MONTREAL – An investigational treatment for recurrent glioblastoma that delivers alternating electric fields through scalp electrodes has shown signs of improved survival in a post hoc analysis of results in particular subgroups of patients enrolled in a phase III trial.
Quality of life outcomes also favored patients who used NovoTTF, compared with those who received chemotherapy.
To date, reports about the device have elicited both antagonistic and enthusiastic reaction from oncologists, with "neither the enthusiasts nor the antagonists having significant basis for either kind of acute reaction," Dr. Zvi Ram said in an interview after presenting the subgroup analyses at the annual meeting of the Society for Neuro-Oncology. "I think it is exciting that we’re getting something completely new – a different, noninvasive modality with no side-effects. I think we should be exhilarated."
The results obtained with a modified version of the device in a separate trial of patients with non–small cell lung cancer (NSCLC) suggest that the therapeutic effects of the device also may extend to other cancers, according to Dr. Ram, professor and chair of neurosurgery at Tel Aviv (Israel) Medical Center.
The device, known as NovoTTF-100A, delivers low-amplitude "tumor treatment fields" ranging from 100 to 300 kHz that have been shown in vitro to slow and reverse tumor cell proliferation by inhibiting mitosis, according to NovoCure Ltd., the manufacturer of the device and sponsor of the trial.
The portable device weighs about 6 pounds and connects to a battery pack. It is designed to be worn almost constantly, with a target of at least 20 hours of use each day.
In a phase III clinical trial presented earlier this year at the American Society of Clinical Oncology, an intent-to-treat analysis comparing NovoTTF vs. best-available chemotherapy found no statistical difference in 1-year overall survival (OS) among 237 recurrent glioblastoma patients randomized to either treatment.
However, a per-protocol analysis (which included only those patients who wore the device for at least 70% of the recommended time during the first month) showed a statistically significant benefit to NovoTTF in 1-year survival, compared with chemotherapy (29.5% vs. 19.1%, respectively; hazard ratio, 0.64; P = .01).
In the new post hoc analysis, a subgroup of 110 patients with a "good prognosis" (aged younger than 60 years, and with a Karnofsky performance status score greater than 80%) showed a "more robust" survival benefit than that seen in the overall intent-to-treat analysis, he said.
In this subgroup, patients who were treated with NovoTTF had a median survival of 9.2 months, compared with 6.6 months in those treated with chemotherapy (P less than .01). However, in the overall intent-to-treat group, median survival was 6.6 months and 6.0 months, respectively, he explained. Moreover, the 1-year OS in this subgroup was significantly higher in the NovoTTF group than in the chemotherapy group (35.2% vs. 20.8%, respectively; P less than .01), whereas the difference was nonsignificant in the larger analysis (23.6% vs. 20.7%).
Another subgroup analysis looked at patients who had previously failed treatment with bevacizumab (roughly 20% of the entire cohort). Both an intent-to-treat analysis and a per-protocol analysis showed significant OS advantages to NovoTTF, Dr. Ram said.
The median OS among 44 patients in the intent-to-treat group was 4 months with NovoTTF vs. 3.1 months with chemotherapy (HR, 0.43; P less than 0.02). NovoTTF also gave a significantly better median OS among 29 patients in the per-protocol analysis for this subgroup (6.3 months vs. 3.3 months; HR, 0.21; P = .02).
"You don’t see this anywhere," he said. "There’s no drug in the world that could produce such response in patients who had already failed" bevacizumab.
The investigators also analyzed a surgery-naive group. "You know these are going to be poor responders, almost identical to [those with] bevacizumab failure," Dr. Ram commented.
In this group of 38 patients, an intent-to-treat analysis showed that overall survival was 9.8 months with NovoTTF vs. 5.5 months with chemotherapy.
Patients also reported significantly better quality of life with NovoTTF than with chemotherapy. On the Quality of Life Symptom Scale, NovoTTF patients scored –34 and –35 on constipation and diarrhea, compared with scores of +77 and +50 for the chemotherapy group. Nausea and vomiting scores were 15 for the NovoTTF group and 61 for the chemotherapy group, and pain scores were –1 for the NovoTTF group and +63 for the chemotherapy group.
A quality of life analysis using the EORT (European Organization of Research and Treatment) QLQ-C30 instrument showed scores of 14 vs. –7 in favor of NovoTTF for cognitive functioning, and scores of 7 vs. 1 in favor of NovoTTF for emotional functioning.
"We do this to all our patients. We intoxicate them," Dr. Ram said. "So even if NovoTTF did not extend survival, if it was equivalent to chemotherapy [for survival], it may still improve quality of life."
Dr. Ram did not know the median length of time that the NovoTTF cohort wore the device, but an earlier phase II study followed some patients up to 59 months. He noted that "70% are still alive; that’s unheard of."
"There were concerns that patients might have more headaches or seizures, but there were none," he added.
In the current study, the rate of adverse events related to the central nervous system was similar (66% for NovoTTF and 67% for chemotherapy). Seizures occurred in 15% of the NovoTTF group and 12% of the chemotherapy group, and headaches occurred in 18% and 13%, respectively.
Dr. Ram also presented evidence to suggest that NovoTTF therapy may have benefits in other forms of cancer. A study reported by his colleagues earlier this year at the European Society for Medical Oncology Congress in Milan showed that NovoTTF therapy (combined with chemotherapy) resulted in significant prolongation of survival in patients with NSCLC, compared with historical controls, he said.
"If this kind of therapy acts against brain cancer cells, it should act also against other tumor types," he reasoned.
In the study, which included 42 NSCLC patients, NovoTTF was delivered with newly designed electrodes placed on the chest and neck of patients with locally advanced, metastatic, stage IIIb and IV disease, he explained. The combination of NovoTTF and pemetrexed increased overall survival to 13.8 months, compared with 8.3 months seen with pemetrexed alone; the 1-year survival rate for the combination was 57%, compared with 30% reported for pemetrexed alone.
"We’re talking about something that appears to be acting against cancer cells regardless of origin," Dr. Ram said. "Action in the lung seems similar to what has been seen in GBM over time: a slow resolution of malignant pleural effusion and masses within the chest."
"It’s very interesting and exciting, even if we do not yet have enough definitive data," commented Dr. Alba B. Brandes, moderator of the session and the chair of medical oncology at Azienda USL, a group of nine hospitals in and around Bologna, Italy.
The study investigators have been criticized for repackaging their nonsignificant intent-to-treat results into per-protocol results that show significance, she said. "An intention-to-treat population and per-protocol population are two different things, and from a statistical point of view, it is sometimes difficult for the oncologic community to accept."
Despite those reservations, she said that the per-protocol observations should not be dismissed, because when they are analyzed in this way the results are highly significant. "I was really surprised to see what happened in the lung cancer. I am a medical oncologist and I have never seen that complete a response. It’s surprising. We have to wonder if all that we know about the treatment of tumors is correct."
Dr. Ram acknowledged that per-protocol analysis of the findings is unconventional, but "there is no precedent for this kind of therapy and I think we may need to redesign the way we assess results in the future. We cannot use the same guidelines and definitions that we were traditionally using."
Dr. Ram disclosed that he is a consultant for NovoCure. Dr. Brandes did not disclose if she had any conflicts of interest at the meeting and did not respond to inquiries about disclosures.