MimiVax and Roswell Park Comprehensive Cancer Center reported topline Phase 2b data on 7 October 2026. The SURVIVE trial tested the survivin peptide vaccine SurVaxM in newly diagnosed glioblastoma. The primary endpoint was overall survival. The full study population did not reach statistical significance. The sponsors report a statistically significant result in a subpopulation aged 65 and under. The release describes that subpopulation as prespecified.
MimiVax now intends to ask the FDA for a meeting. It wants to discuss Breakthrough Therapy Designation and other expedited routes. Accelerated approval is among them. That request sits on a trial whose primary analysis fell short. Newly diagnosed glioblastoma has had no new first-line systemic agent in two decades. That makes the regulatory question matter more than the headline.
What the SURVIVE trial reported
SURVIVE was a randomised, double-blind, placebo-controlled, multicentre Phase 2b study. It evaluated SurVaxM added to standard of care in newly diagnosed glioblastoma. The release reports 233 evaluable patients at 11 US cancer centres. Median overall survival reached 22.8 months on SurVaxM. The placebo figure was 20.3 months. The hazard ratio was 0.75, with a 95 per cent confidence interval of 0.54 to 1.03.
The log-rank p value was 0.076, adjusted for MGMT and KPS stratification. That result does not clear conventional significance. The release says it “did not quite reach the traditional threshold for statistical significance”. The upper confidence bound sits at 1.03, above 1.0. The design target was further away. The trial protocol posted on ClinicalTrials.gov is version 3.2, dated 10 November 2021. It powered the study to detect a hazard ratio of 0.56. The observed figure was 0.75. The significance bar may also have been tighter than 0.05. The protocol planned one interim analysis and set a projected final efficacy threshold of a one-sided p value of 0.01387.
One design detail deserves attention. That protocol planned 265 patients, randomised 3:2 to SurVaxM or placebo. Later amendments may have changed any of these plans. MimiVax said on 15 July 2026 that 249 patients had enrolled. The release reports 233 as evaluable. The protocol counts every eligible randomised patient as evaluable. The release does not account for the other 16. It names no randomised total and no randomisation ratio.
What the subgroup showed in newly diagnosed glioblastoma
The release describes patients aged 65 and under as a prespecified key subpopulation for the primary endpoint. In that group, median overall survival reached 24.2 months. The placebo figure was 20.2 months. The hazard ratio was 0.64, with a confidence interval of 0.44 to 0.93. The p value was 0.019. Progression-free survival moved in the same direction. Its hazard ratio was also 0.64, at a p value of 0.018.
How much weight that p value carries depends on the statistical plan. The release does not say how the trial’s type I error was shared between the two analyses. The posted protocol sets an overall two-sided type I error of 5 per cent. It provides for secondary subgroup analyses of both survival endpoints by demographic and baseline characteristics. It names performance status, MGMT status and IDH-1 status. It names no age cut at 65. It also refers to a formal statistical analysis plan, which LSDN has not seen. A later amendment or that plan may have fixed the age cut.
In the same subgroup, progression-free survival at 12 months reached 44 per cent on SurVaxM. On placebo it reached 24 per cent. The sponsors also report longer-term figures. Roughly 42 per cent of treated patients aged 65 and under were alive at 30 months or beyond. The placebo figure was 18 per cent. The release states that no further deaths occurred between 30 and 54 months.
Two of the three people quoted in the release are trial investigators. Only one tied his assessment to the subgroup. Robert Fenstermaker is co-principal investigator and Chair Emeritus of Neurosurgery at Roswell Park. He is also Chief Medical Officer and Co-founder of MimiVax. He called the result “one of the strongest signals we have seen in newly diagnosed glioblastoma in a long time”. He limited that assessment to patients aged 65 and under.
Ajay Abad is the study principal investigator. He is Assistant Professor of Oncology and Neurology in the Department of Neuro-Oncology at Roswell Park. He said the field had seen “limited progress over the last 20 years”. Against that background, he called an improvement of this magnitude “noteworthy”. His quote does not specify which analysis he meant.
Safety, and what the release does not publish
Tolerability looks unremarkable. Injection-site reactions dominated, almost all Grade 1 or 2. The release gives injection-site reaction at 48 per cent and pain at 41 per cent. Erythema ran at 39 per cent, pruritus at 38 per cent and swelling at 29 per cent. Influenza-like illness ran at 13 per cent.
The less favourable figures sit alongside them. Grade 3 or higher adverse events reached 48 per cent on SurVaxM against 46 per cent on placebo. Serious adverse events ran at 25 per cent against 21 per cent. Adverse events led to discontinuation in 4 per cent of SurVaxM patients. One of those discontinuations was attributed to study treatment. The sponsors explicitly state that no treatment-related deaths occurred in either arm.
Several items are absent from the 7 October release. It publishes no outcome breakdown by MGMT methylation status. MGMT was nonetheless used as a stratification factor. It reports no results for the subpopulation aged over 65. It gives no survivin-specific immune response data, which the release lists among the secondary endpoints. These are LSDN readings of the release as issued, not negative findings.
Why the age cut does not map onto NICE guidance
NICE first-line treatment for newly diagnosed glioblastoma turns on age, but not at 65. NICE guideline NG99, which formally applies in England, sets the thresholds. It was last updated on 29 January 2021. It advises offering radiotherapy at 60 Gy in 30 fractions with concomitant temozolomide. Up to six adjuvant cycles follow. That applies to patients aged around 70 or younger with a Karnofsky performance status of 70 or above.
A shorter schedule applies above that age. For patients aged around 70 and over, NICE recommends 40 Gy in 15 fractions. Concomitant and up to 12 adjuvant temozolomide cycles accompany it. That recommendation covers tumours that are MGMT methylation positive. Best supportive care alone is an option above about 70 with performance status below 70.
So SURVIVE stratified on MGMT status and performance status. Those are two of the three factors NICE uses to choose a regimen. Age is the third, and there the two diverge. A result in patients aged 65 and under does not describe the NICE first-line population. That population extends to roughly 70. Any reader treating the subgroup as the licensable group reads across a five-year gap.
The underlying burden is substantial. Cancer Research UK records 12,533 new brain, other central nervous system and intracranial tumour cases in the UK each year. Those figures cover 2019 and 2021 to 2022. Ten-year survival for brain cancer stands at 19.2 per cent, for 2018. The age gradient is stark. Among those aged 15 to 44, 60.7 per cent of females survive ten years or more. For males the figure is 53.4 per cent. At ages 75 to 99 the figures are 2.9 per cent and 2.6 per cent. All of these cover brain and CNS tumours collectively, not glioblastoma alone.
A control arm above the trial’s own assumption
The placebo arm is the quietly interesting number. Its median overall survival was 20.3 months. The posted protocol had assumed 16 months for the control group. So the comparator outperformed the design assumption by 4.3 months.
The protocol offers part of the explanation. Patients entered only after surgery that left 1 cm³ or less of residual enhancing tumour. They had completed chemoradiation and shown no progression. Survival was then measured from randomisation, not from diagnosis. Each of those features tends to select for longer survival. Fenstermaker said in the release that the placebo arm was added “to account for improvements in standard of care”.
The historical benchmark is not like for like. The trial that set the current regimen for newly diagnosed glioblastoma reported 14.6 months. Roger Stupp and colleagues published that result in the New England Journal of Medicine in March 2005. It randomised 573 patients at 85 centres, before radiotherapy began. Cross-trial comparison across 21 years is not a valid efficacy argument. Patient selection, imaging standards, surgical practice and supportive care have all moved.
The regulatory bar MimiVax has to clear
The FDA publishes its Breakthrough Therapy standard plainly. The designation covers drugs intended to treat a serious condition. It also requires that “preliminary clinical evidence indicates that the drug may demonstrate substantial improvement over available therapy”. That improvement must fall on a clinically significant endpoint. The agency adds that the evidence “should show a clear advantage over available therapy”.
Michael Ciesielski is Chief Executive Officer and Co-Founder of MimiVax. He set out the ask directly. The company wants “to discuss Breakthrough Therapy Designation with the FDA”. It also named “other expedited pathways, including accelerated approval”. Speaking to Fierce Biotech the same day, he said even very small survival increases are very significant to patients.
Accelerated approval is an unusual fit. That pathway ordinarily rests on a surrogate or intermediate endpoint. SURVIVE measured overall survival, the clinical outcome itself.
SurVaxM already carries two designations. The history runs further back than this readout. Roswell Park records an original orphan drug designation in August 2017 for adult glioblastoma. Fast Track designation for newly diagnosed glioblastoma came next, in autumn 2023. It was announced on 12 October that year. A supplemental orphan designation in 2024 broadened coverage to malignant glioma in children and adults.
None of that settles the question. A prespecified subgroup carries more weight than a post-hoc one. It still sits inside a trial whose primary analysis returned a p value of 0.076. The FDA has not commented. No meeting date is public.
The precedent in brain cancer immunotherapy
This shape is familiar in newly diagnosed glioblastoma. Imvax missed its Phase 2b primary endpoint of progression-free survival with IGV-001 in December 2025. It took its overall survival data to the FDA. On 15 September 2026 it said it had met the agency twice this year. The second meeting, in July, was held to align on key elements of a proposed Phase 3 trial. Imvax says it has built the agency’s feedback into the protocol. That study would enrol about 375 patients with unmethylated MGMT tumours. Overall survival would be the primary endpoint. Imvax plans to start in 2027 and is seeking financing.
Celldex Therapeutics offers the older case. Rindopepimut missed overall survival in the Phase 3 ACT IV trial in 2016. It was also tested in newly diagnosed glioblastoma and never reached the market. In the Imvax case, the route forward ran through another randomised trial. The wider field has already been reshaped once this year, after a survival imbalance redrew cancer vaccine trial design.
MimiVax spun out of Roswell Park in Buffalo in 2012, according to Fierce Biotech. It holds an exclusive global licence to commercialise SurVaxM. The release confirms that SurVaxM remains investigational. No regulator has approved it for any indication. The company also runs mid-stage studies in metastatic neuroendocrine tumours and multiple myeloma.
LSDN found no peer-reviewed publication or congress presentation of the SURVIVE results as at 7 October 2026. That is a dated search finding, not a statement about the data. It rests on web search and the sponsors’ own release. The Phase 2a study was published. Ahluwalia and colleagues reported it in the Journal of Clinical Oncology, online on 15 December 2022.
What to watch next
Three things will decide this. The first is whether the FDA grants a meeting. What it then says about a subgroup-led filing matters most. The second is the full dataset, including MGMT breakdowns and the over-65 results. A congress presentation or publication would expose both. The third is whether SurVaxM is ever filed outside the United States. No regulatory step there has been announced.
For British clinicians treating newly diagnosed glioblastoma, the practical question is narrower. A reported benefit at 65 and under sits awkwardly against a guideline drawn at 70. No result above 65 has been published. Our explainer on the accelerated approval pathway sets out how surrogate endpoints and confirmatory trials work at the agency.
