Search

The GLASS-NL consortium, was initiated to gain insight into the molecular mechanisms underlying glioma evolution and to identify markers of progression in IDH-mutant astrocytomas. Here, we present the first results of genome-wide DNA-methylation profiling of GLASS-NL samples. 110 adult patients were identified with an IDH-mutant astrocytoma at first diagnosis. All patients underwent a surgical resection of the tumor at least twice, separated by at least 6 months (median 40.9 months (IQR: 24.0, 64.7). In 37% and 18% of the cases, patients were treated with radiotherapy or chemotherapy respectively, before surgical resection of the recurrent tumor. DNA-methylation profiling was done on 235 samples from 103 patients (102 1st, 101 2nd, 29 3rd, and 3 4th resection). Copy number variations were also extracted from these data. Methylation classes were determined according to Capper et al. Overall survival (OS) was measured from date of first surgery. Of all primary tumors, the methylation-classifier assigned 85 (87%) to the low grade subclass and 10 (10%) to the high grade subclass. The relative proportion of high grade tumors increased ~three-fold at tumor recurrence (32/101, 32%) and even further in the second recurrence (15/29, 52%). Methylation classes were prognostic, both in primary and recurrent tumors. The overall DNA-methylation levels of recurrent samples was lower than that of primary samples. This difference is explained by the increased number of high grade samples at recurrence, since near identical DNA-methylation levels were observed in samples that remained low grade. In an unsupervised analysis, DNA-methylation data derived from primary and first recurrence samples of individual patients mostly (79%) cluster together. Recurrent samples that do not cluster with their primary tumor, form a separate group with relatively low genome-wide DNA-methylation. Our data demonstrate that methylation profiling identifies a shift towards a higher grade at tumor progression coinciding with reduced genome-wide DNA-methylation levels.

INTRODUCTION. Diffuse gliomas are the most frequent and devastating primary CNS tumors in adults. Standard treatment has limited efficacy and without exception these gliomas recur. The evolutionary processes that drive progression in glioma of the IDH-mutant astrocytoma subgroup, remain unclear. The GLASS-NL consortium was initiated to gain insight into the molecular mechanisms underlying glioma evolution and to identify markers of progression in IDH-mutant astrocytomas. Such markers can ultimately assist clinical decision making. Here, we present the results of genome wide DNA-methylation profiling of samples included in the GLASS-NL study. METHODS. 110 adult patients were identified with an IDH-mutant, 1p19q non-codeleted, astrocytoma at first diagnosis. All patients underwent surgical resection of the tumor at least twice, separated by >6 months with a median of 41.9 months (IQR:26.5-65.9). After surgical resection of the initial tumor, 63% and 22% of the patients were treated with radiotherapy or chemotherapy respectively. DNA-methylation profiling was performed on 235 samples from 103 patients, using the Illumina Infinium MethylationEPIC BeadChip array. Copy number alterations (CNAs) were extracted from these data. Methylation subclasses were determined according to Capper et al. (Nature, 2018). Overall survival (OS) was measured from date of initial surgery. RESULTS. Of all primary tumors, 85 (87%) of the tumor samples were assigned to the A_IDH (‘low grade’) methylation subclass and 10 (10%) to the A_IDH_HG (‘high grade’) subclass. The relative proportion of high grade tumors increased ~three-fold at tumor recurrence (32/101, 32%) and even further in the second recurrence (15/29, 52%). The high grade subclass of the recurrent, but not the initial tumor sample, was negatively associated with OS (p < 0.0001). The overall DNA-methylation levels of recurrent samples were lower than that of initial samples. This difference is explained by the increased number of high grade samples at recurrence, since near identical DNA-methylation levels were observed in samples that remained low grade. Analysis of CNAs revealed several chromosomal arms and bands that were more frequently altered in samples of the high grade methylation subclass compared to low grade. In addition, gains and losses of specific genes, such as homozygous deletion (HD) of CDKN2A/B, were more frequent in high grade samples. Overall DNA-methylation levels of recurrent samples with CDKN2A/B HD were lower than that of samples without this deletion. However, CDKN2A/B HD alone does not fully explain DNA-demethylation at malignant progression and other molecular aberrations are likely to contribute as well. CONCLUSION. Longitudinal methylation profiling analysis of IDH-mutant astrocytoma reveals a shift towards a higher grade at tumor recurrence coinciding with reduced genome-wide DNA-methylation levels. Citation Format: Wies Rijan Vallentgoed, Anneke Niers, Karin van Garderen, Martin van den Bent, Erik van Dijk, Kaspar Draaisma, Paul van Eijk, Iris de Heer, Mathilde Kouwenhoven, Johan Kros, Wendy de Leng, Ivonne Martin, Pierre Robe, Marion Smits, Mircea Tesileanu, Hinke van Thuijl, Roel Verhaak, Bart Westerman, Mark van de Wiel, Bauke Ylstra, Pieter Wesseling, Pim French. Methylation analysis of matched primary and recurrent IDH-mutant astrocytoma; an update from the GLASS-NL consortium [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 4020.

The current gold-standard of estimating adverse events of a drug are clinical trials. However, these trials fail to represent real-life practice where patients have additional diseases (comorbidities) and commonly use numerous drugs when entering the clinic. Therefore, there is a rise of interest in combinational therapies, also because effective combinations are expected to prevent therapy resistance. At this moment it is not feasible to predict the adverse events of new combination therapies due to lack of available information both from a dimensionality (i.e., number of adverse events recorded per patient) as well as from a patient-number perspective. When available, this information allows to choose combinational therapies with acceptable adverse events. In this study we developed a preliminary method to predict adverse events of drug combinations in order to select combinations with a mild adverse event profile. We used the FAERS, an FDA post-marketing adverse events registry as data source containing 15 million adverse-event records. First, we developed a method to visualize the adverse events profiles of monotherapy and combination therapy using dimension reduction to accurately represent the relation between adverse events over many patients. These adverse-event profiles are then fed to a convolutional neural network (CNN) to generate an explainable prediction-model for adverse events occurring in combination therapies. The CNN trained on monotherapy is able to learn from the data and recognize adverse event patterns. The learned pattern information, as stored in the so-called latent space, can be converted back onto the original adverse event profiles. This showed a high similarity to the original data, also for unseen combination therapy effects. Furthermore, a t-SNE analysis on the latent space of the CNN is able to separate additive and synergistic adverse event patterns in combination therapy. Our CNN model can successfully learn complex adverse-event patterns for single drugs and their combinations, which are all encoded in the latent space. The developed method is therefore applicable to determine the combinatorial effects of highly complex adverse event profiles. Citation Format: Asli Kucukosmanoglu, Silvia Scoarta, Thomas Wijnands, George Kanev, Bart Westerman, Bert Kiewit, David Noske, Tom Wurdinger. The adverse events atlas, towards a strategy to predict synergistic adverse events of combination therapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6312.