Society for Neuro Oncology annual meeting, Scottsdale Arizona, November 17-20 2016
TOP SCORING ABSTRACTS (Friday morning)
Prospective, multi-center phase III trial of tumor treating fields (Optune) together with temozolomide compared to temozolomide alone in patients with newly diagnosed glioblastoma [abstract LTBK-01]
Early Friday morning, in a late breaking abstract, Roger Stupp presented the long-term analysis of the full dataset of 695 patients enrolled in the EF-14 trial testing Optune in addition to standard therapies for newly diagnosed glioblastoma. These results confirm the findings of the initial report presented at the SNO meeting in 2014. All data was presented for the “intention-to-treat” population.
From randomization (after radiochemotherapy), the addition of Optune to standard treatments improved median progression-free survival by 2.7 months (from 4 to 6.7 months) and median survival was improved by 4.8 months (from 16 to 20.8 months from randomization, or from 19.8 to 24.5 months from diagnosis). The percentage of two-year survivors increased from 30% in the control group to 42.5% in the Optune group, and four-year survivors increased from 10% to 17%. The presentation of this full dataset removes one reason for hesitation regarding Optune shown by some clinicians, as the data can no longer be considered as “only preliminary”.
ACT IV: An international, double-blind, phase 3 trial of rindopepimut in newly diagnosed, EGFRvIII-expressing glioblastoma [abstract ATIM-03]
Following on the heels of Stupp’s presentation, Michael Weller presented the disappointing results of the phase 3 ACT IV trial of rindopepimut for newly diagnosed glioblastoma. Confirming press releases dated March 2016, overall survival in this trial was virtually identical in the control arm and the rindopepimut arm. Median survival in both arms was 20 months from randomization and 23 months from diagnosis. If there is good news resulting from this trial, it’s that median survival for newly diagnosed GBM with standard treatments in phase 3 trial populations can now reach nearly 2 years from diagnosis, significantly longer than in trials conducted in previous decades. This trial recruited only patients positive for the EGFRvIII variant, traditionally thought to be a marker of poorer prognosis for GBM, though the unexpectedly long survival in the control arm of this trial may prompt a re-evaluation of that belief.
CLINICAL TRIALS SESSION 2A (Friday afternoon)
Results of the phase Ib KEYNOTE-028 multi-cohort trial of pembrolizumab monotherapy in patients with recurrent PD-L1-positive glioblastoma [abstract ATIM-35]
Friday afternoon, David Reardon presented results of the KEYNOTE-028, a phase 1 trial of pembrolizumab (Keytruda) for advanced solid tumors with PD-L1 expression, including 26 glioblastoma patients. Data cutoff for this analysis was February 17, 2016. For the glioblastoma patients, median time from diagnosis to enrollment was 20 months. Of 24 evaluable tumors, four achieved >30% decrease in tumor diameter, with another seven having shrinkage of less than 30%, for a total of 46% experiencing some decrease in tumor size. Median overall survival from trial entry was 14 months, the 12-month survival rate was 74%, and six month progression-free survival rate was 44%. Four patients with stable disease continued to receive pembrolizumab at 54-86 weeks (12.4 – 19.8 months). The drug was found to be reasonably well tolerated in this patient population, with the most common side-effects being fatigue and rash (each in 23% of patients), and grade 3-4 treatment-related adverse events occurring in four patients (15%), including one case of grade 3 arthritis, one case of grade 3 lymphopenia, one case of grade 3 syncope (defined as temporary loss of consciousness due to decrease in blood pressure), and one case of grade 4 type 2 diabetes. The survival and progression-free survival results are encouraging, especially the four patients (16%) who are still alive and continuing to receive pembrolizumab beyond the 12 month mark.
Phase 2 study to evaluate the clinical efficacy and safety of MEDI4736 (durvalumab) in patients with glioblastoma: results for cohort B (durvalumab monotherapy), bevacizumab-naive patients with recurrent GBM [abstract ATIM-04]
Reardon then went on to present results for cohort B of the phase 2 trial of MEDI4736 (durvalumab, an antibody against PD-L1) for bevacizumab-naive recurrent GBM. Cohort B consisted of 30 patients receiving durvalumab alone as monotherapy. Data cutoff for this presentation was June 30 2016. Six month progression-free survival, the primary endpoint of the trial, was 20%. Four of the 30 patients in this cohort were known to have IDH1-mutant tumors. Three of the six patients who reached the six month mark without progression had IDH1-mutant tumors. Of the six patients who reached the six months mark without progression, five of these were still without progression at 12 months (the sixth was also progression free at 11 months, the time of last follow-up for that patient). Median progression-free survival was 13.9 weeks (3.2 months), median survival from trial entry was 28.9 weeks (6.7 months), and 12 month survival rate was 44.4%. By modified RANO criteria, of the 30 patients there were 4 partial responses (13%), 14 disease stabilizations (47%) and 12 with progressive disease (40%), for an overall response rate of 13% and a disease control rate of 60%. Durvalumab was well-tolerated, with the most common non-neurological treatment-related side-effect being fatigue in 8 of 31 patients (26%) and decreased lymphocyte counts in 6 patients (19%). Given the observation that three of four IDH1-mutant cases were progression-free at six months, contributing half of the patients to the 20% PFS-6 figure, future investigations of MEDI4736 should examine patient outcomes by IDH status.
Efficacy of a novel antibody-drug conjugate, ABT-414, as monotherapy in EGFR amplified recurrent glioblastoma [abstract ACTR-07]
ABT-414 is an antibody drug conjugate which delivers the cytotoxic agent monomethyl auristatin F (MMAF) to EGFR amplified tumor cells. On Friday afternoon, Martin van den Bent presented results for the 60 patient expansion cohort C. Recurrent GBM patients in this cohort had a median of 2 prior therapies before joining the trial. Median age was 58 years and median KPS was 90.
Six month progression-free survival rate was 28.3% and median survival from trial start was 9 months. Response rate was 5%, and disease stabilization rate was 40%. 52% of patients had continued progressive disease. The PFS-6 of 28.3% was considered to be encouraging, as 56% of patients had already undergone 2-3 prior therapies. Side-effects were mainly ocular, with the most common adverse event being blurred vision, affecting 65% of patients.
AG120, a first-in-class mutant IDH1 inhibitor in patients with recurrent or progressive IDH1 mutant glioma: results from the phase 1 glioma expansion cohorts [abstract ACTR-46]
Ingo Mellinghoff of Memorial Sloan-Kettering Cancer Center presented data on the full cohort of glioma patients included in this trial, including 20 previously reported patients in the dose escalation cohort, as well as 24 patients in the non-enhancing glioma expansion cohort and 22 patients in the enhancing glioma expansion cohort. The entire group of 66 IDH1-mutant gliomas consisted of 14 grade II oligodendrogliomas, 8 grade III oligodendrogliomas, 12 grade II astrocytomas, 6 grade III astrocytomas, 8 grade II oligoastrocytomas, 4 grade III oligoastrocytomas, and 12 glioblastomas. Median age was 41 years. Patients had received a median of two prior therapies, including 74% with prior radiation and 71% with prior temozolomide.
In the non-enhancing expansion cohort, there were 23 patients evaluable for response. Of these, two (9%) had a minor response, 19 (83%) had stable disease, and only 2 (9%) had continuous progression. In this same cohort, volumetric analysis showed a decrease in tumor growth rate in 64% (n=14 of 22). 42% of patients remain on AG-120 as of August 1, 2016 and further follow-up is necessary before mature data can be reported.
No dose-limiting toxicities were observed and the maximum tolerated dose was not reached. There were no serious adverse events related to AG-120 treatment.
Clinical results of the EORTC randomized phase II TAVAREC trial on temozolomide with or without bevacizumab in 1st recurrence of grade II or III glioma without 1p/19q co-deletion [abstract ACTR-09]
Martin van den Bent gave the last presentation for the clinical trials 2A session Friday afternoon. The randomized phase II TAVAREC trial tested the 5/23 temozolomide schedule with or without the addition of Avastin for grade II and III gliomas without 1p/19q codeletion. This trial attempts to answer an important question for lower grade astrocytomas: whether or not to use Avastin at recurrence.
In newly diagnosed glioblastoma trials, the usual outcome of adding Avastin to standard treatments is a several month increase in median progression-free survival, without a corresponding improvement in median overall survival, likely due to patients in the control arm receiving Avastin at recurrence. In the TAVAREC trial, the most remarkable outcome was that Avastin improved neither overall survival nor progression-free survival. The progression-free survival curves for both arms (with and without Avastin) are virtually identical, with a hazard ratio of 0.99 and p value of 0.95, meaning Avastin had virtually no effect on progression-free survival in this patient cohort. The investigators therefore concluded that “results of the TAVAREC trial do not justify further investigation of BEV in this population”. However, an argument may still be made that IDH wild-type lower grade gliomas, with a molecular pathology closer to GBM, could possibly benefit from Avastin. Unfortunately IDH status of patients in this trial wasn’t presented.
CLINICAL TRIALS SESSION 3A (Friday afternoon)
An open label Phase 1b/2 study of orally administered pexidartinib in combination with radiation therapy and temozolomide in patients with newly diagnosed glioblastoma [abstract ACTR-20]
On Friday afternoon, Howard Colman of the University of Utah presented slides for a phase 1b/2 trial of pexidartinib in combination with radiation and temozolomide for newly diagnosed glioblastoma. Pexidartinib is an inhibitor of KIT and CSF1 receptor. The phase 2 dose was 400 mg twice daily. Efficacy results for this trial were compared to the phase 3 RTOG 0525 trial. Median progression-free survival of pexidartinib treated patients was improved compared to historical controls (10 months versus 7.5 months). “While data not fully mature, current estimates of median OS suggest modest benefit in PEX patients versus historical controls”.
Marizomib with bevacizumab in WHO grade IV malignant glioma: Full enrollment results from the phase 1, multicenter, open-label study [abstract ACTR-50]
Daniela Bota of UC Irvine Medical Center presented efficacy results for a phase 1 study of proteasome inhibitor marizomib and Avastin (bevacizumab) for recurrent GBM patients in first or second relapse. 33 patients were evaluable for response. 42% had tumor response, 39% had disease stabilization and 18% had progressive disease. Six month progression-free survival rate was 34% (34% for MGMT unmethylated and 29% for MGMT methylated). Compared to historical controls with Avastin alone, there was an “apparent improvement in PFS and OS in unmethylated MGMT patients”.
A prospective phase II study of everolimus for recurrent adult low grade gliomas [abstract ACTR-32]
Michael Wahl of UCSF presented the results for a trial of the mTOR inhibitor everolimus in recurrent low grade gliomas. 58 patients with an initial low grade diagnosis and histologically confirmed recurrence were treated with everolimus alone. Median age was 45, and 19% had progressed to higher grade gliomas (3 and 4) at the time of enrollment. 40% of patients were IDH-mutant and 1p/19q codeleted (molecular oligodendroglioma) and 40% were IDH-mutant, 1p/19q intact (molecular astrocytoma). Only 9% were IDH wild-type. 41% had undergone prior radiotherapy and 78% had undergone prior chemotherapy. On everolimus, 6 month progression-free survival rate was 84%, and 46% completed one year of treatment without progression. In the printed abstract, at a median follow-up of 3.5 years, median PFS was reported as 1.4 years, and median survival was not yet reached. Additionally, there was a 19% median decrease in capillary density after 6 months of treatment, indicative of anti-angiogenic activity.
Phase 1 trial of genetically modified hematopoietic progenitor cells facilitating bone marrow chemoprotection and enabling TMZ/O6-benzylguanine dose escalation [abstract ACTR-54]
These results were presented by Andrew Sloan, of Case Comprehensive Cancer Center (Cleveland Ohio). The rationale for this trial was to provide chemoprotection for patients’ blood cells by transplanting patients with hematopoietic progenitor cells engineered to express a mutant MGMT gene, leading to normal MGMT activity but low affinity for O6-benzylguanine. O6-benzylguanine, an MGMT inhibitor, was then used to selectively deplete MGMT in the tumor cells. With this improved protection for blood cells, dose escalation of temozolomide and O6-benzylguanine was facilitated. Seven patients with newly diagnosed glioblastoma were treated on this protocol. Survival ranged from 20-36 months, exceeding predicted survival according to recursive partitioning by 1.9 to 3.2-fold. A phase 2 study is being planned.
Phase I study of chimeric antigen receptor-engineered T cells targeting IL13R-alpha-2 for the treatment of glioblastoma [abstract ATIM-13]
Christine Brown, of City of Hope Cancer Center (Duarte, California) presented results for their phase 1 CAR-T cell trial. Patients had recurrent high-grade gliomas, positive for interleukin 13 receptor alpha 2 (IL13R-alpha-2). The therapy was well-tolerated, with no dose-limiting toxicites or severe adverse events. The details of one particular case were presented in detail: a 50 year old male with recurrent multifocal GBM, including metastatic sites in the spine, was treated with 6 weekly infusions of the CAR-T cells into the resection cavity, followed by 10 weekly intraventricular infusions. By day 300, all intracranial and spinal tumors had regressed by 79-97% by maximum area. The man “returned to normal work and life activities” and “regression sustained for 7.5 months”. In addition to this complete response, 5 of 7 patients in the resection arm of the trial had disease stabilization for at least 8 weeks.
A phase 1a/1b, multi-center, open-label dose finding study to assess the safety, tolerability and efficacy of CC-122 administered orally to patients with glioblastoma and other brain tumors [abstract ATIM-28]
Matteo Simonelli, of the Istituto Clinico Humanitas in Milan Italy presented clinical data on CC-122, a thalidomide analog with immune modulation and anti-angiogenic activity. As of the data cutoff (September 29 2016) there were 35 relapsed brain tumor patients evaluable (most of the brain tumors were GBM). Median PFS was 57 days (1.9 months), and six-month progression-free survival rate was 17.4%. Three patients (9%) had durable responses, and 37% had disease stabilization. Two patients (4.3%) were still on treatment at the time of last analysis. CC-122 led to a median 4.57-fold increase in activated CD8+ T-cells for the 3 mg dose group, but only a 1.55-fold median increase for the 5 mg group. Therefore “lower doses of CC-122 may be more optimal for immune modulation”.
Allogeneic tumor lysate / autologous dendritic cell vaccines in newly diagnosed glioblastoma [abstract ATIM-31]
Ian Parney of Mayo Clinic gave this presentation. In this clinical trial, 20 newly diagnosed glioblastoma patients post-resection/radiation/concurrent TMZ were treated with an allogeneic tumor lysate vaccine. The vaccine was prepared by pulsing dendritic cells with allogeneic tumor lysate from two “human GBM cell cultures with defined tumor antigen expression”. Median age was 60.8 years, and five patients (25%) had multifocal tumors (four of these were bilateral). Only one patient (5%) had an IDH mutation, and 30% had methylated MGMT status. The cohort was thus enriched for poor prognostic markers. The vaccine was found to be safe, and led to frequent pseudoprogression. Median PFS was 9.9 months and median survival was 20.5 months, although this could change with further patient follow-up. These results are considered promising in light of the poor prognostic factors of this cohort at the baseline evaluation.
Phase 2 trial of SL-701, a novel immunotherapy comprised of synthetic short peptides against GBM targets in adults with second-line recurrent GBM: Interim results [abstract ATIM-11]
David Reardon presented yet another immunotherapy trial on Friday, this time the preliminary results of the phase 2 trial of the SL-701 peptide vaccine. The trial was divided into 2 stages. Patients in stage 1 were given SL-701 in combination with adjuvants GM-CSF and imiquimod, while patients in stage 2 were given SL-701 with Poly-ICLC and bevacizumab (Avastin). The stage 1 cohort consisted of 46 patients and the stage 2 cohort enrolled 28 patients at the time of analysis.
In the stage 1 trial, there was one partial response of 13 months duration, and response was still ongoing. There were two disease stabilizations of 18 and 20 months, both ongoing.
In the stage 2 trial (SL-701 + Poly-ICLC + Avastin), there were two complete responses and five partial responses. Some of these responses were pending 2nd response assessment. For the 28 patients in stage 2, median PFS is at 5.6 months and median survival was not yet reached. This data is to be considered very preliminary as median follow-up time in this cohort was only 4.1 months at the time of analysis.
Phase IIa clinical trial evaluating dendritic cell vaccine for the treatment of low-grade gliomas [abstract ATIM-32]
Diana Moughon of UCLA presented the results for this exploratory phase 2a trial of autologous tumor lysate-pulsed dendritic cell vaccines for low grade glioma. Only five patients were included in this trial, three with newly diagnosed and two with recurrent low grade gliomas. Four patients had IDH1-mutant tumors, and one 85-year old patient had an IDH wild-type low grade glioma. Surprisingly, the longest time to progression of the five patients was 42.8 months for this 85-year old IDH wild-type patient, who had the worst prognosis of the five. Median time to progression in this trial is 19.8 months, with only one IDH-mutant grade 2 glioma patient still being followed up for time to progression at 33.1 months. Compared to matched controls, there was no improvement in time to progression for patients in this trial, although as acknowledged in the presentation “pseudoprogression may confound imaging studies”, bringing down the median time to progression duration. For two patients, pre- and post-vaccine specimens showed an increase in CD8+, PD-1+ tumor-infiltrating lymphocytes, which could mean that the addition of an anti-PD-1 agent could improve the immune response to the vaccine.
Ten-year follow-up with long term remission in patients with newly diagnosed GBM treated with ICT-107 vaccine (phase I) [abstract ATIM-25]
In the second-to-last presentation of the clinical trials sessions Friday afternoon, Surasak Phuphanich of Cedars-Sinai delivered the 10-year follow up for the initial phase 1 trial of ICT-107 multi-peptide vaccine for newly diagnosed glioblastoma. Data cutoff for this analysis is November 11, 2016. Of the 16 newly diagnosed patients in this trial, 6 (37.5%) are still alive at 100 – 123 months (8.3 – 10.25 years). 3 patients (19%) are without progression at 100 – 118 months (8.3 – 9.8 years). Median progression-free survival is 16.9 months and median overall survival is 38.4 months. Five-year overall survival rate is 50%. Of the six long-term survivors in this trial, five of them (83%) underwent complete resection at the time of diagnosis. All of these patients’ tumors expressed 4 of the 6 antigens targeted by the vaccine (AIM2, TRP-2, HER2, IL13Ra2). The long-term survivors also had elevated interferon-gamma responses relative to short survivors. The abstract for this study calls the 19% long-term (>8 years) remission rate “unprecedented”.
Additional presentations I was unable to attend included: