|
The combination of bevacizumab and irinotecan continues to show significant activity against recurrent malignant gliomas, judging from phase II trial results.
Dr. James J. Vredenburgh, the lead investigator, reported 15 of 35 (43%) patients with grade 4 (glioblastoma multiforme) and 20 of 33 patients (61%) with grade 3 disease were progression free at 6 months. Median progression-free survival was 23 weeks and 42 weeks, respectively.
At 12 months, 13 of the grade 4 patients (37%) and 19 of the grade 3 patients (58%) were still alive. Median overall survival was 40 weeks and 60 weeks, respectively. All told, 59% (40 of 68) of patients in the single-arm study had at least a partial response.
Historically, chemotherapy response rates run about 10% in advanced glioma, according to Dr. Vredenburgh of the Preston Robert Tisch Brain Tumor Center at Duke University in Durham, N.C.
Without comment on the implications, he reported that investigators took 12 patients off treatment after positron emission tomography (PET) scans showed their tumors to be hypometabolic. "They had a cold PET scan," he explained afterward in an interview. "Given the toxicity and the risk involved, we thought that was a reasonable time to stop their treatment. There was no evidence of active tumor."
Five of these patients, he added, have not progressed in more than a year. One young woman was able to return to graduate school.
The trial reported research support from Genentech, the developer of bevacizumab (Avastin), a humanized monoclonal antibody that acts as an antiangiogenesis agent targeting vascular endothelial growth factor (VEGF).
"Malignant gliomas have high concentrations of VEGF. The higher the concentration of VEGF, the poorer the prognosis," Dr. Vredenburgh said. Bevacizumab, he also noted, is synergistic with chemotherapy in colorectal, breast, and lung cancers.
Irinotecan (Camptosar), a topoisomerase 1 inhibitor approved for treatment of metastatic colorectal cancer, has excellent penetration through the blood brain barrier, Dr. Vredenburgh continued. It has shown some activity against recurrent glioblastoma, he noted, citing response rates of 0%-15% and 6-month progression-free survival approaching 20%. "Clearly, it is the bevacizumab making a difference. It's not the irinotecan." he said.
Dr. Vredenburgh listed four mechanisms by which bevacizumab achieves its efficacy: a direct antitumor effect, an antiangiogenic effect, normalization of tumor vasculature with reduced interstitial pressure leading to improvement in hypoxia, and an antitumor stem cell effect.
The trial enrolled glioma patients with measurable disease in two successive cohorts of 32 and 36 patients. About two-thirds of both cohorts were men; nearly half were on enzyme-inducing antiepileptic drugs (EIAEDs). Median age was 49 and 46 years, respectively. More patients in the first cohort had grade 4 gliomas (23 vs. 9 with grade 3 gliomas), whereas patients with grade 3 gliomas predominated in the second group (24 vs. 12 patients with grade 4 gliomas).
The first 32 patients received 125 mg/m² of intravenous irinotecan over 90 minutes every 2 weeks if they were not taking any EIAEDs. If patients were taking phenytoin, carbamazepine, oxcarbazepine, or phenobarbital, the dose was increased to 340 mg/m² on the same schedule. After irinotecan, patients received 10 mg/kg of intravenous bevacizumab over 30-90 minutes.
After seeing acceptable safety results with the first cohort, the investigators changed the regimen. Depending on use of EIAEDs, the second group received 125 mg/m² or 350 mg/m² of irinotecan intravenously over 90 minutes on days 1, 8, 22, and 29. On days 1 and 22, they also received 15 mg/kg of intravenous bevacizumab after irinotecan.
A total of 14 patients (7 in each cohort) finished a year of therapy. Dr. Vredenburgh described the toxicity as manageable, but concluded that giving irinotecan every 2 weeks was optimal.
In the first group, one patient developed an asymptomatic central nervous system hemorrhage after 10 cycles and was taken off therapy. Other adverse events included an arterial thrombotic stroke (one), pulmonary emboli (two), deep venous thrombosis (one), and grade 2 proteinuria (two), which required the patients to come off study. Another three patients came off study because they required surgery, and two patients who complained of fatigue opted for palliative care.
In the second cohort, Dr. Vredenburgh reported 10 patients came off study because of toxicity during the first three cycles. Another six patients required a reduction of their irinotecan dose, and one patient (who started enoxaparin after developing a deep vein thrombosis) had a grade 2 CNS hemorrhage during the ninth treatment cycle. Venous thromboembolism was more common in the second group, which had four cases (11%) and the same number of gastrointestinal toxicities.
Dr. Vredenburgh said the investigators will look for confirmation of safety and efficacy in a recently completed phase II trial comparing bevacizumab to bevacizumab and irinotecan in recurrent glioblastoma. They also hope to include bevacizumab early in the treatment of malignant gliomas, he said.
Goli K.J. et al. Phase II trial of bevacizumab and irinotecan in the treatment of malignant gliomas. Abstract 2003.
Commentary |
The abstract by Goli et al. (of which this commentator is an author) provides further follow-up of a series of patients treated for recurrent malignant glioma with bevacizumab and irinotecan. Reporting on an additional cohort of patients (bringing the total to 35 with glioblastoma multiforme and 33 with grade 3 tumors), the data make clear that the combination of bevacizumab and irinotecan is producing extraordinarily high response rates that also appear to be quite durable. Of particular note is the 6-month progression-free survival of 43% in patients with grade 4 tumors. Accordingly, there is no question that this is a particularly active intervention and perhaps the most active identified to date in patients with recurrent malignant glioma. A series of other studies that have been completed or are in progress will help provide important initial information. One example is a recently completed randomized phase II study of patients with first- or second-relapse glioblastoma multiforme who were treated with bevacizumab or bevacizumab plus irinotecan. Studies such as these will help us try to understand the biological reasons for the marked activity of this treatment, and allow for rational translation into phase III trials in patients with newly diagnosed disease. — Henry S. Friedman, M.D.
|