Assay Aids Patient Selection for Temozolomide in Glioblastoma

June 2011, Vol 2, No 3 -

Chicago, IL—Aprognostic marker that can be identified with a commercial test can identify patients with newly diagnosed glioblastoma who would benefit from life-prolonging temozolomide treatment, according to Mark R. Gilbert, MD, Professor, Department of Neuro-oncology, at the Division of Cancer Medicine, M. D. Anderson Cancer Center, Houston. Dr Gilbert presented data from a new study at a late-breaking session at ASCO 2011.

“There is a pressing need for molecular tests that assist in the identification and monitoring of the most effective treatment for each cancer patient. This is especially true with cancer therapeutics that rely on inducing DNA damage in the tumor,” said Dr Gilbert, lead investigator.

MGMT (O6-methylguanine-DNA methyl transferase) is a key DNA repair enzyme produced by the MGMT gene. When the MGMT gene is methylated, no MGMT is produced. The more MGMT present in the tumor, the more active the DNA repair, which lessens the efficacy of therapies that work by inducing DNA damage.

“The MGMT gene test (MDxHealth) was confirmed in the RTOG 0525 clinical trial to define prognosis for patients with newly diagnosed glioblastoma. Ongoing clinical trials have incorporated this now-validated prognostic marker,” he said.

Two studies demonstrated the benefit of the test. One study (LBA2000) described the development of the MGMT gene test and the consolidation of biomarkers into risk groups. These risk profiles created in the training set were applied to samples from 763 patients in RTOG 0525 (a comparison of dosing schedules of temozolomide), which gave external validation to the test.

The application of the molecular risk classification to RTOG 0525 samples selected patients with significantly (P <.001) improved survival, and it improved on current means of prediction by revealing an additional distinct risk group.

RTOG 0525 was a phase 3 trial that compared standard adjuvant temozolomide with a dose-dense schedule in newly diagnosed glioblastoma patients. Patients were randomized to receive either standard treatment (temozolomide on days 1-4 every 28 days) or the dose-dense regimen (temozolomide on days 1-21 every 28 days) for up to 12 cycles. Clinical outcomes were evaluated according to (1) prognostic recursive partitioning (based on age, performance status, extent of pretreatment surgery, neurologic function, mental status), (2) MGMT status (methylated, unmethylated, indeterminate), and (3) radiation therapy treatment.

The MGMT gene test was able to identify the patients with glioblastoma who benefited most from the treatment. Patients whose tumors demonstrated MGMT methylation had a median survival time of 21.2 months versus 14 months without methylation (P <.001) and longer progression-free survival (8.7 months vs 5.7 months; P <.001). Treatment arm and radiation technique did not predict survival.

“This study confirmed the prognostic significance of MGMT methylation in glioblastoma,” Dr Gilbert said. The results suggested that treatment decisions based on the molecular characteristics of the tumor are not only feasible but make a clinical difference.

MGMT testing is now being provided for numerous phase 2 and phase 3 trials, including the phase 3 CENTRIC trial, in which only MGMT-methylated patients will receive the investigational integrin inhibitor cilengitide added to current chemotherapy for glioblastoma.