An Integrated Analysis of Tumor Purity of Common Central Nervous System Tumors in Children Based on Machine Learning Methods

Background: Tumor purity is defined as the proportion of cancer cells in the tumor tissue, and its effects on molecular genetics, the immune microenvironment, and the prognosis of children’s central nervous system (CNS) tumors are under-researched.Methods: We applied random forest machine learning, the InfiniumPurify algorithm, and the ESTIMATE algorithm to estimate the tumor purity of every child’s CNS tumor sample in several published pediatric CNS tumor sample datasets from Gene Expression Omnibus (GEO), aiming to perform an integrated analysis on the tumor purity of children’s CNS tumors.Results: Only the purity of CNS tumors in children based on the random forest (RF) machine learning method was normally distributed. In addition, the children’s CNS tumor purity was associated with primary clinical pathological and molecular indicators. Enrichment analysis of biological pathways related to the purity of medulloblastoma (MB) revealed some classical signaling pathways associated with MB biology and development-related pathways. According to the correlation analysis between MB purity and the immune microenvironment, three immune-related genes, namely, CD8A, CXCR2, and TNFRSF14, were negatively related to MB purity. In contrast, no significant correlation was detected between immunotherapy-associated markers, such as PD-1, PD-L1, and CTLA4; most infiltrating immune cells; and MB purity. In the tumor purity–related survival analysis of MB, ependymoma (EPN), and children’s high-grade glioma, we discovered a minor effect of tumor purity on the survival of the aforementioned pediatric patients with CNS tumors.Conclusion: Our purity pediatric pan-CNS tumor analysis provides a deeper understanding and helps with the clinical management of pediatric CNS tumors.

AURKA gene polymorphisms and central nervous system tumor susceptibility in Chinese children

Introduction Central nervous system (CNS) tumors comprise 15–20% of all malignancies occurring in childhood and adolescence. Previous researches have shown that overexpression and amplification of the AURKA gene could induce multiple human malignancies, with which the connection of CNS tumor susceptibility has not been extensively studied. Material and methods In this study, we assessed whether and to what extent AURKA gene single nucleotide polymorphisms (SNPs) (rs1047972 C > T, rs2273535 T > A, rs8173 G > C) were associated with CNS tumor susceptibility, based on a case–control analysis in 191 CNS tumor patients and 248 controls. We determined this correlation using odds ratios (ORs) and 95% confidence intervals (CIs). Results AURKA gene rs8173 G > C exhibited a crucial function to CNS tumor susceptibility fall-off (GC/CC vs. GG: adjusted OR = 0.68, 95% CI = 0.46–0.998, P = 0.049). In addition, the combined effect of lowering the risk of developing CNS tumors was more pronounced in carriers with 3 protective genotypes than others (adjusted OR = 0.55, 95% CI = 0.31–0.98, P = 0.044). Further stratification analysis illustrated that the existence of rs8173 GC/CC and three protective genotypes lowered CNS tumor risk in some subgroups. Conclusions Our research suggested that the AURKA gene rs8173 G > C could significantly reduce CNS tumor susceptibility in Chinese children. More functional experiments are needed to explore the role of the AURKA gene rs8173 G > C.

PEDT-1 Integrated diagnoses of pediatric gliomas in our institute by cIMPACT-NOW recommendations

Purpose:Since many genetic abnormalities in glioma have been revealed in recent years, Integrated diagnoses are necessary in the updated fourth edition of the WHO Classification of Tumors of the Central Nervous System(CNS) published in 2016. The Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy (cIMPACT-NOW) was established to provide a forum to evaluate and recommend proposed changes to future CNS tumor classification. We retrospectively classified pediatric gliomas in our hospital in accordance with cIMPACT-NOW recommendations. Methods: This study includes 13 consecutive glioma patients under the age of 18 who underwent surgical resection at our hospital from 2000 to 2021. Histopathological diagnoses and molecular status such as IDH, H3F3A and BRAF were analyzed. Results: There were four females and nine males, ranging in age from 0 to 17 years, median 9.0 years. Three pilocytic astrocytomas (PA) and a polymorphous low-grade neuroepithelial tumor of the young (PLNTY) had BRAF fusion. Four cases were classified as Diffuse midline glioma, H3K27M-altered. Two cases were classified as Diffuse pediatric-type high-grade glioma, H3-wildtype and IDH-wildtype. No genetic alteration was observed in two diffuse astrocytoma cases. An anaplastic oligodendroglioma case with PDGFRA amplification could not be classified as any new entity. All three PA cases with BRAF fusion occurred in cerebellum and all four H3K27M altered cases occurred in midline location such as thalamus, brainstem and cervical spinal cord. Six cases which were classified as pediatric-type diffuse high grade gliomas had poor prognosis. Conclusion:Genetic status is associated to tumor location and patients prognosis. Integrated diagnoses are important in pediatric glioma patients.

From 65 to 70 years old: what is the best approach in the treatment of glioblastoma?

Background: Glioblastoma (GBM) is the most common primary CNS tumor in adults. Between 65-70 years of age, treatment involves the best possible surgical removal followed by radiotherapy (RT), with or without temozolomide (TMZ). After assessing whether patients can tolerate TMZ, doubts regarding RT regimens persist in this age group. This study aimed to compare the overall survival (OS) in GBM patients aged 65-70 years, in two RT regimens with TMZ: Stupp (RT 60 Gy/30 fractions (fx)+TMZ) versus mini-Stupp (RT 40.05 Gy/15 fx+TMZ) and 2 regimens of RT without TMZ: 40 Gy/15 fx versus 25 Gy/5 fx.Methods: All GBM patients, 65-70 years, undergoing RT from 1 January 2014 to 31 December 2020 were retrieved and retrospectively evaluated. Patients were divided into 4 groups: group 1 was Stupp; group 2 was mini-Stupp; group 3 was 40,05 Gy/15 fx without TMZ; and group4 was 25 Gy/5 fx without TMZ.Results: Sixty patients were retrieved with median follow up of 12 months. In the analysis of groups 1 and 2, all variables were comparable (0.21<p<0.6). Median OS was 18 and 15 months, respectively, with no statistically significant difference (p=0.13). The OS at 2 years was 26% and 21% respectively, decreasing to 13% and 0% at 3 years.Analyzing groups 3 and 4, all variables were comparable (0.06<p<0.88). OS had no difference (p=0.5) with 7 months of median OS for both groups.Conclusions: From 65-70 years, if CHT is not viable, the 25 Gy/5 fx should be the standard. When CHT is possible, mini-Stupp appears to be equivalent to Stupp.

BIOM-19. DECIPHERING THE METHYLATION SIGNATURE OF CIRCULATING EXTRACELLULAR VESICLE DNA FOR CNS TUMOR CLASSIFICATION

Genome-wide methylation profiling has recently been developed into a tool that allows subtype tumor classification in central nervous system (CNS) tumors. We previously showed that extracellular vesicle (EV) DNA faithfully reflects the tumor methylation class, including information on the IDH mutation and MGMT promoter methylation status. Furthermore we showed that circulating plasma EVs are elevated in CNS tumor patients in comparison to non-tumor donors (HD) controls with tumor related protein profiles. We now investigated, whether the methylation signatures of circulating DNA (both EV and cfNDA) can be used in liquid biopsy approaches for CNS tumor detection and classification. We isolated DNA from circulating EVs (n=27), cfDNA (n=27) and tumor tissue DNA (n=90) of patients with glioblastoma (GBM), meningioma (MGN) and cerebral metastases (CM). Patients undergoing epilepsy surgery as well as aneurysm clipping were used as non-tumor controls (HD, n= 7). EVs were classified by nanoparticle analysis, immunoblotting, imaging flow cytometry and electron microscopy. Isolated EV-DNA comprised many sorts of molecular weight (up tp &gt;10Kb) in comparison to cfDNA (130-140bp). Healthy donors and tumor patients showed not differences in their DNA size profiles. We performed genome-wide methylation profiling by 850k Illumina EPIC arrays for all DNA analytes and tumor entities. Linear models and empirical Bayes methods identified significant differentially methylated CpGs (GBM vs. HD, MGN, vs HD, CM vs. HD), that revealed tumor specific signatures to detect and discriminate different CNS tumor entities. Visualization of differentially methylated CPGs by dimension reduction (PCA, t-SNE, Umap) verified tumor specific clusters. cfDNA and EV-DNA exhibited distinctive individual CpG profiles. Our study shows that the methylation signature of circulating EV DNA and cfDNA can be used to separate healthy individuals from tumor patients and could potentially complement standard-of-care imaging to improve tumor detection, classification and surveillance.

INNV-27. AN INNOVATIVE VIRTUAL MULTI-INSTITUTIONAL, MULTIDISCIPLINARY NEURO-ONCOLOGY TUMOR BOARD: THE NIH-NOB EXPERIENCE DURING THE COVID-19 PANDEMIC

BACKGROUND The American Academy of Neurology Institute and Society for Neuro-Oncology recommend multidisciplinary tumor board (MTB) meetings as a quality metric in neuro-oncology. With the COVID-19 pandemic resulting in travel restrictions, we expanded our existing MTB by transitioning to a virtual format that maintained our commitment to providing consultation for primary CNS tumor cases. This transition permitted participation by neuro-oncology teams from over 30 Brain Tumor Trials Collaborative (BTTC)/National Cancer Institute-Comprehensive Oncology Network Evaluating Rare CNS Tumors (NCI-CONNECT) centers across the United States. Here, we describe results from opening our MTB remotely to these teams. METHODS We retrospectively reviewed records from remote MTB meetings held between April 2020 and March 2021. To gauge the impact of our MTB on clinical management, we administered a brief survey querying BTTC members. RESULTS Twenty-eight providers presented 41 cases during 24 virtual MTB meetings (range: 1-4 cases per meeting). Two cases (5%) were presented only for educational value. Approximately half (54%) of the cases discussed dealt with diagnosis/management of an NCI-CONNECT rare CNS tumor. During MTB discussions of the 39 cases seeking diagnosis/management recommendations, 32% received clinical trial recommendations, 10% were suggested to enroll in the NCI Neuro-Oncology Branch (NOB) Natural History Study (NCT02851706), 17% received a recommendation to obtain central neuropathology review, and 100% received recommendations for further disease management. Most BTTC survey respondents (83%) found these recommendations impactful in the management/treatment of their presented case or generally useful/informative for their clinical practice. CONCLUSION We describe the feasibility and utility of an innovative virtual multi-institutional MTB. These novel remote meetings allowed for discussion of complex neuro-oncology cases and recommendations from experts, particularly important for those with rare CNS tumors. Our study’s findings during the COVID-19 pandemic of the value of providing remote access to MTBs should apply post-pandemic.

NCOG-23. PATIENT-REPORTED SYMPTOM BURDEN AND INTERFERENCE: A COMPARISON BETWEEN COVID-19 PANDEMIC YEAR AND NORMATIVE DATA IN PATIENTS WITH CENTRAL NERVOUS SYSTEM (CNS) TUMORS

CNS tumor patients are highly symptomatic causing interference with activity and worse quality of life. Social distancing due to the COVID-19 pandemic increased demands on the patient, caregivers, clinicians, and the health care system. The NCI’s Neuro-Oncology Branch Natural History Study (NHS) systematically collected patient-reported outcomes (PROs) provide insight into how these challenges influenced symptom burden and interference during the COVID year. METHODS: Patient and disease characteristic as well as patient-reported symptoms and interference (MDASI-BT/-SP) and general health status (EQ-5D-3L) from 3/2020-2/2021) were compared to NHS normative sample collected prior to 3/2020. RESULTS: The sample (n = 178) was primarily White (82%), male (55%), median age of 45 (range 18 – 79) and KPS ³ 90 (51%). The majority had high-grade (70%) brain (83%) tumors (BT) with ≥ 1 prior recurrence (60%) and 25% were on active treatment. Clinical visits were primarily conducted via telehealth (64%) and 20% of all patients were diagnosed with progression at the time of assessment. Most commonly reported moderate-severe symptoms among BT patients were fatigue (30%), difficulty remembering (28%), feeling drowsy (22%). Among spinal cord tumor patients, fatigue (39%), pain (35%) and numbness/tingling in arms/legs/trunk (35%) were most frequently reported. These symptoms were reported in similar frequencies by the normative sample. Nearly half of the COVID year sample (48%) reported moderate-severe activity-related interference. Reported problems with mobility (38%), self-care (19%), pain/discomfort (40%), and usual activities (50%) were similar in both groups except for increased mood disturbance (53%) was reported during the COVID year. CONCLUSION: These findings support CNS tumor patients remained highly symptomatic with significant impact on health-related quality of life during the COVID year. Clinicians should develop timely individual care plans to help BT patients navigate their disease course. Evaluation of risk associated with more severe symptoms and functional limitations are ongoing.

PATH-46. DIAGNOSTIC IMPACT OF THE CNS TUMOR METHYLATION PROFILING IN A NEUROPATHOLOGY CONSULT PRACTICE

DNA methylation profiling coupled with the application of CNS tumor methylation classifier has contributed to precise and accurate diagnostics for a range of tumor types involving the central nervous system. The impact and characteristics of methylation profiling on tumor diagnosis has not been fully assessed in the setting of neuropathology consultation practice. A consecutive series of 1,258 surgical neuropathology samples obtained primarily in a consultation practice were profiled over 2-year period and analyzed using the DKFZ/Heidelberg CNS tumor methylation classifier. Among the 1,045 cases received from outside institutions for consultation, the classifier was able to refine a histologically diagnosed entity (e.g. medulloblastoma) in 13.3% (n = 139) cases. A substantially new diagnosis was able to be rendered in an additional 17.9% (n = 187) cases, many of which could be confirmed using orthogonal methods. A “suggestive” (0.30-0.84) classifier score was found in 23% (242) cases and we found that complementary methods (UMAP, t-SNE and nearest-neighbors) were able to resolve this uncertainty in 118 cases. We found tumor purity significantly associated with varied classifier score (p = 1.53e-11). Computational tumor purity adjustment by deconvolution on a subset of gliomas provided a proof-of-concept to resolve diagnostics in the setting of low tumor purity. Overall, this work directly assesses the benefit of methylation classification in a set of diagnostically challenging CNS tumors, addresses tumor purity diminished methylation signal and provides complementary approaches to address diagnostics in cases of low-confidence classifier scores.