• brain metastases;
  • metronomic therapy;
  • temozolomide;
  • whole-brain radiotherapy


  1. Top of page
  2. Abstract


Temozolomide (TMZ), an oral methylating imidazotetrazinone, has antitumor activity against gliomas, malignant melanomas, and brain metastasis and is presently administered as a 5-day oral schedule every 4 weeks.


A single-institution phase 2 clinical trial was conducted to determine the efficacy and the safety profile of a new regimen based on a dose-intensified, protracted course of TMZ after whole-brain radiotherapy (WBRT). Patients were eligible if they had at least 1 bidimensionally measurable brain metastasis from breast cancer and nonsmall cell lung cancer (NSCLC). Twenty-seven patients were treated with 30 grays (Gy) of WBRT with concomitant TMZ (75 mg/m2/day) for 10 days, and subsequent TMZ at a dose of 75 mg/m2 per day for 21 days every 4 weeks, for up to 12 cycles.


Two complete responses (7.4%) and 11 partial responses (40.7%) were achieved. The schedule appeared to be well tolerated, with grade 3 toxicity (graded according to National Cancer Institute Common Toxicity Criteria) observed in only 2 patients. The overall median survival was 8.8 months and the median progression-free survival was 6 months.


The concomitant use of WBRT and protracted low-dose TMZ appears to be an active, well-tolerated regimen. The observed antitumor activity suggests the need for further investigation of this schedule in combination with other anticancer agents for the concomitant treatment of brain metastases and primary cancers. Cancer 2008. © 2008 American Cancer Society.

Brain metastases (BMs) represent an important cause of morbidity and mortality for breast cancer and nonsmall cell lung cancer (NSCLC) patients. The incidence of central nervous system (CNS) metastases has been reported to be 15% to 25% in patients with breast cancer and 30% to 40% in patients with NSCLC.1–3 BMs continue to have a negative impact on quality of life and prognosis; in fact, the median overall survival is <3 to 6 months.2, 4 The current standard of care for BM involves surgical resection, whole-brain radiotherapy (WBRT), and control of symptoms with steroids and anticonvulsants.1–6

Conversely, the role of chemotherapy in this setting of patients remains controversial because penetration of the blood-brain barrier (BBB) is a limiting factor for both chemotherapy and target-based agents. Chemotherapy with temozolomide (TMZ) has emerged as a possible alternative approach for patients with recurrent brain metastases. TMZ is an oral alkylating agent that crosses the BBB, achieving effective concentrations in the CNS; moreover, it has demonstrated preclinical activity against several cancers.7, 8

Previous studies have reported the use of single-agent TMZ in the setting of recurrent or progressive BM, with modest results. The study of the Hellenic Cooperative Group enrolled 27 patients with BM and achieved a partial response (PR) rate of 4% using a dose of 150 mg/m2 on Days 1 through 5 every 28 days.9 Abrey et al10 enrolled 41 patients in a phase 2 trial and obtained a disease control rate of 41% (PR + stable disease [SD]) with TMZ at a dose of 150 to 200 mg/m2 on Days 1 through 5 every 28 days. Recently, 2 other phase 2 trials confirmed the lack of efficacy of single-agent TMZ in patients with BM from NSCLC and breast cancer, respectively.11, 12

Conversely, the efficacy and safety of using TMZ concurrently with WBRT was evaluated in 3 other trials. Antonadou et al13 reported that WBRT plus TMZ demonstrated a significant improvement in response rate compared with WBRT alone (96% vs 67%), but that study evaluated only the response rate and data regarding survival were lacking. In another randomized trial, the authors demonstrated that the concomitant treatment was well tolerated and resulted in a significantly better progression-free survival (PFS) for patients with BM.14

A recently reported trial performed at our institution demonstrated that concomitant treatment with WBRT and TMZ (at a dose of 150 mg/m2/day for 5 days every 28 days) is well tolerated, achieving an overall response rate of 44% and a disease control rate of 76.3%. We demonstrated a positive impact of this therapy on the quality of life (QoL) of patients with BMs after 3 and 6 months from the initiation of treatment with TMZ. QoL analysis demonstrated a high level of satisfaction among patients who had been treated with TMZ plus WBRT for BM, providing excellent evidence for patient compliance.15

Myelosuppression is the dose-limiting toxicity when TMZ is administered on a 5-day schedule repeated every 4 weeks16; therefore, to our knowledge few chemotherapy trials to date have been performed based on the combination of TMZ and other currently available agents. At our institution, a regimen based on the combination of TMZ and pegylated liposomal doxorubicin was recently evaluated in the treatment of BMs from solid tumors. This schedule induced a promising clinical response rate, with 3 complete responses (CRs) (15.8%) and 2 PRs (10.5%) reported.17 Another combination of capecitabine and TMZ was investigated in a phase 1 study in which 24 breast cancer patients with either newly diagnosed or recurrent BMs were enrolled; the authors reported 1 CR and 3 PRs. Moreover, 50% of patients had minor responses or SD.18

In the current study, we describe a new regimen based on 30 grays (Gy) of WBRT administered with concomitant TMZ (at a dose of 75 mg/m2/day) for 10 days followed by the administration of TMZ at a dose of 75 mg/m2 per day for 21 days every 4 weeks, for up to 12 cycles. This schedule was designed to use an intensive and protracted (metronomic-like) TMZ dosing schedule that could both induce the described favorable effects of continuous TMZ administration and allow the concomitant use of a second cytotoxic agent for the primary tumor. To our knowledge, this phase 2 trial represents the first study of a protracted, single-agent TMZ schedule in patients with BM from breast cancer and NSCLC.


  1. Top of page
  2. Abstract

Patient Eligibility Criteria

Patients with histologically or cytologically confirmed breast cancer and NSCLC and inoperable BMs were eligible for the study. Patients were eligible if they had at least 1 bidimensionally measurable lesion measuring >20 mm in the largest dimension and assessable by contrast-enhanced computed tomography (CT) scan or gadolinium-enhanced magnetic resonance imaging (Gd-MRI). Patients were required to age ≥18 years and to have a life expectancy of >12 weeks and a Karnofsky performance score (KPS) >60. Other eligibility criteria included adequate hematologic function with a platelet count >100,000/mm2 and hemoglobin >8 g/dL, renal function with a creatinine level ≤2 mg/dL, and adequate liver function with an aspartate aminotransferase level <1.5 × the upper limit of normal. Prior adjuvant hormonal therapy, chemotherapy, or radiotherapy as well as prior regimens for metastatic disease were withdrawn at least 30 days prior to enrollment in the study. Patients who had received prior treatment for BMs, those with severe intercurrent medical illness or symptomatic heart disease, or those who were pregnant or lactating were ineligible. Eligible patients were required to have fully recovered from previous therapy. The Institutional Ethics Committee approved the protocol and patients were required to provide informed consent before beginning the treatment.

Treatment Schedule

Planned conventional WBRT was administered with a daily dose of 3 Gy × 5 days each week for 2 weeks for a total dose of 30 Gy. Planned chemotherapy treatment consisted of TMZ at a dose of 75 mg/m2 administered orally during radiotherapy, followed by 4 weeks off therapy and a subsequent TMZ administration at 75 mg/m2 on Days 1 to 21 every 4 weeks for up to 12 additional cycles. Patients were treated until evidence of progression of BMs.

Dexamethasone and opioids were administrated at the lowest dose needed to preserve neurologic stability. Anticonvulsant agents were only used in patients with seizures.

Patient Evaluation

Pretreatment evaluation included a complete medical history, physical examination, determination of KPS, and hematologic and biochemistry evaluation within 7 days of registration. Evaluation during treatment consisted of physical examinations on Day 1 of each cycle; weekly blood counts were performed for Cycle 1 and on Days 1 and 15 thereafter. Biochemistry was performed on Day 1 of each cycle. The initial examinations had to be performed within 30 days before inclusion in the study. Target lesions were assessed by CT or Gd-MRI. Evaluation of target lesions was performed every 3 cycles according to the World Health Organization (WHO)/ Eastern Cooperative Oncology Group (ECOG) criteria15 as described below.

A CR required the disappearance of all known brain metastases. A PR required a ≥50% decrease in measurable brain lesions or an objective improvement in evaluable, but nonmeasurable brain lesions. It was not necessary for every brain lesion to have regressed, but no brain lesion should have progressed. SD indicated that brain lesions remained unchanged (<50% decrease or <25% increase in the size of measurable lesions). Progressive disease (PD) indicated progression of some or all brain lesions and/or the appearance of new brain lesions. No brain lesion regressed.

Total body CT and/or other examinations if needed were also repeated every 3 cycles for the evaluation of extracranial and systemic disease; all scans were centrally reviewed by 2 blinded radiologists and systemic disease was evaluated according to Response Evaluation Criteria in Solid Tumors (RECIST) criteria.19 The patients who recorded a systemic PD were censored in the study.

All adverse events were recorded and graded according to the National Cancer Institute Common Toxicity Criteria. Neurologic functional status was assessed by 2 examiners at baseline and every 3 cycles as described previously by Antonadou et al.13 In brief, the functional status was assessed by 2 examiners as Level I, fully functional; Level II, fully functional, not able to work; Level III, stays in bed and needs help half the time; and Level IV, requires help all the time.

Statistical Analysis

The primary objective of the current study was to assess the activity, measured as objective BM response, to TMZ given orally on Days 1 to 21 every 4 weeks in 27 evaluable patients with measurable BM from breast cancer and NSCLC. The secondary objectives were to determine overall survival (OS), PFS, and the toxic effects of TMZ when administered in this fashion. Other objectives were assessement of functional status and the amount of symptomatic drug assumption. PFS was measured from the date of registration to the date of documented PD or death. OS was assessed from the time of registration to the date of death resulting from any cause. PFS and OS were both determined by the Kaplan-Meier product limit method.20 A 2-stage Simon accrual design was adopted for this phase 2 trial.21 The minimum target activity level was 20% and early discontinuation of the study was planned in case of no response in the first 12 assessable patients.


  1. Top of page
  2. Abstract

Patient Characteristics

Thirty-two patients with BM from either breast cancer or NSCLC were assessed for eligibility from November 2005 through December 2006. Three patients did not meet the inclusion criteria, 1 patient refused to participate, and 1 patient did not receive the first treatment although he agreed to enter the study. Clinical and demographic characteristics of the remaining 27 patients are listed in Table 1. The average age was 55 years (range, 46-76 years) there were 14 women and 13 men, and 7 patients had KPS <70. Among 27 assessable patients, 15 (55%) had NSCLC and 12 (45%) had breast cancer; 20 had received prior systemic chemotherapy, and 11 had received previous radiotherapy (in extrabrain sites). All patients had BM and 21 (78%) had ≥2 BM; 15 patients (55%) had at least 1 other metastatic site.

Table 1. Patient Characteristics (n=27 Evaluable Patients)
  1. NSCLC indicates nonsmall-cell lung cancer; KPS, Karnofsky performance status; BMs, brain metastases.

 Age, y  
Histologic type  
Previous surgery2075
Previous chemotherapy2074
Previous radiotherapy1244
Site of metastases  
 Lymph nodes1141
No. of BMs  

A total of 114 cycles were administered; the median number of cycles administered was 4.2.

Clinical Activity

Global responses, evaluated every 3 cycles, are summarized in Table 2. Two CRs were achieved in patients with BM from NSCLC and breast cancer, respectively, and lasted 12 months. Both patients had only BMs and underwent surgery for primary cancer. Eleven PRs were recorded in 6 patients with breast cancer and 5 with NSCLC; 6 of these patients remained in PR for up to 6 cycles. Eight of 11 patients who obtained a PR had only BMs. Another 6 patients remained in SD for up to 3 cycles, 3 for up to 6 cycles, and 8 patients demonstrated PD during the treatment. Of these 8 patients 2 were in SD for brain lesions and in PD for visceral sites of disease. At the time of PD they stopped the administration of low-dose TMZ and began a specific treatment for the control of visceral disease. However, they also developed PD at brain sites after 1 month from the suspension of TMZ and the recalculation of time to progression (TTP) of all the patients enrolled in the study considering only PD in the brain gave a median TTP of 8.9 months (which was not significantly different from the previous analysis).

Table 2. Response to Treatment of Brain Metastases
ResponsesNo. of Patients%
  • NSCLC indicates nonsmall-cell lung cancer; CR, complete response; PR, partial response.

  • *

    The responses were also classified according to the primitive tumor from which the brain metastases originated.

Complete responses27.4
Primitive tumor  
 Breast cancer1
Partial responses1140.7
Primitive tumor*  
 Breast cancer6
Objective responses [CR+PR]1348.1
Stable disease622.2
Primitive tumor  
 Breast cancer3
Progressive disease829.7
Primitive tumor  
 Breast cancer2

Another 6 patients remained in SD for up to 3 cycles, 3 for up to 6 cycles, and 8 patients demonstrated PD during the treatment.

As shown in Table 2, the odds ratio (OR) was 48.1% (95% confidence interval [95% CI], 38.7%-56.9%), whereas the disease control rate (CR + PR + SD) was 70.3% (95% CI, 61.7%-82.4%). The overall response rate was 58% (7 of 12 patients) in breast cancer patients and 40% (6 of 15 patients) in NSCLC patients. According to stratification by Radiation Therapy Oncology Group (RTOG) recursive partitioning analysis (RPA), which considers multiple prognostic factors, class I included 11 patients (41%), class II included 10 patients (37%), and class III included 6 patients (22%). Interestingly, in the RPA Class I, 9 (82%) OR and 2 (8%) SD were obtained; conversely, among the 6 patients (22%) included in RPA Class III, no PR + CR was observed (Table 3).

Table 3. RPA Classification Prior to Treatment of the Patients Who Responded or Were Resistant to Treatment at Brain Sites
ResponsesRPA Class I (11 Patients)RPA Class II (10 Patients)RPA Class III (6 Patients)
  1. RPA indicates recursive partitioning analysis.

Complete response2
Partial response74
Stable disease222
Progressive disease44

The median TTP was 6 months (95% CI, 5.1-6.8 months) (Fig. 1A) and the median OS was 8.8 months (95% CI, 6.8-8.9 months) (Fig. 1B). The 1-year survival rate was 18.5% (95% CI, 8.5-22.8%).

thumbnail image

Figure 1. Kaplan-Meier curves of (A) overall survival and (B) and time to disease progression in patients with brain metastases.

Download figure to PowerPoint

Toxicity Evaluation

All patients were evaluable for toxicity. Toxicity was recorded using the National Cancer Institute Common Toxicity Criteria recommendations for the grading of acute and subacute toxicity. TMZ was generally well tolerated and the main toxicities observed were hematologic (Table 4). The toxicities were generally between grade 1 or 2 in severity, although 2 patients had grade 3 events with neutropenia and anemia, respectively. The most common drug-related nonhematologic toxicities were nausea in 9 patients (37%), vomiting in 6 patients (25%), and headache in 5 patients (21%). No liver, renal, or cardiac toxicities were observed in the current series. Treatment interruptions because of toxicity were not observed in our series, and no patients required dose reductions.

Table 4. Number of Adverse Events Recorded Among the Entire Study Classified for Toxicity Grade Accordingly to the NCI-CTC and Number of Patients in Which Each Adverse Event Was Recorded
Adverse EventGrade 1Grade 2Grade 3Grade 4No. of Patients (%)
  1. NCI-CTC indicates National Cancer Institute Common Toxicity Criteria.

 Neutropenia7319 (33)
 Anemia93110 (37)
 Thrombocytopenia537 (26)
 Nausea759 (33)
 Vomiting536 (22)
 Constipation32 (7)
 Diarrhea22 (7)
Infection without neutropenia111 (4)
 Headache535 (29)
 Myalgia112 (7)
 Neuropathic pain1
 Epistaxis212 (7)
 Petechiae/purpura112 (7)

Evaluation of Neurologic Function

Another measure of treatment efficacy was the assessment of neurologic function at baseline and after 3 and 6 cycles of treatment. The effects of treatment on both neurologic function and BM size are listed in Table 5. After 3 cycles of treatment, the percentage of patients in Level I (fully functional) and Level II (fully functional but not able to work) increased from 75% to 85%, whereas the number of patients with Level III status decreased from 25% to 15%.

Table 5. Neurologic Function Assessment Performed on Evaluable Patients During the Study*
TimeLevel I No. of Patients (%)Level II No. of Patients (%)Level III No. of Patients (%)
  • *

    The total number of patients was not the same during the study due to progressive disease developing in some patients.

Baseline>6 (25)>12 (50)>6 (25)
3 cycles>10 (42.5)>10 (42.5)>4 (16)
6 cycles>6 (32)>9 (47)>4 (21)

The efficacy of this treatment was confirmed by the decreased requirement for medication to palliate neurologic symptoms. At baseline, 15 (55%) patients required corticosteroids, 7 (26%) required anticonvulsants, and 8 (29%) required opioids. After 2 months, 4 months, and 6 months of TMZ treatment, these percentages progressively decreased. In fact, at 6 months, 7 (26%) patients required corticosteroids, 4 (15%) required anticonvulsants, and 3 (11%) required opioids. These data confirm the clinical benefit given by the schedule used in the current study in this subset of patients.


  1. Top of page
  2. Abstract

Tumors most commonly metastasizing to brain include those from the lung and breast. Although WBRT is still a standard and active treatment in the management of BM, recent trials of TMZ and WBRT suggest a significant increase in response rate, especially for metastases from lung cancer.22, 23 Unfortunately, using standard schedules (150-200 mg/m2/day for 5 days, every 28 days), TMZ produced a low response rate at other visceral sites derived from either breast cancer or NSCLC.10, 24 In a previous study, we obtained a 45% overall response rate using a schedule containing TMZ at a dose of 150 mg/m2 per day for 5 days, with toxicities including grade 3 neutropenia (15%) and anemia (13%), even if it was safe and tolerable.15 Conversely, the combination of TMZ with other chemotherapy agents (ie, cisplatin) induced a considerable toxicity (most likely because of the addition of other cytotoxic drugs more than because of TMZ itself).11 In another study, 18 patients with NSCLC and BMs received aggressive treatment with WBRT and chemotherapy with vinorelbine, ifosfamide, and cisplatin; they achieved an objective brain response rate of 56%, but all patients reported severe grade 4 neutropenia.25

The possibility of combining TMZ with other agents is shown by the lowering of TMZ dosages that can reduce the occurrence of its side effects. In this view, metronomic chemotherapy may be an alternative to conventional chemotherapy because it provides several favorable effects: 1) delay of the onset of acquired drug resistance; 2) reduction of host toxicity; 3) lack of prolonged drug-free break; and 4) antiangiogenic effects.26 In fact, several studies reported that many cytotoxic drugs, at continuing lower doses, act on the endothelial cell compartment of tumors, inducing a potent antiangiogenic effect.27 In addition, TMZ has been used in protracted low-dose schedules and has been demonstrated to induce antiangiogenic effects in both in vitro and in vivo models.28, 29 In fact, in a preclinical pharmacokinetic and pharmacodynamic study, Zhou et al30 demonstrated that both TMZ metronomic and conventional regimens have similar activities. Using real-time polymerase chain reaction and Western blot analyses, some differences were observed in the serum levels of vascular endothelial growth factor and hypoxia-inducible factor-1‒α, suggesting that the metronomic regimen may be superior to the conventional regimen by preventing the progression of tumors to a proangiogenic state.

Another advantage of a protracted administration of TMZ, even at relatively low daily doses, is shown by a significant and prolonged depletion of the enzyme O6-alkylguanina-DNA alkyltransferase (AGAT) activity, which could enhance the antitumor activity of the agent.31 To our knowledge to date, few clinical studies exist regarding low-dose TMZ administration. A 33% response rate was reported in patients with recurrent brain tumors (melanoma and glioma) who were treated with TMZ at a dose of 75 mg/m2 on a daily schedule for 6 to 7 weeks, which is similar to the response rate reported in patients treated with TMZ on its approved dose schedule.32

A phase 1 trial investigated the combination of TMZ at a dose of 150 mg/m2 per day administered every 2 weeks rather than the conventional 4 weeks (Days 1-7 and 15-21 every 4 weeks), and vinorelbine at escalating doses. A total of 21 patients with recurrent BM were enrolled, with 2 responses (1 PR and 1 minor) and 6 SD with a median survival of 27 weeks.33

In the current study, using a protracted low-dose TMZ schedule, we obtained a promising objective response rate of 48% (13 of 27 patients). A possible explanation for these results is that in the current study, 11 of 27 patients (40%) were included in first RPA class according to RTOG classification,34 and an objective response rate of 82% and 2 (8%) SDs were obtained in this group; conversely, among the 6 patients (22%) included in the third RPA class, we did not observe any objective response. Because both approximately half of the patients included in the study had the brain as the only site of distant metastases and 40% were in RPA class 1, selection bias could have occurred in our series of patients.

The efficacy and safety of TMZ administered concurrently with WBRT for patients with newly diagnosed BMs was also evaluated in recent trials. The largest of these trials demonstrated a significant improvement in the response rate and a trend toward improved survival in the combined therapy arm (8.3 months vs 6.3 months; P = .179); however, it was not statistically significant.35 These results are completely comparable to those of the current study, in which we obtained a median OS of 8.8 months. Conversely, in the current study we used a more intense dose of TMZ per cycle than any previous experience (1575 mg/m2 compared with 750 mg/m2), without any serious toxicity reported.

Although the response rates after WBRT vary, CRs or PRs have been documented in >60% of patients in randomized controlled studies conducted by the RTOG. A revision of 8 different randomized studies in patients with BMs who received WBRT alone demonstrated median survivals ranging from 2.4 to 4.8 months.23 This palliative approach has not translated into a survival benefit.36 In fact, although better than in patients from whom radiotherapy was withheld, the median survival was still poor and the majority of patients still died as a direct result of their brain disease. For example, the median survival for patients with metastatic breast cancer who received WBRT was in the range of 3 to 4 months, with 1-year survival rates at 10% to 20%.37

The addition of protracted courses of TMZ to WBRT did not appear to change the well-documented improvement in neurologic function that can be achieved with radiotherapy alone; the use of symptomatic drugs such as corticosteroids and anticonvulsant agents decreased during the treatment, confirming the activity of this schedule.

To the best of our knowledge, this is the first report specifically discussing the use of protracted low-dose TMZ for 21 days for the treatment of patients with BMs from breast cancer and NSCLC. The results of the current study demonstrate that the combination of this new TMZ schedule and WBRT is active and induces few side effects. The durable responses, favorable TTP, and the improvement in neurologic status observed in this study offer several advantages over conventional therapy, encouraging the design of additional schedules based on the combination of TMZ and other chemotherapeutic agents in the treatment of breast cancer and NSCLC that are metastatic to the brain. The combination with other agents could allow for the control of extrabrain disease sites that are poorly susceptible to TMZ.


  1. Top of page
  2. Abstract