Daniela A. Bota

BACKGROUND: -L) to standard therapy for newly diagnosed glioblastoma. METHODS: After surgery and chemoradiotherapy, patients were randomized (2:1) to receive temozolomide plus DCVax-L (n = 232) or temozolomide and placebo (n = 99). Following recurrence, all patients were allowed to receive DCVax-L, without unblinding. The primary endpoint was progression free survival (PFS); the secondary endpoint was overall survival (OS). RESULTS: For the intent-to-treat (ITT) population (n = 331), median OS (mOS) was 23.1 months from surgery. Because of the cross-over trial design, nearly 90% of the ITT population received DCVax-L. For patients with methylated MGMT (n = 131), mOS was 34.7 months from surgery, with a 3-year survival of 46.4%. As of this analysis, 223 patients are ≥ 30 months past their surgery date; 67 of these (30.0%) have lived ≥ 30 months and have a Kaplan-Meier (KM)-derived mOS of 46.5 months. 182 patients are ≥ 36 months past surgery; 44 of these (24.2%) have lived ≥ 36 months and have a KM-derived mOS of 88.2 months. A population of extended survivors (n = 100) with mOS of 40.5 months, not explained by known prognostic factors, will be analyzed further. Only 2.1% of ITT patients (n = 7) had a grade 3 or 4 adverse event that was deemed at least possibly related to the vaccine. Overall adverse events with DCVax were comparable to standard therapy alone. CONCLUSIONS: Addition of DCVax-L to standard therapy is feasible and safe in glioblastoma patients, and may extend survival. Trial registration Funded by Northwest Biotherapeutics; Clinicaltrials.gov number: NCT00045968; https://clinicaltrials.gov/ct2/show/NCT00045968?term=NCT00045968&rank=1 ; initially registered 19 September 2002.

<h3>BACKGROUND AND PURPOSE:</h3> The World Health Organization has recently placed new emphasis on the integration of genetic information for gliomas. While tissue sampling remains the criterion standard, noninvasive imaging techniques may provide complimentary insight into clinically relevant genetic mutations. Our aim was to train a convolutional neural network to independently predict underlying molecular genetic mutation status in gliomas with high accuracy and identify the most predictive imaging features for each mutation. <h3>MATERIALS AND METHODS:</h3> MR imaging data and molecular information were retrospectively obtained from The Cancer Imaging Archives for 259 patients with either low- or high-grade gliomas. A convolutional neural network was trained to classify <i>isocitrate dehydrogenase 1</i> (<i>IDH1</i>) mutation status, 1p/19q codeletion, and <i>O6-methylguanine-DNA methyltransferase</i> (<i>MGMT</i>) promotor methylation status. Principal component analysis of the final convolutional neural network layer was used to extract the key imaging features critical for successful classification. <h3>RESULTS:</h3> Classification had high accuracy: <i>IDH1</i> mutation status, 94%; 1p/19q codeletion, 92%; and <i>MGMT</i> promotor methylation status, 83%. Each genetic category was also associated with distinctive imaging features such as definition of tumor margins, T1 and FLAIR suppression, extent of edema, extent of necrosis, and textural features. <h3>CONCLUSIONS:</h3> Our results indicate that for The Cancer Imaging Archives dataset, machine-learning approaches allow classification of individual genetic mutations of both low- and high-grade gliomas. We show that relevant MR imaging features acquired from an added dimensionality-reduction technique demonstrate that neural networks are capable of learning key imaging components without prior feature selection or human-directed training.

Importance: Glioblastoma is the most lethal primary brain cancer. Clinical outcomes for glioblastoma remain poor, and new treatments are needed. Objective: To investigate whether adding autologous tumor lysate-loaded dendritic cell vaccine (DCVax-L) to standard of care (SOC) extends survival among patients with glioblastoma. Design, Setting, and Participants: This phase 3, prospective, externally controlled nonrandomized trial compared overall survival (OS) in patients with newly diagnosed glioblastoma (nGBM) and recurrent glioblastoma (rGBM) treated with DCVax-L plus SOC vs contemporaneous matched external control patients treated with SOC. This international, multicenter trial was conducted at 94 sites in 4 countries from August 2007 to November 2015. Data analysis was conducted from October 2020 to September 2021. Interventions: The active treatment was DCVax-L plus SOC temozolomide. The nGBM external control patients received SOC temozolomide and placebo; the rGBM external controls received approved rGBM therapies. Main Outcomes and Measures: The primary and secondary end points compared overall survival (OS) in nGBM and rGBM, respectively, with contemporaneous matched external control populations from the control groups of other formal randomized clinical trials. Results: A total of 331 patients were enrolled in the trial, with 232 randomized to the DCVax-L group and 99 to the placebo group. Median OS (mOS) for the 232 patients with nGBM receiving DCVax-L was 19.3 (95% CI, 17.5-21.3) months from randomization (22.4 months from surgery) vs 16.5 (95% CI, 16.0-17.5) months from randomization in control patients (HR = 0.80; 98% CI, 0.00-0.94; P = .002). Survival at 48 months from randomization was 15.7% vs 9.9%, and at 60 months, it was 13.0% vs 5.7%. For 64 patients with rGBM receiving DCVax-L, mOS was 13.2 (95% CI, 9.7-16.8) months from relapse vs 7.8 (95% CI, 7.2-8.2) months among control patients (HR, 0.58; 98% CI, 0.00-0.76; P < .001). Survival at 24 and 30 months after recurrence was 20.7% vs 9.6% and 11.1% vs 5.1%, respectively. Survival was improved in patients with nGBM with methylated MGMT receiving DCVax-L compared with external control patients (HR, 0.74; 98% CI, 0.55-1.00; P = .03). Conclusions and Relevance: In this study, adding DCVax-L to SOC resulted in clinically meaningful and statistically significant extension of survival for patients with both nGBM and rGBM compared with contemporaneous, matched external controls who received SOC alone. Trial Registration: ClinicalTrials.gov Identifier: NCT00045968.