Glioblastoma is the hardest common cancer to treat. Median survival after diagnosis has sat stubbornly around 15 months for two decades despite surgery, radiation and chemotherapy. Personalised cancer vaccines are one of the more scientifically interesting attempts to change that — and one where the gap between mechanism and proven benefit is currently wide.
Why glioblastoma resists treatment
Four features make it exceptionally difficult.
It grows with diffuse infiltrating margins, sending cells far beyond any visible tumour edge, which means complete surgical removal is impossible in principle rather than in practice.
It is profoundly heterogeneous — different regions of the same tumour carry different mutations, so a therapy targeting one population leaves others to expand.
The blood-brain barrier restricts drug and immune cell access.
And it creates an immunosuppressive microenvironment, actively disabling the immune response around it. The brain is also immunologically distinctive, with its own resident immune cells and limited lymphatic drainage.
What a personalised cancer vaccine is
These are therapeutic, not preventive — they treat existing disease rather than preventing infection.
The approach: sequence the patient’s own tumour and healthy tissue, identify mutations that produce abnormal proteins, predict computationally which of these neoantigens are likely to be displayed on cell surfaces and recognised by T cells, manufacture a vaccine encoding the selected targets — often as mRNA — and administer it to train the immune system against markers found only on that patient’s cancer.
The appeal is precision: neoantigens arise from mutations absent in normal tissue, so the target is genuinely tumour-specific. The obstacles are manufacturing time measured in weeks, cost, and the fact that a tumour this heterogeneous can lose the targeted antigens.
Where the evidence stands
This is the part that requires care, because coverage of early results routinely outruns them.
Early-phase trials in glioblastoma have shown that these vaccines can be manufactured, are generally tolerable, and can generate detectable neoantigen-specific T cell responses — some of which have been found infiltrating the tumour. Individual patients with unusually long survival have been reported.
What has not been shown is a survival benefit in a randomised controlled trial. Phase 1 and 2 studies in glioblastoma typically involve small numbers of selected patients, without a control group, and glioblastoma survival varies enormously with age, performance status, extent of resection and MGMT methylation status — factors that can make a selected group look better than average for reasons unrelated to treatment.
The honest summary: the immunological proof of concept is real and encouraging. The clinical benefit is unproven, and this field has a long history of promising phase 2 results failing in phase 3.
The related area of mRNA cancer vaccines in melanoma has produced more mature randomised data, which is part of why the approach is being pursued in harder tumours (how these vaccines work).
What patients and families should know
Standard care remains maximal safe surgical resection, radiotherapy with temozolomide, and consideration of tumour-treating fields. Vaccines are studied in addition to this, not instead of it.
Access is through clinical trials, listed on ClinicalTrials.gov and available mainly at large academic centres. Eligibility is often restrictive, and tumour tissue must usually be collected at the initial surgery — so raising the question early matters.
Be sceptical of clinics offering personalised cancer vaccines outside trials for payment. Where a therapy is unproven, charging for it is a warning sign rather than an opportunity.
When to see a doctor
Seek urgent medical assessment for new persistent headaches that are worse in the morning or with straining, a first seizure at any age, progressive weakness or numbness on one side, new speech or vision disturbance, or personality and memory change over weeks. These require imaging, not watchful waiting.
Anyone with a glioblastoma diagnosis should ask their neuro-oncologist directly about trial eligibility and about MGMT and IDH testing, which affect both prognosis and treatment options.
The honest bottom line
Personalised neoantigen vaccines are a scientifically sound response to what makes glioblastoma untreatable, and early trials show they can provoke real anti-tumour immune responses. No randomised trial has yet shown they extend survival. That distinction is the whole story right now — the science is genuinely promising, and promising is not the same as proven.
Frequently asked questions
What is a personalized cancer vaccine?
It's a vaccine custom-made for one patient: scientists sequence that person's tumor, identify the unique mutated proteins it carries (neoantigens), and build a vaccine that trains the immune system to hunt cells displaying them.
Does this cure glioblastoma?
No. This was an early-phase trial showing the approach is feasible and encouraging, not a cure. Glioblastoma remains very hard to treat, and much larger trials are needed.
Are cancer vaccines available now?
Personalized cancer vaccines are still experimental and available mainly through clinical trials. They are a fast-moving area of research, not yet standard treatment.