CNS germ cell tumors show striking geographic and genetic disparities

Central nervous system germ cell tumors, rare cancers that arise mainly in the pineal and suprasellar regions of the brain, occur more frequently among East Asian populations than in Western countries, according to a review of international registry studies, clinical cohorts and genetic research. The tumors predominantly affect adolescents and young adults, show a consistent male excess, and may present in atypical locations more often in East Asian patients than in patients treated at Mayo Clinic. Although population data are not uniform and some national comparisons have found no statistically significant difference, evidence reviewed by Hirokazu Takami of the University of Tokyo and Matthew M. K. Shing of the Chinese University of Hong Kong indicates that the geographic disparity is substantial, complex and not simply a division between Asia and the West.

The review reports that CNS germ cell tumors account for roughly 2–5% of pediatric brain tumors in Western populations, compared with approximately 8–20% in East Asia. In the most recent US Central Brain Tumor Registry of the United States report, covering 814 children and adolescents diagnosed from 2017 through 2021, the average annual age-adjusted incidence was 0.20 per 100,000, with rates of 0.27 in males and 0.12 in females. An earlier US analysis covering 2006–2015 similarly found rates of 0.111 per 100,000 in males and 0.038 in females. These are low absolute rates, but CNS germ cell tumors stand out from other central nervous system cancers because their incidence rises sharply around puberty and because they make up a disproportionately large share of brain tumors in children and adolescents.

National figures from East Asia are mixed, illustrating why registry design and case definitions matter. A direct comparison of Japanese and US registries found an age-standardized incidence of malignant CNS germ cell tumors of 0.096 per 100,000 in Japan and 0.075 in the United States, a difference that was not statistically significant. South Korea’s nationwide study, however, identified CNS germ cell tumors as the most common childhood brain tumor subtype among children aged 10–14. Unpublished data presented in the review from Hong Kong registries for 2020–2024 showed an age-standardized incidence of 0.50 per 100,000 among people aged 0–19; rates in the 10–14 and 15–19 groups were 3.2–3.3 times the corresponding US estimates. Taiwan’s cancer registry found 2.17 cases per million person-years among people aged 0–19, while International Incidence of Childhood Cancer data from six Chinese registries yielded estimates of 2.1 per million at ages 0–14 and 2.2 per million at ages 0–19.

The higher frequency is not uniform throughout Asia. A large South Asian series from three Indian tertiary centers found that CNS germ cell tumors represented only 0.43% of all central nervous system tumors, which the investigators described as similar to Western frequencies. More detailed US analyses also show substantial variation within broad ethnic categories. In SEER data from 1992–2010, Asian and Pacific Islander patients had higher intracranial germ cell tumor incidence than White patients, particularly at ages 10–29. A later CBTRUS and SEER analysis found the highest age-adjusted germinoma incidence among Polynesians, at 2.246 per 100,000, followed by a heterogeneous “Other Asian” category at 1.478; the rate was 0.111 among South Asians. The most recent pediatric CBTRUS report found that non-Hispanic Asian or Pacific Islander children had 112% higher incidence than non-Hispanic White children. These small subgroup datasets require caution, but together they suggest that ancestry-specific risk is more informative than a simple national or continental comparison.

Geography is associated not only with how often CNS germ cell tumors occur but also with where they develop. The tumors have a strong midline predilection: in a classic series of 389 patients, 95% arose between the suprasellar and pineal regions, and another 6% involved both sites. Histology also varied by location, with 57% of germinomas arising suprasellarly and 68% of non-germinomatous germ cell tumors arising in the pineal gland. A comparison of 98 Mayo Clinic patients with more than 200 patients in a Japanese consortium found basal ganglia tumors in 8.4% of the Japanese cohort but none in the Mayo group, while bifocal neurohypophyseal and pineal involvement was more frequent at Mayo Clinic, affecting 18.8% rather than 5.8% of patients. The cohorts did not differ significantly in age or sex, leaving the basis for the anatomical disparity unresolved.

Age and sex patterns have nevertheless remained remarkably consistent across regions. Incidence is low in early childhood, increases around puberty and peaks among adolescents and young adults; in the original 389-case series, 68% of diagnoses occurred between ages 10 and 21. Overall male-to-female incidence ratios are generally around 2 to 4, with ratios near 3 to 4 in the recent Hong Kong, Japan and US comparisons, while pineal tumors can show a much larger male excess. Because this sex distribution appears broadly similar in high- and lower-incidence populations, the causes of geographic variation in absolute incidence may differ from the factors governing sex susceptibility. Proposed explanations for male bias include a Y-chromosome-associated microenvironment, the TSPY gene in the gonadoblastoma Y region, and a possible influence of the pubertal hormonal environment, but the biological basis remains incompletely resolved.

Genetic research offers a partial mechanism for the geographic disparity. The first genome-wide association study focused specifically on CNS germ cell tumors included 133 Japanese patients and 762 controls. It identified a common four-base-pair deletion, rs3831846, in an enhancer near BAK1 that was associated with increased disease risk, with an odds ratio of 2.46 and a 95% confidence interval of 1.83–3.31. Reporter assays indicated that the risk allele reduced enhancer activity, potentially lowering expression of BAK1, a pro-apoptotic protein negatively regulated by KIT and KITLG signaling during germ cell development. Reduced BAK1 expression could impair elimination of ectopically migrated primordial germ cells, although that proposed mechanism has not established the cause of the population pattern. The risk allele is more frequent in East Asian populations, at about 49%, than in European populations, at about 20%, but its contribution to the observed incidence differences has not been determined.

A later multi-ancestry genome-wide association study of pediatric germ cell tumors, encompassing testicular, ovarian, central nervous system and extragonadal tumors, included 1,927 cases and 10,601 controls. It confirmed the BAK1 association and identified additional loci involving SPRY4, DMRT1 and DEPTOR for germ cell tumors overall, as well as six loci associated specifically with the CNS subgroup, including regions near KIT, TFAM and LZTR1. Sequencing of 124 CNS germ cell tumors also found frequent, mutually exclusive somatic mutations in the MAPK pathway in 48.4% of cases, most commonly involving KIT, and less frequent mutations in the PI3K pathway. These findings support shared genetic architecture between gonadal and CNS germ cell tumors, but larger ancestry-resolved cohorts are needed to determine how much of the geographic variation BAK1 and related variants explain.

Rarity and inconsistent classification remain major barriers to measuring that variation. Incidence of roughly 0.1–0.2 per 100,000 leaves most national registries with small numbers of cases and wide confidence intervals for subgroups based on age, sex, ethnicity or histology. Site coding can obscure the true distribution of tumors: in one CBTRUS analysis, 31.6% of non-pineal cases were recorded only as “brain, not otherwise specified.” Classification is especially difficult for germinomas and non-germinomatous tumors, which have different prognoses and treatment requirements. In US population data, histological confirmation was reported for 90.8% of germinomas and 86.9% of non-germinomatous cases. By contrast, only 51% of participants in the SIOP CNS GCT 96 trial and 31.4% in the Children’s Oncology Group ACNS0122 trial underwent pretreatment surgery or biopsy; other cases were diagnosed through protocol-defined blood or cerebrospinal fluid marker thresholds. These approaches can avoid unnecessary surgery, but differing thresholds, assays and biopsy practices make international comparisons difficult.

The review’s authors identify harmonized case definitions, consistent anatomical coding and explicit reporting of diagnostic confirmation as priorities for future work. They also call for larger multinational genomic cohorts that preserve distinctions among ancestry, geography and age. Such collaboration is essential because no single registry is likely to have enough cases to resolve modest differences between populations, and because apparent differences can otherwise reflect hospital selection, coverage, diagnostic practices or unstable rates rather than biology. The combined evidence points to elevated CNS germ cell tumor risk in East Asian and some Pacific populations, a persistent male and adolescent pattern, and a plausible contribution from genetic variation. It also shows why more consistent international measurement is essential for determining whether ancestry, environment, ascertainment or additional biological mechanisms account for the striking variation.

Subject of Research: Global epidemiology, geographic variation and genetic susceptibility of central nervous system germ cell tumors

Article Title: Epidemiology of central nervous system germ cell tumors: global patterns, geographic variation, and methodological challenges

Article References: Takami, H., & Shing, M. M. K. (2026). Epidemiology of central nervous system germ cell tumors: global patterns, geographic variation, and methodological challenges. Frontiers in Oncology, 16, Article 1962617. https://doi.org/10.3389/fonc.2026.1962617

Image Credits: AI Generated

DOI: 10.3389/fonc.2026.1962617

Keywords: CNS germ cell tumors, brain cancer, epidemiology, East Asian populations, germinoma, BAK1 variant, genetic susceptibility, cancer registries, adolescent cancer, central, nervous, system