KEY POINTS
- This retrospective single-center study from Taipei Medical University, Taiwan, analyzed 29 children and young adults younger than 30 years treated for central nervous system germ cell tumors between 2015 and 2024. The cohort included 13 pure germinomas, 14 malignant nongerminomatous germ cell tumors (MNGGCTs), and two mature teratomas. Among MNGGCT patients, seven received upfront radiotherapy followed by chemotherapy, and seven received induction chemotherapy followed by radiotherapy. Median follow-up was 6.0, 6.6, and 3.7 years for the germinoma, upfront-RT MNGGCT, and upfront-chemotherapy MNGGCT groups, respectively.
- Pure germinoma patients received upfront whole-ventricular irradiation (WVI) with a tumor-bed boost. WVI doses ranged from 23.4 to 30.0 Gy, followed by a boost to a total tumor-bed dose of 30.0–36.0 Gy. Twelve of 13 patients received no chemotherapy, while one received platinum-based chemotherapy following a partial response to RT. Treatment planning incorporated whole-ventricular and tumor-bed target volumes with image-guided PTV margins of approximately 3 mm.
- Five-year event-free survival (EFS) and overall survival (OS) were both 100% in the 13 pure germinoma patients receiving upfront radiotherapy. No event or death was reported in this group during the analyzed follow-up. These results support the radiosensitivity of CNS germinoma and the feasibility of a radiation-first approach in selected patients, but the cohort did not include a chemotherapy-first germinoma control group and cannot establish equivalence to contemporary combined-modality protocols.
- Among patients with malignant nongerminomatous tumors, observed outcomes differed between treatment sequences. Five-year EFS was 85.7% with upfront RT versus 41.7% with upfront chemotherapy, while corresponding five-year OS was 85.7% versus 60.0%, respectively. Both groups included only seven patients, and all eventually received chemotherapy and radiotherapy. Tumor-bed prescriptions were typically 50.4–55.8 Gy, with WVI, whole-brain irradiation, or craniospinal irradiation selected according to disease characteristics. These descriptive differences do not demonstrate that upfront RT is superior.
- Patterns of failure highlighted the risk of neuraxis dissemination in MNGGCT. One patient receiving upfront whole-brain RT with tumor-bed boost developed spinal/leptomeningeal dissemination and died of progressive disease. In the upfront-chemotherapy group, two patients experienced relapse involving the suprasellar region or spinal axis, and another had radiation necrosis recorded as a treatment-related event. These observations emphasize the importance of neuraxis assessment and careful treatment-volume selection, although the small numbers prevent reliable comparisons between WVI and craniospinal irradiation.
- Endocrine dysfunction was frequent among survivors. Following treatment, growth hormone deficiency was reported in 46.2% of germinoma patients and 57.1% of MNGGCT patients, while central hypothyroidism affected 23.1% and 42.9%, respectively. Secondary adrenal insufficiency occurred in 7.7% and 35.7%. Neuropsychological assessment was available for only nine patients, with limited and inconsistently timed repeated testing, preventing reliable conclusions about long-term cognitive differences between treatment strategies.
- The study is not sufficient to change CNS germ cell tumor treatment sequencing. The retrospective cohort was small, treatment selection was nonrandomized, follow-up differed substantially between MNGGCT groups, and some diagnoses relied on imaging and tumor markers rather than histopathology. The very small number of events prevented robust adjusted comparisons. Furthermore, long-term neurocognitive, endocrine, and secondary malignancy risks remain important considerations when evaluating radiotherapy-first approaches in pediatric and adolescent patients.
CLINICAL TAKEAWAY
Upfront whole-ventricular radiotherapy with a tumor-bed boost achieved excellent observed disease control in a small cohort of patients with pure CNS germinoma, most of whom did not receive chemotherapy. In malignant nongerminomatous disease, numerically better outcomes after upfront RT remain hypothesis-generating because only seven patients were treated per sequence and clinical comparability was not established. Larger prospective studies are required before altering established response-adapted treatment protocols or reducing systemic therapy based on these findings.