Abstract 3934: Quantitative assessment of mtDNA damage differentiates long-term and short-term survivors with glioblastoma

The role of mitochondrial dysfunction in cancer has long been under scientific scrutiny. Emerging evidence suggests that a critical mechanism of anticancer therapy induced apoptosis may be the generation of mitochondrial reactive oxygen species (ROS) that can damage vital biomolecules, including mitochondrial DNA (mtDNA). We have developed methods to quantify mtDNA damage in human cells and undertook this investigation in glioblastoma to determine if there is a correlation between mtDNA damage and patient survival. Fresh-frozen tumor tissue was collected from 17 patients with glioblastoma before initiation of anti-tumor therapy. The patients were classified as short-term survivors (STS, survival 9 months or less) or long-term survivors (LTS, survival 24 months or more). Following initial resection, all of the patients underwent adjuvant therapy, usually consisting of radiation therapy plus temozolomide (1). We modified a protocol to assess mtDNA damage in which quantitative long-range PCR is performed on an 8.9kb segment of the mtDNA circle (2). The amount of amplified DNA generated is inversely proportional to the extent of mtDNA damage that blocks polymerase synthesis. Total DNA was extracted from the samples, quantified and evaluated qualitatively by gel electrophoresis. The amplified mtDNA was normalized to mtDNA copy number, which was determined by Real-Time PCR of the mitochondrial ND1 gene using DNA standards. A statistically significant correlation was found between the level of mtDNA damage and survival. Higher levels of mtDNA damage - less amplifiable mtDNA - were found in the tumors of LTS than in the tumors of STS. The mean amplifiable mtDNA in nanograms per million copies of ND1 was 22.2 for STS tumors and 15.8 for LTS tumors (two-tail t-test P value = 0.038). These results are consistent with the hypothesis that reactive oxygen species must be generated in order to trigger apoptosis in response to chemotherapeutic drugs, or other treatment modalities. LTS glioblastomas have less amplifiable mtDNA, therefore more damage per mtDNA molecule, indicating that these tumors have greater basal ROS-generating metabolism, less effective anti-oxidant buffering, or poor mtDNA repair or elimination mechanisms. In comparison, STS glioblastomas may “shut down” basal mitochondrial production of ROS, possibly by a more pronounced Warburg effect (decrease in oxidative phosphorylation), buffer the ROS via anti-oxidant mechanisms, or more effectively repair or e