Medulloblastoma Prognosis: Survival Rates and Key Factors

Roughly three out of four children diagnosed with medulloblastoma are alive five years later, with ten-year survival approaching 68% in large registry analyses covering both children and adults.¹ Those numbers, though, are averages that mask enormous variation. Depending on the tumor’s molecular identity, where it has spread, and how old the patient is at diagnosis, five-year survival can range from well above 90% down to roughly 40% or lower. Understanding which factors push the odds in which direction matters for treatment decisions, long-term planning, and knowing what questions to ask a care team.

Overall Survival in Children Versus Adults

A common assumption is that adults do worse than children with medulloblastoma. The reality is more nuanced. A SEER database analysis found that five-year survival was about 76% for children and 74% for adults, and at ten years the gap remained narrow, around 68% for children and 67% for adults.¹ A separate analysis confirmed no significant overall survival difference between children and adults, though the difference in excess hazard appeared only after about four years of follow-up, suggesting that late relapses affect adults more.²

The one age group that clearly fares worse is infants. Children under about three years of age at diagnosis have significantly lower survival than both older children and adults.³ Part of this stems from the difficulty of treating very young brains with radiation, which is a cornerstone of medulloblastoma therapy. Clinicians often delay or avoid cranial radiation in infants to protect the developing brain, but this can leave the tumor less well controlled. High-dose chemotherapy with stem-cell rescue has become an alternative strategy for these patients, and it has shown favorable results particularly for infants whose tumors belong to the SHH molecular subtype.⁴

Molecular Subtypes Are the Biggest Prognostic Driver

If there is one lesson from the past two decades of medulloblastoma research, it is that not all medulloblastomas are the same disease. Tumors are now classified into four molecular subgroups, and each carries a strikingly different outlook.

WNT-Activated Tumors

WNT-pathway medulloblastomas have the best prognosis of any subgroup, with survival exceeding 90%.⁵ The outcome is so good that ongoing clinical trials are asking whether treatment intensity can be safely reduced, sparing patients the long-term side effects of full-dose radiation and chemotherapy without sacrificing cure rates. Both children and young adults with WNT tumors share similarly excellent survival.

SHH Tumors

Sonic Hedgehog (SHH) medulloblastomas sit in the middle of the prognostic spectrum, but their outlook varies dramatically based on one genetic detail: whether the tumor carries a TP53 mutation. In children older than five, those with a TP53 mutation had a five-year survival of about 41%, compared to roughly 81% for SHH tumors without the mutation.⁶ TP53 mutations accounted for nearly three-quarters of deaths in that age group’s SHH tumors, making it the single most important risk factor within the subtype. In very young children, by contrast, SHH tumors often have desmoplastic histology and carry a considerably better prognosis.

Group 3

Group 3 tumors carry the worst overall prognosis among the four molecular subgroups. About one in five of these tumors harbors amplification of the MYC gene, a feature strongly linked to treatment failure.⁷ Group 3 tumors are also more likely to present with metastatic spread at diagnosis, compounding the challenge.

Group 4

Group 4 is the most common subtype overall. Its survival typically falls between that of WNT-activated and Group 3 tumors and depends heavily on whether the cancer has spread at the time of diagnosis. As discussed below, Group 4 tumors are also the one subgroup where the extent of surgical removal appears to independently affect progression-free survival.

Histology Still Matters

Before molecular subtyping became routine, pathologists classified medulloblastomas by how the cells looked under a microscope. Those histological categories still provide useful prognostic information, especially when molecular data is unavailable or when clinicians want a second layer of risk stratification.

In one study, five-year progression-free survival was about 82% for desmoplastic and extensively nodular variants, 78% for classic medulloblastoma, and just 44% for the large-cell/anaplastic variant.⁸ An international meta-analysis of young children reinforced this hierarchy: eight-year overall survival was 76% for desmoplastic/nodular tumors, 42% for classic tumors, and just 14% for the large-cell/anaplastic type.⁹ The difference between those numbers highlights how much the “average” survival figure for medulloblastoma glosses over. A child with a desmoplastic tumor and a child with a large-cell tumor are effectively dealing with different diseases.

Metastatic Spread at Diagnosis

Whether the tumor has spread beyond the original site at the time of diagnosis is one of the oldest and most consistent prognostic markers. Medulloblastoma tends to spread through the cerebrospinal fluid to other parts of the brain or spinal cord, and the staging system (M0 through M4) captures the degree of that spread.

Patients with no detectable spread (M0) have substantially better outcomes. In the HIT-2000 trial, children aged four and older with no metastases had a five-year progression-free survival of about 81%, compared to roughly 60% for those with either microscopic cells in the spinal fluid (M1) or visible metastases on imaging (M2/M3).¹⁰ An earlier prospective trial similarly identified metastases as an independent adverse factor, even after accounting for other variables.¹¹ The young-children meta-analysis put numbers on this as well: compared to children with no spread and a complete resection, those with metastatic disease faced roughly double the hazard of relapse or death.⁹

How Much Surgery Helps, and When It Doesn’t

Removing as much tumor as safely possible has long been the first step in medulloblastoma treatment. But a large retrospective study that accounted for molecular subgroup found the picture is more complicated than “more surgery is always better.” After controlling for molecular subtype, the survival benefit of a complete removal compared to a near-total removal essentially disappeared across all four subgroups.¹²

The exception was Group 4 tumors, where complete removal significantly improved progression-free survival compared to a subtotal removal, particularly in patients with metastatic disease. But even in Group 4, the overall survival benefit did not reach statistical significance. The practical takeaway from these findings is important: maximum safe resection remains the standard, but aggressively pursuing that last sliver of tumor is not recommended when the risk of neurological damage is high, because the survival gain from going from near-total to total removal is uncertain for most subtypes.

Radiation and Chemotherapy Considerations

Radiation to the brain and spine (craniospinal irradiation) is a critical component of therapy for most medulloblastoma patients older than about three. Across both children and adults, receiving radiation was the one prognostic factor that consistently predicted better survival regardless of other differences between the two age groups.¹

For patients classified as standard risk (no metastases, near-total or complete resection, favorable biology), clinicians have explored whether the radiation dose can be reduced to limit long-term side effects. Reduced-dose craniospinal radiation, combined with chemotherapy, appears feasible for standard-risk patients as long as staging and radiation delivery are done under optimal conditions.¹³ A more recent analysis found that lower-dose radiation was linked to somewhat higher relapse rates but better overall survival, likely because the addition of chemotherapy compensated for the increased relapse risk while reducing the burden of late toxicity.¹⁴

Proton beam therapy is increasingly used as an alternative to conventional photon radiation. A phase 2 trial found that proton therapy produced survival outcomes similar to conventional radiation with acceptable toxicity.¹⁵ Proton beams are better at sparing healthy tissue surrounding the target, and modeling studies project that this translates into fewer late complications over a patient’s lifetime. One such model estimated that over 30 years, the cumulative incidence of late effects was about 45% with protons versus 68% with conventional photon radiation.¹⁶ Access to proton facilities remains limited in many parts of the world, but where available, proton therapy is becoming a preferred option.

What Happens at Relapse

Relapsed medulloblastoma is one of the hardest situations in pediatric oncology. Prognosis after relapse depends on factors like the tumor’s molecular subgroup, whether the patient previously received radiation, and how soon the relapse occurs after initial treatment. Early relapse (within the first year) and very young age at recurrence are both associated with worse outcomes.¹⁷

For patients who were not irradiated during initial treatment (typically young children), craniospinal irradiation at relapse can meaningfully improve survival. A multicentre cohort found that irradiation at relapse and desmoplastic histology were both associated with longer survival after recurrence, while MYC amplification was associated with dramatically worse outcomes.¹⁸ Re-resection of the tumor at relapse was also linked to longer survival in that analysis.

Among young children with relapsed SHH medulloblastoma who had not been irradiated previously, three-year survival after relapse was about 62% overall and around 75% for those who received salvage craniospinal irradiation.¹⁷ For patients who already received radiation initially, the options narrow considerably, and the evidence for effective salvage strategies is much thinner.¹⁹

Late Effects and Quality of Survival

Surviving medulloblastoma is only part of the story. The treatments that produce cures, especially radiation to the developing brain, leave lasting marks. Common long-term problems include hearing loss, hormone deficiencies (particularly growth hormone), neurological impairments, and, over decades, the risk of second cancers.²⁰

Cognitive effects are among the most impactful. Survivors tend to score below average across multiple thinking domains, and the deficits do not stay stable; they tend to worsen with time. Processing speed, attention, and working memory are the most consistently affected abilities. In one long-term follow-up study of adult survivors, 90% had needed accommodations at school for learning disorders. Younger age at diagnosis predicted lower IQ and academic achievement scores, and longer time since treatment was associated with continued decline in working memory, a pattern that resembles early aging.²¹ Other common health complications in survivors included hearing impairment, second cancers, diabetes, hypertension, and endocrine deficiencies.

The risk of developing an entirely new, treatment-related cancer is a concern that grows over time. A population-based analysis of over 1,800 medulloblastoma patients found that about 6% developed a second primary cancer over a mean follow-up of roughly nine years. The most common sites were the brain, the thyroid, and the blood (leukemia). Among five-year survivors, the risk of dying from a second cancer eventually surpassed the risk of dying from other non-cancer causes beyond 15 years after diagnosis, and mortality after a second cancer diagnosis was about 52%.²² Another analysis of over 2,200 patients found that the rate of second tumors was roughly 3% at ten years and about 5% at fifteen, with female sex and longer overall survival being risk factors.²³ Proton therapy’s lower exit-dose exposure to surrounding tissue may eventually reduce these second-cancer rates, though early comparative data has not yet shown a significant difference between proton and photon patients.²⁴

Liquid Biopsy and Monitoring Without Surgery

One of the more promising developments in medulloblastoma care is the ability to track the tumor through fluid samples rather than repeated brain surgery. Researchers have found that tumor DNA shed into the cerebrospinal fluid can be analyzed to detect mutations, monitor response to treatment, and pick up residual disease that imaging misses.²⁵ In at least two cases described in that study, residual tumor DNA was detected in the spinal fluid even when MRI scans showed no visible disease, suggesting these liquid biopsies can catch minimal residual disease earlier than imaging can.

The approach is still mostly limited to research settings. A comprehensive review of liquid biopsy work in medulloblastoma found that cerebrospinal fluid is the most reliable source of tumor DNA for brain tumors (blood samples yield much less), and that both cell-free DNA and microRNA show promise as complementary biomarkers.²⁶ If validated in larger trials, liquid biopsies could eventually allow more precise treatment adjustments, either intensifying therapy for patients whose tumor DNA persists or de-escalating for those who clear it quickly.

Immunotherapy and Targeted Treatments

Immunotherapy has transformed the treatment of several adult cancers, but medulloblastoma has proven stubbornly resistant. A systematic review of immunotherapy trials in medulloblastoma patients found that clinical benefit was limited overall: progressive disease was the most common outcome, and complete responses were rare, occurring in only three of the patients studied. Median overall survival across immunotherapy trials ranged widely, from about one to 47 months, reflecting the heterogeneity of both the treatments tested and the patient populations.²⁷

CAR T-cell therapy, in which a patient’s own immune cells are engineered to recognize tumor markers, is being explored in early-phase trials. In a small safety study using HER2-targeted CAR T cells delivered directly into the tumor site, none of the three medulloblastoma patients treated experienced severe toxicity, but the results were mixed: two had disease progression and one had stable disease.²⁸ These are very early days, and the field is still working out which immune targets are viable and how to get immune cells past the blood-brain barrier in sufficient numbers. For the time being, immunotherapy for medulloblastoma is a research endeavor, not a proven clinical option.

Disparities in Access and Outcome

Survival statistics drawn from major registries reflect outcomes at centers with the resources to deliver multimodal treatment: surgery, radiation, and chemotherapy in careful sequence. But access to that full treatment chain is not uniform. A study comparing outcomes among children with central nervous system tumors in Southern Israel found that Bedouin children with medulloblastoma had three-year survival of 50%, compared to about 92% for Jewish children in the same region, with an adjusted risk of death more than three times higher after controlling for other variables.²⁹ Geographic distance from specialized treatment centers, delayed diagnosis, and socioeconomic barriers all plausibly contribute to gaps like these. The published survival rates for medulloblastoma are best understood as what is achievable under optimal conditions, not what every patient experiences.