Central Nervous System Germ Cell Tumor

MONDO:0003000 Pathograph 13 Show in embeddings browser central nervous system neoplasm extragonadal germ cell tumor

Central nervous system germ cell tumors are rare, biologically heterogeneous tumors arising mainly in children and adolescents. The family includes germinoma and nongerminomatous germ cell tumors (NGGCTs), the latter comprising teratoma, embryonal carcinoma, yolk sac tumor, choriocarcinoma, and mixed histologies. Cross-sectional tumor profiling supports distinct developmental and molecular states across these groups, but does not by itself establish a single cell of origin, one universal MAPK/PI3K causal chain, or a molecular mechanism of refractory NGGCT. The MONDO disease term covers the CNS umbrella. Most directly applicable tumor and treatment evidence curated here is intracranial; explicitly labeled non-tumor developmental and extracranial germ-cell evidence remains indirect. None should be generalized to rare primary spinal CNS germ cell tumors without site-specific evidence.

Ask OpenScientist

Ask a research question about Central Nervous System Germ Cell Tumor. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).

Submitting...

Do not include personal health information in your question. Questions and results are cached in your browser's local storage.

1
Definitions
7
Pathophys.
6
Phenotypes
3
Hypotheses
3
Gaps
13
Pathograph
4
Genes
4
Medical Actions
4
Subtypes
1
Deep Research
3
Hyp. Reports
🏷

Classifications

Harrison's Part
ONCOLOGY HEMATOLOGY
📘

Definitions

1
CNS germ cell tumor literature scope
A CNS germ cell tumor family that primarily affects children and adolescents and is divided clinically into germinoma and NGGCT; individual components range from mature teratoma to malignant histologies.
OTHER MONDO-level CNS disease-family scope encompassing germinoma and nongerminomatous germ cell tumors. Core disease evidence is intracranial; non-tumor developmental and extracranial comparator studies are explicitly indirect.
Show evidence (1 reference)
PMID:42419530 SUPPORT Other
"Intracranial germ cell tumors (IGCTs) are a group of malignant central nervous system (CNS) tumors primarily affecting children and adolescents, whose pathogenesis remains incompletely understood."
The seed review supports the broad CNS tumor-family scope, typical age group, and unresolved pathogenesis, but its malignant-tumor wording does not capture mature teratoma within the broader ontology class.
◆

Subtypes

4
CNS germinoma MONDO:0002999
The germinoma arm of the CNS germ cell tumor family. Human tumor methylation and transcriptome studies find a primordial-germ-cell-like, globally hypomethylated state, but resemblance to a developmental reference state is not direct lineage evidence.
Show evidence (1 reference)
PMID:28078450 SUPPORT Human Clinical
"We showed that pure germinomas are characterized by global low DNA methylation, a unique epigenetic feature making them distinct from all other iGCTs subtypes."
Genome-wide methylation profiling of human tumors supports a distinct germinoma epigenetic state without proving developmental ancestry.
CNS NGGCT MONDO:0020574
The non-germinoma arm, encompassing several pure and mixed histologies with different differentiation states, biomarkers, treatment responses, and relapse patterns. This umbrella should not be represented as though every component shares one resistance mechanism.
Show evidence (1 reference)
PMID:35137206 SUPPORT Human Clinical
"Genome-wide transcriptome analysis in CNS GCTs indicated that germinoma had a transcriptomic profile representative of primitive cells during early embryogenesis with high meiosis/mitosis potentials, while nongerminomatous GCTs (NGGCTs) had differentiated phenotypes oriented toward tissue..."
Integrative human tumor profiling supports molecular and developmental state differences between germinoma and NGGCT.
CNS embryonal carcinoma MONDO:0018843
The embryonal carcinoma histology within the CNS NGGCT spectrum, previously named only in this entry's prose. It is the least differentiated non-germinomatous component, and its disease-level mechanism is curated in the Embryonal Carcinoma entry rather than duplicated here.
Show evidence (1 reference)
PMID:42115464 SUPPORT Human Clinical
"NGGCTs include embryonal carcinoma, yolk sac tumor, choriocarcinoma, and teratomas and often secrete tumor markers like alpha-fetoprotein (AFP) and beta-hCG, which guide diagnosis and treatment without biopsy."
Names embryonal carcinoma as one of the intracranial NGGCT histologies, supporting it as a subtype of this entry.
CNS teratoma MONDO:0002718
A CNS-specific teratomatous subtype within the NGGCT spectrum. Mature, immature, growing-teratoma-syndrome, and somatic-type-malignancy contexts require separate histopathologic interpretation; this subtype does not imply that every CNS teratoma is malignant.
Show evidence (1 reference)
PMID:37452948 SUPPORT Other
"In the WHO CNS tumor classification, GCTS are classified as germinoma and nongerminomatous GCT (NGGCT) with different subtypes (including teratoma)."
The clinical review places teratoma within the CNS NGGCT classification while separately noting that mature teratoma is not malignant.
◈

Mechanistic Hypotheses

3
Common-Ancestry and Developmental-State Divergence Model
common_ancestry_developmental_state_divergence EMERGING Central Nervous System Germinoma Central Nervous System Nongerminomatous Germ Cell Tumor
Evidence balance 8 support
Histologically distinct components within an individual mixed CNS germ cell tumor may descend from a common tumor ancestor and subsequently diverge into a primordial-germ-cell-like, hypomethylated germinoma state or one of several more differentiated NGGCT states. Cross-sectional resemblance does not identify that ancestor: a mis-migrated primordial germ cell and an endogenous neural stem or progenitor cell are competing cell-of-origin models to be tested rather than assumed. MAPK/PI3K-pathway alterations may cooperate with developmental state and can be shared by distinct components of a mixed tumor, but their temporal position and necessity for initiation remain unresolved.
This is a within-tumor developmental model inferred mainly from cross-sectional human profiles. It does not posit an ancestor shared across patients, discriminate a mis-migrated primordial germ cell from an endogenous neural stem or progenitor cell or another precursor, or generalize germinoma hypomethylation to every NGGCT component.
Deep research
OpenScientist citations 15 citations 2026-07-26T09:20:35.182369
Show evidence (8 references)
PMID:28078450 SUPPORT Human Clinical
"The patterns of methylation strongly resemble that of primordial germ cells (PGC) at the migration phase, possibly indicating the cell of origin for these tumors."
The primary methylation study supports PGC-like resemblance while its explicitly tentative wording preserves the absence of lineage tracing.
PMID:28078450 SUPPORT Human Clinical
"Unlike PGC, however, hypomethylation extends to long interspersed nuclear element retrotransposons."
The human-tumor methylome is PGC-like but not an exact PGC epigenome, so global resemblance cannot by itself identify the normal precursor.
PMID:20582452 SUPPORT In Vitro
"We show here that the lack of methylation of SNRPN that has previously been taken to indicate an origin for GCTs from PGCs is also seen in neural stem cells of mice and humans."
A mouse-and-human neural-stem-cell assay shows that one imprinting feature used to support PGC origin is not PGC-specific. It does not establish a neural-stem-cell lineage for the tumor.
+ 5 more references
NGGCT Failure-State-Specific Persistence and Evolution Model
nggct_failure_state_specific_persistence_and_evolution EMERGING Central Nervous System Nongerminomatous Germ Cell Tumor
Evidence balance 9 support
CNS-NGGCT failures are clinically partitioned into induction progression, marker-negative growing teratoma syndrome (GTS), viable residual or progressive disease, and post-treatment relapse. Clinical separability does not establish molecular separability. GTS may reflect persistence or expansion of a pre-existing teratomatous lineage or treatment-associated maturation and must be distinguished from viable malignant disease; it is not uniformly benign. Malignant progression or relapse may reflect selection of a pre-existing malignant clone or state, acquired genomic or epigenetic changes, reversible treatment-tolerant adaptation, treatment exposure or geography, or sampling. These alternatives must not be pooled or assumed to be therapy-selected without paired, component-resolved longitudinal evidence. No adequately powered diagnosis-to-failure molecular series currently establishes which route is causal.
This model operationally separates GTS, viable malignant progression, and post-treatment relapse without claiming that each category has a distinct molecular cause. It does not promote SHH, MYC/E2F, immune-hot, MAPK, or PI3K status to a clinical resistance biomarker without longitudinal validation.
Deep research
OpenScientist citations 16 citations 2026-07-26T09:35:56.635159
Show evidence (9 references)
PMID:35218656 SUPPORT Human Clinical
"A total of 118 patients experienced a treatment failure. Twenty-four patients had progressive disease during therapy, and additional 11 patients were diagnosed with growing teratoma syndrome (GTS)."
The pooled clinical dataset operationally separates progression during therapy from GTS, but it does not demonstrate molecularly distinct causes.
PMID:35218656 SUPPORT Human Clinical
"Eighty-three individuals experienced disease relapses after treatment ended."
The same pooled cohort separately documents post-treatment relapse, supporting a timing-based clinical partition from progression during therapy and GTS rather than a proven mechanistic partition.
PMID:42488730 SUPPORT Human Clinical
"Subsequent surgical resection revealed a mature teratoma without malignant elements, consistent with growing teratoma syndrome (GTS)."
This single post-chemotherapy resection supports a teratomatous GTS branch in one pineal mixed tumor. It does not distinguish treatment-associated maturation from persistence or expansion of a pre-existing component.
+ 6 more references
miR-214-3p–BCL2L11 Cisplatin-Response Candidate
mir214_bcl2l11_cisplatin_response_candidate EMERGING Central Nervous System Nongerminomatous Germ Cell Tumor
Evidence balance 3 support
In a subset of malignant NGGCT, altered regulation of the miR-199/214 cluster may increase miR-214-3p, reduce the pro-apoptotic protein BCL2L11/BIM, and shift apoptosis and survival after cisplatin exposure. This is a candidate response-modifying chain: the current causal evidence comes from forced miR-214-3p expression and a BCL2L11-knockdown phenocopy in one adult mediastinal mixed-germ-cell-tumor line. It does not establish direct miR-214-3p binding to BCL2L11, endogenous necessity, rescue, or a difference between resistant and sensitive CNS NGGCT in patients.
This is a cisplatin-response candidate, not a validated intrinsic-tolerance mechanism or clinical resistance biomarker. No independent functional replication in a patient-derived intracranial model or resistant-versus-sensitive clinical comparison was retrieved. NCCIT is not a patient-derived CNS NGGCT model; testicular or mediastinal GCT apoptosis and DNA-repair studies and non-germ-cell miR-214 targets remain indirect alternatives, not direct CNS evidence or refutation.
Deep research
OpenScientist citations 17 citations 2026-07-26T09:56:55.186601
Show evidence (3 references)
PMID:29036598 SUPPORT Human Clinical
"The expression levels of 97 genes and 8 miRNAs were correlated with promoter DNA methylation and hydroxymethylation status, such as the miR-199/-214 cluster"
Human tumor multi-omic analysis links the miR-199/214 cluster to methylation state in a cross-sectional NGMGCT-versus-germinoma subtype comparison. Histology and differentiation state therefore confound any resistance inference, and no longitudinal resistant-versus-sensitive comparison was performed.
PMID:29036598 SUPPORT In Vitro
"Overexpresssion of miR-214-3p in NCCIT cells leads to reduced expression of the pro-apoptotic protein BCL2-like 11 and induces cisplatin resistance."
Stable miR-214-3p expression increased NCCIT survival after cisplatin but the authors reported no etoposide difference (data not shown), and BCL2L11 knockdown phenocopied cisplatin survival. This supports a single-cell-line in-vitro response phenotype, but the study did not test endogenous miR-214-3p loss, AGO2 or seed-mutant binding, or miRNA-insensitive BCL2L11 rescue.
PMID:36768818 SUPPORT In Vitro
"Comparing CDDP-resistant vs. -sensitive TGCT cell lines, we identified miR-218-5p, miR-31-5p, miR-125b-5p, miR-27b-3p, miR-199a-5p, miR-214-3p, let-7a and miR-517a-3p as significantly up-regulated"
This extracranial testicular-GCT screen supplies an independent but nonfunctional association. RT-qPCR miR-214-3p direction varied by resistant-versus-sensitive cell-line comparison; miR-214-3p was not among the authors' six-miRNA proposed core biomarker panel, although they suggested possible added predictive value. The study did not test BCL2L11 or perturb miR-214-3p.
?

Discussions and Knowledge Gaps

3
Do CNS germ cell tumors arise from a mis-migrated primordial germ cell, an endogenous neural stem or progenitor cell, or another precursor; when do lineage divergence and global methylation changes occur; and are MAPK/PI3K alterations initiating lesions, later cooperating events, or dispensable markers of particular components?
KNOWLEDGE GAP OPEN gap_cns_gct_developmental_origin_and_driver_timing
Attached to
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
The strongest human evidence is cross-sectional. Methylation profiles in 61 primary tumors and integrated transcriptome/methylome profiles in 84 cases map germinoma and NGGCT to different developmental reference states, while shared MAPK/PI3K variants in microdissected mixed components and component-shared 12p status in a separate cohort support within-tumor common ancestry. SNRPN hypomethylation is also present in mouse and human neural stem cells, and two non-tumor fetal autopsies prompted an author hypothesis of local germinal-matrix derivation. These observations do not trace a tumor lineage. The studies do not observe embryonic migration, establish whether the relevant precursor is a bona fide primordial germ cell, an endogenous neural stem or progenitor cell, or another precursor, or determine whether pathway alterations precede or follow epigenetic state divergence. The 62-tumor sequencing cohort found KIT/RAS or AKT/mTOR lesions in 53%, with KIT mutations restricted to germinomas in that cohort, so one pathway chain cannot be generalized across the family. Direct temporal, lineage, and perturbational evidence that compares competing precursor models is required before adding an established developmental-escape or MAPK/PI3K pathophysiology edge.
Proposed experiments
Spatial single-cell clonal atlas across pure and mixed CNS germ cell tumors
spatial single-cell multi-omic lineage study Relation: this experiment is of type this experiment type This experiment is of type spatial single-cell multi-omic lineage study.
exp_cns_gct_spatial_clonal_developmental_atlas
Prospectively collect treatment-naive pure germinomas, pure NGGCT histologies, and multiple spatially mapped components of mixed tumors. Integrate single-cell DNA, RNA, chromatin accessibility, methylation, and spatial pathology with matched normal DNA and a preregistered human embryonic germ-cell reference atlas.
Model systems
Treatment-naive CNS germ cell tumor cohort
Include age-, sex-, location-, and histology-stratified tumors with central pathology review and separate sampling of every visible mixed component.
Human developmental reference atlas
Use ethically sourced published and prospectively generated reference profiles spanning primordial germ-cell specification and migration, neural stem and progenitor states, and early embryonic differentiation without treating nearest-neighbor similarity as proof of ancestry.
Perturbations
Component-resolved sampling
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
Laser-capture and single-cell profile each component separately while preserving spatial relationships and shared somatic variants.
Readouts
Clonal order and developmental-state assignment
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
Infer trunk and branch lesions, allele-specific copy number, methylation erasure or acquisition, lineage programs, and uncertainty-calibrated similarity to developmental references.
single-cell DNA and RNA sequencing Relation: this readout is measured by this assay This readout is measured by single-cell DNA and RNA sequencing. single-cell methylome and chromatin accessibility profiling Relation: this readout is measured by this assay This readout is measured by single-cell methylome and chromatin accessibility profiling. spatial transcriptomics Relation: this readout is measured by this assay This readout is measured by spatial transcriptomics.
Direction: THRESHOLD DEPENDENT
Controls
Histology and site controls
Analyze germinoma and each NGGCT component separately, include gonadal germ cell tumor comparators, and adjust for purity, immune infiltration, age, sex, anatomic site, and specimen handling.
Clonal-inference controls
Confirm trunk variants by orthogonal targeted sequencing and test alternative phylogenies rather than selecting a lineage tree from transcriptomic similarity alone.
Decision criterion
Common ancestry is supported when spatially distinct mixed components share orthogonally validated trunk lesions and diverge into reproducible component-specific epigenetic states. A particular precursor-of-origin model gains support only if tumor trunk states reproducibly align with a defined developmental interval across cohorts, fit competing precursor models materially less well, and cannot be explained by tumor purity or convergence; developmental similarity alone does not establish lineage. Divergent trunk lesions across components would refute a single common tumor ancestor, whereas component differences arising after a shared trunk are expected under developmental-state divergence and do not refute it. Inconsistent developmental assignments leave cell of origin unresolved rather than refuting common ancestry.
Would support
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
Would refute
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
Stage-specific MAPK/PI3K developmental epistasis
isogenic developmental-stage causal epistasis study Relation: this experiment is of type this experiment type This experiment is of type isogenic developmental-stage causal epistasis study.
exp_cns_gct_stage_specific_driver_epistasis
Introduce recurrent KIT, RAS, AKT1, or mTOR alterations singly and in combinations at defined stages of human primordial-germ-cell-like-cell and embryonic differentiation, then test whether a lesion initiates, maintains, or merely accelerates a tumor-compatible state.
Model systems
Isogenic human candidate-precursor differentiation series
Compare multiple donor backgrounds of primordial-germ-cell-like cells and CNS neural stem or progenitor cells, with inducible editing at matched specification and differentiation stages.
Component-specific CNS germ cell tumor organoids
Establish centrally reviewed germinoma and NGGCT organoids with isogenic correction or introduction of candidate lesions.
Perturbations
Timed driver introduction and correction
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
Activate or correct one candidate pathway lesion at each developmental stage, with matched kinase-dead, vector, and unedited controls.
Readouts
State maintenance and transformation
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
Measure survival without developmental cues, methylation state, differentiation trajectory, clonogenicity, invasion, and tumor formation in an anatomically relevant xenograft.
longitudinal multi-omic profiling Relation: this readout is measured by this assay This readout is measured by longitudinal multi-omic profiling. clonogenic transformation assay Relation: this readout is measured by this assay This readout is measured by clonogenic transformation assay.
Direction: POSITIVE
Controls
Lesion and developmental-stage controls
Compare each alteration across matched primordial-germ-cell-like and neural stem or progenitor backgrounds and developmental stages, include pathway correction and pharmacologic rescue, and analyze germinoma-like and NGGCT-like outputs separately.
Decision criterion
An initiating role requires a lesion to reproducibly create a self-maintaining tumor-compatible state only when introduced at a defined precursor stage, with loss of that phenotype after isogenic correction. A maintenance role is favored if correction collapses established tumor state but introduction does not initiate it. Lack of necessity and sufficiency supports a passenger or context-marker interpretation.
Would support
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
Would refute
mechanistic_hypotheses#common_ancestry_developmental_state_divergence
Show evidence (1 reference)
PMID:28078450 SUPPORT Human Clinical
"Recent exhaustive genomic studies showed that mutations of the genes involved in the MAPK and/or PI3K pathways are common in iGCTs; however, the mechanisms of how different subtypes develop, often as a mixed-GCT, are unknown."
The primary study explicitly identifies the unresolved link between recurrent pathway alterations and subtype development.
Which cellular, lineage, genomic, epigenetic, microenvironmental, or pharmacologic states cause primary refractory disease, viable progression, growing teratoma syndrome, and post-treatment relapse in CNS NGGCT, and which are actionable rather than correlates of histology?
KNOWLEDGE GAP OPEN gap_nggct_refractory_and_relapse_mechanism
Attached to
has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution mechanistic_hypotheses#mir214_bcl2l11_cisplatin_response_candidate
Clinical failure is heterogeneous. A pooled analysis separated 24 progressions during treatment, 11 growing-teratoma-syndrome events, and 83 post-treatment relapses. Large genomic studies predominantly profile primary tumors, and the recent NGGCT cohort sequenced only 12 paired tumor-blood samples; neither design compares diagnosis with residual, progressive, and relapsed viable tumor within patients. A separate study observed early-branched evolution across only two histology-changing longitudinal cases, which supports feasibility but not a recurrent driver. An older serial yolk-sac-tumor case likewise showed increasing copy-number complexity over two recurrences, but its single-case, arm-level design and unreported regimen or exposure details do not establish a resistance locus. Molecular subtypes such as immune-hot, MYC/E2F, and SHH are therefore candidates, not validated resistance states. The miR-214-3p–BCL2L11 result supplies a specific in-vitro cisplatin-tolerance candidate, but lacks a resistant-versus-sensitive clinical cohort, a patient-derived CNS model, and rescue evidence. Histologic mixture, sampling bias, radiation geography, pharmacokinetics, and therapy-selected clonal evolution remain competing explanations. A resistance node, predictive biomarker, or targeted therapy should not be added until longitudinal molecular change is linked to function and outcome.
Proposed experiments
Paired diagnosis-residual-progression-relapse NGGCT evolution cohort
prospective longitudinal tumor evolution study Relation: this experiment is of type this experiment type This experiment is of type prospective longitudinal tumor evolution study.
exp_nggct_paired_failure_evolution_cohort
Build an international prospective cohort with central pathology review, multi-region diagnosis sampling, protocol-level treatment exposure, and tissue at second-look surgery, progression, GTS resection, or relapse. Collect serial plasma and CSF before, during, and after therapy.
Model systems
Longitudinal CNS NGGCT clinical cohort
Enroll pure and mixed NGGCT histologies and preserve separate strata for induction nonresponse, GTS, local viable progression, metastatic progression, and late relapse.
Response-matched comparator cohort
Include durable responders matched on stage, histology, markers, radiation field, age, sex, location, and treatment regimen.
Perturbations
Protocol-defined chemotherapy and radiotherapy exposure
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
Record cumulative agent doses, dose intensity, delays, surgery, radiation volumes and dosimetry, and supportive medications as time-varying exposures.
Readouts
Within-patient clonal and lineage-state evolution
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
Compare somatic variants, copy number, ecDNA, methylation, transcriptome, chromatin accessibility, immune state, and histologic composition across every within-patient time point.
single-cell and bulk multi-omic sequencing Relation: this readout is measured by this assay This readout is measured by single-cell and bulk multi-omic sequencing. spatial pathology Relation: this readout is measured by this assay This readout is measured by spatial pathology.
Direction: THRESHOLD DEPENDENT
Serial molecular residual disease
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
Track patient-specific mutations, copy-number changes, methylation, and validated germ-cell-tumor RNAs in CSF and plasma alongside AFP, beta-hCG, imaging, and clinical response.
cerebrospinal-fluid circulating tumor DNA assay Relation: this readout is measured by this assay This readout is measured by cerebrospinal-fluid circulating tumor DNA assay. serial tumor-marker measurement Relation: this readout is measured by this assay This readout is measured by serial tumor-marker measurement.
Direction: POSITIVE
Controls
Failure-state separation
Do not pool marker-negative GTS, viable malignant progression, and post-treatment relapse. Require central pathology and viable-tumor content estimates for every tissue comparison.
Sampling and treatment controls
Model tumor purity, spatial sampling, histology, treatment intensity, radiation geography, and time to failure; validate candidate alterations in an independent protocol-treated cohort.
Decision criterion
Therapy-selected persistence is supported when a diagnosis-time subclone or state survives, expands, or acquires a reproducible change in viable residual or relapse tumor, precedes failure in CSF, and predicts failure beyond histology, stage, markers, and radiation field. GTS requires a teratomatous lineage result rather than being labeled molecular relapse. Failure to find reproducible within-patient evolution, or associations fully explained by component sampling or treatment geography, refutes a shared molecular-resistance model.
Would support
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
Would refute
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
Functional reconstruction of candidate NGGCT resistance states
patient-derived resistance epistasis and rescue study Relation: this experiment is of type this experiment type This experiment is of type patient-derived resistance epistasis and rescue study.
exp_nggct_resistance_functional_reconstruction
Establish paired diagnosis and failure models from the longitudinal cohort, barcode diagnosis-time cells, and causally test candidate lineage, signaling, DNA-damage-response, and microenvironmental states under clinically matched chemotherapy and radiation schedules.
Model systems
Paired patient-derived NGGCT organoids and xenografts
Preserve component-specific histology and use centrally reviewed paired diagnosis and failure material whenever available.
Isogenic barcoded NGGCT models
Introduce or correct candidate alterations in multiple genetic and histologic backgrounds rather than relying on one cell line.
Perturbations
Clinically matched multimodal therapy
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
Apply carboplatin or cisplatin, etoposide, ifosfamide, and fractionated radiation singly and in protocol-relevant sequences.
Candidate-state ablation and rescue
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
Genetically suppress and rescue only candidates that were observed to expand within patients, including lineage programs, signaling nodes, DNA-repair states, or immune interactions.
miR-214-3p inhibition and BCL2L11 epistasis
mechanistic_hypotheses#mir214_bcl2l11_cisplatin_response_candidate
In models with endogenous miR-214-3p elevation, inhibit miR-214-3p, knock down or restore BCL2L11 independently, and perform concentration-matched rescue under cisplatin and non-platinum therapy.
Readouts
Clonal survival and cross-resistance
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
Measure barcode abundance, viability, apoptosis, DNA damage and repair, differentiation, tumor regrowth, and cross-resistance to chemotherapy and radiation.
clonal barcode tracking Relation: this readout is measured by this assay This readout is measured by clonal barcode tracking. drug and radiation response assay Relation: this readout is measured by this assay This readout is measured by drug and radiation response assay.
Direction: POSITIVE
Controls
Isogenic and histology controls
Include unedited, corrected, and rescue controls; test at least two models per histologic state and compare with durable-response models.
Decision criterion
A candidate is a causal resistance mechanism only if it is enriched longitudinally in patients, increases survival under relevant therapy in more than one model, loses that advantage after specific ablation, and regains it after rescue. Baseline association without within-patient enrichment, or model-only resistance without clinical recurrence, is insufficient.
Would support
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution mechanistic_hypotheses#mir214_bcl2l11_cisplatin_response_candidate
Would refute
mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution mechanistic_hypotheses#mir214_bcl2l11_cisplatin_response_candidate
Show evidence (2 references)
PMID:35218656 SUPPORT Human Clinical
"Due to their rarity, patients with treatment failures remain a poorly characterized group with unfavorable outcomes."
The pooled clinical-trial analysis explicitly identifies the weak characterization of CNS-NGGCT treatment failure.
PMID:39189644 SUPPORT Human Clinical
"Among them, ctDNA before chemotherapy as well as ctDNA clearance were consistently associated with PFS and OS"
This 17-patient retrospective study supports serial CSF ctDNA as a feasible response and residual-disease readout, not as a mediator of resistance; its positivity threshold was optimized within the same small survival dataset.
For localized CNS NGGCT that responds to induction therapy, what radiation field preserves disease control, especially in the spine, while reducing neurocognitive, endocrine, vascular, and second-neoplasm risk when the protocol-defined dose schedule is held constant?
CONTROVERSY OPEN controversy_localized_nggct_radiation_field
Attached to
has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
ACNS1123 gave 30.6 Gy whole-ventricular irradiation with a boost bringing the tumor-bed total to 54 Gy in 66 response-qualified patients. Eight relapsed and every failure involved the spine; central dosimetry review found protocol-conforming targets and dose in 62 of 66 patients, and none of the four patients with deviations progressed. A separate pooled analysis of four prospective trials and two national protocol cohorts found no association between focal or whole-ventricular radiation and metastatic rather than local relapse among patients who had already failed treatment. Conditioning on failure and omitting field-specific denominators for all treated patients means that result cannot estimate field-specific failure incidence or rebut the ACNS1123 pattern. The seed review describes optimal field and dose as controversial and minimizing long-term sequelae while preserving control as an unresolved challenge. Comparative disease-control and late-effect evidence is missing; no universal conclusion that either craniospinal or reduced-field radiation is optimal is warranted. This question is intentionally limited to response-qualified localized NGGCT and field selection at a fixed protocol-defined dose; dose optimization and germinoma protocols require separate evaluation.
Proposed experiments
Risk-adapted radiation field trial with molecular residual-disease surveillance
randomized risk-adapted radiation field trial Relation: this experiment is of type this experiment type This experiment is of type randomized risk-adapted radiation field trial.
exp_localized_nggct_risk_adapted_radiation_trial
In a prospective international trial, centrally confirm localized NGGCT, induction response, histologic composition, serum and CSF markers, and second-look pathology. Compare protocol-defined radiation fields under contemporary protocol consensus and ethical oversight while embedding serial CSF molecular residual-disease testing and lifelong late-effect follow-up.
Model systems
Response-qualified localized CNS NGGCT cohort
Stratify by complete versus partial response, pure versus mixed histology, residual teratoma, marker status, age, sex, location, and protocol-defined molecular residual disease.
Perturbations
Protocol-defined radiation-field assignment
has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
Compare reduced ventricular or ventricular-plus-spinal strategies with a consensus broader-field control, holding the prescribed dose schedule and systemic therapy constant and using centrally reviewed dose-volume constraints.
Readouts
Disease-control and failure geography
has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
Measure event-free and overall survival, local, ventricular, spinal, and disseminated relapse, salvage success, serial tumor markers, and CSF molecular residual disease.
magnetic resonance surveillance Relation: this readout is measured by this assay This readout is measured by magnetic resonance surveillance. cerebrospinal-fluid molecular residual-disease assay Relation: this readout is measured by this assay This readout is measured by cerebrospinal-fluid molecular residual-disease assay.
Direction: THRESHOLD DEPENDENT
Long-term treatment burden
has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
Prospectively assess neurocognition, school and employment function, endocrine replacement, growth, hearing, vascular events, and secondary neoplasms for at least ten years.
longitudinal late-effect assessment Relation: this readout is measured by this assay This readout is measured by longitudinal late-effect assessment.
Direction: NEGATIVE
Controls
Central treatment-quality control
Review pathology, response, second-look surgery, imaging, target contours, dosimetry, and protocol adherence before or during treatment.
Prespecified competing-risk analysis
Lock noninferiority and late-effect estimands, account for salvage and competing mortality, and report every failure location rather than only aggregate progression-free survival.
Decision criterion
A reduced field is acceptable only if it meets a prespecified noninferiority margin for disease control without an excess of spinal or disseminated relapse and produces a clinically meaningful late-effect benefit. If baseline or on-treatment CSF molecular residual disease reproducibly identifies spinal risk, test a biomarker-adapted field strategy prospectively rather than assigning a field from association alone. Similar disease control without measurable long-term benefit does not justify field reduction.
Show evidence (5 references)
PMID:31545689 SUPPORT Human Clinical
"However, the patterns of failure were distinct, with all patients having treatment failure in the spine."
The response-based phase II trial directly identifies the spinal failure pattern after reduced whole-ventricular radiation.
PMID:34990779 SUPPORT Human Clinical
"Eight patients relapsed: 6 with isolated spinal relapse and 2 with disease in the brain and spine."
Central review confirms the number and geography of failures in the reduced-field cohort.
PMID:35218656 SUPPORT Human Clinical
"In our analysis, focal or whole-ventricular radiation therapy was not associated with an increased risk of metastatic relapses."
This case-only failure cohort describes metastatic versus local relapse patterns among patients who failed. Without field-specific denominators among all treated patients, it cannot estimate comparative failure incidence or establish that reduced fields are noninferior.
+ 2 more references
⚙

Pathophysiology

7
Recurrent KIT-RAS Signaling Alterations
Somatic alterations in KIT and other KIT/RAS-pathway genes recur in human intracranial germ cell tumors. They are enriched in germinoma but are not present in every tumor. Cross-sectional sequencing establishes recurrence, not whether these lesions initiate disease or are required to maintain it.
KIT hgnc:6342 HUGO Gene Nomenclature Committee (hgnc) Relation: this pathophysiological event involves this gene This pathophysiological event involves KIT (hgnc:6342). hgnc:6342 is a gene from the HUGO Gene Nomenclature Committee. KRAS hgnc:6407 HUGO Gene Nomenclature Committee (hgnc) Relation: this pathophysiological event involves this gene This pathophysiological event involves KRAS (hgnc:6407). hgnc:6407 is a gene from the HUGO Gene Nomenclature Committee. NRAS hgnc:7989 HUGO Gene Nomenclature Committee (hgnc) Relation: this pathophysiological event involves this gene This pathophysiological event involves NRAS (hgnc:7989). hgnc:7989 is a gene from the HUGO Gene Nomenclature Committee.
Show evidence (2 references)
PMID:24896186 SUPPORT Human Clinical
"We find the KIT/RAS signalling pathway frequently mutated in more than 50% of IGCTs, including novel recurrent somatic mutations in KIT, its downstream mediators KRAS and NRAS, and its negative regulator CBL."
Sequencing of patient tumors establishes recurrent somatic pathway alterations, while neither temporal order nor causal necessity was tested.
PMID:38409885 SUPPORT Human Clinical
"We found that KIT was the most significantly mutated gene (15/47, 32%), which mainly occurred in the germinoma group (13/20, 65%), and less frequently in NGGCT (2/27, 7%)."
A second human tumor cohort supports subtype enrichment and also shows that KIT mutation is neither universal nor confined absolutely to germinoma.
Recurrent AKT-mTOR Signaling Alterations
A subset of intracranial germ cell tumors carries AKT/mTOR-pathway lesions, including focal AKT1 copy-number gain with increased AKT1 expression. Recurrence supports a tumor molecular state but does not establish when the alteration arose or whether it is sufficient for transformation.
AKT1 hgnc:391 HUGO Gene Nomenclature Committee (hgnc) Relation: this pathophysiological event involves this gene This pathophysiological event involves AKT1 (hgnc:391). hgnc:391 is a gene from the HUGO Gene Nomenclature Committee.
Show evidence (1 reference)
PMID:24896186 SUPPORT Human Clinical
"Novel somatic alterations in the AKT/mTOR pathway included copy number gains of the AKT1 locus at 14q32.33 in 19% of patients, with corresponding upregulation of AKT1 expression."
Patient-tumor genomic and expression measurements establish a recurrent AKT1-associated state in a minority of cases, not a universal initiating mechanism.
Germinoma Global DNA Hypomethylation
Pure germinomas show a globally hypomethylated tumor state distinct from other intracranial germ cell tumor subtypes. Its resemblance to migrating primordial germ cells is relevant to developmental-origin hypotheses but is not lineage tracing.
Show evidence (1 reference)
PMID:28078450 SUPPORT Human Clinical
"We showed that pure germinomas are characterized by global low DNA methylation, a unique epigenetic feature making them distinct from all other iGCTs subtypes."
Genome-wide profiling of patient tumors directly supports the subtype-specific methylation state without establishing its causal timing.
Chromosome 12p Gain
Additional 12p copy number occurs across CNS germ cell tumors but is more frequent in NGGCT and in tumors with malignant components. It is a prognostic genomic feature, not a demonstrated universal driver.
Show evidence (2 references)
PMID:34698864 SUPPORT Human Clinical
"A 12p gain was found in 25-out-of-82 cases (30%) and was more frequent in NGGCTs (12% of germinoma cases and 50% of NGGCT cases), particularly in cases with malignant components, such as immature teratoma, yolk sac tumor, choriocarcinoma, and embryonal carcinoma."
Array-based copy-number analysis with FISH validation directly supports the prevalence and subtype distribution of 12p gain in the cohort.
PMID:34698864 SUPPORT Human Clinical
"The presence of 12p gain correlated with shorter progression-free (PFS) and overall survival (OS) (10-year OS: 59% vs. 94%, with and without 12p gain, respectively, P = 0.0002), even with histology and tumor markers incorporated in the multivariate analysis."
Cohort-level outcome analysis supports prognostic association; it does not prove that 12p gain causes treatment failure.
Neurohypophyseal Dysfunction in Suprasellar Germinoma
Suprasellar or neurohypophyseal germinoma can disrupt posterior-pituitary function, producing central diabetes insipidus that may precede tumor diagnosis for more than a year and often persists after tumor treatment.
neurohypophysis UBERON:0002198 Uberon multi-species anatomy ontology (UBERON) Relation: this pathophysiological event occurs in this anatomical location This pathophysiological event occurs in neurohypophysis (UBERON:0002198). UBERON:0002198 is an anatomical location from the Uberon multi-species anatomy ontology.
Downstream
Central Diabetes Insipidus Posterior-pituitary dysfunction impairs vasopressin regulation and produces central diabetes insipidus.
Show evidence (2 references)
PMID:25266413 SUPPORT Human Clinical
"All had symptoms of DI at presentation with a symptom interval above one year in eight cases (42 %)."
The retrospective pediatric suprasellar/bifocal germinoma cohort directly supports prolonged central diabetes insipidus before diagnosis.
PMID:25266413 SUPPORT Human Clinical
"Desmopressin seldom reverses the clinical features of germinoma-induced DI to allow discontinuation, nor does treatment cause bright spot recovery."
Persistence after tumor treatment is directly reported in this small retrospective suprasellar/bifocal germinoma cohort.
Intracranial Tumor Mass Effect and Cerebrospinal Fluid Obstruction
Pineal, ventricular, and other intracranial CNS germ cell tumors can obstruct cerebrospinal-fluid pathways. This location-dependent effect is distinct from tumor histology and from the still-unresolved initiating mechanism.
Downstream
Hydrocephalus Obstruction of cerebrospinal-fluid flow causes hydrocephalus.
Show evidence (1 reference)
PMID:37452948 SUPPORT Other
"Above 50% of intracranial GCTs (IGCTs) present obstructive hydrocephalus."
The clinical review directly supports obstructive hydrocephalus as a common presentation of intracranial disease.
Tumor-Marker Secretion by CNS Germ Cell Tumor Components
AFP and HCG secretion varies by histologic component. Marker elevation can support clinical classification as NGGCT, but absence does not exclude malignant disease and AFP elevation is not specific to a yolk-sac component.
Show evidence (3 references)
PMID:36995447 SUPPORT Human Clinical
"Tumor markers in CNS germ cell tumors (GCTs) include human chorionic gonadotropin (HCG) and alpha fetoprotein (AFP), which have significant diagnostic implications, as elevation of either one leads to clinical diagnosis of non-germinomatous GCTs without histopathological confirmation, justifying..."
The human cohort establishes the clinical marker-to-classification relationship while the entry preserves histology-specific caveats.
PMID:36995447 SUPPORT Human Clinical
"AFP was often elevated in GCTs without a yolk sac tumor component, especially immature teratoma."
The same cohort directly shows why AFP elevation cannot be treated as a component-specific readout in every CNS germ cell tumor.
PMID:32642701 SUPPORT Human Clinical
"The current biomarkers alpha-fetoprotein (AFP) and human chorionic gonadotropin (HCG) have limited sensitivity/specificity for diagnosing malignant germ cell tumors (GCTs) and "marker-negative" patients require histological confirmation for diagnosis."
Background to the four-case intracranial biomarker series supports the explicit limitation of AFP/HCG-negative classification.
⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.
Pathograph: causal mechanism network for Central Nervous System Germ Cell Tumor Interactive directed graph showing how pathophysiology mechanisms, phenotypes, genetic factors and variants, experimental models, environmental triggers, and treatments relate through causal and linked edges.
●

Phenotypes

6
Endocrine 1
Central Diabetes Insipidus OCCASIONAL HP:0000863 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Central diabetes insipidus (HP:0000863). HP:0000863 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
PMID:39035740 SUPPORT Human Clinical
"Presenting symptoms were headaches (n = 24, 50%), visual disturbances (n = 17, 35.4%), vomiting (n = 12, 25%), nausea (n = 8, 16.7%), and diabetes insipidus (n = 7, 14.6%)."
This 48-patient CNS-GCT cohort directly supports an occasional disease-family frequency band for diabetes insipidus.
PMID:25266413 SUPPORT Human Clinical
"All had symptoms of DI at presentation with a symptom interval above one year in eight cases (42 %)."
The site-selected germinoma cohort supports prolonged diagnostic delay among patients whose suprasellar or bifocal tumors caused DI; it does not supply an umbrella-disease frequency.
Eye 1
Visual Impairment FREQUENT HP:0000505 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Visual impairment (HP:0000505). HP:0000505 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:39035740 SUPPORT Human Clinical
"Presenting symptoms were headaches (n = 24, 50%), visual disturbances (n = 17, 35.4%), vomiting (n = 12, 25%), nausea (n = 8, 16.7%), and diabetes insipidus (n = 7, 14.6%)."
The 35.4% cohort estimate supports a frequent broad visual-impairment phenotype; the source does not resolve a uniform anatomic mechanism.
Metabolism 1
Elevated Circulating Alpha-Fetoprotein Concentration HP:0006254 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Elevated circulating alpha-fetoprotein concentration (HP:0006254). HP:0006254 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:36995447 SUPPORT Human Clinical
"AFP was often elevated in GCTs without a yolk sac tumor component, especially immature teratoma."
Direct serum/CSF marker and histopathology correlation supports diagnostic use while showing that AFP is not component-specific.
Nervous System 1
Hydrocephalus FREQUENT HP:0000238 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Hydrocephalus (HP:0000238). HP:0000238 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:37452948 SUPPORT Other
"Above 50% of intracranial GCTs (IGCTs) present obstructive hydrocephalus."
The review supplies a quantitative intracranial presentation estimate that maps to the frequent band while preserving the anatomic scope.
Other 2
Precocious Puberty VERY_RARE HP:0000826 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Precocious puberty (HP:0000826). HP:0000826 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:39035740 SUPPORT Human Clinical
"Two patients with NGGCT presented with precocious puberty."
Two of 48 cohort patients had this presentation, supporting a very-rare band at the umbrella-disease level without establishing mechanism.
Elevated Circulating Beta Chorionic Gonadotropin Concentration HP:6000485 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Elevated circulating beta chorionic gonadotropin concentration (HP:6000485). HP:6000485 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:36995447 SUPPORT Human Clinical
"HCG was elevated only in cases with a germinoma or choriocarcinoma component, and there existed a clear cut-off HCG value between the two."
Direct marker-to-histopathology correlation supports HCG as a diagnostic readout while preserving component-dependent interpretation.
🧬

Genetic Associations

4
KIT (Recurrent germinoma-enriched somatic mutation)
Gene: KIT hgnc:6342 HUGO Gene Nomenclature Committee (hgnc) Relation: this disease-associated gene is this gene This disease-associated gene is KIT (hgnc:6342). hgnc:6342 is a gene from the HUGO Gene Nomenclature Committee. variant_origin: SOMATIC
Show evidence (1 reference)
PMID:38409885 SUPPORT Human Clinical
"We found that KIT was the most significantly mutated gene (15/47, 32%), which mainly occurred in the germinoma group (13/20, 65%), and less frequently in NGGCT (2/27, 7%)."
Human tumor sequencing establishes recurrence and subtype enrichment, not universal causality.
KRAS (Recurrent somatic KIT-RAS-pathway mutation)
Gene: KRAS hgnc:6407 HUGO Gene Nomenclature Committee (hgnc) Relation: this disease-associated gene is this gene This disease-associated gene is KRAS (hgnc:6407). hgnc:6407 is a gene from the HUGO Gene Nomenclature Committee. variant_origin: SOMATIC
Show evidence (1 reference)
PMID:24896186 SUPPORT Human Clinical
"We find the KIT/RAS signalling pathway frequently mutated in more than 50% of IGCTs, including novel recurrent somatic mutations in KIT, its downstream mediators KRAS and NRAS, and its negative regulator CBL."
Patient-tumor sequencing supports recurrent KRAS mutation within a heterogeneous pathway-level pattern.
NRAS (Recurrent somatic KIT-RAS-pathway mutation)
Gene: NRAS hgnc:7989 HUGO Gene Nomenclature Committee (hgnc) Relation: this disease-associated gene is this gene This disease-associated gene is NRAS (hgnc:7989). hgnc:7989 is a gene from the HUGO Gene Nomenclature Committee. variant_origin: SOMATIC
Show evidence (1 reference)
PMID:24896186 SUPPORT Human Clinical
"We find the KIT/RAS signalling pathway frequently mutated in more than 50% of IGCTs, including novel recurrent somatic mutations in KIT, its downstream mediators KRAS and NRAS, and its negative regulator CBL."
Patient-tumor sequencing supports recurrent NRAS mutation within a heterogeneous pathway-level pattern.
AKT1 (Recurrent somatic copy-number gain with increased expression)
Gene: AKT1 hgnc:391 HUGO Gene Nomenclature Committee (hgnc) Relation: this disease-associated gene is this gene This disease-associated gene is AKT1 (hgnc:391). hgnc:391 is a gene from the HUGO Gene Nomenclature Committee. variant_origin: SOMATIC
Show evidence (1 reference)
PMID:24896186 SUPPORT Human Clinical
"Novel somatic alterations in the AKT/mTOR pathway included copy number gains of the AKT1 locus at 14q32.33 in 19% of patients, with corresponding upregulation of AKT1 expression."
Patient-tumor genomic and expression data establish a recurrent alteration in a minority of cases without defining its temporal role.
💊

Medical Actions

4
Response-Adapted CarboPEI Chemotherapy and Whole-Ventricular Radiation
Action: ChemoradiotherapyNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Chemoradiotherapy (NCIT:C94626). NCIT:C94626 is a clinical intervention from the NCI Thesaurus. NCIT:C94626
For localized germinoma in SIOP-CNS-GCT-II, four carboPEI courses were followed by 24 Gy whole-ventricular radiation, with a 16 Gy tumor boost for residual disease. This subtype- and response-specific protocol should not be generalized to NGGCT or treated as randomized proof of one optimal regimen.
Show evidence (1 reference)
PMID:42234858 SUPPORT Human Clinical
"Patients with localized germinoma received four courses of 'carboPEI' chemotherapy, then 24Gy WVRT if in complete remission (CR), with an additional 16Gy tumor boost if residual disease was present."
The prospective protocol report directly establishes the response-adapted localized-germinoma regimen.
Response-Qualified Chemotherapy and Reduced-Field Radiation for Localized NGGCT
Action: ChemoradiotherapyNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Chemoradiotherapy (NCIT:C94626). NCIT:C94626 is a clinical intervention from the NCI Thesaurus. NCIT:C94626
Agent: carboplatin CHEBI:31355 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses carboplatin (CHEBI:31355). CHEBI:31355 is a therapeutic agent from Chemical Entities of Biological Interest. etoposide CHEBI:4911 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses etoposide (CHEBI:4911). CHEBI:4911 is a therapeutic agent from Chemical Entities of Biological Interest. ifosfamide CHEBI:5864 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses ifosfamide (CHEBI:5864). CHEBI:5864 is a therapeutic agent from Chemical Entities of Biological Interest.
ACNS1123 used alternating carboplatin/etoposide and ifosfamide/etoposide, followed in complete or partial responders by reduced whole-ventricular radiation and a tumor-bed boost. The phase-II result was accompanied by spinal failures, so field adequacy remains an open question rather than a universal standard.
Show evidence (2 references)
PMID:31545689 SUPPORT Human Clinical
"Patients received six cycles of chemotherapy with carboplatin and etoposide alternating with ifosfamide and etoposide, as used in the Children's Oncology Group predecessor study (ACNS0122; ClinicalTrials.gov identifier: NCT00047320)."
The phase-II protocol directly establishes the alternating chemotherapy backbone for localized NGGCT.
PMID:31545689 SUPPORT Human Clinical
"Patients who achieved a complete response (CR) or partial response (PR) with or without second-look surgery were eligible for reduced RT, defined as 30.6 Gy whole ventricular field and 54 Gy tumor-bed boost, compared with 36 Gy craniospinal irradiation plus 54 Gy tumor-bed boost used in ACNS0122."
The trial defines the response qualification and reduced radiation field; it does not settle whether that field is adequate for all localized NGGCT.
Myeloablative Chemotherapy with Autologous Stem Cell Transplantation
Action: hematopoietic stem cell transplantationNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is hematopoietic stem cell transplantation, annotated with Hematopoietic Cell Transplantation (NCIT:C15431). NCIT:C15431 is a clinical intervention from the NCI Thesaurus. Ontology label: Hematopoietic Cell Transplantation NCIT:C15431
High-dose chemotherapy with autologous stem-cell rescue is a selected salvage option for relapsed or progressed CNS germ cell tumors. Evidence comes from a small prospective study and does not establish a subtype-independent standard.
Show evidence (2 references)
PMID:23824533 SUPPORT Human Clinical
"Eleven patients with non-germinomatous germ cell tumors and nine patients with germinomas were enrolled."
The 20-patient sample bounds the generalizability of this salvage strategy.
PMID:23824533 SUPPORT Human Clinical
"In conclusion, HDCT/autoSCT was feasible, and survival rates were encouraging. Further study with a larger cohort of patients is needed to elucidate the role of HDCT/autoSCT in the treatment of relapsed or progressed CNS-GCTs."
The investigators explicitly frame feasibility as encouraging but insufficient to define the treatment's role.
Resection of Persistent or Growing Teratomatous Disease
Action: surgical procedure on nervous systemNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is surgical procedure on nervous system, annotated with Neurosurgical Procedure (NCIT:C15656). NCIT:C15656 is a clinical intervention from the NCI Thesaurus. Ontology label: Neurosurgical Procedure NCIT:C15656
Surgical resection can define and remove enlarging teratomatous tissue after chemotherapy, including growing teratoma syndrome. Case-level evidence does not establish one universal timing or extent of resection.
Show evidence (1 reference)
PMID:42488730 SUPPORT Human Clinical
"Subsequent surgical resection revealed a mature teratoma without malignant elements, consistent with growing teratoma syndrome (GTS)."
This single case demonstrates resection in a post-chemotherapy pineal GTS context without establishing a general treatment rule.
🔬

Biochemical Markers

4
Alpha-Fetoprotein (Variable elevation in serum or cerebrospinal fluid)
Show evidence (1 reference)
PMID:36995447 SUPPORT Human Clinical
"AFP was often elevated in GCTs without a yolk sac tumor component, especially immature teratoma."
Direct marker-to-histopathology correlation supports AFP as an analyte while bounding component-specific interpretation.
Human Chorionic Gonadotropin (Variable elevation in serum or cerebrospinal fluid)
Show evidence (1 reference)
PMID:36995447 SUPPORT Human Clinical
"HCG was elevated only in cases with a germinoma or choriocarcinoma component, and there existed a clear cut-off HCG value between the two."
Direct marker-to-histopathology correlation supports HCG as an analyte with component-dependent interpretation.
Circulating miR-371a-3p
Show evidence (1 reference)
PMID:32642701 SUPPORT Human Clinical
"Serum and/or CSF miR-371a-3p levels distinguished those with intracranial malignant GCTs from LCH and, if known in real time, could have helped clinical management."
The small clinical series is a direct intracranial signal but remains hypothesis-generating rather than a validated diagnostic test.
Cerebrospinal Fluid Circulating Tumor DNA
Show evidence (1 reference)
PMID:39189644 SUPPORT Human Clinical
"Among them, ctDNA before chemotherapy as well as ctDNA clearance were consistently associated with PFS and OS (p < .05)."
The retrospective association supports further evaluation of CSF ctDNA for monitoring but does not establish a causal resistance mechanism.
🔬

Diagnosis

2
Serum and Cerebrospinal Fluid Tumor-Marker Assessment
Measure AFP and HCG in both serum and CSF to support CNS germ cell tumor classification. Marker elevation can permit a clinical NGGCT diagnosis in some settings, while marker-negative malignant tumors may require tissue confirmation.
Markers: AFP, HCG
Show evidence (2 references)
PMID:36995447 SUPPORT Human Clinical
"HCG was elevated only in CSF in 3-of-52 cases, and AFP was elevated only in serum in 7-of-49 cases, emphasizing the potential utilization of both serum and CSF studies."
Direct paired-compartment observations support testing both serum and CSF.
PMID:32642701 SUPPORT Human Clinical
"The current biomarkers alpha-fetoprotein (AFP) and human chorionic gonadotropin (HCG) have limited sensitivity/specificity for diagnosing malignant germ cell tumors (GCTs) and "marker-negative" patients require histological confirmation for diagnosis."
Background to the proof-of-concept series bounds marker-negative interpretation and the need for histology.
Neuraxis Staging with MRI and Cerebrospinal Fluid Cytology
Before treatment, contrast-enhanced MRI of brain and whole spine plus CSF cytology evaluates macroscopic and cytologic dissemination. This is staging, not a substitute for histology when tumor markers are non-diagnostic.
Show evidence (1 reference)
PMID:37452948 SUPPORT Other
"Staging work-up includes CSF cytology for tumor cells and contrast-enhanced MRI of brain and spine for macroscopic metastasis before treatment commences."
The clinical review directly describes pre-treatment neuraxis staging.
📊

Prevalence

1
Children aged 0-14 years in Kumamoto Prefecture, Japan
Annual Incidence 0.45 per 100,000 1–9 per 1,000,000
Age-adjusted annual incidence over a 23-year regional survey; this regional estimate should not be treated as a global rate.
Show evidence (1 reference)
PMID:24751890 SUPPORT Human Clinical
"The age-adjusted annual incidence rate was 0.45 cases (boys: 0.64, girls: 0.28) per 10(5) children."
The population-based regional survey directly supplies an annual pediatric incidence estimate and sex-stratified rates for Kumamoto.
{ }

Source YAML

click to show 
name: Central Nervous System Germ Cell Tumor
creation_date: "2026-07-26T08:57:31Z"
synonyms:
- CNS germ cell tumor
description: >-
  Central nervous system germ cell tumors are rare, biologically heterogeneous
  tumors arising mainly in children and adolescents. The family includes
  germinoma and nongerminomatous germ cell tumors (NGGCTs), the latter
  comprising teratoma, embryonal carcinoma, yolk sac tumor, choriocarcinoma, and
  mixed histologies. Cross-sectional tumor profiling supports distinct
  developmental and molecular states across these groups, but does not by
  itself establish a single cell of origin, one universal MAPK/PI3K causal
  chain, or a molecular mechanism of refractory NGGCT. The MONDO disease term
  covers the CNS umbrella. Most directly applicable tumor and treatment evidence
  curated here is intracranial; explicitly labeled non-tumor developmental and
  extracranial germ-cell evidence remains indirect. None should be generalized
  to rare primary spinal CNS germ cell tumors without site-specific evidence.
categories:
- Central Nervous System Neoplasm
- Germ Cell Neoplasm
- Pediatric Brain Tumor
parents:
- central nervous system neoplasm
- extragonadal germ cell tumor
disease_term:
  preferred_term: central nervous system germ cell tumor
  term:
    id: MONDO:0003000
    label: central nervous system germ cell tumor
classifications:
  harrisons_chapter:
  - classification_value: ONCOLOGY_HEMATOLOGY
definitions:
- name: CNS germ cell tumor literature scope
  definition_type: OTHER
  scope: >-
    MONDO-level CNS disease-family scope encompassing germinoma and
    nongerminomatous germ cell tumors. Core disease evidence is intracranial;
    non-tumor developmental and extracranial comparator studies are explicitly
    indirect.
  description: >-
    A CNS germ cell tumor family that primarily affects children and
    adolescents and is divided clinically into germinoma and NGGCT; individual
    components range from mature teratoma to malignant histologies.
  evidence:
  - reference: PMID:42419530
    reference_title: Advances in diagnosis and treatment of intracranial germ cell tumors and future directions.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Intracranial germ cell tumors (IGCTs) are a group of malignant central
      nervous system (CNS) tumors primarily affecting children and adolescents,
      whose pathogenesis remains incompletely understood.
    explanation: >-
      The seed review supports the broad CNS tumor-family scope, typical age
      group, and unresolved pathogenesis, but its malignant-tumor wording does
      not capture mature teratoma within the broader ontology class.
prevalence:
- population: Children aged 0-14 years in Kumamoto Prefecture, Japan
  measure_type: ANNUAL_INCIDENCE
  prevalence_class: BAND_1_9_PER_1000000
  rate_per_100000: 0.45
  notes: >-
    Age-adjusted annual incidence over a 23-year regional survey; this regional
    estimate should not be treated as a global rate.
  evidence:
  - reference: PMID:24751890
    reference_title: "Incidence of primary central nervous system germ cell tumors in childhood: a regional survey in Kumamoto prefecture in southern Japan."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The age-adjusted annual incidence rate was 0.45 cases (boys: 0.64, girls:
      0.28) per 10(5) children.
    explanation: >-
      The population-based regional survey directly supplies an annual
      pediatric incidence estimate and sex-stratified rates for Kumamoto.
has_subtypes:
- name: Central Nervous System Germinoma
  display_name: CNS germinoma
  subtype_term:
    preferred_term: central nervous system germinoma
    term:
      id: MONDO:0002999
      label: central nervous system germinoma
  description: >-
    The germinoma arm of the CNS germ cell tumor family. Human tumor
    methylation and transcriptome studies find a primordial-germ-cell-like,
    globally hypomethylated state, but resemblance to a developmental reference
    state is not direct lineage evidence.
  evidence:
  - reference: PMID:28078450
    reference_title: Genome-wide methylation profiles in primary intracranial germ cell tumors indicate a primordial germ cell origin for germinomas.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      We showed that pure germinomas are characterized by global low DNA
      methylation, a unique epigenetic feature making them distinct from all
      other iGCTs subtypes.
    explanation: >-
      Genome-wide methylation profiling of human tumors supports a distinct
      germinoma epigenetic state without proving developmental ancestry.
- name: Central Nervous System Nongerminomatous Germ Cell Tumor
  display_name: CNS NGGCT
  subtype_term:
    preferred_term: central nervous system nongerminomatous germ cell tumor
    term:
      id: MONDO:0020574
      label: central nervous system nongerminomatous germ cell tumor
  description: >-
    The non-germinoma arm, encompassing several pure and mixed histologies with
    different differentiation states, biomarkers, treatment responses, and
    relapse patterns. This umbrella should not be represented as though every
    component shares one resistance mechanism.
  evidence:
  - reference: PMID:35137206
    reference_title: Transcriptome and methylome analysis of CNS germ cell tumor finds its cell-of-origin in embryogenesis and reveals shared similarities with testicular counterparts.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Genome-wide transcriptome analysis in CNS GCTs indicated that germinoma
      had a transcriptomic profile representative of primitive cells during
      early embryogenesis with high meiosis/mitosis potentials, while
      nongerminomatous GCTs (NGGCTs) had differentiated phenotypes oriented
      toward tissue formation and organogenesis.
    explanation: >-
      Integrative human tumor profiling supports molecular and developmental
      state differences between germinoma and NGGCT.
- name: Central Nervous System Embryonal Carcinoma
  display_name: CNS embryonal carcinoma
  subtype_term:
    preferred_term: embryonal carcinoma of the central nervous system
    term:
      id: MONDO:0018843
      label: embryonal carcinoma of the central nervous system
  description: >-
    The embryonal carcinoma histology within the CNS NGGCT spectrum, previously
    named only in this entry's prose. It is the least differentiated
    non-germinomatous component, and its disease-level mechanism is curated in
    the Embryonal Carcinoma entry rather than duplicated here.
  evidence:
  - reference: PMID:42115464
    reference_title: "Germinomas and Non-germinomatous Germ Cell Tumors."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      NGGCTs include embryonal carcinoma, yolk sac tumor, choriocarcinoma, and
      teratomas and often secrete tumor markers like alpha-fetoprotein (AFP) and
      beta-hCG, which guide diagnosis and treatment without biopsy.
    explanation: >-
      Names embryonal carcinoma as one of the intracranial NGGCT histologies,
      supporting it as a subtype of this entry.
- name: Central Nervous System Teratoma
  display_name: CNS teratoma
  subtype_term:
    preferred_term: central nervous system teratoma
    term:
      id: MONDO:0002718
      label: central nervous system teratoma
  description: >-
    A CNS-specific teratomatous subtype within the NGGCT spectrum. Mature,
    immature, growing-teratoma-syndrome, and somatic-type-malignancy contexts
    require separate histopathologic interpretation; this subtype does not imply
    that every CNS teratoma is malignant.
  evidence:
  - reference: PMID:37452948
    reference_title: Brain and Spinal Tumors Originating from the Germ Line Cells.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      In the WHO CNS tumor classification, GCTS are classified as germinoma and
      nongerminomatous GCT (NGGCT) with different subtypes (including teratoma).
    explanation: >-
      The clinical review places teratoma within the CNS NGGCT classification
      while separately noting that mature teratoma is not malignant.
pathophysiology:
- name: Recurrent KIT-RAS Signaling Alterations
  description: >-
    Somatic alterations in KIT and other KIT/RAS-pathway genes recur in human
    intracranial germ cell tumors. They are enriched in germinoma but are not
    present in every tumor. Cross-sectional sequencing establishes recurrence,
    not whether these lesions initiate disease or are required to maintain it.
  genes:
  - preferred_term: KIT
    term:
      id: hgnc:6342
      label: KIT
  - preferred_term: KRAS
    term:
      id: hgnc:6407
      label: KRAS
  - preferred_term: NRAS
    term:
      id: hgnc:7989
      label: NRAS
  evidence:
  - reference: PMID:24896186
    reference_title: Novel somatic and germline mutations in intracranial germ cell tumours.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      We find the KIT/RAS signalling pathway frequently mutated in more than
      50% of IGCTs, including novel recurrent somatic mutations in KIT, its
      downstream mediators KRAS and NRAS, and its negative regulator CBL.
    explanation: >-
      Sequencing of patient tumors establishes recurrent somatic pathway
      alterations, while neither temporal order nor causal necessity was tested.
  - reference: PMID:38409885
    reference_title: Whole-exome sequencing has revealed novel genetic characteristics in intracranial germ cell tumours in the Chinese.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      We found that KIT was the most significantly mutated gene (15/47, 32%),
      which mainly occurred in the germinoma group (13/20, 65%), and less
      frequently in NGGCT (2/27, 7%).
    explanation: >-
      A second human tumor cohort supports subtype enrichment and also shows
      that KIT mutation is neither universal nor confined absolutely to
      germinoma.
- name: Recurrent AKT-mTOR Signaling Alterations
  description: >-
    A subset of intracranial germ cell tumors carries AKT/mTOR-pathway lesions,
    including focal AKT1 copy-number gain with increased AKT1 expression.
    Recurrence supports a tumor molecular state but does not establish when the
    alteration arose or whether it is sufficient for transformation.
  genes:
  - preferred_term: AKT1
    term:
      id: hgnc:391
      label: AKT1
  evidence:
  - reference: PMID:24896186
    reference_title: Novel somatic and germline mutations in intracranial germ cell tumours.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Novel somatic alterations in the AKT/mTOR pathway included copy number
      gains of the AKT1 locus at 14q32.33 in 19% of patients, with corresponding
      upregulation of AKT1 expression.
    explanation: >-
      Patient-tumor genomic and expression measurements establish a recurrent
      AKT1-associated state in a minority of cases, not a universal initiating
      mechanism.
- name: Germinoma Global DNA Hypomethylation
  subtypes:
  - Central Nervous System Germinoma
  description: >-
    Pure germinomas show a globally hypomethylated tumor state distinct from
    other intracranial germ cell tumor subtypes. Its resemblance to migrating
    primordial germ cells is relevant to developmental-origin hypotheses but is
    not lineage tracing.
  evidence:
  - reference: PMID:28078450
    reference_title: Genome-wide methylation profiles in primary intracranial germ cell tumors indicate a primordial germ cell origin for germinomas.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      We showed that pure germinomas are characterized by global low DNA
      methylation, a unique epigenetic feature making them distinct from all
      other iGCTs subtypes.
    explanation: >-
      Genome-wide profiling of patient tumors directly supports the
      subtype-specific methylation state without establishing its causal timing.
- name: Chromosome 12p Gain
  subtypes:
  - Central Nervous System Germinoma
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  description: >-
    Additional 12p copy number occurs across CNS germ cell tumors but is more
    frequent in NGGCT and in tumors with malignant components. It is a
    prognostic genomic feature, not a demonstrated universal driver.
  evidence:
  - reference: PMID:34698864
    reference_title: 12p gain is predominantly observed in non-germinomatous germ cell tumors and identifies an unfavorable subgroup of central nervous system germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      A 12p gain was found in 25-out-of-82 cases (30%) and was more frequent in
      NGGCTs (12% of germinoma cases and 50% of NGGCT cases), particularly in
      cases with malignant components, such as immature teratoma, yolk sac
      tumor, choriocarcinoma, and embryonal carcinoma.
    explanation: >-
      Array-based copy-number analysis with FISH validation directly supports
      the prevalence and subtype distribution of 12p gain in the cohort.
  - reference: PMID:34698864
    reference_title: 12p gain is predominantly observed in non-germinomatous germ cell tumors and identifies an unfavorable subgroup of central nervous system germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The presence of 12p gain correlated with shorter progression-free (PFS)
      and overall survival (OS) (10-year OS: 59% vs. 94%, with and without 12p
      gain, respectively, P = 0.0002), even with histology and tumor markers
      incorporated in the multivariate analysis.
    explanation: >-
      Cohort-level outcome analysis supports prognostic association; it does not
      prove that 12p gain causes treatment failure.
- name: Neurohypophyseal Dysfunction in Suprasellar Germinoma
  subtypes:
  - Central Nervous System Germinoma
  description: >-
    Suprasellar or neurohypophyseal germinoma can disrupt posterior-pituitary
    function, producing central diabetes insipidus that may precede tumor
    diagnosis for more than a year and often persists after tumor treatment.
  locations:
  - preferred_term: neurohypophysis
    term:
      id: UBERON:0002198
      label: neurohypophysis
  downstream:
  - target: Central Diabetes Insipidus
    description: >-
      Posterior-pituitary dysfunction impairs vasopressin regulation and
      produces central diabetes insipidus.
    causal_link_type: INDIRECT_KNOWN_INTERMEDIATES
    intermediate_mechanisms:
    - Disruption of posterior-pituitary vasopressin regulation.
  evidence:
  - reference: PMID:25266413
    reference_title: "Diabetes insipidus in pediatric germinomas of the suprasellar region: characteristic features and significance of the pituitary bright spot."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      All had symptoms of DI at presentation with a symptom interval above one
      year in eight cases (42 %).
    explanation: >-
      The retrospective pediatric suprasellar/bifocal germinoma cohort directly
      supports prolonged central diabetes insipidus before diagnosis.
  - reference: PMID:25266413
    reference_title: "Diabetes insipidus in pediatric germinomas of the suprasellar region: characteristic features and significance of the pituitary bright spot."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Desmopressin seldom reverses the clinical features of germinoma-induced DI
      to allow discontinuation, nor does treatment cause bright spot recovery.
    explanation: >-
      Persistence after tumor treatment is directly reported in this small
      retrospective suprasellar/bifocal germinoma cohort.
- name: Intracranial Tumor Mass Effect and Cerebrospinal Fluid Obstruction
  description: >-
    Pineal, ventricular, and other intracranial CNS germ cell tumors can obstruct
    cerebrospinal-fluid pathways. This location-dependent effect is distinct
    from tumor histology and from the still-unresolved initiating mechanism.
  downstream:
  - target: Hydrocephalus
    description: Obstruction of cerebrospinal-fluid flow causes hydrocephalus.
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:37452948
    reference_title: Brain and Spinal Tumors Originating from the Germ Line Cells.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Above 50% of intracranial GCTs (IGCTs) present obstructive hydrocephalus.
    explanation: >-
      The clinical review directly supports obstructive hydrocephalus as a common
      presentation of intracranial disease.
- name: Tumor-Marker Secretion by CNS Germ Cell Tumor Components
  description: >-
    AFP and HCG secretion varies by histologic component. Marker elevation can
    support clinical classification as NGGCT, but absence does not exclude
    malignant disease and AFP elevation is not specific to a yolk-sac component.
  evidence:
  - reference: PMID:36995447
    reference_title: "Impact of tumor markers on diagnosis, treatment and prognosis in CNS germ cell tumors: correlations with clinical practice and histopathology."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Tumor markers in CNS germ cell tumors (GCTs) include human chorionic
      gonadotropin (HCG) and alpha fetoprotein (AFP), which have significant
      diagnostic implications, as elevation of either one leads to clinical
      diagnosis of non-germinomatous GCTs without histopathological confirmation,
      justifying intensified chemotherapy and irradiation.
    explanation: >-
      The human cohort establishes the clinical marker-to-classification
      relationship while the entry preserves histology-specific caveats.
  - reference: PMID:36995447
    reference_title: "Impact of tumor markers on diagnosis, treatment and prognosis in CNS germ cell tumors: correlations with clinical practice and histopathology."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      AFP was often elevated in GCTs without a yolk sac tumor component,
      especially immature teratoma.
    explanation: >-
      The same cohort directly shows why AFP elevation cannot be treated as a
      component-specific readout in every CNS germ cell tumor.
  - reference: PMID:32642701
    reference_title: Clinical utility of circulating miR-371a-3p for the management of patients with intracranial malignant germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The current biomarkers alpha-fetoprotein (AFP) and human chorionic
      gonadotropin (HCG) have limited sensitivity/specificity for diagnosing
      malignant germ cell tumors (GCTs) and "marker-negative" patients require
      histological confirmation for diagnosis.
    explanation: >-
      Background to the four-case intracranial biomarker series supports the
      explicit limitation of AFP/HCG-negative classification.
mechanistic_hypotheses:
- hypothesis_group_id: common_ancestry_developmental_state_divergence
  hypothesis_label: Common-Ancestry and Developmental-State Divergence Model
  status: EMERGING
  applies_to_subtypes:
  - Central Nervous System Germinoma
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  description: >-
    Histologically distinct components within an individual mixed CNS germ cell
    tumor may descend from a common tumor ancestor and subsequently diverge
    into a primordial-germ-cell-like, hypomethylated germinoma state or one of
    several more differentiated NGGCT states. Cross-sectional resemblance does
    not identify that ancestor: a mis-migrated primordial germ cell and an
    endogenous neural stem or progenitor cell are competing cell-of-origin
    models to be tested rather than assumed. MAPK/PI3K-pathway alterations may
    cooperate with developmental state and can be shared by distinct components
    of a mixed tumor, but their temporal position and necessity for initiation
    remain unresolved.
  evidence:
  - reference: PMID:28078450
    reference_title: Genome-wide methylation profiles in primary intracranial germ cell tumors indicate a primordial germ cell origin for germinomas.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The patterns of methylation strongly resemble that of primordial germ
      cells (PGC) at the migration phase, possibly indicating the cell of origin
      for these tumors.
    explanation: >-
      The primary methylation study supports PGC-like resemblance while its
      explicitly tentative wording preserves the absence of lineage tracing.
  - reference: PMID:28078450
    reference_title: Genome-wide methylation profiles in primary intracranial germ cell tumors indicate a primordial germ cell origin for germinomas.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Unlike PGC, however, hypomethylation extends to long interspersed nuclear
      element retrotransposons.
    explanation: >-
      The human-tumor methylome is PGC-like but not an exact PGC epigenome, so
      global resemblance cannot by itself identify the normal precursor.
  - reference: PMID:20582452
    reference_title: Variable methylation of the imprinted gene, SNRPN, supports a relationship between intracranial germ cell tumours and neural stem cells.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: >-
      We show here that the lack of methylation of SNRPN that has previously
      been taken to indicate an origin for GCTs from PGCs is also seen in
      neural stem cells of mice and humans.
    explanation: >-
      A mouse-and-human neural-stem-cell assay shows that one imprinting feature
      used to support PGC origin is not PGC-specific. It does not establish a
      neural-stem-cell lineage for the tumor.
  - reference: PMID:41190468
    reference_title: "Primordial Germ Cells in the Human Fetal Brain: Origins of Intracranial Germ Cell Tumours."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The distribution seems to favour the hypothesis that PGCs may be derived
      from germinal matrix neural stem cells rather than necessarily from
      extracranial midline migration to the brain.
    explanation: >-
      This is the authors' distribution-based inference from two non-tumor
      fetal autopsies. The observation establishes neither local
      neural-stem-cell derivation nor tumor ancestry, but it keeps that origin
      alternative open.
  - reference: PMID:28078450
    reference_title: Genome-wide methylation profiles in primary intracranial germ cell tumors indicate a primordial germ cell origin for germinomas.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Histologically and epigenetically distinct microdissected components of
      mixed-GCTs shared identical somatic mutations in the MAPK or PI3K
      pathways, indicating that they developed from a common ancestral cell.
    explanation: >-
      Shared mutations across microdissected mixed components support common
      ancestry and component-specific epigenetic divergence. Their relative
      timing is inferred rather than directly observed, and the data neither
      identify the normal precursor nor establish initiation by the pathway
      lesion.
  - reference: PMID:34698864
    reference_title: 12p gain is predominantly observed in non-germinomatous germ cell tumors and identifies an unfavorable subgroup of central nervous system germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The 12p copy number status was shared among histological components in
      mixed GCTs.
    explanation: >-
      FISH-validated human-tumor copy-number analysis supplies an orthogonal
      component-shared feature in sampled mixed tumors. Shared status does not
      identify the normal precursor, establish temporal order, or prove
      universal clonality across mixed tumors.
  - reference: PMID:35137206
    reference_title: Transcriptome and methylome analysis of CNS germ cell tumor finds its cell-of-origin in embryogenesis and reveals shared similarities with testicular counterparts.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Co-analysis with the transcriptome of human embryonic cells revealed that
      germinomas had expression profiles similar to those of primordial germ
      cells, while the expression profiles of NGGCTs were similar to those of
      embryonic stem cells.
    explanation: >-
      Cross-reference transcriptome analysis supports distinct developmental
      state resemblance, not direct observation of tumor initiation.
  - reference: PMID:24896186
    reference_title: Novel somatic and germline mutations in intracranial germ cell tumours.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Overall, 53% of the tumors harbored somatic mutations in at least one of
      the genes involved in KIT/RAS or AKT/mTOR pathways
    explanation: >-
      Sequencing of 62 human tumors establishes recurrent pathway alterations,
      but prevalence is incomplete, subtype distribution is unequal, and the
      cross-sectional design does not establish developmental timing.
  notes: >-
    This is a within-tumor developmental model inferred mainly from
    cross-sectional human profiles. It does not posit an ancestor shared across
    patients, discriminate a mis-migrated primordial germ cell from an
    endogenous neural stem or progenitor cell or another precursor, or
    generalize germinoma hypomethylation to every NGGCT component.
- hypothesis_group_id: nggct_failure_state_specific_persistence_and_evolution
  hypothesis_label: NGGCT Failure-State-Specific Persistence and Evolution Model
  status: EMERGING
  applies_to_subtypes:
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  description: >-
    CNS-NGGCT failures are clinically partitioned into induction progression,
    marker-negative growing teratoma syndrome (GTS), viable residual or
    progressive disease, and post-treatment relapse. Clinical separability does
    not establish molecular separability. GTS may reflect persistence or
    expansion of a pre-existing teratomatous lineage or treatment-associated
    maturation and must be distinguished from viable malignant disease; it is
    not uniformly benign. Malignant progression or relapse may reflect selection
    of a pre-existing malignant clone or state, acquired genomic or epigenetic
    changes, reversible treatment-tolerant adaptation, treatment exposure or
    geography, or sampling. These alternatives must not be pooled or assumed to
    be therapy-selected without paired, component-resolved longitudinal
    evidence. No adequately powered diagnosis-to-failure molecular series
    currently establishes which route is causal.
  evidence:
  - reference: PMID:35218656
    reference_title: "Pattern of treatment failures in patients with central nervous system non-germinomatous germ cell tumors (CNS-NGGCT): A pooled analysis of clinical trials."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      A total of 118 patients experienced a treatment failure. Twenty-four
      patients had progressive disease during therapy, and additional 11
      patients were diagnosed with growing teratoma syndrome (GTS).
    explanation: >-
      The pooled clinical dataset operationally separates progression during
      therapy from GTS, but it does not demonstrate molecularly distinct causes.
  - reference: PMID:35218656
    reference_title: "Pattern of treatment failures in patients with central nervous system non-germinomatous germ cell tumors (CNS-NGGCT): A pooled analysis of clinical trials."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Eighty-three individuals experienced disease relapses after treatment
      ended.
    explanation: >-
      The same pooled cohort separately documents post-treatment relapse,
      supporting a timing-based clinical partition from progression during
      therapy and GTS rather than a proven mechanistic partition.
  - reference: PMID:42488730
    reference_title: Growing Teratoma Syndrome Arising from a Pineal Mixed Germ Cell Tumor in a 12-Year-Old Boy.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Subsequent surgical resection revealed a mature teratoma without malignant
      elements, consistent with growing teratoma syndrome (GTS).
    explanation: >-
      This single post-chemotherapy resection supports a teratomatous GTS branch
      in one pineal mixed tumor. It does not distinguish treatment-associated
      maturation from persistence or expansion of a pre-existing component.
  - reference: PMID:38668041
    reference_title: Successful Multimodal Treatment of Intracranial Growing Teratoma Syndrome with Malignant Features.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      A radical resection via the occipital transtentorial approach was
      performed, and histopathological examination revealed a teratoma with
      malignant features. Methylation classifier analysis confirmed the
      diagnosis of teratoma, and DMRT1 loss and 12p gain were identified by copy
      number variation analysis, potentially elucidating the cause of growth and
      malignant transformation of the teratoma.
    explanation: >-
      This single intracranial GTS case prevents treating the teratomatous branch
      as uniformly inert or benign. The alterations are hypothesis-generating
      observations, not established causes of growth, transformation, or
      treatment failure.
  - reference: PMID:41675560
    reference_title: Pre-treatment journey and outcome for children with intracranial non-germinomatous germ cell tumors-the Shanghai experience.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Whole-exome sequencing was performed on 12 paired tumor-blood samples to
      characterize molecular alterations.
    explanation: >-
      The recent NGGCT cohort adds diagnosis-time molecular observations, but
      paired tumor-blood samples are not paired diagnosis-relapse tumors and
      therefore cannot identify acquired resistance.
  - reference: PMID:38430549
    reference_title: Novel molecular subtypes of intracranial germ cell tumors expand therapeutic opportunities.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Three distinct subtypes associated with unique genomic and clinical
      profiles were identified with transcriptome analysis: Immune-hot,
      MYC/E2F, and SHH.
    explanation: >-
      Molecular-state heterogeneity supplies candidates for future resistance
      testing but does not show that a state is selected by therapy or causes
      relapse.
  - reference: PMID:38409885
    reference_title: Whole-exome sequencing has revealed novel genetic characteristics in intracranial germ cell tumours in the Chinese.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Clonal evolution analysis revealed an early branched evolutionary pattern
      in two IGCT patients who underwent changes in the histological subtype or
      degree of differentiation during disease surveillance.
    explanation: >-
      Two longitudinal human cases support clonal continuity and state
      selection as a possibility, but the sample is too small and
      histology-changing surveillance is not equivalent to a replicated
      refractory-NGGCT resistance mechanism.
  - reference: PMID:11005262
    reference_title: Comparative genomic hybridization in pineal germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Fifteen primary pineal germ cell tumors (8 germinomas, 4 mixed
      teratomas-germinomas, 2 immature teratomas, and 1 yolk sac tumor) and 2
      recurrences of the yolk sac tumor were studied by comparative genomic
      hybridization (CGH).
    explanation: >-
      The cohort contains one primary yolk-sac tumor and its two recurrences,
      establishing serial genomic material but not a treatment-resistance
      mechanism.
  - reference: PMID:11005262
    reference_title: Comparative genomic hybridization in pineal germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      the first recurrence showed 7 (4 gains vs 3 losses), the second 13
      imbalances (8 gains vs 5 losses).
    explanation: >-
      Copy-number complexity increased between the two recurrences. This is
      direct serial progression evidence, but the small observation lacks
      reported regimen or exposure details, gene-level resolution, a
      comparator, and functional validation, so it cannot identify a resistance
      driver.
  notes: >-
    This model operationally separates GTS, viable malignant progression, and
    post-treatment relapse without claiming that each category has a distinct
    molecular cause. It does not promote SHH, MYC/E2F, immune-hot, MAPK, or PI3K
    status to a clinical resistance biomarker without longitudinal validation.
- hypothesis_group_id: mir214_bcl2l11_cisplatin_response_candidate
  hypothesis_label: miR-214-3p–BCL2L11 Cisplatin-Response Candidate
  status: EMERGING
  applies_to_subtypes:
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  description: >-
    In a subset of malignant NGGCT, altered regulation of the miR-199/214
    cluster may increase miR-214-3p, reduce the pro-apoptotic protein
    BCL2L11/BIM, and shift apoptosis and survival after cisplatin exposure. This
    is a candidate response-modifying chain: the current causal evidence comes
    from forced miR-214-3p expression and a BCL2L11-knockdown phenocopy in one
    adult mediastinal mixed-germ-cell-tumor line. It does not establish direct
    miR-214-3p binding to BCL2L11, endogenous necessity, rescue, or a difference
    between resistant and sensitive CNS NGGCT in patients.
  evidence:
  - reference: PMID:29036598
    reference_title: Global DNA methylation analysis reveals miR-214-3p contributes to cisplatin resistance in pediatric intracranial nongerminomatous malignant germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The expression levels of 97 genes and 8 miRNAs were correlated with
      promoter DNA methylation and hydroxymethylation status, such as the
      miR-199/-214 cluster
    explanation: >-
      Human tumor multi-omic analysis links the miR-199/214 cluster to
      methylation state in a cross-sectional NGMGCT-versus-germinoma subtype
      comparison. Histology and differentiation state therefore confound any
      resistance inference, and no longitudinal resistant-versus-sensitive
      comparison was performed.
  - reference: PMID:29036598
    reference_title: Global DNA methylation analysis reveals miR-214-3p contributes to cisplatin resistance in pediatric intracranial nongerminomatous malignant germ cell tumors.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: >-
      Overexpresssion of miR-214-3p in NCCIT cells leads to reduced expression
      of the pro-apoptotic protein BCL2-like 11 and induces cisplatin
      resistance.
    explanation: >-
      Stable miR-214-3p expression increased NCCIT survival after cisplatin but
      the authors reported no etoposide difference (data not shown), and
      BCL2L11 knockdown phenocopied cisplatin survival. This supports a
      single-cell-line in-vitro response phenotype, but the study did not test
      endogenous miR-214-3p loss, AGO2 or seed-mutant binding, or
      miRNA-insensitive BCL2L11 rescue.
  - reference: PMID:36768818
    reference_title: "Integrated Microarray-Based Data Analysis of miRNA Expression Profiles: Identification of Novel Biomarkers of Cisplatin-Resistance in Testicular Germ Cell Tumours."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: >-
      Comparing CDDP-resistant vs. -sensitive TGCT cell lines, we identified
      miR-218-5p, miR-31-5p, miR-125b-5p, miR-27b-3p, miR-199a-5p, miR-214-3p,
      let-7a and miR-517a-3p as significantly up-regulated
    explanation: >-
      This extracranial testicular-GCT screen supplies an independent but
      nonfunctional association. RT-qPCR miR-214-3p direction varied by
      resistant-versus-sensitive cell-line comparison; miR-214-3p was not among
      the authors' six-miRNA proposed core biomarker panel, although they
      suggested possible added predictive value. The study did not test
      BCL2L11 or perturb miR-214-3p.
  notes: >-
    This is a cisplatin-response candidate, not a validated intrinsic-tolerance
    mechanism or clinical resistance biomarker. No independent functional
    replication in a patient-derived intracranial model or
    resistant-versus-sensitive clinical comparison was retrieved. NCCIT is not
    a patient-derived CNS NGGCT model; testicular or mediastinal GCT apoptosis
    and DNA-repair studies and non-germ-cell miR-214 targets remain indirect
    alternatives, not direct CNS evidence or refutation.
phenotypes:
- category: Endocrine
  name: Central Diabetes Insipidus
  frequency: OCCASIONAL
  description: >-
    Central diabetes insipidus is a location-dependent presentation of
    suprasellar or bifocal germinoma and may precede diagnosis for a prolonged
    interval. The frequency band is based on one broader regional CNS-GCT cohort
    and should not be generalized to every site or histology.
  phenotype_term:
    preferred_term: Central diabetes insipidus
    term:
      id: HP:0000863
      label: Central diabetes insipidus
  evidence:
  - reference: PMID:39035740
    reference_title: "Primary central nervous system germ cell tumors in Central America and the Caribbean Region: an AHOPCA 20-year experience."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Presenting symptoms were headaches (n = 24, 50%), visual disturbances (n =
      17, 35.4%), vomiting (n = 12, 25%), nausea (n = 8, 16.7%), and diabetes
      insipidus (n = 7, 14.6%).
    explanation: >-
      This 48-patient CNS-GCT cohort directly supports an occasional
      disease-family frequency band for diabetes insipidus.
  - reference: PMID:25266413
    reference_title: "Diabetes insipidus in pediatric germinomas of the suprasellar region: characteristic features and significance of the pituitary bright spot."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      All had symptoms of DI at presentation with a symptom interval above one
      year in eight cases (42 %).
    explanation: >-
      The site-selected germinoma cohort supports prolonged diagnostic delay
      among patients whose suprasellar or bifocal tumors caused DI; it does not
      supply an umbrella-disease frequency.
- category: Nervous System
  name: Hydrocephalus
  frequency: FREQUENT
  description: >-
    Obstructive hydrocephalus is a common presentation of intracranial CNS germ
    cell tumors, especially lesions near ventricular cerebrospinal-fluid
    pathways. This does not describe rare primary spinal tumors.
  phenotype_term:
    preferred_term: Hydrocephalus
    term:
      id: HP:0000238
      label: Hydrocephalus
  evidence:
  - reference: PMID:37452948
    reference_title: Brain and Spinal Tumors Originating from the Germ Line Cells.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Above 50% of intracranial GCTs (IGCTs) present obstructive hydrocephalus.
    explanation: >-
      The review supplies a quantitative intracranial presentation estimate that
      maps to the frequent band while preserving the anatomic scope.
- category: Eye
  name: Visual Impairment
  frequency: FREQUENT
  description: >-
    Visual disturbance occurs in CNS-GCT cohorts, with mechanism and severity
    depending on tumor site. The cohort wording does not establish a single
    visual pathway or distinguish impairment present at diagnosis from a
    specific ophthalmologic syndrome.
  phenotype_term:
    preferred_term: Visual impairment
    term:
      id: HP:0000505
      label: Visual impairment
  evidence:
  - reference: PMID:39035740
    reference_title: "Primary central nervous system germ cell tumors in Central America and the Caribbean Region: an AHOPCA 20-year experience."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Presenting symptoms were headaches (n = 24, 50%), visual disturbances (n =
      17, 35.4%), vomiting (n = 12, 25%), nausea (n = 8, 16.7%), and diabetes
      insipidus (n = 7, 14.6%).
    explanation: >-
      The 35.4% cohort estimate supports a frequent broad visual-impairment
      phenotype; the source does not resolve a uniform anatomic mechanism.
- category: Endocrine
  name: Precocious Puberty
  frequency: VERY_RARE
  description: >-
    Precocious puberty is an uncommon presentation reported in NGGCT,
    potentially associated with secretory tumor components. The available cohort
    observation does not establish HCG as the mediator in those two patients.
  phenotype_term:
    preferred_term: Precocious puberty
    term:
      id: HP:0000826
      label: Precocious puberty
  subtypes:
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  evidence:
  - reference: PMID:39035740
    reference_title: "Primary central nervous system germ cell tumors in Central America and the Caribbean Region: an AHOPCA 20-year experience."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Two patients with NGGCT presented with precocious puberty.
    explanation: >-
      Two of 48 cohort patients had this presentation, supporting a very-rare
      band at the umbrella-disease level without establishing mechanism.
- category: Blood
  name: Elevated Circulating Alpha-Fetoprotein Concentration
  diagnostic: true
  description: >-
    Elevated serum or CSF AFP supports NGGCT clinical classification, but it is
    not specific to a yolk-sac component and can occur with immature teratoma.
  phenotype_term:
    preferred_term: Elevated circulating alpha-fetoprotein concentration
    term:
      id: HP:0006254
      label: Elevated circulating alpha-fetoprotein concentration
  subtypes:
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  reports_on:
  - target: Tumor-Marker Secretion by CNS Germ Cell Tumor Components
    relationship: READOUT_OF
    direction: POSITIVE
    endpoint_context: DIAGNOSTIC
    interpretation: AFP elevation is a tumor-marker readout, not a unique histology label.
  evidence:
  - reference: PMID:36995447
    reference_title: "Impact of tumor markers on diagnosis, treatment and prognosis in CNS germ cell tumors: correlations with clinical practice and histopathology."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      AFP was often elevated in GCTs without a yolk sac tumor component,
      especially immature teratoma.
    explanation: >-
      Direct serum/CSF marker and histopathology correlation supports diagnostic
      use while showing that AFP is not component-specific.
- category: Blood
  name: Elevated Circulating Beta Chorionic Gonadotropin Concentration
  diagnostic: true
  description: >-
    Elevated serum or CSF HCG is a diagnostic tumor-marker readout across
    selected CNS germ cell tumor components; it does not by itself resolve every
    histologic distinction.
  phenotype_term:
    preferred_term: Elevated circulating beta chorionic gonadotropin concentration
    term:
      id: HP:6000485
      label: Elevated circulating beta chorionic gonadotropin concentration
  reports_on:
  - target: Tumor-Marker Secretion by CNS Germ Cell Tumor Components
    relationship: READOUT_OF
    direction: POSITIVE
    endpoint_context: DIAGNOSTIC
    interpretation: HCG elevation is a tumor-marker readout used in clinical classification.
  evidence:
  - reference: PMID:36995447
    reference_title: "Impact of tumor markers on diagnosis, treatment and prognosis in CNS germ cell tumors: correlations with clinical practice and histopathology."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      HCG was elevated only in cases with a germinoma or choriocarcinoma
      component, and there existed a clear cut-off HCG value between the two.
    explanation: >-
      Direct marker-to-histopathology correlation supports HCG as a diagnostic
      readout while preserving component-dependent interpretation.
biochemical:
- name: Alpha-Fetoprotein
  biomarker_term:
    preferred_term: Alpha-Fetoprotein
    term:
      id: NCIT:C16278
      label: Alpha-Fetoprotein
  presence: Variable elevation in serum or cerebrospinal fluid
  subtypes:
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  notes: >-
    AFP is a diagnostic analyte whose elevation depends on histologic
    composition and is not specific to a yolk-sac component.
  evidence:
  - reference: PMID:36995447
    reference_title: "Impact of tumor markers on diagnosis, treatment and prognosis in CNS germ cell tumors: correlations with clinical practice and histopathology."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      AFP was often elevated in GCTs without a yolk sac tumor component,
      especially immature teratoma.
    explanation: >-
      Direct marker-to-histopathology correlation supports AFP as an analyte
      while bounding component-specific interpretation.
- name: Human Chorionic Gonadotropin
  biomarker_term:
    preferred_term: Human Chorionic Gonadotropin
    term:
      id: NCIT:C2275
      label: Human Chorionic Gonadotropin
  presence: Variable elevation in serum or cerebrospinal fluid
  notes: >-
    HCG is a diagnostic analyte whose elevation varies with tumor component and
    specimen compartment.
  evidence:
  - reference: PMID:36995447
    reference_title: "Impact of tumor markers on diagnosis, treatment and prognosis in CNS germ cell tumors: correlations with clinical practice and histopathology."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      HCG was elevated only in cases with a germinoma or choriocarcinoma
      component, and there existed a clear cut-off HCG value between the two.
    explanation: >-
      Direct marker-to-histopathology correlation supports HCG as an analyte
      with component-dependent interpretation.
- name: Circulating miR-371a-3p
  subtypes:
  - Central Nervous System Germinoma
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  notes: >-
    Candidate serum/CSF biomarker supported by a four-case proof-of-concept
    series containing three intracranial malignant GCTs and one Langerhans-cell
    histiocytosis comparator. No validated sensitivity, specificity, or
    teratoma-detection claim is assigned.
  evidence:
  - reference: PMID:32642701
    reference_title: Clinical utility of circulating miR-371a-3p for the management of patients with intracranial malignant germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Serum and/or CSF miR-371a-3p levels distinguished those with intracranial
      malignant GCTs from LCH and, if known in real time, could have helped
      clinical management.
    explanation: >-
      The small clinical series is a direct intracranial signal but remains
      hypothesis-generating rather than a validated diagnostic test.
- name: Cerebrospinal Fluid Circulating Tumor DNA
  subtypes:
  - Central Nervous System Nongerminomatous Germ Cell Tumor
  notes: >-
    Candidate monitoring biomarker from a retrospective 17-patient pediatric
    NGGCT cohort. The study's data-derived 6% threshold and small sample prevent
    treating ctDNA as a validated residual-disease or prognostic standard.
  evidence:
  - reference: PMID:39189644
    reference_title: Monitoring pediatric CNS non-germinomatous germ cell tumors via cerebrospinal fluid circulating tumor DNA.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Among them, ctDNA before chemotherapy as well as ctDNA clearance were
      consistently associated with PFS and OS (p < .05).
    explanation: >-
      The retrospective association supports further evaluation of CSF ctDNA
      for monitoring but does not establish a causal resistance mechanism.
genetic:
- name: KIT
  gene_term:
    preferred_term: KIT
    term:
      id: hgnc:6342
      label: KIT
  association: Recurrent germinoma-enriched somatic mutation
  variant_origin: SOMATIC
  notes: >-
    KIT alteration is recurrent but non-universal and is not established as the
    initiating event in every CNS germ cell tumor.
  evidence:
  - reference: PMID:38409885
    reference_title: Whole-exome sequencing has revealed novel genetic characteristics in intracranial germ cell tumours in the Chinese.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      We found that KIT was the most significantly mutated gene (15/47, 32%),
      which mainly occurred in the germinoma group (13/20, 65%), and less
      frequently in NGGCT (2/27, 7%).
    explanation: >-
      Human tumor sequencing establishes recurrence and subtype enrichment, not
      universal causality.
- name: KRAS
  gene_term:
    preferred_term: KRAS
    term:
      id: hgnc:6407
      label: KRAS
  association: Recurrent somatic KIT-RAS-pathway mutation
  variant_origin: SOMATIC
  evidence:
  - reference: PMID:24896186
    reference_title: Novel somatic and germline mutations in intracranial germ cell tumours.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      We find the KIT/RAS signalling pathway frequently mutated in more than
      50% of IGCTs, including novel recurrent somatic mutations in KIT, its
      downstream mediators KRAS and NRAS, and its negative regulator CBL.
    explanation: >-
      Patient-tumor sequencing supports recurrent KRAS mutation within a
      heterogeneous pathway-level pattern.
- name: NRAS
  gene_term:
    preferred_term: NRAS
    term:
      id: hgnc:7989
      label: NRAS
  association: Recurrent somatic KIT-RAS-pathway mutation
  variant_origin: SOMATIC
  evidence:
  - reference: PMID:24896186
    reference_title: Novel somatic and germline mutations in intracranial germ cell tumours.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      We find the KIT/RAS signalling pathway frequently mutated in more than
      50% of IGCTs, including novel recurrent somatic mutations in KIT, its
      downstream mediators KRAS and NRAS, and its negative regulator CBL.
    explanation: >-
      Patient-tumor sequencing supports recurrent NRAS mutation within a
      heterogeneous pathway-level pattern.
- name: AKT1
  gene_term:
    preferred_term: AKT1
    term:
      id: hgnc:391
      label: AKT1
  association: Recurrent somatic copy-number gain with increased expression
  variant_origin: SOMATIC
  evidence:
  - reference: PMID:24896186
    reference_title: Novel somatic and germline mutations in intracranial germ cell tumours.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Novel somatic alterations in the AKT/mTOR pathway included copy number
      gains of the AKT1 locus at 14q32.33 in 19% of patients, with corresponding
      upregulation of AKT1 expression.
    explanation: >-
      Patient-tumor genomic and expression data establish a recurrent alteration
      in a minority of cases without defining its temporal role.
diagnosis:
- name: Serum and Cerebrospinal Fluid Tumor-Marker Assessment
  markers: AFP, HCG
  description: >-
    Measure AFP and HCG in both serum and CSF to support CNS germ cell tumor
    classification. Marker elevation can permit a clinical NGGCT diagnosis in
    some settings, while marker-negative malignant tumors may require tissue
    confirmation.
  evidence:
  - reference: PMID:36995447
    reference_title: "Impact of tumor markers on diagnosis, treatment and prognosis in CNS germ cell tumors: correlations with clinical practice and histopathology."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      HCG was elevated only in CSF in 3-of-52 cases, and AFP was elevated only in
      serum in 7-of-49 cases, emphasizing the potential utilization of both
      serum and CSF studies.
    explanation: >-
      Direct paired-compartment observations support testing both serum and CSF.
  - reference: PMID:32642701
    reference_title: Clinical utility of circulating miR-371a-3p for the management of patients with intracranial malignant germ cell tumors.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The current biomarkers alpha-fetoprotein (AFP) and human chorionic
      gonadotropin (HCG) have limited sensitivity/specificity for diagnosing
      malignant germ cell tumors (GCTs) and "marker-negative" patients require
      histological confirmation for diagnosis.
    explanation: >-
      Background to the proof-of-concept series bounds marker-negative
      interpretation and the need for histology.
- name: Neuraxis Staging with MRI and Cerebrospinal Fluid Cytology
  description: >-
    Before treatment, contrast-enhanced MRI of brain and whole spine plus CSF
    cytology evaluates macroscopic and cytologic dissemination. This is staging,
    not a substitute for histology when tumor markers are non-diagnostic.
  evidence:
  - reference: PMID:37452948
    reference_title: Brain and Spinal Tumors Originating from the Germ Line Cells.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Staging work-up includes CSF cytology for tumor cells and contrast-enhanced
      MRI of brain and spine for macroscopic metastasis before treatment
      commences.
    explanation: >-
      The clinical review directly describes pre-treatment neuraxis staging.
treatments:
- name: Response-Adapted CarboPEI Chemotherapy and Whole-Ventricular Radiation
  description: >-
    For localized germinoma in SIOP-CNS-GCT-II, four carboPEI courses were
    followed by 24 Gy whole-ventricular radiation, with a 16 Gy tumor boost for
    residual disease. This subtype- and response-specific protocol should not be
    generalized to NGGCT or treated as randomized proof of one optimal regimen.
  treatment_term:
    preferred_term: Chemoradiotherapy
    term:
      id: NCIT:C94626
      label: Chemoradiotherapy
  evidence:
  - reference: PMID:42234858
    reference_title: Final report of the SIOP-CNS-GCT II protocol for germinoma-omitting radiotherapy boost is safe in localized disease in complete remission after chemotherapy.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Patients with localized germinoma received four courses of 'carboPEI'
      chemotherapy, then 24Gy WVRT if in complete remission (CR), with an
      additional 16Gy tumor boost if residual disease was present.
    explanation: >-
      The prospective protocol report directly establishes the response-adapted
      localized-germinoma regimen.
- name: Response-Qualified Chemotherapy and Reduced-Field Radiation for Localized NGGCT
  description: >-
    ACNS1123 used alternating carboplatin/etoposide and
    ifosfamide/etoposide, followed in complete or partial responders by reduced
    whole-ventricular radiation and a tumor-bed boost. The phase-II result was
    accompanied by spinal failures, so field adequacy remains an open question
    rather than a universal standard.
  treatment_term:
    preferred_term: Chemoradiotherapy
    term:
      id: NCIT:C94626
      label: Chemoradiotherapy
    therapeutic_agent:
    - preferred_term: carboplatin
      term:
        id: CHEBI:31355
        label: carboplatin
    - preferred_term: etoposide
      term:
        id: CHEBI:4911
        label: etoposide
    - preferred_term: ifosfamide
      term:
        id: CHEBI:5864
        label: ifosfamide
  evidence:
  - reference: PMID:31545689
    reference_title: "Phase II Trial of Response-Based Radiation Therapy for Patients With Localized CNS Nongerminomatous Germ Cell Tumors: A Children's Oncology Group Study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Patients received six cycles of chemotherapy with carboplatin and
      etoposide alternating with ifosfamide and etoposide, as used in the
      Children's Oncology Group predecessor study (ACNS0122; ClinicalTrials.gov
      identifier: NCT00047320).
    explanation: >-
      The phase-II protocol directly establishes the alternating chemotherapy
      backbone for localized NGGCT.
  - reference: PMID:31545689
    reference_title: "Phase II Trial of Response-Based Radiation Therapy for Patients With Localized CNS Nongerminomatous Germ Cell Tumors: A Children's Oncology Group Study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Patients who achieved a complete response (CR) or partial response (PR)
      with or without second-look surgery were eligible for reduced RT, defined
      as 30.6 Gy whole ventricular field and 54 Gy tumor-bed boost, compared
      with 36 Gy craniospinal irradiation plus 54 Gy tumor-bed boost used in
      ACNS0122.
    explanation: >-
      The trial defines the response qualification and reduced radiation field;
      it does not settle whether that field is adequate for all localized NGGCT.
- name: Myeloablative Chemotherapy with Autologous Stem Cell Transplantation
  description: >-
    High-dose chemotherapy with autologous stem-cell rescue is a selected
    salvage option for relapsed or progressed CNS germ cell tumors. Evidence
    comes from a small prospective study and does not establish a
    subtype-independent standard.
  treatment_term:
    preferred_term: hematopoietic stem cell transplantation
    term:
      id: NCIT:C15431
      label: Hematopoietic Cell Transplantation
  evidence:
  - reference: PMID:23824533
    reference_title: "Myeloablative chemotherapy and autologous stem cell transplantation in patients with relapsed or progressed central nervous system germ cell tumors: results of Korean Society of Pediatric Neuro-Oncology (KSPNO) S-053 study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Eleven patients with non-germinomatous germ cell tumors and nine patients
      with germinomas were enrolled.
    explanation: >-
      The 20-patient sample bounds the generalizability of this salvage strategy.
  - reference: PMID:23824533
    reference_title: "Myeloablative chemotherapy and autologous stem cell transplantation in patients with relapsed or progressed central nervous system germ cell tumors: results of Korean Society of Pediatric Neuro-Oncology (KSPNO) S-053 study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      In conclusion, HDCT/autoSCT was feasible, and survival rates were
      encouraging. Further study with a larger cohort of patients is needed to
      elucidate the role of HDCT/autoSCT in the treatment of relapsed or
      progressed CNS-GCTs.
    explanation: >-
      The investigators explicitly frame feasibility as encouraging but
      insufficient to define the treatment's role.
- name: Resection of Persistent or Growing Teratomatous Disease
  description: >-
    Surgical resection can define and remove enlarging teratomatous tissue after
    chemotherapy, including growing teratoma syndrome. Case-level evidence does
    not establish one universal timing or extent of resection.
  treatment_term:
    preferred_term: surgical procedure on nervous system
    term:
      id: NCIT:C15656
      label: Neurosurgical Procedure
  evidence:
  - reference: PMID:42488730
    reference_title: Growing Teratoma Syndrome Arising from a Pineal Mixed Germ Cell Tumor in a 12-Year-Old Boy.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Subsequent surgical resection revealed a mature teratoma without malignant
      elements, consistent with growing teratoma syndrome (GTS).
    explanation: >-
      This single case demonstrates resection in a post-chemotherapy pineal GTS
      context without establishing a general treatment rule.
discussions:
- discussion_id: gap_cns_gct_developmental_origin_and_driver_timing
  prompt: >-
    Do CNS germ cell tumors arise from a mis-migrated primordial germ cell, an
    endogenous neural stem or progenitor cell, or another precursor; when do
    lineage divergence and global methylation changes occur; and are MAPK/PI3K
    alterations initiating lesions, later cooperating events, or dispensable
    markers of particular components?
  kind: KNOWLEDGE_GAP
  status: OPEN
  attaches_to:
  - mechanistic_hypotheses#common_ancestry_developmental_state_divergence
  rationale: >-
    The strongest human evidence is cross-sectional. Methylation profiles in 61
    primary tumors and integrated transcriptome/methylome profiles in 84 cases
    map germinoma and NGGCT to different developmental reference states, while
    shared MAPK/PI3K variants in microdissected mixed components and
    component-shared 12p status in a separate cohort support within-tumor common
    ancestry. SNRPN hypomethylation is also present in mouse and human neural
    stem cells, and two non-tumor fetal autopsies prompted an author hypothesis
    of local germinal-matrix derivation. These observations do not trace a
    tumor lineage. The studies do not observe embryonic migration, establish
    whether the relevant precursor is a bona fide primordial germ cell, an
    endogenous neural stem or progenitor cell, or another precursor, or
    determine whether pathway alterations precede or follow epigenetic state
    divergence. The 62-tumor sequencing cohort found KIT/RAS or AKT/mTOR lesions
    in 53%, with KIT mutations restricted to germinomas in that cohort, so one
    pathway chain cannot be generalized across the family. Direct temporal,
    lineage, and perturbational evidence that compares competing precursor
    models is required before adding an established developmental-escape or
    MAPK/PI3K pathophysiology edge.
  proposed_experiments:
  - experiment_id: exp_cns_gct_spatial_clonal_developmental_atlas
    name: >-
      Spatial single-cell clonal atlas across pure and mixed CNS germ cell tumors
    description: >-
      Prospectively collect treatment-naive pure germinomas, pure NGGCT
      histologies, and multiple spatially mapped components of mixed tumors.
      Integrate single-cell DNA, RNA, chromatin accessibility, methylation, and
      spatial pathology with matched normal DNA and a preregistered human
      embryonic germ-cell reference atlas.
    experiment_type:
      preferred_term: spatial single-cell multi-omic lineage study
    model_systems:
    - name: Treatment-naive CNS germ cell tumor cohort
      description: >-
        Include age-, sex-, location-, and histology-stratified tumors with
        central pathology review and separate sampling of every visible mixed
        component.
    - name: Human developmental reference atlas
      description: >-
        Use ethically sourced published and prospectively generated reference
        profiles spanning primordial germ-cell specification and migration,
        neural stem and progenitor states, and early embryonic differentiation
        without treating nearest-neighbor similarity as proof of ancestry.
    perturbations:
    - name: Component-resolved sampling
      target: mechanistic_hypotheses#common_ancestry_developmental_state_divergence
      description: >-
        Laser-capture and single-cell profile each component separately while
        preserving spatial relationships and shared somatic variants.
    readouts:
    - name: Clonal order and developmental-state assignment
      target: mechanistic_hypotheses#common_ancestry_developmental_state_divergence
      description: >-
        Infer trunk and branch lesions, allele-specific copy number, methylation
        erasure or acquisition, lineage programs, and uncertainty-calibrated
        similarity to developmental references.
      assays:
      - preferred_term: single-cell DNA and RNA sequencing
      - preferred_term: single-cell methylome and chromatin accessibility profiling
      - preferred_term: spatial transcriptomics
      direction: THRESHOLD_DEPENDENT
    controls:
    - name: Histology and site controls
      description: >-
        Analyze germinoma and each NGGCT component separately, include gonadal
        germ cell tumor comparators, and adjust for purity, immune infiltration,
        age, sex, anatomic site, and specimen handling.
    - name: Clonal-inference controls
      description: >-
        Confirm trunk variants by orthogonal targeted sequencing and test
        alternative phylogenies rather than selecting a lineage tree from
        transcriptomic similarity alone.
    decision_criterion: >-
      Common ancestry is supported when spatially distinct mixed components
      share orthogonally validated trunk lesions and diverge into reproducible
      component-specific epigenetic states. A particular precursor-of-origin
      model gains support only if tumor trunk states reproducibly align with a
      defined developmental interval across cohorts, fit competing precursor
      models materially less well, and cannot be explained by tumor purity or
      convergence; developmental similarity alone does not establish lineage.
      Divergent trunk lesions across components would refute a single common
      tumor ancestor, whereas component differences arising after a shared
      trunk are expected under developmental-state divergence and do not refute
      it. Inconsistent developmental assignments leave cell of origin
      unresolved rather than refuting common ancestry.
    would_support:
    - mechanistic_hypotheses#common_ancestry_developmental_state_divergence
    would_refute:
    - mechanistic_hypotheses#common_ancestry_developmental_state_divergence
  - experiment_id: exp_cns_gct_stage_specific_driver_epistasis
    name: Stage-specific MAPK/PI3K developmental epistasis
    description: >-
      Introduce recurrent KIT, RAS, AKT1, or mTOR alterations singly and in
      combinations at defined stages of human primordial-germ-cell-like-cell and
      embryonic differentiation, then test whether a lesion initiates,
      maintains, or merely accelerates a tumor-compatible state.
    experiment_type:
      preferred_term: isogenic developmental-stage causal epistasis study
    model_systems:
    - name: Isogenic human candidate-precursor differentiation series
      description: >-
        Compare multiple donor backgrounds of primordial-germ-cell-like cells
        and CNS neural stem or progenitor cells, with inducible editing at
        matched specification and differentiation stages.
    - name: Component-specific CNS germ cell tumor organoids
      description: >-
        Establish centrally reviewed germinoma and NGGCT organoids with
        isogenic correction or introduction of candidate lesions.
    perturbations:
    - name: Timed driver introduction and correction
      target: mechanistic_hypotheses#common_ancestry_developmental_state_divergence
      description: >-
        Activate or correct one candidate pathway lesion at each developmental
        stage, with matched kinase-dead, vector, and unedited controls.
    readouts:
    - name: State maintenance and transformation
      target: mechanistic_hypotheses#common_ancestry_developmental_state_divergence
      description: >-
        Measure survival without developmental cues, methylation state,
        differentiation trajectory, clonogenicity, invasion, and tumor formation
        in an anatomically relevant xenograft.
      assays:
      - preferred_term: longitudinal multi-omic profiling
      - preferred_term: clonogenic transformation assay
      direction: POSITIVE
    controls:
    - name: Lesion and developmental-stage controls
      description: >-
        Compare each alteration across matched primordial-germ-cell-like and
        neural stem or progenitor backgrounds and developmental stages, include
        pathway correction and pharmacologic rescue, and analyze
        germinoma-like and NGGCT-like outputs separately.
    decision_criterion: >-
      An initiating role requires a lesion to reproducibly create a
      self-maintaining tumor-compatible state only when introduced at a defined
      precursor stage, with loss of that phenotype after isogenic correction.
      A maintenance role is favored if correction collapses established tumor
      state but introduction does not initiate it. Lack of necessity and
      sufficiency supports a passenger or context-marker interpretation.
    would_support:
    - mechanistic_hypotheses#common_ancestry_developmental_state_divergence
    would_refute:
    - mechanistic_hypotheses#common_ancestry_developmental_state_divergence
  evidence:
  - reference: PMID:28078450
    reference_title: Genome-wide methylation profiles in primary intracranial germ cell tumors indicate a primordial germ cell origin for germinomas.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Recent exhaustive genomic studies showed that mutations of the genes
      involved in the MAPK and/or PI3K pathways are common in iGCTs; however,
      the mechanisms of how different subtypes develop, often as a mixed-GCT,
      are unknown.
    explanation: >-
      The primary study explicitly identifies the unresolved link between
      recurrent pathway alterations and subtype development.
- discussion_id: gap_nggct_refractory_and_relapse_mechanism
  prompt: >-
    Which cellular, lineage, genomic, epigenetic, microenvironmental, or
    pharmacologic states cause primary refractory disease, viable progression,
    growing teratoma syndrome, and post-treatment relapse in CNS NGGCT, and
    which are actionable rather than correlates of histology?
  kind: KNOWLEDGE_GAP
  status: OPEN
  attaches_to:
  - has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
  - mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
  - mechanistic_hypotheses#mir214_bcl2l11_cisplatin_response_candidate
  rationale: >-
    Clinical failure is heterogeneous. A pooled analysis separated 24
    progressions during treatment, 11 growing-teratoma-syndrome events, and 83
    post-treatment relapses. Large genomic studies predominantly profile
    primary tumors, and the recent NGGCT cohort sequenced only 12 paired
    tumor-blood samples; neither design compares diagnosis with residual,
    progressive, and relapsed viable tumor within patients. A separate study
    observed early-branched evolution across only two histology-changing
    longitudinal cases, which supports feasibility but not a recurrent driver.
    An older serial yolk-sac-tumor case likewise showed increasing copy-number
    complexity over two recurrences, but its single-case, arm-level design and
    unreported regimen or exposure details do not establish a resistance
    locus.
    Molecular subtypes such as immune-hot, MYC/E2F, and SHH are therefore
    candidates, not validated resistance states. The miR-214-3p–BCL2L11 result
    supplies a specific in-vitro cisplatin-tolerance candidate, but lacks a
    resistant-versus-sensitive clinical cohort, a patient-derived CNS model,
    and rescue evidence. Histologic mixture, sampling bias, radiation
    geography, pharmacokinetics, and therapy-selected clonal evolution remain
    competing explanations. A resistance node, predictive biomarker, or
    targeted therapy should not be added until longitudinal molecular change is
    linked to function and outcome.
  proposed_experiments:
  - experiment_id: exp_nggct_paired_failure_evolution_cohort
    name: >-
      Paired diagnosis-residual-progression-relapse NGGCT evolution cohort
    description: >-
      Build an international prospective cohort with central pathology review,
      multi-region diagnosis sampling, protocol-level treatment exposure, and
      tissue at second-look surgery, progression, GTS resection, or relapse.
      Collect serial plasma and CSF before, during, and after therapy.
    experiment_type:
      preferred_term: prospective longitudinal tumor evolution study
    model_systems:
    - name: Longitudinal CNS NGGCT clinical cohort
      description: >-
        Enroll pure and mixed NGGCT histologies and preserve separate strata for
        induction nonresponse, GTS, local viable progression, metastatic
        progression, and late relapse.
    - name: Response-matched comparator cohort
      description: >-
        Include durable responders matched on stage, histology, markers,
        radiation field, age, sex, location, and treatment regimen.
    perturbations:
    - name: Protocol-defined chemotherapy and radiotherapy exposure
      target: mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
      description: >-
        Record cumulative agent doses, dose intensity, delays, surgery,
        radiation volumes and dosimetry, and supportive medications as
        time-varying exposures.
    readouts:
    - name: Within-patient clonal and lineage-state evolution
      target: mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
      description: >-
        Compare somatic variants, copy number, ecDNA, methylation, transcriptome,
        chromatin accessibility, immune state, and histologic composition across
        every within-patient time point.
      assays:
      - preferred_term: single-cell and bulk multi-omic sequencing
      - preferred_term: spatial pathology
      direction: THRESHOLD_DEPENDENT
    - name: Serial molecular residual disease
      target: mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
      description: >-
        Track patient-specific mutations, copy-number changes, methylation, and
        validated germ-cell-tumor RNAs in CSF and plasma alongside AFP, beta-hCG,
        imaging, and clinical response.
      assays:
      - preferred_term: cerebrospinal-fluid circulating tumor DNA assay
      - preferred_term: serial tumor-marker measurement
      direction: POSITIVE
    controls:
    - name: Failure-state separation
      description: >-
        Do not pool marker-negative GTS, viable malignant progression, and
        post-treatment relapse. Require central pathology and viable-tumor
        content estimates for every tissue comparison.
    - name: Sampling and treatment controls
      description: >-
        Model tumor purity, spatial sampling, histology, treatment intensity,
        radiation geography, and time to failure; validate candidate alterations
        in an independent protocol-treated cohort.
    decision_criterion: >-
      Therapy-selected persistence is supported when a diagnosis-time subclone
      or state survives, expands, or acquires a reproducible change in viable
      residual or relapse tumor, precedes failure in CSF, and predicts failure
      beyond histology, stage, markers, and radiation field. GTS requires a
      teratomatous lineage result rather than being labeled molecular relapse.
      Failure to find reproducible within-patient evolution, or associations
      fully explained by component sampling or treatment geography, refutes a
      shared molecular-resistance model.
    would_support:
    - mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
    would_refute:
    - mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
  - experiment_id: exp_nggct_resistance_functional_reconstruction
    name: Functional reconstruction of candidate NGGCT resistance states
    description: >-
      Establish paired diagnosis and failure models from the longitudinal
      cohort, barcode diagnosis-time cells, and causally test candidate
      lineage, signaling, DNA-damage-response, and microenvironmental states
      under clinically matched chemotherapy and radiation schedules.
    experiment_type:
      preferred_term: patient-derived resistance epistasis and rescue study
    model_systems:
    - name: Paired patient-derived NGGCT organoids and xenografts
      description: >-
        Preserve component-specific histology and use centrally reviewed paired
        diagnosis and failure material whenever available.
    - name: Isogenic barcoded NGGCT models
      description: >-
        Introduce or correct candidate alterations in multiple genetic and
        histologic backgrounds rather than relying on one cell line.
    perturbations:
    - name: Clinically matched multimodal therapy
      target: mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
      description: >-
        Apply carboplatin or cisplatin, etoposide, ifosfamide, and fractionated
        radiation singly and in protocol-relevant sequences.
    - name: Candidate-state ablation and rescue
      target: mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
      description: >-
        Genetically suppress and rescue only candidates that were observed to
        expand within patients, including lineage programs, signaling nodes,
        DNA-repair states, or immune interactions.
    - name: miR-214-3p inhibition and BCL2L11 epistasis
      target: mechanistic_hypotheses#mir214_bcl2l11_cisplatin_response_candidate
      description: >-
        In models with endogenous miR-214-3p elevation, inhibit miR-214-3p,
        knock down or restore BCL2L11 independently, and perform
        concentration-matched rescue under cisplatin and non-platinum therapy.
    readouts:
    - name: Clonal survival and cross-resistance
      target: mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
      description: >-
        Measure barcode abundance, viability, apoptosis, DNA damage and repair,
        differentiation, tumor regrowth, and cross-resistance to chemotherapy
        and radiation.
      assays:
      - preferred_term: clonal barcode tracking
      - preferred_term: drug and radiation response assay
      direction: POSITIVE
    controls:
    - name: Isogenic and histology controls
      description: >-
        Include unedited, corrected, and rescue controls; test at least two
        models per histologic state and compare with durable-response models.
    decision_criterion: >-
      A candidate is a causal resistance mechanism only if it is enriched
      longitudinally in patients, increases survival under relevant therapy in
      more than one model, loses that advantage after specific ablation, and
      regains it after rescue. Baseline association without within-patient
      enrichment, or model-only resistance without clinical recurrence, is
      insufficient.
    would_support:
    - mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
    - mechanistic_hypotheses#mir214_bcl2l11_cisplatin_response_candidate
    would_refute:
    - mechanistic_hypotheses#nggct_failure_state_specific_persistence_and_evolution
    - mechanistic_hypotheses#mir214_bcl2l11_cisplatin_response_candidate
  evidence:
  - reference: PMID:35218656
    reference_title: "Pattern of treatment failures in patients with central nervous system non-germinomatous germ cell tumors (CNS-NGGCT): A pooled analysis of clinical trials."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Due to their rarity, patients with treatment failures remain a poorly
      characterized group with unfavorable outcomes.
    explanation: >-
      The pooled clinical-trial analysis explicitly identifies the weak
      characterization of CNS-NGGCT treatment failure.
  - reference: PMID:39189644
    reference_title: Monitoring pediatric CNS non-germinomatous germ cell tumors via cerebrospinal fluid circulating tumor DNA.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Among them, ctDNA before chemotherapy as well as ctDNA clearance were
      consistently associated with PFS and OS
    explanation: >-
      This 17-patient retrospective study supports serial CSF ctDNA as a
      feasible response and residual-disease readout, not as a mediator of
      resistance; its positivity threshold was optimized within the same small
      survival dataset.
- discussion_id: controversy_localized_nggct_radiation_field
  prompt: >-
    For localized CNS NGGCT that responds to induction therapy, what radiation
    field preserves disease control, especially in the spine, while reducing
    neurocognitive, endocrine, vascular, and second-neoplasm risk when the
    protocol-defined dose schedule is held constant?
  kind: CONTROVERSY
  status: OPEN
  attaches_to:
  - has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
  rationale: >-
    ACNS1123 gave 30.6 Gy whole-ventricular irradiation with a boost bringing
    the tumor-bed total to 54 Gy in 66 response-qualified patients. Eight
    relapsed and every failure involved the spine; central dosimetry review
    found protocol-conforming targets and dose in 62 of 66 patients, and none
    of the four patients with deviations progressed.
    A separate pooled analysis of four prospective trials and two national
    protocol cohorts found no association between focal or whole-ventricular
    radiation and metastatic rather than local relapse among patients who had
    already failed treatment. Conditioning on failure and omitting
    field-specific denominators for all treated patients means that result
    cannot estimate field-specific failure incidence or rebut the ACNS1123
    pattern. The seed review describes optimal field and dose as controversial
    and minimizing long-term sequelae while preserving control as an unresolved
    challenge. Comparative disease-control and late-effect evidence is missing;
    no universal conclusion that either craniospinal or reduced-field radiation
    is optimal is warranted. This question is intentionally limited to
    response-qualified localized NGGCT and field selection at a fixed
    protocol-defined dose; dose optimization and germinoma protocols require
    separate evaluation.
  proposed_experiments:
  - experiment_id: exp_localized_nggct_risk_adapted_radiation_trial
    name: >-
      Risk-adapted radiation field trial with molecular residual-disease surveillance
    description: >-
      In a prospective international trial, centrally confirm localized NGGCT,
      induction response, histologic composition, serum and CSF markers, and
      second-look pathology. Compare protocol-defined radiation fields under
      contemporary protocol consensus and ethical oversight while embedding
      serial CSF molecular residual-disease testing and lifelong late-effect
      follow-up.
    experiment_type:
      preferred_term: randomized risk-adapted radiation field trial
    model_systems:
    - name: Response-qualified localized CNS NGGCT cohort
      description: >-
        Stratify by complete versus partial response, pure versus mixed
        histology, residual teratoma, marker status, age, sex, location, and
        protocol-defined molecular residual disease.
    perturbations:
    - name: Protocol-defined radiation-field assignment
      target: has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
      description: >-
        Compare reduced ventricular or ventricular-plus-spinal strategies with
        a consensus broader-field control, holding the prescribed dose schedule
        and systemic therapy constant and using centrally reviewed dose-volume
        constraints.
    readouts:
    - name: Disease-control and failure geography
      target: has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
      description: >-
        Measure event-free and overall survival, local, ventricular, spinal, and
        disseminated relapse, salvage success, serial tumor markers, and CSF
        molecular residual disease.
      assays:
      - preferred_term: magnetic resonance surveillance
      - preferred_term: cerebrospinal-fluid molecular residual-disease assay
      direction: THRESHOLD_DEPENDENT
    - name: Long-term treatment burden
      target: has_subtypes#Central Nervous System Nongerminomatous Germ Cell Tumor
      description: >-
        Prospectively assess neurocognition, school and employment function,
        endocrine replacement, growth, hearing, vascular events, and secondary
        neoplasms for at least ten years.
      assays:
      - preferred_term: longitudinal late-effect assessment
      direction: NEGATIVE
    controls:
    - name: Central treatment-quality control
      description: >-
        Review pathology, response, second-look surgery, imaging, target
        contours, dosimetry, and protocol adherence before or during treatment.
    - name: Prespecified competing-risk analysis
      description: >-
        Lock noninferiority and late-effect estimands, account for salvage and
        competing mortality, and report every failure location rather than only
        aggregate progression-free survival.
    decision_criterion: >-
      A reduced field is acceptable only if it meets a prespecified
      noninferiority margin for disease control without an excess of spinal or
      disseminated relapse and produces a clinically meaningful late-effect
      benefit. If baseline or on-treatment CSF molecular residual disease
      reproducibly identifies spinal risk, test a biomarker-adapted field
      strategy prospectively rather than assigning a field from association
      alone. Similar disease control without measurable long-term benefit does
      not justify field reduction.
  evidence:
  - reference: PMID:31545689
    reference_title: "Phase II Trial of Response-Based Radiation Therapy for Patients With Localized CNS Nongerminomatous Germ Cell Tumors: A Children's Oncology Group Study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      However, the patterns of failure were distinct, with all patients having
      treatment failure in the spine.
    explanation: >-
      The response-based phase II trial directly identifies the spinal failure
      pattern after reduced whole-ventricular radiation.
  - reference: PMID:34990779
    reference_title: "A Phase 2 Trial of Response-Based Radiation Therapy for Localized Central Nervous System Germ Cell Tumors: Patterns of Failure and Radiation Dosimetry for Nongerminomatous Germ Cell Tumors."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Eight patients relapsed: 6 with isolated spinal relapse and 2 with disease
      in the brain and spine.
    explanation: >-
      Central review confirms the number and geography of failures in the
      reduced-field cohort.
  - reference: PMID:35218656
    reference_title: "Pattern of treatment failures in patients with central nervous system non-germinomatous germ cell tumors (CNS-NGGCT): A pooled analysis of clinical trials."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      In our analysis, focal or whole-ventricular radiation therapy was not
      associated with an increased risk of metastatic relapses.
    explanation: >-
      This case-only failure cohort describes metastatic versus local relapse
      patterns among patients who failed. Without field-specific denominators
      among all treated patients, it cannot estimate comparative failure
      incidence or establish that reduced fields are noninferior.
  - reference: PMID:42419530
    reference_title: Advances in diagnosis and treatment of intracranial germ cell tumors and future directions.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Current treatment predominantly relies on surgery, radiotherapy, and
      chemotherapy, though the optimal radiation field and dose remain
      controversial.
    explanation: >-
      The seed review explicitly identifies field and dose as unresolved; a
      review-level statement establishes the gap, not comparative efficacy.
  - reference: PMID:42419530
    reference_title: Advances in diagnosis and treatment of intracranial germ cell tumors and future directions.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      As the survival rate of patients with IGCTs improves, we are also
      confronted with new clinical challenges: how to minimize treatment-related
      long-term sequelae while maintaining excellent therapeutic outcomes
    explanation: >-
      The review separately frames late-effect reduction as an unresolved
      objective that must be measured alongside disease control.
📚

References & Deep Research

Deep Research

1
OpenScientist ▸
Key Findings
openscientist-autonomous 33 citations 2026-07-26T09:01:09.593845

Key Findings

1. Germinoma is defined by global DNA hypomethylation resembling migrating primordial germ cells

Genome-wide methylation profiling of 61 intracranial GCTs (Fukushima et al., 2017) established that pure germinomas are characterized by global low DNA methylation, a unique epigenetic feature distinguishing them from all other iGCT subtypes. The methylation landscape closely mirrors that of primordial germ cells at the migration phase, and hypomethylation extends beyond the PGC signature into LINE retrotransposons. This is the strongest single molecular argument for a PGC-related state of origin for germinoma. [Human/computational]

Importantly, the same study showed that histologically and epigenetically distinct microdissected components of mixed GCTs shared identical somatic MAPK/PI3K mutations, indicating they developed from a common ancestral cell that subsequently diverged in developmental state. This is direct evidence that developmental-state divergence within mixed tumors is a real, clonally-anchored phenomenon rather than the co-incidence of independent tumors.

"pure germinomas are characterized by global low DNA methylation, a unique epigenetic feature making them distinct from all other iGCTs subtypes. The patterns of methylation strongly resemble that of primordial germ cells (PGC) at the migration phase, possibly indicating the cell of origin for these tumors" — PMID: 28078450

2. Marked geographic, age, and sex predilection

CNS GCTs show a striking region-specific prevalence, comprising 15.3% of pediatric CNS tumors in some Asian populations versus 3.6% in North America (PMID: 34074342). Incidence is bimodal, peaking in the first months of life and again in adolescence, with a male predominance that is most pronounced for pineal tumors. The common intracranial sites are the pineal region, neurohypophysis/suprasellar region, bifocal pineal–neurohypophyseal disease, basal ganglia, and cerebral ventricles. More than 50% of intracranial GCTs present with obstructive hydrocephalus, and primary spinal tumors are rare — a distinction that must be preserved, since the evidence base is overwhelmingly intracranial. [Human/clinical]

"There are two age peaks of incidence distribution at the first few months of life and in adolescence." … "Above 50% of intracranial GCTs (IGCTs) present obstructive hydrocephalus. Spinal tumors are rare." — PMID: 37452948

3. KIT/RAS/MAPK and AKT/mTOR are the dominant molecular drivers; KIT enriched in germinoma

The landmark genomic study of 62 intracranial GCTs (Wang et al., 2014, Nature) found the KIT/RAS signaling pathway mutated in >50% of IGCTs, including recurrent somatic mutations in KIT, KRAS, NRAS, and the negative regulator CBL; novel AKT/mTOR alterations, notably AKT1 copy-number gain at 14q32.33 in 19% of patients with AKT1 upregulation; loss-of-function BCORL1 mutations; and enriched rare germline variants in the histone demethylase JMJD1C. [Computational/genomic]

A Chinese whole-exome cohort (Huang et al., 2024, n=47) confirmed KIT as the most significantly mutated gene (15/47, 32%), predominantly in germinoma (13/20, 65%) versus NGGCT (2/27, 7%). NF1 mutation was associated with shorter OS/PFS, and clonal-evolution analysis revealed an early branched pattern accompanying histologic-subtype changes — reinforcing the common-ancestor-then-diverge model. KRAS codon 12/13/61 mutations have been independently documented, and chromosomal instability produces a characteristic 12p gain.

"We find the KIT/RAS signalling pathway frequently mutated in more than 50% of IGCTs … Novel somatic alterations in the AKT/mTOR pathway included copy number gains of the AKT1 locus at 14q32.33 in 19% of patients" — PMID: 24896186

"KIT was the most significantly mutated gene (15/47, 32%), which mainly occurred in the germinoma group (13/20, 65%), and less frequently in NGGCT (2/27, 7%)" — PMID: 38409885

Pathway / feature Frequency Subtype skew Source
KIT/RAS/MAPK activation >50% of IGCTs germinoma-enriched PMID: 24896186
KIT mutation 32% overall; 65% germinoma vs 7% NGGCT germinoma PMID: 38409885
AKT1 gain (14q32.33) 19% — PMID: 24896186
12p gain / chromosomal instability characteristic all PMID: 38012690
NF1 mutation — worse OS/PFS PMID: 38409885

4. Localized germinoma is highly curable with chemotherapy plus reduced whole-ventricular irradiation

The SIOP-CNS-GCT-II trial (Calaminus et al.) treated 166 localized germinoma patients with four courses of "carboPEI" chemotherapy, then 24 Gy whole-ventricular radiotherapy (with a 16 Gy boost only if residual disease persisted), achieving 5-year EFS 0.94 ± 0.02 and OS 0.98 ± 0.01. Metastatic germinoma (n=61) treated with craniospinal radiotherapy reached 5-year EFS 0.98, OS 1.00. Notably, omitting the radiotherapy boost was safe in patients in complete remission after chemotherapy (PMID: 42234858). The EANO/SNO/EURACAN consensus affirms >90% 5-year EFS for localized germinoma via chemotherapy followed by whole-ventricular irradiation with local boost, while NGGCT 5-year EFS exceeds 70%. [Human/clinical]

"With more than 90% 5-year event-free survival (EFS), localized germinomas can be managed without aggressive surgery, and benefit from chemotherapy followed by whole ventricular irradiation with local boost" — PMID: 34724065

5. Circulating miR-371a-3p is a sensitive biomarker for malignant GCTs — but blind to teratoma

MicroRNAs of the miR-371~373 and miR-302/367 clusters are over-expressed in all malignant GCTs; miR-371a-3p is elevated in serum and CSF at diagnosis and outperforms AFP and β-hCG on sensitivity/specificity. In intracranial cases, CSF miR-371a-3p has preceded histologic diagnosis by up to 2 years and detected relapse when conventional markers were below threshold (PMID: 32642701). A critical caveat: miR-371a-3p is expressed in undifferentiated GCT but not in teratoma, so it cannot detect mature teratoma components — the same blind spot that underlies growing teratoma syndrome. [Human/clinical]

"Circulating miR-371a-3p, which is expressed in undifferentiated TGCTs but not in teratomas, is a promising biomarker for TGCTs" — PMID: 38396829

6. Growing teratoma syndrome is a distinct marker-negative phenomenon

Growing teratoma syndrome (GTS) is the paradoxical enlargement of teratomatous components during or after chemo-/radiotherapy despite normalized or negative tumor markers, typically with honeycomb/cystic imaging. It reflects therapy selecting for and unmasking differentiated, low-proliferation teratoma rather than treatment failure. In one pineal mixed GCT, the Ki-67 index fell from 25% at diagnosis to 5% after resection, confirming differentiation to mature teratoma (PMID: 42488730). Methylation classifiers can confirm the teratoma diagnosis, and up to 45% of presumed immature-teratoma patients experience growing disease during treatment (PMID: 42095539). Surgical resection is the mainstay of GTS management. [Human/clinical]

"It manifests as paradoxical growth of teratomatous components, with multiple cystic lesions on cranial imaging despite normalized tumor markers" — PMID: 39109622

7. Klinefelter syndrome and sex-chromosome aneuploidy are established genetic risk factors

Males with Klinefelter syndrome (47,XXY) have an elevated incidence of pineal and suprasellar germinomas. A FISH study of 13 male intracranial GCT patients found KS in 15% and statistically significant X and Y chromosome polyploidies in tumor versus non-tumor tissue (PMID: 18758161). X-chromosome polyploidy and X hypomethylation have been proposed as transformation mechanisms. A birth-defect/GCT case-control study (Schraw et al., 552 cases vs 6,380 controls) found GCT risk increased among children with any birth defect (OR 1.7; 95% CI 1.3–2.4) and markedly so with syndromic defects (OR 10.4; 95% CI 4.9–22.1) (PMID: 37366624). [Human/clinical]

"KS was found in 15% of the cases, demonstrating that this constitutive aneuploidy may be related to carcinogenesis. When tumor and non-tumor tissues were compared, statistically significant X and Y chromosome polyploidies in tumors were revealed" — PMID: 18758161

8. Germinoma has an immune-cell-rich microenvironment with high PD-1/PD-L1 expression; immune balance is prognostic

Germinoma frequently shows massive immune infiltration. In 100 germinomas, PD-1 (PDCD1) was expressed by immune cells in 93.8% and PD-L1 (CD274) in tumor cells in 73.5%; higher immune infiltration (lower tumor-cell content) predicted longer PFS (P = 0.03) (PMID: 31179566). In a 90-patient CNS GCT cohort, germinomas had higher CD4+/Foxp3+ infiltration and CTLA-4 than NGGCT, PD-1/PD-L1 in >90%, and PD-1 expression was an independent prognostic factor for PFS/RFS (PMID: 39958339). PD-L1 tumor-cell ratio has also been associated with faster tumor growth. These data provide a rationale for checkpoint-inhibitor trials. [Human/clinical]

"PD1 (PDCD1) was expressed by immune cells present in most germinomas (93.8%), and PD-L1 (CD274) expression was found in tumour cells in the majority of germinomas examined (73.5%)" — PMID: 31179566

9. Platinum hypersensitivity depends on p53/apoptotic response; resistance involves miR-371-373, OCT4 loss, and PI3K/AKT (largely testicular/in vitro evidence)

In testicular embryonal carcinoma cell lines, cisplatin triggers a p53-dominant transcriptional response (~54% of upregulated genes are p53 targets), and p53 knockdown confers relative resistance (PMID: 15940259). Sensitivity reflects DNA-repair deficits (interstrand crosslink / homologous recombination) plus hypersensitive p53-mediated apoptosis (Noxa/Puma/Fas via p73/Sp1). Resistance mechanisms include OCT4 down-regulation, failure to induce Puma/Noxa, altered microRNAs (miR-17/-106b, miR-302a, miR-371–373), elevated MDM2, cytoplasmic p21, and PDGFRβ/PI3K/pAKT activation (PMID: 25546083). Evidence-type caveat: these are predominantly testicular and in-vitro data (including cell lines such as NCCIT), not direct patient CNS evidence, and must be labeled as such. [In vitro / testicular surrogate]

"changes in the expression levels of micro-RNAs such as miR-17/-106b, miR-302a, or miR-371 to -373; elevated levels of MDM2 and cytoplasmic translocation of p21 by phosphorylation; and activation of the PDGFRβ/PI3K/pAKT pathway" — PMID: 25546083

10. Mouse models center on the 129-strain testicular teratoma and germ-cell pluripotency genes

The 129 mouse strain spontaneously develops testicular teratomas; the Ter mutation in the Dnd1 gene is a potent modifier of tumor incidence (PMID: 23784831). Additional models include the 129-Chr19(MOLF) chromosome-substitution strain and conditional Dmrt1 and Pten alleles. Teratomas arise from germ cells via misregulation of pluripotency genes (Oct4, Sox2, Nanog). Model limitation: these are gonadal (testicular) models; no faithful model of the intracranial midline GCT microenvironment currently exists, and the role of somatic/physiologic context in teratoma sensitivity remains unknown. [Model organism]

"Leroy Stevens identified the 129 mouse strain as a model of spontaneous testicular teratoma and later isolated a substrain carrying the Ter mutation, a potent modifier of tumor incidence" — PMID: 23784831

11. Clinical presentation is location-dependent

Suprasellar/neurohypophyseal germinomas present with central diabetes insipidus (polyuria/polydipsia), hypopituitarism, growth failure, and visual defects; DI can precede diagnosis by >1 year (42% with symptom interval >1 yr) and is accompanied by loss of the posterior pituitary "bright spot" on MRI (PMID: 25266413). Pineal lesions cause Parinaud syndrome (upgaze palsy) and obstructive hydrocephalus. Germinoma constitutes 50–65% of cerebral GCTs. Bifocal (pineal + suprasellar) disease is treated as locoregional rather than metastatic (PMID: 16530340). [Human/clinical]

"All had symptoms of DI at presentation with a symptom interval above one year in eight cases (42 %)" — PMID: 25266413

12. Diagnosis integrates tumor markers, MRI of brain and spine, CSF cytology, and often biopsy

Serum and/or CSF AFP (yolk sac tumor / immature teratoma) and β-hCG (choriocarcinoma / syncytiotrophoblast) help identify and subclassify GCTs; markedly elevated markers permit marker-based diagnosis without biopsy (e.g., β-hCG >50 IU/L, AFP >25 ng/mL thresholds in the CNS sGCT pilot). Pure germinoma is typically marker-negative or low β-hCG. Staging requires contrast-enhanced MRI of brain and whole spine plus CSF cytology (PMID: 37452948). A key pitfall: intracranial dysgerminoma/germinoma can mimic inflammatory/demyelinating disease (oligoclonal bands, steroid-responsive) and be marker-negative (PMID: 31712009). Emerging minimally-invasive tools include CSF cfDNA methylation classifiers and miR-371a-3p. [Human/clinical]

"Staging work-up includes CSF cytology for tumor cells and contrast-enhanced MRI of brain and spine for macroscopic metastasis before treatment commences." — PMID: 37452948

13. Incidence and demographics quantified

A Kumamoto (Japan) survey reported a pediatric CNS-GCT age-adjusted annual incidence of 0.45/100,000 children (boys 0.64, girls 0.28; M:F 2.29:1), versus CBTRUS 0.18, SEER 0.15, and Germany 0.10 per 100,000 (PMID: 24751890). GCTs were 44.3% of cases aged 0–14; germinoma 64.5% vs nongerminoma 35.5%; pineal location 45.2%. Historically incidence is 5–8× higher in Japan/East Asia than Western countries, with a pubertal peak and overall M:F ~3–4:1 (higher for pineal) (PMID: 24896186). [Human/clinical]

"The age-adjusted annual incidence rate was 0.45 cases (boys: 0.64, girls: 0.28) per 10(5) children. At 2.29, the ratio of CNS-GCTs was higher in these boys than girls." — PMID: 24751890

14. Relapse prognosis is subtype-dependent; salvage with HDCT + autologous SCT cures a subset

In KSPNO S-053, relapsed/progressed CNS-GCT treated with myeloablative high-dose chemotherapy and autologous stem cell transplant (± radiotherapy) achieved 3-year OS 59.1 ± 11.2% overall, markedly better for germinoma (88.9 ± 10.5%) than NGGCT (36.4 ± 14.5%; P = 0.028) (PMID: 23824533). Radiotherapy — particularly craniospinal — was associated with better outcome. Late spinal relapses have occurred 8–18 years after remission, mandating prolonged surveillance. [Human/clinical]

"The probability of 3-year overall survival was 59.1 ± 11.2 % (36.4 ± 14.5 % for NGGCTs vs. 88.9 ± 10.5 % for germinomas, P = 0.028)" — PMID: 23824533

15. Radiation FIELD, not just dose, controls germinoma relapse

Yamasaki et al. (57 iGCTs, mostly local irradiation) found that for pure germinomas 8 of 9 relapses occurred OUTSIDE the irradiation fields, with local RT alone giving 5-yr PFS 75% ± 8.8% — insufficient without intensification (PMID: 32398600). Whole-ventricular field coverage reduced recurrence dramatically (HR 0.060; 95% CI 0.012–0.312; p < 0.001) (PMID: 42243616). Kortmann established that chemotherapy converts macroscopic to microscopic disease, permitting dose reduction to the tumor and ventricular system while maintaining field coverage — and that chemotherapy alone cannot replace radiotherapy (PMID: 24224870). This cleanly separates the field question (must cover ventricles) from the dose question (can be reduced). [Human/clinical]

"8 of 9 relapses from 24 PGNs occurred outside irradiation fields, with a 5-year progression-free survival (5-year PFS) of 75%±8.8%" — PMID: 32398600

16. Origin models: ectopic PGC (germinoma) vs embryonic/pluripotent cell (NGGCT/teratoma)

Two co-existing theories persist. The germ-cell theory points to germinoma's PGC-like methylation/transcriptome, KIT expression, and PGC-marker overlap. The embryonic cell theory holds that IGCTs arise from pluripotent embryonic cells that escape normal migration and differentiation, better explaining non-germinomatous and teratomatous elements (PMID: 42419530). Pineal-region tumors are thought to arise from ectopic PGCs and cells of adjacent structures (PMID: 37831207). Because mixed-tumor components share driver mutations from a common ancestral clone and show early branched divergence, resemblance must be interpreted as cell-state similarity, not lineage tracing. [Human/computational]

"The embryonic cell theory suggests that IGCTs may originate from pluripotent embryonic cells that escape normal migration and differentiation during embryonic development" — PMID: 42419530

17. Survivors face substantial treatment-related late morbidity

Because germinoma is highly curable, the clinical focus has shifted to reducing sequelae: permanent hypopituitarism/diabetes insipidus (often irreversible — the bright spot does not recover), radiation-induced cavernous malformations years after whole-ventricular/craniospinal RT causing hemorrhage and neurologic deficit (PMID: 40347128), and endocrine/visual dysfunction and loss of social independence after higher-dose or repeat radiation (PMID: 36610798). Late spinal relapse up to 18 years mandates lifelong surveillance. [Human/clinical]

"an intracranial germinoma treated with whole-ventricular irradiation. Three years after treatment, the patient developed a symptomatic hemorrhagic RICM" — PMID: 40347128

Section-by-Section Disease Characterization

Section 1 — Disease Information

CNS GCT (MONDO:0003000) is an umbrella for germ-cell-derived neoplasms of the CNS, overwhelmingly intracranial and midline (pineal, suprasellar/neurohypophyseal, bifocal, basal ganglia, ventricular; spinal primaries rare). Subtypes: germinoma (dysgerminoma equivalent) and NGGCT (embryonal carcinoma, yolk sac tumor, choriocarcinoma, teratoma [mature/immature], mixed). Synonyms: intracranial germ cell tumor (IGCT), primary CNS GCT, intracranial germinoma. Identifiers: MeSH "Neoplasms, Germ Cell and Embryonal"; ICD-O germ-cell histology codes; Orphanet intracranial GCT entries. Evidence is a mix of aggregated disease-level resources and clinical cohort/registry data (PMID: 37452948).

Section 2 — Etiology

Primary drivers are somatic KIT/RAS/MAPK and AKT/PI3K/mTOR activation plus chromosomal instability (12p gain). Genetic risk: Klinefelter syndrome (47,XXY), sex-chromosome aneuploidy, birth defects/syndromes (OR up to 10.4). Rare germline variants in JMJD1C. No confirmed environmental or infectious cause; no established protective factors. Gene–environment interaction data are lacking (PMID: 24896186, PMID: 18758161, PMID: 37366624).

Section 3 — Phenotypes

Location-dependent: central DI (HP:0000863), hypopituitarism (HP:0040075), growth delay (HP:0001510), hydrocephalus (HP:0000238), Parinaud/upgaze palsy (HP:0000602), visual impairment (HP:0000505), precocious puberty (HP:0000826, β-hCG-secreting). Onset childhood/adolescent; progression subacute-to-chronic; DI frequently precedes diagnosis by >1 year (diagnostic delay). QoL impact dominated by endocrine and visual sequelae (PMID: 25266413).

Section 4 — Genetic/Molecular Information

Recurrent somatic drivers: KIT, KRAS, NRAS, CBL (MAPK); AKT1 gain, PI3K/mTOR; BCORL1 LoF; NF1 (poor prognosis). Germline: JMJD1C enrichment, KS/aneuploidy. Epigenetics: germinoma global hypomethylation (signature). Chromosomal: 12p gain, X/Y polyploidy. Somatic > germline for drivers (PMID: 24896186, PMID: 38409885, PMID: 28078450).

Section 5 — Environmental Information

No robustly established environmental, lifestyle, or infectious cause. This section is not applicable / not established for CNS GCT beyond the genetic/developmental risk factors above.

Section 6 — Mechanism / Pathophysiology

Upstream: developmental mis-location of a germ-cell/pluripotent progenitor + MAPK (GO:0000165) or PI3K/AKT (GO:0043491)/mTOR (GO:0031929) driver → proliferation. Germinoma retains a PGC-migration-phase state with DNA demethylation (GO:0080111) and an immune-rich, PD-1/PD-L1-high microenvironment. NGGCT differentiates along embryonal/extraembryonic lineages, secreting AFP/β-hCG. Downstream clinical manifestations arise from location and mass effect (hydrocephalus, DI). Cell types: primordial germ cell (CL:0000670), pluripotent stem cell (CL:0002248), infiltrating T cells (CL:0000084) (PMID: 28078450, PMID: 31179566).

Section 7 — Anatomical Structures Affected

Primary: pineal gland (UBERON:0001905), neurohypophysis/posterior pituitary (UBERON:0002198), hypothalamus (UBERON:0001898), third/lateral ventricles (UBERON:0002285/0002286), basal ganglia (UBERON:0002420); spinal cord (UBERON:0002240) rare. Body system: nervous/endocrine. Lateralization: often midline/bilateral (bifocal) (PMID: 37452948).

Section 8 — Temporal Development

Onset pediatric/adolescent, bimodal (infancy + adolescence); insidious-to-subacute. Germinoma highly curable; NGGCT more aggressive. Course: treatment-induced remission common; late relapse (spinal) up to 18 years. Critical intervention window is at diagnosis and during marker/imaging surveillance (PMID: 37452948, PMID: 42243616).

Section 9 — Inheritance and Population

Incidence 0.45/100,000 children (Japan) vs 0.10–0.18 (West); M:F ~2.3–4:1. Mostly sporadic somatic; heritable risk via KS/aneuploidy and syndromic birth defects. No classical Mendelian inheritance pattern (PMID: 24751890, PMID: 18758161).

Section 10 — Diagnostics

Serum/CSF AFP + β-hCG; MRI brain + whole spine; CSF cytology; biopsy when markers non-diagnostic. Emerging: CSF cfDNA methylation classifier, miR-371a-3p (blind to teratoma). Differential: inflammatory/demyelinating disease (PMID: 37452948, PMID: 32642701, PMID: 31712009).

Section 11 — Outcome/Prognosis

Localized germinoma >90% 5-yr EFS; metastatic germinoma near 100% OS with CSI; NGGCT >70%. Relapse: germinoma salvage OS ~89% vs NGGCT ~36%. Prognostic factors: subtype (germinoma vs NGGCT), NF1 mutation, PD-1 expression, immune infiltration, extent of RT field. Late morbidity substantial (PMID: 34724065, PMID: 23824533).

Section 12 — Treatment

Germinoma: platinum-based chemotherapy (carboPEI/carboplatin+etoposide; CHEBI: carboplatin CHEBI:31355, etoposide CHEBI:4911) + whole-ventricular RT (MAXO:0000009) with dose reduction. NGGCT: intensified chemo + CSI/boost ± second-look surgery. Salvage: high-dose chemo + autologous SCT ± CSI. Emerging: KIT and PI3K/AKT/mTOR targeted therapy; PD-1/PD-L1 checkpoint blockade. Surgery (MAXO:0000006) for GTS/residual teratoma; endocrine hormone replacement (PMID: 34724065, PMID: 39959669).

Section 13 — Prevention

No primary prevention (no modifiable cause). Secondary prevention = early detection via marker/imaging surveillance and awareness of DI as a sentinel symptom. Tertiary = reducing RT field/dose to limit late effects; lifelong surveillance for late relapse and second tumors (PMID: 25266413, PMID: 36610798).

Section 14 — Other Species / Natural Disease

Human disease primarily; comparative biology via murine testicular teratoma (NCBI Taxon 10090). Orthologous genes: Kit, Kras, Akt1, Dnd1, Dmrt1, Pten. No significant naturally-occurring intracranial GCT reported in companion animals. Not zoonotic (PMID: 23784831).

Section 15 — Model Organisms

Mouse (129 strain, Ter/Dnd1, Dmrt1, Pten conditionals); teratomas via pluripotency-gene misregulation. Recapitulates teratoma initiation but not intracranial location, germinoma hypomethylation, or NGGCT secretion. Resources: MGI, IMSR. In-vitro surrogates: testicular EC cell lines (e.g., NCCIT) for chemosensitivity/resistance (PMID: 23784831, PMID: 25546083).

Mechanistic Model / Interpretation

The findings cohere into a developmental-origin model in which a single mis-located progenitor cell acquires a KIT/RAS/MAPK or PI3K/AKT/mTOR driver mutation and then diverges into distinct developmental states that define the histologic subtypes:

   Embryonic development
    │
   Ectopic/mis-migrated progenitor  ── acquires KIT/RAS/MAPK or AKT/PI3K driver
    │                              (± 12p gain, chromosomal instability)
    ▼
   ┌────────────────────────── COMMON ANCESTRAL CLONE ──────────────────────────┐
   │        (early BRANCHED divergence into different developmental STATES)      │
   ▼                                                                             ▼
 GERMINOMA                                                   NON-GERMINOMATOUS GCT
 • PGC-migration-phase state                                 • embryonic/pluripotent state
 • GLOBAL DNA HYPOMETHYLATION                                • EC / yolk sac / choriocarcinoma /
 • KIT-enriched, marker-poor                                   teratoma / mixed
 • immune-rich (PD-1/PD-L1 high)                             • AFP/β-hCG secreting
 • exquisitely radiosensitive                               • teratoma = miR-371 blind spot
    │                                                             │
    ▼                                                             ▼
 Chemo + whole-VENTRICULAR RT (dose-reduced)                Intensified chemo + CSI/boost
 >90% 5-yr EFS                                              ~70–90% 5-yr EFS
    │                                                             │
    ▼                                                             ▼
 Relapse mostly OUT-OF-FIELD                                GROWING TERATOMA SYNDROME
 (field, not dose, matters)                                (marker-negative, Ki-67 falls,
 Salvage HDCT+SCT OS ~89%                                   surgery is mainstay)
                                                   Salvage HDCT+SCT OS ~36%

Three post-treatment phenomena must be kept conceptually separate:

Phenomenon Markers Biology Management
Growing teratoma syndrome Negative/normalized Therapy unmasks differentiated, low-Ki-67 teratoma Surgical resection
Chemo-selection of viable malignant component May rise Resistant malignant clone survives therapy Intensified systemic therapy
True relapse Variable (marker or miR-371 rise) Regrowth of malignant clone, often out-of-field Salvage HDCT + SCT ± CSI

The upstream trigger is developmental mis-location plus a MAPK/PI3K driver; the downstream clinical manifestations (hydrocephalus, DI, Parinaud syndrome) are consequences of tumor location and mass effect. Germinoma's global hypomethylation is both a diagnostic signature and a plausible mechanistic link to its PGC-like state and immune-rich microenvironment.

Evidence Base

PMID Topic Supports
28078450 Genome-wide methylation of iGCTs Germinoma hypomethylation, PGC state, common ancestral clone
24896186 Novel mutations (Wang, Nature) KIT/RAS >50%, AKT1 gain 19%, incidence, sex ratio
38409885 WES in Chinese iGCTs KIT 32% (germinoma 65% vs NGGCT 7%), NF1, clonal evolution
38012690 Genetics/epigenetics/immune review Dual-pathway activation, 12p gain
39959669 Genomic diagnostics/therapeutics MAPK activation, KIT as target
34724065 EANO/SNO/EURACAN consensus >90% EFS localized germinoma; NGGCT >70%
42234858 SIOP-CNS-GCT-II final report EFS 0.94/OS 0.98; boost omission safe
42243616 Long-term outcomes/recurrence Whole-ventricular field HR 0.060
32398600 Local RT + IT MTX/HDCT 8/9 relapses out-of-field
24224870 Management (Kortmann) Chemo converts macro→micro; dose reduction
32642701 miR-371a-3p in iGCT Sensitive biomarker
38396829 microRNAs / teratoma challenge miR-371 teratoma blind spot
39109622 / 42488730 Growing teratoma syndrome Marker-negative growth; Ki-67 25%→5%
42095539 Presumed immature teratoma 45% growing disease during treatment
18758161 / 37366624 KS/aneuploidy; birth defects Genetic risk factors
31179566 / 39958339 Immune landscape PD-1/PD-L1, prognostic infiltration
25546083 / 15940259 Cisplatin sensitivity/resistance p53 hypersensitivity; miR/PI3K resistance (testicular/in vitro)
23784831 Testicular teratoma models 129-strain, Dnd1/Ter, Dmrt1, Pten
25266413 / 16530340 DI/bright spot; bifocal Presentation; bifocal-as-locoregional
37452948 / 37831207 Reviews Staging, presentation, origin
24751890 Kumamoto incidence survey 0.45/100,000, M:F 2.29:1
23824533 KSPNO S-053 salvage Relapse OS 89% germinoma vs 36% NGGCT
36610798 / 40347128 Late effects Endocrine/visual morbidity; radiation cavernoma
42419530 Advances/future directions Embryonic-cell origin theory
34074342 External metastasis / review Geographic prevalence 15.3% vs 3.6%

Ontology Term Suggestions

  • Disease: MONDO:0003000 (CNS germ cell tumor).
  • Anatomy (UBERON): pineal gland (UBERON:0001905), posterior pituitary (UBERON:0002198), hypothalamus (UBERON:0001898), third ventricle (UBERON:0002285), lateral ventricle (UBERON:0002286), basal ganglia (UBERON:0002420), spinal cord (UBERON:0002240).
  • Cell types (CL): primordial germ cell (CL:0000670), pluripotent stem cell (CL:0002248), T cell (CL:0000084), regulatory T cell (CL:0000815).
  • Biological process (GO): MAPK cascade (GO:0000165), PI3K/AKT signaling (GO:0043491), TOR signaling (GO:0031929), DNA demethylation (GO:0080111), germ cell migration (GO:0008354), apoptotic process (GO:0006915).
  • Phenotype (HPO): Central diabetes insipidus (HP:0000863), Hypopituitarism (HP:0040075), Hydrocephalus (HP:0000238), Ophthalmoplegia (HP:0000602), Precocious puberty (HP:0000826), Growth delay (HP:0001510), Visual impairment (HP:0000505).
  • Chemicals (CHEBI): cisplatin (CHEBI:27899), carboplatin (CHEBI:31355), etoposide (CHEBI:4911), ifosfamide (CHEBI:5864).
  • Treatments (MAXO): radiotherapy (MAXO:0000009), chemotherapy (MAXO:0000058), surgical resection (MAXO:0000006), hematopoietic stem cell transplantation, hormone replacement therapy.

Limitations and Knowledge Gaps

  1. CNS umbrella vs intracranial evidence base. Virtually all molecular, treatment, and outcome data derive from intracranial cohorts. Primary spinal CNS GCTs remain essentially uncharacterized and must not be assumed to share the intracranial biology.
  2. Origin is unresolved. Both PGC and embryonic-cell models rest on state resemblance (methylation, transcriptome, markers), not lineage tracing. No experiment has directly demonstrated the human cell of origin.
  3. Cross-context evidence conflation. Much resistance biology (p53, miR-371–373, PI3K/AKT, OCT4) comes from testicular tumors and cell lines (e.g., NCCIT) rather than patient CNS tissue; direct CNS-GCT resistance data are sparse.
  4. No faithful intracranial model. Available mouse models are gonadal (testicular teratoma); they do not reproduce the intracranial midline microenvironment, germinoma hypomethylation, or NGGCT secretion.
  5. Field-versus-dose not fully resolved. Although field coverage clearly matters, optimal ventricular field boundaries and minimal effective dose remain formally controversial.
  6. Biomarker gaps. miR-371a-3p is blind to teratoma; no circulating marker reliably detects mature teratoma or GTS.
  7. Statistical fragility. Rarity yields small cohorts; some regional incidence differences and single-variant associations (e.g., KRAS Q61L) rest on very few cases.

Proposed Follow-up Experiments / Actions

  1. Spinal-primary characterization. Assemble a dedicated cohort of primary spinal CNS GCTs for methylation/WES to test whether intracranial biology generalizes.
  2. Lineage-discriminating single-cell/spatial studies. Apply single-cell multi-omics and spatial transcriptomics to mixed tumors to map the branched clonal trajectory and directly test PGC-state vs embryonic-state origin rather than bulk resemblance.
  3. CNS-specific resistance modeling. Derive intracranial GCT organoids/patient-derived models to test whether testicular resistance mechanisms (miR-371–373, PI3K/AKT, OCT4 loss) operate in CNS disease.
  4. Prospective CSF liquid biopsy. Validate combined CSF cfDNA methylation classifier + miR-371a-3p for diagnosis, minimal-residual-disease monitoring, and discrimination of GTS vs true relapse (noting the teratoma blind spot).
  5. Targeted-therapy trials. Test KIT inhibitors in KIT-mutant germinoma, PI3K/AKT/mTOR inhibitors in AKT1-altered tumors, and PD-1/PD-L1 checkpoint blockade in immune-rich germinoma, ideally as radiation-sparing strategies.
  6. Prospective field-vs-dose randomization. Formally test ventricular field boundaries and dose de-escalation to minimize late morbidity while preserving out-of-field control.
  7. Long-term survivorship registries. Systematically capture endocrine, neurocognitive, vascular (cavernoma), and second-tumor outcomes, with surveillance extending ≥18 years to capture late spinal relapse.

Report compiled from 18 confirmed findings across 70 reviewed papers over 5 investigation iterations. Evidence types are labeled throughout: [Human/clinical] (cohorts, trials, registries), [Computational/genomic] (methylation/genomic classifiers), [In vitro / testicular surrogate] (cell lines), and [Model organism] (mouse). Molecular resemblance to primordial germ cells is interpreted as cell-state similarity rather than proven lineage.

Artifacts