II study of arsenic trioxide (ATO) in 170–171. combination with dexamethasone (Dex) and ascorbic acid (VITC) 2 Shen ZX, Chen GQ, Ni JH, Li XS, Xiong SM, Qiu QY et al. Use of in patients with relapsed/refractory multiple myeloma. Blood arsenic trioxide (As2O3) in the treatment of acute promyelocytic 2003; 102: 5271a. leukemia (APL): II. Clinical efficacy and in 19 Bahlis NJ, McCafferty-Grad J, Jordan-McMurry I, Neil J, relapsed patients. Blood 1997; 89: 3354–3360. Reis I, Kharfan-Dabaja M et al. Feasibility and correlates 3 Soignet SL, Maslak P, Wang ZG, Jhanwar S, Calleja E, Dardashti LJ of arsenic trioxide combined with ascorbic acid-mediated et al. Complete remission after treatment of acute promyelocytic depletion of intracellular glutathione for the treatment of leukemia with arsenic trioxide. N Engl J Med 1998; 339: relapsed/refractory multiple myeloma. Clin Cancer Res 2002; 8: 1341–1348. 3658–3668.

Leukemia