An investigation into the impact of next generation sequencing on the use of targeted treatments in glioblastoma.

Abstract

2035 Background: Next-generation sequencing (NGS) provides clinicians immense amounts of information about a patient’s cancer. NGS allows the detection of a wide range of gene alterations undetectable by older techniques. This has the promise to help guide clinicians in their decision-making as genetic alterations can be both prognostic and predictive. NGS may reveal new treatment options for patients with glioblastoma. The goal of this project was to evaluate how often NGS results influence the use of targeted treatment agents in glioblastoma. This study did not investigate the impact of NGS on reclassification of cancers, prognosis, or treatment decisions outside of the use of targeted therapy. Methods: We conducted a retrospective chart review to see if adult glioblastoma patients received treatments for potentially actionable gene alterations and for fusion variants because of NGS. Here we looked at alterations labeled as actionable from the Tempus company’s NGS results. We collected diagnosis and treatment information from the Electronic Medical Record and from the Electronic Data Warehouse. We examined if patients after receiving NGS would receive targeted treatment, as that likely indicated these treatments were given due to NGS indication. We excluded cytotoxic chemotherapy agents and bevacizumab as these therapies are utilized regardless of a targeted gene indication. This analysis excluded treatments received in clinical trials in which enrollment was independent of NGS. This analysis looked at the proportion of patients with actionable alterations that were treated with targeted agents. Results: 261 glioblastoma patients were found on NGS to have mutations which were potentially actionable. Thirty-three of these patients (12.6%) received a respective targeted therapy, and the other 228 patients (87.4%) did not receive an NGS guided targeted therapy. 97 patients had EGFR copy number gains or EGFR gain of function mutations of which 21 were treated with depatuxizumab mafodotin, and one additional patient who was treated with ABBV 321. Of the NGS treatment guided therapies, depatuxizumab mafodotin was the most used targeted agent in our sample set. Additionally, there were 30 patients with fusion variants and 11 of them were FGFR3-TACC3 fusions. Of these 11, 4 received targeted regimens (36.4%) with either TAS-120, pemigatinib, erdafitinib, or erdafitinib with ponatinib. The other fusion variants were not acted upon. Conclusions: These preliminary results suggest that NGS data currently does not frequently impact treatment decisions for glioblastoma. Although this study is limited by being a single institution study, future efforts include expanding this analysis to multiple institutions. With improvements in therapeutics as well as with wider availability of NGS testing, sequencing data may impact a larger percentage of glioblastoma patients.