Radiation-induced midbrain fibrosarcoma is a rare, highly aggressive tumor, which is associated with poor prognosis. We present the case of a 48-year old man with brainstem fibrosarcoma 20 years following radiation therapy received for a pituitary tumor. We discuss this case in the context of the diagnostic criteria for these tumors, and previous reports of secondary and primary sarcomas of the central nervous system.

e13535 Background: Glioblastoma (GBM), the most aggressive brain tumor in human patients, is highly heterogeneous and intensively vascularized. Glioma stem/initiating cells (GSC) are found to play a crucial role by increasing cancer aggressiveness and by promoting resistance to therapy. Recently, crosstalk between GSCs and vascular endothelial cells that line capillaries has been shown to considerably promote GSC self-renewal and tumor progression. GSCs have been shown to also transdifferentiate into bona-fide vascular endothelial cells (GEC). GECs inherit mutations present in GSCs and are resistant to traditional anti-angiogenic therapies. Methods: We develop a multispecies mathematical model to investigate the 3D spatiotemporal dynamics of vascularized GBM progression and response to cancer therapies. Results: The model predicts GSCs drive invasive fingering and that GECs spontaneously form a network within the hypoxic core, consistent with published experimental findings. We demonstrate that standard-of-care treatments using DNA-targeted therapy (radiation/chemo) together with anti-angiogenic therapies reduce GBM tumor sizes but increase invasiveness. Anti-GEC treatments block the GEC support of GSCs and reduce tumor sizes but can lead to increased invasiveness. Anti-GSC therapies that promote differentiation or disturb the stem cell niche effectively reduce tumor invasiveness and sizes, but are ultimately limited in reducing tumor sizes because GECs can maintain GSCs. Anti-GEC therapies are required to remove the tumor completely. Conclusions: Our results suggest that a combinatorial regimen targeting the vasculature, GSCs and GECs, using drugs already approved by the FDA, can reduce both tumor sizes and invasiveness and could lead to tumor eradication without recurrence when the treatment is stopped.

2057 Background: The optimal role of chemotherapy and radiation (RT) in adult low-grade glioma (LGG, WHO grade 1 & 2) is unclear. We conducted a systematic review and study-level meta-analysis of the literature on overall survival (OS) and progression free survival (PFS) in patients with LGG. Methods: Pubmed was queried with MeSH terms. All comparative studies of adults with newly diagnosed, supratentorial LGG were included. Comparisons of interest were OS and PFS at 2, 5, and 10 years in chemotherapy versus no chemotherapy and early RT versus delayed or no RT. Data were extracted from studies and synthesized with a random effects model. Quality of evidence was determined by American Academy of Neurology criteria and further analysis was performed, separating high quality (class I and II) from low quality (class III and IV) evidence. Numbers needed to treat (NNT) were determined from the risk difference. Results: 1531 articles were screened; 18 studies were included. Chemotherapy was not associated with a significant survival advantage compared to control. However, an analysis of high quality data revealed a survival advantage at 10 years associated with chemotherapy compared to control with NNT 5 (relative risk death chemo vs control 0.69 [0.56-0.86] p = 0.0006). Furthermore, NNT to prevent one progression with chemotherapy at 5 and 10 years was 6 and 3, respectively. Early RT was not associated with an OS advantage compared to control. However, early RT had progression benefit at all time points, with NNT of 10, 6, and 5 at 2, 5, and 10 years. Conclusions: Further study will be needed to confirm the optimal role of chemotherapy and RT. Caution must be used in interpretation as much of the literature consists of low-quality studies. [Table: see text]

2047 Background: Standard GBM therapy is associated with early progression and poor overall survival (OS). DC-ATA AV-GBM-1, a personal vaccine consisting of autologous DC pulsed with ATA, was investigated in a multicenter trial in patients with newly diagnosed GBM. Methods: Key eligibility criteria for surgical collection of tumor were clinical suspicion of new primary GBM & age 18-70 years. ATA lysate was prepared from irradiated tumor cells that were self-renewing in serum-free media. Autologous monocytes (MC) were collected by leukapheresis. Prior to initiating concurrent radiation therapy (RT) and temozolomide (TMZ), patients were enrolled with intent-to-treat (ITT) with DC-ATA after RT/TMZ. Eligibility included confirmation of primary GBM, availability of ATA & MC, KPS > 70 and plans for RT/TMZ. DC-ATA was manufactured during RT/TMZ. MC were differentiated into DC by culturing with IL-4 & GM-CSF, then DC were incubated with ATA. DC-ATA was suspended in 500 mg GM-CSF just prior to s.c. injections at weeks 1, 2, 3, 8, 12,16, 20, & 24 (8 doses). Patients were not excluded based on apparent disease progression or PFS. Standard adjuvant TMZ regimens were started after the 3 weekly injections. Primary endpoint was > 75% OS 14.6 months from ITT enrollment. Secondary endpoints included median OS & progression-free survival (PFS). Results: Cell line and MC collection were successful for 97% of patients. Median age of the 60 ITT enrollees was 59 years. 3 patients withdrew before starting DC-ATA; 57 received 392 injections; 68% received all 8. Most common AE attributed to DC-ATA were local injection site reactions (16%) & flu-like symptoms (10%), but 33% experienced seizures. After 3 years of follow up, OS at 14.6 mos is 52.7% (95% CI 39.8,65.8), median OS 16.0 mos (95% CI 12.9,21.7) & median PFS 10.4 mos (95% CI 8.6,11.6). OS rates at 1, 2, & 3 years are 70.1%, 32.4%, & 23.2%. Longer OS was associated with 8 DC-ATA doses (p < 0.0001), on < 2 mg/day dexamethasone (dex) at start of DC-ATA (p = 0.005), > 6 cycles of adjuvant TMZ (p = 0.0054), & KPS 90 or 100 (p = 0.010) at enrollment. Independent variables per multivariate Cox regression analysis were 8 DC-ATA doses, dex dose, IDH mutated, TMZ > 6 cycles, & MGMT promoter methylated. Concurrent TMZ regimens included TMZ alone (n = 28), TMZ + anti-VEGF(n = 14), & TMZ + tumor treating fields (TTF) (n = 10); 8 received no concurrent TMZ. OS was longer in patients treated with concurrent TMZ alone compared to no TMZ (p = 0.0003), TMZ + anti-VEGF (p = 0.045) or TMZ + TTF (p = 0.045). The only common features among 7 patients progression-free at 3 years are 8 DC-ATA injections, age < 60, & < 2 mg dex. Conclusions: DC-ATA was reliably produced and injections well-tolerated in combination with various TMZ-based regimens, but the primary OS objective was not achieved. PFS was encouraging, but did not translate into improved OS, perhaps because DC-ATA was limited to 8 injections. Clinical trial information: NCT03400917 .

Recurrent glioblastoma multiforme (GBM) is a highly aggressive cancer with poor prognosis, and an overall survival of 6 to 7 months with optimal therapies. The NovoTTF-100A™ System is a novel antimitotic cancer therapy recently approved for the treatment of recurrent GBM, based on phase III (EF-11) trial results. The Patient Registry Dataset (PRiDe) is a post-marketing registry of all recurrent GBM patients who received NovoTTF Therapy in a real-world, clinical practice setting in the United States between 2011 and 2013. Data were collected from all adult patients with recurrent GBM who began commercial NovoTTF Therapy in the United States between October 2011 and November 2013. All patients provided written consent before treatment was started. Overall survival (OS) curves were constructed for PRiDe using the Kaplan-Meier method. Median OS in PRiDe was compared for patients stratified by average daily compliance (≥75% v<75% per day) and other prognostic variables. Adverse events were also evaluated. Data from 457 recurrent GBM patients who received NovoTTF Therapy in 91 US cancer centers were analyzed. More patients in PRiDe than the EF-11 trial received NovoTTF Therapy for first recurrence (33% v 9%) and had received prior bevacizumab therapy (55.1% v 19%). Median OS was significantly longer with NovoTTF Therapy in clinical practice (PRiDe data set) than in the EF-11 trial (9.6 v 6.6 months; HR, 0.66; 95% CI, 0.05 to 0.86, P = .0003). One- and 2-year OS rates were more than double for NovoTTF Therapy patients in PRiDe than in the EF-11 trial (1-year: 44% v 20%; 2-year: 30% v 9%). First and second versus third and subsequent recurrences, high Karnofsky performance status (KPS), and no prior bevacizumab use were favorable prognostic factors. No unexpected adverse events were detected in PRiDe. As in the EF-11 trial, the most frequent adverse events were mild to moderate skin reactions associated with application of the NovoTTF Therapy transducer arrays. Results from PRiDe, together with those previously reported in the EF-11 trial, indicate that NovoTTF Therapy offers clinical benefit to patients with recurrent GBM. NovoTTF Therapy has high patient tolerability and favorable safety profile in the real-world, clinical practice setting.