- Email: [email protected]
ADO study
original article
Annals of Oncology 17: 1592–1597, 2006 doi:10.1093/annonc/mdl148
Dose-intensified bi-weekly temozolomide in patients with asymptomatic brain metastases from malignant melanoma: a phase II DeCOG/ADO study D. Schadendorf1*, A. Hauschild2, S. Ugurel1, A. Thoelke1, F. Egberts2, M. Kreissig3, R. Linse4, U. Trefzer5, T. Vogt6, W. Tilgen7, P. Mohr8 & C. Garbe3 1 Skin Cancer Unit, German Cancer Research Center & University Hospital Mannheim, Department of Dermatology; 2University Hospital, Kiel; 3University Hospital, Tu¨ bingen; 4Community Hospital, Erfurt; 5Charite Hospital, Humboldt University, Berlin; 6University Hospital, Regensburg; 7University Hospital, Homburg; 8 Elbe Hospital, Buxtehude, Germany
Received 17 March 2006; revised 9 May 2006; accepted 22 May 2006
original article
approval to treat brain tumours. The purpose of this study was to test a dose-intensified regimen of temozolomide in melanoma patients with brain metastases in a prospective, open-label, multicentre phase II trial. Patients and methods: Forty-five patients with asymptomatic brain metastases from melanoma were stratified into arm A (no prior chemotherapy; n = 21) and arm B (previous chemotherapy; n = 24). Patients received oral temozolomide either 150 mg/m2/day (arm A) or 125 mg/m2/day (arm B), days 1–7 and 15–21, every 28 days. The primary study end point was objective response, and secondary end points were overall survival and safety. Results: Two patients (4.4%) achieved a partial response (PR) in brain metastases (one in each arm), one of them (2.2%) also showing a PR in extracerebral disease. An additional five patients (11.1%; two in arm A, three in arm B) showed disease stabilisation (SD) in brain and other sites. However, 82% revealed progressive disease (PD) already evident 8 weeks after therapy initiation. Median survival time from therapy onset was 3.5 months (range 0.7–8.3; arm B) and 4.3 months (range 1.6–11.8; arm A), P = 0.43. Dose modifications and prolongations of therapy cycles due to toxicity were required in 20% of patients. Grade 3/4 toxicity was observed in one patient only (2.2%). Conclusions: Oral administration of temozolomide given bi-weekly is well-tolerated in melanoma patients with cerebral involvement. However, the efficacy is limited, with lower than 5% objective responses observed in brain and extracerebral metastases. Key words: temozolomide, brain metastases, metastatic melanoma
introduction Malignant melanoma in its advanced stage of metastatic disease has a poor prognosis with a median survival time of approximately 9 months [1]. Furthermore, melanoma is the third most common cause of metastases in the brain after carcinomas of the lung and breast [2]. Initially, up to 20% of patients with stage IV melanoma suffer from brain metastases [3]. Metastases of the CNS develop in nearly half of advanced melanoma patients [2, 4] during the progression of disease dominating prognosis of those patients. Surgical excision or radiosurgery are the first choice for patients with solitary brain metastases. In the case of multiple or inoperable CNS metastases, whole brain irradiation (WBI) is considered as the current standard of care, although only neurological symptoms *Correspondence to: Dr D. Schadendorf, Skin Cancer Unit, German Cancer Research Center & University Hospital Mannheim, Theodor Kutzer Ufer 1, D-68135 Mannheim, Germany. Tel: +49-621-383-2126; Fax: +49-621-383-2163; E-mail: [email protected]
ª 2006 European Society for Medical Oncology
are improved and no survival advantage has been shown so far. In most cases of advanced melanoma, the presence of brain metastases is associated with widespread visceral involvement. The failure of systemic drug therapy in metastatic melanoma may be related to their lacking ability to cross the blood–brain barrier. Within the group of cytotoxic drugs known to have some activity in melanoma, fotemustine, a nitrosourea analogue evaluated in several phase II open label studies, was shown to have an activity on cerebral metastases as well as other metastatic sites [5, 6]. Temozolomide is a novel alkylating, orally available agent with some activity in metastatic melanoma [7, 8]. Furthermore, it was demonstrated that temozolomide penetrates the blood–brain barrier. Low-dose continuous temozolomide over 6 weeks combined with WBI was shown to improve overall survival in patients with glioblastoma compared with WBI alone [9]. Clinical responses and improvement of neurological symptoms were shown in other cancers as well [10–12]. In retrospective analyses of patients with advanced
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
Background: Temozolomide has shown some efficacy in metastatic melanoma and recently received extended
original article
Annals of Oncology
melanoma, a reduced relapse risk in the brain after temozolomide treatment compared with dacarbazine was suggested [13–15]. Here we report the results of a first prospective, open-label, phase II trial evaluating the safety and efficacy of a doseintensified regimen of temozolomide in patients with brain metastases of malignant melanoma.
patients and methods patients
protocol design and treatment This national multicentre, open-label phase II study of temozolomide included melanoma patients who had had newly diagnosed (<6 weeks), asymptomatic cerebral metastases. Patients were stratified into two arms, those who had or had not received previous chemotherapy for metastatic melanoma. Temozolomide was given orally in the fasting state, once a day for 7 days/week, repeated every other week, at a dose of 150 mg/m2/day (previously untreated, arm A) or 125 mg/m2/day (previously treated, arm B) on days 1–7 and days 15–21 of every cycle month. Temozolomide was administered until the occurrence of either unacceptable toxicity or evidence of progression of cerebral disease for a maximum of 1 year from initial treatment. Antiemetics were administered at the discretion of the treating physician. Primary study end points were the efficacy of temozolomide at week 8 and week 20 in patients in terms of clinical response rate and toxicity in patients who had previously been untreated for the presentation of brain metastases. Both patient groups who had received prior or no prior chemotherapy for extracerebral melanoma disease were evaluated separately.
Volume 17 | No. 10 | October 2006
response and toxicity assessment All patients had routine evaluations before treatment, including complete medical history, physical examination, complete tumour imaging [brain computed tomography (CT) or Gd-magnetic resonance imaging (MRI), chest X-ray, sonography of the abdomen or CT scan], complete blood cell count and blood chemistry. At week 8 and week 20 after beginning study treatment, patients were restaged using the same imaging technique for evaluating the rate of objective responses (CR, PR), which was the primary objective of the study. Tumour response evaluation was based on RECIST criteria [16]. Responses were confirmed 4 weeks later. Intracerebral and extracerebral responses were recorded separately. Toxicity was recorded on day 1 of each cycle according to the Common Toxicity Criteria (CTC) together with a complete medical history and a physical examination. A neurological assessment and a documentation of vital signs were also done. Laboratory assessment (complete blood chemistry, blood count and serum chemistry) was performed on a weekly basis.
dose modifications Toxicity led to dose reduction based on the most severe grade of any single toxicity observed. Unacceptable toxicity resulted in drug withdrawal. Upon resolution of the toxicity, patients continued treatment at one dose level below the dose level administered. If more than a two-dose level reduction (in previously untreated patients; level I, 125 mg/m2; level II, 100 mg/m2) or more than a one-dose level reduction (in previously treated patients) were required for continuing treatment of any patient, then the patient was withdrawn from the study. Therefore, patients who required dose reductions to a dose level of <100 mg/m2/day of temozolomide were taken off the study. If acute neutrophil count (ANC) <1500 cells/ll (‡grade 2) and/or platelet count <100 000/ll (‡grade 1) occurred at any given time while on therapy, therapy was delayed until haematological recovery (ANC ‡1500 and/or platelet count ‡100 000). If restart of therapy was not possible within 21 days after last drug application, the patient went off study. Two or more weeks of therapy delay required a dose reduction by one level. ANC <500 cells/ll for 5 days and/or platelet count <25 000 delayed restarting of therapy until haematological recovery (ANC ‡1500 and/or platelet count ‡100 000). A subsequent dose reduction by one dose level was mandatory. If restart of therapy was not possible within 21 days after last drug application, the patient went off study. Febrile neutropenia (ANC <500 with fever temperature >38.3C) also led to patient withdrawal from the study. For CTC grade 2 or less non-haematological toxicity, no dose reductions were required. For grade 3 or 4 non-haematological toxicity (including GI toxicity unresponsive to standard therapy), a delay in dosing was recommended until toxicity had resolved to baseline or grade 1. If restart of therapy was not possible within 21 days after last drug application, then the patient went off study. All non-haematological CTC grade 2, 3 and 4 toxicities must have resolved to at least CTC grade 1 (or for baseline CTC grade 2 elevations in transaminases, bilirubin and/or alkaline phosphatase to the level defined by the inclusion criteria) prior to repeat dosing.
statistical analyses The primary end point was to assess the cerebral and extracerebral objective response rate (defined as complete responses, partial response and stable disease) in week 8 and week 20 in melanoma patients with asymptomatic brain metastases. Patients were accrued in this study using a two-stage Simon’s design [17] in order to improve the probability of observing the real response rate. Therefore, if there were three or more responders among the
doi:10.1093/annonc/mdl148 | 1593
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
Patients were enrolled with histological confirmed malignant melanoma and newly diagnosed (within 6 weeks prior to study treatment) brain metastases, which were irresectable and neurologically asymptomatic. Eligible patients who met the following eligibility criteria were included in the study: age ‡18 years, a life expectancy greater than 3 months, Karnofsky performance status ‡70% and adequate blood cell counts (polynuclear neutrophils ‡1500/mm3, platelets ‡100 000/mm3 and haemoglobin ‡8 g/dl). Patients with or without prior chemotherapy [except temozolomide or disease progression during previous dacarbazine (DTIC) therapy] were accepted. Patients were required to have no chemotherapy within the previous 4 weeks and no previous cerebral radiotherapy. Measurable disease according to RECIST criteria, i.e. at least one ‘target’ lesion ‡1 cm by MRI or CT of the brain, was required. No steroid medication was allowed during treatment and prior treatment had to be stopped at least 7 days before onset of therapy. Previous chemotherapy for metastatic extracerebral disease was permitted and patients were stratified according to previous treatment status. Patients with previously treated cerebral metastases with stereotactic radio- or neuro-surgery and newly developing brain metastases were allowed onto the study. Patients were ineligible if they suffered from metastases of an ocular or mucosal melanoma, had previously received temozolomide or progressed using DTIC. All acute toxicities from previous treatment must have resolved. Patients were excluded from participation in this study if they had: a history of epilepsy or any kind of partial/generalized seizures treated by anticonvulsant medication, any physical conditions requiring glucocorticosteroids for systemic therapy, acute infection requiring intravenous antibiotics, frequent vomiting or medical conditions that could interfere with oral medication intake (e.g. partial bowel obstruction), uncontrolled, medically severe disturbance of cardiac, renal and liver function. All patients provided written informed consent. The study was approved by the ethics committee of each participating centre.
Secondary end points included overall and progression-free survival, as well as time to develop cerebral symptoms.
original article
Annals of Oncology
first 21 evaluable patients per arm, that arm would be extended to a total of 66 evaluable patients. Since a drop-out rate of 15% was assumed, a total of approximately 80 patients/arm was calculated in a positive event. Statistical analyses for baseline demographics, response rates and adverse events were descriptive. Overall survival and time to cerebral progression were analysed using the Kaplan–Meier method.
results patients’ characteristics Between January 2002 and July 2004, 45 patients were included in the study with 21 patients previously treated with chemotherapeutics (arm B) and 24 patients previously untreated (arm A) in eight German centres. All eligible patients were evaluated in the ITT population. The clinical characteristics of all patients are summarised in Table 1. Among the 21 patients who had received prior chemotherapy for systemic extracranial disease, 16 had received one prior
Variable Age, years Median Range Gender Male Female Time from initial diagnosis of melanoma, years £1 1–2 2–3 3–4 4–5 5–10 Prior therapies None 1 ‡2 No. of baseline brain metastases £3 4–5 >5 Pretreatment of brain lesions Yes No Extracerebral metastases None 1 organ only 2–3 organs involved >4 organs affected Distribution of extracranial metastases Lung Liver Bones Skin Lymph nodes Others
1594 | Schadendorf et al.
Patients characteristics 54.5 30–76 29 16 5 13 12 8 2 4 24 16 5 15 14 16 8 37 0 9 16 16 28 21 11 17 27 17
treatment administration A total of 89 cycles of temozolomide were applied with 40 started and 34 completed cycles in arm A patients and 49 started and 41 completed cycles in arm B patients. Only two patients in both arms received three or more cycles. The mean duration of treatment was 48 days in both arms. The reasons for treatment discontinuation were progressive disease in 35 patients and patients’ refusal in six patients. response evaluation and survival The objective brain response was evaluated at week 8 in 40 patients demonstrating a progressive brain disease in 33 (73.3%) of the patients. Two partial responses (PR; one in each arm) and five stable diseases (SD; two prior untreated and three prior treated patients) could be observed in brain metastases (Table 2). Extracerebral disease was evaluated in 37 patients and was found not to be controlled in the majority of the patients (n = 31; 68.9%). Since most of the patients with missing tumour assessment data (Table 2) had clinically progressive disease (PD) before scheduled scans, investigators decided not to perform scans, patients refused scans, or progressed more quickly than anticipated leading to a medically unstable condition not allowing scans, and early deaths, the rate of PD is even higher. Only one patient experienced a PR in extracerebral disease (also responding in the brain) and five patients had disease stabilization. Patients responding in the brain (SD/PR) had a higher likelihood of also responding systemically. Two out of three patients of the untreated group
Volume 17 | No. 10 | October 2006
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
Table 1. Patient demographics and baseline characteristics
therapy, four had two previous regimens and one had three prior treatments including DTIC, platinum derivatives, vincaalkaloids and fotemustine, as well as interferons and IL-2. The median time from diagnosis of primary melanoma to brain metastases was 2 years and the median time from entry into stage IV to brain metastases was 4 months. The majority of patients presented more than one cerebral lesion and >90% of patients had additional, extra-cerebral metastases. Only five (11.1%) patients exhibited one single brain lesion (one in arm A; four in arm B), and in 10 patients (22.2%) two to three brain metastases were detected and classified as not operable (four in arm A; six in arm B). In 14 (31.1%) patients, four to five metastases were counted in the brain (six in arm A; eight in arm B) and in the remaining 15 (33.3%) more than five brain lesions were detected (10 in arm A; five in arm B). Systemic extracranial disease was located in five or six organs in three patients, respectively, and in four organs in 10 patients at the start of therapy. In seven patients metastases had infiltrated into three organs and nine patients each suffered from metastases in one or two organs. In only four patients (two in each arm) was the brain the only location of metastases, which placed these four patients into recursive partitioning analysis (RPA) class I according to Gaspar et al. [18]. All remaining 41 patients (19 in arm A and 22 in arm B with prior chemotherapy) were classified as RPA class II (presence of extracranial disease, Karnofsky index ‡70). Organ systems most often involved were lungs (28 patients) and lymph nodes (27 patients), followed by liver (21 patients), skin and other sites (each 17 patients) and bones (11 patients).
original article
Annals of Oncology
Table 2. Summary of clinical responses in brain and extracerebral metastases after 8 weeks Parameter
Response in brain No. of patients %
Extracerebral response No. of patients %
Partial response Stable disease Progressive disease Missinga Total
2 5 33 5 45
1 5 31 8 45
4.4 11.1 73.3 11.1 100
2.2 11.1 68.9 17.8 100
a
Indicates patients who did not have a clinical response recorded by the investigator. Primary reasons were disease progression before scheduled scans; investigator decided not to perform scan, patients refused scans or patients progressed more quickly than anticipated and died or were not medically stable to undergo scan procedures.
safety results Myelosuppression was the most severe adverse event; however, grade 3/4 toxicity occurred only rarely in 2.4% of patients with one patient experiencing grade 4 leukopenia, neutropenia and thrombocytopenia (Table 3). In addition, grade 3/4 toxicities such as bilirubinemia, rise in liver enzymes, constipation, sensory or motor deficits as well as consciousness were only observed in very few patients (one to two in each category). Only two patients discontinued treatment because of toxicity. One patient stopped because of grade 2 thrombocytopenia; the other because of grade 3 leukopenia. In three patients a one-level dose reduction was necessary and in three other patients a delay in therapy was needed. Treatment discontinuation because of disease progression was the most common reason.
discussion Brain metastases are the most frequent neurological complication in cancer. Melanoma is one of the most frequent neoplasias causing cerebral metastases [10, 19]. Prognosis for patients with brain metastases is poor, independent of the primary tumour. Gaspar et al. [18] and the Radiation Therapy Oncology Group (RTOG) identified a three-tier classification scheme—the recursive partitioning analysis (RPA) classes with a high survival prediction value. Median survival ranged from
Volume 17 | No. 10 | October 2006
Figure 1. Kaplan–Meier estimate of overall survival in months for patients who had not received prior chemotherapy (black line) and for patients with previous chemotherapy (broken line).
7.1 months for RPA class I patients (controlled brain disease and without extracranial involvement) to 4.2 months for RPA class II patients (KI >70%, uncontrolled brain disease; extracerebral disease and/or age <65 years) and 2.3 months for RPA class III patients with reduced Karnofsky index <70. This risk classification has been confirmed by various other groups [20–22]. In a British study, the prognostic index of the RPA classes was confirmed in a retrospective analysis of 112 melanoma patients with brain metastases following palliative whole brain radiotherapy suggesting that patients with RPA class III, leptomeningial involvement or patients with RPA class II with poor prognostic index were unlikely to benefit from palliative whole brain radiotherapy (WBRT) [23]. No standard systemic therapy exists for patients with CNS melanoma who suffer in most cases from additional extracerebral disease. Of the available drugs, temozolomide and fotemustine have been used most frequently in this situation with limited success. In a recent prospective randomized multicentre trial in Europe, 76 patients were randomised receiving either fotemustine or fotemustine plus WBRT [6]. There was no significant difference in brain response, control rates or overall survival time (86 days versus 105 days). Several phase II trials suggest that single-agent temozolomide has modest activity in patients with recurrent or progressive brain metastases [10]. The most widely used dosing schedule of temozolomide is a dose of 150–200 mg/m2/day for 5 sequential days every 4 weeks. In the largest, so far, conducted two-arm phase II study in patients with melanoma brain metastases, patients with no prior chemotherapy treated with temozolomide as outlined above yielded a median survival of 3.5 months and previously pretreated patients survived 2.2 months [24]. Using the same two-arm phase II strategy with a two-step Simon’s design as in Agarwala et al. [24], in the present study patients with melanoma metastases were treated with a more dose-intense temozolomide regimen. In contrast to
doi:10.1093/annonc/mdl148 | 1595
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
who had a systemic SD had a corresponding SD/PR in the brain with time-to-progression of 14 weeks in those two patients (compared with 6.5 weeks in PD patients in the untreated group). Similarly, three out of four pretreated patients with objective brain response (one PR, three SD) exhibited either a PR (one) or a SD (two) systemically with a median time-to-progression of 14.6 weeks (compared with 6 weeks in progressive patients). Therefore, only a small subset of patients in both arms achieved a clinical benefit. The Kaplan–Meier estimates of survival are shown in Figure 1. Median overall survival was 4.1 months (range 0.7–11.8 months) for the entire study population; 3.6 months (range 0.7– 8.3 months) in patients with prior chemotherapy and 4.3 months (range 1.6–11.8 months) in untreated patients.
original article
Annals of Oncology
Table 3. Summary of adverse events in 42 patients Adverse event
Haematologic Anaemia Thrombocytopenia Neutropenia Leukopenia Non-haematologic Nausea Vomiting Diarrhoea Obstipation Liver enzymes Neurologic
No chemotherapy All grades No. of patients
%
Prior chemotheray All grades No. of patients
%
Grade 3/4 No. of patients
%
0 0 0 0
0 0 0 0
9 7 1 4
21.4 16.7 2.4 9.5
0 1 1 1
0 2.4 2.4 2.4
0 0 0 0 0 0
0 0 0 0 0 0
8 3 0 6 9 4
19.0 7.1 0 14.3 21.4 9.5
0 0 0 0 0 2
0 0 0 0 0 4.8
%
Grade 3/4 No. of patients
6 1 1 3
14.3 2.4 2.4 7.1
4 2 2 6 4 0
9.5 4.8 4.8 14.3 9.5 0
1596 | Schadendorf et al.
prospective studies should evaluate further this potential beneficial effect of temozolomide, possibly in combination with other agents.
references 1. Eigentler TK, Caroli UM, Radny P, Garbe C. Palliative therapy of disseminated malignant melanoma: a systematic review of 41 randomised clinical trials. Lancet Oncol 2003; 4: 748–759. 2. Douglas JG, Margolin K. The treatment of brain metastases from malignant melanoma. Sem Oncol 2002; 29: 518–524. 3. Cattell E, Kelly C, Middelton MR. Brain metastases in melanoma: a European perspective. Sem Oncol 2002; 29: 513–517. 4. Sampson JH, Carter JH, Friedman AH, Seigler HF. Demographics, prognosis and therapy in 702 patients with brain metastases from malignant melanoma. J Neurosurg 1998; 88: 11–20. 5. Jacquillat C, Khayat D, Banzet P et al. Final report of the French multicenter phase II study of the nitrosourea fotemustine in 153 evaluable patients with disseminated malignant melanoma including patients with cerebral metastases. Cancer 1990; 66: 1873–1878. 6. Mornex F, Thomas L, Mohr P et al. A prospective randomized multicentre phase III trial of fotemustine plus whole brain irradiation versus fotemustine alone in cerebral metastases of malignant melanoma. Melanoma Res 2003; 13: 97–103. 7. Middelton MR, Grob JJ, Aaronson N et al. Randomized phase III study of temozolomide versus dacarbacine in the treatment of patients with advanced metastatic malignant melanoma. J Clin Oncol 2000; 16: 158–166. 8. Kaufmann R, Spieth K, Leiter U et al. Temozolomide in combination with interferon alfa versus temozolomide alone in patients with advanced metastatic melanoma: A randomized phase III multicenter study of the Dermatologic Cooperative Oncology Group (DeCOG). J Clin Oncol 2005; 23: 9001–9007. 9. Stupp R, Mason WP, van den Bent MJ et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352: 987–996. 10. Langer CJ, Metha MP. Current management of brain metastases, with a focus on systemic options. J Clin Oncol 2005; 23: 6207–6219. 11. Antonadou D, Paraskevaidis M, Serris G et al. Phase II randomized trial of temozolomide and concurrent radiotherapy in patients with brain metastases. J Clin Oncol 2002; 20: 3644–3650. 12. Christodoulou C, Bafaloukis D, Kosmidis P et al. Hellenic Cooperative Oncology Group. Phase II study of temozolomide in heavily pretreated cancer patients with brain metastases. Ann Oncol 2001; 1: 249–254.
Volume 17 | No. 10 | October 2006
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
the approved dosing schedule with a weekly dose intensity of 250 mg/m2 the temozolomide scheme given in this study (7 days on, 7 days off) mounted to a dose intensity of 525 mg/m2 with an increased cerebral drug concentration [25]. Although tolerability of this dose-intensified schedule was good, clinical efficacy was unfortunately limited with a median survival of 3.6 months in pretreated and 4.3 months in prior untreated patients and only two patients with short-lasting PR. Therefore, an increased dose intensity did not lead to a significant clinical benefit and prognosis seemed to be dictated by the underlying RPA class. The results indicate that as long as cerebral and extracerebral disease cannot be controlled, a significant survival benefit cannot be expected in the majority of patients. Our results are in contrast to Siena et al. [26] using the same temozolomide dosing schedule which reports a 40% PR/SD rate in 21 melanoma patients. However, since a full report has not been published so far, one can only speculate about the reasons leading to the different results with patient selection being the most likely factor. The observations and conclusions of this report, however, are in line with studies by the American Cytokine Working Group, which tested similar combination schemes of WBRT and temozolomide in melanoma patients who showed an improved survival in patients with glioblastoma [9]. The American Cytokine Working Group [13] tested the prolonged temozolomide dosing regimen (75 mg/m2 for 6 weeks)—in a second trial in combination with thalidomide [27]—plus WBRT in 31 and 40 patients, respectively. In both trials, three patients experienced an objective response with a median survival between 4 and 6 months. In conclusion, temozolomide was well tolerated in this doseintensified phase II study in patients with melanoma metastases in the brain. Dose intensification did not lead to a significant increase in response rates in brain or extracerebral disease with only modest antitumour activity in a small subset of patients. Other treatment approaches are clearly necessary for this patient population. Since temozolomide has some activity in metastatic melanoma comparable to DTIC [1, 7, 8] and as smaller retrospective analyses suggest that temozolomide might have a preventive effect on cerebral metastases [14, 15], future
Annals of Oncology
Volume 17 | No. 10 | October 2006
21. Harrison BE, Johnson JL, Clough RW, Halperin EC. Selection of patients with melanoma brain metastases for aggressive treatment. Am J Clin Oncol 2003; 26: 354–357. 22. Nieder C, Nestle U, Motaref B et al. Prognostic factors in brain metastases: should patients be selected for agressive treatment according to recursive partitioning analysis (RPA) classes? Int J Radiat Oncol Biol Phys 2000; 46: 297–302. 23. Morris SL, Low SH, A’Hern RP et al. A prognostic index that predicts outcome following palliative whole brain radiotherapy for patients with metastatic malignant melanoma. Br J Cancer 2004; 91: 829–833. 24. Agarwala SS, Kirkwood JM, Gore M et al. Temozolomide for the treatment of brain metastases associated with metastatic melanoma: a phase II study. J Clin Oncol 2004; 22: 2101–2107. 25. Stupp R, Gander M, Leyvraz S, Newlands E. Current and future developments in the use of temozolomide for the treatment of brain tumours. Lancet Oncol 2001; 2: 552–560. 26. Siena S, Landonio G, Baietta E et al. Multicenter phase II study of temozolomide therapy for brain metastasis in patients with malignant melanoma, breast cancer, and non-small cell lung cancer. ASCO 2003; Abstr 407. 27. Atkins MB, Sosman J, Agarwala SS et al. A Cytokine Working Group phase II study of temozolomide (TMZ), thalidomide (THAL) and whole brain radiation therapy (WBRT) for patients with brain metastases from melanoma. ASCO 2005; Abstr 7552.
doi:10.1093/annonc/mdl148 | 1597
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
13. Margolin K, Atkins MB, Thompson JA et al. Temozolomide and whole brain irradiation in melanoma metastatic to the brain: a phase II trial of the Cytokine Working Group. J Cancer Res Clin Oncol 2002; 128: 214–218. 14. Paul MJ, Summers Y, Calvert AH et al. Effect of temozolomide on central nervous system relapse in patients with advanced melanoma. Melanoma Res 2002; 12: 175–178. 15. Weber RW, O’Day S, Rose M et al. Low-dose outpatient chemobiotherapy with temozolomide, granulocyte-macrophage colony stimulating factor, interferonalpha2b, and recombinant interleukin-2 for the treatment of metastatic melanoma. J Clin Oncol 2005; 23: 8992–9000. 16. Therasse P, Arbuck SG, Eisenhauer EA et al. New 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. 17. Simon R. Optimal two-stage designs for phase II clinical trials. Control Clin Trials 1989; 10: 1–10. 18. Gaspar L, Scott C, Rotman M et al. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 1997; 4: 745–751. 19. Gupta G, Robertson AG, Mackie RM. Cerebral metastases of cutaneous melanoma. Br J Cancer 1997; 76: 256–259. 20. Fife KM, Colman MH, Stevens GN et al. Determinants of outcome in melanoma patients with cerebral metastases. J Clin Oncol 2004; 22: 1293–1300.
original article
Annals of Oncology 17: 1592–1597, 2006 doi:10.1093/annonc/mdl148
Dose-intensified bi-weekly temozolomide in patients with asymptomatic brain metastases from malignant melanoma: a phase II DeCOG/ADO study D. Schadendorf1*, A. Hauschild2, S. Ugurel1, A. Thoelke1, F. Egberts2, M. Kreissig3, R. Linse4, U. Trefzer5, T. Vogt6, W. Tilgen7, P. Mohr8 & C. Garbe3 1 Skin Cancer Unit, German Cancer Research Center & University Hospital Mannheim, Department of Dermatology; 2University Hospital, Kiel; 3University Hospital, Tu¨ bingen; 4Community Hospital, Erfurt; 5Charite Hospital, Humboldt University, Berlin; 6University Hospital, Regensburg; 7University Hospital, Homburg; 8 Elbe Hospital, Buxtehude, Germany
Received 17 March 2006; revised 9 May 2006; accepted 22 May 2006
original article
approval to treat brain tumours. The purpose of this study was to test a dose-intensified regimen of temozolomide in melanoma patients with brain metastases in a prospective, open-label, multicentre phase II trial. Patients and methods: Forty-five patients with asymptomatic brain metastases from melanoma were stratified into arm A (no prior chemotherapy; n = 21) and arm B (previous chemotherapy; n = 24). Patients received oral temozolomide either 150 mg/m2/day (arm A) or 125 mg/m2/day (arm B), days 1–7 and 15–21, every 28 days. The primary study end point was objective response, and secondary end points were overall survival and safety. Results: Two patients (4.4%) achieved a partial response (PR) in brain metastases (one in each arm), one of them (2.2%) also showing a PR in extracerebral disease. An additional five patients (11.1%; two in arm A, three in arm B) showed disease stabilisation (SD) in brain and other sites. However, 82% revealed progressive disease (PD) already evident 8 weeks after therapy initiation. Median survival time from therapy onset was 3.5 months (range 0.7–8.3; arm B) and 4.3 months (range 1.6–11.8; arm A), P = 0.43. Dose modifications and prolongations of therapy cycles due to toxicity were required in 20% of patients. Grade 3/4 toxicity was observed in one patient only (2.2%). Conclusions: Oral administration of temozolomide given bi-weekly is well-tolerated in melanoma patients with cerebral involvement. However, the efficacy is limited, with lower than 5% objective responses observed in brain and extracerebral metastases. Key words: temozolomide, brain metastases, metastatic melanoma
introduction Malignant melanoma in its advanced stage of metastatic disease has a poor prognosis with a median survival time of approximately 9 months [1]. Furthermore, melanoma is the third most common cause of metastases in the brain after carcinomas of the lung and breast [2]. Initially, up to 20% of patients with stage IV melanoma suffer from brain metastases [3]. Metastases of the CNS develop in nearly half of advanced melanoma patients [2, 4] during the progression of disease dominating prognosis of those patients. Surgical excision or radiosurgery are the first choice for patients with solitary brain metastases. In the case of multiple or inoperable CNS metastases, whole brain irradiation (WBI) is considered as the current standard of care, although only neurological symptoms *Correspondence to: Dr D. Schadendorf, Skin Cancer Unit, German Cancer Research Center & University Hospital Mannheim, Theodor Kutzer Ufer 1, D-68135 Mannheim, Germany. Tel: +49-621-383-2126; Fax: +49-621-383-2163; E-mail: [email protected]
ª 2006 European Society for Medical Oncology
are improved and no survival advantage has been shown so far. In most cases of advanced melanoma, the presence of brain metastases is associated with widespread visceral involvement. The failure of systemic drug therapy in metastatic melanoma may be related to their lacking ability to cross the blood–brain barrier. Within the group of cytotoxic drugs known to have some activity in melanoma, fotemustine, a nitrosourea analogue evaluated in several phase II open label studies, was shown to have an activity on cerebral metastases as well as other metastatic sites [5, 6]. Temozolomide is a novel alkylating, orally available agent with some activity in metastatic melanoma [7, 8]. Furthermore, it was demonstrated that temozolomide penetrates the blood–brain barrier. Low-dose continuous temozolomide over 6 weeks combined with WBI was shown to improve overall survival in patients with glioblastoma compared with WBI alone [9]. Clinical responses and improvement of neurological symptoms were shown in other cancers as well [10–12]. In retrospective analyses of patients with advanced
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
Background: Temozolomide has shown some efficacy in metastatic melanoma and recently received extended
original article
Annals of Oncology
melanoma, a reduced relapse risk in the brain after temozolomide treatment compared with dacarbazine was suggested [13–15]. Here we report the results of a first prospective, open-label, phase II trial evaluating the safety and efficacy of a doseintensified regimen of temozolomide in patients with brain metastases of malignant melanoma.
patients and methods patients
protocol design and treatment This national multicentre, open-label phase II study of temozolomide included melanoma patients who had had newly diagnosed (<6 weeks), asymptomatic cerebral metastases. Patients were stratified into two arms, those who had or had not received previous chemotherapy for metastatic melanoma. Temozolomide was given orally in the fasting state, once a day for 7 days/week, repeated every other week, at a dose of 150 mg/m2/day (previously untreated, arm A) or 125 mg/m2/day (previously treated, arm B) on days 1–7 and days 15–21 of every cycle month. Temozolomide was administered until the occurrence of either unacceptable toxicity or evidence of progression of cerebral disease for a maximum of 1 year from initial treatment. Antiemetics were administered at the discretion of the treating physician. Primary study end points were the efficacy of temozolomide at week 8 and week 20 in patients in terms of clinical response rate and toxicity in patients who had previously been untreated for the presentation of brain metastases. Both patient groups who had received prior or no prior chemotherapy for extracerebral melanoma disease were evaluated separately.
Volume 17 | No. 10 | October 2006
response and toxicity assessment All patients had routine evaluations before treatment, including complete medical history, physical examination, complete tumour imaging [brain computed tomography (CT) or Gd-magnetic resonance imaging (MRI), chest X-ray, sonography of the abdomen or CT scan], complete blood cell count and blood chemistry. At week 8 and week 20 after beginning study treatment, patients were restaged using the same imaging technique for evaluating the rate of objective responses (CR, PR), which was the primary objective of the study. Tumour response evaluation was based on RECIST criteria [16]. Responses were confirmed 4 weeks later. Intracerebral and extracerebral responses were recorded separately. Toxicity was recorded on day 1 of each cycle according to the Common Toxicity Criteria (CTC) together with a complete medical history and a physical examination. A neurological assessment and a documentation of vital signs were also done. Laboratory assessment (complete blood chemistry, blood count and serum chemistry) was performed on a weekly basis.
dose modifications Toxicity led to dose reduction based on the most severe grade of any single toxicity observed. Unacceptable toxicity resulted in drug withdrawal. Upon resolution of the toxicity, patients continued treatment at one dose level below the dose level administered. If more than a two-dose level reduction (in previously untreated patients; level I, 125 mg/m2; level II, 100 mg/m2) or more than a one-dose level reduction (in previously treated patients) were required for continuing treatment of any patient, then the patient was withdrawn from the study. Therefore, patients who required dose reductions to a dose level of <100 mg/m2/day of temozolomide were taken off the study. If acute neutrophil count (ANC) <1500 cells/ll (‡grade 2) and/or platelet count <100 000/ll (‡grade 1) occurred at any given time while on therapy, therapy was delayed until haematological recovery (ANC ‡1500 and/or platelet count ‡100 000). If restart of therapy was not possible within 21 days after last drug application, the patient went off study. Two or more weeks of therapy delay required a dose reduction by one level. ANC <500 cells/ll for 5 days and/or platelet count <25 000 delayed restarting of therapy until haematological recovery (ANC ‡1500 and/or platelet count ‡100 000). A subsequent dose reduction by one dose level was mandatory. If restart of therapy was not possible within 21 days after last drug application, the patient went off study. Febrile neutropenia (ANC <500 with fever temperature >38.3C) also led to patient withdrawal from the study. For CTC grade 2 or less non-haematological toxicity, no dose reductions were required. For grade 3 or 4 non-haematological toxicity (including GI toxicity unresponsive to standard therapy), a delay in dosing was recommended until toxicity had resolved to baseline or grade 1. If restart of therapy was not possible within 21 days after last drug application, then the patient went off study. All non-haematological CTC grade 2, 3 and 4 toxicities must have resolved to at least CTC grade 1 (or for baseline CTC grade 2 elevations in transaminases, bilirubin and/or alkaline phosphatase to the level defined by the inclusion criteria) prior to repeat dosing.
statistical analyses The primary end point was to assess the cerebral and extracerebral objective response rate (defined as complete responses, partial response and stable disease) in week 8 and week 20 in melanoma patients with asymptomatic brain metastases. Patients were accrued in this study using a two-stage Simon’s design [17] in order to improve the probability of observing the real response rate. Therefore, if there were three or more responders among the
doi:10.1093/annonc/mdl148 | 1593
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
Patients were enrolled with histological confirmed malignant melanoma and newly diagnosed (within 6 weeks prior to study treatment) brain metastases, which were irresectable and neurologically asymptomatic. Eligible patients who met the following eligibility criteria were included in the study: age ‡18 years, a life expectancy greater than 3 months, Karnofsky performance status ‡70% and adequate blood cell counts (polynuclear neutrophils ‡1500/mm3, platelets ‡100 000/mm3 and haemoglobin ‡8 g/dl). Patients with or without prior chemotherapy [except temozolomide or disease progression during previous dacarbazine (DTIC) therapy] were accepted. Patients were required to have no chemotherapy within the previous 4 weeks and no previous cerebral radiotherapy. Measurable disease according to RECIST criteria, i.e. at least one ‘target’ lesion ‡1 cm by MRI or CT of the brain, was required. No steroid medication was allowed during treatment and prior treatment had to be stopped at least 7 days before onset of therapy. Previous chemotherapy for metastatic extracerebral disease was permitted and patients were stratified according to previous treatment status. Patients with previously treated cerebral metastases with stereotactic radio- or neuro-surgery and newly developing brain metastases were allowed onto the study. Patients were ineligible if they suffered from metastases of an ocular or mucosal melanoma, had previously received temozolomide or progressed using DTIC. All acute toxicities from previous treatment must have resolved. Patients were excluded from participation in this study if they had: a history of epilepsy or any kind of partial/generalized seizures treated by anticonvulsant medication, any physical conditions requiring glucocorticosteroids for systemic therapy, acute infection requiring intravenous antibiotics, frequent vomiting or medical conditions that could interfere with oral medication intake (e.g. partial bowel obstruction), uncontrolled, medically severe disturbance of cardiac, renal and liver function. All patients provided written informed consent. The study was approved by the ethics committee of each participating centre.
Secondary end points included overall and progression-free survival, as well as time to develop cerebral symptoms.
original article
Annals of Oncology
first 21 evaluable patients per arm, that arm would be extended to a total of 66 evaluable patients. Since a drop-out rate of 15% was assumed, a total of approximately 80 patients/arm was calculated in a positive event. Statistical analyses for baseline demographics, response rates and adverse events were descriptive. Overall survival and time to cerebral progression were analysed using the Kaplan–Meier method.
results patients’ characteristics Between January 2002 and July 2004, 45 patients were included in the study with 21 patients previously treated with chemotherapeutics (arm B) and 24 patients previously untreated (arm A) in eight German centres. All eligible patients were evaluated in the ITT population. The clinical characteristics of all patients are summarised in Table 1. Among the 21 patients who had received prior chemotherapy for systemic extracranial disease, 16 had received one prior
Variable Age, years Median Range Gender Male Female Time from initial diagnosis of melanoma, years £1 1–2 2–3 3–4 4–5 5–10 Prior therapies None 1 ‡2 No. of baseline brain metastases £3 4–5 >5 Pretreatment of brain lesions Yes No Extracerebral metastases None 1 organ only 2–3 organs involved >4 organs affected Distribution of extracranial metastases Lung Liver Bones Skin Lymph nodes Others
1594 | Schadendorf et al.
Patients characteristics 54.5 30–76 29 16 5 13 12 8 2 4 24 16 5 15 14 16 8 37 0 9 16 16 28 21 11 17 27 17
treatment administration A total of 89 cycles of temozolomide were applied with 40 started and 34 completed cycles in arm A patients and 49 started and 41 completed cycles in arm B patients. Only two patients in both arms received three or more cycles. The mean duration of treatment was 48 days in both arms. The reasons for treatment discontinuation were progressive disease in 35 patients and patients’ refusal in six patients. response evaluation and survival The objective brain response was evaluated at week 8 in 40 patients demonstrating a progressive brain disease in 33 (73.3%) of the patients. Two partial responses (PR; one in each arm) and five stable diseases (SD; two prior untreated and three prior treated patients) could be observed in brain metastases (Table 2). Extracerebral disease was evaluated in 37 patients and was found not to be controlled in the majority of the patients (n = 31; 68.9%). Since most of the patients with missing tumour assessment data (Table 2) had clinically progressive disease (PD) before scheduled scans, investigators decided not to perform scans, patients refused scans, or progressed more quickly than anticipated leading to a medically unstable condition not allowing scans, and early deaths, the rate of PD is even higher. Only one patient experienced a PR in extracerebral disease (also responding in the brain) and five patients had disease stabilization. Patients responding in the brain (SD/PR) had a higher likelihood of also responding systemically. Two out of three patients of the untreated group
Volume 17 | No. 10 | October 2006
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
Table 1. Patient demographics and baseline characteristics
therapy, four had two previous regimens and one had three prior treatments including DTIC, platinum derivatives, vincaalkaloids and fotemustine, as well as interferons and IL-2. The median time from diagnosis of primary melanoma to brain metastases was 2 years and the median time from entry into stage IV to brain metastases was 4 months. The majority of patients presented more than one cerebral lesion and >90% of patients had additional, extra-cerebral metastases. Only five (11.1%) patients exhibited one single brain lesion (one in arm A; four in arm B), and in 10 patients (22.2%) two to three brain metastases were detected and classified as not operable (four in arm A; six in arm B). In 14 (31.1%) patients, four to five metastases were counted in the brain (six in arm A; eight in arm B) and in the remaining 15 (33.3%) more than five brain lesions were detected (10 in arm A; five in arm B). Systemic extracranial disease was located in five or six organs in three patients, respectively, and in four organs in 10 patients at the start of therapy. In seven patients metastases had infiltrated into three organs and nine patients each suffered from metastases in one or two organs. In only four patients (two in each arm) was the brain the only location of metastases, which placed these four patients into recursive partitioning analysis (RPA) class I according to Gaspar et al. [18]. All remaining 41 patients (19 in arm A and 22 in arm B with prior chemotherapy) were classified as RPA class II (presence of extracranial disease, Karnofsky index ‡70). Organ systems most often involved were lungs (28 patients) and lymph nodes (27 patients), followed by liver (21 patients), skin and other sites (each 17 patients) and bones (11 patients).
original article
Annals of Oncology
Table 2. Summary of clinical responses in brain and extracerebral metastases after 8 weeks Parameter
Response in brain No. of patients %
Extracerebral response No. of patients %
Partial response Stable disease Progressive disease Missinga Total
2 5 33 5 45
1 5 31 8 45
4.4 11.1 73.3 11.1 100
2.2 11.1 68.9 17.8 100
a
Indicates patients who did not have a clinical response recorded by the investigator. Primary reasons were disease progression before scheduled scans; investigator decided not to perform scan, patients refused scans or patients progressed more quickly than anticipated and died or were not medically stable to undergo scan procedures.
safety results Myelosuppression was the most severe adverse event; however, grade 3/4 toxicity occurred only rarely in 2.4% of patients with one patient experiencing grade 4 leukopenia, neutropenia and thrombocytopenia (Table 3). In addition, grade 3/4 toxicities such as bilirubinemia, rise in liver enzymes, constipation, sensory or motor deficits as well as consciousness were only observed in very few patients (one to two in each category). Only two patients discontinued treatment because of toxicity. One patient stopped because of grade 2 thrombocytopenia; the other because of grade 3 leukopenia. In three patients a one-level dose reduction was necessary and in three other patients a delay in therapy was needed. Treatment discontinuation because of disease progression was the most common reason.
discussion Brain metastases are the most frequent neurological complication in cancer. Melanoma is one of the most frequent neoplasias causing cerebral metastases [10, 19]. Prognosis for patients with brain metastases is poor, independent of the primary tumour. Gaspar et al. [18] and the Radiation Therapy Oncology Group (RTOG) identified a three-tier classification scheme—the recursive partitioning analysis (RPA) classes with a high survival prediction value. Median survival ranged from
Volume 17 | No. 10 | October 2006
Figure 1. Kaplan–Meier estimate of overall survival in months for patients who had not received prior chemotherapy (black line) and for patients with previous chemotherapy (broken line).
7.1 months for RPA class I patients (controlled brain disease and without extracranial involvement) to 4.2 months for RPA class II patients (KI >70%, uncontrolled brain disease; extracerebral disease and/or age <65 years) and 2.3 months for RPA class III patients with reduced Karnofsky index <70. This risk classification has been confirmed by various other groups [20–22]. In a British study, the prognostic index of the RPA classes was confirmed in a retrospective analysis of 112 melanoma patients with brain metastases following palliative whole brain radiotherapy suggesting that patients with RPA class III, leptomeningial involvement or patients with RPA class II with poor prognostic index were unlikely to benefit from palliative whole brain radiotherapy (WBRT) [23]. No standard systemic therapy exists for patients with CNS melanoma who suffer in most cases from additional extracerebral disease. Of the available drugs, temozolomide and fotemustine have been used most frequently in this situation with limited success. In a recent prospective randomized multicentre trial in Europe, 76 patients were randomised receiving either fotemustine or fotemustine plus WBRT [6]. There was no significant difference in brain response, control rates or overall survival time (86 days versus 105 days). Several phase II trials suggest that single-agent temozolomide has modest activity in patients with recurrent or progressive brain metastases [10]. The most widely used dosing schedule of temozolomide is a dose of 150–200 mg/m2/day for 5 sequential days every 4 weeks. In the largest, so far, conducted two-arm phase II study in patients with melanoma brain metastases, patients with no prior chemotherapy treated with temozolomide as outlined above yielded a median survival of 3.5 months and previously pretreated patients survived 2.2 months [24]. Using the same two-arm phase II strategy with a two-step Simon’s design as in Agarwala et al. [24], in the present study patients with melanoma metastases were treated with a more dose-intense temozolomide regimen. In contrast to
doi:10.1093/annonc/mdl148 | 1595
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
who had a systemic SD had a corresponding SD/PR in the brain with time-to-progression of 14 weeks in those two patients (compared with 6.5 weeks in PD patients in the untreated group). Similarly, three out of four pretreated patients with objective brain response (one PR, three SD) exhibited either a PR (one) or a SD (two) systemically with a median time-to-progression of 14.6 weeks (compared with 6 weeks in progressive patients). Therefore, only a small subset of patients in both arms achieved a clinical benefit. The Kaplan–Meier estimates of survival are shown in Figure 1. Median overall survival was 4.1 months (range 0.7–11.8 months) for the entire study population; 3.6 months (range 0.7– 8.3 months) in patients with prior chemotherapy and 4.3 months (range 1.6–11.8 months) in untreated patients.
original article
Annals of Oncology
Table 3. Summary of adverse events in 42 patients Adverse event
Haematologic Anaemia Thrombocytopenia Neutropenia Leukopenia Non-haematologic Nausea Vomiting Diarrhoea Obstipation Liver enzymes Neurologic
No chemotherapy All grades No. of patients
%
Prior chemotheray All grades No. of patients
%
Grade 3/4 No. of patients
%
0 0 0 0
0 0 0 0
9 7 1 4
21.4 16.7 2.4 9.5
0 1 1 1
0 2.4 2.4 2.4
0 0 0 0 0 0
0 0 0 0 0 0
8 3 0 6 9 4
19.0 7.1 0 14.3 21.4 9.5
0 0 0 0 0 2
0 0 0 0 0 4.8
%
Grade 3/4 No. of patients
6 1 1 3
14.3 2.4 2.4 7.1
4 2 2 6 4 0
9.5 4.8 4.8 14.3 9.5 0
1596 | Schadendorf et al.
prospective studies should evaluate further this potential beneficial effect of temozolomide, possibly in combination with other agents.
references 1. Eigentler TK, Caroli UM, Radny P, Garbe C. Palliative therapy of disseminated malignant melanoma: a systematic review of 41 randomised clinical trials. Lancet Oncol 2003; 4: 748–759. 2. Douglas JG, Margolin K. The treatment of brain metastases from malignant melanoma. Sem Oncol 2002; 29: 518–524. 3. Cattell E, Kelly C, Middelton MR. Brain metastases in melanoma: a European perspective. Sem Oncol 2002; 29: 513–517. 4. Sampson JH, Carter JH, Friedman AH, Seigler HF. Demographics, prognosis and therapy in 702 patients with brain metastases from malignant melanoma. J Neurosurg 1998; 88: 11–20. 5. Jacquillat C, Khayat D, Banzet P et al. Final report of the French multicenter phase II study of the nitrosourea fotemustine in 153 evaluable patients with disseminated malignant melanoma including patients with cerebral metastases. Cancer 1990; 66: 1873–1878. 6. Mornex F, Thomas L, Mohr P et al. A prospective randomized multicentre phase III trial of fotemustine plus whole brain irradiation versus fotemustine alone in cerebral metastases of malignant melanoma. Melanoma Res 2003; 13: 97–103. 7. Middelton MR, Grob JJ, Aaronson N et al. Randomized phase III study of temozolomide versus dacarbacine in the treatment of patients with advanced metastatic malignant melanoma. J Clin Oncol 2000; 16: 158–166. 8. Kaufmann R, Spieth K, Leiter U et al. Temozolomide in combination with interferon alfa versus temozolomide alone in patients with advanced metastatic melanoma: A randomized phase III multicenter study of the Dermatologic Cooperative Oncology Group (DeCOG). J Clin Oncol 2005; 23: 9001–9007. 9. Stupp R, Mason WP, van den Bent MJ et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005; 352: 987–996. 10. Langer CJ, Metha MP. Current management of brain metastases, with a focus on systemic options. J Clin Oncol 2005; 23: 6207–6219. 11. Antonadou D, Paraskevaidis M, Serris G et al. Phase II randomized trial of temozolomide and concurrent radiotherapy in patients with brain metastases. J Clin Oncol 2002; 20: 3644–3650. 12. Christodoulou C, Bafaloukis D, Kosmidis P et al. Hellenic Cooperative Oncology Group. Phase II study of temozolomide in heavily pretreated cancer patients with brain metastases. Ann Oncol 2001; 1: 249–254.
Volume 17 | No. 10 | October 2006
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
the approved dosing schedule with a weekly dose intensity of 250 mg/m2 the temozolomide scheme given in this study (7 days on, 7 days off) mounted to a dose intensity of 525 mg/m2 with an increased cerebral drug concentration [25]. Although tolerability of this dose-intensified schedule was good, clinical efficacy was unfortunately limited with a median survival of 3.6 months in pretreated and 4.3 months in prior untreated patients and only two patients with short-lasting PR. Therefore, an increased dose intensity did not lead to a significant clinical benefit and prognosis seemed to be dictated by the underlying RPA class. The results indicate that as long as cerebral and extracerebral disease cannot be controlled, a significant survival benefit cannot be expected in the majority of patients. Our results are in contrast to Siena et al. [26] using the same temozolomide dosing schedule which reports a 40% PR/SD rate in 21 melanoma patients. However, since a full report has not been published so far, one can only speculate about the reasons leading to the different results with patient selection being the most likely factor. The observations and conclusions of this report, however, are in line with studies by the American Cytokine Working Group, which tested similar combination schemes of WBRT and temozolomide in melanoma patients who showed an improved survival in patients with glioblastoma [9]. The American Cytokine Working Group [13] tested the prolonged temozolomide dosing regimen (75 mg/m2 for 6 weeks)—in a second trial in combination with thalidomide [27]—plus WBRT in 31 and 40 patients, respectively. In both trials, three patients experienced an objective response with a median survival between 4 and 6 months. In conclusion, temozolomide was well tolerated in this doseintensified phase II study in patients with melanoma metastases in the brain. Dose intensification did not lead to a significant increase in response rates in brain or extracerebral disease with only modest antitumour activity in a small subset of patients. Other treatment approaches are clearly necessary for this patient population. Since temozolomide has some activity in metastatic melanoma comparable to DTIC [1, 7, 8] and as smaller retrospective analyses suggest that temozolomide might have a preventive effect on cerebral metastases [14, 15], future
Annals of Oncology
Volume 17 | No. 10 | October 2006
21. Harrison BE, Johnson JL, Clough RW, Halperin EC. Selection of patients with melanoma brain metastases for aggressive treatment. Am J Clin Oncol 2003; 26: 354–357. 22. Nieder C, Nestle U, Motaref B et al. Prognostic factors in brain metastases: should patients be selected for agressive treatment according to recursive partitioning analysis (RPA) classes? Int J Radiat Oncol Biol Phys 2000; 46: 297–302. 23. Morris SL, Low SH, A’Hern RP et al. A prognostic index that predicts outcome following palliative whole brain radiotherapy for patients with metastatic malignant melanoma. Br J Cancer 2004; 91: 829–833. 24. Agarwala SS, Kirkwood JM, Gore M et al. Temozolomide for the treatment of brain metastases associated with metastatic melanoma: a phase II study. J Clin Oncol 2004; 22: 2101–2107. 25. Stupp R, Gander M, Leyvraz S, Newlands E. Current and future developments in the use of temozolomide for the treatment of brain tumours. Lancet Oncol 2001; 2: 552–560. 26. Siena S, Landonio G, Baietta E et al. Multicenter phase II study of temozolomide therapy for brain metastasis in patients with malignant melanoma, breast cancer, and non-small cell lung cancer. ASCO 2003; Abstr 407. 27. Atkins MB, Sosman J, Agarwala SS et al. A Cytokine Working Group phase II study of temozolomide (TMZ), thalidomide (THAL) and whole brain radiation therapy (WBRT) for patients with brain metastases from melanoma. ASCO 2005; Abstr 7552.
doi:10.1093/annonc/mdl148 | 1597
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 25, 2015
13. Margolin K, Atkins MB, Thompson JA et al. Temozolomide and whole brain irradiation in melanoma metastatic to the brain: a phase II trial of the Cytokine Working Group. J Cancer Res Clin Oncol 2002; 128: 214–218. 14. Paul MJ, Summers Y, Calvert AH et al. Effect of temozolomide on central nervous system relapse in patients with advanced melanoma. Melanoma Res 2002; 12: 175–178. 15. Weber RW, O’Day S, Rose M et al. Low-dose outpatient chemobiotherapy with temozolomide, granulocyte-macrophage colony stimulating factor, interferonalpha2b, and recombinant interleukin-2 for the treatment of metastatic melanoma. J Clin Oncol 2005; 23: 8992–9000. 16. Therasse P, Arbuck SG, Eisenhauer EA et al. New 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. 17. Simon R. Optimal two-stage designs for phase II clinical trials. Control Clin Trials 1989; 10: 1–10. 18. Gaspar L, Scott C, Rotman M et al. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 1997; 4: 745–751. 19. Gupta G, Robertson AG, Mackie RM. Cerebral metastases of cutaneous melanoma. Br J Cancer 1997; 76: 256–259. 20. Fife KM, Colman MH, Stevens GN et al. Determinants of outcome in melanoma patients with cerebral metastases. J Clin Oncol 2004; 22: 1293–1300.
original article