Our next speaker is Dr. Ozdemir, who's going to present the analysis of mutational processes in 23 adult hemispheric diffuse gliomas, identifies DNA damage repair deficiency as a major contributor to glioma genesis. Dr. Ozdemir is the recipient of the Brain Lab Neurosurgery Award. Congratulations. Mr. and Mrs. Chairman, dear colleagues, first of all, I must express my gratitude for this award to the tumor section. Thank you for acknowledging it. And today I'm going to talk to you about our analysis of mutational processes in gliomas. I have no disclosures. As you know, the causality of gliomas remains largely unknown to date. The only firmly established risk factor is previous radiation exposure. Our aim in the study was to come with a hypothesis that a mutational process common to many or most gliomas is existent. And to analyze this, we analyzed the signatures of mutational processes that contributed to the total mutational burden in each tumor case. How we did this, we started looking at the base changes in the whole exome studies that we have done in our cases. As you know, our genetic code is consistently altered by endogenous and exogenous sources to result in changes in the nucleotides. These nucleotide changes result in different possibilities of nucleotide changes. If you want to analyze this in a systematic manner, this one single nucleotide change results in six different possibilities. If you take this in the context of the surrounding bases, this can result in 96 possible changes for one single nucleotide, considering the base before and the base after it. Having looked at these changes systematically, other laboratories have come up with 30 different mutational signatures in different tumors located inside the body. But these mutational processes are not exclusive. The tumors or the normal cells inside the body are exposed to these mutational processes throughout an extended period of time. And each of these mutational processes have a different strength causing these changes in the mutational signature. So in the end, we end up having a mixture of the different mutational signatures within one single tumor that we are analyzing. But there are ways of telling these different mutational processes from each other and telling what type of mutational processes have played a role in the formation or in the genesis of that single individual tumor. This has been applied for many tumors and one universal feature of most of these tumors was that they was an effect signature based on aging of the DNA in these tumors. To analyze or to look into our glioblastoma and diffuse glioma cases, we analyzed 25 hemisteric diffuse gliomas, basically excluded thalamic and brainstem gliomas. The median age was 50 years, we excluded pediatric cases and these tumors belonged into different molecular subsets including IDH mutant ones, TERT mutant, TERT only ones and the double negatives which also included the histone tree mutated gliomas. We did paired blood and tumor sample whole exome studies and identified mutations in each of these tumors and analyzed the weights of these mutational signatures in each of these patients using a bioinformatics tool which basically works completely in silico. We also analyzed clinical and radiological parameters to tell these tumors and make associations between these clinical parameters and the signatures that we were seeing. We looked at the age at presentation, the pathology, the molecular subset, the mutational load in each patient, anatomic location, anatomical phylogenetic compartment and the multifacality of these tumors. When we broke down these tumors into such a table, we identified actual patterns that were clearly visible among these cases. First of all, there was marked variation between each tumor. Not any tumor was the same as the next one. There were some outliers but you could easily tell that most of the tumors did have this yellow signature that you are seeing which is a result of the process of aging of the DNA. In addition to that, most of these tumor types, whatever their genetic subset may be, always included these DNA damage repair deficiency-related signatures among the signatures that they were carrying. These were very common in all glioma molecular subsets with 100% of the double negatives being affected, over two-thirds of the third mutant cases affected and over 60% of the IDH mutant gliomas affected. We compared our findings with a much larger cohort of the TCGA studies and compared to our cohort of 25 patients, this cohort of the TCGA including lower and higher grade gliomas contained 560 cases and we found a comparable percentage of DNA damage repair-related signatures in those samples. We also looked at the change of these patterns throughout time and we also found that the signature 1, which is related to aging of the DNA, continuously increased in time both in our cohort and also in the TCGA cohort. In contrast to this, the DNA damage repair-related signatures actually were more prevalent in younger patients, both in our cohort and also in the TCGA cohort. And when we looked into detail, into more detail, looking and analyzing all the DNA damage repair-related proteins in these cases, we found at least one somatic change in 72% of these 25 cases and at least one germline risk single nucleotide polymorphism, again in 72% of the cases that we have analyzed. There was at least one somatic or germline event in 100% of the cases that we have studied in this analysis. So, are these findings clinically relevant? Actually, when one purely looks into clinical observations in these cases, we see that gliomas are associated with radiation exposure. Gliomas clinically have a response to DNA damaging agents, including alkylating agents or radiation therapy. They quickly acquire new genetic alterations, which is called the mutator phenotype. They progress into higher grades over time and in many of our cases we see bone marrow suppression when we are using alkylating agents. Most of these findings, most of these clinical observations, actually point to a possible DNA damage repair lack or deficiency in these cases. Interestingly, in a recently published study of the TCGA, six of the eight germline risk polymorphisms reported in glioblastomas are actually DNA damage repair related genes. This is also true for lower grade gliomas, where four of the six germline risk polymorphisms are actually DNA damage repair related genes. So, this indicates that heredity is likely important in our glioma cases. So, in summary, each glioma has a unique collection of mutational signatures. Signatures common to most gliomas are attributed to aging and DNA damage repair related defects, and DNA damage repair deficiency related signatures are more prevalent in younger patients and they may be inherited or somatically acquired. Thank you very much for your patience and in the end I would like to acknowledge all the contributors who have worked in this project. Thank you very much.

mutational processes

gliomas

DNA damage repair deficiency

mutational signatures

heredity