Temozolomide rechallenge in recurrent malignant glioma by using a continuous temozolomide schedule

The “rescue” approach


  • James R. Perry MD,

    Corresponding author
    1. Crolla Family Brain Tumour Research Unit, Department of Medicine, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Ontario, Canada
    • Department of Medicine, University of Toronto, Sunnybrook Health Sciences Centre, A402, 2075 Bayview Avenue, Toronto, Ontario, Canada M4N 3M5
    Search for more papers by this author
    • Fax: (011) 416-480-5054

  • Philippe Rizek,

    1. Crolla Family Brain Tumour Research Unit, Department of Medicine, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author
  • Rosemary Cashman RN,

    1. BC Cancer Agency, Vancouver, British Columbia, Canada
    Search for more papers by this author
  • Meredith Morrison RN,

    1. Crolla Family Brain Tumour Research Unit, Department of Medicine, Sunnybrook Health Sciences Center, University of Toronto, Toronto, Ontario, Canada
    Search for more papers by this author
  • Tara Morrison MD

    1. Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, Pennsylvania
    Search for more papers by this author



Despite advances in first-line therapy, there are few data on treatment of glioblastoma multiforme (GBM) at recurrence. Temozolomide (TMZ) is well tolerated and may have activity despite prior TMZ exposure if novel dose schedules are used.


The authors reviewed their experience with a continuous TMZ schedule (50 mg/m2 daily), given at progression after conventional 5-day TMZ. Patients were reported in 3 groups: 1) GBM after progression on conventional TMZ; 2) GBM at first recurrence after completion of standard concomitant and adjuvant TMZ; and 3) patients with other anaplastic gliomas at second relapse on conventional TMZ.


In Group 1, 21 patients with a median age of 54 years (range, 33 years-68 years) received a median of 3 cycles (range, 2-12 cycles) of continuous TMZ at 50 mg/m2. Overall clinical benefit (complete response, partial response, and stable disease) was 47%, with 6-month progression-free survival (PFS) of 17%. In Group 2, 14 patients with GBM, median age 52 years (range, 38 years-62 years) received continuous TMZ at progression after initial TMZ/radiotherapy (RT) and adjuvant TMZ. The median interval after adjuvant TMZ was 3 months (range, 2 months-10 months). A median of 5 cycles of TMZ was given, and 6-month PFS was 57%. In Group 3, 14 patients with a median age of 49 years (range, 34 years-56 years) received continuous TMZ; 2 partial responses and 6 with stable disease were seen, with a 6-month PFS of 42%. Toxicities were mild and well tolerated; lymphopenia was common but no serious opportunistic infections were identified.


Although retrospective, our results demonstrate that continuous daily administration of TMZ is an active regimen despite prior TMZ therapy. The excellent tolerability of this regimen may allow future combination with other alkylating agents or with novel therapies. Cancer 2008. © 2008 American Cancer Society.

Temozolomide (TMZ) is an oral alkylating agent indicated for the treatment of malignant glioma at relapse and, more recently, as part of the standard of care for patients with newly diagnosed glioblastoma multiforme (GBM). When administered daily, concurrent with radiotherapy (RT) and followed by adjuvant treatment, temozolomide improves 2-year survival of patients with newly diagnosed GBM from 11.2% to 27.3%, and 4-year suvival from 3.8% to 12.9%.1, 2 The benefit of TMZ in patients with GBM has been correlated with epigenetic silencing of the DNA repair enzyme 06-methylguanine-DNA-methyltransferase (MGMT) in tumor tissue retrieved at the initial surgical diagnosis,3 and MGMT status may be an important predictor of prolonged survival.4 Despite these advances, GBM remains an essentially incurable disease and improvements in both front-line and salvage therapies are urgently needed.

There is no consensus as to the optimal second-line therapy for patients with recurrent malignant glioma, although lomustine, procarbazine, etoposide, carboplatin, tamoxifen, and irinotecan are all commonly considered and used.5 Unfortunately, these therapies have a limited response rate and, even when effective, the durability of response is modest.

There is growing interest in the use of therapies targeted at the robust and aberrant vasculature in malignant glioma. Metronomic chemotherapy, a term used to describe chemotherapy delivered on a continuous or near-continuous basis, but below the typical maximum tolerated dose, is believed to function through an antiangiogenic mechanism.6 Lower daily doses of chemotherapy on a protracted schedule may inhibit endothelial cell recovery and restrict mobilization and activity of bone marrow-derived circulating endothelial precursors, thus contributing to antiangiogenic activity.7 Few clinical trials have explored metronomic chemotherapy in the setting of malignant glioma.

Temozolomide appears to be an ideal drug for metronomically delivered regimens. It is an oral agent, well tolerated, and can be given in a variety of dosing schedules. For example, in addition to the conventional approved dosing of 150 mg/m2 to 200 mg/m2 given 5 days of each 28-day cycle, novel schedules such as 75 mg/m2 given 21 of 28 days and 150 mg/m2 7 days on/7 days off have been described. These schedules allow greater TMZ dose intensity per cycle of delivery; however, these are not continuous regimens and may not be truly metronomic, as they may allow endothelial and tumor cell recovery during treatment gaps. In vitro studies of continuously delivered TMZ found 10-fold increased sensitivity of endothelial cells compared with glioma cells in vitro.7 Both conventional and continuously delivered TMZ have been shown to have similar pharmacokinetic properties, linear kinetics, and similar cytotoxic activity when tested in xenograft models of glioma.8

Short-term continuous delivery of TMZ (75 mg/m2 daily for 6 weeks) has been shown to be effective and well tolerated during initial therapy when combined with fractionated external beam RT for GBM.1 After completion of this chemoradiation regimen, patients are offered the conventional adjuvant TMZ 150 mg/m2 to 200 mg/m2 regimen on 5 days of each 28-day cycle. At the time of progression of disease, either during or after completion of adjuvant treatment, patients are usually offered treatment other than TMZ under the assumption that drug resistance to TMZ is present.9 The concept of rechallenging a patient at recurrence with the same chemotherapy agent is new in neuro-oncology, but has some precedent in other diseases such as ovarian and colorectal cancer.10, 11 If the available alternative schedules of TMZ work through different mechanisms of action than conventional TMZ, it may be possible to offer TMZ rechallenge in patients with disease progression during or after their initial therapy.

The rationale for TMZ rechallenge in previously TMZ-treated patients includes not only the previously discussed potential antiangiogenic mechanism of action of continuously delivered drug, but also the potential for novel schedules to overcome TMZ resistance. One of the principal mechanisms of TMZ resistance is the MGMT repair pathway. TMZ induces an alkylating lesion at the O-6 position of methylguanine and, left unrepaired, causes cytotoxicity and apoptosis. High levels of intracellular MGMT inhibit the therapeutic effect of TMZ by removing the methyl group resulting from temozolomide alkylation. However, in the process of DNA repair, MGMT is consumed as a so-called “suicide protein.” The MGMT pathway therefore represents a target for therapeutic modulation; perhaps especially for patients in whom epigenetic silencing of the MGMT promoter region is not present, and high levels of MGMT protein remain expressed.12 Novel protracted schedules of administration of TMZ deplete MGMT activity detected in peripheral blood mononuclear cells13; thus, these chemotherapy schedules may overcome TMZ resistance and restore chemosensitivity and cytotoxic damage in previously resistant cells.

Another rationale for the use of alternative TMZ regimens is dose intensity. The conventional 5-day TMZ regimen will deliver 750 mg/m2 to 1000 mg/m2 per month, whereas other schedules allow higher, well-tolerated dose delivery. For example, the 21-day schedule delivers 1575 mg/m2 per month, and the 7 day on/off schedule delivers 2100 mg/m2 per month. In this study, we report our experience with a 50-mg/m2 schedule, delivered daily and continuously throughout therapy; the dose intensity of this schedule is 1400 mg/m2 per 28 days and therefore delivers 1.4 to 1.9 times more TMZ than the conventional regimen, in addition to having potential antiangiogenic and MGMT-depleting activity.

We reviewed our experience with continuous TMZ, delivered at the time of progression after failure of conventional 5-day TMZ in patients with malignant glioma, to determine the potential efficacy and tolerability of this regimen for further evaluation in a prospective study.


Research ethics board approval was obtained for the purposes of a retrospective chart review of patients treated with TMZ at Toronto-Sunnybrook Regional Cancer Centre between January 2001 and July 2005. We identified all patients treated with the continuous 50-mg/m2 daily regimen and extracted demographic, treatment, and outcome details. Patients were given TMZ at 50 mg/m2 daily, rounded to the nearest 5 mg and seen monthly in clinical follow-up. Dose reduction by 25% was planned for grade 3 or 4 myelotoxicity. Imaging with brain computed tomography or magnetic resonance imaging was obtained after every 2 cycles of therapy. Responses were recorded according to the method of Macdonald et al,14 and duration of response was determined as time from the start of therapy to the time of progression. Antiemetic and infectious prophylaxis was not routinely prescribed.

We grouped patients into 3 categories. The first group was composed of patients with histologically confirmed GBM who had surgery and conventional external beam RT, and were followed without concurrent or adjuvant chemotherapy until their first recurrence. These patients were treated before local adoption of the TMZ chemoradiation strategy. At first recurrence, they received conventional TMZ given as 150 mg/m2 to 200 mg/m2 5 days of each 28-day cycle. When these patients progressed, during conventional TMZ treatment, they were offered a change to TMZ in the 50-mg/m2 daily regimen.

The second group was composed of patients with GBM who received surgery, external beam RT combined with concurrent TMZ, and then adjuvant TMZ for a complete 6 cycles. At first progression these patients were offered rechallenge with continuous TMZ at 50 mg/m2. The interval between discontinuation of adjuvant TMZ and time of relapse was recorded.

The third group consisted of patients with anaplastic gliomas (anaplastic astrocytoma, anaplastic oligodendroglioma, or anaplastic oligoastrocytoma) treated at the time of second relapse during conventional TMZ treatment. This group underwent surgery and RT, and then was followed until relapse, when they received TMZ at 150 mg/m2 to 200 mg/m2 on 5 days of each 28-day cycle. At the time of progression, during conventional TMZ treatment, they were offered continuous 50-mg/m2 TMZ treatment.


Patients With Recurrent GBM at Second Relapse While Receiving Conventional TMZ Chemotherapy

There were 21 patients in this group of patients with GBM who were treated with initial RT alone before adoption of the RT/TMZ strategy. Median age was 54 years (range, 33 years-68 years), 11 had biopsy alone, and 10 had subtotal surgical resections. All received 50 Gray to 60 Gray external beam RT with conventional fractionation. Patients experienced their first progression at a mean of 4.5 months (range, 3 months-15 months) after completion of RT. At progression, 150 mg/m2 to 200 mg/m2 TMZ was delivered on 5 days of each 28-day cycle for a median of 3 cycles (range, 2-12 cycles) before second progression was confirmed on imaging studies. These patients were then treated with temozolomide at 50 mg/m2 daily until progression. The overall clinical benefit rate was 47% with no complete responses, 3 partial responses (14.3%), and 7 patients having stable disease (33%) after at least 2 cycles of therapy. Six-month progression-free survival (PFS) after second relapse was 17%.

Patients With Recurrent GBM at First Progression After Concurrent Chemoradiation With TMZ and Adjuvant TMZ

There were 14 patients in this group. Median age was 52 years (range, 38 years-62 years), 9 had subtotal resections, and 5 had a biopsy to establish the diagnosis of GBM. All received concurrent TMZ with RT followed by 6 cycles of conventionally administered adjuvant TMZ and had at least stable disease at the time of treatment completion. Patients were followed with serial neuroimaging studies until first progression, at which time treatment with TMZ at 50 mg/m2 given continuously was offered. The median number of months since completion of adjuvant TMZ was 3 months (range, 2 months-10 months). The median number of cycles of continuous TMZ delivered was 5 months, and we observed 2 (14%) partial responses and 9 (64%) instances of stable disease. Six-month PFS from the time of first relapse was 57%.

Patients With Recurrent Anaplastic Glioma at Second Relapse While Receiving Conventional TMZ Chemotherapy

There were 14 patients in this group; 3 had anaplastic astrocytoma, 4 had mixed anaplastic oligoastrocytomas, and 7 had anaplastic oligodendrogliomas. Six of the 7 patients with anaplastic oligodendrogliomas had loss of heterozygosity of the critical regions of chromosomes 1p and 19q. Median age of these patients was 49 years (range, 34 years-56 years); all were treated with surgery (11 subtotal resections, 3 biopsy alone) and initial RT. No adjuvant chemotherapy was used. Patients were followed until first progression, at which point they received adjuvant chemotherapy with conventional TMZ at 150 mg/m2 to 200 mg/m2 given on 5 days of each 28-day cycle. The median number of conventional TMZ cycles administered was 6 (range, 4-12). At the time of second progression, while still on conventional temozolomide, a change to continuous treatment with TMZ at 50 mg/m2 was offered. We observed 2 (14%) partial responses, and stable disease in 6 (43%), for an overall clinical benefit rate of 57%. Six-month PFS from the time of second relapse was 42%.


No nonhematological toxicities were detected during continuous TMZ treatment. Hematological toxicity, in particular lymphopenia, was frequent (Table 1). Twelve (24%) patients had grade 1, 14 (29%) patients had grade 2, and 5 (10%) patients had grade 3 lymphopenia. No patients received antibiotic prophylaxis, and no serious opportunistic infections occurred. Six patients received treatment for oral thrush during continuous TMZ treatment, and 2 patients were recorded as having a herpes zoster exacerbation.

Table 1. Hematological Toxicity Observed During Treatment With Continuous Temozolomide at 50 mg/m2
ParameterToxicity GradeNo. (%) of Patients
  1. ALC indicates absolute lymphocyte count; ANC, absolute neutrophil count.

Lymphopenia, ALC112 (24)
 214 (29)
 35 (10)
Neutropenia, ANC13 (6)
 22 (4)
Thrombocytopenia, platelet count14 (8)
 21 (2)


The ease of administration and tolerability of TMZ makes this drug suitable for testing in a wide variety of schedules. Our retrospective study of patients treated with continuous TMZ at 50 mg/m2 demonstrates that this novel “metronomic” schedule is well tolerated and has activity both in patients refractory to conventionally delivered TMZ and when used as a rechallenge after a TMZ treatment break.

Rechallenge with TMZ at the time of tumor progression has been previously reported in small case series. For example, Franceschi et al retrospectively identified 14 patients with various types of glioma with a history of response to conventional 5 of 28-day TMZ.15 At the time of recurrence, their patients were rechallenged with the same 5 of 28-day regimen at a median interval of 12 months (range, 4 months-39 months) after the conclusion of treatment. They observed an overall clinical benefit rate of 43% and 6-month PFS of 36%. These results are comparable to our similar group of patients rechallenged with dose-dense metronomic TMZ, although our response rate (64%) and 6-month PFS (57%) were higher. Jauch et al recently reported their experience with 54 patients, half with GBM, who were rechallenged with a variety of TMZ schedules (35 with 5 of 28 days, 7 with 7 days on/off, 4 with 21 of 28 days, and 7 with continuous treatment).16 The overall clinical benefit rate in GBM patients was 65%, and median time to progression 20 weeks. Of patients who progressed during initial conventional TMZ therapy, 8 of 12 responded to administration of 1 of the alternative TMZ schedules.

Although many alternative TMZ schedules appear to be well tolerated and have activity against malignant glioma, the optimal dosing regimen is unclear. The continuous TMZ regimen reported in our study is of interest for several reasons. Protracted TMZ regimens may overcome MGMT-mediated drug resistance and offer a higher dose intensity per month of delivery, and the continuous regimen reported in our study may add the potential antiangiogenic activity of metronomically delivered chemotherapy.

Kong et al described 12 consecutive patients offered daily TMZ at a dose of 40 mg/m2 at the time of first progression after initial therapy.17 Their patients previously received RT with or without radiosurgery boost followed by a variety of chemotherapy regimens, including exposure to TMZ in conventional 5 of 28-day cycles. They observed 2 patients with a partial response after changing to the continuous schedule and a median PFS of 3 months. No grade 3 or 4 toxicities were observed in their study.

TMZ is a generally well-tolerated medication. In conventional dosing schedules, clinically significant hematological toxicity is uncommon and usually easily managed. Our experience with continuous TMZ reveals lymphopenia to be the most common toxicity with this regimen. Over half of our patients experienced some degree of lymphopenia, although no serious opportunistic infections were encountered. Immunosuppression is a common problem in patients with malignant glioma, likely because of multifactorial and converging risk factors such as the presence of the disease itself, prolonged use of corticosteroid therapy, and the use of TMZ early in the course of disease. TMZ is associated with CD4+ specific lymphopenia, and the degree of iatrogenic immunosuppression caused by TMZ requires careful monitoring of these patients, who are at risk for the complications of immune deficiency.18 Although it is used in some centers, we did not routinely offer prophylaxis for infections such as Pneumocystis carinii pneumonia, as very few cases have occurred in patients from our population base.

Our retrospective series has several limitations, including potential issues with selection bias, and incomplete data from the clinical records. We also did not test for MGMT promoter methylation or other biomarkers associated with response and improved prognosis. As this is not a uniform patient population, the reported rates of response PFS are uncontrolled observations. Nonetheless our experience suggests that a 50-mg/m2 regimen of daily TMZ has activity in the setting of recurrent TMZ-treated malignant glioma, including GBM. We saw responses and promising PFS in patients with GBM who were rechallenged with continuous TMZ after a treatment break, and we saw responses in patients who progressed while on the 5 of 28-day regimen. Continuous administration of TMZ at 50 mg/m2 daily may act by depleting tumor MGMT, a hypothesis that must be tested in a prospective setting, ideally with correlation of serum MGMT activity and direct measurement of MGMT expression in tumor specimens. Continuous TMZ may also work through an antiangiogenic mechanism, and this may be an advantage when compared with the other common alternative schedules, because these others are noncontinuous and may allow endothelial and tumor cell recovery between cycles. Correlation of TMZ activity in the daily schedule with markers of angiogenesis (such as serum vascular endothelial growth factor or measurement of circulating endothelial precursors) would lend support to this hypothesis.

TMZ is the drug of choice for malignant gliomas such as anaplastic astrocytoma at recurrence, and is standard first-line treatment for glioblastoma multiforme. Optimal second-line therapy after TMZ failure in these patients has not been defined. The response rates and 6-month PFS in our patients treated with continuous TMZ are similar to other conventional palliative therapies such as lomustine.

Continuous daily TMZ administration appears to be safe and well tolerated. Alternative dosing regimens of TMZ may well have a role as second-line therapy of choice after TMZ failure and, given the tolerability of the regimen, this schedule can be tested in combination with other cytotoxic and novel therapies. These results are part of the rationale for an ongoing Canadian Brain Tumor Consortium 120-patient prospective phase II study testing daily TMZ in patients with recurrent GBM during or after standard chemoradiation.