Alternating Electric Fields for the Treatment of Glioblastoma.

Abstract

Glioblastomais themostcommonmalignant tumorof thecentral nervous system,1 and one of themost difficult cancers to treat. Standard therapy includes maximal surgical resection, high-dose external-beam radiation therapy, and regional or systemic chemotherapy. Still, 5-year survival rates are only 5%.1 Current therapies lack specificity, fail toaddress tumor heterogeneity, or are limitedbecauseofan inability topenetrate theblood-brainbarrier. Because conventional strategies lack success, investigational approaches,2-6 with varying degrees of supportive evidence, are oftenused.Historically, advances in this field have been separated by decades of failure. Life expectancy for patients with glioblastoma has changed little during the past decade, andmedian survival remains at less than 15months.7 In this issue of JAMA, Stupp and colleagues8 report their evaluation of a novel approach to the treatment of glioblastomausing transcutaneousdelivery of low-intensity intermediate-frequency alternating electric fields (AEFs), also referred to as tumor-treating fields (TTFields). In this multisite randomized clinical trial, adult patients with supratentorial glioblastoma who had no evidence of tumor progression following the completion of standard chemoradiotherapy were randomized (2:1) to receivemaintenance treatmentwitheither TTFields and temozolomide (n = 466) or temozolomide alone (n = 229),withmedian time fromdiagnosis to randomization of 3.8 months in both groups. The study was not blinded because a sham treatment was considered inappropriate. The trial was terminated as a result of a planned interim analysis demonstrating a benefit in progression-free survival (PFS) in the intent-to-treat population, which was the primaryendpoint of the study.The interimanalysis included210 patients randomized to theTTFields and temozolomidegroup and105patients randomizedto the temozolomidealonegroup. Median PFS in the intent-to-treat population was 7.1 months in theTTFields and temozolomide groupvs 4.0months in the temozolomide group (hazard ratio, 0.62 [98.7% CI, 0.430.89]; P = .001). Overall survival, a powered secondary end point prespecified to be based on the per-protocol population, was also significantly enhancedwhen study accrual was halted.Medianoverall survivalwas20.5months intheTTFields and temozolomide group vs 15.6 months in the temozolomide alone group (hazard ratio, 0.64 [99.4% CI, 0.42-0.98]; P = .004). The robustness of this interim analysis was supported by additional analyses on all 695 patients randomized in the study todate (with final analysisplannedafter follow-up data collection is completed in theentire studypopulation).As expected froma locallydelivered therapy, thedeliveryofAEFs was not associated with any significant increase in systemic toxic effects, though there was a higher incidence of scalp irritation, anxiety, confusion, insomnia, and headaches. The incidence of seizures was not increased. It has been hypothesized that the polarity of AEFs at specific frequencies candisrupt spindle formationduring cell divisionand lead tomitotic arrest.WhenAEFsare applied tohumanand rodent cancer cell lines in vitrousing insulatedwires fixed to the bottom of standard cell culture dishes, there was a significant inhibition of cellular proliferation across all cell lines tested.9 Microphotography showed examples of prolongedmitoseswithnuclear rotation, proliferation arrest, and apoptotic cell death supporting thehypothesis. Inmice, AEFs generated by intradermal insulated wires placed alongside subcutaneous tumors inhibited growth.9 Similarly, growth of orthotopic rat gliomas was inhibited using 3 electrodes positioned on the head that generated a calculated 1 to 2 V/cm at the tumor when accounting for the impedance of the electrode insulation and the rat head.10 Further studies, again by the same group,11 demonstrated an additive effect with some chemotherapies, although temozolomide, the specific chemotherapyused to treat patients in this study,wasnot tested. Overall, even though the ability of AEFs to inhibit cell growth is apparent in some situations and appears to be dependent on the frequency, intensity, anddirectionof theelectrical field, the specificity of this effect for cancerous cells has not been established and the exact mechanism still remains unclear. Earlyhumanstudies supportedbyNovocureLtd, thecompany marketing the TTFields device, were encouraging. In a single group pilot trial of 10 patients with recurrent glioblastoma, the scalps of patientswere coveredwith insulated electrodes and 50 Vwas applied.10 This was designed to produce 1 to 2V/cmdeepwithin thebrain basedon3-Dphantomsimulations that were apparently validated in 1 patient undergoing surgery for hydrocephalus. A frequency of 200 kHz was used because thiswas thought to be the optimal frequency in rat and human experimental glioma models. The field directionsalternated in2perpendiculardirections.ThemedianPFS and overall survival for treated patients was encouraging at 26.1 weeks (range, 3-124weeks) and 62.2 weeks (range, 20.3124.0 weeks), respectively. A subsequent phase 3 trial also in patients with recurrent glioblastoma was not as convincing. Patients were randomized to receive TTFields using the NovoTTF-100A device (20-24 h/d) or a physician’s choice of chemotherapy. The deviceproducedno increase inoverall survival in this studywith a power of 80% to detect a hazard ratio for death of 0.63, althoughmorepartialorcompleteradiological responseswereobAuthor Audio Interview at jama.com