Atypical behavior of recurrent glioblastoma tumor cells with a highly adherent radial glial phenotype

Accurate prognostication may aid in the selection of patients who will benefit from surgery at recurrent WHO grade 4 glioma. This study aimed to evaluate the role of serial tumour volumetric measurements for prognostication at first tumour recurrence. We retrospectively analyzed patients with histologically-diagnosed WHO grade 4 glioma at initial and at first tumour recurrence at a tertiary hospital between May 2000 and September 2018. We performed auto-segmentation using ITK-SNAP software, followed by manual adjustment to measure serial contrast-enhanced T1W (CE-T1W) and T2W lesional volume changes on all MRI images performed between initial resection and repeat surgery. Thirty patients met inclusion criteria; the median overall survival using Kaplan-Meier analysis from second surgery was 10.5 months. Seventeen (56.7%) patients received treatment post second surgery. Univariate cox regression analysis showed that greater rate of increase in lesional volume on CE-T1W (HR = 2.57; 95% CI [1.18, 5.57]; p = 0.02) in the last 2 MRI scans leading up to the second surgery was associated with a higher mortality likelihood. Patients with higher Karnofsky Performance Score (KPS) (HR = 0.97; 95% CI [0.95, 0.99]; p = 0.01) and who received further treatment following second surgery (HR = 0.43; 95% CI [0.19, 0.98]; p = 0.04) were shown to have a better survival. Higher rate of CE-T1W lesional growth on the last 2 MRI images prior to surgery at recurrence was associated with increase mortality risk. A larger prospective study is required to determine and validate the threshold to distinguish rapidly progressive tumour with poor prognosis.

Target population These recommendations apply to adult patients with recurrent WHO grade 2 infiltrative diffuse glioma (oligodendroglioma, astrocytoma).Questions and Recommendations:Imaging Q1: In adult patients with suspected recurrence of histologically proven WHO grade 2 diffuse glioma, do advanced imaging techniques using magnetic resonance spectroscopy, perfusion weighted imaging, diffusion weighted imaging or PET provide superior assessment of tumor recurrence and histologic progression compared to standard MRI neuroimaging?Recommendation Level III: In adult patients with suspected recurrence of histologically proven WHO grade 2 diffuse glioma, advanced imaging techniques using magnetic resonance spectroscopy, perfusion weighted imaging, diffusion weighted imaging or PET are suggested for identification of tumor recurrence or histologic progression.Pathology Q1: In adult patients with suspected recurrence of histologically proven WHO grade 2 diffuse glioma, is molecular testing for IDH-1, IDH-2, and TP53 Mutations and MGMT promotor methylation mutation warranted for predicting survival and formulating treatment recommendations?Recommendation Level III: It is suggested that IDH mutation status be determined for diagnostic purposes. TP53 mutations occur early in WHO grade 2 diffuse glioma pathogenesis, remain stable, and are not suggested as a marker of predisposition to malignant transformation at recurrence or other measures of prognosis. Assessment of MGMT status is suggested as an adjunct to assessing prognosis. Assessment of CDK2NA status is suggested since this is associated with malignant progression of WHO grade 2 diffuse gliomas.Q2: In adult patients with suspected recurrence of histologically proven WHO Grade 2 diffuse glioma, is testing of proliferation indices (MIB-1 and/or BUdR) warranted for predicting survival and formulating treatment recommendations?Recommendation Level III: It is suggested that proliferative indices (MIB-1 or BUdR) be measured in WHO grade 2 diffuse glioma as higher proliferation indices are associated with increased likelihood of recurrence and shorter progression free and overall survival.Chemotherapy Q1: In adult patients with suspected recurrence of histologically proven WHO grade 2 diffuse glioma, does addition of temozolomide (TMZ), other cytotoxic agents or targeted agents to their treatment regimen improve PFS and/or OS?Recommendation Level III: Temozolomide is suggested in the therapy of recurrent WHO grade 2 diffuse glioma as it may improve clinical symptoms. PCV is suggested in the therapy of WHO grade 2 diffuse glioma at recurrence as it may improve clinical symptoms with the strongest evidence being for oligodendrogliomas. TMZ is suggested as the initial choice for recurrent WHO grade 2 diffuse glioma. Carboplatin is not suggested as there is no significant benefit from carboplatin as single agent therapy for recurrent WHO grade 2 diffuse gliomas. There is insufficient evidence to make any recommendations regarding other agents in the management of recurrent WHO grade 2 diffuse glioma.Radiotherapy Q1: In adult patients with suspected recurrence of histologically proven WHO grade 2 diffuse glioma, does addition of radiotherapy to treatment regimen improve PFS and/or OS?Recommendation Level III: Radiation is suggested at recurrence if there was no previous radiation treatment. Q2: In adult patients with suspected recurrence of histologically proven WHO grade 2 diffuse glioma after previous radiotherapy, does addition of re-irradiation or proton therapy to the treatment regimen improve PFS and/or OS?Recommendation Level III: It is suggested that re-irradiation be considered in the setting of WHO grade 2 diffuse glioma recurrence as it may provide benefit in PFS and OS.Surgery Q1: In adult patients with suspected recurrence of histologically proven WHO grade 2 diffuse glioma, does surgical resection improve PFS and/or OS?. There is insufficient evidence to make any new specific recommendations regarding the value of surgery or extent of resection in relationship to survival for recurrent WHO grade 2 diffuse glioma.

OBJECTIVE This is the first clinical outcomes report of NRG Oncology RTOG 0539, detailing the primary endpoint, 3-year progression-free survival (PFS), compared with a predefined historical control for intermediate-risk meningioma, and secondarily evaluating overall survival (OS), local failure, and prospectively scored adverse events (AEs). METHODS NRG Oncology RTOG 0539 was a Phase II clinical trial allocating meningioma patients to 1 of 3 prognostic groups and management strategies according to WHO grade, recurrence status, and resection extent. For the intermediate-risk group (Group 2), eligible patients had either newly diagnosed WHO Grade II meningioma that had been treated with gross-total resection (GTR; Simpson Grades I-III) or recurrent WHO Grade I meningioma with any resection extent. Pathology and imaging were centrally reviewed. Patients were treated with radiation therapy (RT), either intensity modulated (IMRT) or 3D conformal (3DCRT), 54 Gy in 30 fractions. The RT target volume was defined as the tumor bed and any nodular enhancement (e.g., in patients with recurrent WHO Grade I tumors) with a minimum 8-mm and maximum 15-mm margin, depending on tumor location and setup reproducibility of the RT method. The primary endpoint was 3-year PFS. Results were compared with historical controls (3-year PFS: 70% following GTR alone and 90% with GTR + RT). AEs were scored using NCI Common Toxicity Criteria. RESULTS Fifty-six patients enrolled in the intermediate-risk group, of whom 3 were ineligible and 1 did not receive RT. Of the 52 patients who received protocol therapy, 4 withdrew without a recurrence before 3 years leaving 48 patients evaluable for the primary endpoint, 3-year PFS, which was actuarially 93.8% (p = 0.0003). Within 3 years, 3 patients experienced events affecting PFS: 1 patient with a WHO Grade II tumor died of the disease, 1 patient with a WHO Grade II tumor had disease progression but remained alive, and 1 patient with recurrent WHO Grade I meningioma died of undetermined cause without tumor progression. The 3-year actuarial local failure rate was 4.1%, and the 3-year OS rate was 96%. After 3 years, progression occurred in 2 additional patients: 1 patient with recurrent WHO Grade I meningioma and 1 patient with WHO Grade II disease; both remain alive. Among 52 evaluable patients who received protocol treatment, 36 (69.2%) had WHO Grade II tumors and underwent GTR, and 16 (30.8%) had recurrent WHO Grade I tumors. There was no significant difference in PFS between these subgroups (p = 0.52, HR 0.56, 95% CI 0.09-3.35), validating their consolidation. Of the 52 evaluable patients, 44 (84.6%) received IMRT, and 50 (96.2%) were treated per protocol or with acceptable variation. AEs (definitely, probably, or possibly related to protocol treatment) were limited to Grade 1 or 2, with no reported Grade 3 events. CONCLUSIONS This is the first clinical outcomes report from NRG Oncology RTOG 0539. Patients with intermediate-risk meningioma treated with RT had excellent 3-year PFS, with a low rate of local failure and a low risk of AEs. These results support the use of postoperative RT for newly diagnosed gross-totally resected WHO Grade II or recurrent WHO Grade I meningioma irrespective of resection extent. They also document minimal toxicity and high rates of tumor control with IMRT. Clinical trial registration no.: NCT00895622 (clinicaltrials.gov).

PurposeIn primary brain tumors, the efficacy of immune-modulating therapies is still under investigation as inflammatory responses are restricted by tight immunoregulatory mechanisms in the central nervous system. Here, we measured soluble PD-L1 (sPD-L1) in the plasma of patients with recurrent glioblastoma (GBM) and recurrent WHO grade II–III glioma treated with bevacizumab-based salvage therapy.MethodsThirty patients with recurrent GBM and 10 patients with recurrent WHO grade II–III glioma were treated with bevacizumab-based salvage therapy at the Medical University of Vienna. Prior to each treatment cycle, EDTA plasma was drawn and sPD-L1 was measured applying a sandwich ELISA with a lower detection limit of 0.050 ng/ml. Leukocyte counts and C-reactive protein (CRP) levels were measured according to institutional practice.ResultsMedian number of sPD-L1 measurements was 6 per patient (range: 2–24). At baseline, no significant difference in sPD-L1 concentrations was observed between WHO grade II–III glioma and GBM. Intra-patient variability of sPD-L1 concentrations was significantly higher in WHO grade II–III glioma than in GBM (p = 0.014) and tendentially higher in IDH-mutant than in IDH-wildtype glioma (p = 0.149) In WHO grade II–III glioma, sPD-L1 levels were significantly lower after one administration of bevacizumab than at baseline (median: 0.039 ng/ml vs. 0.4855 ng/ml, p = 0.036). In contrast, no significant change could be observed in patients with GBM.ConclusionsChanges in systemic inflammation markers including sPD-L1 are observable in patients with recurrent glioma under bevacizumab-based treatment and differ between WHO grade II–III glioma and GBM.

Introduction: Gliomas are the most common primary malignant brain tumors in adults, with high recurrence rates despite multimodal treatment. While previous multi-institutional genomic studies have revealed molecular differences between primary and recurrent tumors, the AACR Project GENIE dataset provides another valuable resource to investigate these distinctions. Methods: We analyzed 8, 959 adult glioma cases from Project GENIE v16.1 (2024), selecting cases with complete mutation and copy number alteration (CNA) data. Tumors were stratified by Isocitrate dehydrogenase (IDH) mutation status, as IDH-wildtype and IDH-mutant gliomas have distinct molecular profiles and patterns of progression. The IDH-mutant cohort (n=2, 635) comprised 79.9% primary tumors (n=2, 097) and 20.1% recurrent tumors (n=538). The IDH-wildtype cohort (n=5, 885) included 89.8% primary tumors (n=5, 290) and 10.2% recurrent tumors (n=595). We examined the top 10 mutated somatic genes and compared mutational frequencies and copy number variations between primary and recurrent tumors using Fisher's exact test with Benjamini-Hochberg correction. Results: In IDH-mutant gliomas, recurrent tumors showed increased mutations across multiple genes, particularly in KMT2D (7.8% to 20.6%, OR 3.06, p<0.001), NF1 (5.25% to 13.8%, OR 2.79, p<0.001) and NOTCH1 (12.4% to 23.6%, OR 2.18, p<0.001). In IDH-wildtype gliomas, the most notable changes were observed in CIC (21.65% to 32.71%, OR 2.12, p<0.001) and SMARCA4 (6.82% to 13.75%, OR 2.36, p<0.001). Oncogenic drivers like TP53 (p>0.05) and PTEN (p>0.05) maintained stable mutation frequencies between primary and recurrent tumors in both IDH cohorts. Copy number alterations for all examined genes showed no significant differences between primary and recurrent tumors in either IDH group. Conclusion: Analysis of this large real-world dataset reveals molecular evolution patterns that align with previous observations, particularly in the stability of copy number alterations in driver genes between primary and recurrent tumors across both IDH groups. Similarly, oncogenic drivers like TP53 and PTEN maintained stable mutation frequencies, indicating preservation of critical growth-promoting pathways during glioma progression. Recurrent IDH-mutant tumors showed high mutational variance in chromatin regulators like KMT2D as well as developmental pathway gene NOTCH1. Recurrent IDH-wildtype gliomas showed more modest increases in mutational frequencies compared to primary counterparts, indicating less selective pressure from treatment in these more aggressive tumors. While the unbalanced distribution of primary versus recurrent samples and potential sampling biases inherent in real-world data limit the generalizability of these findings, these patterns may inform future therapies. Further validation in independent cohorts is warranted. Citation Format: Abhishek Bhattacharya, Anand Kornepati. Molecular evolution patterns in primary versus recurrent adult gliomas: A multi-institutional study from Project GENIE [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1133.

MR perfusion depicts angiogenesis as a key factor for growth and malignancy in gliomas by means of increased regional cerebral blood volume (rCBV). The rCBV increase is not limited to the tumour area, but may also produce a stripe-like pattern of peritumoural rCBV increase that we defined as the "striate sign". We evaluated if prior radiochemotherapy influences perfusion values and pattern in and adjacent to malignant gliomas comparing rCBV of treated recurrent gliomas with untreated gliomas. Ninety-three patients with primary or recurrent WHO grades II-IV glial tumours underwent T2*-weighted dynamic susceptibility-weighted contrast-enhanced (DSC)-MRI. Differences of normalised rCBV and rCBV(max) were evaluated using Kruskal-Wallis analysis with post hoc tests. The number of cases showing a hot spot of rCBV (rCBV(max)) and/or a peritumoural striate pattern of rCBV increase (striate sign) was assessed and evaluated by Fisher's exact test. Significance level was determined as p < 0.05. Normalised rCBV, rCBV(max) and number of cases with the striate sign were significantly lower in recurrent (rCBV = 3.24 +/- 1.22, rCBV(max) = 5.05 +/- 2.27 and striate sign = 10/24) compared to primary WHO grade IV tumours (rCBV = 4.44 +/- 1.39, rCBV(max) = 7.31 +/- 3.0 and striate sign = 17/21, respectively). There were fewer cases with a striate sign in treated recurrent WHO grade III tumours than in untreated malignant transformed WHO grade II tumours. The pattern and degree of rCBV increase in and around gliomas differ between untreated and previously treated tumours. These differences might be due to post-therapeutic changes of the tumour-associated microvasculature by radiochemotherapy. Spectroscopic and susceptibility-weighted MR imaging may provide further insights into the tumour biology.

Background: Glioblastoma (GBM) is the most frequent and the most malignant brain tumor in adults. Despite aggressive therapy, overall survival of GBM patients remains poor due to intrinsic or acquired resistance to combined radiation and chemotherapy. Underlying biology behind GBM progression after first line therapy is mostly unknown and thus, there are only limited targeted therapeutic options for the 2nd line treatment. The aim of this study was to investigate genetic landscape in paired samples of primary versus recurrent tumors and to validate prognostic and predictive value of the most significant differences in independent cohort of GBMs. Patients and methods: 43 paired tumor samples were obtained during the 1st surgery followed by radiochemotherapy and after tumor recurrence during the 2nd palliative surgery. 24/43 paired samples were analyzed for whole genome copy number variations (CNV), 23/43 by fluorescent in situ hybridization for determination of EGFR, p53, RB1, MDM2, CDKN2A, 1p, 19q and 10p statuses. Methylation-specific PCR was used for analysis of MGMT promoter methylation and competitive amplification of differentially melting amplicons PCR for IDH1/2 mutations detection. The impact of most relevant alterations identified in paired GBMs was then validated in independent cohort of unpaired 104 tumors. Results: The most significant differences induced by chemoradiotherapy and/or GBM progression in primary versus recurrent tumors were EGFR amplification (19/24 primary tumors, 16/24 recurrent tumors), CDKN2A loss (18/24 primary tumors, 16/24 recurrent tumors), MDM2 gain (4/24 primary tumors), PDGFRA gain (3/24 recurrent tumors) and GSTT1 loss in primary tumors (7/24) and GSTM1 gain in recurrent tumors (7/24). Univariate analysis of overall survival for significant copy number alterations shown significantly worse prognosis for patients with GSTT1 deletion (HR = 5.356, CI = 1.161-24.701, p = 0.031) in primary tumor and trend for better prognosis in patients with CDKN2A deletion (HR = 0.391, CI = 0.124-1.229, p = 0.108) and GSTM1 amplification (HR = 0.341, CI = 0.090-1.288, p = 0.112) in recurrent tumors. Significantly worse prognosis for GBM patients with aberrations of GSTT1 gene was also confirmed in independent validation cohort of unpaired GBMs in multivariate analysis adjusted for age and Karnofsky score (gain: HR = 1.95, CI = 1.116-3.260, p = 0.011; loss: HR = 2.15, CI = 1.185-3.900, p = 0.012). Conclusion: Molecular genetic profiling of 24 recurrent GBM showed association between recurrence and status of glutathione S-transferases genes. Significance of GSTT1 gene aberration was validated in independent group of glioblastomas and indicates involvement of redox enzymes in GBM chemoradiation response and tumor progression. Acknowledgment: The financial support of grants TE02000058, IGA UP LF_2017_013 is gratefully acknowledged. Citation Format: Magdalena Houdova Megova, Zuzana Sporikova, Radek Trojanec, Ondrej Kalita, Jana Vrbkova, Jiri Drabek, Vladimira Koudelakova, Marian Hajduch. Mapping of molecular landscape underlying drug resistance and recurrence in glioblastoma: Paired analysis of primary and recurrent tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3695.

The molecular and genetic basis of tumor recurrence is complex and poorly understood. RIPK3 is a key effector in programmed necrotic cell death and, therefore, its expression is frequently suppressed in primary tumors. In a transcriptome profiling between primary and recurrent breast tumor cells from a murine model of breast cancer recurrence, we found that RIPK3, while absent in primary tumor cells, is dramatically reexpressed in recurrent breast tumor cells by an epigenetic mechanism. Unexpectedly, we found that RIPK3 knockdown in recurrent tumor cells reduced clonogenic growth, causing cytokinesis failure, p53 stabilization, and repressed the activities of YAP/TAZ. These data uncover a surprising role of the pro-necroptotic RIPK3 kinase in enabling productive cell cycle during tumor recurrence. Remarkably, high RIPK3 expression also rendered recurrent tumor cells exquisitely dependent on extracellular cystine and undergo necroptosis upon cystine deprivation. The induction of RIPK3 in recurrent tumors unravels an unexpected mechanism that paradoxically confers on tumors both growth advantage and necrotic vulnerability, providing potential strategies to eradicate recurrent tumors.

BACKGROUND Glioblastomas are highly vascular tumours that overexpress vascular endothelial growth factor (VEGF), however the anti-VEGF antibody bevacizumab has not demonstrated significant overall survival benefit for patients. Vasculogenic mimicry (VM), the ability of tumour cells to form functional microvasculature lacking an endothelial lining, may compensate for insufficient or inhibited angiogenesis. VM has been reported in primary glioblastoma tissue, but not investigated at progression in recurrent tumours. Prostate specific membrane antigen (PSMA) is expressed in abnormal vessels within primary glioblastoma tumours. We aimed to determine whether VM occurs in primary and recurrent glioblastoma tumours and whether VM vessels also express PSMA. MATERIAL AND METHODS Thirty-five matched pairs of primary and recurrent formalin-fixed paraffin-embedded glioblastoma tissue were immunohistochemically labelled for the endothelial cell marker CD34 and PSMA, and counterstained with periodic acid-Schiff (PAS). Vascular structures were categorised as endothelial vessels (CD34+/PAS+) or VM (CD34-/PAS+). Vessels were counted in ten regions within each tumour and a vessel density/mm2 for each vessel category was determined. RESULTS Data are expressed as median values. Endothelial vessels accounted for most blood vessels observed. There was a significant decrease in endothelial vessel density in recurrent (41.34 vessels/mm2) compared to primary (86.98 vessels/mm2) glioblastomas (p &amp;lt;0.001). VM was observed in 15/35 primary tumours (42.86%) and 10/35 recurrent tumours (28.57%). VM accounted for only a small proportion of the overall vasculature and VM density was not different between the primary and recurrent groups (p = 0.266). Kaplan-Meier analyses with log rank tests indicated that VM-tumours had longer median OS time than VM+ tumours, regardless of whether VM status was determined based on the primary or recurrent tumour. However, there was no significant difference in survival distribution between VM+ and VM-tumours (primary χ2(1) = 0.381, p = 0.537; recurrent χ2(1) = 3.552, p = 0.059). PSMA is expressed in endothelia lined vessels based on serial section labelling for PSMA and CD34. PSMA expression significantly decreased in recurrent (median H-score = 0.22) compared to primary (median H-score = 0.44) tumours (one-tailed p = 0.031; n = 32) mirroring the CD34+ vessel density. CONCLUSION VM contributes to a relatively small proportion of the overall vasculature in glioblastoma tumours and was not associated with OS nor the expression of PMSA. Since VM- tumours had longer OS than VM+ tumours, assessment of VM in recurrent tumours in a larger cohort may determine if there is a clinical significance for VM in recurrent glioblastoma.

2060 Background: We completed a study evaluating a single intratumoral delivery of PVSRIPO in recurrent WHO grade IV MG patients (N Engl J Med. 2018 Jul 12;379(2):150-161). Some patients who originally benefitted from the infusion of PVSRIPO demonstrated tumor recurrence, and we hypothesized that retreatment could trigger an immune recall effect, further extending their survival. We now report the impact of second and third intratumoral reinfusion of PVSRIPO in patients treated in the original dose finding study. Methods: Eligible patients were adults with recurrent supratentorial WHO grade IV MG who were experiencing disease recurrence after having benefitted from the first infusion of PVSRIPO. Additional eligibility criteria included: solitary tumor 1-5.5cm in diameter; ≥4 weeks after chemotherapy, bevacizumab or study drug; adequate organ function; KPS≥70%; and positive anti-polio titer. One patient each was retreated at 1 x 107 TCID50 and 1 x 1010 TCID50, and three patients were retreated on the identified phase 2 dose of 5 x 107 TCID50. Results: As of 2/09/2019, five patients have received a second intratumoral dose of PVSRIPO on study, one of which received a total of 3 doses. The patients who received two infusions of PVSRIPO were retreated 72 months, 43 months, 34 months, and 6 months after the first infusion. One additional patient received a second infusion of PVSRIPO 60 months after the first infusion and a third infusion of PVSRIPO 78 months after the first infusion. All patients demonstrated soap bubble degeneration on imaging, and two patients demonstrated tumor contraction. No grade 3 or higher adverse events related to PVSRIPO were observed after retreatment. Three of these patients remain alive more than 81, 80 and 52 months following the first PVSRIPO infusion and more than 9, 20 and 18 months after the second infusion, respectively. Two patients died 63 months and 20 months after the first infusion of PVSRIPO and 19.6 and 14 months after the second, respectively. The patient treated 3 times received the third infusion more than 2 months ago. Conclusions: Intratumoral reinfusion of PVSRIPO via CED is safe, and encouraging efficacy results have been observed. Clinical trial information: NCT01491893.

BACKGROUND Surgery for recurrent glioblastoma (GBM) has the potential of prolonging patient survival. However, the impact of the timepoint of re-resection after tumor suspicion on post-surgical survival and neurological deterioration remains unclear. MATERIAL AND METHODS Patients undergoing first re-resection for recurrent WHO grade 4 IDH-wildtype GBM were analyzed retrospectively. Post-surgical neurological deterioration and post-surgical survival of patients undergoing early surgery (within 3 weeks) after suspected recurrence were compared to those receiving surgery later than 3 weeks. In addition, volumetric analyses were used to determine tumor burden and location. RESULTS 103 consecutive patients (mean age: 59 years; 64 males and 39 females) undergoing first re-resection for recurrent GBM from 2018 to 2022 were included. Gross total resection (GTR) had been achieved in 93 patients (91%) at first diagnosis and was achieved in 96 patients (93%) at repeat surgery. Intraoperative MRI-guided resection was performed in 84 (82%) cases and 5-ALA was applied in 8 (8%) cases, with neuromonitoring utilized in 14 (14%) cases. The median time interval between first and second resection was 341 days (IQR 343 days) while the median time interval between suspected tumor recurrence and re-resection was 20 days (IQR 23 days). Surgery was delayed beyond 3 weeks after suspected recurrence in 49 cases (mean interval 11 in the ‘early’ vs 187 days in the ‘late’ group, p&amp;lt;0.001) due to salvage chemotherapy/targeted therapy in 18 cases, suspicion of pseudo-progression in 7 cases and patient preference in 24 cases. Compared to patients receiving ‘early’ repeat surgery, this delay was not associated with a higher rate of immediate (14/54 vs 10/49 patients, p=0.508) or permanent (3/54 vs 4/49 patients, p=0.600) post-operative neurological deterioration. Median survival after repeat surgery was 220 days and did not significantly differ between early and late surgery (212 in the late vs 243 days in the early group, p=0.682). Volumetric analyses revealed no significant differences in tumor volume or localization pertaining to eloquence. CONCLUSION Timing of surgery for suspected first recurrence of GBM did not impact on functional outcomes and post-surgical survival. In general, satisfactory rates of postoperative neurological deficits and post-surgical survival in this contemporary cohort of patients with repeat surgery for recurrent GBM could be achieved.

2095^ Background: No standard treatment exists for recurrent WHO grade 3 malignant glioma patients. Bevacizumab (BV) with irinotecan has significant anti-tumor activity for recurrent glioblastoma. Carboplatin has anti-glioma activity and can potentiate the cytotoxicity of irinotecan. We evaluated the progression-free survival (PFS) of BV in combination to irinotecan and carboplatin in recurrent WHO grade 3 malignant gliomas, as well as its safety. Methods: Adult patients with measurable recurrent WHO grade 3 malignant glioma, ≥12 weeks after radiation therapy, ≥4 weeks after chemotherapy, with adequate organ function, KPS ≥70%, no prior failure or grade ≥3 toxicity to the agents, and no contraindications to BV were eligible for study. Patients received BV at 10 mg/kg with irinotecan on days 1 and 15 of a 28-day cycle. The dose of irinotecan was 125 mg/m2 for patients not on enzyme inducing anti-epileptics (EIAEDs) and 340 mg/m2 for patients on EIAEDs. All patients received carboplatin at an AUC of 4 on day 1 of each cycle. MRIs were obtained every 8 weeks. Results: As planned, 39 WHO grade III malignant glioma patients were enrolled on study. Median age was 47 (range, 26-71). At a median follow up of 14 months, the 6-month PFS is 69% and the median PFS is 11 months. A median of 8 cycles were given. Seven patients completed the planned course of treatment (12 cycles) with hypometabolic PET scan and nine patients remain on study. Fifteen patients came off study due to disease progression and eight due to toxicity. As of 1/25/2012, 22 patients are still alive and 17 have died. Grade 3-4 toxicities included: neutropenia (grade 3, n=12; grade 4, n= 1), thrombocytopenia (grade 3, n=6; grade 4, n=4), nausea (grade 3, n=7), diarrhea (grade 3, n=2), deep venous thrombosis (grade 3, n=2), febrile neutropenia (grade 3, n=1; grade 4, n=1), fatigue (grade 3, n=8; grade 4, n=1), cerebral infarction (grade 4, n=3), intracranial hemorrhage (grade 4, n=1), posterior reversible encephalopathy syndrome (grade 3, n=1). Conclusions: The combination of bevacizumab, irinotecan and carboplatin for WHO grade 3 malignant glioma patients is effective and with no more than expected toxicity.

BACKGROUND Despite an overall favorable prognosis, early recurrences occur in a significant proportion of WHO grade 1 meningiomas following surgical resection. Given a wide variety of institutional preferences, individualized patient circumstances, and a lack of level 1 evidence, there are no standardized guidelines for the treatment of recurrent WHO grade 1 meningiomas, in particular to the requisiteness of radiotherapy and its timing. Our study assessed the local tumor control after postoperative radiotherapy in recurrent molecularly high-risk WHO grade 1 meningiomas. MATERIAL AND METHODS Patients with primary WHO grade 1 meningioma and DNA methylation profiles (n=224) were retrospectively screened for recurrences (n=31). The local 3-year progression-free survival (PFS) after primary resection, as well as recurrence resection, was determined using Kaplan-Meier analysis. Integrated risk scores were calculated for primary and recurrent WHO grade 1 meningiomas. RESULTS Primary WHO grade 1 meningiomas were completely resected (GTR) in 81.3% (182/224). The 3-year PFS was 95.4% for molecularly prognostically favorable (IntS-low) versus 71.1% for prognostically unfavorable (Int-S-int/high) primary WHO grade 1 meningiomas. None of the patients received postop-RT after primary resection. After recurrence resection (n=31, Int-S-low: 14/31, Int-S-int/high: 17/31), 85% of subtotally resected meningioma recurrences received postop-RT. Notably, the local 3-year PFS for clinically-molecularly prognostically unfavorable recurrences (STR, Int-S-int/high) after postop-RT was 85.9%. CONCLUSION Postop-RT in molecularly prognostically unfavorable, recurrent WHO grade 1 meningiomas after STR may lead to improved local tumor control. Future matched-pair analyses with patients with recurrent WHO grade 1 meningiomas are warranted to further illustrate the benefit of postop-RT after recurrence resection.

e22538 Background: The purpose of the study was to determine DNA content and distribution of cells in cell cycle phases by flow cytometry in patients with primary and recurrent soft tissue sarcomas. Methods: 60 patients with soft tissue sarcomas (STS) were recruited: 30 with primary tumors and 30 with recurrent ones. Mean age of patients with primary STS was 56±5.4 years, with recurrent STS – 55±6.7 years. DNA content was determined using the BD Facs Cantoo II flow cytometer with CycleTEST PLUS DNA Reagent Kit (Becton Dickinson). The data were processed using ModFit LT program. Results: Comparative analysis of the cell cycle kinetics showed an increase in the percentage of cells in G2+M phase by 2 times in diploid and by 2.1 times in aneuploid recurrent tumors in comparison with primary ones (1.8±0.5% vs. 0.9±0.1% for diploid tumors; 5.4±2.2% vs. 2.6±0.7% for aneuploid tumors). An increase in the percentage of aneuploid tumors was found in recurrent G2 and G3 tumors (from 50% in primary to 66.7% in recurrent G2 tumors and from 63.25% in primary to 85% in recurrent G3 tumors). Mean content of aneuploid cells in recurrent G2 tumors was 2.2 times higher (p≤0.05), while the differences in primary and recurrent G3 tumors were not significant. The percentage of aneuploid tumors depended on the disease stage and increased in stages IIb and III in recurrent tumors, compared to primary ones (from 37.5% to 71.4% in recurrent st. IIb; and from 65% in primary st. III to 72.7% in recurrences) (p≤0.05). Conclusions: DNA analysis by flow cytometry demonstrated a high biologic potential of both primary and recurrent tumors. Some values in the mitotic cycle in recurrent tumors were probably associated with adjuvant therapy, as well as influenced by the coefficient of two parameters – the percentage of cells in G2+M phase and the cell loss factor determining a high malignant potential of these tumors.

Patients with glioblastoma almost always suffer a recurrence of an aggressive treatment resistant tumour and succumb within months highlighting the need to develop treatment options for recurrent glioblastoma. One potential target is prostate specific membrane antigen (PSMA) expressed on the new vessels in primary glioblastoma tumours. This study compared the expression of PSMA in primary and recurrent glioblastoma tumours. Formalin-fixed paraffin-embedded sections of primary and matched recurrent tumours from 13 patients with glioblastoma were processed for PSMA immunohistochemical labelling. PSMA expression was scored from digitally scanned sections using a categorical system (Total PSMA expression = PSMA label intensity (0–3) x percent PSMA-positive blood vessels). Little PSMA labelling was observed in any adjacent ‘unaffected’ brain tissue. PSMA was localised to blood vessels including those displaying microvascular proliferation in 12/13 primary and 9/13 recurrent glioblastoma tumours. Total PSMA expression scores ranged from 0 – 230 (maximum score = 300) and was divided into low and high expression based on a median split (median = 57) of PSMA expression across all tumours. Regression analysis showed a significant difference in PSMA expression between primary and recurrent glioblastoma (p = 0.04) with PSMA being highly expressed in ~70% of primary and only 31% of recurrent glioblastoma. Three cases displayed high expression in both primary and recurrent glioblastoma and one case had no PSMA expression at all. In conclusion, whilst higher expression of PSMA was detected in more primary tumours, 70%+ of all primary and recurrent tumours expressed PSMA to some extent adding evidence that this may be a useful target for treatment of glioblastoma. Larger cohorts and other techniques for detecting PSMA are needed to provide further evidence for PSMA’s utility as a target and to further define which glioblastoma patients are most likely to benefit from this approach.