Fax: (713) 745-1046
A phase II trial of arsenic trioxide in patients with metastatic melanoma
Article first published online: 29 AUG 2005
Copyright © 2005 American Cancer Society
Volume 104, Issue 8, pages 1687–1692, 15 October 2005
How to Cite
Kim, K. B., Bedikian, A. Y., Camacho, L. H., Papadopoulos, N. E. and McCullough, C. (2005), A phase II trial of arsenic trioxide in patients with metastatic melanoma. Cancer, 104: 1687–1692. doi: 10.1002/cncr.21386
- Issue published online: 3 OCT 2005
- Article first published online: 29 AUG 2005
- Manuscript Accepted: 23 MAY 2005
- Manuscript Revised: 29 APR 2005
- Manuscript Received: 30 NOV 2004
- Cell Therapeutics, Inc
- Phase II;
- arsenic trioxide;
Arsenic trioxide induces growth inhibition and apoptosis in human melanoma cell lines. Therefore, a Phase II trial was conducted to evaluate the efficacy and toxicity of single-agent arsenic trioxide in patients with Stage IV melanoma.
Twenty patients, 10 with metastatic melanoma of cutaneous origin and 10 with metastatic melanoma of choroidal origin, received arsenic trioxide 0.25 mg/kg/day for 5 days, followed by a maintenance dose of 0.35 mg/kg/day twice a week. All patients with melanoma of cutaneous origin and four patients with melanoma of choroidal origin had received prior therapy.
Single-agent arsenic trioxide did not induce clinical response in this patient population. Eight patients (five with melanoma of cutaneous origin, and three with melanoma of choroidal origin) had disease stabilization for at least six weeks. The median overall survival duration for patients with melanoma of cutaneous origin was 7.9 months, and that of patients with melanoma of choroidal origin has not been reached at a median follow-up duration of 11.8 months. Grade 3 toxicity included neutropenia, fatigue, abdominal pain, and arthralgia. Grade 4 toxicity did not occur.
Single-agent arsenic trioxide was generally well tolerated; however, no tumor regression was observed in this patient population. Future clinical trials should evaluate arsenic trioxide in combination with other anticancer drugs that may improve its clinical activity in melanoma. Cancer 2005. © 2005 American Cancer Society.
Stage IV metastatic melanoma is associated with poor prognosis; 5- and 10-year patient survival rates are 7–19% and 2–16%, respectively.1 Despite the development of new modalities of therapy, such as immunotherapy and biochemotherapy, the outcome for patients with advanced disease remains poor. Current standard treatment includes dacarbazine as a single agent or in combination with multiple chemotherapeutic agents. Combination therapies such as dacarbazine, cisplatin, carmustine, and tamoxifen (the Dartmouth regimen), bleomycin, vincristine, lomustine, and dacarbazine (BOLD), and cisplatin, vinblastine, and dacarbazine (CVD) improve clinical responses but do not improve median survival duration.2, 3 In addition, combination therapy is associated with more toxicity.
The management of advanced melanoma of choroidal origin remains most problematic; most systemic treatment regimens have overall response rates below 10%. Although the combination of dacarbazine, carmustine, vincristine, and bleomycin with or without interferon-alfa showed a response rate > 15%,4, 5 a recent multicenter trial could not confirm these results.6 Therefore, a new effective drug for patients with advanced melanoma is greatly desired.
Arsenic trioxide (Trisenox, Cell Therapeutics, Seattle, WA) was recently approved by the U.S. Food and Drug Administration for induction of remission and consolidation in patients with relapsed or refractory acute promyelocytic leukemia (APL). Preclinical studies have shown that arsenic trioxide induces apoptosis in cells from a variety of different tumor types, and Phase I/II studies have shown clinical activity of arsenic trioxide in multiple myeloma and myelodysplastic syndrome.7, 8 Several pathways have been identified through which arsenic trioxide induces apoptosis, including 1) specific inhibition of antiapoptotic pathways, and 2) mitochondrial membrane depolarization and activation of downstream apoptotic pathways by indirect targeting of mitochondria through generation of reactive oxygen species or direct targeting of the mitochondria.9, 10 In vitro exposure of six human melanoma cell lines to arsenic trioxide resulted in inhibition of proliferation and growth associated with induction of apoptosis in five of these cell lines.11 Furthermore, sodium arsenite induced apoptosis in melanoma cells by way of the tumor necrosis factor (TNF)-α–mediated pathway.12 Melanoma cells are frequently constitutively activated for NF-κB, a transcription factor known to inhibit apoptosis.13 Arsenic trioxide inhibits NF- κB activity through its interaction with IκB, a NF- κB inhibitor that regulates translocation of NF- κB to the nucleus.14, 15 Furthermore, because melanoma cells operate under higher levels of endogenous oxidative stress than melanocytes, these cells may be sensitive to exogenous oxidative stress, thereby providing a selective target for arsenic trioxide.16, 17
In the Phase I dose escalation study of arsenic trioxide for solid tumor, the most common dose-limiting toxicities were fluid retention, dyspnea, and fatigue.18 These toxicities were more common at dose levels of 0.3 and 0.35 mg/kg/dose administered for 5 days every 4 weeks. The pharmacokinetic study suggested that a loading period followed by a maintenance period with less frequent dosing, considering the long half-life of arsenic trioxide, may provide adequate exposure to arsenic trioxide. The recommended arsenic trioxide induction dose was 0.25–0.3 mg/kg/day for 5 days every 4 weeks. On the basis of in vitro findings and results of Phase I and pharmacokinetic studies, we conducted a Phase II trial to test the efficacy and toxicity of arsenic trioxide in patients with metastatic melanoma. For our Phase II study, a loading dose of 0.25 mg/kg/day for 5 days, followed by a maintenance dose of 0.35 mg/kg twice weekly, was used. This dosing schedule was chosen because it was expected to produce peak plasma concentrations similar to those observed in pediatric APL patients who achieved a response with arsenic trioxide.19 This dosing schedule was expected to minimize toxicities seen with higher doses of arsenic trioxide.
MATERIALS AND METHODS
Patients with histologically confirmed Stage IV metastatic cutaneous or choroidal melanoma were enrolled in the current study. To participate in the current study, patients with metastatic melanoma of cutaneous origin must have had disease progression after at least one prior chemotherapy (excluding biologic and hormonal drugs). Patients with metastatic melanoma of choroidal origin may have had either no or up to three prior systemic treatments. Patients must have unidimensionally measurable disease defined by the Response Evaluation Criteria in Solid Tumors (RECIST) Committee.20 Patients in whom the brain was the only site of metastasis or with symptomatic central nervous system (CNS) involvement were excluded. However, patients with small asymptomatic brain metastases were eligible provided they did not require immediate treatment. Preexisting Grade 2 or greater neurotoxicity and/or neuropathy, absolute QTc interval of greater than 460 milliseconds in the presence of potassium ≥ 4.0 mEq/L, and magnesium 1.8 mg/dL were among exclusion criteria. The protocol for the current study was approved by the M. D. Anderson Cancer Center Institutional Review Board, and all patients gave written informed consent to participate before enrollment, in accordance with institutional and government guidelines.
Arsenic trioxide (Trisenox) was supplied by Cell Therapeutics Inc. (Seattle, WA). Arsenic trioxide was administered at 0.25 mg/kg/day for 5 days during the first week as an induction regimen, followed by a maintenance dose of 0.35 mg/kg/day twice a week, starting the second week. The drug was administered intravenously over 1–2 hours. One course was defined as 6 weeks of treatment. If patients were given a rest period of 2 or more weeks, they had to undergo reloading with arsenic trioxide (0.25 mg/kg/day for 5 consecutive days).
Toxicities were evaluated and recorded according to the National Cancer Institute's Common Toxicity Criteria (CTC), version 2.0, for adverse event reporting. Patients were required to keep a daily diary of adverse events and change in concomitant medications. All patients had blood tests on Day 5 and then every following week to monitor laboratory abnormality. An electrocardiogram was administered within 1 week before treatment, on Days 5 and 15, and then every 2 weeks to monitor QTc prolongation.
Clinical responses were evaluated in this study with new international criteria proposed by the Response Evaluation Criteria in Solid Tumors (RECIST) Committee. Responses were evaluated after 6 weeks (1 course) of treatment and after every treatment course based on guidelines set by the RECIST Committee. The time to progression (TTP) and overall survival duration were measured from start of treatment and recorded for each patient.
The statistical design of the current study was based on a modified two-stage Simon design.21 Response data from patients with melanoma of cutaneous origin and from those with melanoma of choroidal origin were analyzed separately. A response rate of 20% or greater was targeted, with a 5% rejection error. The design called for initial treatment of 10 patients in each group. If no responses were seen among the first 10 patients, the trial would be terminated for that group. If at least 1 response were to occur among the first 10 patients treated, the study would continue for that group to an enrollment total of 29 subjects.
Between November 2002 and March 2004, 20 patients were enrolled in the current study; 10 patients were diagnosed with melanoma of cutaneous origin and the remaining 10 with melanoma of choroidal origin. The pretreatment characteristics of these patients are listed in Table 1. All 20 patients were evaluable for response and toxicity. All patients with melanoma of cutaneous origin had prior systemic therapy (median, 1.5; range, 1–2); 9 patients received prior dacarbazine- or temozolomide-based therapy, and 7 received prior interleukin-2–based combination therapy. Four of the patients with melanoma of choroidal origin had prior therapy; one patient with BOLD plus interferon-alfa, and three with cisplatin-based hepatic arterial chemoembolization only.
|Characteristic||No. of patients (%)|
|Total patients enrolled in the study||20|
|Evaluable patients||20 (100)|
|ECOG performance status|
|Origin of primary tumor|
|Serum lactate dehydrogenase level|
|Higher than upper normal limit||7 (35)|
|Prior systemic treatment|
|Patients with melanoma of cutaneous origin||10 (100)|
|Biologic therapy||1 (10)|
|Adjuvant interferon therapy||4 (40)|
|Patients with melanoma of choroidal origin||10 (100)|
|Hepatic artery chemoembolization||3 (30)|
|Site of metastases|
|Patients with melanoma of cutaneous origin||10 (100)|
|Lymph node||8 (80)|
|Other viscera||2 (20)|
|Patients with melanoma of choroidal origin||10 (100)|
|Soft tissue (maxillary sinus)||1 (10)|
The 20 evaluable patients received a total of 38 courses of arsenic trioxide, with a median of 1 course per patient (range, 1–5). Twelve patients completed only 1 course; 4 patients received 2 courses; 2 patients received 3 courses; 1 patient received 4 courses; and 1 patient received 5 courses. Disease progression was the reason for treatment discontinuation in all patients.
Table 2 lists adverse events. Arsenic trioxide was well tolerated in general. No Grade 4 toxicity was observed. Only two patients developed Grade 3 hematologic toxicity (neutropenia), and thrombocytopenia was minimal (3 patients with Grade 1 toxicity). Nonhematologic Grade 3 toxicities included fatigue in two patients, and abdominal pain and arthralgia in one patient each. Grade 2 toxicity occurring in ≥ 20% of patients was limited to neutropenia and headache. Other common toxicities included skin rash, nausea and/or vomiting, constipation or diarrhea, cough, insomnia, tachycardia, dyspnea, fever and/or sweating, sore throat, and edema. All these adverse events were reversed upon discontinuation of treatment. Grade 2 QTc prolongation was detected in one patient, and seven patients had Grade 1 QTc prolongation at least one time during the treatment course. Asymptomatic sinus tachycardia occurred in 11 (55%) patients. One patient with known coronary artery disease had atrial fibrillation that required medical cardioversion. One patient developed asymptomatic, bilateral, patchy, pulmonary infiltrates after 5 courses of therapy, necessitating interruption of therapy.
|Toxicity (n = 20)||Grade 1/2 No. (%)||Grade 3 No. (%)|
|Neutropenia||9 (45)||2 (10)|
|Fatigue||13 (65)||2 (10)|
|Abdominal pain||13 (65)||1 (5)|
|Arthralgia||6 (30)||1 (5)|
|Skin rash||13 (65)||—|
|Fever/night sweat||12 (60)||—|
|Sore throat||9 (45)||—|
|QTc prolongation||8 (40)||—|
|Peripheral neuropathy||7 (35)||—|
|Nasal congestion||4 (20)||—|
Although no patients had a clinical response, eight (40%) patients had disease stabilization for at least 6 weeks (median TTP among these 8 patients, 4.1 mos). With a median follow-up duration of 13.8 months, 4 patients remained alive for > 7.3months to > 20.2 months. The median TTP among all 20 patients was 1.5 months (range, 1.2–7.6 mos), and the median overall survival duration had not been reached at a median follow-up duration of 8.6 months (range, 2.1 mos to > 20.2 mos). Progression-free survival and overall survival for all patients are shown in Figures 1 and 2, respectively.
Among 10 patients with melanoma of cutaneous origin, 5 had disease stabilization. The TTP for these 5 patients was 2.8, 4.1, 5.6, 7.2, and 7.6 months, respectively. Among patients with melanoma of choroidal origin, three had disease stabilization. The TTP for these patients was 2.8, 3.3, and 4.1 months, respectively, and the median overall survival had not been reached with a median follow-up time of 11.8 months. There was no clear relation between the prognostic factors, (such as gender or serum lactate dehydrogenase levels) and clinical benefit (such as TTP or overall survival duration).
This Phase II trial was conducted to determine the clinical efficacy of arsenic trioxide in patients with advanced melanoma based on in vitro studies demonstrating arsenic trioxide-induced inhibition of cell growth and the induction of apoptosis in melanoma cell lines. No clinical response was observed among 20 patients with advanced melanoma. Thus, single-agent arsenic trioxide was not efficacious in this patient population. The discrepancy between the in vitro activity and the clinical efficacy of an anticancer drug is a recurring dilemma in new drug development for patients with advanced melanoma, and this situation calls for more thorough approaches to preclinical drug testing, including examination of relevant molecular mechanisms of oncogenesis, tumor–stromal interaction, and use of more realistic model systems.
One of the likely explanations of clinical inefficacy of arsenic trioxide in patients with advanced melanoma is that metastatic melanoma cells have several activated antiapoptotic pathways in addition to NF-κB as mentioned above. For example, loss of PTEN is associated with aberrant cell growth and escape from apoptosis.22 Mutational activation of BRAF results in constitutive activation of the MAP kinase signaling pathway and subsequent growth and invasion of melanomas.23 Alteration of these signaling pathways may protect cells against arsenite-induced oxidative stress and apoptosis in melanoma cells. Specifically, suppression of the PI-3-kinase–AKT, MAP kinase, and JNK pathways was shown to dramatically accelerate arsenite-induced apoptosis.12 Furthermore, arsenic trioxide-induced apoptosis is modulated by the N-(N-L-gamma-glutamyl-L-cysteinyl)glycine (GSH)-dependent redox system: malignant cell lines that are very sensitive to arsenic trioxide have consistently low levels of GSH, whereas arsenic trioxide-resistance is associated with increased GSH levels.24 Agents that down-regulate intracellular levels of GSH, such as L-buthionine-(S,R)-sulfoximine (BSO) and ascorbic acid, potentiate the apoptotic effect of arsenic trioxide in vitro.25, 26
Therefore, arsenic trioxide in combination with drugs that inhibit other prominent antiapoptotic pathways may be a promising strategy to treat patients with advanced melanoma.
In addition, it is possible that the dosing schedule in this study was not optimal to induce clinical responses in patients with advanced melanoma. Although we determined the dosing schedule based on the previous Phase I and pharmacokinetic results to achieve a clinically active serum arsenic trioxide concentration as in the pediatric APL population, we did not perform a pharmacokinetic study in this trial. Considering that toxicities were modest with the arsenic trioxide dose used in this trial, a clinical trial testing a new dosing schedule of a higher arsenic trioxide dose in this patient population can be entertained.
Arsenic trioxide was generally well tolerated in this trial, with Grade 3 adverse events being only neutropenia, fatigue, abdominal pain, and arthralgia, each of which occurred in ≤ 10% of patients. We did not observe any clinically significant QTc interval prolongation or sinus tachycardia. Therefore, arsenic trioxide is likely to be safe in combination with other anticancer drugs that may improve its clinical activity in melanoma. Recent preclinical studies have demonstrated a significant synergistic effect of arsenic trioxide and ascorbic acid on induction of apoptosis in several leukemia, lymphoma, and myeloma cell lines.24, 27, 28 To test clinical activity of this combination, we are currently conducting a Phase II study in patients with metastatic melanoma.
- 8Trials of arsenic trioxide in multiple myeloma. Cancer Control. 2003; 10: 370–374..
- 11Arsenic trioxide induces apoptosis of human melanoma cell lines in-vitro [abstract]. Proc Am Soc Clin Oncol. 2001; 20: 360a., :
- 18Clinical and pharmacologic study of arsenic trioxide (As2O3) in patients with solid tumors [abstract]. Proc Am Soc Clin Oncol. 2000; 19: 201a., , , et al.:
- 19Phase 1 trial and pharmacokinetic study of arsenic trioxide in children with refractory leukemia [abstract]. Proc Am Soc Clin Oncol. 2002; 21: 396a., , , et al.
- 20New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst. 2000; 92: 205–216., , , et al.