The addition of tumor-treating fields to maintenance temozolomide chemotherapy significantly extended PFS and OS among patients with glioblastoma who completed standard chemoradiation, according to interim results from a randomized clinical trial.
Glioblastoma has a high mortality rate, with the majority of patients dying within 2 years of diagnosis, according to study background.
“Tumor-treating fields [Optune, Novocure Ltd.] are an antimitotic treatment that selectively disrupts the division of cells by delivering low-intensity, immediate-frequency alternating electric fields via transducer arrays applied to the shaved scalp,” Roger Stupp, MD, professor and chair of oncology at University Hospital Zurich, and colleagues wrote. “In a randomized phase 3 trial in which tumor-treating fields were compared with chemotherapy in 237 patients with recurrent glioblastoma, the use of tumor-treating fields did not prolong PFS or OS, but the therapy was associated with better quality of life without the typical chemotherapy-associated toxic effects.”
Stupp and colleagues sought to observe the safety and efficacy of tumor-treating fields in combination with maintenance temozolomide (Temodar, Merck) following chemoradiation in 695 patients with glioblastoma from 83 international treatment centers.
Researchers randomly assigned patients to maintenance treatment with temozolomide alone (n = 229) or with tumor-treating fields (n = 466).
Patients received 150 mg/m2 to 200 mg/m2 daily temozolomide for 5 days of each 28-day treatment cycle for six to 12 cycles. Delivery of tumor-treating fields occurred continuously (> 18 hours per day) via four transducer arrays placed on the shaved scalp and connected through a portable medical device.
PFS served as the primary endpoint. OS served as a secondary endpoint.
Median follow-up at the time of the interim analysis — which included data from 315 patients (tumor-treating fields plus temozolomide, n = 210; temozolomide monotherapy, n = 105) — was 38 months (range, 18-60).
Patients who received tumor-treating fields experienced significantly longer median PFS (7.1 months vs. 4 months; HR = 0.62; 98.7% CI, 0.43-0.89) and OS (20.5 months vs. 15.6 months; HR = 0.64; 99.4% CI, 0.42-0.98).
Tumor-treating fields did not appear to significantly increase the occurrence of adverse events compared with temozolomide alone. Forty-three percent of patients assigned tumor-treating fields reported mild to moderate skin irritation, with grade 3 skin reactions in 2% of patients.
Twelve patients died (tumor-treating fields, n = 8; temozolomide, n = 4) due to unrelated causes while receiving adjuvant therapy.
Based on these results, the study was terminated early. These data served the basis for the FDA’s approval of tumor-treating fields in combination with temozolomide for newly diagnosed patients with glioblastoma in October 2015.
Because tumor-treating fields therapy continued after progression, researchers acknowledged these data could be limited by a reporting bias for second-line therapies after tumor progression.
“Treatment failure patterns, effects of second-line therapies, and additional molecular analyses on baseline tumor biopsies will allow for better understanding of the clinical effects of this novel treatment modality,” Stupp and colleagues wrote. “With the last patient randomized on Nov. 29, 2014, these data are not expected before the end of 2016.”
Due to this efficacy, more work needs to be done to examine how tumor-treating fields work, John H. Sampson, MD, PhD, MBA, MHSc, professor and chief of neurosurgery at Duke University School of Medicine, wrote in an accompanying editorial.
“The tumor-treating fields device produces locally delivered alternating electric fields that are purported to arrest mitosis in tumor cells deep inside the brain,” Sampson wrote. “The mechanisms whereby this novel approach can treat tumors and leverage chemotherapy, however, remain unclear. Given the survival benefit reported in this study, it should now be a priority to understand the scientific basis for the efficacy of tumor-treating fields.”
However, the design of the study — including that there was not a placebo arm, and that patients in the control group received less adjuvant chemotherapy — may allow doubts about this therapy to persist, Sampson added.
“If tumor-treating field therapy fails to be adopted, will this decision be attributed to professional parochialism or to data that are not trusted?” Sampson wrote. “The current study provides additional important data on a novel device for the treatment of glioblastoma, but will not completely resolve that debate.” – by Cameron Kelsall
Disclosure: The study was funded by Novocure. Stupp reports travel expenses from Novocure, as well as personal fees from Merck, Novartis and Roche/Genentech. Please see the full study for a list of all other researchers’ relevant financial disclosures. Sampson reports personal fees from Brainlab, Bristol-Myers Squibb and Celldex Therapeutics, as well as patents with Annias Immunotherapeutics and Celldex Therapeutics.
Ashley L. Sumrall
In 2005, Roger Stupp, MD,
and colleagues revolutionized the management of newly diagnosed glioblastoma. In 2015, with the cooperation of neuro-oncologists around the world, he further advanced the care of patients with this deadly disease. The prolongation of PFS and OS demonstrated in this report has not been seen with any other therapy tested in this setting for a decade. Despite multiple attempts by many talented clinicians using a wide variety of pharmacologic interventions, glioblastoma remains a formidable opponent that requires novel treatment approaches to control tumor growth.
Although prior reports by Kirson and colleagues suggested safe combination of cytotoxics with tumor-treating fields (TTFields) in animal models and patients with recurrent glioblastoma, these numbers were rather small. It is reassuring to finally see the safety data from over 200 patients. No significant increase in seizures or infections were noted. Reported severe skin reactions were quite low, which is reassuring for patients who have just completed radiation. Many clinicians have expressed concern that this group of patients could be at higher risk for skin complications due to this.
There are some possible confounding factors in this trial, such as selection bias and the lack of a placebo arm. Crossover and broad choices of second-line therapies — including use of bevacizumab (Avastin, Genentech/Roche) — also make data interpretation challenging. Further, it is noteworthy that the temozolomide group received fewer cycles of chemotherapy than the combination therapy arm. It is unclear how this impacted results.
Additional limitations of this trial involve radiographic interpretation. There was central review of MRI, but the older Macdonald criteria were used. The phenomenon of “treatment effect” on initial scans while using TTFields has been observed in previous clinical trials and is poorly understood. This process is akin to the changes clinicians identify as pseudoprogression for MGMT
-methylated tumors, but has a different mechanism. It is not clear how this was addressed in this trial, and if therapy was prematurely changed or interrupted due to misinterpretation of imaging findings by clinicians.
Lastly, it is worth mentioning that 43% of patients demonstrated “mild to moderate skin irritation” in the combination arm. Only 2% experienced severe skin reactions. In previous reports of patients undergoing therapy with TTFields, dermatologic adverse events were much lower (EF-11 trial, 16%; post-marketing experience, 21.8%). The Common Terminology Criteria for Adverse Events dermatologic toxicity grading system is a poor metric for these lesions. The skin reactions produced by TTFields are due to ongoing contact between the array, hydrogel and scalp for extended periods, and they vary significantly in evaluation and management. They are best described using criteria proposed by Lacouture and colleagues in 2014 as allergic and irritant dermatitis, mechanical lesions, ulcers and skin infection. Therapy must often be interrupted for these conditions, and patients may experience multiple subtypes of skin toxicity. The incidence and description of skin toxicity, the primary adverse event for TTFields, is poorly described.
This is particularly important as skin toxicity affects compliance with therapy. It is recommended that patients use TTFields for an average of 18 of 24 hours per day to obtain the benefit in survival.
These issues aside, it is always exciting to broaden our treatment options for glioblastoma. This is one more option to present to patients. Moving forward, we will need to more aggressively incorporate TTFields into clinical trials. Eligibility criteria will need to be carefully considered, and decisions to include or continue TTFields in patients with progression will prove to be important.
As a community, we will eagerly look forward to more data on this novel therapy.
Kirson ED, et al. Proc Natl Acad Sci USA
Lacouture ME, et al. Semin Oncol