Temozolomide and Radiotherapy for Brain Metastases 

Abstract & Commentary 

Synopsis: The treatment of choice for multiple brain metastases is whole-brain radiotherapy. Patients with medically or technically inoperable solitary brain metastases also receive radiotherapy. Efforts to enhance the effects of radiotherapy via concomitant systemic therapy have been unsuccessful. Investigators in Greece reported their phase 2 trial results showing a significant increase in the response rate of lesions in patients who received combined modality therapy using radiotherapy and temozolomide.

Source: Antonadou D, et al.J Clin Oncol. 2002;20:3644-3650.

Antonadou and colleagues at the metaxas cancer hospital in Piraeus, Greece, randomized 52 patients with previously untreated brain metastases to radiotherapy alone vs radiotherapy with temozolomide. Patients with life expectancies exceeding 3 months were accrued from October 1999 through June 2000. All patients had breast cancer or lung cancer, including either small cell or non-small cell carcinoma (NSCLC). Every patient had brain metastases as documented on CT or MRI scan. Demographic and functional status was well balanced between the 2 arms. Temozolomide was selected because it is an alkylating agent known to readily penetrate the blood brain barrier and may act as a radiosensitizer. Twenty-three patients received RT alone, and 25 were randomized to combined therapy. Two patients from the initial 52 randomized to either group (n = 4) refused treatment, leaving 48 patients. Median age was 61-62 years. Thirty percent of the RT patients and 20% of the combined therapy patients had known metastases elsewhere. Two thirds of patients in each group had NSCLC. Multiple brain metastases were identified in 70% (n = 16) of the RT arm and 76% (n = 19) of the combined-therapy arm.

All patients received 40 Gy whole brain radiotherapy using 12MV photons at 2 Gy per fraction. During RT, the combined therapy patients were given TMZ 75 mg/m2/d, followed by 200 mg/m2/d per 5-day cycle for 6 cycles as consolidation. The lowest dose of steroids necessary to maintain neurologic stability was used throughout RT, and then tapered. Anticonvulsants and anti-emetics were prescribed as needed.

There were 45 evaluable patients. Median follow-up was 4 months. Eighty-seven percent (n = 42) of patients died by the conclusion of the study. No patient received chemotherapy aside from TMZ following RT. Patients were assessed via monthly CT or MRI and neurologic examination. Responses were determined based on patient status at 2-months post-RT, and were confirmed 1 month later. Two radiologists blinded to the randomization centrally reviewed all scans.

Five parameters were used to compare treatment efficacy. Those were the objective response rate on imaging, including complete and partial responses; improvement in neurologic status; diminished reliance on anticonvulsants and anti-emetics; median overall survival; and cause of death. In the RT-alone arm, there were 7 complete responses (33%) and 7 partial responses (33%), along with 5 patients who had stable disease (24%). The combined therapy group exhibited a statistically significantly higher objective response rate (P = 0.017) with 9 CRs (38%) and 14 PRs (58%), along with 1 patient who had stable disease. Despite the higher proportion of combined therapy patients who manifested an improvement in their neurologic status (15% vs 9%), the change in symptomatology was not great enough to declare a statistical difference. In terms of reliance on steroids and anticonvulsants, RT-alone patients used 9% less steroids and 14% less anticonvulsants by the time of their final evaluation, in comparison to 33% less steroids and 40% less anticonvulsants in the combined-therapy arm. Side effects from the TMZ included headache, nausea, vomiting, and mild-to-moderate, reversible, myelosuppression in a small number of patients. Median overall survival was not significantly different between groups at 7 months for the RT-alone arm and 8.6 months for the RT/TMZ arm (P = 0.45). Finally, while 86% of RT alone patients died of systemic disease, so did 90% of combined therapy patients.

Antonadou et al concluded that daily temozolomide was safe and well-tolerated when combined with radiotherapy, and that its use led to a statistically significant improvement in the objective response rate. Because of their poor prognosis, it was difficult to demonstrate an improvement in survival in the patient cohort studied. The role of chemotherapy in patients with brain metastases remains controversial, and phase 3 randomized trials are needed.

Comment by Edward J. Kaplan, MD

Temozolomide is known to penetrate the blood brain barrier, and was recently shown to have anti-tumor activity as a single-agent therapy against brain metastases by Abrey, who published phase 2 results from work done at Memorial Sloan Kettering.1 Patients received 150-200 mg/m2/d for 5 days per 28 day cycle, just as in Antonadou et al’s paper. Overall median survival for the 34 patients assessed was 6.6 months. Forty-one percent of patients achieved a partial response or stable disease. In another phase 2 trial published earlier this year by the Cytokine Working Group,2 31 patients with malignant melanoma metastatic to brain were treated with RT + 75 mg/m2/d TMZ, just as in the Greek study. Additional TMZ was continued for 6 week cycles followed by 10 weeks off. Median survival was 6 months, and toxicities were limited.

The Antonadou paper was an intriguing study because its results look promising, and there were no competing therapies that could cloud the results, such as might be seen in a crossover design. The most obvious criticism is the study’s small size. Even if there was a significant difference in improvement in symptomatology, it would be very difficult to detect with such a tiny sample size. Another issue that comes to mind is whether patients were evenly split between MRI and CT imaging. For example, if more patients in one or the other group were imaged with MRI rather than CT, a skewed CR/PR rate might result since MRI is typically more sensitive to detecting finer detail. In addition, multiple metastases can sometimes be mistaken for solitary metastases on CT, which tends to be less of a problem with MRI. Since TMZ is an oral agent that can be associated with nausea and vomiting, some patients may not have received effective amounts of TMZ despite being in the combined therapy arm. Regarding the RT, it is somewhat unusual to treat the brain with photons as high as 12 million electron volts. Usually, a brain is treated with 6 MV or similar machines. I also feel that the inclusion of patients with solitary brain lesions, who made up more than a quarter of the study population, probably diluted the results because these would seem to be the patients least likely to benefit from a synergistic effect related to TMZ. That is, in general, patients with solitary metastases have fewer symptoms and may have a longer life expectancy than patients with multiple metastases, thus making it harder to show an improvement. Finally, with almost all patients dying of systemic disease, and no patient receiving cytotoxic chemotherapy targeting extracranial disease, it will probably be impossible to demonstrate a survival advantage resulting from better CNS control until the systemic disease is better controlled.

Interestingly, an abstract of a phase 3 trial conducted by the same researchers was presented at the October 2002 ASTRO meeting in New Orleans.3 It was well received by the attendees. The basic format was the same as in the phase 2 trial, except that the RT was delivered as 30 Gy in 10 fractions. One hundred thirty-four patients participated. Again, no survival advantage was demonstrated (8.3 vs 6.3 mo; P = 0.18), but the objective response rate was significantly better in the combined therapy arm (53% vs 33%; P = 0.04).

I am optimistic about the role of TMZ in the management of patients with brain metastases. Although it may take substantially longer to conduct, a trial looking at patients with isolated brain metastases would be the best way to assess the contribution of TMZ to their therapy. 

Dr. Kaplan is Acting Chairman, Department of Radiation Oncology, Cleveland Clinic Florida, Ft. Lauderdale, FL; Medical Director, Boca Raton Radiation Therapy Regional Center, Deerfield Beach, FL.


1. Abrey LE, et al. J Neurooncol. 2001;53:259-265.

2. Margolin K, et al. J Cancer Res Clin Oncol. 2002;128: 214-218.

3. Antonadou D, et al. Int J Radiat Oncol Biol Phys. 2002;54:93-94.