Society for Neuro Oncology annual meeting, San Francisco California, November 16-19 2017
Phase III trial of CCNU/temozolomide (TMZ) combination therapy vs. standard TMZ therapy for newly diagnosed MGMT-methylated glioblastoma patients: the CeTeG/NOA-09 trial [abstract ACTR-58]
Ulrich Herrlinger of the Neurooncology Working Group (NOA) of the German Cancer Society presented results of the CeTeG trial (also known as NOA-09). CeTeG is a randomized phase 3 trial for newly diagnosed glioblastoma with methylated MGMT promoter, testing CCNU (lomustine) combined with temozolomide (TMZ) versus TMZ alone. This trial was conducted at 17 centers in Germany and was a follow-up to a non-randomized phase 2 trial which published results in 2006 and 2009. The CeTeG trial was relatively small for a phase 3 trial, with a sample size calculation of 128 patients total, and this sample size was based on expectations of a significant increase in survival rate at 2 years as seen in the phase 2 trial compared to historical controls.
Patients in this trial had relatively good prognosis, with high rates of complete resection (61% of all patients had complete resection) and high average KPS (82% of patients had KPS of 90-100). Overall the arms were well balanced, with the only significant imbalance between the two arms being gender, which was not prognostically relevant.
In the combination arm receiving CCNU + TMZ, cycles were 6 weeks in length, with 100 mg/m2 oral CCNU given on day 1 of each cycle and TMZ on days 2-6 of each cycle, with a starting TMZ dose of 100 mg/m2 and possible escalation up to 200 mg/m2 in later cycles. Cycle 1 starts at the same time as radiation.
In the control arm of TMZ alone, cycles were 4 weeks in length, and used the standard TMZ schedule (daily at a dose of 75 mg/m2 during radiation, and 150 mg/m2 on days 1-5 of the first adjuvant cycle and possible escalation up to 200 mg/m2 in later cycles.
Importantly, this trial achieved its primary endpoint of increased overall survival. Survival in the CCNU + TMZ arm was statistically superior to TMZ alone, with a p value of 0.049. Hazard ratio for death from any cause was 0.6 in the CCNU + TMZ arm.
Median reported survival was 46.9 months for CCNU + TMZ versus 30.4 months for TMZ alone, a difference of 16.5 months. As seen in the Kaplan-Meier survival estimates, the curves did not separate until after the 2 year mark. 1, 2, 3, 4, and 5 year survival rate was 88.8, 71.4, 57.4, 48.8 and 34% in the CCNU + TMZ arm versus 84.4, 65.4, 42.3, 31.4 and 27.7% in the TMZ arm. Differences in the survival rate between the two arms were greatest at the 3 and 4 year mark, with 15% more patients surviving to 3 years and 17.4% more patients surviving to 4 years in the combination arm versus the TMZ alone arm. The survival curves then tightened up at 5 years with only 6.3% more patients surviving in the CCNU + TMZ arm.
Given the significant overall survival differences, it's surprising to note that progression-free survival was not significantly different between the two arms (p=0.41), although the progression-free survival curves separated somewhat at about 2 years (a phenomenon also seen in the overall survival curves), after which time CCNU + TMZ shows a slight superiority over TMZ alone. Some potential explanations given by the authors for the lack of a strong PFS signal included: "problems with PFS assessment according to RANO?" including potential undetected pseudoprogressions; and "long-term effects of CCNU?", noting that in studies of low grade gliomas treated with CCNU, delayed responses were sometimes seen months or years after the end of therapy.
The percentage of patients receiving further lines of therapy after progression was similar in both arms (59.1% in the CCNU + TMZ arm, 63.5% in the TMZ arm). The only salvage therapy that favored the CCNU + TMZ arm was re-resection: more patients underwent re-resection in the CCNU + TMZ arm (31.8% versus 22.2%). More patients received re-irradiation in the TMZ alone arm (23.8% versus 18.2%). More patients in the TMZ alone arm received further chemo or targeted agent therapy (60.3% versus 48.5%). The percentage of patients receiving bevacizumab after progression was similar in both arms (27% in the TMZ alone arm, 30.3 % in the TMZ + CCNU arm). The authors concluded that differences in treatment after progression are not an explanation for the superior survival in the TMZ + CCNU arm.
Combination therapy with TMZ + CCNU approximately doubled the rate of low-grade (but not high-grade) hematoxicity (including neutropenia and thrombocytopenia) and nausea. However no deaths due to treatment toxicity were observed, and no severe infections, liver failure, or lung fibrosis. More brain edema was observed in the combination arm, and more low-grade alopecia (patchy hair loss).
The authors concluded by noting that acute toxicity of the combination treatment was rare, and importantly, "the primary aim of CeTeG/NOA-09 was achieved: the OS superiority of CCNU/TMZ for MGMT promoter methylated newly diagnosed GBM could be demonstrated".
This was certainly one of the most significant trial outcomes reported at the 2017 SNO conference, in terms of immediate clinical applicability based on randomized prospective trial results, and the presentation was awarded the Adult Clinical Research Award for this year . CCNU (lomustine) has been in use for brain tumors since the 1970s and should be available at every brain tumor center. Since temozolomide was approved for newly diagnosed glioblastoma in 2005, it has been exceedingly rare for a phase 3 trial in the newly diagnosed GBM setting to achieve statistically significant prolongation of survival with a novel regimen. More work needs to be done to explain why progression-free survival benefit was more modest than overall survival benefit in this trial. The authors do not feel the overall survival outcomes can be explained by differences in salvage therapies post-progression. Pseudoprogression may be more widespread in MGMT methylated glioblastoma with the combination chemotherapy than it is with temozolomide-based therapy alone, which could potentially help explain a narrowing of the progression-free survival curves. A hypothesis for the improved survival results for the combination therapy is a possible synergistic interaction between TMZ and CCNU, whereby cells escaping sensitivity to TMZ through mismatch repair defects are thereby rendered more sensitive to the CCNU treatment (Stritzelberger et al. 2017, Reference 1). Now that the results of CeTeG have been reported to the international neuro-oncology community, the mechanisms behind the improved outcomes with the combination chemotherapy will likely become the subject of more intense investigation. It will be interesting to see how quickly these phase 3 trial results will be translated into changes in clinical practice for MGMT-methylated glioblastoma and also MGMT-methylated lower grade gliomas.
Reference 1: Acquired temozolomide resistance in human glioblastoma cell line U251 is caused by mismatch repair deficiency and can be overcome by lomustine. Strizelberger et al. 2017. Pubmed link
AG-120, a first-in-class mutant IDH1 inhibitor in patients with recurrent or progressive IDH1 mutant glioma: updated results from the phase 1 non-enhancing glioma population [abstract ACTR-46]
Ingo Mellinghoff of Memorial Sloan Kettering Cancer Center presented results of a phase 1 trial of AG-120 (ivosidenib), a mutant IDH1 inhibitor, for IDH1-mutant cancers. The presentation specifically focused on a subset of patients in the phase 1 trial: those with non-enhancing (no contrast enhancement on MRI images) IDH1-mutant gliomas. This analysis included 11 patients in the dose escalation phase, and an additional 24 patients from the dose expansion phase, a total of 35 patients. The primary study objective was to evaluate safety and tolerability of AG-120, and determine the maximum-tolerated dose and/or the recommended phase 2 dose.
28 out of 35 patients (80%) of patients in this analysis were treated with 500 mg of AG-120 daily. The majority (24/35, or 69%) of patients in this non-enhancing glioma cohort were WHO grade 2 gliomas. An additional 23% were WHO grade 3. Only one grade IV glioma (3%) was included.
Most patients in this analysis had been previously treated with either radiation (57%) or chemotherapy (69%) and the median number of prior systemic therapies was 2.
AG-120 was well tolerated, and the maximum tolerated dose was not reached. The majority of adverse events were low grade, and only 20% of patients experienced a grade 3 or higher adverse event.
Pharmacodynamic analysis of tumor tissue in two patients revealed that AG-120 treatment strongly suppressed 2-hydroxyglutarate levels in the tumors. 2-hydroxyglutarate is the oncometabolite produced by the mutant IDH1 enzyme.
By far the most common response to AG-120 treatment in this cohort was stable disease, which was achieved in 83% of patients. Only two patients (5.7%) achieved a minor response, including one grade 2 and one grade 3 glioma. Only four out of the 35 patients (11%) had progressive disease with neither stabilization or response.
More important is the duration of stable disease without progression. Median duration of AG-120 treatment for all 35 patients is 16 months. For the grade 2 gliomas, representing nearly 70% of the population of this study, median progression-free survival has not yet been reached, and looks to be at least 19 months at the time of the analysis (data cutoff May 12 2017).
When volumetric growth rates pre- and post AG-120 treatment were calculated by imaging studies for the 24 patients in the dose expansion group, the mean percentage change in tumor volume per six months was found to be 24% prior to treatment, and 11% after AG-120 treatment. In the 1p/19q intact (that is, the astrocytoma) subgroup of 15 patients, before and after AG-120 growth rates per six months were 38% and 14%. This confirms that the primary effect of AG-120 in this group of non-enhancing (mostly) lower grade gliomas is to significantly slow tumor progression, which was deemed to be a disease stabilization in the majority of cases.
A different drug by Agios Pharmaceuticals, called AG-881, is a dual inhibitor of mutant IDH1 and IDH2, is more brain penetrant than AG-120, and is also being studied in clinical trials for IDH mutant gliomas.
Compliance and treatment duration predict survival in a phase 3 EF-14 trial of Tumor Treating Fields with temozolomide in patients with newly diagnosed glioblastoma
This presentation was given by Zvi Ram of Tel Aviv Medical Center. The positive results of the randomized phase 3 EF-14 trial, leading to the FDA approval of Tumor Treating Fields (Optune) for newly diagnosed glioblastoma have been published, and updates have been presented in press releases by Novocure.
The Optune device “has an internal log file which allows the calculation of subject compliance with treatment”. Compliance is simply the average percentage of time each patient is wearing the Optune device per day. Patients wearing Optune are instructed to aim for at least 75% compliance (an average of 18 hours per day). This presentation by Zvi Ram provided important new data on the importance of compliance to survival outcomes.
According to the published EF-14 trial results (Stupp et al. 2015), “TTFields could be continued until the second radiological progression, or clinical deterioration, for a maximum of 24 months”, though some patient continued wearing the device for longer than 2 years. In the Optune arm of the trial, patients received treatment with Optune for a median of 8.2 months, with 13% still wearing the device at 2 years, 3% at 3 years, and only 1% at 4 years.
In a multivariate analysis which included other important prognostic variables such as extent of resection, MGMT status, age, and KPS, compliance (at least 75% versus less than 75%) was an independent, statistically significant predictor of survival.
In the Optune arm of the trial, median compliance was in the 70-80% range, with 209 of 450 (46%) patients achieving at least 80% compliance and 43 of 450 (9.6%) patients achieving at least 90% compliance. Baseline characteristics of each compliance subgroup (0-30%, >30-50%, >50-60%, >60-70%, >70-80%, >80-90%, and >90-100%) were shown in the slides and were “well balanced across the different compliance groups”.
Median progression-free survival (PFS) and overall survival (OS) for each compliance group were also shown. The lowest compliance subgroups (less than 50% compliance) did not have significantly improved PFS or OS compared to TMZ alone, and a threshold compliance of at least 50% (wearing Optune for at least 12 hours a day) was required to show an extension of PFS and OS compared to TMZ alone.
The presentation then zoomed in on the 43 patients who achieved over 90% compliance. In this subgroup, median PFS from randomization was 8.2 months, compared to 6.7 months in the entire Optune arm, and 4 months for the TMZ-only arm. Median survival from randomization in the >90% compliance group was 24.9 months, compared to 20.9 months in the entire Optune arm, and 16 months for the TMZ-only arm. Survival at 1, 2, 3, 4, and 5 years from randomization was compared between the >90% compliance group and the TMZ-only group. 29.3% of the >90% compliance group survived to three years, compared to 16.3% in the TMZ-only group. This 29.3% survival was maintained through to five years for the >90% compliance group, while survival in the TMZ-only group declined to 4.5% at the five year mark.
Using previously published data, it is possible to make further calculations. 5-year survival in the entire Optune arm of 466 patients was reported to be 13%, or approximately 61 patients. 5-year survival in the >90% compliance subgroup (n=43 patients) was 29.3%, or approximately 13 patients. According to these calculations, the >90% compliance group made up less than 10% of the entire Optune arm, but contributed over 20% of the 5-year survivors.
Comparing baseline characteristics in the 90% compliance group versus the TMZ-only group, we find a slightly younger median age in the 90% compliance group (52 versus 57), same median KPS (90), a slightly higher percentage of gross total resections (63% versus 54% in the TMZ only arm), and a very similar percentage of MGMT-methylated patients (40.5% versus 41.6%).
It is interesting to note that median KPS (Karnofsky performance status) in all of the compliance subgroups was 90, with the exception of the lowest compliance subgroup (0-30% compliance) which had a median KPS of 80. This makes it unlikely that the group achieving the highest level of compliance with Optune did so because they were less incapacitated by their disease than the groups with lower compliance: at least half the patients in all the compliance subgroups (save the lowest one) had KPS performance scores of 90-100 and were able to carry on normal daily activities with only minor symptoms of disease (as per the definition of Karnofsky performance status).
A randomized trial prospectively assigning patients to different levels of Optune compliance is both unlikely and impractical, even ethically questionable. As we now know, 33% of patients in the Optune arm of EF-14 did not manage to achieve over 70% compliance, despite the recommendation for at least 75% compliance. In the absence of such a trial, which will likely never happen, we are left with post-hoc analysis such as the present one, showing better PFS and survival outcomes with higher levels of Optune compliance, as well as compliance on Optune being independently associated with better survival in multivariate analysis adjusted for other known prognostic variables such as MGMT status, age, and extent of resection. On the basis of these results, it may be reasonable to suggest that the target for compliance on Optune should no longer be “75%”, but “as high as possible”, although lower levels of compliance (over 50%) may still provide therapeutic benefits compared to not wearing the device at all.
Durable responses observed in IDH1 wildtype and mutant recurrent high grade glioma (rHGG) with Toca 511 & Toca FC treatment [abstract ATIM-02]
Timothy Cloughesy of UCLA presented updated results for the phase 1 trial of tumor resection followed by injection of Toca 511 into the tumor resection cavity. This trial (NCT01470794) treated 56 patients with recurrent glioblastoma or anaplastic astrocytoma (WHO grades 3 and 4).
Pooling results across all three phase 1 dose escalation trials (n=127 patients), Toca 511 was found to be well tolerated, with only 25% of patients experiencing grade 1 or 2 treatment-related adverse events (including fatigue in 11%), and 7% of patients experiencing a grade 3 or higher treatment-related adverse event (including one case of headache, one case of fatigue, and two cases of vasogenic cerebral edema). Administration of the fluorouracil prodrug, Toca FC, was also associated with a low incidence (3.3%) of grade 3 or higher treatment-related adverse events. Only three out of 122 patients discontinued therapy due to adverse events associated with Toca FC.
In the NCT01470794 trial of resection followed by injection of Toca 511 into the resection cavity, 46/56 (82%) of patients had recurrent glioblastoma, and the remainder had anaplastic (grade 3) or other gliomas. 50% were at first recurrence, 23% were at second recurrence, and 27% were at third or more recurrence.
This trial included the endpoint of durable response rate, defined as a response lasting at least 24 weeks (5.5 months). Remarkably, all responders (6 out of 53 patients evaluable for efficacy, 11.3%) are in complete response and still alive, with a median duration of response of at least 35.1 months (nearly 3 years). An additional 18.9% had stable disease as best response, for a clinical benefit rate of 30.2% (response plus stable disease).
In the subset of patients (n=23) treated with higher doses of Toca 511 and meeting the eligibility criteria for the ongoing phase 3 Toca 5 trial (at first or second recurrence, no prior Avastin, and tumor no larger than 5 cm), there were 5 out of 23 patients (21.7%) with durable complete response lasting for a median of 35.7+ months. An additional 21.7% had stable disease as best response, for a total clinical benefit rate (response + stable disease) of 43.5%. In this subgroup of 23 patients, three of the complete responders had wild-type IDH1, and two had anaplastic astrocytoma with mutant IDH1. These responses occurred gradually over 6-19 months, consistent with an immunologic mechanism of action. Median survival from Toca 511 treatment in this subgroup is 14.4 months, which compares favorably with historical benchmarks of 9-10 months for recurrent GBM with standard treatments.
The safety profile, increased median survival relative to historical benchmarks, and relatively high rate of complete durable responses lasting for a median of at least three years is encouraging. The randomized phase 2/3 Toca 5 trial for recurrent glioblastoma or anaplastic astrocytoma (NCT02414165) recently re-opened in November 2017 to new recruitment and the phase 1 Toca 7 trial for newly diagnosed high grade glioma (NCT02598011) is scheduled to open sometime in 2018.
Presentation may be downloaded from the Tocagen website
Intravenous delivery of Toca 511 in patients with high grade glioma results in quantifiable expression of cytosine deaminase in tumor tissue [abstract ATIM-21]
Tobias Walbert of Henry Ford Hospital presented results of this phase 1 trial, which tested administration of the Toca 511 virus by intravenous injection. Toca 511 is a retroviral replicating vector that transmits a gene to infected tumor cells. This gene, cytosine deaminase (CD), causes the cell to convert orally administered Toca FC (an extended release version of 5-fluorocytosine) into the active chemotherapy agent 5-fluorouracil. Following intravenous injection for 1, 3, or 5 days, tumors were resected and additional Toca 511 was injected into the tumor cavity walls. This pre-resection intravenous administration allowed investigators to analyze tumor tissue for detection of cytosine deaminase (CD) within tumors.
17 patients were included in this trial. 14 of these patients (82%) had diagnoses of GBM and the remaining three had grade 3 gliomas. 47% were at first recurrence while 53% were at second or more recurrence.
11 of 17 (65%) resected tumors were positive for cytosine deaminase, showing that intravenously administered Toca 511 successfully entered into tumors in the majority of cases. Analysis of resected tumor tissue following intravenous Toca 511 injection found that tumors with high T-cell infiltrate did not limit the ability of Toca 511 to enter tumor cells. Conversely, the presence of immunosuppressive regulatory T-cells (Tregs) were also not required to allow entry of Toca 511 into tumor cells.
The maximum tolerated dose of Toca 511 was not defined and grade 3 or higher adverse events were rare, occurring in only 3 patients (17.6%).
Stable disease was achieved in 3 of 17 patients (17.6%) and late onset (>12 months after treatment) radiologic responses were seen in two of 17 patients (11.8%), consistent with an immunologic mechanism of action. The responding patients were an IDH wild-type anaplastic astrocytoma patient at 3rd recurrence, whose response was noted on MRI 9 months after discontinuing Toca FC and was on no other anti-cancer therapy; and an IDH1-mutant GBM patient at first recurrence who had onset of response 13 months after initiating Toca FC. Complete response was eventually achieved in this patient, and response has lasted for 16 months and counting. This patient remains on Toca FC and no other anti-cancer therapy. Median survival from trial entry for this group of 17 patients, 53% of whom were at second or more recurrence, is 13.6 months.
Currently recruiting patients is the Toca 6 trial for recurrent solid tumors, and the randomized phase 2/3 Toca 5 trial for recurrent malignant glioma, which has just reopened as of November 2017. A phase 1 trial of Toca 511/ Toca FC for newly diagnosed malignant glioma (Toca 7) is scheduled to open in 2018.
Allogeneic tumor lysate / autologous dendritic cell vaccines in newly diagnosed glioblastoma: Results of clinical trial MC1272 [abstract ATIM-13]
Ian Parney of Mayo clinic presented the results of this phase 1 trial (NCT01957956). The vaccine in this study was composed of autologous dendritic cells matured from each patient’s CD14+ monocytes, pulsed with an allogeneic tumor lysate derived from two GBM cell lines with defined tumor antigen expression. This vaccine differs from autologous vaccines such as DCVax in that those vaccines use both autologous dendritic cells as well as autologous tumor lysate created from each patient’s tumor, while this vaccine uses an allogeneic tumor lysate made from GBM cells that are not from the patient’s own tumor. An optimized technique has been developed by this group for generating mature CD83+ dendritic cells from each patient’s monocytes.
20 newly diagnosed GBM patients were included in this study, and this group was enriched in patients with poor prognostic markers. 25% (n=5) patients had multifocal tumors. Only 1 patient (5%) tested positive for IDH1 mutation. 30% (n=6) were positive for MGMT methylation, and 55% (n=11) underwent gross total resection.
Median progression-free and overall survival were 9.7 and 18.3 months. More impressive than these median values, is the long tail of patients still progression-free past two years. According to the Kaplan-Meier chart shown in this presentation, there were still five patients without progression at the time of data cutoff. For these five patients, median follow-up was over two years. Three patients made it to two years without progression, and two patients had made it to three or nearly three years without progression at the time of last follow-up. At least three patients out of 20 (15%) still without progression at two years is suggestive of activity of the vaccine in this subpopulation. The prolonged PFS in these patients “corresponds to tumor antigen-specific responses vs. gp100”.
Full enrollment results from the phase 1/2, multicenter, open-label study of marizomib (MRZ) +/- bevacizumab (BEV) in recurrent WHO grade IV malignant glioma [abstract ACTR-71]
Daniela Bota of UC Irvine presented these results. The trial was divided into three parts: Part 1 was a dose escalation and dose expansion trial of the brain penetrant proteasome inhibitor marizomib added to bevacizumab and consisted of 36 patients. Part 2 was a trial of single-agent marizomib consisting of 30 patients. Part 3 was an intra-patient dose escalation of marizomib combined with bevacizumab consisting of 35 patients.
In Part 1, 27 out of 36 patients (75%) were treated with 0.8 mg/m2, while 9 patients (25%) were treated at lower doses in the dose escalation phase. Patients in Part 1 also received standard dose bevacizumab. The overall response rate for the 36 patients receiving marizomib and bevacizumab was 16/36 or 44.4%. 11/36 or 31% had stable disease. Response plus stable disease rate is 27/36 or 75%. Median progression-free survival was 3.9 months and median overall survival was 9.4 months.
In Part 2, 30 patients were treated with marizomib alone. In this cohort there was only 1 partial response (1/30 or 3%) and 6 with stable disease (6/30 or 20%), for a response plus stable disease rate of 23%.
Patients in Part 1 experiencing central nervous system related adverse events (including ataxia, balance disorder, dizziness, fall, gait disturbance, hallucination) had increased PFS and OS compared to those who did not suffer from these side-effects, and this observation provided a justification for Part 3 of the study, an intra-patient dose escalation of marizomib combined with bevacizumab. In Part 3, 10 out of 35 patients were escalated to 1 mg/m2 of marizomib, but only 1 patient was able to tolerate this dose. No patient reached 1.2 mg/m2. The intra-patient dose escalation can therefore be said to have not succeeded, as the next higher dose of 1 mg/m2 was largely not tolerated.
It appears that single agent marizomib at 0.8 mg/m2 has some activity, but only for a minority of patients, while adding bevacizumab to marizomib leads to much higher response rates and increased survival at 12 months, but still leads to an average median overall survival statistic for recurrent glioblastoma.
This page is a team effort by the volunteers of the Musella Foundation, and is edited by Stephen Western
If you would like to suggest any changes to improve this page, use our contact form
This page was created on 12/20/2017 and last updated on 12/20/2017